C. Gopalan Auth., Prof. H. K. A. Visser, Dr. J. G. Bindels Eds.

cmLD NUTRITION IN SOUTH EAST ASIA
Eighth Nutricia Symposium
CHILD NUTRITION
IN SOUTH EAST ASIA
Yogyakarta, 4-6 April 1989
Edited by
H.K.A. VISSER and J.G. BINDELS
Kluwer Academic Publishers

DORDRECHT / BOSTON / LONDON
Library of Congress CataIoging-in-PubJication Data
Nutricia Symposium (8th: 1989: Yogyakarta, Indonesia)

Child nutrition in South East Asia: Eighth Nutricia Symposium /
edited by H.K.A. Visser and J.O. Bindels.
p. cm.
1. Children-Asia, Southeastern-Nutrition-Congresses.
2. Pregnant women-Asia, Southeastern-Nutrition-Congresses.
3. Malnutrition in children-Asia, Southeastern-Congresses.
I. Visser, H.K.A. (Hendrik Klaas Aldert) II. Bindels, J.O.
III. Title.
RJ206.N816 1989
618.92'39'00959-<ic20 90-4133
ISBN-13: 978-94-010-7393-6 e-ISBN-13: 978-94-009-1996-9

DOl: 10.1007/978-94-009-1996-9
Published by Kluwer Academic Publishers,

P.O. Box 17,3300 AA Dordrecht, The Netherlands.
Kluwer Academic Publishers incorporates

the publishing programmes of
D. Reidel, Martinus Nijhoff, Dr. W. Junk and MTP Press.
Sold and distributed in the U.S.A. and Canada

by Kluwer Academic Publishers,
101 Philip Drive, Norwell, MA 02061, U.S.A.
In all other countries, sold and distributed

by Kluwer Academic Publishers Oroup,
P.O. Box 322, 3300 AH Dordrecht, The Netherlands.
Printed on acid-free paper
All Rights Reserved

© 1990 by Kluwer Academic Publishers
Softcover reprint of the hardcover 1st edition 1990
No part of the material protected by this copyright notice may be reproduced or utilized in any form or by
any means, electronic or mechanical, including photocopying, recording or by any information storage
and retrieval system, without written permission from the copyright owner.
Preface
Exactly 25 years after the first Nutricia Symposium was held in Groningen, the
Netherlands, dealing with fundamental aspects of neonatology, the 8th Nutricia
Symposium was held at nearly the opposite end of the globe and dealt with a number
of most important aspects of child nutrition in South East Asia.
It was the intention of this symposium to bring together a respectable number of
specialists in this field from the countries involved, complemented by a number of
experts from outside the area chosen because of their recognised contributions to the
main topics of the symposium. In this way, a platform for discussion and cooperation
was provided to bring the specific problems of child nutrition in South East Asia
closer to a solution. From the contents of the papers and also the valuable discus-
sions, it should be clear to the reader that the goal of this symposium has been
attained.
It would have been difficult to find a place which was better suited for this
symposium than Yogyakarta, Indonesia. The excellent work done by the several
local committees in organising this symposium is greatly acknowledged, especially
the efforts of Prof. H. Iskandar Wahidiyat, Prof. Darwin Karyadi, Prof. I.G.N. Gde
Ranuh, Prof. Moelyono S. Trastotenoyo, Prof. S.H. Pudjiadi, Dr. Samsudin, Dr.
Suharyono, Dr. Yati Sunarto and Dr. A.G. Soemantri need to be mentioned.
Since the proceedings of this symposium consist mainly of contributions by local
specialists in the field, it seems logical that this book will find its way allover South
East Asia. However, one would be foolish to think that this book is only of limited
value for the other parts of the world. Many problems indicated here have a much
wider relevance than to South East Asia only. For instance, it is of great interest to
find in this book a large number of realistic descriptions of the actual practice of
breast feeding in this region. The fact that one of the largest infant food manufac-
turers, Nutricia (fully endorsing the WHO code) made this symposium possible
should be interpreted as an awareness of its great responsibility in this respect for
child health in South East Asia, as well as in other parts of the world.
March 1990 THE EDITORS

Contents
Preface v
List of participants (+ photo) x
List of authors and editors Xlll
OPENING OF THE SYMPOSIUM xv

H.K.A. Visser
SESSION 1
GENERAL ASPECTS OF CHILD NUTRITION IN SOUTH EAST ASIA
Chairman: H. Iskandar Wahidiyat
C. Gopalan, Childhood malnutrition in South East Asia: Concepts and

strategies for control 3
D. Karyadi, Vitamin A problems in South East Asia with special reference
to its impact on child health 21
L. Mata, Infection and malnutrition 35
S. Surbakti, Y.K. Husaini and M.A. Husaini, The pattern of feeding and the
nutritional status of infants and children in Indonesia 45
Discussion 55
SESSION 2
NUTRITION OF PREGNANT MOTHER AND INFANT
Chairman: Moelyono S. Trastotenoyo
J.A. Kusin, S. Kardjati, C. de With and W.M. van Steenbergen, The East
Java pregnancy study: Effect of prenatal energy supplementation on
mother and infant 63
S.H. Pudjiadi, Weaning foods in infant nutrition 89
J.I. Counsilman, Ethnic differences in breast-feeding among well-to-do
Singaporeans 95
M. Gracey and H. Sullivan, A prospective study of growth and nutrition of
Aboriginal children from birth to two years in North-West Australia 107
Discussion 117
viii
SESSION 3
EARLY CHILDHOOD MALNUTRITION AND ITS LATE EFFECTS,
WITH EMPHASIS ON PREVENTION
Chainnan: I.G.N. Gde Ranuh
C.A Powell and S.M. Grantham-McGregor, Selective review of studies on

the behavioural effects of childhood malnutrition 125
P.F. Heywood and AH. Heywood, Protein-energy malnutrition in Papua New
Guinea: Its functional significance 141
A Alisjahbana, R. Peeters, M. Maertens, S. Tanuwidjaja, H. Tjokrohusodo, W.
Ngantung and D.A Primana, Mother and child health at village level: The
effect of the mother and child card. 149
Discussion 159
SESSION 4
NUTRITIONAL DEFICIENCIES IN CHILDREN
Chainnan: D. Karyadi
AG. Soemantri Hardjojoewono, Iron deficiency anemia and its problems 169
T. Sadjimin, Iron deficiency and infant development 177
P. Bhaskaram, Infection and immunity of vitamin A and iron deficient children 185
I. Sudigbia, Supplementary feeding in childhood diarrhea 199
Discussion 207
SESSION 5
DIARRHEA AND MALABSORPTION IN CHILDREN
Chainnan: Yati Soenarto
M.G.M. Rowland, Diarrhea, growth and feeding practices 217

AM. Molla and Ayesha Molla, Dietary management of acute diarrhea: A
scientific basis 227
Suharyono, Chronic diarrhea and its management: Breast-feeding and short
chain peptide 237
C.-Y. Yeung, F.H.W. Wong and A.Y.C. Tam, Breath hydrogen test for lactose
malabsorption in Chinese children and infants 243
Discussion 249
ix
SESSION 6
MEDICAL TEACHING ON ClllLD NUTRITION
Chairman: A.S. Ongkie
Pitono Soeparto, Integrated medical teaching curriculum on child nutrition 257
Discussion 267
CLOSING REMARKS 273

H.K.A. Visser
INDEX OF SUBJECTS 279

List of participants
Dr. (Mrs.) Anna Alisjahbana, Bandung, Indonesia Prof. A.M. Molla, Karachi, Pakistan
Dr. (Mrs.) P. Bhaskararn, Hyderabad, India Prof. Muzief Munir, Manado, Indonesia
Dr. Dradjat Budiman, Yogyakarta, Indonesia Prof. A.S. Ongkie, Manado, Indonesia
Dr. Kou Liang Chang, Taipei, Taiwan Dr. (Mrs.) Christine A. Powell, Mona, St.
Dr. 1.1. Counsilman, Clovis, California, USA Andrew, Jamaica
Dr. Edhi Dharma, Yogyakarta, Indonesia Prof. S.H. Pudjiadi, Jakarta, Indonesia
Dr. S. Drahaman, Colombo, Sri Lanka Prof. Abdul Waheed Qureshi, Lahore, Pakistan
Prof. RE. Eeckels, Dacca, Bangladesh Dr. (Mrs.) Nursida Raid, Medan, Indonesia
Dr. (Mrs.) C. Geefhuysen, Yogyakarta, Indonesia Prof. I.G.N. Gde Ranuh, Surabaya, Indonesia
Dr. C. Gopalan, New Dehli, India Dr. M.G.M. Rowland, Cambridge, England
Prof. M. Gracey, Perth, Australia Dr. Tonny Sadjimin, Yogyakarta, Indonesia
Dr. (Mrs.) Endang D. Hamid, Medan, Indonesia Dr. Saraswati Boerhan, Surabaya, Indonesia
Prof. R Hariyono, Semarang, Indonesia Dr. Satriono, Ujung Pandang, Indonesia
Prof. P.F. Heywood, Herston, Queensland, Dr. Teluk Sebodo, Yogyakarta, Indonesia
Australia Prof. D. Sinniah, Kuala Lumpur, Malaysia
Dr. Boerhan Hidayat, Surabaya, Indonesia Dr. A.G. Soemantri Hardjojoewono, Semarang,
Prof. D. Hull, Nottingham, England Indonesia
Dr. Mahdin A. Husaini, Bogor, Indonesia Dr. Yati Soenarto, Yogyakarta, Indonesia
Prof. Ismangoen, Yogyakarta, Indonesia Dr. (Mrs.) Pitono Soeparto, Surabaya, Indonesia
Dr. (Mrs.) Sri Kardjati, Surabaya, Indonesia Dr. Soenarto, Yogyakarta, Indonesia
Dr. D.R Karunaratne, Colombo, Sri Lanka Dr. Dwi A. Soeyoso, Surabaya, Indonesia
Prof. Darwin Karyadi, Bogor, Indonesia Dr. Deddy Subardja, Bandung, Indonesia
Dr. (Mrs.) Darwin Karyadi, Bogor, Indonesia Dr. I. Sudigbia, Semarang, Indonesia
Prof. Ashfaq Ahmed Khan, Peshawar, Pakistan Dr. Suhartati, Jakarta, Indonesia
Prof. Zafarullah Kundi, Rawalpindi, Pakistan Dr. Suharyono, Jakarta, Indonesia
Prof. (Mrs.) J.A. Kusin, Amsterdam, Dr. (Mrs.) Emelia Suroto-Hamzah, Bandung,
The Netherlands Indonesia
Dr. Wen Chang Liu, Kaohsiung, Taiwan Dr. Purnomo Suryantoro, Yogyakarta, Indonesia
Dr. (Mrs.) Anita Li Ming Cheng, Hong Kong Dr. Achmad Suryono, Yogyakarta, Indonesia
Dr. Chairuddin P. Lubis, Medan, Indonesia Dr. 1.0. Suwenda Pare, Kediri, Indonesia
Dr. Bondan Lukito, Jakarta, Indonesia Prof. Moelyono S. Trastotenoyo, Semarang,
Dr. Sidharta Martohusodo, Malang, Indonesia Indonesia
Prof. L. Mata, Guadelupe, Costa Rica Prof. H.K.A. Visser, Rotterdam, The Netherlands
Dr. (Mrs.) Moenginah P.A., Yogyakarta, Prof. A. Samik Wahab, Yogyakarta, Indonesia
Indonesia Prof. H. Iskandar Wahidiyat, Jakarta, Indonesia
Dr. (Mrs.) Moersintowarti Navrendra, Surabaya, Dr. Sabdo Waluyo, Solo, Indonesia
Indonesia Prof. Chap-Yung Yeung, Hong Kong
Participants Nutricia
Dr. J.G. Bindels, Zoetermeer, The Netherlands Dr. L.T. Leksana, Jakarta, Indonesia
Ir. (Mrs.) Amalia Ernawati, Jakarta, Indonesia Mrs. O. Maasland, Zoetermeer, The Netherlands
Drs. E.I. van der Hagen, Zoetermeer, The Dr. Ir. A.K. Muntjewerf, Zoetermeer, The
Netherlands Netherlands
Mr. A. Istanto, Iakarta, Indonesia Ir. M.O.M. van Overbeek, Zoetermeer, The
Mr. O. Kreukniet, Jakarta, Indonesia Netherlands
List of authors and editors
Dr. Anna Alisjahbana, Department of Child Health, Medical School Unpad, Hasan
Sadikin General Hospital, JL Pasteur 38, Bandung, Indonesia.
Co-authors: R. Peeters, M. Maertens, S. Tanuwidjaja, H. Tjokrohusodo, W.
Ngantung and D.A. Primana.
Dr. P. Bhaskaram, National Institute of Nutrition, Indian Council of Medical
Research, Jamai Osmania PO, Hyderabad, 500007, India.
Dr. J.G. Bindels, Nutricia Research, P.O. Box 1, 2700 MA Zoetermeer, The
Netherlands.
Dr. J.J. Counsilman, Department of Zoology, National University of Singapore,
Republic of Singapore. Present address: 3700 Plainsman 54, Bryan TX 77802,
USA.
Dr. C. Gopalan, The Nutrition Foundation of India, B-37, Gulmohar Park, New
Delhi 110049, India.
Prof. Michael Gracey, Gastroenterology & Nutr. Res. Unit, Princess Margaret
Children's Medical Research Foundation Unit (inc), Perth, Western Australia.
Present address: Aboriginal Health Unit, Health Department of Western
Australia, 189 Royal St. East Perth, Western Australia 6004.
Co-author: H. Sullivan.
Prof. P.P. Heywood, Papua New Guinea Institute of Medical Research. Present
Address: Community Nutrition Program, University of Queensland, Clinical
Sciences Building, Royal Brisbane Hospital, Herston, Queensland, Australia
4006.
Co-author: A.H. Heywood.
Dr. Mahdin A. Husaini, Nutrition Research and Development Centre, Kompleks
Gizi, JL DR Semeru 63, Bogor 1612, Indonesia.
Co-authors: Soedarti Surbakti and Yayah K. Husaini.
Prof. Darwin Karyadi, Nutrition Research and Development Centre, Kompleks Gizi,
JL DR Semeru 63, Bogor 1612, Indonesia.
Prof. J.A. Kusin, Nutrition and Agrotechnology Section, Royal Tropical Institute,
Mauritskade 63,1092 AD Amsterdam, The Netherlands.
Co-authors: Sri Kardjati, C. de With and W.M. van Steenbergen.
Prof. L. Mata, Institute for Health Research (INISA), University of Costa Rica, P.O.
Box 212, 2100 Guadalupe, Costa Rica.
Prof. A.M. Molla, Aga Khan University Hospital, Stadium Road, PO Box 3500,
Karachi 74800, Pakistan.
Co-author: Ayesha Molla.
Dr. Christine A. Powell, Tropical Metabolism Research Unit, University of the West
Indies, Mona, St. Andrew, Jamaica.
Co-author: S.M. Grantham-McGregor.
xiv
Prof. S.H. Pudjiadi, Child Health Department, Medical Faculty University of

Indonesia, JL Salemba 6, Jakarta, Indonesia.
Dr. M.G.M. Rowland, Dunn Nutritional Laboratory, Cambridge, United Kingdom.
Present Address: East Anglian Regional Health Authority & PHLS Com-
municable Disease Surveillance Centre, Union Lane, Chesterton, Cambridge CB4
lRF, United Kingdom.
Dr. Tonny Sadjimin, Clinical Epidemiology and Biostatistics Unit, Faculty of
Medicine, Gadiah Mada University, Yogyakarta.
Dr. A.G. Soemantri Hardjojoewono, Department of Pediatrics, Diponegoro Univer-
sity, RS DR Kariadi, JL DR Sutomo 16, Semarang, Indonesia.
Dr. Pitono Soeparto, RS DR Soetomo, Bagian Ika, FK University Airlangga, JL
Dharma Husada 6-8, Surabaya, Indonesia.
Dr. I. Sudigbia, Department of Pediatrics, Diponegoro University, RS DR Kariadi,
JL DR Sutomo 16, Semarang, Indonesia.
Dr. Suharyono, Department of Pediatrics, Division Gastroenterology, Medical
Faculty, University of Indonesia, JL Salemba Raya 6, Jakarta, Indonesia.
Dr. Soedarti Surbakti, Central Bureau of Statistics, Jakarta, Indonesia.
Co-authors: Yayah K. Husaini and Mahdin A. Husaini.
Prof. H.K.A. Visser, Department of Pediatrics, Sophia Children Hospital/Erasmus
University, P.O. Box 70029, 3000 LL Rotterdam, The Netherlands.
Prof. Chap-Yung Yeung, Department of Pediatrics, Queen Mary Hospital, Hong
Kong.
Co-authors: F.H.W. Wong and A.Y.C. Tam.
Opening of the Symposium
Ladies and gentlemen,
It is indeed a great pleasure to welcome all of you at this 8th Nutricia symposium on
behalf of the Advisory Board. It was in the early 60s that Nutricia started this series
of symposia. The first five were organized in the Netherlands, by Professor Jonxis
and myself (the subjects were in the fields of neonatology and nutrition); number 6
was organized by Professor Eeckels in Leuven (the subject was tropical pediatrics),
number 7 by Professor Silk in the U.K. (gastroenterology and nutrition in internal
medicine) and now number 8 in Indonesia on 'Child Nutrition in South East Asia'.
Mr. Van der Hagen, the president of Nutricia, mentioned yesterday evening at the
opening dinner, that there is a mutual relationship between industry and science.
Nutricia is supporting science in several ways: by giving money to organize these
symposia, but also by supporting fellowships and research projects. This year, a
Nutricia Research Foundation has been established and at the end of this year the
first fellowships and grants will be awarded. Moreover, once every three years a
Nutricia Award for most distinguished service to nutrition will be presented to a
scientist, elected by an international board of advisors. We are very grateful to
Nutricia for this generous support to nutritional science.
The subject of this meeting is of great interest. Child nutrition is a very important
subject. There is strong evidence that nutritional deficiencies and undernutrition in
children may have long-lasting - even permanent - effects.
Nutrition of the pregnant mother and infant has great priority in governmental
strategies. Nutritional deficiencies and infections, and particularly diarrhea in
children, have great impact on child health in large parts of the world, including
South East Asia. Pediatrics and nutrition are disciplines on their own, clinically and
scientifically. But they are strongly related. This Nutricia symposium brings together
workers from different disciplines, and this may help to bridge the gap between
clinicians and nutritionists.
This is a meeting where most of the papers will be presented by investigators who
themselves work in this part of the world or who have recently worked there. I think
H.K.A. Visser and J.O. Bindels (eds.) Child Nutrition in South East Asia, xv-xvi.
© 1990 Kluwer Academic Publishers.
xvi
this is the right approach. Those of us present here who work in the so-called
industrialized northern countries are guests and we wish to listen to you and to learn.
We may contribute a little to the discussion. The members of the Advisory Board are
very grateful indeed that so many excellent workers from this part of the world have
been willing to contribute to this meeting.
We look forward not only to a meeting with good papers and good discussions but
also to a meeting where people from different parts of the world can meet and
become friends. This will not be so difficult, we belong to one great family, a family
which is devoted to children.
I wish you a very good meeting.
H.K.A. VISSER
Session 1
General aspects of child nutrition in South East Asia
Chainnan: H. Iskandar Wahidiyat

Childhood malnutrition in South East Asia:
Concepts and strategies for control
C.GOPALAN
Introduction
During the last four decades there have been some major changes in our concepts of
priorities in child nutrition and these have naturally been reflected in changing
strategies for combating childhood malnutrition. Some examples of such changing
concepts which have influenced our strategies, though not always beneficially, have
been listed in Table 1. For our present purpose, earlier concepts which have now
been either widely accepted, or largely discarded as untenable, need only to be
mentioned very briefly for their historical interest. Our major concern in this paper
will be with such concepts and strategies which are still current, controversial and
being debated.
It was only after the wide recognition of PEM (protein-energy malnutrition) with
its florid clinical manifestations like kwashiorkor and marasmus, and of nutritional
blindness from vitamin A deficiency, as major public health problems, that the
emphasis in child nutrition in Asia shifted from the school child to the pre-school
child. With the increasing recognition of the late effects of early malnutrition (i.e.
malnutrition in the under-threes), the "under-three" age-group came to be considered
as the critical phase of development which required special focus.
PEM throughout the fifties and early sixties was looked upon as a manifestation
of primary protein deficiency. The era of the PAG (Protein-Advisory Group) held
sway for about two decades. Fish-protein concentrates, fabricated largely by bureaus
of commercial fisheries in North America and Europe, were sought to be shipped to
the Third World as the panacea for PEM. All this came to an end finally in the great
"protein fiasco", when it became abundantly clear, largely on the basis of work
carried out in Asia, that PEM as seen in Asian populations subsisting on cereal-
legume dietaries was primarily the result of calorie deficiency, protein deficiency
being an incidental secondary by-product. It is now agreed that if habitual cereal-
legume diets could be fed to children at levels which would fully meet their caloric
needs, their protein needs would also be met [1]. The challenge now is to achieve
adequate caloric density (overcoming the "bulk factor") in the conventional cereal
H. K.A. Visser and J. G. Bindels (eds.) Child Nutrition in South East Asia. 3-20.
4
Table 1. Some examples of changing concepts regarding childhood malnutrition
1. Shift of major focus from the "school-age" period (prior to the sixties) to the
pre-school age period.
2. The era of the PAG and "protein concentrates" followed by the "great protein
fiasco" - and the recognition of calorie deficiency as the primary problem.
3. Predominant concern over "How to reach the pre-school child"; "On-the Spot"
feeding programmes to "prevent sharing of food supplement". - Relative
neglect of nutrition in pregnancy and lactation; erosion of concept of maternal
health/nutrition as a major determinant of infant/child nutrition. - A short-lived
aberration.
4. The pushing of the simplistic slogan "GOBI" by some international agencies to
the detriment of broad-based integrated Child Care.
5. Misuse of the concept of "adaptation"; "Small is healthy?"; and "appropriate
weight for height".
6. The use of technological solutions for specific problems - nutritional blindness,
goitre and anaemia.
7. Supplementary feeding.
based diets in order that they can be fed to young children in quantities adequate to
meet their full caloric needs.
Also, the earlier postulates which sought to divide kwashiorkor and marasmus (of
the pre-school child) into two neat and tidy separate compartments attributable to
two distinct and different dietary etiologies - kwashiorkor being the result of primary
protein deficiency and marasmus being the result of caloric deficiency - were shown
to be misconceptions [2]. Kwashiorkor, marasmus and marasmic kwashiorkor as
seen in the pre-school child are now recognised as different clinical manifestations
with a common dietary etiology and, therefore, calling for a common strategy for
their prevention and control.
Paediatricians and nutrition scientists from Asia often fail to appreciate the
enormous practical significance of the above developments. The problem of
childhood malnutrition in Asia would have been far more formidable if some of the
earlier concepts were in fact true. At least some of the credit for dispelling these
earlier erroneous concepts must go to Asian scientists.
The sixties in this century also witnessed yet another unfortunate aberration. With
the recognition of the importance of nutrition in the pre-school period, the major
concern revolved around the question: "How to reach the pre-school child". "On-the-
spot" feeding of under-fives was widely resorted to as the strategy for combating
childhood malnutrition. Children of the poor were gathered in village squares at an
appointed time of the day and were offered food supplements (often donated by
foreign agencies) in amounts roughly equivalent to a third o.f their daily food
requirement; they were expected to consume the ration on the spot. The anxiety here
was to prevent the mother from letting her other children share the supplement! Thus
the mother instead of being enlisted as a powerful ally and beneficiary, became
someone to be guarded against! The concept that improved maternal nutrition is an
essential requisite for infant and child nutrition came to be forgotten. The mother
receded to the background and out of focus. Nutritional care of women in pregnancy
and lactation was largely neglected although the importance of breast-feeding was
5
being extolled. The futility of strategies based on such misconceptions came to be

recognised in due course.
A major beneficial development during the last three decades has been the
increasing recognition of the importance of breast feeding and of the superiority of
breast-milk over all the infant food formulas and the consequent large-scale return to
breast-feeding all over the world - especially in the developed countries. This
development has also not been without some deleterious side-effects. The shrinking
market for commercial milk foods in Europe has apparently led to the temptation on
the part of at least some commercial milk food manufacturers to aggressively
promote such foods in the unwary Third World - a process sometimes unfortunately
abetted, or at least not adequately resisted, by some Third World paediatricians and
health scientists. Commercial milk foods have a place in special situations - but the
poor overcrowded unhygienic metropolitan urban slums of the Third World certainly
do not represent "such special situations" for "legitimate use" of such commercial
milk foods.
In this article, six major questions regarding strategies for combating childhood
malnutrition currently under discussion, and the concepts underlying them, are
briefly considered.
Improvement of maternal nutrition - an essential part of a

comprehensive strategy for combating childhood malnutrition
There are several studies which indicate that improvement of the nutritional status of
the mother, apart from being obviously beneficial to the mother, is also an essential
requisite for uplifting the nutritional status of the offspring, both in infancy and in
early childhood. That improvement of maternal health/nutritional status must
become the cornerstone of programmes for combating childhood malnutrition and
that this could well be the most rewarding strategy for combating childhood
malnutrition is a concept that is gaining wide acceptance. While I am aware that
there are quite a few studies on this subject, I may be permitted here to draw special
attention to two Indian studies which serve to highlight the relevance and importance
of maternal nutritional upliftment as a strategy for combating childhood malnutrition.
Studies of Shanti Ghosh et al. Shanti Ghosh and her colleagues in India [3] in 1969
initiated an extensive longitudinal study on growth and development of children of
different socio-economic groups, whom they had followed from birth, for periods
now extending to nearly 18 years. These studies showed that with respect to both
height and weight, infants who started with the initial handicap of low birth weight in
spite of their being full-term (small for gestational age - SGA) never fully recovered
from their initial handicap unlike preterm infants with weights appropriate to their
gestational age (AGA). They continued to grow in a lower growth trajectory (Tables
2 and 3). It may be argued that since a great majority of SGA infants may be
6
Table 2. Height measurements from birth to 14 years of premature, appropriate for gestational age (AGA).
full-term. small for gestational age (SGA), and full-term, AGA infants in India
Age PretermAGA Full-term SGA Full-term AGA

(em) (em) (em)
Birth 44.6 43.4 49.3

1 year 69.3 67.9 73.7
2 years 79.3 77.5 83.0
3 years 87.0 85.4 90.7
4 years 94.3 92.6 96.5
5 years 100.5 96.9 102.3
6 years 106.6 100.3 107.4
7 years 112.9 109.6 115.8
8 years 121.0 114.5 121.5
9 years 126.4 119.7 124.3
10 years 131.6 125.3 126.4
11 years 136.6 131.6 131.7
12 years 142.6 137.0 140.0
13 years 148.4 145.0 146.9
14 years 154.5 147.0 156.4
Source: S. Ghosh, S.K. Bhargava and V. Kapani (unpublished observations).
Table 3. Weight Measurements from birth to 14 years of premature. appropriate for gestational age
(AGA). full-term, small for gestational age (SGA), and full-term. AGA infants in India
Age PretermAGA Full-term SGA Full-term AGA

(kg) (kg) (kg)
Birth 1.8 1.7 3.0

1 year 7.3 6.7 8.8
2 years 9.4 8.7 10.7
3 years 11.4 10.5 12.4
4 years 12.9 12.0 13.7
5 years 14.3 13.1 15.2
6 years 15.8 15.1 16.5
7 years 18.6 16.2 19.1
8 years 20.6 17.6 21.5
9 years 23.2 19.8 23.8
10 years 24.1 22.0 26.9
11 years 26.4 24.9 28.4
12 years 27.2 28.9 30.5
13 years 33.3 34.0 35.9
14 years 42.0 35.6 43.6
Source: S. Ghosh. S.K. Bhargava and V. Kapani (unpublished observations).
expected to come from poor households, their poorer growth performance could be
due to their substandard postnatal nutrition and not necessarily a reflection of their
initial handicap of low birth weight. In order to be able to answer this question, the
authors had also compared the growth performance of SGA infants of both the poor
7
length in em
110
105
100
95
90
85
80 -B- birth weight >2500 g
75 income p.cap. >200Rs
70 -+- birth weight .;2500 g
I
60 birth weight >2500 g
55 income p.cop. :s;SORs
50 -*- birth weight ~2500 g
45 income p.cap. ~50Rs
40
0 6 12 18 24 30 36 42 48 54 60
age in months
Figure 1. Mean length/standing height by per capita income, birth weight and age. Source: S. Ghosh, S.K.
Bhargava and V. Kapani (unpublished data).
weight in kg
16
14
12
10
1:birth weight >2500
2:birth weight .;2500

3:birth weight >2500

4
income p.cop ~50 Rs
4:birth weight ,;;2500

2
income p.cap (50 Rs
0+----,---,----,---,----,---,----,---,----,---,---
o 6 12 18 24 30 36 42 48 54 60
age In months
Figure 2. Mean weight by per capita income, birth weight and age. Source: S. Ghosh, S.K. Bhargava and
V. Kapani (unpublished data).
8
Table 4. Maternal height and incidence of low birth weight (LBW) in offspring
Maternal height (cm) Income group Incidence ofLBW
Less than 145 cm Less than Rs. 50/- per person per month 35.5%
More than 145 cm Less than Rs. 50/- per person per month 24.2%
More than 145 cm More than Rs. 200/- per person per month 15.0%
Source: S. Ghosh, S.K. Bhargava and V. Kapani (unpublished observations).
as well as the relatively well-to-do households, and found that in each income group,
the growth performance of SGA infants was poorer than that of infants with normal
birth weights in the same income group. The worst growth performance was seen in
infants who belonged to the poor group who also had the additional handicap of low
birth weight to begin with (Figures 1 and 2).
The same authors also observed a direct relationship between stunting in the
mothers and the occurrence of low birth weight in their offspring (Table 4).
Recent Studies of the Nutrition Foundation of India. The Nutrition Foundation of

India recently completed a study on maternal nutritional status and infant growth
among poor communities inhabiting the slums of the metropolitan cities of Bombay,
Calcutta and Madras [4]. The study indicated that the incidence of low birth weight
weight in kg
4
"""*" NCHS
-+- O! 45 kg (meon:50.19)
3 ~ 38 kg (meon:45.60)
-B- < 38 kg (meon:34.14)
2
0 234 5 6
age in months
Figure 3. Average weights of infants in different maternal weight categories. Source: NFl Scientific
Report 9,1988.
9
weight in kg
4
-+ NCHS
3
-+- ;> 3 kg
2.5 - 3 kg
--B- < 2.5 kg
2
0 2 3 4 5 6
age In months
Figure 4. Average weights of infants in different birth weight categories. Source: Nfl Scientific Report 9,
1988.
weight in kg
Wt;>38kg & Ht>145cm
-+- Wt;::38kg & Ht< 145cm
-+ Wt<38kg & Ht<145cm
2~--,------,------,------,------,------,------,---
o 2 3 4 5 6
age in months
Figure 5. Average weights of infants in different maternal weight and height categories. Source: Nfl
Scientific Report 9, 1988.
10
deliveries in women who started their pregnancy with body weights of 45 kg and
above was substantially less than in those who started their pregnancies with lower
body weights (Figure 3). As against an incidence of about 30-33% of low birth
weight deliveries in poor communities in general, the incidence among mothers with
weights of over 45 kg at the start of pregnancy was no more than 17%. It appeared,
therefore, that if women of poor communities could be helped to achieve a body
weight of around 50 kg at the height of their pregnancy (assuming a modest gain of
around 6 kg during pregnancy) it would be possible to achieve a substantial reduc-
tion in the quantum of low birth weight deliveries in poor communities.
Even more importantly, the study showed that infants of mothers who achieved a
body weight of around 50 kg at the height of their pregnancy grew satisfactorily until
at least the end of four months of infancy to achieve body weights exceeding 90% of
the 50th percentile and lengths exceeding 98 % of the N CHS standard (Figure 4).
The study also suggested that the mother's low body weight was a more important
determinant of low birth weight in the offspring than low maternal height; and that
even in the case of mothers who were significantly stunted, achievement of body
weights of around 50 kg by the end of pregnancy could result in substantial reduction
in the incidence of low birth weight deliveries (Figure 5), This is a conclusion of
considerable practical significance. Stunting in the present generation of mothers
cannot be corrected through nutrition intervention, but augmentation of weight gain
during pregnancy, even in stunted mothers, can be achieved through nutrition
intervention, improvement of diets, rest, and restriction of energy expenditure during
the last weeks of pregnancy.
The study also provided suggestive evidence that a major contribution to the
"smallness" of the low birth-weight baby may be taking place in the third trimester.
While low birth-weight infants of poor communities studied were about 68% of the
NCHS standard with respect to their body weights, they were above 92% of the
NCHS standard with respect to their lengths (asymmetric), indicating that poor fat
deposition in the third trimester may be the major factor. The practical importance of
this finding is that where resources are limited, nutrition supplementation, even
during the third trimester, could prove beneficial. In the same study it was also found
that iron supplementation to mothers, apart from raising their haemoglobin levels
also had beneficial effects indicative of augmentation of lactation and better infant
growth, but the evidence on this point was not sufficiently strong to be considered
conclusive.
On the basis of this study, the Nutrition Foundation of India recommended that
mothers must become the central participants and beneficiaries and indeed the
starting points of child welfare/development operations in the country. It was also
recommended that the ICDS (Integrated Child Development Service) programme in
the country should be restructured to better focus on the mother.
Nutritional upliftment of the mothers-to-be - adolescent girls. In most Asian

countries today, the average age of girls when they marry is low; and before the girl
11
attains the age of 25 years she has generally had her three or more children. The
girls, because of under-nutrition and poor access to basic health services, are already
under-nourished, even at the time of their marriage. They arrive at motherhood,
practically illiterate having dropped out of schools after the first or second year of
schooling, and are totally unprepared and ill-equipped for the demanding tasks of
motherhood that await them. Once married and once thus caught up in household
and maternity chores, they are largely "unreachable" to the conventional health
services. It is for this reason that the ICDS system in India today is able to capture
only a small fraction of young pregnant women and nursing mothers.
In view of these considerations, the Nutrition Foundation of India proposed the
adoption of a strategy whereby the focus in health care and national development
will shift to the mothers-to-be, namely adolescent girls [5]. The NFl proposal
advocates a broad-based programme of "non-formal education for better-living",
improved health care, and vocational training, aimed at adolescent girls in rural
areas. Such a programme could provide them with the benefit of basic education in
personal hygiene, environmental sanitation, nutrition, child rearing and sex educa-
tion, in addition to providing them with vocational skills that would facilitate income
generation and improvement in economic status. These girls will then be better
equipped for motherhood and will be more responsive to health services and
developmental programmes. Antenatal programmes and family-planning
programmes would then become more effective.
It is the adolescent girls of today who will usher in the generation of the.21 st
century; it is their health, nutritional status and education that will largely determine
the quality and the calibre of the next generation. A broad-based programme of
education, health care and vocational training aimed at poor rural adolescent girls
must, therefore, find an important place in any meaningful comprehensive strategy
for combating childhood malnutrition in Asia.
"Child Survival" and "GOBI"
Nutrition has apparently been a fertile field for "slogans" and "catch-phrases", which
have often found uncritical acceptance in the Third World. Two such slogans which
have thus far gained a wide usage are "Child Survival" and "GOBI".
"Child Survival". "Child Survival" has now almost displaced old-fashioned

"Maternal and child health/nutrition" in the nutrition vocabulary. Concern for child
survival is obviously laudable; but the implications behind the substitution of "Child
survival" in place of "Child nutrition" as the target for developing countries, have
hardly been appreciated.
The strategies that are needed to ensure just "child survival" fall far short of
strategies needed to ensure child health and nutrition. Through crisis management
technologies - like oral rehydration for instance, and prompt treatment of acute
12
Table 5. Birth rates and death rates in U.P. and KeraIa
State Birth Death Infant Child

rate rate mortality mortality
0-4 years
KeraIa 23.3 6.5 31.3 10.2

Uttar Pradesh 37.6 15.8 141.6 54
Source: Sample Registration System, 1985, Vital Statistics Division, Office of the Registrar General of
India.
Table 6. NNMB percentage distribution of 1-5 year old children according to Gomez's classification -
1975-1979
State Normal Grade of malnutrition

Mild Moderate Severe
Kerala 16.4 45.3 33.5 4.8

Ultar Pradesh 18.8 48.8 26.8 6.1
Source: NNMB Report for the years 1975-1979, NIN, Hyderabad, 1980.
infections - better child-survival can be ensured; but ensuring child nutrition calls for
a lot more - the eradication of poverty, dietary improvement in poor households,
improved environmental sanitation, and access to a safe water supply. The adoption
of "child-survival" as the target in preference to "child health/nutrition" among poor
popUlations implies a lowering of our sights; it implies concern with "survival" as
such, and not with the "quality" of the survivors. It provides a soft option to the
policy-makers and could retard movements for the speedy elimination of current
glaring inequities in the socio-economic order in many Third World countries, which
lie at the root of the malnutrition problem. An ever-expanding pool of substandard
survivors cannot but further erode the quality of Human Resources of any country.
These implications might have been totally unintended by the slogan-coiners, but all
the same we must clearly recognise the limitations of the slogan. Programmes for
improved child-survival must be part of a larger strategy for the improvement of
child health and nutrition and must not be pursued in isolation as a disaster-relief
operation.
In an earlier publication [6] it was shown that the remarkably better rate of child-
survival witnessed today in the State of Kerala in the South of India as compared to
the State of Uttar Pradesh in the north of the country, is not reflected in better growth
performance of children in Kerala (Tables 5 and 6 and Figure 6). Kerala has
achieved better child survival because of higher levels of female literacy and better
health care facilities; but poverty has not been eliminated and diets have not
improved. As was pointed out in the earlier publication: "literacy may motivate the
13
height in em
140
135
130
125
120
115
110
105
100 ,-----
6 7 8 9 10 11 12
age in years
Figure 6. Mean heights of school children in V.P. and Kerala. Source: NNMB Report for the years
1975-1979, NIN, Hyderabad, 1980.
poor to use available health facilities to combat illnesses and infections in children
and the women to space their births; but, literacy, unfortunately, cannot be a
substitute for food or an antidote to poverty".
"GOBI". There is nothing basically wrong with the simplistic GOBI slogan
(subsequently expanded) as long as it is viewed not as the total agenda for child
health/nutrition operations of a country, but as no more than a limited agenda for
action which UNICEF has set for itself - the agenda being limited to such items as
UNICEF considers operationally feasible and within its competence and resources;
thus, it is at best a "departmental" or "institutional" agenda and not a national one.
Caution is necessary in order to avoid overstating the case for these selective items,
and more importantly, to avoid the temptation to project GOBI as a national agenda
for child nutrition for Third WorId countries.
Growth-monitoring, Growth-monitoring has unfortunately been pushed in many

countries, as an end in itself, without much thought being given to building up the
infrastructure essential for the follow-up action which alone can make this expensive
and time-consuming operation, meaningful. Growth-monitoring is only a means and
not an end in itself. Growth-monitoring, instead of being used to detect early growth-
faltering, is often misused to identify the most undernourished children in the
14
community who would qualify for feeding programmes, as in the World Bank
assisted Tamil Nadu Nutrition Programme in India. These are not arguments against
growth-monitoring as such but against the manner in which this operation is being
currently widely misused. Since this subject has been discussed at some length in
earlier publications [7,8] it will not be necessary to deal with it in greater detail here.
Oral rehydration. In recent years, 'oral rehydration' has been vigorously promoted in
a manner likely to give the impression to the unwary that it is almost the total answer
to the problem of childhood diarrhoea [9]. Oral rehydration is not a preventive
procedure; even as a curative procedure it addresses only the problem of dehydration
and nothing more. It certainly facilitates better child survival, but does not solve the
basic factors in the community responsible for the problem of diarrhoea. The
essential requisites for ensuring proper environmental sanitation and personal
hygiene and for promotion of good nutrition cannot be bypassed. Care is necessary
to ensure that excessive reliance on oral rehydration as the answer to the diarrhoea
problem does not lead to a relative neglect of other essential programmes that alone
could address the root causes of the problem and not just its symptoms. This subject
has also been discussed in an earlier paper [9].
Immunisation. The drive for universal immunisation is welcome. However, it must

be remembered that immunisation is just one component of health-care and by itself
cannot work wonders [10]. It must be part of an integrated package of health services
and must not be pursued in isolation to the detriment of other elements in the health-
care package. If the drive for universal immunisation is wisely used, not for the
narrow purpose of achieving immunisation targets but for spearheading a vigorous
movement for the activation of "sleepy" health systems in some Asian countries, it
would have served its purpose.
Breast-feeding. Promotion of breast-feeding as the mass-anchor of infant nutrition is

unexceptionable. However, as far as the poor in Asia are concerned, they do not need
to be told about the need for breast-feeding. Economic necessity and cultural
preferences favour breast-feeding. The important practical question is at what stage
in infancy should supplements be introduced. This subject was discussed briefly in
an earlier paper [11]. I argued that as far as abjectly poor communities are con-
cerned, recommendations with regard to the optimal time for the introduction of
supplements, and the type and nature of such supplements must be based not just on
theoretical estimates of breast-milk output in relation to nutrient requirements, but
also on wider practical considerations - including in particular the risks of infections.
In poor communities living under bad hygienic conditions with no access to a safe
water supply, the theoretical benefits of early supplementation (by the 3rd month)
may be more than offset by the inevitable earlier onset of diarrhoeal episodes. It is
possible that under such circumstances, infants with early supplementation may
actually arrive at their sixth month in a worse state of health and nutrition than
15
infants not receiving such early supplementation. I had argued, therefore, that the
balance of advantage in such extremely poor communities may well lie in delaying
the introduction of supplements until the infant is six months old. This of course will
not be considered ideal for infants in better environments. Cowen had actually shown
that with exclusive breast-feeding for six months, it is possible for some rural
communities in India to achieve satisfactory growth-performance in infants [12].
However, this will also logically imply good nutritional support to the mother, at
least during the first six months of lactation - a point often neglected and not
sufficiently emphasised even by ardent advocates of breast feeding.
"Small is healthy for the poor?" - The "adaptation" misconcept
In recent years, considerable confusion has been caused by the loose manner in
which the concept of adaptation to under-nutrition has been misused. So-called
"adaptation" associated with dietary caloric intake equivalent to just about 70% of
the mean recommended level, and characterised by a restriction of physical activity,
stunting and small body-size has been approvingly equated with normality. It has
even been suggested that "smallness" is an appropriate and welcome attribute of
poor people consistent with their good health [13], the activity pattern and
"adjustment" of life-styles imposed by poor diets are explained away as "cultural
adaptation" [14]. Asian nutrition scientists are being advised not only to not use
"international standards" of growth (as this would yield "over-estimation of under-
nutrition"); indeed Seckler [13] goes even further and advises that even the "best
indigenous standard" of the high socio-economic groups of the country would also
be "abnormally large" for the majority of the population who are poor. Payne [14]
actually argues that even if it is the case that children of developing countries have
the same genetic potential for growth as those of the more fortunate countries of
Europe and U.S.A., they could settle for a lower level of growth in keeping with their
"economy and ecology". It is not surprising that this plea for acquiescence in growth
retardation in the poor has been sharply rejected as an exercise in "perpetuation of
under-nutrition".
Paediatricians and health scientists of developing countries must firmly reject
these counsels of despair. We will do well to heed the note of caution by Du Nony
[16]: "Adaptation is not progressive but defensive ... Being adapted does not
contribute to evolution. Adaptation is never a goal but only a means". Adaptation has
to be viewed not as normality but as a "strategic retreat from normality" and as "a
state of siege".
I have discussed this question in some detail in a number of earlier publications
and have cautioned against the danger of the misuse of the concept of adaptation in a
manner likely to promote social and political indifference to (and acquiescence in)
so-called "moderate under-nutrition" in children [17-21].
I do not therefore wish to elaborate further on this subject.
16
Stunting and "weight-for-height". It has been argued that a stunted child with a
weight appropriate to his height could be considered "normal" and "adapted". I see
no physiological basis whatsoever for the view that is being widely propagated that it
is not so much height-for-age or weight-for-age that matters but that it is the weight-
for-height ratio that is the crucial indicator of normality. This convenient and curious
hypothesis could lend legitimacy to the "small is healthy for the poor" concept.
There is not only no scientific evidence in support of this hypothesis; there is indeed
evidence pointing to a near-linear relationship between deficits in height-for-age or
weight-for-age and functional impairment irrespective of weight-for-height.
The enormous cost that "stunting" implies to the society and the State has not
been appreciated. A community in which a considerable part of the population is
stunted is a community with high infant and child mortality, high levels of morbidity
in children. It is also a community where children have lost valuable learning time
and where mothers have lost a large part of their already meagre daily wages. The
removal of factors in the socio-economic and health scene which prevent children
from attaining their full genetic potential for growth and development must be the
central objective of any national health and nutrition policy directed to the improve-
ment of the quality of a country's Human Resources. We should not be satisfied with
normal weight-for-height in our children; we must really aim at "normal weight-for-
normal-height" for a particular age.
Are international growth standards applicable to Asian children?
The above considerations raise the question as to whether the genetic potential for
growth and development of Asian children is similar to those of the USA (NCHS) or
whether there are significant genetic differences. I recently discussed this question at
some length at the Fourth Annual Scientific Conference of the Nutrition Society of
Malaysia [21]. I illustrated that on the basis of Japanese data and on data from
affluent sections of the Indian population that the NCHS standard, now in wide use,
is in fact appropriate for Japanese and possibly to Indian children - for boys up to the
15th year and for girls up to the lIth or 12th year (Figures 7 and 8). If we are
justified in projecting these conclusions based on Japanese and Indian (affluent) data
to most (if not to all) Asian countries (and I believe, we are) we may conclude that
the NCHS standard is applicable to most Asian children at least up to the 12th year
in the case of girls and the 15th year in the case of boys. The area of doubt is with
respect to the applicability of the NCHS standard for the older age groups - the
adolescents and the adults. This doubt will be dispelled only when it is absolutely
clear that the secular trend with respect to height in Japan has come to a complete
halt - which is not the case as yet. It should not be forgotten that Asian countries
today are in a state of developmental flux; there is an active secular trend and growth
standards developed now on the basis of even the observed growth performance of
17
height in em
180
160
140
120
100
NCHS (50th peretile)
-+- Japan
o 2 4 6 8 10 12 14 16 18
age in years
Figure 7. Comparison of mean heights of boys. Sources: I. WHO - International Growth References
(NCHS) - 50th percentile (Nut/78.1). 2. SEAMIC Health Statistics (1987).
height in em
160
155
150
145
140
NCHS
135 -* Delhi
--B- Japan
130
8 9 10 11 12 13 14 15 16 17
age in years
Figure 8. Comparison of mean heights of girls. Sources: I. WHO - International Growth References
(NCHS) (Nut/78.1). 2. NFl - Scientific Report 10, Growth in Affluent Indian Girls During Adolescence
(unpublished). 3. SEAMIC Health Statistics (1987).
18
currently affluent sections may be valid only for the next 10 years or so. Japan
provides the pointer in this regard.
Technological solutions to specific nutrition problems
There are at least two specific nutritional problems - Vitamin A deficiency and
iodine deficiency which lend themselves to simple technological solutions and which
do not call for great improvement in the economic status of populations.
Vitamin A deficiency. The approach of prevention and control of nutritional blind-

ness through the periodic oral administration of synthetic vitamin A concentrate was
pioneered by the National Institute of Nutrition in India, nearly two decades ago.
New dimensions of the problem of vitamin A deficiency and possible other benefi-
cial effects of vitamin A supplementation with respect to improved child survival
have been suggested by subsequent work in Indonesia. It would seem unwise,
however, for Asian countries to perpetually rely on synthetic vitamin A (the
technical know-how for the manufacture of it lies with a European firm) for the
solution of one of their major public health problems. It is not just prudence and
considerations of self-reliance that should also caution us against such reliance on
synthetic vitamin A; there are valid scientific reasons as well. We must make wide
use of the rich natural resources of B-carotene which are abundant in most Asian
countries with their lush green vegetations. The Red Palm, one of the richest sources
of B-carotene is very much an Indonesian plant, which was taken from the Botanical
garden at Bogor to Malaysia where it is now being extensively grown. The soil and
climate in many Asian countries favour the cultivation of red palm. The potential
rich harvest of B-carotene from red palm oil is now being largely wasted; it is the oil
that is attracting all the attention and not the B-carotene. An imaginative programme
for large-scale production of B-carotene, and for increasing reliance on this source
rather than on synthetic vitamin A, must now form an important item on the agenda
of Asian countries including in particular Indonesia, Bangladesh, India and Nepal.
There is now evidence that with respect to prevention and control of certain types of
neoplasms, B-carotene may actually be superior to synthetic vitamin A.
Vigorous attempts must be made to include green leafy vegetables in the diets of
young children and mothers. Some imaginative programmes in this regard initiated
in Nepal, though on a small scale, need to be encouraged and widely adopted.
Iodine deficiency. The obvious answer to the problem of iodine deficiency is

fortification of common salt with iodate - a proven procedure which has yielded
highly satisfactory results in many parts of the world. The programme has unfor-
tunately been poorly implemented; the implications of iodine deficiency with respect
to mental development have now been better appreciated.
Nutritional blindness and iodine deficiency are at least two specific nutritional
deficiencies which Asian countries can totally eradicate before the tum of the
century.
19
Supplementary feeding programmes
In many Asian countries, there are vast pockets of poverty with a high prevalence of
manifest clinical under-nutrition in children. Such situations certainly do call for
relief operations which can help to at least partially mitigate the severity of under-
nutrition. Supplementary feeding programmes wherein generally one third of the
day's food requirement are freely distributed to young children are being undertaken
in many such areas. The food used is generally supplied by foreign donor agencies.
Many Asian countries like Bangladesh and parts of India, for instance, are also
frequently plagued by floods and cyclones which render several hundreds of poor
families homeless. Such populations also need immediate emergency relief in the
form of food supplements to ward off starvation.
Thus supplementary feeding programmes are unavoidable in some situations, but
the limitations of such programmes as a durable, and as a cost-effective answer to the
problem of childhood malnutrition must be clearly understood. Firstly, it is futile to
expect that a programme in which just one third of the daily food requirement alone
is provided and that, also not very regularly, will make a significant contribution to
nutritional upliftment; all it can do is to help to mitigate acute hunger. Secondly, to
the extent to which such programmes help to push the problems of poverty and
under-nutrition under the carpet and retard the institution of more durable and
effective long-range programmes for eradication of poverty, employment and
income-generation, elimination of illiteracy, improvement of environmental
sanitation and improvement of diet in poor households, they would be doing an
actual disservice. Many of these programmes are inefficiently implemented with the
result that only a fraction of the enormous inputs actually reach the intended
beneficiaries. Politically vested interests are generated around these populist
measures which, though ostensibly based on "free" foreign food donations, actually
use up a good part of the country's material and manpower resources. Expensive and
time-consuming "growth-monitoring" operations, for instance, are being used for
choosing "beneficiaries" for such programmes (clearly a waste of resources) under
the mistaken belief that the food supplement is being "effectively targeted". Asian
countries cannot hope to solve their problems of poverty and under-nutrition through
hand-outs and charity operations. It is not through generating dependence but
through encouraging self-reliance and self-help and providing the communities with
the means by which they can help themselves that poverty and under-nutrition can be
combated.
Clearly, there is a need for a judicious combination of short-term measures (like
supplementary feeding) which will address the immediate needs, and long-term
measures which alone can provide the durable answer to the problem of under-
nutrition. Where supplementary feeding programmes are resorted to, they must be
used as the "entry point" for a comprehensive programme of development, including
health and nutritional upliftment of the community. Such programmes must not be
give-away charity operations but must involve the community with respect to
responsibility for organisation, implementation and cost-sharing.
20
The problems of childhood malnutrition which beset Asia are fortunately of far
less magnitude than those that afflict parts of Africa. While the "population problem"
in Asia, especially South East Asia, is undoubtedly formidable, Asia has abundant
food resources which have, as yet, not been fully exploited. Asia can feed its people
with its own internal resources, provided these are efficiently hamessed and provided
the glaring inequities in distribution of available resources are corrected and
developmental plans and programmes are efficiently and vigorously implemented.
Asia's real deficiencies today are not so much with respect to the availability of food
resources, but as regards the way in which these resources are being exploited.
References
1. Gopalan C, Narasinga Rao BS: Proc Nutr Soc India 10: Ill, 1971.
2. Gopalan C: In: McCance RA and Widdowson EM (eds) Calorie deficiencies and protein deficiencies
- Proceedings of a colloquium held in Cambridge, April 1967. J & A Churchill Ltd, London, 1968,
pp.49-58.
3. Ghosh S, Bhargava SK, Kapani V (unpublished data).
4. Scientific Report 9: Nutrition Foundation of India, 1988.
5. Gopalan C: Nfl Bulletin 5(1):5-7,1984.
6. Gopalan C: Nfl Bulletin 5(1):1-3,1984
7. Gopalan C and Chatterjee M: Special Publication Series 2, Nutrition Foundation of India, 1985.
8. Gopalan C: Nfl Bulletin 8(2):1-4,1987.
9. Gopalan C: Nfl Bulletin 6(4):8, 1985
10. Gopalan C: Nfl Bulletin 9(3):8,1988
11. Gopalan C: NFl Bulletin April issue: 4-5, 1981.
12. Das D, Dhania J and Cowan B: Nfl Bulletin April issue: 3-5, 1982.
13. Seckler D: In: PV Sukhatme (ed) Newer concepts in nutrition and their implications for policy.
Maharashtra Association for the Cultivation of Science, 1982, p. 127.
14. Payne P: In: A Pacey and P Payne (eds) Agricultural development and nutrition. Hutchinson, 1985.
15. Jaya Rao KS: Economic Political Weekly 21:24,1986.
16. Jaya Rao, KS: Quart J Surg Sc, September and December issue: 306-313, 1977.
17. Gopalan C: NFl Bulletin October issue: 5-6, 1982.
18. Gopalan C: NFl Bulletin October issue: 1-5, 1983.
19. Gopalan C: In: John C Waterlow (ed) Linear growth retardation in less developed countries. Nestle
Nutrition Workshop Series Vol 14, Nestec Ltd, Vevey!Raven Press Ltd, New York, 1988,
pp. 265-284.
20. Gopalan C: Paper presented at WIDER, Helsinki, Finland 1988 (to be published).
21. Gopalan C: Key-note address at the Third Annual Scientific Conference of the Nutrition Society of
Malaysia, Kuala Lumpur, 1989 (to be published).
Vitamin A problems in South East Asia with special
reference to its impact on child health
DARWIN KARYADI
Introduction
On the basis of data available, Asia has been traditionally considered the area at
highest risk from xerophthalmia. To determine what constitutes the public health
problem of xerophthalmia, incidence rates have been extrapolated from the com-
prehensive 1977-79 research study in Indonesia [1].
Sommer [2] has attempted to estimate, on the basis of an extensive prevalence
survey in Indonesia and then extrapolating these figures to at least four countries:
India, Bangladesh, the Philippines, and Indonesia itself, that at least 500,000 pre-
school age children every year develop active xerophthalmia involving the cornea
(X2/X3). About half of this number will be blind and a very high proportion,
probably in excess of 60%, will die. The annual figures for these same countries of
non-corneal xerophthalmia are many times higher, probably in the order of around 5
million.
Globally, South East Asia remains the focus of vitamin A deficiency; therefore,
with the overall objective of reducing xerophthalmia prevalence to below 0.5% by
the year 2000, nearly all countries in the region have now formulated and are
implementing national programmes for the prevention of blindness as an integral part
of primary care activities. Since the World Health Assembly Resolution in 1984, and
research studies in Indonesia, it has been accepted that vitamin A deficiency is a
systemic disease which has systemic effects on human health, which exceed
xerophthalmia or blindness as a serious public health problem because of implica-
tions for child survival.
This report will highlight the latest advances in the impact of vitamin A on child
health (especially on vision, morbidity and mortality), on practical aspects of
treatment, intervention strategy, and recent progress in the diagnosis of vitamin A
deficiency (clinical, biological, ophthalmological assessments) based on South East
Asia WHO - Government of Indonesia Regional Meeting in Jakarta, November 3-5,
1988 [3].
In addition, since 1985 the UN has launched a 10 year action programme to
H.K.A. Vi.vser and J.G. Bindels (eds.) Child Nutrition in South East Asia, 21-34.
22
control and prevent vitamin A deficiency, xerophthalmia and nutritional blindness.

FAO prepared a plan to complement the primary short-term interventions by WHO
and Unicef in the medical field (primary vitamin A capsule distribution) with
projects aiming at providing long-term solutions from the agricultural sector through
increased production and consumption of vitamin A and carotene (pro-vitamin A)
rich foods [4]. This report will also highlight this new development.
Functional and clinical aspects of vitamin A deficiency
Functional consequences of vitamin A deficiency
The first classical focus as a functional consequence of vitamin A deficiency based

on research results has been undertaken on impaired vision as nightblindness, which
goes back to antiquity with accounts in the literature of ancient Egypt, China,
Indonesia, and Japan. It was during the last three decades that the widespread
occurrence of xerophthalmia leading to blindness led to extensive as well as
intensive studies generally recognized from a public health point of view and which
were carried out under the auspices of the WHO.
A second important consequence of vitamin A deficiency leads to a keratinization
of the epithelium system i.e. alimentary, respiratory, and genito-urinary tracts. In
addition to keratinization, which means a violation of the integrity of the epithelial
barrier, the epithelial cells lack normal secretions, with a resulting loss of cilia to
keep the surface clean. Therefore it is not surprising that vitamin A deficiency is
associated with diarrheal and respiratory infections, presumably related to the
disintegrative condition of the respective intestinal or pulmonary lining [2, 5-7].
A further specific role of vitamin A deficiency in a functional immune system has
been appreciated since the early recognition of vitamin A as an essential nutrient.
The present interest in vitamin A as an immunoregulatory agent stems from two
sources: the vulnerability of vitamin A deficient humans and animals to infection and
the stimulatory effects of vitamin A, retinoic acid and synthetic retinoids on various
immune parameters [8]. In general, the conclusions are not consistent [9-12].
Thus vitamin A clearly plays an important, fascinating role in the nutrition of the
human species. Not only is it involved in key functions such as vision, growth and
protection from infection, but also in large doses it serves as a useful therapeutic
agent for cancer and skin diseases. Much interest is also presently being shown in the
possible independent action of carotenoids in fertility and as free radical and singlet
oxygen scavengers. Thus, with its many functions and possible roles in health and
disease, vitamin A is assured of continued attention from the scientific community in
the coming years [13].
As an illustration shown in Figure I, it seems clear that once the cycle of vitamin
A deficiency and infections commence, they impact on each other. Indonesian
studies have confirmed that vitamin A deficiency is a key variable in xerophthalmia,
23
Increased metabolic demands

Interruption normal feeding
Infection Vitamin A
deficiency
Reduced resistance to infection

Loss of epithelial integrity
Impaired immune system
Figure 1. Cycle of disease and deficiency
it further suggests that the role of vitamin A has expanded in its influence on child
morbidity and mortality. In general, this has become obvious from laboratory and
animal experiments and limited field trials. Replication and further studies are now
underway, one of which is currently being undertaken by the University of Gajah
Mada and Johns Hopkins University in the Yogyakarta area.
Clinical aspects and treatment of vitamin A deficiency
In Jakarta a series of scientific meetings on xerophthalmia have been convened in

1974, 1980, and 1988, respectively, which were sponsored by WHO, IVACG and the
Government of Indonesia to standardize the clinical classification of xerophthalmia
for assessment and to evaluate the purposes and criteria for public health sig-
nificance. The current usage of the classification is shown in Table 1.
In 1988 WHO [14] issued a guide to their use in the treatment and prevention of
vitamin A deficiency and xerophthalmia.
As can be seen from Table 2 in the annexes, the therapeutic schemes are for
children over 1 year and under 6 years old. Also important to note is that for a
woman of reproductive age, pregnant or not, with nightblindness or Bitot's spot, a
daily dose of 10,000 ill of vitamin A should be orally administered (1 sugar coated
tablet) for 2 weeks. Special considerations are addressed to the disease target
distribution (Table 3) which involves the administration of a high dose of vitamin A
24
Table I a. Clinical classification of xerophthalmia and public health criteria (WHO, 1981 revision)
Night blindness (XN)

Conjunctival xerosis (XIA)
Bitot's spot (XIB)
Corneal xerosis (X2)
Corneal ulcerationlkeratomalacia < 1/3 corneal surface (X3A)
Corneal ulcerationlkeratomalacia;::: 1/3 corneal surface (X3B)
Corneal scar (XS)
Xerophthalmia fundus (XF)
Table lb. Prevalence criteria for a xerophthalmia problem of public health significance (WHO, 1981
revision)
Nightblindness (XN) 1.0%

Bitot's spot (XIB) 0.5%
Corneal xerosis/corneal ulcerationlkeratomalacia (X2/X3A/X3B) 0.01%
Corneal scar (XS) 0.05%
Plasma vitamin A < 10 ug/dl 5.0%
Table 2. Xerophthalmia treatment schedule for children over I year and under 6 years old
Immediately on diagnosis 200 000 IV vitamin A orally'

The following day 200 000 IV vitamin A orally
4 weeks later 200 000 IV vitamin A orally
• If there is persistent vomiting or profuse diarrhoea, an intramuscular injection of 100,000 IU of water-

miscible vitamin A (but not an oil based preparation) may be substituted for the first dose. The use of
sterile syringes and needles is, of course, essential.
Children under I year old and children of any age who weigh less than 8 kg: TREAT with half the
doses shown above.
- 200,000 IU - which will give protection from 3 to 6 months to individuals at

special risk of developing vitamin A deficiency. Infants and children with infections
such as acute or prolonged diarrhea, acute lower respiratory infections, or severe
protein-energy malnutrition who present themselves for treatment at a health centre
are the most important groups requiring vitamin A supplementation in any targeted
programmes.
The following presentation highlights the recent advances from the South Asia
Regional Meeting, November 3-5, 1988 in Jakarta [3].
Different diagnostic techniques
New techniques are being developed, tested and increasingly used. These are
included in Tables 4, 5, 6.
Each country in the region should determine the magnitude and severity of
vitamin A deficiency by choosing indicators which can meet programme goals.
25
Table 3. Disease-targeted prevention schedule for pre-school children at high risk (e.g. those presenting
themselves at a health centre with measles', severe protein-energy malnutrition, acute or prolonged
diarrhea, or acute lower respiratory infections)
Children over 1 year and under 200,000 IV of vitamin A orally at the time of first contact
6 years old with health worker for each episode of iIlness b
Infants under 1 year old 100,000 IU of vitamin A orally at the time of first contact
and children of any age with health workers for each episode of iIIness b
who weigh less than 8 kg
• In areas where measles is a particularly severe disease, with a high mortality and a high risk of
blindness, as in Africa, it is appropriate to apply the full treatment for xerophthalmia.
b This dose should not be given to children who have already received a high dose vitamin A supplement
within the preceding month.
These indicators should be used to establish cut-offs for programme management

and to make decisions which are dependent upon the purpose, policy, the resources
of the country and the considerations of the chosen indicators' strengths and
limitations.
Procedures for data collection analysis and interpretation will depend upon the
local situation, but in general two types of information for sound programme
management were identified.
First level - scientific reports that give details of the design and procedures used
in implementing and evaluation programmes.
Second level - summary programme performance reports for policy makers in
terms that are understandable, brief, and capable of being used for making decisions.
Whenever possible the same indicators should be used when comparing data from
population assessment and monitoring.
Research priorities in vitamin A deficiency in the region
It has been accepted that vitamin A deficiency has a systemic effect on human
health, which exceeds xerophthalmia or blindness, and is a public health problem.
The main causes of death in childhood in the region are diarrhea, protein energy
malnutrition, and acute respiratory infection. These diseases are associated with
vitamin A deficiency. At this stage much more needs to be learned, and research
activities are suggested to elucidate this relationship. In addition, evidence is needed
to explain the relationship between community and behavioural practices with
vitamin A deficiencies. It has also been suggested that how health services respond
to vitamin A deficiency should be studied. It has been noted by the group that within
the region there has been a varying degree of government support and concern
devoted to vitamin A programmes. Vitamin A programmes undoubtedly compete
with other health programmes and with other government investments.
26
Table 4. Indicators for assessing vitamin A deficiency
Specific indicators
Clinical diagnosed xerophthalmia"
Biochemical
- Serum vitamin A (distribution curve)"
- Relative dose response (RDR), modified relative dose response (MRDR)"
- The shift of serum vitamin A in population (.1. Prop Serum VA < 20 mg/dl)
- Isotope dilution
- Liver biopsy
Physiological
- Dark adaptation test (DAT)
- Relative dark adaptation test (RDAT)
- History of night blindness (XN)"
Cytology
- Conjunctival impression cytology (CIC-A)"
- Buccal smear (need further study)
- Skin
General indicators
Dietary";
- Quantitative
- Semi quantitative
- Qualitative
Proxy indicators
Child mortality rates b
Infectious disease incidence/prevalence
Infestation prevalence
Age-Specific breastfeeding patterns
Stunting malnutritionb
Poverty indicators
Sanitation levels
Community access factors
Agricultural productionb
Disaster proneness
Cultural practices
" Recommended aids for consideration in public health programme.

b Data which may be available from secondary sources by province.
The explanation of success and failure of vitamin A campaigns are not yet clearly
defined. There is a possibility that convergence of various programmes could
influence the goal of the vitamin A programmes.
The emphasis of health-related research on vitamin A deficiency should focus on
three components, i.e. health service research, biological/technology related research,
and communitylbehaviouraI!psychological research.
The discussions and deliberation on issues by the participants were centered
around these three aspects. The outcome of the discussions was formulated as areas
of research important in developing effective vitamin A programmes. Efforts have
27
Table 5. Purpose for which diagnostic indicators of vitamin A risk may be needed.
Purpose Specific indicators General indicators
I. Population assessment Clinical xerophthalmia Dietary intake

Biochemical:
- serum vitamin A
- RDR
-MRDR
Physiology:
- history of XN
Cytology:
- CIC
2. Individual Clinical xerophthalmia

Physiology:
- history of XN
Biochemical:
- relative dose response
3. Evaluation Clinical xerophthalmia Dietary intake

Biochemical:
- shift ( ) population
serum vitamin A
- relative dose response
Cytology:
- CIC
4. Monitoring/surveillance Clinical xerophthalmia Dietary intake

Physiology: Proxi indicators:
- history of XN - market food price
- programme progress
indicated (input/output)
- other proximations
5. Research Many of the above indicators can be applied depending on the

purpose and the available resources.
Develop new diagnostic techniques
been made to establish research priorities. Three questions have been used in order to
detennine these priorities, namely:
1. What research-generated infonnation does the policy maker need in the region?
2. What are the major optional research questions?
3. What could be surrogate indicators of vitamin A deficiency?
It is agreed that each member country has its own priorities. However, the outcome
of the discussions is useful as generalized guidelines can be found to give countries
priority for vitamin A deficiency control programmes.
The following tables summarize the area of priority and interest in research on
vitamin A that could be adopted by member countries in the region (see Table 7).
28
Table 6. Strengths and limitations of recommended diagnostic aids for vitamin A deficiency.
Diagnostic aid Strengths Limitation
Clinical - Non invasive - Large samples needed

xerophthalmia - Widely used - Trained observer
- Standardized - Does not detect marginal status
- Established - Variable, specificity
Biochemical - Highly specific - Invasive

- Detect marginal VA status - Sophisticated
- Cut off points established - Highly trained technicians
- Easy to interpret - Requires special equipment and
handling specimens
CIC-A - Detects marginal VA status - Undermined specificity

- Cut off points established - Interpretation not yet standardized
- Minimum invasive - Limitation for children below two years
- Stability of specimens
- Minimum laboratory
Dietary - Widely used - Inaccurate

- Non invasive - Incomplete food tables
- Minimum training required - Does not measure status
- Inexpensive
- Educational benefit
- Information on cause
- Many methods
History of - Minimum training - Variable specificity

night blindness - Frequently used - Subjective
- Specific local term - Recall bias
- Early sign - Observer error
- Cheap
- Fast
The role of different interventions and delivery mechanisms in vitamin

A supplementation
Major intervention strategies in preventing and controlling vitamin A deficiency are:

Dietary improvement through nutrition education,
Distribution of large periodic doses of vitamin A and
Fortification.
Various means to implement major strategies are:

Through primary health care,
- Using social marketing techniques,
- Targeted/specific agricultural programme and food policy.
Each strategy needs prerequisites to maximize its usefulness. Prerequisites for
dietary improvement through nutrition education are:
29
Table 7a. Research priorities in vitamin A deficiency in relation to community behaviour
Question related to
Community behaviour Research Operational Surrogate

generated research indicators
information
1. Role of non-health sectors in H H

controlling vitamin A deficiency:
- Industry
- Religious leader
- Education system
- NOO's
- Communication industry
2. Cultural and behavioural determinants L H
of dietary intake of vitamin A rich foods
3. Intra family determinant of inadequate L H
feeding practices
4. Positive deviance in high risk area H
5. Identification of successful community L H
programmes of prevention and control of
vitamin A deficiency
6. Causes of gross vitamin A deficiencies, L H
massive campaigns taken place
7. Behavioural and socio-economic indicators L H
of high risk families and communities
8. What are the components of the health model L M M
that are amenable to change and their impact
on vitamin A status
- Internal education
- Housing (density of housing)
- Sanitary practice
- Price subsidization
- Vulnerable group feeding programmes
- Permissiveness in feeding practice
in child rearing
H =high priority; M =medium priority; L =low priority.

a need for a national policy on nutrition education with the focus on vitamin A
issues,
identification of locally available foods and processing methods,
availability of funding, manpower, management, capability and technology,
the need for collaboration among multisectoral agencies based on effective
communication strategies, especially involving communications specialists,
anthropologists, educators and informal community leaders.
Vitamin A deficiency control through diet modification can only be achieved if there
are:
increased production of vitamin A rich foods
improved food preparation methods and serving practice and
an increased intake of available (and culturally appropriate) vitamin A rich foods.
30
Table 7b. Research priorities in vitamin A deficiency in relation to health system
Question related to
Health system Research Operational Surrogate

infonnation
1. Situation analysis of effectiveness of the H H

health system dealing with Vitamin A deficiency
2. Improving programme acceptability by H H
high risk population
3. Vitamin A programme in disaster, drought, H H H
refuge situation
4. Financing mechanism for vitamin A programme H H
5. Attribution of successful voluntary community
participation H H
6. Policy analysis of government levels H H
to vitamin A partiCipation
7. Agricultural sector approach H H
8. Cost effectiveness of alternative intervention H
strategies as compared with other governments
9. Assessment of the magnitude of the problem H
10. Development of an effective vitamin A programme H L H
- Surveillance
- Monitoring
- Evaluation
- Cost analysis
I 1. Manpower and training requirement for L H
successful vitamin A programme
12. Surrogate indicators of high risk population L M H
H =high priority; M =medium priority; L =low priority.
For vitamin A capsule distribution, the prerequisites are:

a need for a national policy decision based on appropriate information on the
magnitude of the problem collected through national population based surveys,
updating information during programme implementation which covers changes in
magnitude, changes in methods or strategies and incorporating results of new
research,
availability of technology and expertise to support the interventions,
adequate funding, manpower, infrastructures on effective delivery systems, and
the need for operational research on programme management.
Prerequisites for vitamin A fortification are a need for specific national policy based
on adequate research and technology particular to each country in the region which
covers the magnitude of the problem, appropriateness of the intervention, availability
of fortifiable foods, funding, manpower, technology and management capability.
The programme goal is that "All Countries in the region should be free from
vitamin A deficiency as a public health problem by the year 2000".
The target groups of vitamin A intervention are normally all pre-school children
(under 6 years old). If there is a need to set priorities, children with clinical signs of
vitamin A deficiency should be addressed first, and children with measles, diarrhea,
31
Table 7e. Research priorities in vitamin A deficiency in relation to biological technology
Question related to
Related biological technology Research Operational Surrogate

infonnation
1. Impact on mortality among pre-school children H

of vitamin A supplementation programmes
where support is weak
2. Impact of vitamin A supplementation on H
future productivity and social functioning
3. Feasibility of combined supplementation H H
intervention
4. Impact on morbidity among pre-school children
of vitamin A supplementation programmes
- ARI H
- Diarrhea H
- Malaria L
- Growth and development M
- Xerophthalmia L
- Measles and other immunicable diseases M
- Iron status L
- Immuno competence M
- Other parasitic infection L
5. System for extended delivery of H M
vitamin A supplementation
6. Vitamin A deficiency among school children H L
7. New diagnostic test for vitamin A deficiency H H
8. Detenninants of severity of vitamin A deficiency
among province/countries in the region
- Dietary intake L H
- Nutritional status L H H
- Socio-economic status L H
- Seasonality L
- Comorbidity L
- Response to vitamin A supplementation L
- Other ecological factors L
9. Effects of vitamin A status in response M
to therapeutic agent
10. Effect of vitamin A status in response M
to environmental toxicants
11. Mechanism of action of vitamin A on L L
morbidity condition
12. Mechanism of action of morbidity on L L
vitamin A deficiency
13. Vitamin A deficiency among older adults L L
- Pregnant women and low birth weight
H = high priority; M =medium priority; L =low priority.
acute respiratory tract infections and protein energy malnutrition would be secon-
dary; thirdly, pregnant and lactating mothers should be included in prevention
programmes. High doses of vitamin A may also be given to the mother within two
32
months after delivery, when new pregnancy is unlikely; however, pregnant mothers
should never be given a large dose of vitamin A due to potential toxicity, and
teratogenic effects. The recommended maximum safe intake for pregnancy is 10,000
IU daily.
Resources for vitamin A intervention include:
continuous supplies of vitamin A capsules
established effective delivery systems
adequate availability of vitamin A rich foods and oil as well as other absorption
enchancers
available training programmes based on the country's needs,
available materials for information, education and communication (1EC) and its
development through operational research, and
tools for monitoring and evaluation.
There is a strong need for regional networking which aims at:
regular meetings for the exchange of experience and information on the specific
issues in vitamin A intervention (i.e. Social marketing),
exchange of expertise,
clearing house/information pool, and
training exchanges.
New prospects for prevention through horticultural and educational

intervention strategy
As mentioned above, since 1985 FAO, together with other UN agencies, has been
undertaking a complementary agricultural intervention strategy providing long-term
solutions to increase production and consumption of vitamin A and carotene
(provitamin A) rich foods.
Over the years, in spite of large scale massive oral dose intervention in some
countries, e.g. in India (BRD-Medical College, Gorakpur, India), an increasing trend
of xerophthalmia among pre-school children has been reported. In order to view the
problem in the right perspective, Singh [15], in the light of past experience, ques-
tioned massive oral dose intervention and pointed to weaknesses due to: (1) insecure
national efforts and uncertain welfare programmes based on unstable economics, (2)
dependence on foreign help subject to variable politico-economic pressures, (3)
unreliable infrastructures under the supervision of an inexperienced bureaucratic
machinery which lacks full understanding of the problems at hand, (4) the fact that in
many cases, the recipient population, often poor social segments, are treated as
experimental subjects by new intervention techniques. If not successful, such
intervention may only help commercial producers of the specific vitamin A supple-
ment and administrators of the actual programmes. The history of intervention
failures clearly indicates that the prevention and the cure for vitamin A deficiency
must look much more to solutions based on local self-reliant social and technical
approaches.
33
Table 8. Effect oCred palm oil on vitamin A in serum and )!;erophthalmia prevalency.
Village Group Date of Vitamin A Xerophthalmic

enmination in serum (ugldl) (%)
GunungPati Red palm oil Dec 63 14.7±O.61 7.3

(4g1day) June 64 17.8±0.84 2.0
March 65 19.4±0.56 0.5
KedungPane Decolorized Dec 63 13.0±0.64 3.5

palm oil June 64 14.0± 1.04 1.5
March 65 13.0± 1.44 1.9
Table 9. Vitamin A and carotene serum content before and after treatment with palm oil.
Before Vitamin A Recovery After

deficiency period
Child Vit.A Caro- symptoms (day) Vit.A Caro-
name IJ.gldl tene IJ.gldl tene
1. Djaenudin 22 Hem,Bit 7.50 22 37

2. Nurdin 9 46 Hem,Bit 7.30 8 28
3. Sodat 12 18 Hem,Bit 30.50 13 40
4. Mulyana 5 21 Hem,Bit 3.30 23 36
5. Udin 12 26 Bit .60 15 30
6. Udjang 5 20 Hem,Bit 7.70 19 25
7. Rachmat 4 7 Hem,Bit 3.30 8 25
8. TitinH. 7 23 Hem,Bit 3.30 13 41
9. Romli 5 5 Hem,Bit 7.36 20 64
10. Priyatni 12 51 Keratomalacia Staphyloma 23 66
11. Kardi 12 23 Keratomalacia Staphyloma 20 36
12. Engkus 12 35 Hem,Bit 7.30 20 28
13. Somad 11 13 Hem, Bit 2.21 15 39
14. Emi 4 11 Hem, Bit 2.30 8 17
15. Aang 10 36 Hem,Bit 2.30 23 32
Mean 8.5 23.8 16.6 36.5

S.E. 0.83 1.34 3.34
A recent study of a village near Varanasi [16] gives appropriate attention to the
potential of green leafy vegetables as a practical source of vitamin A in human diets.
The following argument serves to show that the promotion of green leaves, in their
natural state or processed, is a workable and self-reliant approach to reducing the
incidence of vitamin A deficiency.
It is also simultaneously desirable to concentrate on efforts to improve overall
nutritional, socio-economic and living conditions by initiating long-term sustainable
family welfare programmes to alleviate poverty. In Indonesia an attempt was made
in the sixties, and is now being again explored, to use red palm oil which contains
60,000 ug/l00 g B-carotenes.
Oey et al. [17] and Karyadi et al. [18] have shown the preventive and curative
effect of oral administration of the crude palm oil, as shown in Tables 8 and 9.
34
Further study is needed to look at the possibilities of administering it in a con-

centrated form by the use of posyandus once a month.
The Ministry of Agriculture, together with the Ministry of Health, will launch an
integrated health-oriented agriculture programme through gardening programmes for
the provision of green leafy vegetables in the posyandus (village level primary health
care services) in the forthcoming Five Year Development Plan.
This strategy of prevention requires a need for policy programme development
involving multisectoral collaboration, especially involving community leaders as
well as proper nutrition education for mothers.
References
1. Ministry of Health, Indonesia: Nutritional blindness prevention project. Characterization of vitamin

A deficiency and xerophthalmia and the design of effective intervention program. Final Report
MOH, Jakarta, Indonesia, 1980.
2. Sommer A: Nutritional blindness xerophthalmia. Oxford Univ Press, London and New York, 1982.
3. WHO. Regional Office for South East Asia and Ministry of Health, Republic of Indonesia: Latest
advances in the control of vitamin A deficiency and its impact on health, Jakarta, Indonesia,
November 3-5, 1988.
4. Darwin Karyadi: Establishment of a regional network to strengthen national vitamin A deficiency
control programme. Mission findings and recommendation, FAO, STC Report, 1988.
5. Helen Keller International: The effect of diarrheal disease on the absorption and utilization of
vitamin A. Information Paper 2 by Donna Nager, 1979.
6. Sommer A: The Lancet 1:1411,1980.
7. Sommer A et al.: Am J Clin Nutr 40: 1090, 1984.
8. Nauss KM: In: Bauernfeind JC (ed) Vitamin A deficiency and its control, Academic Press Inc,
Orlando, 1986, p. 207.
9. Bhaskaram C and Reddy V: B Med J 3:522, 1975.
10. Brown KH et al.: Am J Clin Nutr 33:212,1980.
11. Ifekwunigwe AE et al. Am J Clin Nutr 33:621,1980.
12. Keusch GT: Paediatric Annals 11:1004, 1982.
13. Olson JA: In: Bauernfeind JC (ed) Vitamin A deficiency and its control, Academic Press Inc,
Orlando, 1986, p. 57.
14. WHO, Unicef, IVACG. Task Force: Vitamin A supplements. A guide to their use in the treatment
and prevention of vitamin A deficiency and xerophthalmia, 1988.
15. Singh N: Xerophthalmia and the green leaves. Gardens for Development Vol. 2, No. I, 1988.
16. Xerophthalmia Club. Bulletin No. 36, 1987.
17. Oey KL, Liem IT, Rose CS, Prawiranegara DD and Gyorgy P: Am J Clin Nutr 20: 1267, 1967.
18. Karyadi et al.: In: Penilaian keadaan gizi anak penderita defisiensi vitamin A dengan latar be1akang
sosial ekonomi dan pengobatan dengan minyak kelapa sawit (Elais guineensis Jacq), Prosiding
Kongres Nasional Pertama, Persatuan Ahli Mata, 1968, p. 169.
Infection and malnutrition
LEONARDO MATA
Introduction
The simultaneous occurrence of famine and pestilence has been recorded by man
since the beginning of written history. While the presence of infectious diseases was
recorded in the early descriptions of kwashiorkor, the role of infection in the
causality of malnutrition was overlooked for about three decades. Recognition of
infection-malnutrition interactions in medical literature, was made by Scrimshaw,
Taylor and Gordon, in the late 1950s [1]. They proposed the existence of synergistic
and antagonistic interactions between infection and malnutrition [1]. In synergism
there is an exacerbation of the effect of infection, while infection aggravates the
nutritional deficiency: the outcome of such an interaction is assumed to be greater
than the sum of the two factors. Synergism is frequent in developing countries where
infectious diseases are highly prevalent, and diets are often deficient.
The other kind of interaction, antagonism, results when one or more nutrients
become deficient, impairing replication of the infectious agent, whether a parasite,
bacterium or virus [1]. Antagonism has been demonstrated in several animal models
[1], and additional reports appear intermittently [2]. However, it rarely occurs in
man, and when it does, it is under extreme nutritional deprivation [3].
Long-term field studies in rural societies living with deprivation, showed the
importance of infection-nutrition interactions in determining morbidity, growth
failure, acute malnutrition, and mortality [4-7]. In such societies, many infants
exhibit growth faltering generally at about 3 to 6 months, if they are breast-fed;
infants living under deprivation, and who are prematurely weaned or are bottle-fed
shortly after birth, show growth faltering at an earlier date [5,7,8]. It is not yet clear
whether stunting is primarily related to inadequate food supplementation when the
mother's milk becomes insufficient, or to the occurrence of repeated infectious
diseases, or to both. While a decrease in the supply of human milk without the
benefit of adequate food supplements is the rule in many developing societies, the
striking event during weaning is the recurrence of infectious diseases [5].
H. K.A. Visser and J. G. Bindels (eds.) Child Nutrition in South East Asia. 35-44.
36
Studies of infection-malnutrition interactions
Animal models
Animals severely deprived of either calories, protein, vitamins, amino acids, and
trace elements, or of several of these factors, show an increased clinical response to
individual viruses, rickettsias, bacteria and protozoa, when compared with well-
nourished control animals. The course of infection in deprived animals is more
severe, more prolonged, and more lethal than in the controls [1,3]. The enhanced
clinical responses are attributed to alterations in the host-specific and non-specific
immune competence. On the other hand, infectious organisms are capable of altering
the nutritional state of animals inoculated by natural or unnatural routes. Animal
studies are fundamental for understanding infection-malnutrition interactions, but
they may not be extrapolated to human situations because (a) experimental nutri-
tional deficiencies leading to host resistance in animals do not easily occur in human
populations; and (b) the type, dose, specificity, and route of infection used in animal
models may not be the same when observed naturally in humans.
Humans
Clinical studies in confined young adult volunteers demonstrated that most metabolic
pathways become altered during the course of infection. Viral, bacterial, and
parasitic infections diminish appetite, increase catabolism after the anabolic phase,
and induce metabolic alterations consisting of nutrient over-utilization, nutrient
wastage, nutrient sequestration and nutrient diversion [9, 10]. Abnormalities occur
even with mild or asymptomatic infections. Fever, anxiety and other forms of stress,
and the need to satisfy gluconeogenesis induce hormonal alterations, with mobiliza-
tion of amino acids from the muscle, and with nitrogen loss. There are losses of
sodium, potassium, bicarbonate, chloride, and phosphate in intestinal, respiratory and
systemic infections [9, 10]. There is increased synthesis of liver enzymes; of foreign
protein, lipid, and carbohydrate; and of acute-phase reactant proteins (nutrient
diversion). Also, there is a decrease of circulating trace elements (zinc, iron and
copper) with concentration in hepatic cells (nutrient sequestration). Since zinc plays
a role in cell-mediated immunity, the sequestration of this and other trace elements
during infection possibly alters immune responses. Last, but not least, infected
children studied in hospitals are noted for their losses of urinary nitrogen and plasma
vitamin A.
Effect of infection on nutrition and growth
Long term studies in impoverished traditional societies show that infection is the
most important factor inducing malnutrition and growth retardation [4-6, 11]. High
37
rates of infection are, in tum, detennined by deficient social, cultural and economic
conditions which favor deficient sanitation and personal hygiene [5].
Maternal infection during pregnancy
In the pregnant woman, viruses, bacteria, and protozoa may reach the placenta and
fetus. The risk of maternal infection is enhanced by physiologic alterations during
gestation. The risk for the fetus is low, with no more than 2 percent of infants born
infected in industrial countries [12]. Such rates may increase during epidemics, for
instance, of rubella, and is particularly troublesome in populations affected by
acquired immunodeficiency syndrome (AIDS) [13]. Intrauterine or intrapartum
infection may not induce clinical manifestations in the mother or fetus. In a few
cases, however, there is interruption of pregnancy, pre-tenn delivery, fetal growth
retardation, embryopathy, overt disease in the infant or acute or chronic sequelae
(12]. Pathogenic mechanisms include decreased placental blood flow, penneability
to metabolites and antigens, decreased cell multiplication, cell proliferation,
inflammation, and necrosis (12].
Maternal infection is significantly more frequent in traditional rather than in
industrial societies (5, 14, 15]. In poor urban and rural areas, high levels of cord IgM
as well as intrauterine infections, occur in greater frequency than in populations with
better conditions (5, 14]. For instance, in Guatemala, Peru, and Colombia, a high
proportion of newborn babies have high levels of serum IgM (5, 15, 16], suggesting
common fetal antigenic stimulation; this, in tum, might be related to fetal growth
retardation. Such greater occurrence, however, does not explain the high incidence of
prematurity and fetal growth retardation observed in deprived populations, which
may reach as much as 40 percent (5]. The finding, however, deserves scrutiny, as an
alternative hypothesis to that which purports that feeding food supplements during
gestation has the most important effect on the outcome of pregnancy.
Early neonatal infection
During exclusive breast-feeding, infections with Giardia, Entamoeba histolytica,

Shigella, and Salmonella were rare or few, and usually asymptomatic (5]. With
weaning, infection increased to attain high rates by the end of the first year and
especially during the second and third years of life. Multiple and chronic infections
were more often seen at these ages. Virus shedding occurred from the first weeks of
life (5, 15], and increased strongly in the second semester, and in the second and
third years constituted a virtual "viral flora" [5].
The implications of excessive infections are (a) precocious development of serum
IgG and IgM to attain very high levels in the first year of life; and (b) damage to the
intestinal mucosa resulting in inflammation and malabsorption [5, 11].
38
Infectious disease
The interaction of recurrent morbidity with growth is evident in individual weight

and length charts of children corresponding to the first 2 years of life (Figure 1).
Infectious diseases are exceedingly common: in the Guatemalan village of Santa
Maria Cauque, children had about seven episodes per year in the first 3 years of life
[5, 11]. Respiratory and diarrheal disease were commonest in infancy and in the
second year of life. Diarrhea increased with age to reach the highest values in the
second year of life.
Common infections induce anorexia, nutrient losses, metabolic alterations,
hormonal imbalances and alterations in the immune function [9, 10]. These events
may lead to wasting, stunting, acute and chronic malnutrition, and death.
Reduced food consumption
This is overt in enteric disease, in which anorexia, fever, diarrhea, and vomiting are
important causes of reduced food consumption by village children [4, 15]. Food
withdrawal and inappropriate treatment and feeding during convalescence are
implications derived from cultural tradition, taboos and beliefs. Energy and protein
consumption are reduced by 20 to 60 percent during infectious disease [11, 17]. As
much as 16 percent of the overall dietary calories and 18 percent of the protein may
not be consumed by weaned children as a result of diarrhea [15]. This is more
significant in regions where food consumption is already below 80 percent of the
90 90
80Y 040 80Y 002
B W" 2711 COl B.W." C!369 til
B. L ., -477 em S.L., 4S.lcm
80 80
O(Z3)
£.",.101,""
".~/.
70 OIBI ., 70
,
0(3) Alctl".
'"a:
W
,
0(20)
T,icll"fI,
'"a:
W
.... ....
,
D~I 0(8)
W 0(111 $1,.11"/0/,,.. W Sill• • "

:::E :::E E_"'·*'TfI~.
~ ~
Z Z
w w
u u
50
I~~~~""II'
Dill
cec
40 40
Ir ...' .lIk Ir...' .lIk
0
I
3
I
6 9
I~~ISA
I I 9 15
INI A
12 15 18 21 24 0 3 6 12 IB 21 24
MONTHS MONTHS
Figure I. Body length of two children from birth to two years of age related to enteric infections and
diarrhea episodes. The panel on the right corresponds to a child with fetal growth retardation; on the left.
data for a child with adequate intrauterine growth is shown.
39
recommended level. Anorexia is found in acute respiratory disease, Whooping cough,

and measles, but reduced food consumption is observed even with mild infections
like the common cold.
Altered digestive-absorptive processes
Diarrhea accelerates transit through the intestinal tract, restricting consumption and
interfering with digestion. The mucosal surface can be covered and altered by
Giardia or bacteria, or it may be damaged by invading viruses, bacteria, and
parasites [18]. Microscopic mucosal lesions concurrent with bacterial colonization
are seen in persons living in the tropics [19]; jejunitis and the accompanying
malabsorption of sugars, fats, and vitamins disappear spontaneously in persons
settling in an environment with better sanitation [20]. Enterotoxins stimulate
formation of cAMP and cGMP, impairing absorption of sodium and water [21].
Certain bacteria split bile salts, reducing micelle formation and fat absorption and
increasing the bile acid pool, an irritant to the mucosa; microbial overgrowth in the
small intestine is related to malnutrition in children [22]. Diarrhea is an abnormal
hypersecretory state, with losses of fluid, electrolytes, and protein [21]. Hypersecre-
tion is caused by imbalances in digestive-absorptive processes; changes in hydros-
tatic pressure, enterotoxins, bile and fatty acids, hormones and neurotransmitters; and
greater calcium permeability [21]. Dehydration is a serious consequence of diarrhea
and other febrile processes, rendering the host acutely malnourished within a matter
of hours. Rehydration results in a prompt return to normality and has a profound
influence on recuperation of nutritional sequelae [23].
Altered immune response
Infectious agents may alter one or more cell types and functions of the immune
system [1,2,5, 7, 24, 25]. Many viral infections have been known for their depres-
sant effect on the immune system. More recent examples are the retroviruses, known
for their lymphotropic and lympholytic effects. The most dramatic effect is that of
the human immunodeficiency virus (HIV) on CD4-lymphocytes. Progressive
depletion of lymphocytes favors development of opportunistic agents and tumors,
resulting in AIDS [26].
Wasting and stunting effect
Repetitive bouts of infection lead to progressive deterioration of the nutritional

status, evident as weight faltering and impaired linear growth. These changes were
obvious in the growth curves of breast-fed Cauque children, who were observed
40
prospectively from birth to school age [5, 27]. In general, growth velocity during
exclusive breast-feeding (3 to 6 months) was comparable to that of reference curves
of the National Center for Health Statistics (NCHS), but after the onset of supplemen-
tary feeding, there was growth faltering. Infants maintained at the breast beyond six
months of age, and for prolonged periods, suffer from subtle starvation. However,
the most striking negative impact on growth is associated, more often than not, with
infectious disease episodes during the weaning process. Many infections relate to
weaning foods contaminated with pathogens, or to contaminated hands, water, and
utensils [28].
Figure 1 shows body length of two boys from the village of Santa Maria Cauque,
Guatemala, from birth to two years of age, compared to the 50th percentile of the
curve of the National Center for Health Statistics. All episodes of diarrhea and
enteric agents experienced by the children in the first two years of life, are also
recorded. The study was conducted in 1964-1969 [29]. The boy at the left had
adequate intrauterine growth and adequate weight and length at birth. The child at
the right suffered from fetal growth retardation (about 40 percent of infants in that
village have low birth weights). All diarrheas were associated with periods of arrest
in linear growth, regardless of infectious etiology [18, 29]. The negative effect of
diarrhea on growth was more marked in the child who had fetal growth retardation.
Acute weight loss (5-10 percent) was common [30]. Wasting (deficit of weight for
height greater than 80 percent persisted for weeks or months, contributing to the
genesis of marasmus [5], and was complicated in some instances by inadequate
feeding during convalescence. Wasting and stunting also occur in conjunction with
other diseases, notoriously measles and pertussis. Impaired growth was greater if
infants had fetal growth retardation or had been permanently weaned. Similar
observations have been obtained by field and hospital studies of children in other
latitudes [4, 5, 6, 31, 32]. Furthermore, effective treatment of intestinal helminthiasis
resulted in improved child growth (33).
Severe malnutrition
Undernourished children become critically ill after acute watery diarrhea, dysentery,
measles, pertussis, malaria and other infectious diseases [4, 5, 17], or social and
psychological stress or neglect [34]. Sudden falls in serum albumin after measles,
hookworm and diarrhea, may precipitate kwashiorkor [17]; a protein-loosing
enteropathy has been recognized in rotavirus and Shigella diarrheas [35]. Severe
malnutrition may develop several weeks after the occurrence of bouts of diarrhea,
measles or malaria, often quite independent of food availability [5].
Infectious disease particularly eliminates children who are weak, who are born
premature or with fetal growth retardation, or who suffer from immunological,
genetic, or other degenerative disorders. When there is inadequate or deficient
maternal (family) technology, insufficient health services like immunization, and
41
poor coverage with oral rehydration therapy (ORT), then infectious diseases become
the main determinants of malnutrition and death of children in developing countries
[29,30).
Effect of nutrition on resistance
Intrauterine growth retardation impairs immune competence, and small-for-

gestational-age infants exhibit a higher incidence of infection and mortality in the
first years of life [5, 29]. Breast-feeding is the most important defense barrier against
infection in the first months of life [8]. Other defense mechanisms are gastric acidity,
intestinal motility, intestinal microflora, and immune response and its amplification
[36, 37]. Well-nourished individuals better withstand dehydration, nutrient losses,
and other infection-induced alterations.
In postnatal undernutrition, whether of nutritional origin (as in famine) or due to
nutrition-infection interaction (the most common form in developing countries) there
are alterations in integrity of skin and mucosae, hypochlorhydria, and abnormalities
in intestinal microbiota. Severely malnourished children may exhibit decreased
lymphoid cells, impaired function of T and B lymphocytes, and diminished synthesis
of complement and secretory IgA [36, 37]. Undernourished children experience
greater severity, chronicity, and mortality due to infections. The duration of disease
and the carrier state of certain agents is generally longer in undernourished as
compared to well-nourished individuals [1,5].
Under severe food deprivation, depletion of the immune system may rather
suppress infection, and starved individuals can develop acute infectious diseases
upon nutritional recuperation [13]. On the other hand, infections either enhance
immunity or induce secondary immunodeficiency or immunological paralysis [37].
The public health implications are enhanced symptoms, duration, and augmented
mortality from infectious disease [1,4,5,38].
Discussion
There is less evidence now that malnutrition increases susceptibility to infection.

Rather, the onset of an infection relates to the frequency of host exposure to the
infectious agent. This is supported by the high incidence of enteric infection and
diarrhea among well-nourished travellers to tropical regions, which incriminates host
and environmental hygiene much more than the nutritional status in the outcome of
infectious diseases.
The negative effects of infection start with the "generalized acute phase metabolic
response" [9, 10], which is rather stereotyped in adults infected experimentally, and
very likely in the child, judged by the striking similarity in nitrogen balance and
clinical effects observed in children with natural infections. The response in children
42
is probably more serious, and additional variables must be taken into account, such
as the nature of infection, the ecosystem of the child, added risk factors like stress or
child abuse, chronic infectious or parasitic diseases, a background of fetal growth
retardation and similar factors.
The adverse effect of infection relates to this acute-phase metabolic response,
mediators of which are interleukin 1 and cachectin [39,40], hormone-like substances
released into the blood stream by tissue macrophages and blood mononuclear under
the stimulus of an infection. With interleukin-l, the immediate response in the brain
is the onset of anorexia; in the bone marrow, the release of neutrophils for inflamma-
tion. In the liver, there is an increased uptake of aminoacids and trace elements and
an increased synthesis of acute-phase reactant proteins; in skeletal muscle, an
increased breakdown of protein and utilization of released aminoacids for energy. In
the pancreas, there is a release of insulin and glucagon and enhanced glucose
utilization for maintenance of fever; this, in tum, diminishes utilization of free fatty
acids and ketones for energy sources. Clinical findings of persons suffering from
infections relate to the actions unleashed by interleukin-l, for instance, fever,
anorexia and malaise, the first responses to infection. There is an acceleration of
metabolism of body cells, resulting in marked loss of muscle, fatigue and weakness;
there is leukocytosis. In chronic infections there is marked weight loss and anemia.
Cachexia may appear in some individuals harboring certain infectious agents, which
are found to be associated with the release of tumor necrosis factor or cachectin.
Infection has negative effects on the pregnant woman, fetus, infant, and young
child. The result of infection is growth retardation, morbidity, malnutrition, and
death, occurring at any stage of fetal or child development. The role of infection as a
determinant of malnutrition in developing countries is now clear. Even when poor
village children consume adequate amounts of calories and protein during disease-
free episodes, the onset of infection leads to reduced food consumption [11, 17].
The nutritional damage inflicted by infection is enhanced by inappropriate child
care, lack of prompt ORT and treatment, and inadequate feeding during convales-
cence [30]. Recurrent diarrhea, anorexia, and fever - coupled with poor diets and
hygiene - induce marasmus and kwashiorkor. The risk of death increases for pre-
term and small for gestational age infants [5], particularly if they have become
wasted and/or stunted [4, 5, 38]. The well-nourished child, however, is also at risk of
death from infectious disease if the environment is hostile. Survivors tend to remain
stunted, and women previously undernourished may in tum give birth to pre-term
and retarded neonates, in a vicious circle.
Our field studies emphasize the need to improve environmental sanitation,
personal hygiene, and child-rearing practices, promoting child nutrition and health
within a holistic approach. Populations suffering from acute food shortages are
exceptional, and in such cases food distribution should be given high priority [29].
A practical way of knowing whether infection or diet is the best target for
intervention in a given ecosystem, is to assess the prevalence of undernutrition and
mortality. If an excess occurrence of malnutrition and death is observed in all ages
43
and social classes, then food is probably the limiting factor, for instance, in famines
[29]. The common constraint in developing countries, however, does not seem to be
a deficient food intake, because even in very poor villages, children frequently
consume at least 80 to 90 percent of the FAO/WHO recommended calories and
protein. In that instance, undernutrition is generally restricted to infants and young
children, and occasionally to the very old and feeble. In the latter, a greater impor-
tance of infection is apparent.
Epidemiological evidence shows that control and prevention of infectious disease,
particularly diarrhea, correlates with improved nutrition, growth, and survival [29]. It
is expected that widespread application of ORT, improved water supplies, expanded
primary health services (immunization, family planning), and health education will
improve nutrition worldwide, without unjustified emphasis on food distribution
programs. This comment, however, does not apply to populations affected by a
proven limitation of food availability.
Acknowledgement
Supported by the University of Costa Rica, the National Council for Scientific
Research and Technology (CONlCIT), the· Ministry of Health (Costa Rica) and
Nutricia (the Netherlands).
References
1. Scrimshaw NS, Taylor CE and Gordon JE: Am J Med Sci 237:367-403, 1959.
2. TetzlaffCL, Carlomagno MA and McMurray ON: Med Microbiol ImmunoI177:305-315, 1988.
3. Murray MJ, Murray AB: Lancet 1:123-125, 1977.
4. McGregor lA, Rahman AK, Thomson AM et al. Trans R Soc Trop Med Hyg 64:48-77, 1970.
5. Mata U: The children of Santa Maria Cauqu6. A prospective field study of health and growth, MIT
Press, Cambridge, 1978.
6. GuerrantRL, Kirchoff LV, Shields OS et al. JInfect Dis 148:986-997, 1983.
7. WhiteheadRG, Coward WA, LunnPG etal. Trans R SocTrop Med Hyg71:189-195, 1977.
8. Jellife DB, Jellife EFP: Human milk in the modern world. Psychosocial, nutritional and economic
significance, Oxford University Press, Oxford, 1978.
9. Beisel WR: Am J Clin Nutr 30:1236-1247, 1977.
10. Beisel WR: Annu Rev Med 26:9-20,1975.
II. Mata U, Kronmal RA, Urrutia JJ, et al. Am J Clin Nutr 30:1215-1227, 1977.
12. Elliott K and Knight J: Intrauterine infections. Ciba Found Symp 10 (new series), Elsevier, Excerpta
Medica, Amsterdam, 1974.
13. HoffR, Berardi VP, Weiblen BJ et al: N Eng J Med, 318:525-530, 1988.
14. Alford CA, FoCt JW, Blankenship WJ et al: J Pediat 75:1167-1178,1969.
15. Mata L, Urrutia JJ, Serrato G et al: Am J Clin Nutr 30:1834-1842, 1977.
16. McMurray ON, De Aly AC, Rey H: Bull Pan Am Health Organ 14:376-385, 1980.
17. Whitehead RG: In: Isliker H and Schurch B (eds) The impact of malnutrition on immune defense in
parasitic infestation, Hans Huber Pub, Bern, 1981, p. 15.
18. Mata L: In: Bellanti JA (ed) Acute diarrhea: Its nutritional consequences in children. Nestl6,
Vevey/Raven Press, New York, 1983, p. 85.
19. Gangarosa EJ, Beisel WR, Benyajati C, et al: Am J Trop Med 9:125-135,1971.
44
20. Lindebaum J, Gerson CD, Kent TH; Ann Intern Med 74:218-222, 1971.
21. Field M: In: McClung HJ (ed) Etiology, pathophysiology, and treatment of acute gastroenteritis,
Ross Lab, Columbus, 1978, p. 57,
22. Mala U, Jimenez F, Cord6n M, Schneider RE, Viteri F, Rosales R, and Prera E: Am J Clin Nutr
25:1118-1126,1972.
23. Molla A, Molla AM, Sarker SA, Khatoon M and Rahaman MM: In: Chen LC and Scrimshaw NS
(eds) Diarrhea and malnutrition. Plenum Pub, New York, 1983, p. 143.
24. World Health Organization: Bull WId Hlth Org 46:537-546, 1972.
25. Chandra RK and Newberne PM: Nutrition, immunity and infection. Mechanisms of interactions.
Plenum, N.Y., 1977.
26. Gotlieb MS: In: Cooney TG, Ward TT (eds) AIDS and other medical problems in the male
homosexual. W.B. Saunders Co, Philadelphia, 1978, p. 651.
27. Mata U,Urrutia JJ and Lechtig A: Am J Clin Nutr24:249-259, 1971
28. Mata L: In: Bellanti JA (ed) Acute diarrhea Its nutritional consequences in children. Nestle,
Vevey/Raven Press, N.Y., 1983, p. 3.
29. Mata L: In: MacKenzie JS (ed) Viral diseases in South-East Asia and the Western Pacific, Academic
Press, Sydney, 1982, p. 56.
30. Mata L: Proc Nutr Soc 38:29-40, 1979.
31. Tomkins AM, Garlick PJ, Schofield WN and Waterlow JC: Clin Sci 65:313-324,1983.
32. Black RE, Brown KH and Becker S: Pediatrics 73:799-805, 1984.
33. Stephenson LS, Latham MC, Kurz KM, Kinoti SN and Brigham H: Am J Trop Med Hyg 41:78-87,
1989.
34. Goodall J: Proc Nutr Soc 38: 17-28, 1979.
35. Wahed A, Rahaman MM, Gilman RH et al: Dacca, ICDDR,B, 1981, 12 pp.
36. DuPont HL and Pickering LK: Infections of the gastrointestinal tract. Microbiology, pathophysiol-
ogy and clinical features. Plenum, N.Y., 1980.
37. Targett GAT: In: Isliker Hand Schurch B (eds), The impact of malnutrition on immune defense in
parasitic infestation, Hans Huber Pub, Bern, 1980, p. 158.
38. Chen LC, Alauddin-Chowdhury AKM and Huffman SL: Am J Clin Nutr 33:1836-1845,1980.
39. DinarelIo CA: N Eng J Med 311:1413-1418, 1984.
40. Beutler B and Cerami A: N Eng J Med 316:379-385,1987.
The pattern of feeding and the nutritional status of infants
and children in Indonesia
SOEDARTI SURBAKTI, YAYAH K. HUSAINI and

MAHDIN A. HUSAINI
Introduction
Health is essential to the satisfaction of human needs and to an improved quality of

life. The first years of life are crucial in laying the foundation of good health. At this
time certain specific biological and psychosocial needs must be met to ensure the
survival and healthy development of the child and future adult. One of these vital
needs is the nutrition of the infant.
Malnutrition is the most widespread and serious problem affecting young
children; combined with infection it is the major cause of death and is responsible for
the retarded growth and development of a large number of children [1].
Infant feeding varies widely among different cultures in the quantity and quality
of the supplementary foods and in the manner in which the endogenous food is
provided. In most areas in Indonesia, apart from breast milk, supplementation
already starts at the age of two months in the form of biscuit soaked in tea, or in the
form of fruit - either banana, papaya, or orange juice. By the age of four or five
months the first major solid food is introduced, i.e., a porridge consisting of rice
flour and cane sugar. At the age of five to six months another solid food, more
complex than the porridge, is introduced. This dish is called nasi tim; nasi means rice
and tim means soft-prepared. The ingredients of nasi tim are rice, vegetables
(spinach or tomato or carrot), animal protein (chicken liver, sliced meat, fish) or
vegetable protein (tahu, tempe). Ketchup or bouillon is added as flavouring in order
to make the dish more palatable. The above regimen is continued until the child
reaches the age of 12 months and even longer until the child starts to share the menu
of other family members, with the exception of highly spiced dishes [2].
Malnutrition is more than a medical problem. It is related to the conditions of
families and women; it has social, economic and political roots; and it is closely
associated with poverty [3]. Regarding the complex background to the problem of
malnutrition, a national survey has been conducted to provide information on infant
feeding and growth. This global information is urgently needed for national politics
and programme development to deal effectively with the problem.
H.K.A. Visser and J.G. Bindels (eds.) Child Nutrition in South East Asia, 45-54.
46
Material and methods
In 1986, a nationwide survey was undertaken to investigate infant-feeding practices

and their nutritional status through SUSENAS (National Social Economic Survey)
activities. SUSENAS has two main data: 1) The main data on the characteristics of
the population, such as the name of the family member, the relationship with the
head of the family, sex, age, marital status, education, and employment; 2) Modules
which form the direction of the survey, such as health, nutrition, houses, welfare,
criminals, etc. The first main data are always collected in every SUSENAS survey,
but the second main data or modules, depending on the particular information, are
only collected according to the aim of the survey conducted at that time. The
sampling of the families was designed to be based on "population block census" to
cover children from birth to three years living in urban and rural areas and including
all main ethnic, socio-economic, and religious groups. There was a random selection
within each target area. All houses in which the children lived were identified and
numbered during a pre-survey visit to a sample village. During the survey, each
marked house was visited by a trained enumerator, and a responsible adult was
interviewed. The data source was a questionnaire. When a home was visited, the
information was always obtained from the person who fed the child. Most often this
was the mother, and only in very few cases was this the grandmother or another
person.
An interview was conducted using a standard questionnaire to obtain information
on the following: level of education, the mother's occupation and age, total expendi-
ture of the family, type of infant feeding, duration of breast feeding, and the age at
which breast-feeding was supplemented with other foods. Ages were determined by
asking directly the date of birth or with the assistance of a local events calendar or
with the child's growth chart which is usually kept in the home.
After interviewing the mothers or responsible persons, the trained statistician field
workers measured the weight of the children (either naked or very lightly clothed) on
"dacin" scales. The Harvard anthropometric standard was used in the analyses. The
cut-off identifying mild and moderate PCM was 80 percent from the median weight
for age, and severe PCM was 60 percent from the median weight for age.
Results
Table I shows the results of a retrospective investigation with infants aged 5-23
months, concerning the types of feeding used when the children were 0 to 5 months
of age, as related to the family's expenditure per month in rural and urban areas.
More than 95 percent of infants were breast-fed, except for families with expenditure
above Rp 300.000,- (US $190); about 15 percent were not breast-fed. There was a
tendency for more infants to be given breast-milk substitutes when the family's
expenditure was increasing. During the first five months of life, more than 50 percent
Table 1. Percentage of children receiving breast-milk and supplementary foods during the age 0-5 months, according to the household expenditures.
Household expendi- Areas Number of Breast- Breast milk Soft foods Solid Adult
tures per month children milk subsitutes foods foods
(1) (2) (3) (4) (5) (6) (7) (8)
(Rp) (N) (%) (%) (%) (%) (%)
<25.000 Indonesia 421,384 99.57 6.43 54.68 9.60 0.48

Rural 403,391 99.83 5.98 54.59 10.03 0.50
Urban 17,993 93.89 16.51 56.82
25.000- 49.999 Indonesia 3,208,710 99.10 9.24 55.10 9.93 0.79
Rural 2,836,121 99.17 8.78 54.63 9.89 0.81
Urban 372,689 98.53 12.80 58.72 10.31 0.64
50.000- 74.999 Indonesia 2,261,823 98.28 13.78 52.69 9.60 0.43
Rural 1,743,164 98.83 11.05 51.90 10.12 0.36
Urban 518,659 96.41 22.98 55.33 7.84 0.69
75.000- 99.999 Indonesia 1,205,439 97.29 17.40 51.58 7.55 0.30
Rural 759,174 98.05 13.52 48.73 7.78 0.30
Urban 446,265 95.98 24.01 56.42 7.16 0.46
100.000-149.999 Indonesia 528,766 95.41 26.70 58.99 9.69 0.68
Rural 255,928 95.57 20.22 56.30 9.65 0.74
Urban 272,838 95.26 32.78 61.51 9.74 0.61
150.000-199.999 Indonesia 167,181 96.42 31.93 67.47 12.76 0.39
Rural 68,896 99.19 23.49 70.80 9.86
Urban 98,285 94.48 37.85 65.13 14.80 0.66
200.000-299.999 Indonesia 71,134 95.31 37.47 57.52 3.89
Rural 22,169 95.83 39.56 71.87 7.35
Urban 48,965 95.07 36.53 51.03 2.32
2:300.000 Indonesia 18,326 85.64 56.37 46.19 7.98
Rural 6,020 74.53 44.45 55.D7
Urban 12,306 91.07 62.21 41.85 11.85 .j::.
-..l
US$ 1 equal to Rp 1.600,-.
00
"""
Table 2. Relationship between the duration of breast feeding and the education of the mothers
Education of mother Duration of breast-feeding
<3mos 3-5mos 6-8 mos 9-11 mos 12-17 mos 18-24mos >24mos
(%) (%) (%) (%) (%) (%) (%)
A. Separate from mother 8.84 10.92 12.43 9.76 32.10 16.81 9.13
B. Lives with mother:

None 0.39 0.31 1.92 5.93 23.42 26.15 41.88
Primary non completed 0.98 1.71 2.47 6.39 21.16 41.52 41.52
Primary completed 1.35 1.73 4.27 9.84 25.75 23.55 33.51
Junior High School completed 1.63 4.10 4.60 11.91 24.43 22.66 30.66
Senior High School completed 5.45 6.22 8.50 11.74 25.75 26.37 15.97
Diploma I/lI/lII
(One to three years 12.87 15.35 18.44 9.04 17.06 11.54 5.71
after high school)
University completed 20.75 12.28 32.60 13.66 15.00 2.88 2.83
Number of children 65,480 91,146 155,691 365,541 978,424 1,009,618 1,474,883

49
of infants were given soft foods such as ripe mashed banana ("pi sang ambon") or
soft-boiled rice (bubur beras encer) or porridge made of rice flour (bubur tepung).
The number of infants who were given "nasi tim" (see the Introduction supra) has
also been recorded at the age of 0 to 5 months. However, the number of infants who
were fed nasi tim was less than 10 percent, and as daily food less than 1% at the age
of 0-5 months.
When approaching 12 months (9 to 11 months), about 25 percent of infants were
still being given soft food, and this level declined further when the infants became
older. The number of infants who were given solid foods (nasi tim) increased. At the
age of 1 to 5 months this was only 10 percent, but at 9 to 11 months more than 30
percent of infants were given solid foods. In rural areas as well as in urban areas the
numbers of infants who were fed soft foods and solid foods were not influenced by
household expenditures. By the age of 9 to 11 months, 5 to 15 percent of infants
were receiving adult foods. The numbers of infants receiving adult foods decreased
in the higher household expenditure groups.
When infants' breast-feeding histories were reconstructed by asking mothers who
had children of 24 to 35 months of age, it was found that overall only 1.71 percent of
infants had never received breast-milk (in urban areas it was 3.89 percent and in rural
areas it was 1.05 percent. The numbers of infants who had never been breast-fed
were not influenced by the education of the mothers. The mothers who had received
the highest education still breast-fed their infants, but the duration of breast-feeding
was shorter, as shown in Table 2. Those breast-fed for more than 12 months were
markedly common in lower education groups of mothers, particularly in rural areas.
Figure 1 clearly demonstrates the relationship between the level of education and
the duration of breast-feeding. The number of infants who were breast-fed for the
duration of 24 months were found to be 40 percent for mothers who had no formal
education, and only 13 percent for mothers who had education at University level.
On the other hand, the number of mothers who breast-fed their infants for the
duration of three months was only 2 percent for mothers with no education; in
marked contrast it was found to be 24 percent for the mothers with university
education.
An analysis of the nutritional status of the children against the educational
standards of the mothers and fathers are shown in Figures 2 and 3. Mothers with a
higher education tend to have children with better nutritional status (Figure 2). A
similar association between the number of well-nourished children with the level of
education of the fathers is also demonstrated in Figure 3, although this tendency was
less significant compared to the mothers. The studies revealed that these association
curves are similar between rural and urban areas.
Household expenditure was also associated with the numbers of well-nourished
children as shown in Figure 4. The families with the higher household expenditures
tended to have more well-nourished children. This pattern is more clearly
demonstrated in urban areas rather than in rural areas. In rural areas, although there
was a tendency of increasing household expenditures followed by an increase in the
50
percentage af children
40 -B- breast-fed <3 months
-+- breast-fed >24months
30
20
10
no education prim.incompl. prim.compl. jr.high schaol sr.high school university

education of mothers
Figure I. The relationship between the education of mothers and the percentage of children who were
breast-fed for less than 3 months and above 24 months.
percentage of wellinourished children
90
80
70
60
50
40
30
20~--'-------~--------r-------~-------r------~----
no education prim.incompl. prim.compl. jr.high school sr.high school university

education of mothers
Figure 2. Percentage of well-nourished children, age 3-4 years, according to the education of the mothers.
51
percentage of wellnourished children
70
60
50
40
urban
-+- urban + rural

-* rural
30~--,-------,-------,-------,------'r------'-
no education prim.incompl. prim.campl. jr.high school sr.high school university
education of fathers
Figure 3. Percentage of well-nourished children, age 3-4 years, according to the education of the fathers.
percentage of wellnourished children
55
45
35
25~--,------~------.-------~----~----~------~----~---
< 25 25-49 50-74 75-99 100-149 150-199 200-299 > 300

household expenditure per month x1000Rp
Figure 4. Percentage of well-nourished children, age 3-4 years, based on household expenditures.
52
number of well-nourished children, this association remained consistent only up to a

certain level of household expenditure (Rp 150.000,- to Rp 199.000,- per month).
Above this level of expenditure, the numbers of well-nourished children decrease
with the increasing household expenditures.
Discussion
The present study demonstrates that the mothers still entertained strong positive
feelings about the importance of breast-feeding. In Indonesia, breast-feeding is still
generally practised in both rural as well as urban areas - over 95% of mothers still
manage to breast-feed their infants for the first three months. Only 1.71 percent of
children had never been breast-fed. In urban communities, the percentage of children
who had never been breast-fed was 3.89 percent and in rural areas it was 1.05
percent, but these numbers were still comparatively low. However, the number of
infants receiving both breast-milk and breast-milk substitutes were high. In this
situation the function of breast-milk is starting to be replaced by breast-milk
subtitutes, with the results that the infants could be earlier weaned.
There is a difference in breast-feeding patterns between urban and rural com-
munities. Mothers in urban communities breast-feeding their infants were likely to
be for shorter periods of time than rural mothers. This supports the hypothesis that
breast-feeding is made more difficult by many of the women from urban com-
munities [4]. However, it is particularly interesting that the duration of breast-feeding
for the families from rural areas with household expenditures above Rp 200.000;
(equal to US$ 125) per month was shorter than the urban areas of the same level of
household expenditure. Although there is no sound information on this evidence, it is
thought that the mothers with higher socio-economic status in rural areas, have
tended to wean their infants earlier. This situation is likely to be associated with the
prevalence of undernourished children in those groups. Although the percentage of
well-nourished children rose with the higher household expenditures, above the level
of Rp 200.000,- (US$ 125), however, the percentage of well-nourished children is
decreasing.
There is a tendency that the higher the education of the mothers, the shorter the
duration of breast-feeding. As in many other developing countries [1], the higher
education of mothers and the higher income of families were associated with a
shorter duration of breast-feeding. Findings in the present study show that the
percentage of the mothers who breast-fed their infants for a duration of less than
three months was becoming higher with the higher education of the mothers. On the
other hand, the numbers of mothers who breast-fed for more than 24 months, was
decreasing with the higher education of mothers.
As Table 2 indicates, nearly 50 percent of the mothers still breast-fed their infants
after 12 months, therefore breast-milk continued to make a significant contribution to
the calorie and nutrients intake of the child.
53
Calorie and nutrients intake is likely to vary with the educational and economic
background of the parents. The educational level of the mothers and fathers may
have positive effects on the calorie and nutrients intake of children because those
parents with a higher education are more responsive to the needs of the children [5].
This present study indicates that the percentage of well-nourished children rises in
accordance with the higher education of mothers as well as that of the fathers.
There was no marked difference in feeding patterns between rural and urban
communities, although there was a tendency for more rural mothers to introduce
adult foods to their infants. Feeding adult foods to infants may have a negative effect
on the nutrients intake of the children [3,6]. At the time when infants begin to need
foods other than breast-milk to satisfy their nutrient needs, they are not yet ready
physiologically for an adult diet; their stomachs are still small and they cannot chew.
This present study indicates that 15 percent of infants, age 9-11 months started
eating adult diets. And the previous study shows that the cereal grain is often
replaced by starchy foods, such as roots and cassava, which are more bulky [7].
Supplementary foods were given to infants as early as one to two months [7]. In
the present study, more than half the mothers were giving supplementary foods to
their infants during the first five months of life. Recent work by Martorell and Klein
[8] has confirmed that weaning foods are the major vehicle for the transmission of
fecal pathogens during infancy in developing countries. The problem may arise
because of unclean cooking vessels and utensils or because of poor personal hygiene
in food handling. Incomplete cooking may fail to kill all pathogens and if the foods
are allowed to stand, even for relatively short periods of time, a proliferation of
bacteria will follow. In a study conducted by Soekirman [9] of a number of infants
belonging to low economic working mothers in Semarang, Central Java, he notes
that at three months and older, 80 to 91 percent (N = 238) of the infants had been
exposed to solid foods. The early introduction of solid food increased the risk of
diarrhea. Among the four types of beikosts studied (solid food, water, tea and fruit
juice), solid foods (mashed banana and rice, or rice-flour porridge) and tea were
likely to be negatively associated with nutritional status. The percentage of infants
who were mild, moderate, and severe PEM in the group of infants who had been
given solid food was double the percentage of similar nutritional status of those who
had not been given solid food. The reason for this effect is likely to be that foods and
utensils were easily contaminated by a diarrhea-causing organism. Therefore,
emphasizing the education of mothers on how to prepare and use hygienically
acceptable supplementary foods from relatively inexpensive but nutritious locally
available foods is seriously needed. Government policies should address this issue,
especially for low income groups, in order to educate and facilitate them to be able to
feed their children properly.
References
1. WHO/UNICEF: Infant and young child feeding current issues, Geneva, 1981.
2. Husaini MA: Biological aspects: experience and needs oflocal weaning food in Indonesia. Presented
54
at the Workshop on Low Cost Weaning Foods, Jakarta, 11-12 July 1988.
3. Underwood BA: Weaning practices in deprived environments: the weaning dilemma. Presented at
the Workshop on Current Issues in Feeding the Normal Infant, Palm Springs, CA, April 8-11, 1984.
4. Anyanwu RC and Enwonwu CO, Nigeria Food and Nutrition Bulletin 7:33, 1985.
5. Husaini MA, Husaini YK and Surbakti S, Pola Makanan dan gizi bayi dan anak BALITA. (Feeding
pattems and nutritional status of infants and under five children). Presented at the Second Congress
of Indonesian Statistician Association, Jakarta, 29-30 June, 1987.
6. Tin Tin 00 and Khin Maun Naing, Food and Nutrition Bulletin 7:47, 1985.
7. Husaini, MA and Karyadi D, Bulletin IImu Pengetahuan dan Teknologi LIPI (Bulletin of the
National Institute of Sciences), 4:35, 1983.
8. Martorell R and Klein RE, Nutr Rep Int 21:447,1980.
9. Soekirman, The effect of maternal employment on nutritional status of infants from low income
households in Central Java, Ph D Thesis, Cornell University, Ithaca, 1983.
Discussion: Session 1
Paper by C. Gopalan
M.S. Trastotenoyo:
Please allow me to make a comment on "GOBI" as a slogan. In my opinion, GOBI should be
understood as an integrated program and not as a slogan. For instance, GOBI is always linked with the
three F's: family planning, female feeding supplementation, and female education. Therefore, if we
implement this program carefully and in line with the development of the country and the local
situation, I still think it will be a relevant aid to improve child nutrition and welfare.
L.Mata:
With respect to "GOBI", I would like to point to one of the several shortcomings of this program. The
environmental effect is completely disregarded. I will deal with this extensively in my paper.
l.A. Kusin:
In connection with this discussion on GOBI, to my knowledge GOBI-FFF was indeed brought up as a
package but I know only very few countries which implement the three F's. Also worth mentioning
here is the discussion and correspondence which appeared in the last few months in The Lancet in
which UNICEF has replaced GOBI by the Burmaco initiative which means comprehensive primary
health care within rural development. This probably means that it was admitted that the approach of
GOBI is too narrow.
A.W. Qureshi:
I would like to comment on oral rehydratation not being the solution for control of the diarrheal
disease. The problems we face as clinicians are major constraints like the budgets and lack of
education, and likewise the poor environmental conditions. As long as we cannot achieve these
objectives, I think we have to resort to the ORS program to prevent the increasing mortality and
morbidity caused by diarrheal disease.
A.A. Kahn:
In the report on one of the recent studies of the Nutrition Society of India, it was mentioned that a
major contribution to the "smallness" of the low birthweight baby may take place in the last trimester.
In my opinion, the first trimester is at least as important in this respect since that is the time when the
growth of the brain takes place.
C.Gopalan:
I am in full agreement with this remark. In fact I would even go further than the first trimester of
pregnancy, we should go back to the girl, which I also have stated in my paper. The point I was
making on supplementation of pregnant women during the third trimester of pregnancy was that even
if you are able to reach only the last trimester, you will still be doing some benefit. Since the actual
H.KA. Visser and l.G. Bindels (eds.) Child Nutrition in South East Asia, 55-60.
56
situation in Asian counuies is that presently we are only able to reach them at the third trimester, or in
the middle of the second one, nuuitional supplements will still be worthwhile, but that this is not the
ideal situation, I entirely agree.
l.A. Kusin:
Please allow me to give a reaction to Dr. Kahn's remark. The experiments of Dr. Widdowson in the
early fifties have clearly shown that up to 28 weeks, when the fetus reaches a weight of about 1 kg, the
mother does not need too much extra energy or protein and brain development is fully protected.
Therefore, from a physiological point of view, I would still prefer to consider the third trimester as the
primary target for nuuitional supplementation, which is also supported by the data of Noy.
Paper by D. Karyadi
C.Gopa/an:
Dr. Karyadi could you please comment on the claims from Indonesia with respect to the other
dimensions of the problem of vitamin A deficiency and possible other beneficial effects of vitamin A
supplementation on child survival. To my knowledge this still seems to be a controversial area.
D.Karyadi:
At the present state of the art, indeed those other dimensions of vitamin A brought up by the famous
"Aceh"-study, launched by our group jointly with the group from the Johns Hopkins Hospital,
Baltimore, are still controversial. In my paper I already mentioned that the study to confirm this
controversy is underway. Moreover, also in the Gambia, Nepal and India they are replicating the study
to understand me mechanism of me decreased morbidity and mortality. We have just published a study
in the American Journal of Clinical Nuuition in which vitamin A and vitamin E status were studied
together with me response to oral doses of both vitamins. We found that vitamin E inadequacy, which
impairs vitamin A absorption and storage, may well contribute to the high incidence of clinical
vitamin A deficiency in West Javan children. Now we are investigating some immunological aspects
in the children in that study which have been supplemented with vitamin E compared wim the control
group. In conclusion, I have good hopes mat at least parts of the controversies wim respect to me
dramatic effect of vitamin A on mortality will be cleared up at the end of mis year during the meeting
of the International Vitamin A Consultative Group in Nepal. We do recognise mat the dramatic
decrease in mortality is not singularly caused by vitamin A alone. A challenging question we are now
working on is what role vitamin E may have in mis respect.
P. Bhaskaram:
I would like to point out that in the acute spread of infections, diarrhea and particularly measles, the
cause for corneal lesions may be entirely different. Although corneal lesions associated wim
malnutrition in the post-infectious period which may occur may be prevented by vitamin A, me
etiology of eye lesions in diseases like diarrhea and measles is entirely different. Therefore, though the
vitamin A levels are lower in these acute infectious periods, the main mechanism involved in
producing corneal lesions are more likely to be due to the disease per se, rather than to vitamin A
deficiency. So, I am afraid that if we treat children with corneal lesions only with vitamin A, we are
ignoring other aspects by which we can prevent blindness in these children.
D. Karyadi:
Yes, I agree with you. Because of the two factors of pathogenesis, we should not intervene in the
controls wim vitamin A alone. We should aim for a comprehensive approach of interventions.
l.A. Kusin:
If we talk about dietary means to obtain sufficient vitamin A, only dark green leafy vegetables are
mentioned and sometimes red palm oil. In semi-arid zones, however, the availability of dark green
leafy vegetables is very limited and seasonally conditioned. Since one of me staple foods in these
regions is yellow maize. we expected it to be a rich source of beta-carotene. Unfortunately, we only
found very little beta-carotene in the maize strains from Madura; the yellow colour is mainly due to
xanthophyls. Would it not be a good idea to look for maize strains rich in beta-carotene?
57
D.Karyadi:
This is a very useful observation and I agree that we should especially focus on locally available
vitamin sources. The search for corn variants rich in beta-carotene is just an example. We should look
for plant resources rich in vitamin A as well as in other nutrients coming from the fat such as vitamin
E, and also alternative sources for vitamin C should be assessed.
A.M. Molla:
Although in many Asian Countries dark green leafy vegetables are available, we still see a high degree
of vitamin A defiCiency in countries like Bangladesh. In spite of the consumption of dark green leafy
vegetables, this is difficult to explain. Other factors like the state of the intestine, chronic malabsorp-
tion, state of diarrhea or amount of fat consumed along with the dark green leafy vegetables might also
be involved.
Paper by L. Mata
S.H. Pudjiadi:
Dr. Mata, am I right in observing that you did not use a local standard growth curve but the Harvard
one?
L.Mata:
In our first studies the only standard growth curve we had available was the Jackson-Kelly standard,
which is the Harvard one for the first two years and the Iowa one afterwards. Later we adopted the
NCHS standard. I believe that the growth potential for children in Guatemala and Costa Rica is the
same and in general I do not think that it is appropriate to have a smaller standard for Latin America
than for North America.
E. Suroto-Hamzah:
Could you please comment about treatment of infections in malnourished children in a hospital setting,
especially with respect to the question when to start giving antibiotics. From one side you want to
protect the child and from the other side you do not want to raise resistency because of the misuse of
antibiotics.
L.Mata:
Actually, the treatment of acute severely malnourished children was not my primary field of interest; I
am more interested to prevent this situation. However, I would mention some cases where infections in
children were successfully treated with colostrum and human milk. My view is that you have to treat
children with antibiotics if it is very obvious, but not by giving them in a preventive way. Further, the
child with acute malnutrition should be managed with a hypercaloric diet which could be simply
supplementation with milk.
M.S. Trastotenoyo:
Do you have breast milk banks in your hospitals or do you just collect it?
L. Mata:
We began doing so in 1977 when Dr. Largea from Argentina showed that by just giving full fresh
colostrum and milk, the neonatal mortality could be tremendously lowered without the use of
antibiotics. Nowadays we have to cope with the problem of AIDS which can be transmitted via breast
milk. Although there is no good evidence for this transmission, we are now advising that the milk be
pasteurized when it is pooled.
M.Gracey:
There have been some important changes in the incidence of infection and the nutritional status of the
children in Santa Maria Cauque since you have studied them. Could you please comment on this.
58
L.Mata:
After the earthquake in 1976 the whole construction pattern of the village was changed. Houses do
now have tiles or cement floors which helps hygiene and children now sleep in separate beds because
the houses are warmer. However, the problem with the water still remains, so still many children
suffer from diarrhea. Because of these slight changes, the weight of the women has increased and also
the birth weight of the infants that have been born recently, although we still have many low birth
weight infants. In my opinion we could bring about more pronounced changes if we capitalise on the
knowledge that the people in the village have to improve on the way the mother handles everything
around the baby, to curtail infection. I would call this maternal technology: the mother has to know
how to adequately store and handle the water for drinking and cooking, she has to learn the correct
handling of human and animal faeces, she has to learn about oral rehydration (preferably with respect
to cereal-based ORT preparations) and she has to learn about reproductive behavior.
Paper by S. Surbakti, Y.K. Husaini and M.A. Husaini

A.W. Qureshi:
Please allow me to ask two questions. When you say breast feeding, do you mean only breast feeding
up to 5 months of age with no introduction of bottle feeding? We regularly see that soon after a child
is born, within one week or even the first day, some mothers give water and introduce the bottle. In
some places they do not even give colostrum, they just throw it away.
M.A. Husaini:
With respect to your first question, we only asked the mothers about breast feeding. Whether this
relates to exclusive breast feeding cannot be deduced from our data. About the second part: we also
see this experience here in which to even one day old children, water is given or even some honey.
M.S. Trastotenoyo:
Could you please give the exact definition of breast feeding for this study.
M.A. Husaini:
In this survey we just asked the mothers whether they breast fed their infants. When they already
stopped breast feeding we asked when they did stop and how long they actually did breast feed. We
did not ask the mothers on the frequency of nursing or whether they breast fed their infants at least
once a day.
c.-Y. Yeung:
Before I would like to ask you to what extent the infants you studied in your survey followed standard
growth curves, please allow me to make some general remarks on how the effects of tradition and
cultural practice influence the mother's choice of food for their infants, and as a result their growth.
Recently, very striking differences were found in the growth curves of Chinese infants either living in
China Town Toronto or in Scarborough, which is a new town near Toronto. In the first location, where
mothers adhere to the very traditional Chinese habits of infant feeding, essentially the same growth
curves were found as in much older studies with Chinese infants: about the age of 4-5 months, height
starts to falter off. In the latter location, however, where Chinese mothers adopted the usual North
American habit of feeding infants, growth rates up to 30 months were identical to the North American
growth rates. Also, for instance, when the child is ill, according to traditional Chinese habits of infant
feeding it is fed only with very diluted, very watery rice. Therefore, I believe that cultural habits are
the reason why the growth curve for traditionally reared Chinese children is lower compared to those
brought up in other locations.
M.A. Husaini:
We found in our survey with ten thousand children that the growth curves follow the Harvard standard
until 6 months of age after which they deviate. For the children in the well-off families we see that the
Harvard standard is followed much longer. We are planning to investigate this in more detail and also
pay attention to the diversification and regional aspects of infant feeding practices.
59
S.H. Pudjiadi:
I can add to this in that we found in Jakarta that the growth curves depend entirely on the social
economic condition and the education of the mothers. We have seen that until the age of 6 years,
growth is comparable to Harvard standards.
P. Bhaskaram:
Looking at your data, I observe that well-clothed and educated mothers breast fed their babies for
electively less time, but, however, subsequently it was found that the percentage of well-nourished
children was higher in these families. In contrast, you also show that poor uneducated mothers fed
their children on the breast for a longer time and subsequently malnutrition in these children was much
higher. Can these two be related and what would be your message regarding the weaning.
MA. Husaini:
We found in our study that infant feeding in quality and quantity is influenced by a number of factors,
for instance by the education of the mothers and also by income. Interestingly, in rural areas we see
that the education of the mothers and their income are not so closely related as in the urban regions.
Further, in rural areas mothers with higher income stop breast feeding earlier but this does not result in
a better outcome for the infants as was found in the urban areas. Therefore we would stress here the
importance of a government policy to improve the power of the mothers to feed the children
appropriately, especially in the rural areas. The nutritional status of the infants and children in our
studies is the result of a number of confounding factors and not only breast feeding. In our study we
found that the better nutritional status of the infants is related to the mothers' education, as is the fact
that they stop breast feeding earlier. It would, however, be foolish to conclude that the improved
nutritional status of the infants is caused by the fact that they have been breast fed for a shorter period.
A.A. Kahn:
Since you just answered the question I wanted to ask, please let me make a general remark that with
breast feeding the growth pattern depends on the quality of the mother's milk, in particular its fat
content. Many babies do very well on breast milk alone for a long time and I think in the developing
countries we must encourage prolonged breast feeding.
C. Gopalan:
Mister chairman, I am pleased to hear the very last remark of the speaker which has put the whole
issue in its proper perspective by bringing in the multifactoral factors. Otherwise, the interpretation
could be wrong.
General discussion: Session 1

I.A. Kusin:
If you compare the growth charts from the exclusively breast fed child with that of the supplemented
and weaned child during the first 6 months of life, for instance those shown by Dr. Mata, you see that
growth in these infants is exactly the same. Further, if you carefully read those many articles quoting
exclusively breast feeding as the ideal way of breast feeding, you find out that this only occurs in 25%
of the mothers. Among others, Dr. Rowland said that they probably represent a quite selective group
of good performers. In our study in Madura we found that only 5% of the infants were exclusively
breast fed, the others were force fed from the first or the second day onwards. My question is why do
these children also follow the same standard growth charts up to 4 months, and do not have diarrhea
for the first 4 months.
L.Mata:
My experience is that in very undisturbed peaceful societies breast feeding is universal and exclusive
for at least 3 months, in many cases even for 6-7 months. The growth in these infants is fully
compatible with standard growth curves such as the NCHS. If supplementation is started early, the
child can still show normal growth, provided it is surrounded with a marvellous protected environment
like we, for instance, have seen in our study in Costa Rica. However, I do not think that you can do
60
this in places where cholera, diarrhea and other infectious diseases are prevalent. In those regions
weaning foods are very dangerous if they cannot be prepared properly. In this respect we should also
remind ourselves that even if given in only small amounts to the infant, human milk still has
significant nutritional and health promoting effects. Because of its protecting and anti-inflamatory
properties based on the action of secretory IgA as well as the several other active substances, such as
the recently discovered hormonal and growth stimulating factors, human milk, even if given in minor
amounts, has a very important function for the optimal growth of the infant.
M.A. Husaini:
From our studies that we did 10 years ago in a very poor area, we indeed obtained similar data as you
mentioned from Madura showing adequate growth up to 6 months. I would speculate that perhaps next
to immunity the infant received from breast milk, also some immunity is obtained from the mother
during the period of pregnancy which might influence this prevention of infection up to a certain age.
ALi Ming Cheng:

In view of the ongoing discussion about breast feeding, I would like to make some comments about
the situation in Hong Kong. In the sixties, more than 50% of the mothers exclusively breast fed their
infants. This figure dropped to only 5% in 1979. Since then there have been efforts to promote breast
feeding and about three years ago two surveys were done to investigate its success. In our study we
found that beyond one month only about 7% of the mothers still breast fed their infants completely.
The other study reported a higher proportion, 28%, but beyond one month this figure was reduced to
about 8%. We found that there were three reasons why mothers did not breast feed: inconvenience,
embarrassment and work. Inconvenience was indicated because of social and dietary restrictions on
the mother, embarrassment is something we gathered from western culture where breasts are
associated with sex symbols and because of the widespread use of TV and advertisements. Work
should not really be an argument since women have maternity leave for 4-6 months.
W.C.Liu:
The situation in Taiwan is highly comparable to that in Hong Kong. In 1940 to 1950 we had 60-70%
breast feeding and now this figure is about 2%. We have made tremendous efforts to promote breast
feeding with several programs, but unfortunately without any success.
D.R. Karunaratne:
From the above comments on the very low rates of breast feeding in Hong Kong and Taiwan, I am
getting somewhat confused. Our main objective in the management of infant feeding is in my opinion
to reduce infant mortality and the other is to have ideal growth in these infants. If I consider the very
low rates of infant mortality in these countries, I think that they have achieved this goal. So should we
continue to promote breast feeding and try to decrease infant mortality in this way? Or should we aim
to follow the process Hong Kong and Taiwan has undergone which in fact succeeded in obtaining very
low infant mortality and ideal growth? I think that these are quite different messages and we should
make it clear which we should follow.
C. Gopa/an:
I think the message is already clear, that there has been a tremendous movement towards breast
feeding in the first world. I think the advantages of breast feeding, despite what you have seen with
respect to growth in the first few months, have been well documented. Its superiority has been
accepted very well by nutrition scientists and health scientists, so I do not think that you should just go
by parameters of the type of growth performance. Although, it has never been denied that with
artificial foods, if given in appropriate amounts, you can obtain satisfactory growth performance, I
think the message is still that:
- breast feeding is best,
- breast feeding should be exclusive for at least 4 months and,
- in the really poorest communities in the slum areas, breast feeding should be exclusive until 6
months, since the advantage of supplementation by the fourth month may be offset by the greater
chances of diarrhea.
Session 2
Nutrition of pregnant mother and infant
Chairman: Moelyono S. Trastotenoyo

The East Java pregnancy study: Effect of prenatal energy
supplementation on mother and infant
lA. KUSIN, SRI KARDJATI, C. DE WITH and

W.M.VANSTEENBERGEN
Introduction
In the Maternal and Child Health services of most developing countries more
emphasis is put on the child than on the mother.
Maternal nutritional health is even more neglected. Usually the only routine
activity is the distribution of iron-folate tablets to pregnant women. Nutrition
guidance is either not given or given in passing. Yet, the poor nutritional status of
women must encompass their capacity to meet the vigorous demands of their
multiple roles as mothers and productive workers. There is sufficient evidence to
show that in communities where malnutrition is prevalent among pre-school
children, women are undernourished as well. Such observations are not surprising. If
food availability is limited, the low priority accorded to women in intra-family food
distribution and traditional food practices, particularly during pregnancy, result in
low dietary intakes.
Nutrition Surveys in East Java showed that 18-42% of non-pregnant, non-
lactating women had a weight-for-height less than 90% of the Indonesian standard
[1]. The habitual diet provided on average 1500 kcal and 40 g protein, irrespective of
the physiological state [2]. Already at the age of 0-5 months, the percentage of
infants with a weight-for-age ofless than 80% of the Harvard standard was very high
[3].
Based on the results of these Nutrition Surveys it was assumed that low
birthweight and inadequate breast-milk yield due to undernutrition during pregnancy
and lactation are the key variables influencing infant survival and growth.
To assess the relation between maternal nutrition and reproductive performance in
terms of the bearing and nurturing of offspring, an experimental trial was conducted
(the East Java Pregnancy study = EJPS). Since energy was the most limiting in the
habitual diet [2], the hypotheses were that energy supplementation in the last
trimester of pregnancy will improve fetal growth and increase maternal fat stores.
Indirectly, the beneficial effect may be reflected in an increase in breast-milk yield
and postnatal growth.
H.K.A. Visser and J.C. Bindels (eds.) Child Nutrition in South East Asia. 63-87.
64
Subjects and methods
The study area
Three villages were selected from the region of Sampang, Madura, known to have
the poorest health indicators in the East Java province. They are located along the
southern coast in a semi-arid zone. The population, about 7,SOO individuals in 1981,
is a traditional Moslem community. Living conditions were modest, water was taken
from springs and latrines were non-existent. The main occupations are agriculture
and fishing. In the dry season salt production provides additional income. With good
rainy seasons, three crops of maize can be harvested. In the more fertile parts of the
villages one agricultural cycle can yield two crops of maize and one crop of rice.
Cassava and sweet potatoes are cultivated after the "long" rainy season in
March-June. Some pulses are intercropped with maize.
The habitual diet consists of a mixture of maize and rice. In the lean period (the
dry season) rice is replaced by cassava. The main side dish is fresh fish. Since the
rainy season lasts only 60-90 days per year, leafy vegetables are available for a short
period. Gourd-type of vegetables are therefore usually eaten.
Experimental design
After a census, from August 1981 onwards all newly married couples and mothers
with children from O-S years were visited monthly at home to diagnose a pregnancy
as early as possible.
A baseline period (August 1981-August 1982) was included to assess the habitual
dietary intake of pregnant women and the birthweight distribution. The study design
was a controlled, randomized trial.
For the allocation of pregnant women to the high or low energy supplement,
random number tables were used. Sample size was calculated to detect a difference
in mean birthweight of 100 g between the experimental groups at a significance level
of S% with a power of 80%, using a two-tailed test (N;;;; ISO). Due to the large
variation in quantities of supplement consumed (compliance), the experimental
period was extended to increase the sample size to about 2S0 per experimental group
(birth cohorts September 1982-December 1984). Women were reminded to visit the
antenatal clinic monthly. About 2S% never or only once came to be examined before
28 weeks gestation. Some women were discovered too late in pregnancy. These
women were not included in the intervention trial and are referred to as non-com-
pliers.
Supplements
One daily portion i.e. one package of the high energy (HE) supplement (composition
65
50% fat, 40% glucose, 10% casein) contained 465 kcal and 7.1 g protein. One
package of the low energy (LE) supplement (composition 28% glucose, 50% casein)
contained 52 kcal and 6.2 g protein. Both supplements resembled a coffee creamer
and were served as a hot beverage. To improve acceptance and to avoid sharing,
"jamu" was added. This is a bitter-tasting extract of herbs, traditionally taken during
pregnancy to maintain health and to ensure an easy delivery. Female fie1dworkers,
residents from the villages, delivered one package of the supplement each day at the
respective homes. They prepared the beverage in the quantity the women wanted to
drink, then waited till the women had consumed their daily portion. Left-overs were
returned to the field-station to be weighed.
Anthropometry
The schedule of data collection is presented in Table 1. Pre-pregnant weight was

measured during the monthly home visits with a bathroom scale. When women were
found to be pregnant, they were referred to the project's clinic in each village for an
obstetric interview and examination. Anthropometric measurements were taken
according to standardized techniques [4].
Weight was measured with a eMS scale, height with a microtoise, skinfolds with
a Harpenden caliper and upper arm circumference with a Martin flexible steel
measuring tape. Birthweight was taken within 24 hours after birth with a Salter scale,
to the nearest 50 g.
In the postnatal period, weight (Salter scale) and length (locally made length
board) were taken at weekly intervals in the first month and at 4-weekly intervals at
the age of 1-12 months. Quality control [5] and the checking of equipment was done
at 3-monthly intervals.
Table 1. Schedule of activities and data collection
Type Period Frequency Method
Enrolment of pregnant women B+E Monthly Home visit

Obstetric history and B*+ E Monthly Interview & examination
examination at project's clinic
Anthropometry:
- pre-pregnant weight B*+ E Monthly Home visit
- height (lx, weight,
upper arm circumference,
4 skinfolds B*+ E Monthly Clinic
Home diet B+E 4 x 3-day food weighing
Distribution of supplement E Daily Home visit
Birthweight B+E Within 24 hours Home visit
Socio-economic variables E Ix Clinic or home visit
Infant weight & length E Monthly Home visit
B =baseline period, E =experimental period, * =incomplete coverage

66
Food consumption
Foods consumed by women were measured by weighing during 3 consecutive days

at the pregnancy month 5-6 and monthly thereafter. Snacks eaten between meals
were recorded as well. The amounts of raw foods eaten per day were converted into
energy and nutrients, using the Indonesian and the FAO Food Composition Tables
[6].
Socio-economic status of households
The socio-economic score for households was based on a summation of scores for
the education of the husband and the wife, the type of housing and property. The
maximum score is 79 points [7].
Breast-milk intake
Breast-milk intake was measured in a subs ample of infants born to multiparous

mothers who had consumed the HE or LE supplement. The groups were matched for
sex, age, birth month and location in the village. Twins and infants with a
birth weight of less than 2500 g were excluded.
Sample size was calculated to detect a difference of 100 g in mean breast-milk
intake per 24 hours between the two groups at a significance level of 5% and a
power of 80% with a one-tailed test. The required sample size of 38 was increased to
60 per group to allow for drop-outs or incomplete observations.
Age at enrolment was either 2 weeks (± I week) or 6 weeks (± I week).
Breastmilk intake was measured on 4 successive occasions at 8 week intervals by
test weighing during 48 hours at the homes of the mothers. Trained female field
workers stayed at the mother's home during three days and two nights.
Data analysis
Birthweight. Pregnant women participated on a voluntary basis which resulted in a

wide variation of compliance. Apart from testing the outcome in the HE and LE
groups, the degree of compliance was taken into account by dividing women into
three sub-categories: LE-l or HE-I consumed less than 45 packages, LE-2 or HE-2
consumed 45-89 packages and LE-3 or HE-3 consumed 90 packages or more. The
extra energy consumed by sub-categories HE-I, HE-2 and HE-3 corresponds to less
than 20.925 kcal, 20.925-41.849 kcal and 41.850 kcal or more respectively.
In the final analysis of effect on birthweight a stepwise, multiple regression
Table 2. Characteristics of pregnant women by experimental groups
Characteristic High energy group Low energy group Non-compliers

(N=276) (N=266) (N=20S)
N Mean SD N Mean SD N Mean SD
Age, years 274 25 5 265 25 6 191 27 6

Parity I 51 19% 65 21% 39 29%
2-7 206 76% 192 73% 131 70%
8+ 17 6% 17 6% 20 10%
Height, cm 271 150.8 4.9 265 150.6 5.2 69 149.8 4.8
Pre pregnant
weight (kg) 206 42.0 4.5 198 43.0 5.2 141 43.3 4.7
Pre pregnant WH2 206 18.4 1.9 198 18.9 2.1 69 19.0 1.9
Education score" 274 3.0 2.3 261 2.9 2.2 204 2.6 2.4
Total SES scoreb 268 31 8.5 259 31 7.S 202 28 9.0
" An education score of 3 means Primary School completed.

b Socio-economic score (SES) based on education parents, housing, property: less 30 = poor.
0"\
-.J
68
stari experiment
I
energy (kcal) protein (g)
2000 j -+-- energy-intake (keol)

50
1900 -- protein-intake (g) 48
1800 46
1700 44
1600 42
1500 40
1400 38
1300 36
1200 34
1100 32
Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct Jan Apr Jul Oct
I 81 I 82 I 83 I 84
Figure 1. Energy and protein intake by calendar month over the whole study period in the 3 villages
combined, parity and infant sex combined: Baseline and total experiment groups.
analysis was perfonned per experimental group. The variables under consideration
were the sex of the infant, parity (primi versus multiparae), the mother's height (cm),
pre-pregnant weight (kg), mean energy intake from home diet, supplements (HE or
LE) and the degree of compliance in five sub-categories: less than 45 packages and
then increasing by 10 packages (45-54; 55-64 etc.).
Maternal anthropometry. Similarly, the effect on weight gain in the third trimester of
pregnancy and on four weeks postpartum weight and body mass index (BMI = WH2)
were tested in a step by step multiple regression with as independent variables the
sex of the newborn, the mother's height, pre-pregnant BMI, parity, energy, protein
and fat intake from home diet, vectors representing treatment, compliance and the
interaction between the two and the socio-economic score of the households.
Breast-milk intake. To assess the factors influencing breast-milk intake the independ-
ent variables in the multiple regression analysis were sex, age and birthweight of the
infant, the mode of feeding (exclusive breast feeding or mixed feeding), breast
feeding frequency, the mother's pre-pregnant and 4 week postpartum BMI, and the
mother's time spent on infant care and experimental groups (HE versus LE).
Postnatal growth. As a first step, attained weight and length by age were calculated
over the age period of 0-51 weeks. To test the effect of prenatal energy supplementa-
69
Table 3. Mean daily energy and protein intake from home diets by duration of pregnancy
Gestational age (weeks) Energy (kcal) Protein (g)
Mean SD Mean SD
20-24 1574 452 44.8 12.3

29-32 1615 459 42.4 12.4
33-36 1599 466 41.4 12.1
37-40 1572 409 41.6 11.4
Data from 393 pregnant women with 4 food weighings at monthly intervals.
tion on postnatal growth, attained weight and length on 3 occasions in the first year
were considered i.e. at the age of 16-19 weeks, 24-27 weeks and 48-51 weeks. Per
variable and per age-group a standard Anova was done for non-compliers versus LE
versus HE groups and within the LE group LE-l versus LE-2 versus LE-3, within the
HE group HE-I versus HE-2 versus HE-3. When the result was significant a
DUNCAN multiple range test was subsequently performed.
Results
The characteristics of pregnant women in the experimental period are shown in

Table 2. The differences between the subcategories of the LE- and HE groups and
the non-compliers were not significant.
Home dietary intake
In the study popUlation food availability is determined by the rainy seasons. Erratic
rains will result in poor harvests and less opportunity for salt production. This
situation occurred in the baseline period and is reflected in a very low intake of
energy and protein in 9 of the 13 months. Good seasons coincided with the start of
the intervention in August-September 1982, which continued throughout 1983. The
year 1984 was less favourable (Figure 1). Energy and protein intake from home diets
were not related to gestational age from 20 weeks onwards (Table 3).
Home dietary intake was also not influenced by the type of supplement
(differences between the HE 1-3 and LE 1-3 sub-categories were not significant),
indicating that the supplement was not a substitute for foods habitually consumed
(Table 4).
Effect on birth weight
Details of the effect of the intervention on birthweight have been published [8]. In
the baseline period 468 singleton babies were born alive. Birthweight was recorded
70
Table 4. Mean daily energy and protein intake from home diets by experimental groups and compliance
Experimental Energy (kcal) Protein (g) Fat (g)

groups
N Mean SD Mean SD Mean SD
HE-l 164 1740 409 44.3 9.8 19.64 7.0

HE-2 464 1608 394 42.1 10.7 17.05 7.2
HE-3 321 1591 363 41.7 9.7 16.06 6.7
LE-l 163 1651 389 44.6 11.1 18.53 8.8

LE-2 361 1648 350 43.6 9.8 17.61 7.8
LE-3 392 1542 302 40.4 8.3 16.45 5.6
N = Number of food weighing periods, HE = High energy group, LE = Low energy group.
1 consumed less than 45 packages, 2 consumed 45-89 packages, 3 consumed 90 packages or more.
for 419 babies (87%). The mean birthweight was 2835 g (standard deviation 363 g).
The percentage oflow birthweight (less than 2,500 g) was 12.2% (Table 5).
A total of 753 infants were born in the experimental period. Among the 717
liveborn singleton babies birthweight was measured in 687 (95%). The mean
birthweight of all babies was 2933 g, i.e. about 100 g higher than in the baseline
period, but there was no difference between the LE- and the HE-groups (Table 5). A
sub-category breakdown illustrates that in the LE-group the mean birth weight in sub-
category LE-3 was lower than in LE-l and LE-2, while mean birthweight in
subcategory HE-3 was higher than in HE-l and HE-2 (Table 6).
In the step by step multiple regression analysis the best equation for the LE-group
included only the mother's height (r = 0.26). For the HE-group pre-pregnant weight
and quantity class contributed significantly to birthweight. An increment in mean
birthweight of 54 g per 10,000 kcal can be expected from the following equation:
HE birthweight = 1913 + 22.0* pre-pregnant weight
+ 27.1 * quantity class.
However, energy supplementation only explained a minor amount of the variation in
birthweight (r = 0.13).
The difference in the incidence of low birthweight was not significant viz 12.2%
in the baseline period 11.2% in the HE-group and 9.5% in the LE-group
Effect on maternal anthropometry
Antenatal care was not felt to be a need in the community. Even with active
guidance, women visited the clinic irregularly, when time pennitted. To make full
use of the data available, the means and standard deviations of anthropometric
measurements were calculated from all records, irrespective of the number per
woman.
Table 5. Birth weight distribution of singleton live births in the baseline and experimental period
Birth Baseline Experimental period

weight period
class (kg) Total High energy Low energy Non-compliers
N Cum.% N Cum.% N Cum.% N Cum.% N Cum.%
<2.0 6 1.4 12 1.7 4 1.6 3 1.2 5 2.8

2.0- 45 12.2 53 9.5 25 11.2 21 9.5 7 6.8
2.5- 201 60.4 257 46.9 95 48.1 91 45.6 71 46.9
3.0- 144 95.0 304 91.1 ll8 93.8 109 88.9 77 90.4
3.5-4.1 21 100.0 61 100.0 16 100.0 28 100.0 17 100.0
All classes
N 417 687 258 252 177
Mean (g) 2835 2933 2908 2948 2948
SD (g) 363 407 397 392 440
-....J
72
Table 6. Mean and standard deviation of birth weight by sub-categories of experimental groups
Sub-categories High energy Low energy
N Mean SD N Mean SD
< 45 packages 51 2919 389 50 2962 482

(HE-lor LE-l)
45-89 packages 126 2852 423 98 2973 386
(HE-2 or LE-2)
~ 90 packages 81 2990 346 104 2917 350
(HE-3 or LE-3)
Pre-pregnant weight was 42.7 kg and weight one month before delivery was 49.3
kg (72% of the total sample measured at each point). Weight gain (about 6.6 kg) was
low. Compared to the curve for primiparae in Aberdeen [9], the deviation in the
study population already started at 20-24 weeks of gestation (Figure 2). The pattern
of change in skinfold thickness was quite different from that in well-nourished
women [l0].
It seems that, in general, maternal fat stores are used to subsidize the energy cost
of pregnancy in the last trimester (Figure 3).
Multiple regression equations for the 3 dependent variables and the set of
independent variables including treatment groups and compliance are given in
Table 7.
weight gain (kg)
13
12
*. E.J.P.S,
--B- Aberdeen Women
11
10
9
8
7
6
.. *
5
'*
... *'
"
4 .. "
3
*"
..
*'
'*
2
*'/
.' ~.
"
O~--r-----'-----------'-----'-----'-----'-----'---
16 20 24 28 32 36 40 44
gestational age (weekS)
Figure 2. Mean weight gain in pregnancy of EJPS and Aberdeen women. All measurements included.
73
changes in skinfold thickness (mm)
-* E.JP.S.
-B- Aberdeen
a)
-*--**
10 20 30
gestational age (weeks)
chonges in skinfold thickness (mm)
-* E.J.P.S.
-B- Aberdeen
b)
._*-._. .
10 20 30 40
changes in skinfald thickness (mm)
c)
-B-- Aberdeen
10 20 30
changes in skinfold thickness (mm)

d)
-*. E.J.P.S.
--B- Aberdeen
_... _._-_._._... _._.- . __ .. _._*

10 20 30
gestationa I age (weeks)
Figure 3. Changes in skinfold thickness (mm) in pregnancy in EJPS and Aberdeen women. (a) supra ileac,
(b) subscapular, (c) triceps, (d) biceps.
74
Table 7. Multiple regression equations
Variable 2nd trimester 3rd trimester 4 weeks post £artum

weight gain weight gain weight BMI
Pre-pregnant Wt 0.63c 0.278c

Mother's Ht 4.88 12.48c -19.28c
Pre-pregnant BMI -
Parity -0.13 0.06
S.E.S.
Sex 4.46 -O.82c
Kcal intaked - 0.002" - 0.001"
Protein intaked O.04b O.lOb 0.48b
Fata,d
Compliance -0.61" 0.24" 0.46" 0.20
Experim. group
Exp. group by
compliance
Constant -6.32 3.08 - 2.33 3.55
AdjustedR2 0.066 0.072 0.623 0.537
Multiple
correlation 0.31 0.28 0.79 0.74
SE of prediction 1.89 1.67 l.l 2.49
Fforthe
regression 3.36 10.21 100.13 58.91
Df.= 6.142 2.234 5.295 6.294
p< 0.004 0.001 0.001 0.001
a p < 0.05, b P < 0.01, c p < 0.001, d from home diet.
In weight gain in the third trimester viz. during the intervention, only the infant's
sex and compliance contributed significantly. The sex effect was such that mothers
who produced male babies had higher mean gain rates than those who gave birth to
female babies (228 g and 164 g respectively; P < 0.01). Interestingly, the greatest
gains were observed for the low compliers and least gains for the high compliers.
The best equations for 4 weeks postpartum weight and BMI included pre-
pregnant weight, height, home diet and compliance, rather than energy supplementa-
tion.
Breast-milk intake
The characteristics of mothers and infants, who were enrolled in the substudy on
breast-milk intake are shown in Table 8.
Pre-pregnant weight and BMI, as well as BMI at four weeks post partum, were in
general higher than those of the total population. As mentioned earlier, low
birthweight babies were excluded. Hence, mean birthweight was also higher than
that of the total sample.
75
Table 8. Breast-milk study: Characteristics of mothers and infants and feeding pattern by experimental
groups
Characteristics HE group' LEgroup'
N Mean SD N Mean SD
Mother
Age, years 53 27 5 54 27 7
Height, cm 53 150.7 4.6 53 151.0 5.4
Prepregnant 42 40
- weight, kg 41.7 3.5 42.0 5.3
- WHb 18.5 1.5 18.5 2.0
_ WH b.c 22.7 1.8 22.7 2.5
4 weeks post-partum
- weight, kg 53 42.9 3.1 55 43.9 5.1
- WHb 53 18.9 1.4 53 19.3 2.1
_ WHb,c 53 23.2 1.6 53 23.7 2.5
Consumption during
pregnancy
- energy intake,
kcaVdayb 49 1570 428 49 1617 443
- protein intake,
g/dayb 49 41 11 49 43 13
- supplement,
number of
packagesC 53 84 23 54 76 36
Time spent on
babycare (%) 53 18.9 3.9 54 18.5 4.5
Infant
Sex: male 28 32
female 25 23
Birthweight, g 53 3010 291 55 3056 298
Breast-feeding:
only breast-milk 10 12
breast-milk+food 43 43
Energy intake from
additional food,
KcaVday-range 43 133-200 43 123-194
• Enrolled in High Energy (HE) group N =53, in Low Energy (LE) group N =55.
b Average intake per day of 2-4 observation periods.
C Total intake in 3rd trimester of pregnancy.
The HE- and LE-groups were, however, comparable for the mother and infant
characteristics.
Breast-milk intake ranged from 689 to 744 g per 24 hours in the total group,
values comparable to those of well-nourished populations [11]. No difference could
be demonstrated between the HE- and the LE-groups (Table 9). The supplement was
also not a significant factor in the multiple regression analysis.
76
Table 9. Breast-milk output in gJ24 h by experimental group and age of the infant: Tow sample
Age in days Tow group HE group' LEgroup'
N Mean SO N Mean SO N Mean SO
8- 29 52 694 153 29 704 154 23 683 155

30- 59 49 740 141 21 765 161 28 722 123
60- 89 55 705 120 29 708 128 26 702 114
90-119 44 744 123 20 743 130 24 744 120
120-149 57 705 114 29 727 73 28 682 143
150-179 43 694 123 22 681 75 21 708 160
180-209 58 692 134 31 695 128 27 687 14
210-239 35 689 101 14 686 93 21 691 109
• HE and LE =High or Low energy supplement groups Wilcoxon test (two sided) HE vs LE:p => 0.25
Effect on postnatal growth
The intervention trial was not designed to test effects on postnatal growth but infants
were followed to assess their growth in relation to infant feeding. This follow-up
offered the possibility to compare growth of babies born to mothers in the different
sub-categories in the LE- and HE-groups and the non-compliers.
It should be stressed that there was no intervention for infants, except for the
routine dietary guidance and referral for vaccinations.
Figure 4 shows that the weight curve of infants belonging to mothers who were
not cooperative (non-compliers) was superior to that of infants born to mothers in the
HE-group and this in turn was superior to that of infants born to mothers in the LE-
group (NC > HE > LE at P < 0.05). There was no difference in length curves
between the 3 groups.
To assess growth performance, mean weight (Figure 5) and mean length (Figure
6) of the 3 groups are expressed as a percentage of the median values of the NCHS
reference.
The pattern of the curves for the non-compliers (NC), HE- and LE-groups is
(except of weight in the LE group) strikingly similar for both weight and length,
though at different levels.
Weight in the NC and HE-groups was better than at birth till 20-23 weeks, but
faltering started at 8-11 weeks in the NC group and at 12-15 weeks in the HE-group.
In the LE-group faltering already began in the first month after birth. From 20 weeks
growth progressively lagged behind.
Length, on the other hand, faltered particularly between 4 and 24 weeks - the
deviation slowing down until the age of 36-39 weeks, after which it became stable.
The effect of treatment by compliance on postnatal growth was tested at 3 age
periods, viz. 16-19 weeks, 24-27 weeks and 48-51 weeks.
Although there was a downward trend in weight at the age of 24-27 weeks and
48-51 weeks in LE-l, LE-2 and LE-3 sub-categories, the differences were not
77
length (em)
76
74
72
a) 70
68
66
64
62
60
58
56
54 -8-
.*
N.C.H.S .
52 E.J.P.S.
50
0 4-7 12-15 20-23 28-31 36-39 44-47
age (weeks)
weight (kg)
10
8
b)
7
5 -8- N.C.H.s.
-+- Non compliers
4 -8- High energy group
"* Low energy group
3~~--'---1---'---~---r---r---'---+---'---+---'--~--
o 8-11 16-19 24-27 32-35 40-43 48-51

age (weeks)
Figure 4. Length and weight curves of infants, (a) Length, (b) Weight.
78
weight as % of NCHS-reference
100 --+-- Non compliers
-B- High energy group
-j(-- Low energy group

95
90
85
80
75~~--~--~--~---+---+---+---+---+---4---4--~~--r-
o 4 8 12 16 20 24 28 32 36 40 44 48
oge (weeks)
Figure 5. Mean weight as percentage of NCHS reference by experimental groups.
lenght as % of NCHS-reference
100
-t- Non compliers
-a-- High energy group

99
-* Low energy group
98
97
96
95
94
93 ~,----
4 8 12 16 20 24 28 32 36 40 44 48
age (weeks)
Figure 6. Mean length as percentage of NCHS reference by experimental groups.

79
Table 10. Postnatal growth: Weight (g) by experimental groups and compliance at 3 age periods
Sub- Age (weeks)

cate-
gories 16-19 24-27 48-51
N Mean SD N Mean SD N Mean SD
LE-l 41 5968 948 37 6866 1036 28 7946 1076

LE-2 75 5725 888 73 6666 1060 56 7844 1058
LE-3 82 5951 859 73 6514 836 63 7663 1102
HE-l 41 5868 842 40 6746 956 28 8302 1059

HE-2 96 5917 831 92 6746 808 73 7909 1029
HE-3 65 6278 739 60 7109 945 49 8156 1046
LE-l vs LE-2 vs LE-3: NS at all age periods:

HE-3 > HE-l HE-3 > HE-2 HE-3 >HE-2 HE-l vs HE-2 vs HE-3:NS
p<O.OI p < O.ot > HE-l
p <0.01
Table 11. Postnatal growth: Length (cm) by experimental groups and compliance at 3 age periods
Sub- Age (weeks)

cate-
gories 16-19 24-27 48-51
N Mean SO N Mean SO N Mean SO
LE-l 33 60.4 2.7 31 62.9 3.4 23 70.2 2.9

LE-2 70 59.9 2.7 67 63.3 2.8 51 69.7 2.9
LE-3 75 60.3 3.1 66 62.9 2.6 52 69.8 2.7
HE-l 36 60.9 3.3 32 63.7 3.2 17 69.5 3.3

HE-2 81 60.7 2.7 68 63.3 2.7 59 70.2 3.1
HE-3 60 61.1 3.0 55 64.5 2.5 44 70.5 3.2
Differences LE-l vs LE-2 vs LE-3
HE-l vs HE-2 vs HE-3

} NS at all age periods.
significant. On the other hand, at the age of 16-19 weeks and 24-27 weeks weight in
sub-category HE-3 was significantly higher than in sub-categories HE-I and HE-2
(Table 10). At none of the age periods was there a significant difference in length in
the 6 sub-categories (Table 11).
More important in considering the effect on postnatal growth is the degree to
which prenatal energy supplementation can maintain an adequate nutritional status.
If 95% of the NeHS reference for length and 90% for weight are taken as the cut-off
points, a comparison of the most compliant sub-categories LE-3 and HE-3 underlines
the benefit observed in mean values (Table 12).
80
Table 12. Percentage of children with adequate weight and length in the LE-3 and HE-3 sub-categories
Age, Percentage with Percentage with

weeks WA;::90%NCHS HA;::95%NCHS
LE-3 HE-3 LE-3 HE-3
N % N % N % N %
4-7 76 65.8 58 69.0 75 73.3 57 84.2

8-11 90 60.0 64 73.5 85 76.5 60 80.0
12-15 85 61.1 67 79.1 79 69.6 64 75.0
16--19 82 56.1 65 80.0 75 64.0 60 73.4
20--23 82 48.7 63 69.8 77 58.5 59 64.4
24-27 73 36.1 60 68.3 66 45.5 55 64.5
28-31 72 32.8 59 48.0 67 47.8 45 65.8
32-35 76 23.6 59 44.1 69 42.1 52 57.7
36--39 79 21.6 53 45.3 65 32.3 44 56.8
40--43 75 13.3 56 41.1 67 35.9 47 57.4
44-47 70 14.3 54 36.8 53 26.4 45 55.6
48-51 63 15.8 49 26.1 52 26.9 44 52.3
N =total number of infants per age-group, LE-3 =low energy group, mothers consumed 90 or more
packages per day, HE-3 =high energy group, idem.
The percentage of infants with adequate weight in the LE-3 and HE-3 sub-
categories was comparable at the age of 4-7 weeks. Thereafter, a clear dissociation
between the two sub-categories emerges. At 24-27 weeks about twice as many HE-3
infants had an adequate weight than LE-3 infants. At 48-51 weeks the difference
remained significant, viz. 26.1 % for HE-3 infants and 15.8% for LE-3 infants.
The difference in adequate length between HE-3 and LE-3 infants existed
throughout infancy. Contrary to weight, the difference was larger after 24-27 weeks
viz. at 48-51 weeks 52.3% of HE-3 infants and 26.9% of LE-3 infants had an
adequate length.
Discussion
The East Java Pregnancy Study (EJPS) was conceived in 1980 because the effect of
feeding trials during pregnancy on birth weight was inconsistent.
In the past years it has become increasingly clear that chronic energy deficiency is
difficult to define [12, 13]. Long-term energy balance appears to exist on seemingly
low energy intake [14, 15].
Were EJPS mothers undernourished to such a degree that energy supplementation
during pregnancy will have a positive effect on both the offspring and mother?
Table 13 shows the characteristics of populations from the intervention trials in

Guatemala [16-18], Taiwan [15, 19-21], Colombia (Bogota) [22-24], The Gambia
[14,25-27] and EJPS.
Table 13. Characteristics of populations in 5 intervention trials
Characteristics Guatemala Taiwan Bogota, Columbia Gambia" East Java, EJPS b
Home diet: kcal/day 1558 1800 1586 1350/l700 1600

protein, glday 45 40 35 41/62 42
Height (cm) 149 155 150 157 150
Pre-pregnant weight (kg) 42 5404 54.3c 54 42
Pre-pregnant BM 1 18.9 20.3 24.1c 21.1 18.7
Weight gain 7 7.6 ? 6.8no4 6.6
Mean birth weight (g)
unsupplemented 2980 3068d 2950 2720/3020 2948
% LBW unsupplemented 21 6 11 35/13 11
" Wet and dry season; b Experimental period; C At end of first trimester; d First study infants.
Table 14. Interventions and outcome in 5 intervention trials
Characteristics Guatemala Taiwan Bogota, Colombia Gambia East Java, EJPS
Study period 1969-1977 1967-1973 1976-1978 1980-1982 1981-1985

Target population pregnant, lactating pregnant, lactating households with pregnant women, pregnant women
women women malnourished from 10th week last trimester
children 0-7 years infants 0-12 m. children 0-5 years
Outcome:
Mean birth weight 30 glIO.OOO kcal males Ist-2nd= + 161 g males+95 g wet season only + 224 g 50/10.000 kcal
Weight gain in r=0.241 none yes, if more than yes, in wet season none
pregnancy 13 weeks supplemented
Postnatal growth yes" no yes" yes yes
0-12 months
00
.....
a Infants also supplemented
82
Home dietary intake in energy and protein of EJPS pregnant women was within
the range of the other populations studied, except for Taiwan which in retrospect was
not undernourished. It was lower than the habitual intake of the pregnant women
among whom under-nutrition was improbable [28]. In the latter collaborative study
of energy requirements during pregnancy in Scotland, Holland, Thailand and the
Philippines, habitual diets provided 1750-2140 kcal per day.
Interestingly, there was a wide variation in BMI before pregnancy or in early
pregnancy, suggesting different levels of energy expenditure. Weight gain was low
in all groups, certainly lower than the 8.5-8.9 kg of healthy Thai and Philippino
women from the collaborative study mentioned above. Mean birthweight (excluding
Taiwan) was comparable in the 4 populations. The high percentage of low
birth weight babies in Guatemala suggest that the birthweight distribution was
skewed to the left, indicating that a relatively large deprived group exists.
From these data one may conclude that if the populations in Guatemala, Bogota
and the Gambia are considered undernourished, the EJPS population certainly
belongs to this category and similar positive effects may be expected.
Was the EJPS intervention comparable to that in the other trials?
The duration of supplementation in EJPS was the shortest among the 5 intervention
trials (Table 14). The supplements were offered only in the last 12 weeks of preg-
nancy. In the other studies the intervention started at the end of the first trimester.
Except for the Gambia, supplementation continued after birth for mothers and
children.
How do outcomes in EJPS compare with those o/the other intervention trials?
It seems that the modest effect on birth weight is comparable to that in Guatemala.
The effect may have been larger, if the home diets were poorer (as in the baseline
period or in the Gambia) or if supplementation was targetted to women with a low
pre-pregnant weight (less than 40 kg).
Energy supplementation in the last trimester was not sufficient to improve the
mother's weight or fat stores in this period, neither 4 week postpartum weight and
BMI. It may explain the similar breast-milk intake of infants, born to the LE- and
HE-groups. No finn conclusions can be drawn because in retrospect the study design
should have taken compliance into account. No studies are known which assessed
the effect of energy supplementation during pregnancy on breast-milk output.
Supplementing mothers during lactation in the Gambia did not result in better yields
[29].
The most striking benefit of prenatal energy supplementation in EJPS was during
postnatal growth of infants.
Interpretation of the results should take the degree of compliance into account.
With the data available it was not possible to differentiate household profiles or
mother's characteristics in the groups, non-compliers versus compliers and in the
83
Table 1S. Mean weights (g) of supplemented and unsupplemented infants in Bogota, Colombia compared
to weight of infants in low and high energy sub-categories in East Java.
Age, Bogota East Java

months
Unsuppl. Suppl. Difference LE-3 HE-3 Difference
(1) (2) (2-1) (1) (2) (2-1)
Birth 2956 3024 68 2917 2990 73

3 4979 5176 197 5413 5651 238
6 6204 6580 376 6514 7108 594
9 7178 7550 372 7300 7585 285
12 7944 8366 422 7950 8243 293
LE =low energy supplement, HE =high energy supplement, LE-3/HE-3: mothers consumed 90 packages
or more.
Table 16. Mean length (cm) of supplemented and unsupplemented infants in Bogota, Colombia compared
to length of infants in low and high energy sub-categories in East Java
Age, Bogota East Java

months
Unsuppl. Suppl. Difference LE-3 HE-3 Difference
(1) (2) (2-1) (I) (2) (2-1)
Birth 47.8 47.8 0

3 56.8 57.3 0.5 58.6 59.1 0.5
6 6l.9 62.8 0.9 62.9 64.5 l.6
9 65.3 66.4 1.1 66.9 68.1 l.2
12 68.3 69.7 1.4 70.7 71.1 0.4
LE =low energy supplement, HE =high energy supplement, LE-3/HE-3: mothers consumed 90 packages
or more.
sub-categories LE 1-3 or HE 1-3. However, our observations suggest that the non-
compliers generally belong to the more affluent social and economic stratum of the
popUlation in local terms. The same applies to the LE-1 and HE-I sub-categories
who complied least. There were indications that the LE-3 and HE-3 sub-categories
were the most deprived: their home diets were the poorest; in sub-category LE-3
mean birthweight and third trimester weight gain [7] were the lowest.
Hence, one would expect better growth among infants born to non-compliant
mothers compared to those born to LE- or HE-mothers, as confirmed by the results.
If differences in growth can be attributed to the intervention during pregnancy, the
largest differences should be found between the LE-3 and HE-3 sub-categories,
which indeed were illustrated by the results.
A comparison with data on postnatal growth from Bogota [24J, the Gambia [30]
and Guatemala [31J shows that the ecological environment determines the outcome.
With comparable mean birth weight, the effect of prenatal supplementation on
infant weight in EJPS was similar to that in the Gambia. The EJPS results were more
84
% of infants occording to NCHS- reference

85
80 * --*- length/age EJPS
75 -[3-- weight/age EJPS
70
0
* length/age Bogota
65 -B- weight/age Bagoto
60
55
50 * -*
45
40
35
30
25
20
15
0 3 6 9 12
age (monthS)
Figure 7. Percentage of infants with good weight-for-age (WA) and length-for age (LA) in Bogota,
Colombia and EJPS. WA ~ 90% NCHS reference, LA ~ 95% NCHS reference.
impressive than in Guatemala and Bogota, where mothers and infants were supple-
mented. As an example, weight and length of the supplemented and unsupplemented
groups in Bogota are compared with those of the HE-3 and LE-3 subcategories of
EJPS in Tables 15-16.
The difference in weight at birth was comparable for the 2 locations. At 3 and 6
months the difference between HE-3 and LE-3 sub-categories in EJPS was greater
than between the supplemented and unsupp1emented infants in Bogota. At 9 and 12
months the differences were in the opposite direction, suggesting that supplementa-
tion of infants had a benefit in the weaning period. In EJPS mothers object to length
measurement at birth. At 3-12 months of age differences between HE-3 and LE-3
were either equal or larger than between the 2 groups in Bogota, except at the age of
12 months.
The superior growth of EJPS infants as compared to Bogota infants is illustrated
in (Figure 7). The percentage of infants with good weight is 15-30% higher in EJPS
infants at the age of 0-9 months. The corresponding figures for good length are
25-40% at the age of 0-12 months.
Prenatal supplementation may thus be a better input to prevent malnutrition than
supplementing the infants directly, at least in our setting.
85
Conclusions
1. The East Java Pregnancy Study suggests that for pregnant women a threshold
in home dietary intake of 1400-1500 kcal and 40 g of protein per day exists,
below which fetal growth is impaired. Such low intakes occur in the semi-arid
zone of Madura once in 3-4 years, but it may be more frequent in other rain
dependent zones of Indonesia.
2. In the relatively better seasons, reproductive performance in terms of
birthweight and breast-milk yield was remarkably good. Such a performance
was, however, achieved at the expense of the mother's nutritional status. The
average pre-pregnant weight of 42 kg at a mean height of 150 cm is far less
than the desirable weight of 47 kg mentioned by the FAO/WHO/UNU Expert
Groups [32].
3. Based on the literature and our findings it is reasonable to assume that prenatal
energy supplementation is beneficial for fetal and infant growth.
4. The duration of supplementation, limited to the last 12 weeks of pregnancy
was, however, not sufficient to improve the mothers' nutritional status.
Policy implications
In the 1989-94 Five Year Plan the reduction ofIMR to 50 per 1000 livebirths and of
the incidence of LBW to 7% are formulated as goals by the Ministry of Health of
Indonesia. To reach these targets equal emphasis should be put on the nutritional
health of mothers and children. What can be done? A four dimension approach has
been suggested for the care of Under-Fives: epidemiological, technical, organiza-
tional and political dimensions [33]. Applied to maternal and child nutrition, the
following can be considered:
Epidemiologic dimension
Indonesia is a large country, composed of different ecological zones and cultural

entities. There is a wide variation by province and within provinces in socio-
economic status, food availability, IMR and prevalence of under-nutrition. No
community based data of birthweight distribution are available apart from the few
research projects. Proper allocation of limited funds needs a sound information
system. One or two measurements (women's weight and height) may be included in
the existing Household Budget Surveys. Weight of mothers should be measured
during the growth monitoring sessions for pre-school children to select the group for
special attention, viz. mothers with a non-pregnant weight of 40 kg or less. Univer-
sities can be involved in a more in-depth situational analysis in their regions.
86
Technical dimension
No specific technologies exist at present in Indonesia to improve maternal nutrition.
Two approaches are potentially feasible and affordable:
• Food rations or a food pricing policy during lean seasons in risk communities.
• Energy-enrichment of "jamu". "Jamu" is the local name for a traditional medicine
or roborant, which consists of a mixture of herbs.
Specific "jamu's" are being sold for a wide range of conditions, from maintenance of
health during pregnancy, stimulation of breastmilk production, treatment of illnesses
to stimulation of libido.
In our research project, we have used a fat-enriched "jamu" which had a shelf-life
of one year at a cost of US $ 0,25 per daily portion of 500 kcal in 1981.
Organizational dimension
The food security approach should be part of the existing nutrition surveillance or
early warning system. These systems should be extended to areas which are periodi-
cally exposed to low levels of food availability.
Targetting food to women at risk is in practice almost impossible. In our com-
munity it would be socially unacceptable, while it is a common belief that one should
reduce eating during pregnancy to prevent difficult labor.
If supplementary feeding is considered justified as an additional, short-term
intervention, the fat-enriched "jamu" is potentially acceptable and cost-beneficial.
Sharing will never occur, if supplied as "jamu" for pregnant women. The replace-
ment effect will be negligible, since it is not considered a food. The existing
distribution channel of the "jamu" industry reaches even the most remote villages in
Java, while it can be distributed via the Integrated Health Posts (POSYANDU)
throughout Indonesia.
It is a matter of policy to involve local industry in nutrition interventions. The
choice has already been made to enrich salt with iodine and Ajinomoto
(monosodium glutamate) with vitamin A. Subsidies can be given for areas where the
increase in price of the enriched "jamu" may be prohibitive for the community.
Political dimension
Nutrition has always received much attention in Indonesia through the Family
Nutrition Improvement Programme (UPGK) and recently the Integrated Health Post
(POSYANDU). The focus was, however, almost exclusively on the child. Within the
framework of "Safe Motherhood", one may expect a balance of approaches towards
the intricate mother-infant relationship and the quality of life. The indications are
favourable, as illustrated by the national Mother's Welfare Symposium in June 1988,
followed by the national workshop in August 1988.
87
Acknowledgement
We wish to thank Mr. Schofield for the analysis of the maternal anthropometry data.
The study was financed by the Netherlands Ministry for Development Cooperation.
Responsibility for the contents and for the opinions expressed rests solely with the
authors.
References
1. Kusin JA, Sri Kardjati and de With C: Trop Geogr Med 31:571, 1979.
2. Kardjati Sri, Kusin JA and de With C: NutrRep Intern 28:1341,1983.
3. Kardjati Sri, Kusin JA, de With C and Sudibia: Trop Geogr Med 30:359,1978.
4. Werner JS and Lourie JA: Blackwell Scientific Publications, Oxford, 1969, p. 27.
5. Habicht JP: Pan American Health Organization, Washington DC, 1973.
6. Food and Agriculture Organization FAO, Rome, 1972.
7. Kardjati Sri: Ph.D. thesis, Airlangga University, Surabaya, 1985, p. 61.
8. Kardjati Sri, Kusin JA and de With C: Br J Obstet Gynaec 95:783, 1988.
9. Hytten FE and Leitch I: In: The physiology of human pregnancy, Blackwell Scientific Publication,
London, 1971,p.281.
10. Taggart NR, Holliday RM, Billewics WZ, Hytten FE and Thomson AM: Br J Nutr21:439, 1967.
11. World Health Organization, The quantity and quality of breastmilk, WHO, Geneva, 1985.
12. Margen S: In: Pollitt E and Amante P (eds) Energy intake and activity, Alan R Liss Inc, New York,
1984, p. 57.
13. James WPT: In: Schiirch B and Scrimshaw NS (eds) Chronic energy deficiency: consequences and
related issues, Nestle Foundation, Lausanne, 1987, p. 7.
14. Prentice AM: In: Pollitt E and Amande P (eds) Energy intake and activity, Alan R Liss Inc, New
York, 1984, p. 3.
15. Adair LS: Ibid, 1984 p. 33.
16. Habicht JP, Yarbrough C, Lechtig A and Klein RE: In: Winick M (ed) Nutrition and fetal
development, Wiley Interscience Publication, New York, 1974, p. 127.
17. Lechtig A, Habicht JP, Delgado H, Klein RE, Yarbrough C and Martorell R: Pediat 56:508,1975.
18. Lechtig A, Martorell R, Delgado H, Yarbrough C and Klein RE: J Trop Pediatr 24:217, 1978.
19. Blackwell RQ, Chow BF, Chinn KS, Blackwell BM and Hsu SC: Nutr Rep Intern 7:517,1973.
20. Adair LS, Pollitt E, and Mueller WH: Br J Nutr 51:357,1984.
21. Adair LS and Pollitt E, Am J Clin Nutr 41 :94, 1985.
22. Mora JO, de Paredes B, Wagner M, de Navarro L, Suescun J, Christiansen N and Herrera MG: Am J
Clin Ntr 32:455, 1979.
23. Mora JO: In: Underwood BA (ed) Nutrition intervention. Strategies in naturial development.
Academic Press, New York, 1980, p. 79.
24. Mora JO, Herrera MG, Suescun J, de Navarro L and Wagner M: Am J Clin Nutr 34:1885,1981.
25. Prentice AM, Whitehead RG, Watkinson M, Lamb WH and Cole TJ: Lancet 1:489, 1983.
26. Prentice AM, Cole TJ, Foord FA, Lamb WH and Whitehead RG: Am J Clin Nutr 46:912, 1979.
27. Lawrence M, Coward WA, Lawrence F, Cole TJ and Whitehead RG: Am J Clin Nutr45:1442, 1987.
28. Durnin JVGA: In: Schiirch Band Schrimshaw NS (eds) Chronic energy reficiency: consequences
and related issues, Nestle Foundation, Lausanne, 1987, p. 135.
29. Prentice AM, Roberts SB, Prentice A et al.: Hum Nutr Clin Nutr 37 C:53, 1983.
30. Prentice AM, Cole TJ, Lamb WH and Whitehead RG, Proceedings Fourth Asian Congress of
Nutrition, Bangkok, 1983, p. 302.
31. Delgado HL, Valverde VE, Martorell R and Klein RE: Early Hum Devt 6:273, 1982.
32. Energy and protein requirements, WHO Technical Report Series, Geneva, 1985, p. 183.
33. Samhari Baswedan H: Prisma, Jakarta 40:3, 1986.
Weaning foods in infant nutrition
SOLIHIN H. PUDJIADI
Introduction
The word 'weaning' is used in different ways: it is used to indicate that breast-
feeding has totally stopped; it is used when breast-feeding is being gradually
withdrawn or when other foods are being introduced without withdrawing breast-
feeding.
What we in Indonesia mean with weaning, is the introduction of any kind of food
other than the mother's milk or starting formula. We emphasize that breast-feeding
should be continued if possible until 2 years of age.
Human milk is without doubt the best food for infants, as nature intended it. There
are many reasons that support this statement: the composition of the nutrients in
human milk is such that it provides all the basic needs for the growing infant.
Furthermore, human milk has anti-infective properties, which are both humoral and
cellular; protective factors in human milk are: immunoglobulins, lysozyme, bifidus
factor and lactoferrins that limit the availability of iron for intestinal bacteria,
especially E.coli. Other protective factors in human milk are macrophages, lym-
phocytes, neutrophiles and epithelial cells (motile), amoeboid and phagocytic [1-6].
Alongside this, breast-feeding is considered to lengthen the period of the mother's
infertility, thereby ensuring better child-spacing.
Furthermore, breast-feeding is much cheaper than any other infant formula, and
spares the family budget of the low income group.
There are circumstances however, in which babies do not get human milk at all;
e.g.:
the mother dies soon after delivery,
the mother has to be hospitalized for a long period,
the mother is not able to produce milk,
- there is contra-indication for the mother to feed,
there is contra-indication for the baby to obtain human milk (e.g. congenital
lactase deficiency).
90
In such cases the baby should be fed with a formula of which the composition is
close to that of mother's milk.
After a certain age, human milk is no longer sufficient to cover the needs of the
rapid growing body. Also the amount of milk that the mother produces may
decrease. In these circumstances the diet should be supplemented with other food.
The amount of milk produced by the mother depends on many factors, such as the
nutritional status of the mother, the extra food given to the lactating mother or
psychologic stress.
When should supplementary feeding start?
It is impossible to give general advice for the age at which weaning should be
started. It not only depends on the quantity of milk produced, but also on the
individual needs of the baby, and that varies very much in basic requirements [7].
Factors that affect individual energy expenditure are age, surface area (related to
height and weight), sex factor, temperature of the environment and the activity of the
child itself. The younger the baby, the more active is its metabolism and the more
energy is required.
One should realize that weaning is considered to have negative effects on the
amount of milk that will be produced by the mother. The number and duration of
suckling periods will fall, plasma prolactin levels will decrease and menstruation and
fertility will return earlier.
Next to this, introducing weaning food already in the first few months of life can
have more negative side-effects. The proper time for the introduction of solid food
into the diet should be based on the following factors:
1. Nutritional considerations
2. Educational considerations
3. Psychological considerations
4. Socio-economical considerations
It is generally accepted that solid foods should not be commenced until the age of
4-6 months [8-10]. Several hazards created by the early introduction of solid food
are: high solute load, which means a high level of salt, introducing hyperosmolality;
obesity, caused by a high caloric intake; food allergy; hazards introduced by harmful
natural compounds such as gluten, nitrates, sucrose and salt or by food additives and
contaminants.
In giving recommendations on weaning, in Indonesia we have to envisage two
situations; one in which parents belong to the high and middle socio-economic class,
and one in which the parents belong to the low socio-economic class:
91
Introduction of weaning food to middle and high socio-economic class
Babies are born to mothers who know how to feed themselves during pregnancy and
can afford to buy supplementary food and regularly see the obstetrician. Normally,
the babies are born mature with a normal weight, which is higher than of children
born socio-economic class from mothers in the low-income groups. Milk production
is normally adequate, the nutritional status is good and some extra food is used by
the mother in the lactation period. Because of the good hygienic conditions at home
and the regular check-ups with the necessary immunization, such infants are rarely
sick and can in general grow well in the first 4-6 months. For those infants solid
food should not be given before the age of 4 months. If there are signs of overweight,
the introduction of solid food should be postponed until the infant is 6-7 months old.
What kind of solid food should be given?
The mothers of this class of infants are in general busy; either actively involved in
business, employed, or indulging in many social activities. For those mothers, 'ready
for use' food is a welcome relief.
Obtainable at the market are the so-called pre-cooked milk cereals that can be
given to the infant by adding water only. Instant cereal, which must be mixed with
milk and sugar to get the right amount of protein and energy can also be bought. For
those who prefer the traditional steamed rice for their babies but do not have the time
to prepare it, ready-for-use baby chicken rice is available at the supermarkets.
How should solidfood be given?
Solid food must be gradually introduced. Starting from the age of 4-5 months to 6-7
months, every month one serving of solid food should replace one serving of breast
milk or infant formula, so that at 7 months of age the baby gets his given solid food
three times daily and twice breast-fed or given follow-up formula with additional
fruit or fruit juice. This menu can be maintained until the age of 12 months. Starting
from 10 months of age, ground rice can be replaced by soft cooked rice.
Introduction of weaning food in the low socio-economic class
In Indonesia many mothers in this class start giving complementary food very early:
sometimes in the first month chewed rice, blended rice-banana or honey is given.
Unsanitary conditions under which the supplementary food is prepared, stored or fed
will increase the chance of gastroenteritis causing impaired growth [11]. Another
negative aspect is the previously mentioned influence of weaning on the amount of
92
Table I. Recipes for the preparation of weaning foods (Nutrition Subdivision of the Child Health
Department)
Milk cereal Soft boiled rice
Ingredients Flour" 15 g Rice 25g

Milk powder 25g Meat 20g
Sugar 109 Tempe 15g
Vegetable 40g
Water 200ml Waterb 300-400ml
Cooking time 5 min. 15-30 min.
a Rice, corn, oats. b Depending how soft the baby likes.
milk produced by the mother, and its relation with early returning fertility [12]. For
infants in this low socio-economic class, as a general rule, weaning should also not
start before the age of 4 months. There are cases in which weaning food can be
introduced earlier; e.g. when breast-milk is not sufficiently produced, the infant
formula is too expensive for the parents, while growth already starts to decline. We
should make every effort to reduce the number of these cases by strongly emphasiz-
ing the value of breast-feeding and give mothers some practical advice.
Mothers should be given education about how to prepare quality solid food in
hygienic conditions [13]. The ingredients of these self-prepared weaning foods
should be available at the local market for a relatively low price, without causing a
heavy burden on their budget. Table 1 shows some recipes for the preparation of
different kinds of solid food, practised by the Nutrition Subdivision of our Child
Health Department.
For this under-privileged group tin-packed pre-cooked solid food is beyond their
reach. Adequate alternatives, replacing expensive cow's milk with a mixture of plant
protein, such as soyabean and mungbean, are available. This food is packed in small
sachets for one serving, well-adapted to unsanitary conditions.
Suitability of rice-plant protein formulas
Before being released, a low-cost rice-plant protein formula (Promina) was inten-
sively studied with healthy infants and malnourished children in the ward of the
Child Health Department of the University of Indonesia as well as in the Outpatient
Department. These infants were compared with those who received milk-cereal
formula as to the acceptability, tolerance, growth velocity and blood changes. The
following results were obtained:
Hospitalized patients:
Thirty patients, recovering from different diseases were given the two kinds of
formula in random order:
Acceptability (based on intake of more than 50% of each serving):
93
- rice-plant protein porridge: 73%

- milk-cereal porridge: 83%
Older infants and those used to having milk-cereal at home preferred milk-cereal
porridge more. No significant difference in tolerance was assessed on comparing the
consumption of those two kinds of porridge.
Malnourished patients:
The two formulas were given in random order to thirty PEM patients:
Acceptablity:
- rice-plant protein: 67%
- milk-cereal: 83%
Tolerancy: vomiting occurred in 5 patients during plant protein serving; none during
milk-cereal porridge serving.
Outpatient department:
Fifty-five infants (mean age 3.3 months) on rice-plant protein formula versus 50
infants (mean age 3.6 months) on milk-cereal formula:
Acceptability: in general more infants preferred milk-cereal porridge, but the

difference was not significant.
Side-effects: abdominal distension, diarrhea, vomiting or allergic manifestations
were not found in both groups.
Growth rate: no difference in weight and height increase was observed.
Blood-picture (Hb, Ht, MeHC): no significant changes were seen during the trial
in the rice-plant protein nor in the milk-protein group.
It can be concluded that rice-plant protein porridge can safely be given to healthy as
well as to sick infants. These products are of special interest for parents from the low
socio-economic class, because the price is lower than the milk-cereal formula. A
comparison is made in Table 2.
The rice-plant protein formula can also be consumed by patients suffering from
cows'-milk protein allergy or lactose deficiency.
Recommendations
Human milk is without doubt the best food for infants. Every effort should be made
to stimulate breast-feeding during the whole period of infancy. When human milk is
not available, an appropriate infant formula can be given as a substitute.
Fruit juice can be given from the age of 3 months, but solid food not earlier than 4
months. Parents from the middle and high socio-economic class should be aware of
signs of overweight in their infants. If there are signs of overweight, introducing
supplementary food should be postponed until 6-7 months.
94
Table 2. Rice/plant protein formulas versus milk cereal formulas
Composition per 100 g Rice/plant protein formula Milk cereal formula
Protein (g) 15 15
Fat (g) 5-18 9
Carbohydrate (g) 68-74 72
Minerals (g) 2-3 3.6
Energy (kcal) 400-500 417
Recommended serving size (g) 25-50 40
Price per 300 g (Rp) 1200-2000 2400-4000
Unfortunately, it happens that less privileged infants for a number of reasons

cannot sufficiently be fed with breast-milk. Since infant-fonnulas are out of
economic reach, we have to face the situation that weaning foods are introduced
earlier, certainly when there are signs that growing faltering starts.
Mothers should be instructed to cook quality baby food from ingredients available
at the local market, without disturbing their budget. Supplementary food that is
prepared, stored and fed under unsanitary conditions will lead to increased morbidity
due to infection.
A low price cereal-plant protein fonnula that has been tried on infants, recovering
patients and PEM children was found to be adequate for use.
References
I. Bullen CL, Teerle PV and Willis AT: 1 Med Microbiol9:325, 1976.
2. leliffe DB and leliffe EFP: Human milk in the modem world; psychosocial, nutritional and
economic significance, Oxford University Press, London, 1978.
3. Fomon Sl and Filer U: In: Fomon Sl (ed) Infant nutrition, Saunders, Philadelphia, 1974, p. 359.
4. Donald EG and Lebenthal E: In: Lebenthal E (ed) Textbook of gastroenterology and nutrition in
infancy. Vol. I: gastrointestinal development and perinatal nutrition, Raven Press, New York, 1981,
p.295.
5. Garza C, SchanIerRJ, Butte NF and Motil KJ: Clinics in Perinatology 1:11, 1987.
6. lensen RG, Ferris AM, Lammikeefe CJ and Henderson RA, Nutrition Today 6:20, 1988.
7. Poskitt EME: In: Ballabriga A and Rey 1 (eds) Weaning: Why, what and when? Raven Press, New
York, 1987, p. 45.
8. American Academy of Pediatrics, Committee on Nutrition, Pediatrics 6: 1178, 1980.
9. ESPGAN Committee on Nutrition: Acta Paediatr Scand, 1981, Supp1287.
10. World Health Organization: Weaning, from breast milk to family food, WHO, Geneva, 1988.
11. Vis HL and Hennart P: In: Ballabriga A and Rey 1 (eds) Weaning: Why, what and when? Raven
Press, New York, 1987, p. 169.
12. Ramachandran P: In: Ballabriga A and Rey 1 (eds) Weaning: Why, what and when? Raven Press,
New York, 1987, p. 187.
13. Rowland MGM: Acta Paediatr Scand, Suppl 323, 1986, p. 33.
Ethnic differences in breast-feeding among
well-to-do Singaporeans
J. J. COUNSILMAN
Introduction
In Singapore, the present rate for initiating breast-feeding with or without supple-
ments is 60%, and the rate at 3 months is less than 20% [1-3]. These estimates are
much lower than those obtained by the first studies of infant feeding during the early
1950s, that is, over 90% for initiation and over 60% at 3 months [4-7], though not as
low as those recorded in 1971, that is, 40% for initiation and 5% at 3 months [8].
Within this pattern of decline and partial resurgence, Singapore's 2 basic economic
classes (viz. upper and lower income families) and its 3 major ethnic communities
(viz. Chinese, Malays and Indians) have shown important secondary patterns.
During the early 1950s, upper and lower income families (for all 3 ethnic groups
combined) differed little for the initiation of breast-feeding but markedly for its
continued use. Over 95% of poor mothers and 90% of well-to-do mothers started
feeding by breast; but by 3 months the rate for the upper class had dropped by nearly
half while that for the lower class had dropped by only a quarter [6,7]. By 1971 the 2
classes differed substantially for both aspects, with 51 % of poor mothers and 28% of
well-to-do mothers initiating this method, and 43% of the lower class and 10% of the
upper class continuing its use for 1 month [8]. Ten years later only the rate of
initiation differed very much, but this time with well-to-do mothers leading the way.
A large sample in 1980 revealed that 71 % of wealthier mothers breast-fed for 1 week
and 21 % continued for 1 month, while 49% of poorer mothers breast-fed for 1 week
and 27% continued for 1 month [9, 10]. The same relative differences between the 2
classes probably exist today [1-3].
Data on ethnic variations are fewer and more variable [11]. Nevertheless, they
indicate that all 3 communities contributed to the initial decline in breast-feeding in
the 1950s and 1960s and to its subsequent resurgence in the 1970s and 1980s, and to
differences between upper and lower income families. They specifically indicate that
Chinese mothers led the trend away from breast-feeding and have lagged behind in
its revival. For both economic classes combined, percentages of mothers initiating
breast-feeding at various times since 1951 illustrate this point. During the 1950s,
H.K.A. Visser and J.O. Bindels (eds.) Child Nutrition in South East Asia. 95-106.
© 1990 Kluwer Academic Publishers. Printed in the Netherlands.
96
about 90% of Chinese and 95% of Malays and Indians started feeding their babies by
breast [6,7]. By 1960, recorded values were 65% for Chinese, 63% for Malays
(which was probably an underestimate) and 82% for Indians [12,13]. Twenty years
later, they were 41 %, 90% and 63% for Chinese, Malays and Indians, respectively
[9]. And, most recently, they were 58%, 98% and 94% in 1986 [3].
Similar patterns have been observed in Malaysia for the same ethnic groups over
the period of 1950 to 1977 [14].
Reasons for the overall decline have been identified as the increasing affluence
and 'Westernization' of Singapore as a whole [11, 15]. These factors appear to be
largely synonymous with those detennined by the World Health Organization as
operating in developing countries worldwide [16]. They are (i) the gradual erosion of
traditional lifestyles as a result of modernization breaking down the supportive
family structure and (ii) the imitation of wealthy subgroups by poor subgroups,
including the initial imitation of the Western elite by the non-Western elite. Regard-
less of which condition, or whether both, prevailed among well-to-do Singaporeans
in the early 1950s, the low income of poor mothers then (and before) probably
directly prevented them from bottle-feeding. For example, the high incidence of
malnutrition among low income mothers during this period [4, 5] suggests that they
could not afford substitutes for breast milk. During the 1960s, milk substitutes (such
as condensed milk and milk powder) undoubtedly became increasingly available,
and affordable by nearly all Singaporean families. Some influence other than income
[15], such as modem working conditions or imitation of wealthy mothers, or both
[12], must have then worked against breast-feeding within the lower as well as upper
economic class. The renewed interest in this method during the 1970s and 1980s
supports the imitation hypothesis since working conditions favourable to breast-
feeding have apparently not increased substantially over the past 15 years [3]. On the
other hand, differences in education between the economic classes may have
replaced income as the primary factor detennining differences in feeding behaviour.
In Singapore [2,10], as in a number of Western countries (e.g. [17,18]), the propor-
tion of breast-feeders increases with increasing education.
At least among poor mothers, ethnic differences in infant feeding are apparently
not related to physical characters, education or occupation [1]. Counsilman and
Viegas [11] believe that for both economic classes they are related to differences in
the salutary nature of a number of modem and traditional beliefs and practices. In
general Malay and Indian families inherit traditional practices that promote breast-
feeding, or practices that at least help them resist the modem trend towards bottle-
feeding; while Chinese families display an unfortunate combination of modem and
traditional practices that promote bottle-feeding. For example, assistance with the
care of the young helps Malay and Indian mothers sustain breast-feeding, whereas it
has the opposite effect on Chinese mothers, that is, along with bottle-feeding it gives
them greater freedom from the home [10].
My previous reports concentrated on differences between breast- and bottle-
feeders within the lower [1] and upper [2] economic groups separately, and only
97
briefly considered ethnic feeding patterns and their causes. This study examines
ethnic variations and their origins in much greater detail, though only for well-to-do
families. To identify differences among the communities, subjects from each ethnic
class are characterized for a small number of features of breast- and bottle-feeding.
To identify the causes of these differences, the ethnic classes are then compared for a
large number of features of the mother, her environment and behaviour and her
pregnancies, births and children.
Methods
Data were obtained from 2 interviews with women that gave birth to single live
babies at the Kandang Kerbau Hospital during the period of 10 March to 29 April
1985. The first interview was conducted in the hospital with 242 postpartum women;
and the second was conducted by phone 6 to 9 weeks later with 197 of the same
subjects. Only subjects whose total family income exceeded the national average
were included. Their ethnic make-up (73% Chinese, 17% Malays and 10% Indians)
approximately matched that of Singapore as a whole [19].
The primary 'study' variable, ethnic class, was compared with 44 'descriptive'
variables. All comparisons were analyzed for frequencies of subjects from each
ethnic class falling within various categories of each descriptive variable. The
number and nature of categories belonging to a descriptive variable depended on (i)
the answers provided and (ii) the distribution of frequencies (because some combina-
tions of answers or too many caused invalid tests of significance). Most comparisons
were analyzed with chi-square. More powerful statistical techniques were not
appropriate for such a mixture, as I had, of categorical, ordinal and ratio variables
[20].
Table 1. Statistical comparisons of ethnic class with 10 features of infant feeding
Variable X2 df N p<
1. Initial feeding method 21.6 4 242 0.0001

2. Period of feeding solely by breast" 32.0 6 197 0.0001
3. Total period of breast-feeding" 36.5 6 197 0.0001
4. Period of breast-feeding previous child 15.4 4 108 0.01
5. Reasons for breast-feeding 11.0 4 176 0.05
6. Reasons for weaning" 3.8 3 94 ns
7. Use of drugs to suppress lactation" 12.8 2 156 0.01
8. Prior knowledge of breast-feeding 4.7 4 132b ns
9. Frequency of breast-feeds a day" 1.9 4 131 ns
10. Breast-feeding success 3.4 4 132 ns
a Data from follow-up interview; all other data from hospital interview.
b Primiparas only.
98
Results
Ethnic differences
Table 1 summarizes the statistical comparisons of ethnic class with 10 features of

infant feeding. The ethnic classes differed highly significantly and in the same
manner for all 4 measures of the initiation and duration of breast-feeding. For the
initial choice of a feeding method, proportions of subjects choosing breast-feeding
were 57% for Chinese, 90% for Indians and 94% for Malays. At the time of the
second interview, proportions of all subjects that were feeding their babies by breast
alone (Le. without supplements) were 8% for Chinese, 20% for Indians and 28% for
Malays. When feeding with supplements was included, the proportions rose to 13%,
20% and 44%, respectively. And, lastly, for mUltiparous mothers that breast-fed their
most recent previous child for at least 3 days the average duration of suckling that
child (with or without supplements) was 13.7 days (N = 82) for Chinese, 39.4 days
(N = 9) for Indians and 63.3 days (N = 17) for Malays. There was therefore a clear
order of preference among the communities, with breast-feeding favoured most by
Malays and least by Chinese.
Since nearly all bottle-feeders (96%) were Chinese, reasons for bottle-feeding
apply chiefly to this class (and thus were not analyzed statistically). The types of
answers were numerous but could be grouped into medical problems with the baby
(5%), returning to work soon (6%), personal motives (e.g. embarrassed by breast-
feeding) (28%), medical problems with mother (29%) and convenience (32%). In
contrast, both variables, 'reasons for breast-feeding' and 'reasons for weaning',
could be examined statistically, though only the former was significant. A majority
of women in all 3 ethnic classes gave health of baby as their primary reason for
feeding by breast; but disproportionately more Chinse gave this reason, while more
Indians gave bonding with baby and more Malays gave tradition. For weaning, 'no
milk' (38%) was the most common reply, followed by returning to work soon (19%),
inconvenience (16%), medical problems with mother (14%), baby refused breast
(8%) and medical problems with baby (5%).
Although most subjects (73%) stopped lactating naturally, there were still
significant differences among the classes for the use of drugs to suppress lactation. In
particular, of the 42 subjects that used drugs, 41 were Chinese (of which 26 were
bottle-feeders and 15 breast-feeders); 1 was a Malay (a breast-feeder). The most
commonly used modem drug was bromocryptine, though 2 Chinese took stiloestrol.
Two other Chinese consumed an unidentified oral powder obtained from Chinese
herbalists; and the Malay applied 'Malay topical herbs' to her breasts.
Influences on ethnic differences
Table 2 gives the statistical comparisons of ethnic class with 32 physical, health,
environmental and behavioural variables.
99
Table 2. Statistical comparisons of ethnic class with selected physical, health, environmental and
behaviour variables.
Variable X2 df N p<
Physical characters
I. Age 18.5 8 240 0.05
2. Height 13.0 8 234 ns
3. Pre-pregnancy weight 14.9 8 236 0.10
4. Length of pregnancy 0.9 6 242 ns
5. Type of delivery 1.2 4 242 ns
6. Parity 0.9 2 242 ns
7. Weight of baby 11.9 8 237 ns
8. Sex of baby 0.9 2 242 ns
Health measures
1. Physical health during pregnancy 3.4 4 242 ns
2. Emotional problems during pregnancy 4.2 4 242 ns
3. Past reproductive failures 3.2 2 242 ns
4. Mother's physical health after birth" 0.8 2 197 ns
5. Baby's health after birth" 0.5 2 197 ns
Environmental conditions
I. Number of adults in home 5.4 4 242 ns
2. Number of children in home l.l 2 242 ns
3. Lifestyle 1.9 2 242 ns
4. Type of occupation 12.4 8 241 ns
5. Family income 20.4 6 242 0.01
6. Education 7.4 4 242 ns
7. Weeks before returning to work" 10.3 6 197 ns
8. Husband's attitude towards breast-feeding 17.6 4 242 0.01
9. Mother-in-Iaw's attitude towards breast-feeding 32.0 6 219 0.0001
10. Feeding method of subject's mother 15.6 4 242 0.01
II. Influences on subject's feeding choice 4.7 4 242 ns
12. Husband's smoking 4.2 4 241 ns
13. Help with care of baby' 9.8 2 196 0.01
14. Behaviour of baby' 0.1 2 196 ns
Behavioural attributes
I. Timing of pregnancy 0.6 2 242 ns
2. Preparation for birth 2.8 2 132 b ns
3. Time of first contact with baby after birth 7.3 6 229 ns
4. Foods avoided during pregnancy 13.4 4 239 0.01
5. Foods avoided after birth 7.1 2 239 0.05
• Data from follow-up interview; all other data from hospital interview.
b Primiparas only.
Physical characters. Only age and pre-pregnancy weight varied significantly or

nearly so among the ethnic classes. On average, Chinese were 1.8 years older than
either Malays or Indians, and were 1.5 kg lighter than Malays and 4.6 kg lighter than
Indians.
100
Health measures. None of these variables differed significantly among the com-
munities.
Environmental conditions. Except for total family income, the 3 groups did not differ
for any of the more pervasive environmental conditions, namely, number of adults in
the home, number of children in the home, perceived lifestyle, type of occupation,
education and weeks before retuming to work (including not working). Chinese
families earned about S$400 more a month than did Indian or Malay families.
The other 4 significant variables were more directly related to infant care. For
both the husband's and mother-in-Iaw's attitude towards breast-feeding, more
Chinese than expected statistically had relatives whose attitude was neutral, negative
or not known; while disproportionately more Malays, and to a slightly lesser extent
more Indians, had relatives whose attitude was favourable. For example, 53% of
Chinese, 79% of Indians and 83% of Malays had husbands that favoured breast-
feeding. Most subjects (69%) indicated that they themselves had been fed by breast
rather than by bottle, but disproportionately few were Chinese. A similar pattern was
observed for help with the care of the baby: 67% of all subjects received daily help,
but again fewer than expected were Chinese. I did not ascertain whether help was
from relatives or paid workers. Since Chinese families more commonly employ
servants (personal observation), the aid Chinese mothers receive may differ in kind
as well as amount from that received by other mothers.
Behavioural attributes. Only 2 variables in this group, foods and drinks avoided
during pregnancy and those avoided during the first month after the birth, differed
among the communities.
During pregnancy, 62% of Chinese, 44% of Malays and 39% of Indians avoided
foods and drinks that were considered 'cooling' or were believed to result in an
unhealthy baby. Cooling foods included many types of fruit (e.g. papaya, water-
melon), vegetables (e.g. Chinese cabbage, cucumber) and meat (e.g. duck, prawns).
Foods believed to affect the foetus adversely (by, for instance, causing jaundice or
fits) included mutton, pineapple, ginger, durians and strong spices. Altogether 45
specific items from both groups were identified. In addition, a small number of
subjects avoided some general categories of foods in either or both groups; namely,
fruit, vegetables, meat, seafood (including freshwater fish), sweet foods, cold foods
or cold drinks. Nearly all the Malay and Indian mothers that restricted their diet
during pregnancy abstained from foods believed to adversely affect the foetus.
During the first month after the birth, 55% of Chinese, 33% of Malays and 44%
ofIndians restricted their intake of foods believed to prevent healing by mothers (e.g.
pork, chilli), to interfere with breast-feeding in some way (e.g. ginger, sesame oil,
Chinese wine) or were cooling (see above). For all 3 groups of foods, 24 specific
items were identified. In addition, a few of the following general categories of foods
and drinks were avoided by a number of subjects: carbonated drinks, meat, fruit,
vegetables, spices, white-coloured food, cold drinks, canned foods and sweet foods.
101
Fruit and vegetables were less frequently avoided at this time than during the
pregnancy, but meat, spices and fluids (including tea, coffee and water) more often
were. Malays and Indians avoided a wider range of items at this time than during
pregnancy, but Chinese avoided a narrower range.
Nearly all subjects (95%) took prescribed medicines (e.g. iron tablets) during
pregnancy - and consequently this variable was not tested statistically - but all 9 that
consumed non-prescribed medicines (e.g. 'Chinese herbs') were Chinese.
Discussion
My major finding was that among Singapore's well-to-do families breast-feeding

was favoured most by Malay mothers and least by Chinese mothers, and Indian
mothers showed preferences in between but closer to those of Malays. This pattern
was the same for several different measures of feeding recent babies, namely, initial
choice of a feeding method, feeding solely by breast for at least 6 to 9 weeks and
feeding by breast with or without supplements for at least 6 to 9 weeks; it also held
for the average duration of suckling the most recent previous child (see Table 1). The
same order of preference has probably existed for both poor and well-to-do families
since at least the early 1950s (see Introduction). A lower incidence of breast-feeding
among Chinese compared with non-Chinese has perhaps existed throughout
Singapore's history, since wealthy Chinese mothers have traditionally used wet
nurses (e.g. [21,22]).
Many short-tenn studies of rural and urban populations in Malaysia have
observed similar patterns for the same 3 ethnic communities (e.g. [23-26]). One
extensive study, the Malaysian Family Life Survey, found consistent relative
differences among the communities from 1950 to 1977, though for Malays there was
no decline in the duration of breast-feeding and only a small drop in initiation [14].
For example, during the period of 1950 to 1969 about 74% of Chinese, 84% of
Indians and 97% of Malays initiated breast-feeding; during the period of 1975 to
1977, the proportions were 54%,85% and 96%.
I also found a number of features of the subjects themselves and their environ-
ments and behaviour that varied significantly among the ethnic classes and therefore
could have been responsible for differences in feeding behaviour among the
communities (see Table 2). However, only a few appeared to be important.
Since breast- and bottle-feeders did not differ for any physical or health measures
[1, 2], it is unlikely that the age and weight differences among ethnic classes had
much effect on their feeding behaviour. The significantly higher family income of
Chinese mothers can apparently be dismissed as a cause of their greater use of bottle-
feeding since this method is more common among poor mothers [2]. This conclusion
is consistent with an absence of ethnic variations in the other general environmental
conditions that I studied (see Results).
If breast-feeding is considered a good health practice (e.g. [27]) and abstention
102
from specific or whole categories of common foods and drinks a poor one, my
findings for the remaining significant variables show a high degree of consistency.
For the 4 environmental conditions, Malays made up the highest proportions of
subjects that (i) received help with the care of the baby, (ii) had husbands with
positive attitudes towards breast-feeding, (iii) had mother-in-Iaws with positive
attitudes towards breast-feeding, and (iv) had themselves been fed by breast; while
Chinese made up the lowest proportions of subjects with these features. For the 2
behavioural attributes, foods avoided during the pregnancy and foods avoided during
the first month after the birth, Chinese mothers were predominant. The only
inconsistency was that fewer Indians abstained from eating some foods during the
pregnancy while fewer Malays abstained after the birth.
My findings all together indicate that among well-to-do Singaporeans the ethnic
communities did not differ for factors that would be expected to act more or less
similarly on all 3 groups if applied equally (e.g. education, mother's health during
pregnancy). Rather, they suggest that the communities varied for factors that, while
having significant effects on infant feeding, are really reflections of more fundamen-
tal, culturally determined, attitudes and practices. For example, I believe that breast-
feeding is profoundly influenced by the amount of help a mother receives with the
care of the baby, which in tum is appreciably influenced by a family's cultural
heritage. Therefore, to account for differences in feeding patterns among Singapore's
ethnic communities, it is necessary to probe into the origins of their basic cultural
differences.
Within Malaysia and Singapore, nearly all Malays are Muslims, and their
adherence to Islam is one of the most important factors distinguishing them from
non-Malays [28, 29]. Few non-Malays, for instance, participate each year in
ramadan, the month long ritual of abstention from eating, drinking or having sexual
intercourse during daylight hours. Of direct relevance to my study is Islam's specific
commandment that Muslim mothers should breast-feed each of their children for 2
years (Koran: chapter 2, verse 233; chapter 31, verse 14). At least a general notion of
the link between religion and breast-feeding is common among Malays. In a popular
book on culture in Malaysia and Singapore Malay mothers are reported to believe
that breast-milk 'strengthens the child's spirit as well as his body, that it develops his
faith and his character, that it makes a bond so spiritual that nothing can destroy the
relationship of the child and his mother' [30, pp. 116-117]. A very recent survey of
breast-feeding in Singapore found that significantly more Malay mothers (52.6%)
than Chinese (2.1 %) or Indian (21.4%) mothers believed that their religion had
something to say about infant feeding [3].
It is commonly presumed in Singapore that this country's Malay community
possesses the same 'rural view' of life as does the predominantly rural Malay
community in Malaysia. Since breast-feeding is more prevalent in the countryside
than cities throughout Southeast Asia [29], the perception of rural origins could be a
second major factor promoting this feeding method among Malays. In both this study
and my previous one on poor mothers disproportionately more Malays than Chinese
103
or Indians gave tradition as a reason for breast-feeding; and more had relatives with
positive attitudes towards breast-feeding. In this study and a later one [3] Malay
mothers received the most help with the care ofthe baby (see also [10]).
As mentioned, Indian mothers in Singapore and Malaysia show patterns of breast-
feeding in between those of the other 2 communities though closer to those of
Malays. Nearly as many initiate breast-feeding as do Malay mothers, but they
continue feeding for shorter periods [9, 14,26]. They received help with the care of
children nearly as often, which suggests that Indian families, like Malay families,
possess some traditional, rural-derived practices. For example, a large proportion of
Indian males in Singapore have spent much of their lives in rural environments [29].
Few mothers of this community, however, benefit from religious support for breast-
feeding.
In contrast, the Chinese community is the most modem in the sense of readiness
to follow current Western trends in, for instance, consumer goods, leisure activities
and degree of mobility. Although my sample did not reveal ethnic differences in
perceived lifestyles, Wong and Ng's study [29] of fertility patterns in Southeast Asia
found that among ethnic groups within the 5 countries surveyed (viz. Indonesia,
Malaysia, the Philippines, Singapore and Thailand) the Chinese ranked at the top or
near the top for all 7 of their measures of modernity (except radio-listening, for
which they substituted TV-watching). This community is also the wealthiest in all 5
countries, and the one most often associated with urban businesses. For infant
feeding in particular, the Chinese apparently prefer bottle-feeding, both as a symbol
of their modernity and for its convenience.
The conclusion of Wong and Ng is supported by the findings of several other
studies. In Singapore, Chung [10] observed that significantly more Chinese than
Malays or Indians believed that bottle-feeding is more modem than breast-feeding.
He also found that more Chinese mothers presumed that breast-feeding ruins the
figure; and in this they are perhaps being modem in having a greater concern for the
sexual rather than the nurturing function of breasts. Convenience as a major reason
for bottle-feeding by Chinese has been recorded by several Singaporean surveys [6,
9, 10], as well as mine [1] and a British study [31]. In a local report [10] nearly all
the mothers bottle-feeding for convenience gave freedom from the home as the
reason for bottle-feeding's convenience. In a similar vein, Chinese parents in
Singapore are more likely than parents in the other communities to consider
restrictions on their freedom as one of the disadvantages of having many children
[29].
Notwithstanding their adoption of numerous modem attitudes and practices, many
Chinese have retained the traditional Chinese view of the pregnancy, birth and first
month after the birth as periods of extreme danger to the mother's health [32]. In my
study, Chinese mothers more often used prescribed drugs during the pregnancy,
during the birth (Viegas, personal observation) and to suppress lactation. Chung [10]
observed that significantly more Chinese believed that breast-feeding weakens the
mother. A final example comes from my study of poor mothers where Chinese made
104
up 66% of the sample but 85% of subjects that consumed health tonics during
pregnancy [1].
This view of pregnancy, birth and lactation is of course derived from the tradi-
tional Chinese view of medicine. The body is in a constant state of change brought
about by the interaction of 2 complementary forces, the yang (or male, creative,
heaven, etc.) principle and the yin (or female, passive, earth, etc.) principle. Health is
maintained by the correct balance of yang and yin foods and drinks in the diet. To
achieve this, every illness and debilitating condition is classified by its yangness or
yinness, and more of one type of food and drink and less of the other are then
consumed [32]. In practice, most ill-health and weaknesses are considered as
resulting from excessive yin. During pregnancy, for example, excessive yin is
counterbalanced by the avoidance of yin foods and drinks, such as tea and cooling
fruits and vegetables (see Results), and by the increased consumption of yang foods
and drinks, such as eggs, chicken soup and bird's nest soup [22, 33]. After the birth,
excessive yin makes the mother particularly susceptible to contamination by evil
spirits, and she and the baby must remain secluded within the home for a month or so
[34]. Foods specifically identified as interfering with healing (e.g. most seafoods) are
also avoided.
Both the Malay and Indian communities also believe that the mother and baby are
in physical and spiritual danger during the first month or so after the birth - 28 days
for Indians [30] and 44 days for Malays [35]. But their food restrictions are in
general less severe than those of the Chinese; the range of restricted items, for
example, is much smaller, especially during pregnancy (see Results). Millis [35]
contends that dietary restrictions by Malays during pregnancy are for personal not
traditional reasons.
Although it is likely that these dietary practices were adaptive in the past, their
present consequences are probably not beneficial for the baby. Because of the
modem alternative to breast-feeding, one practice in particular has become maladap-
tive, that is, the refusal of some mothers to breast-feed at all because it would require
a change in their diet. Chen [9], for example, found that 15% of her sample were
unwilling to give up ginger or sesame oil to breast-feed jaundiced babies - both these
foods are believed to help the mother's recovery but to be detrimental to babies. At
least a few women in my study chose not to breast-feed because they did not wish to
give up foods considered bad for even healthy babies (e.g. ginger, Chinese herbs) or
considered as interfering with milk production (e.g. Indian herbs).
There are other possible consequences of dietary taboos during pregnancy and
lactation. Many yin foods are high in vitamin C [32], and their avoidance could be
partly responsible [4,5,6,32,33,35] for the very high incidence of lactation failure
among Singaporean and Malaysian women [4, 5, 6, 7, 9, to, 12,24,26]. Elsewhere
[2], I have reported that among breast-feeders a greater proportion of those practising
food taboos (43%) breast-fed for less than 15 days than did those on an unrestricted
diet (19%). The restriction of fluids could be a more serious problem [35].
Although an incidence of 60% among well-to-do mothers for initiating breast-
105
feeding may appear to be high for a developing country, the incidence for poor
mothers is only about 36% [1]. Moreover, I estimate that the percentage of babies
from both economic classes and all 3 ethnic groups presently being fully or partially
fed by breast for at least 3 months is at most 20% [2]. In 1980 only 13% of a large
sample of mostly poor mothers breast-fed up to 2 months [9]. Both this study and the
companion work on poor mothers had identical percentages - 16% - of multiparous
mothers that had breast-fed their most recent previous child for at least 3 months
(Counsilman, unpublished).
I believe that it follows from my findings that improving the incidence and
duration of breast-feeding in this country will largely involve strengthening or
redirecting the influence of several traditional practices and weakening the influence
of several modem ones. I do not believe that national campaigns or teaching
programs for medical staff or mothers should concentrate on the physical benefits of
breast-feeding. These are reasonably well-known, at least among well-educated
women [3]; but the culturally based attitudes and practices that may promote or
interfere with breast-feeding are not. Recommendations based on these findings will
be presented in a future report.
Summary
Within my sample of well-to-do Singaporean mothers, 57% of Chinese, 90% of

Indians and 94% of Malays initiated breast-feeding. Several other measures (e.g.
total suckling period) also indicated that breast-feeding was favoured most by
Malays and least by Chinese. This pattern could not be attributed to ethnic variations
in physical characters (e.g. weight), health measures (e.g. past reproductive failures)
or general environmental conditions (e.g. education). Rather, it appeared to reflect
basic cultural attributes. Breast-feeding among Malays is promoted by their religion
and by support from relatives, which is perhaps a traditional practice derived from
the recent rural origins of this community. In contrast, bottle-feeding among Chinese
is advanced by their greater tendency to adopt Western habits. In addition, at least
one traditional dietary practice stops some Chinese mothers from breast-feeding, and
several others may reduce its success. Indians are similar to Malays in diet and aid
from relatives, but most do not benefit from religious support.
Acknowledgements
I thank S. Chua of the Department of Obstetrics and Gynaecology, National

University Hospital, for collecting the data; and o. Viegas of the same Department
for obtaining pennission from Kandang Kerbau Hospital for the study.
106
References
1. Counsilman 11, Chan SY, Haiyon H, Rahim NA, Salim R, Tai TY, Tan ML, Zainy Z and Viegas 0:
J Trop Pediatr 32:310, 1986.
2. Counsilman 11, Chua S and Viegas 0: J Trop Pediatr 32:313, 1986.
3. Counsilman 11 and Yong A: Unpublished, 1986.
4. Millis J: Med J Australia 1:322,1954.
5. Millis J: J Pediatr 45:697, 1954.
6. Millis J: Med J Malaya 10:157,1955.
7. Millis J: Med J Malaya 13:145, 1958.
8. Wong HB: Breast Feeding in Singapore, Choon Kee Press, Singapore, 1971.
9. Chen AJ: Singapore Community Health Bu1122:32, 1981.
10. Chung SY: M Sc thesis, National University of Singapore, 1979.
11. Counsilman JJ and Viegas 0: Trop Biomed 2:161,1985.
12. Wong HB, Paramathypathy K and Tham NB: J Singapore Paediatr Soc 5:89,1963.
13. Wong HB: J Singapore Paediatr Soc 3:14,1961.
14. Haaga JG: Am J Public Health 76:245, 1986.
15. Tan KL: Annals Ac Med 12:609, 1983.
16. World Health Organization, Division of Family Health: World Health Stat Q 35:92,1983.
17. Counsilman JJ, Mackay EV and Copeland RM: Aust N Z J Obstet Gyn 23:208, 1983.
18. Klackenberg G and Klackenberg-Larsson I: Acta Paediatrica Scandinavica Suppl 187:94, 1968.
19. Sudderruddin KI: Singapore 1985, Information Division, Ministry of Communications and
Information, Singapore, 1985.
20. Siegel S: Nonparametric Statistics for the Behavioral Sciences, McGraw-Hill, New York, 1956.
21. Kao YE: Acta Paediatr 36:233, 1948.
22. Platt BS and Gin SY: Arch Dis Child 13:343, 1938.
23. Dugdale AE: Far East Med J 6:230, 1970.
24. Teoh SK: Med J Malaysia 30:175,1975.
25. Pathmanathan I: Med J Malaysia 33: 113, 1978.
26. Chen ST: Med J Malaysia 33:120,1978.
27. Counsilman JJ and Mackay EV: Aust N Z J Obstet Gyn 25:101,1985.
28. New Encyclopaedia Britannica 11 :368, 1978.
29. Wong AK and Ng SM: Ethnicity and Fertility in Southeast Asia, Research Notes and Discussions
Paper No. 50, Institute of Southeast Asian Studies, 1985.
30. Craig J: Culture Shock, Times Books International, Singapore, 1979.
31. Goel, KM, House F and Shanks RA: Brit Med J 2: 1181, 1978.
32. Ling S, King J and Leung V: Amer J Chinese Medicine 3:125, 1975.
33. Millis J: J Trop Pediatrics 2: 103, 1956.
34. Koh THG: J Trop Pediatrics 27:88,1981.
35. Millis J: Med J Malaya 13:139,1958.
A prospective study of growth and nutrition of Aboriginal
children from birth to two years in North-West Australia
MICHAEL GRACEY and HELEN SULLIVAN
Introduction
TIl-health is very prevalent in Australian Aborigines whose infants and children are
often undernourished and have high rates of infections, particularly of the respiratory
and gastrointestinal tracts [1-4]. The high levels of morbidity which exist in young
Aborigines are demonstrated by the hospital admission rates of Aboriginal infants for
gastroenteritis in Western Australia being 16 to 20 times those for non-Aboriginal
infants [5]. Despite the seriousness of these problems, there is little systematic
information about the factors which exist in Aboriginal communities to cause
unfavourable poor growth in their infants and young children which is usually
associated with high rates of infections. The "Kimberley Mothers' and Babies'
Project" was undertaken in several remote communities in the Kimberley region of
the tropical far north of Western Australia, which has one of the largest populations
of full-blood Aborigines in Australia, to investigate these circumstances at the
community level. The present paper reports some aspects of the project which
document maternal and environmental factors which were associated with growth
and health outcomes in a prospectively studied cohort of 48 infants born in the
region.
Material and methods
The study was done in six autonomously run Aboriginal communities in the Fitzroy
River Valley district in the far north of Western Australia. The area is at 17-18°
latitude south, 123-127° longitude east, in semi-arid, sparsely populated savannah
country. The climate is tropical with extremely hot, monsoonal wet summers
(December-April) and very dry, warm winters (May-November). The total
Aboriginal population of the communities (which are spread over 300 km) of about
1500 lives in groups of 120 to 430 persons in conventional housing with ablution
facilities which are provided by government agencies. Communities are serviced by
108
doctors and public health nurses from the Health Department of Western Australia
who conduct regular clinics and provide emergency services. Pregnant women are
transferred at 38 weeks' gestation to the Regional Hospital at Derby, on the coast, for
their confinements. All pregnant women from the communities who were due to
deliver babies in the 12 months from July 1984 to June 1985 were invited to take
part and involve their infants in a study of their health, growth and nutrition. In the
enrolment period there were 54 pregnant women who were eligible; of these, two
women declined to take part, one miscarried, one child was fostered out and one
family left the area. The remaining 49 infants in the original cohort (24 boys, 25
girls) were studied, at monthly intervals, to 24 months of age except for one infant
girl who died from viral myocarditis at 11 months of age.
Families live in government-provided houses for which regular rent is from
family income or, mostly, from social security payments which are provided, usually
every fortnight (Le. two weeks). Houses have running water, flush toilets, electricity,
most have wood-burning stoves and some have LPG gas. Each community has its
own community-run food store and all have access to town stores which can be up to
2 hours' drive away. Fifty four percent of families used community stores; the rest
used town shops for their supplies. Stoves were provided but only 58% used them
and 54% of families cooked their food, in their customary way, in fires on the
ground. Fourteen percent of toilets were blocked when environmental information
was regularly collected; only 27% of toddlers used toilets or had their nappies
(diapers) adequately disposed of. Family and community rubbish collections were
made regularly in 90% of households and 8% of families had rubbish collected only
every 2 weeks. Households had an average of 9.5 persons including 2.2 under-fives
and 2.0 schoolchildren. Households had, on average, 5.2 dogs, cats and other animals
which included lizards and pet birds. Information about mothers' pregnancies and
deliveries was obtained from routine clinic and hospital records. Laboratory
investigations were done by the State Health Laboratory Services using standard
methods. Personal information and data about smoking and alcohol consumption
during pregnancy were obtained by one of us (H.S.) by direct questioning. Detailed
community-based information about socio-economic and living conditions was
gathered every 3 months throughout the 2-year study period.
Statistical methods
Data were analysed using multiple regression, MANOVA [6] or ANOVA except for
qualitative (ranked) variables for which the Chi-square test, with Yates' correction,
was used. Descriptive statistics were obtained using the Statistical Package for the
Social Sciences (SPSS).
Each child was weighed unclothed and measured within 5 days of the due date of
their age each month during the first year and within 7 days during the second year
of life. Subjects were weighed on standardized, Seca beam balance scales (to the
109
15
12
kg
12 18 24
months
Figure 1. Fitted growth curve of a girl in the 'good' nutritional group. The 95th, 90th, 50th, 10th and 5th
centiles of the NCHS (1977) reference values are shown.
15
12
kg
12 18 24
months
Figure 2. Fitted growth curve of a child from the 'moderate' nutritional group which included more than
half the children and characteristically showed growth faltering and 'crossing centiles' after 6 months of
age.
110
15
12
kg
12 18 24
months
Figure 3. Fitted growth curve of a child from the 'bad' nutritional group; these children had obvious
growth faltering and failure to thrive.
nearest 10 g) and recumbent length measured (to the nearest 5 mm) with specially
designed boards using standard techniques [7, 8].
Nutritional status of children was assessed with the method described by Count
[9] in which:
y = c + at +b log t
This provides fitted growth curves from one month to 2 years of age (Figures 1-3).
Parameter (a) for weight was ranked, median values obtained, 95% confidence levels
determined and then divided into 3 groups; viz, (i) above, (ii) within, and (iii) below
confidence limits. Using this method, nutritional status can be described as 'bad',
'moderate' or 'good' [9-11].
Results
The average age of mothers in the study was 24 years 4 months and their average
height was 160 cm. Their prior obstetric experience was: gravida 1 (33%), gravida 2
(31 %), gravida 3+ (36%); para 0 (33%), para 1 (34%), para 2+ (33%). Of 65
previous infants, 17 (26%) were of low birthweight « 2.5 kg) including 7 infants
who weighed less than 2 kg at birth. Forty-eight percent of mothers were anaemic at
their first antenatal visit during the present pregnancies and two-thirds were recorded
111
as being anaemic during pregnancy despite 87.5% of the women being prescribed
iron preparations. Ten of the women (21 %) had urinary tract infections detected
during pregnancy. Five mothers had glycosuria during pregnancy, 4 had elevated
blood pressure (systolic> 130 and/or diastolic> 90 mm Hg) on 2 or more occasions,
3 had pre-eclamptic toxaemia, 6 had proteinuria of undetermined cause(s), one-third
of the pregnancies were induced and 8 of the women had forceps deliveries. Two
women had one (or less) antenatal visits, 20 had 2 to 5 antenatal visits. Weight gain
during pregnancy was < 3 kg in 4 women and > 12 kg in ten women.
Mean weights (3.13 kg) and lengths (50.5 cm) of infants at birth correlated
positively with normal maternal systolic and diastolic blood pressures, controlled
weight gain and the absence of oedema or proteinuria during pregnancy, and with
uncomplicated vaginal deliveries. Other positive correlations included absence of
maternal anaemia, regular iron medication, regular antenatal clinic supervision and
classes, the number of visits to the doctor during pregnancy, abstinence from alcohol
and lack of smoking or stopping smoking during pregnancy. Babies of mothers who
smoked through pregnancy (n=12) were lighter (2.69 kg) and shorter (49.3 cm) than
those of non-smokers (3.23 kg and 49.9 cm). Birthweights of infants of the 39
mothers who did not drink alcohol or stopped by 12 weeks' gestation were sig-
nificantly greater (3.23 kg) than those five mothers who stopped drinking between 12
and 30 weeks' gestation (2.65 kg) or those who stopped later or continued to drink
throughout their pregnancies (3.11 kg).
Gestational ages of infants were significantly greater in mothers with more
antenatal attendances to doctors or clinic nurses and with mothers whose systolic or
diastolic blood pressures were not elevated.
At birth Kimberley boys were, on average, 190 g lighter and Kimberley girls were
60 g lighter than median international reference values [II, 12]. Only 19% had
satisfactory growth and nutrition to two years of age, 25% had obvious growth
faltering and failure to thrive and 56% had intermediate degrees of growth faltering
and undernutrition (Figures 1-3). Between 3 and 6 months of age rates of weight
gain and linear growth fell below those expected and after 9 months of age
Aboriginal infants were lighter than expected reference values and remained so until
2 years of age. By then the Kimberley children were more than 1 kg lighter and 4.5
cm (boys) and 3.6 cm (girls) shorter than the international reference values. Most of
the children had episodes of growth faltering or failure to thrive in the second half of
the first year of life and/or during the second year of life.
Maternal Hb> I1g%, regular attendance at antenatal classes, uncomplicated
vaginal deliveries and characteristics which correlated with improved birthweights
all correlated with satisfactory growth to 24 months of age. Favourable growth
correlated with families being within walking distance of food stores, income being
regular every fortnight, the children being from households which did not run out of
food, where hand-washing before food preparation and after going to the toilet was
practised and where infants' soiled napkins (diapers) were adequately disposed of.
Families with well-nourished children also had significantly more rubbish bins.
112
Good growth was also consistently associated with households having fewer major
consumer items; there were significantly fewer refrigerators, freezers, air-con-
ditioners, clothes washing machines, television sets, video cassette recorders (VCRs)
and cassette tape players in families of children with 'good' nutrition. Overall, 49%
of families had refrigerators, 22% had food freezers, 11 % had air-conditioners, 48%
had clothes washing machines, 20% had VCRs and 20% had television. Overall,
there was only 1 knife for every second household member, 0.5 forks/person, 0.9
pannikins (cups)/person and less than 1 plate/person/household. Families of ade-
quately nourished children had significantly more spoons, knives and forks per
person.
There were more cats, dogs and other animals and more cockroaches in the 'bad'
nutrition group than in the 'good' nutrition group. Leaking household taps, inade-
quate nappy washing (to 12 months) and lack of adequate rubbish bins (to 18
months) were significantly associated with poor growth outcomes.
Overall, 84% of households had hot water, 90% washed their dishes adequately,
86% had food storage facilities, 55% had lockable space for food and 25% had tin
trunks in which food could be kept securely. Average family income* was $422.74
each fortnight; fifty-one percent of mothers gambled regularly; on average, this cost
$11.50 per week.
Sixty percent of the infants were admitted to hospital in their first six months of
life. Main reasons for their first admission in that period were respiratory infections
(17% of the group), gastroenteritis or vomiting (13%) and social reasons; con-
junctivitis was an important secondary cause of admissions. One-third of children
were re-admitted, mostly for respiratory, gastrointestinal and skin infections and
because the infants were abandoned or for other social reasons.
In the second 6 months 65% of infants were hospitalized; main reasons were
respiratory infections, social reasons or poor feeding practices, gastroenteritis, fever
or febrile convulsions, skin and eye infections, and weight loss or failure to thrive.
Second and later admissions in that period were mostly for respiratory infections,
fever, gastroenteritis and failure to thrive. More than two-thirds of the children were
hospitalized in the 12-18 months' age period. Respiratory and gastrointestinal
infections were prominent causes for admission but failure to thrive, social causes,
nutritional anaemia and urinary tract infections and intestinal parasites became more
important causes for admissions; a similar pattern occurred in the second half of the
second year of life. On average, the children were admitted 5.1 times and spent 32
days in hospital in the first two years of life mostly for infections, failure to thrive
and social reasons.
Discussion
This prospective study of 48 Aboriginal infants in North-West Australia showed that

* One Australian dollar is worth approximately US$ 0.76 (1990).
113
less than 20% of them grew satisfactory in the first two years of life, 25% had
obvious growth faltering and failure to thrive while the rest had intermediate degrees
of undernutrition and growth failure. This confirms the pattern of widespread mild to
moderate undernutrition which has been documented many times in Aboriginal
children over the past two decades [1, 3, 9, 13-17].
Birthweights were positively correlated with mothers' good health during
pregnancy, regular antenatal supervision and lack of drinking alcohol or smoking.
Satisfactory growth in the first two years of life correlated with these maternal
characteristics and with absence of maternal anaemia postnatally and with better
access to food supplies, better financial management at home and improved stan-
dards of personal and family hygiene. 'Good' nutritional outcomes were consistently
associated with households with fewer expensive consumer items such as television,
air-conditioners and video cassette recorders. These items were probably purchased
by households of children with unsatisfactory growth from fortnightly social security
payments thus jeopardizing the household's ability to ensure a regular food supply
until the next social security payment became available. Basic food costs are 35% to
45% higher in these communities than in southern Australian cities [18] so with
fixed, limited incomes and poorly developed budgeting skills, many families have
difficulty maintaining adequate and nutritious food supplies; in some households this
is made worse by alcohol abuse and gambling [19]. Children had high levels of
morbidity as judged by their attendance at public health clinics and/or hospital
outpatient clinics or by patterns of admission to hospital; hospital outpatient
attendances for illnesses and dressings ranged from 13 per child over the 2 years in
the 'moderate' nutritional group to 19 in the 'bad' nutritional group.
More than half the children were admitted to hospital in each of the four 6-month
periods in their first two years. As with outpatient and clinic attendances, admissions
were not significantly less frequent in the children with 'bad' nutrition compared
with the other two groups. This might be because children with the lowest standard
of nutrition have more frequent and serious infections and families of children with
the best nutritional standard present them more often to hospital because of concern
for their health while most children who were in the 'moderate' nutritional group
came from families who might not appreciate the potential seriousness of minor
illnesses and disturbances of their children's health. It is probably also because of the
high frequency of minor infections and trauma in Aboriginal children rather than the
severe and life-threatening infections which are more characteristic of children with
severe malnutrition in developing countries [20]. Most of the hospital admissions
experienced by these children were for infections, particularly respiratory and
gastrointestinal; this is similar to Mata's prospective study of children growing up in
Central America [21] as well as in undernourished populations elsewhere [22] and
with other studies of young Aborigines [23,24]. Gastrointestinal infections are very
prevalent in Aboriginal children with prominent infecting agents being en-
terotoxigenic E. coli, Aeromonas, Salmonella, Shigella and Campylobacter [25].
The present study has the advantages of being prospective, longitudinal, monthly
114
and community-based. It, therefore, gives a better perspective about health, nutrition,
growth and illness in these children than could be provided by retrospective or cross-
sectional data from clinic or hospital material. Another important aspect of this study
relates to the potential for growth in Aboriginal children; this is a controversial
subject which has been discussed elsewhere [26]. Cocldngton [27] has shown that
Aboriginal children who lived in social circumstances similar to Caucasian
Australian children had growth patterns similar to those of other Australian children.
In the present study nine of the children (19% of the cohort) had growth patterns
very similar to, and statistically indistinguishable from, the 50th centile curve of the
NCHS reference values [12]. This suggests that the growth potential of Aboriginal
children may be similar to that of other racial groups and that the NCHS reference
values are appropriate to the assessment of growth and nutrition of young
Aborigines.
Acknowledgements
This work was supported by grants from the Australian Institute of Aboriginal
Studies and the Research and Development Grants Advisory Committee of the
Australian Department of Health. We thank the Commissioner of the Health
Department of Western Australia for approval for the work and for the assistance of
his field staff in the Kimberley. Special thanks are due to the Community Councils
for their approval of the project and for the help of their health workers Stella Hale,
Edna Mungulu, Mary Nellie, Mary Shandley, Mary Spinks and Mary Vanbee for
their work.
References
1. Gracey M. Murray H. Hitchcock NE. Owles EN and Murphy BP: Nutr Res 3: 133. 1983.
2. Gracey M: Med J Aust 143:S43. 1985.
3. Roberts D. Gracey M and Spargo RM: Med J Aust 148:68. 1988.
4. ThomsonN: Aust NZJ Med 14:705. 1984.
5. Gracey M and Anderson CM: Aust Paediatr J 25:230. 1989.
6. Rao CR: Advanced statistical methods in biometric research. Wiley. New York. 1952.
7. Jelliffe DB: The assessment of the nutritional status of the community (Geneva: World Health
Organization). 1966.
8. United Nations: How to weigh and measure children. Assessing the nutritional status of young
children in household surveys. U.N. Department of Technical Cooperation for Development and
Statistical Office. New York. 1986. (DP/UN/lNT-81-04l/6E).
9. Count EW: Human Biology 15:1. 1943.
10. Berkey CS and Reed RB: Human Biology 59:973. 1987.
11. Gracey M and Sullivan H: Annals of Human Biology 15:375. 1988.
12. National Center for Health Statistics: NCHS growth curves for children. birth to 18 years.
Washington. D.C.; DHEW Publication no. (PHS) 70-1650.1977.
13. Kettle ES: Med J Aust 1:972. 1966.
14. Propen DN. Edmonds R and Parsons PA: Aust Paediatr J 4:134.1968.
15. Kirke DK: Med J Aust 2:1005. 1969.
115
16. Maxwell GM and Elliott RB: Am J Clin Nulr 22:716, 1969.

17. Gracey M: In: Wahlqvist ML and Truswell AS (eds) Clinical Nutrition. John Libbey, London, 1986.
18. Sullivan H, Gracey M and Hevron V: Med J Aust 147:334,1988.
19. Hunter EM, Spargo RM: Med J Aust 149:668, 1988.
20. Scrimshaw NS, Taylor CE and Gordon JE: Am J Med Sc 237:367,1959.
21. Mata L: The children of Santa Maria Cauque: A prospective field study of health and growth. MIT
Press, Cambridge, Massachusetts, 1978.
22. Black RE, Brown KH, Becker Sand Yunus M: Am J EpidemioII15:305, 1982.
23. Shannon TGD and Gracey M: Med J Aust Special Suppl 11, 1977.
24. McNeilly J, Cicchini C, Oliver D and Gracey M: Med J Aust 21 :547 (1983).
25. Gracey M, Burke V and Robinson J: Ann Trop Paediatr 3:35, 1983.
26. Gracey M and Sullivan H: Ann Hum BioI 15:395, 1988.
27. Cockington RA: Aust NZ J Med 10:199, 1980.
Paper by J.A. Kusin, Sri Kardjati, C. de With and

W.M. van Steenbergen
D. Karyadi:
There are three things I would like you to comment on. Firstly, do you also have data on the mother's
activities, because this will have an influence on their energy balances. Secondly, do you have
collected data on morbidity and mortality of the children in addition to the 12 month period of follow-
up investigation. Finally, did you also consider micronutrients in your study, such as iron and other
trace elements or vitamins.
fA. Kusin:
With respect to your first question, we now indeed have more data available on the activities of the
women. These data were collected by the field workers who stayed for 72 hour periods with the
mothers and who also recorded food intake or the breast-milk intake of the infants. It was observed
that the mothers were for more than 50% of the time resting. They just sat down or were doing other
work which did not really need much energy; perhaps this can be attributed as an adaptation to the
very low intakes. Concerning your second question on morbidity and mortality of the infants I would
like to refer you to Dr. Kardjati.
S. Kardjati:
Please let me summarise our preliminary findings of the second phase of our studies in which
morbidity and its effect on food consumption were recorded. During one year (June 1987-July 1988)
50% of birth cohorts 1985, 1986 and 1987 in two villages were visited at home at weekly intervals.
Cautiously estimated, we found that the total number of days with sickness was between 70 and 140
days .a year. This means that for about one third of the time during that period the children were
recorded as sick. The amount of food consumption during these periods was reduced by about 20%
compared to periods when the infants were not recorded as being sick.
I.A. Kusin:
We observed each child when it was sick and measured its food and breast-milk intake and its activity
pattern. Subsequently, we returned to this child when it was healthy and performed the same
measurements. The breast-milk intake did not differ but the children did eat significantly less, in spite
of being offered enough food. It should be stressed here that there is no taboo about feeding these
children while they are ill; they just did not want to eat.
As far as mortality is concerned, the figure was 120 per 1000 during the first year, of which 38%
died during the first months and 70% died before the age of 6 months. We found a clear relationship
with birthweight. Although we do not dare to conclude that it is caused by supplementation, we found
a significant reduction in perinatal mortality in the high energy supplemented group.
H.KA. Visser and J.O. Bindels (eds.) Child Nutrition in South East Asia, 117-122.
118
Your last question can be answered very briefly: we did not include any micronutrients like iron or
other trace elements in the supplement
C. Capa/an:
In many supplementation studies, a confounding factor is that the supplement influences the intake
during the normal meals. I could imagine that intake of the high energy supplement appears to be
somewhat "filling". This is, for instance, also seen in some school meal programs where because of
the school lunch, the home dietary intake turns out to be reduced. Could you please indicate how you
controlled this factor.
I.A. Kusin:
I fully agree that this is one of the key points in all supplementation studies and although we have
indicated this already in our paper, I am pleased to give it some more emphasis in this place. The
beauty of the Jamu is that it is not considered a food. We gave it very early before mothers go to work
and poor women do not have breakfast. If you look at figure 5, and compare the intakes, excluding
those derived from the supplements. in the respective compliance sub-groups of the high energy and
the low energy group. you will see that they are highly comparable. We also compared these intakes
with those recorded at the same calender months over a period of 5 years and could not detect
significant differences. So we are reasonably confident because of the consistency of the data over
several years and because of the similarity of the intakes in both groups. that there was no replace-
ment. To conclude, it is also necessary to state that Jamu for pregnant women. which is very specific
to Indonesia. will never be shared with anyone because it is too bitter for children and men.
Paper by S.H. Pudjiadi

M.S. Trastonenoyo:
Please allow me to start with a comment about the early introduction of solid food. I think that it is
rather dangerous, in the case of failure to breast-feed. to give solid foods straightaway. We should first
try to find the cause for these problems and to see whether we can solve them because if you directly
give solid foods then breast-feeding will certainly stop.
S.H. Pudjiadi:
Indeed. you are right. if you give solid foods too early when the baby is exclusively breast-fed.
lactation will gradually stop. Generally. I recommended in my paper that if possible solid foods should
not be given before the baby is 4 months old. If the baby does not receive enough breast-milk. we
advise the mothers to feed them more often and to increase the duration of the feedings to at least 5
times a day and at least 20 minutes per feeding. However. if this fails and the mother obviously cannot
buy any commercial infant formula, what alternative do we have when also considering that growth
already starts faltering.
A.A. Kahn:
I wonder whether the milk insufficiency syndrome really does exist. or whether it is only present in the
minds of the mothers because we were unable to communicate with them in time to motivate them for
successful breast-feeding.
S.H. Pudjiadi:
In our experience there are definitely mothers who just cannot produce sufficient breast-milk although
they are highly motivated to do so and carefully followed all our advice to increase their milk flow.
M.S. Trastonenoyo:
Could I perhaps ask Dr. Kusin about her experiences on this SUbject?
IA. Kusin:
In our study we found that poor mothers in general produce breast-milk in quite sufficient quantities
119
from 0 to 12 months onwards. With the introduction of solid foods, breast-milk production was not
reduced because the mothers continued to breast-feed their infants about 12-20 times a day. However,
please bear in mind that this refers to a rural population; in urban populations mothers probably just do
not have the time or the opportunity to breast-feed their children so often.
M.A. Husaini:
About weaning food, there are, in my opinion, two important aspects. The first is to provide adequate
nutrients for growth and health, and the second is to learn and teach the child to eat more diversified
foods. This, in my experience also influences its eating habits later in adulthood. Since our govern-
ment policy to improve the nutritional status focuses on better diversification, I think that we should
already start at the weaning age.
SR. Pudjiadi:
It is rather difficult to change the habits of the population but I think by teaching them, we can achieve
the position that a number of less favorable habits will be changed. For this reason we give much
advice to the mothers as to what kinds of locally available ingredients, cheap but good, they can use to
prepare nutritionally adequate, diversified food for their children.
C. Gopalan:
I would like to counteract somewhat on the general impression this paper made on me. Although the
presence of numerous commercial infant and weaning foods in our region cannot be overlooked, I
think they simply cannot playa relevant role in solving the main problem we are facing and that is
malnutrition. We have been trying for the last 2 or 3 decades to popularise and to advise poor
populations to use the foods which are available within a radius of 3-4 kilometers from their own
homes and how to fashion nutritious recipies out of those foods, which are within their economic as
well as their geographical reach.
Paper by J.J. Counsilman

C.-y. Yeung:
In Hong Kong we have a very similar situation as with the Chinese in Singapore. I am aware of two
surveys to investigate the reasons for this. The first, questioning mothers of the University teaching
staff, revealed that social freedom was the most heard reason for not breast-feeding. The second
survey, performed in a less privileged group of the community, revealed embarrassment in public as
the most imminent reason. In Hong Kong many families live in small appartments (3-400 square feet)
housing 6 or more persons, of which quite often 2 are grand-parents. Another important reason
mentioned was working requirements. Although traditionally mothers are not allowed to go to work
before the end of the first month, the situation appears to be changing as some mothers start working
earlier, most likely because of financial reasons.
JJ. Counsilman:
Let me comment on social freedom. In Singapore this is exactly the same; mothers indicate the
importance of having the freedom to leave the baby with other relatives. Concerning embarrassment, I
do not know what can be done about that. Perhaps one could encourage the mother to use a light cover
over the baby when it is being breast-fed. If lactation has already started, the mother could express the
milk into a bottle.
D.Sinniah:
First, I would like to state that the observations made about Malaysian breast-feeding practices in
Singapore are very similar to those in Malaysia. Secondly, I would like to mention a survey on breast-
feeding practices among nurses and doctors in Malaysia. We found that although the incidence of
breast-feeding soon after birth was highest with doctors, this fell within one month to very low levels,
lower than the national average. The first of the reasons given was work, since maternity leave is only
about 6 weeks. The women who did not elect to breast-feed felt that they were not convinced that
120
breast-feeding was best. They felt that they could feed their babies satisfactOrily by preparing the food
hygienically and without difficulty. Obviously, doctors themselves need to be re-educated!
E. Suroto-Hamzah:
I think that it would be useful in this respect to make a distinction between two definitions: the
initiation and the willingness to breast-feed, respectively. I think that willingness is influenced by
cultural opinions, while initiation is influenced by hospital policies. Could you therefore comment on
the general hospital pOlicies in Singapore with respect to the efforts which are made to persuade the
mothers to breast-feed their infants.
J J. Counsilman:
It used to be the practice that the mother and baby were separated during the first 24 hours, the
traditional Chinese rest period. In 1986 the situation was that 58% of the mothers did not even touch
their babies within 24 hours. In the new University Hospital, this has now changed quite drastically
and a considerable part of the staff now strongly encourages the mothers to breast-feed. Some doctors
even put the babies to the breast right away. This situation mainly applies to the University hospital
which is mostly used by the well-to-do Chinese popUlation. I arn confident, however, that this
changing attitude will be adopted all over Singapore.
M.GM. Rowland:
I think we have hit some fairly positive notes which we should not lose. Once lactation is establiShed,
and I suggest that this is a process which is completed in the first few weeks of life, it is in fact
difficult to stop. Therefore, even in countries where women are going out to work, where there are
constraints during the day, once a woman has established lactation it is possible to maintain it. This is
true whether or not there is initiation of weaning with semi-solid or solid food. The important thing for
the pediatric and the allied professions is the establishment of lactation. This is the most fragile period
and here we should make sure that we get it right.
J J. Counsilman:
I completely agree with you, you just cannot overestimate the importance of the role health workers
have in stimulating the initiation of lactation. For instance, in Singapore it was found in one study that
79% of the women did not know that lactation may take some time to become established. In another
study it was found that around 80% of the mothers did not know that there exists a difference between
colostrum and mature milk.
Paper by M. Gracey and H. Sullivan

D.Karyadi:
I would like to know whether you have included in your study approach some anthropologists or
sociologists because I think that intervention studies based on an anthropological approach might
result in a completely different outcome.
M.Gracey:
You are right and as a matter of fact we did include an anthropologist in our studies. The perspective
of the Aborigines with respect to health and well-being are indeed totally different to our own ones.
For many years we have tried to get health messages across in a way we felt was appropriate, but all
the time we used the wrong words, words which to them were meaningless; even the word "health"
has no particular meaning to these people. I think that there are many cross-cultural differences and
messages that we have to learn before we can really manage to improve the prospects of health, well-
being and longevity in these people.
A.A. Kahn:
I arn interested to know whether you investigated to what extent the people in your study could
understand the simple health messages you gave them.
121
M. Gracey:
The question concerning the actual status of health knowledge with the Aborigines is a very important
one. Nowadays, education is provided by the Australian government, even in the most remote areas, in
their own language next to English and in an appropriate way. Many young Aborigines that are now
coming out of school are really quite well-educated in a theoretical sense. But theoretical knowledge is
not the same as attitudes and practices and I think that this is the hard core where we have to try to
change behavior before we will see the improvements in health we wish to accomplish.
J J. Counsilman:
Could you please speculate on the levels and types of morbidity in your group now compared to that
before settlement times.
M.Gracey:
Knowledge of Australian Aborigines before contact with Europeans is obviously very limited. From
skeletal remains and so forth, we know that they had excellent teeth and also became old enough to
develop osteoarthritis. From the fact that many of the large collections of skeletons contain large
proportions of children'S skeletons we may deduce that infant mortality and morbidity were probably
really quite high. After European settlement in 1788, the health status of the Aborigines can simply be
described as poor, dominated by infectious diseases such as sexually transmitted diseases (syphilis),
respiratory infections (tuberculosis) and gastrointestinal infections (typhoid fever). After the
referendum in 1967, the Aborigines became stationary in great numbers. They clustered on the fringes
of towns in groups that previously would not have lived together. The living conditions were very
poor, very poor hygiene and the real environment for the transmission of infectious diseases. The main
causes of morbidity in our Aboriginal children were respiratory infections, gastroenteritis, failure to
thrive, malnutrition, child abandonment, social problems and what generally could be called child
abuse, often associated with parental alcohol abuse.
D.R. Karunaratne:
I would like to ask you whether you have followed your group of infants for even longer than 2 years.
I wonder if catch-up growth would be possible once the children become adequately nourished so that
at the age of 16 or 18 they might have caught up with those who did not suffer from faltering between
6 months and 2 years.
M.Gracey:
We followed one group for 5 years and they showed no catch-up growth whatsoever. We have also
analysed cross-sectional data up to the age of 16 years from which it can be suggested that the
nutritional insult that occurs in early life in these infants may be permanent in many of them.

M.G.M. Rowland:
In many of the papers presented until now, the growth of populations has been presented in relation to
the NCHS standards making the point that we should have one international standard and not double
standards. I fully endorse that. A number of speakers simply mentioned that during the early part of
infancy growth is almost identical. If we observe these curves in detail, for instance those presented by
Dr. Gracey, we see that during the early months in infancy growth really is explosive. After the first
few months curves flatten off, possibly still following centile lines but almost permanently depressed.
My question to the panel is how do we identify growth faltering when it first occurs, and when we
might need to intervene on an individual basis.
M. Gracey:
The obtained growth charts, whether the Harvard standard, the NCHS references or local reference
values are used, all have limitations. Because of the extreme difficulties in the normal range of
variability in growth patterns. I think that for the individual child. growth velocity is much more
important than attained growth.
122
I.A.Kusin:
According to our experiences it will be extremely difficult if you really want to follow an individual
child. If you have individual data. the intake data often do not go along with the growth rates. Also
there is sometimes a dissociation in weight faltering and length faltering which cannot be explained;
why do they still grow well in weight when they start faltering in length?
L. Mata:
Personally, I am not convinced that we can really intervene in this growth faltering in the tropics. I
think that the problem with this faltering is not because of a shortage of food but because children are
not hungry. This anorexia is in my opinion closely related to the concept of infection. Therefore, I
think that the main thing we could do to prevent growth faltering is to provide an environment in
which the children will have less infections.
I.A. Kusin:
In the villages where we performed our study all the fully breast-fed infants after 6 months have
returned to the breast because of anorexia, which is 30% of the total population. Some mothers
provide them with food but they refuse and only want to suckle. So we observed a higher percentage
of exclusive breast-feeding at the end of infancy than at an earlier phase caused by a very persistent
anorexia.
DR. Karunaratne:
One of the things that discourages mothers to breast-feed is that they are working or are employed. In
Sri Lanka this problem has been overcome by the government insisting that mothers obtain 3 months
paid maternity leave, both in the government and in the private sector. I think if this example would be
followed in, for instance, Singapore and/or Hong Kong, mothers should be able to feed at least for 3 to
4 months.
M.Gracey:
I appreciate that situation and I believe that it is very important in those countries you just mentioned.
For the population we studied, however, the situation is completely different. The mothers do not have
to work, they get money from the government anyway.
l.A. Kusin:
We should remember that 80% of the mothers do perform productive work but are not paid.
Traditionally, they return after 6 weeks, the culturally sacred period, to work. Although we have been
advising mothers for 8 years not to leave their child at home after those six weeks to work in the fields,
we have not succeeded yet. To me this means that there is simply a need to work culturally and that
lactating women in that culture obtain no protection.
For a change I would also like to raise a question. Dr Mata, can you explain why in Guatemala the
incidence of low birth-weight is 42% while the dietary intake of pregnant women is higher than, lets
say, the Gambia or our studies.
L.Mata:
We really do not know the answer, but I think one possible explanation is methodological. To obtain
reliable figures you must cover the field completely, every single birth in a village, and you have to
include also the infants which die in the first few minutes or hours after birth. Therefore, I think that
the incidence of low birth-weight infants in parts of Africa and India would also be around 40% if the
proper methodology is followed.
R.E. Eec1(eis:
If we use the cut off point of 2.5 kg for low birth weight infants, we might run a danger of overestimat-
ing the relevance of the figures of incidence on prevalence. I think that for the countries we are dealing
with here, the definition for low birth weight as is used in the western countries is not the most
appropriate one. I would pragmatically define this point as the level from which, if you go further
down, mortality will increase very sizably and very suddenly. According to this definition the point
would definitely not be 2.5 kg, but much lower - up to approximately 2. I kg.
Session 3
Early childhood malnutrition and its late effects,

with emphasis on prevention
Chairman: LG.N. Gde Ranuh

Selective review of studies on the behavioural effects of
childhood malnutrition
c.A. POWELL and S.M. GRANTHAM-McGREGOR
Introduction
Protein-energy malnutrition is a somewhat misleading name for childhood malnutri-

tion. Other nutrient deficiencies besides protein and energy, such as trace elements
and vitamins, are frequently present [1]. Therefore in this paper we shall omit the
prefix 'protein-energy' and just refer to 'childhood' malnutrition. The criteria
commonly used for the diagnosis of malnutrition depends on weight deficits and the
presence of oedema [2]. This fails to distinguish between severity and duration, and
includes children who have been mildly undernourished for long periods of time
beginning from gestation, and those who have suffered a more short severe episode.
Also children who have suffered from repeated bouts of infection are combined with
others who have been predominantly affected by insufficient nutrient intake. It is not
clear to what extent these factors may affect behaviour differently. The social context
in which malnutrition occurs as well as the stage of development of the child also
modify the impact of malnutrition.
In 1963 Stoch and Smythe [3] reported the results of the first longitudinal study of
survivors of severe malnutrition and their mental development. A large number of
studies have subsequently been carried out and it has become clear that the relation-
ship between malnutrition and poor mental development is complex. In this chapter,
we will give a selective review with some emphasis on our research in Jamaica.
Transient effects
It is helpful to divide the behavioural effects of undernutrition into transient and long
term ones. Transient ones may be due to metabolic changes and are present only as
long as the insult is present. With the removal of the insult the behaviour returns to
normal. Torun and Viteri [4] showed in a small group of children that when energy
intake was reduced their energy expenditure was rapidly reduced. Unfortunately they
did not report what happened when normal energy intake was restored. They
H.K.A. Visser andJ.G. Bindels (eds.J Child Nutrition in South East Asia, 125-140.
126
100
>-
......
J- 60
>
......
J-
w
<
40
Adm. 1wk. Dis.
Figure 1. Means and standard errors of activity ratings observed on admission to hospital, one week later
and when ready for discharge. ~ =18 severely malnourished children, D =14 adequately nourished
children.
50
til
l!J
Z
j:: 40
<C
0:::
:z
o
...... 30
~
0:::
o
.....J
a..
X 20
IJ.J
Adm. Dis.
Figure 2. Means and standard errors of exploration ratings observed on admission to hospital, one week
later and when ready for discharge. ~ = 17 severely malnourished children, D =14 adequately
nourished children.
127
attributed the reduction in energy expenditure to reduced activity.

The behaviour of severely malnourished children in the acute stage is another
example of transient behavioural changes. In Jamaica, we studied 18 severely
malnourished children while they recovered in hospital and compared them with 14
adequately nourished children in the same hospital who were suffering from other
diseases [5]. The children were observed using time scheduled observations, once a
week. They were observed alone in their cribs (unstructured) for three ten minute
sessions, and when presented with a standard set of toys (structured) for 15 minutes.
On admission to hospital, the malnourished children were much less active than
the adequately nourished children in the unstructured observations. This difference
disappeared one week later (Figure 1). The malnourished children were also
significantly more apathetic than the adequately nourished group, who were in tum
more distressed. These differences had also disappeared one week later.
In the structured observations the malnourished children explored significantly
less than the adequately nourished group (Figure 2), and handled fewer toys. The
different types of exploratory behaviour were also observed such as pushing, pulling,
banging, throwing. Although these differences in exploration were reduced one week
later the groups were still significantly different. However, by the time the children
left hospital all differences had disappeared except that the malnourished group still
handled significantly fewer toys.
The changes shown in exploration, activity and affect were obviously transient. It
is unclear which nutrient deficiencies or metabolic abnormalities were responsible
for these changes.
Transient changes in cognitive function also occur when adequately nourished
children miss breakfast [6, 7]. In Jamaica, we have recently demonstrated that
transient behavioural changes due to short term food deprivation vary, depending on
the underlying nutritional status of the children [8]. In this study we examined the
effects of missing breakfast on the cognitive functions of three groups of nine and
ten year old children; stunted (low height-for-age), non-stunted controls and
previously severely malnourished. Both malnourished groups had lower scores in
tests of fluency (a measure of generation of ideas) and coding, whereas the control
groups had higher scores in arithmetic and efficiency of problem solving when they
missed breakfast. The children were then divided into wasted (low weight-for-
height) and non-wasted. Wasted children were adversely affected in digit span
backwards tests and the wasted members of the two malnourished groups were also
adversely affected in the efficiency of problem solving.
The malnourished children whether stunted, wasted or previously severely
malnourished demonstrated different transient behavioural changes from those
shown by the non-stunted children.
This phenomenon of an interaction between current food deprivation and the
nutritional status of the children needs further investigation. It is possible that other
nutrient deficits such as iron and iodine may also interact with nutritional status.
128
Further, the possibility of interactions between two or more concurrent deficiencies

should be considered.
Long term effects
Mechanism
The mechanism linking malnutrition with long term effects is not established. There
are several hypotheses. The deficiencies causing transient behavioural changes, such
as reduced exploration and poor attention, may persist for long periods or occur
repeatedly. A substantial developmental lag may then accrue, secondary to the
transient changes. The developmental lag thus sustained may be difficult or impos-
sible to correct.
Undernourished children's apathetic behaviour may induce reciprocal behavioural
changes in their caretakers. Chavez and Martinez [9] showed in a Mexican village,
that the parents of children who received nutritional supplementation were more
stimulating towards them, than parents of unsupplemented children. The caretakers'
behaviour could therefore contribute to the children's poor development.
There is also the possibility that permanent neuroanatomical and metabolic
changes to the brain, are responsible for the children's poor development. Rats which
are malnourished in early life and then rehabilitated have persistent but subtle
changes in their brains [to]. However, there is no evidence directly linking these
changes to behaviour.
It is possible that there is an interaction between malnutrition and social back-
ground. Pollitt [11] has suggested that malnourished children behave similarly to
small birth weight ones. These children develop poorly in deprived homes but show
little disadvantage in better homes. Studies in developed countries of children who
were malnourished secondary to other diseases, are often cited as evidence for an
interaction effect. These children have failed to show any marked disadvantage in
their long term mental development. However, there are many problems in interpret-
ing these studies [12]. The most important is that the children have rarely been as
severely malnourished as those in poor countries. Further, the duration of malnutri-
tion was usually relatively short. In contrast, malnourished children in developing
countries usually suffer from inadequate nutrition for many years following the acute
episode.
Studies of long term effects
Childhood malnutrition is usually found in situations of extreme poverty. Some of

the characteristics of poverty, such as unemployment [13], low levels of maternal
education, and poor housing [14] themselves have a detrimental effect on child
129
development. Therefore even without severe malnutrition a child from a poor

environment is at risk of poor intellectual development.
The main problem with studies of malnutrition and mental development, is
separating the effects of poor social background from those of malnutrition. The very
early studies in this area failed to control for social background. In later studies,
however, various strategies were used. Index children have been compared with their
siblings or controls matched for social background.
Studies of children with a history of early severe malnutrition have been reviewed
in detail elsewhere [15-17]. In this paper, we will only give a brief summary. Only
those studies in which sufficient details are given to allow evaluation and the groups
were reasonably well-matched, will be mentioned.
Sibling studies
A list of the studies in which survivors of severe malnutrition were compared with
their siblings and in some cases also with matched controls, is given in Table 1. The
children were compared on a number of measures which included tests of IQ, and
specific cognitive functions, school achievement and behaviour.
In five studies measures of intelligence were used. In the studies from Mexico
[18, 19], Jamaica [20, 21], Lebanon [22] and Nigeria [23] the previously mal-
nourished children performed significantly worse than their siblings. In these studies
the WISC or Stanford-Binet Tests were used. The South African study [24] was the
only one in which intelligence tests were used and no difference was found between
the groups. Even in this study the index children had lower scores on a drawing test.
In the Jamaican study, although the children had lower scores on the WISC, there
was no difference between them and their siblings in school achievement.
In three other studies school grades were used as the outcome measure. Pereira
and colleagues [25] found significantly worse school grades in the index children
whereas in the other two studies [26, 27] there was no difference between the two
groups of children. In another South African study there was no difference between
the index children and their siblings on grip strength, motor development and
coordination [28].
Studies with matched controls

Matched controls were used in addition to siblings in four of the studies in Table I.
Controls were also used in a further five studies (Table 2). The previously mal-
nourished children performed significantly worse than the controls in eight out of the
nine studies summarised. These were the studies from India [29], Mexico [19],
Uganda [30], Barbados [31, 32], Jamaica [21, 33], Nigeria [23] and Lebanon [22].
There was no difference between the index children and their matched yardmates in
the study from South Africa [28].
t ..l
o
-
Table 1. Studies of survivors of severe malnutrition compared with siblings and matched controls
Reference Subjects TypeofPEM Siblings or comparisons Results of index group
Birch et al., 1971 [18] 37, aged 5-13 yrs. Kwashiorkor 37 siblings Lower scores on
Cravioto & Delicardie nearest in age verbal, performance &
1975[19] fullscale of WISC.
(Mexico) Poorer on inter-
sensory integration
Evans et aI., 1971 [24] 40, aged 9-15 yrs. Kwashiorkor 40 siblings No difference on New
(S. Africa) S. African Scales.
Lower on Goodenough
Drawing Test
Hertzig et aI., 1972 [20] 74 males, aged Mixed, severe 38 male siblings Lower scores on
Richardson et aI., 1973 [21] 5-lOyrs. PEM 6-12 yrs. verbal & fullscale
(Jamaica) of WISC, no difference
in school achievement
74 classmates Lower scores on verbal
or yardmates performance & fullscale
of WISC. Lower school
grades
McLaren et al., 1973 [22] 14, S stimulated Marasmus 14 sibs. of S S. Us & U all had
(Lebanon) in hospital 15 sibs. of Us lower scores on
15 Us not stimulated 15 sibs. ofU Stanford Binet than
15 U moderate under- 15 control their sibs. All scored
nutrition 3-5 yrs. 10werthanC
Table 1. (continued)
Reference Subjects TypeofPEM Siblings or comparisons Results of index group
Nwuga, 1977 [23] 52, aged 9-10 yrs. Kwashiorkor 34 siblings Lower scores in WISe:
(Nigeria) visual memory, picture
arrangement, block design
and Ravens Scale.
32 classmates Lower in all of the above
& shape discrimination.
38 upper class children Upper class group had
higher scores in all subtests.
Bartel et aI., 1978 [28] 31, aged 6-12 yrs. Kwashiorkor 31 siblings No differences in grip strength,
(S. Africa) or marasmic- motor development, finger
kwashiorkor tapping and motor speed.
31 yardmates No difference in above
(except 2 of 93 items).
No effect of age of
hospital admission.
Graham & Adrianzen, 110,6-12 yrs. Marasmus or 188,6-12 yrs. No difference in school
1979 [26] marasmic- siblings grade.
(Peru) kwashiorkor
Pereira et ai., 1979 [25] 130, 6-12 yrs. Kwashiorkor 88 siblings Significantly worse
(India) or marasmic- school grades.
kwashiorkor
Moodie et al., 1980 [27] 116, 15 yrs. Kwashiorkor 89 siblings No difference in school grades,
(S. Africa) after acute or employment and
episode social adjustment.
....,
-
Table 2. Studies of survivors of severe malnutrition compared with matched controls
w
N
Reference Subjects TypeofPEM Comparisons Results of index group
Champakam et al., 19,8-11 yrs. Kwashiorkor 53 classmates matched for age, Lower scores on tests of
1968 [29] religion, caste, SES, family size neurosensory integration and
(India) birth order, parental education intelligence
Lowest in tests of perceptual and

abstract abilities
Cravioto & Delicardie, 19, from birth to 58 mths. Mixed, severe malnutrition 19 same cohort, matched for birth Fewer bipolar concepts
1975 [19] weight, length and gestation
(Mexico)
Hoorweg & Stanfield, 60,11-12 yrs. Mixed, severe malnutrition 20 clinic attenders, matched Lower scores in tests of motor
1976 [30] for age, sex and SES.
(Uganda) functions, block design, memory
for design, Ravens matrices,
and incidental learning. No
difference in arithmetic vocabulary,
Proteus mazes, knox cubes
Galler, 1984 [31] 129,5-11 yrs. Underweight or marasmus Classmates, yardmates Lower scores in verbal perfonnance
(Barbados) matched for age and full scale of WISC, poorer
school grades more neurological
soft signs
Galler, 1986 [32] 9-15 yrs. Same Same No catch up in WISC

(Barbados)
Grantham-McGregor et aI. 17,7-8 yrs. Mixed, severe malnutrition 18 children in hospital Lower scores on Griffiths and
1987 [33] matched for age and SES Stanford Binet
(Jamaica)
133
Comments on studies
Differences between the index and comparison groups were more readily
demonstrated with cognitive function or IQ measures than with school achievement
measures. School achievement reflects more than the children's aptitudes. Parental
attitudes and the amount and quality of teaching are also important. In general
differences between index and control children were more often found with matched
controls than with siblings. This is not surprising because Cravioto and Delicardie
[34] showed that although groups may be matched on macro-environmental
measures marked differences may exist in the micro-environment. Richardson [35]
maintains that it is impossible to match for all socio-cultural factors which may
affect development.
There are also problems with sibling studies. Although siblings are probably the
best controls for the home environment, there may be differences in intra-family
relationships which result in different treatment to some children. More importantly,
siblings are likely to have been undernourished themselves. It is possible that the
only difference in their nutritional history is that they did not suffer from a severe
episode of malnutrition. The relative importance of duration and severity of malnutri-
tion, when considering mental development is unknown.
In order to explore this concept, we have compared the developmental levels of
children who had recently suffered from severe malnutrition with those of children
who were similarly stunted but had not suffered from a severe episode. Stunting was
used as a proxy for duration of malnutrition [36].
The developmental levels of the previously severely malnourished children were
only slightly and not significantly lower than those of the group matched for
stunting. Both groups were significantly lower than a third group of children of
normal height-for-age, from similar social backgrounds [37].
The implications of these findings are that the poor developmental levels usually
found in severely malnourished children could be due mainly to factors associated
with stunting rather than to the episode of severe malnutrition. Further, the lack of
difference found between sibling and index children in some studies may be
explained by both groups having a similar duration of malnutrition.
Although there is no consensus between the studies, on severe malnutrition, it is
reasonable to conclude that an episode of severe malnutrition in early childhood
superimposed on conditions of poor socio-economic backgrounds is associated with
an increased risk of long term impaired intellectual function and behaviour disorders.
There are no consistent findings of specific cognitive deficits. In several studies the
children were found to have behavioural differences resembling those of children
with attention deficit disorders [38].
Rehabilitation
It is possible that the poor development associated with severe malnutrition can be
ameliorated by providing enriched environments. Malnourished Korean girls who
134
were adopted by North American families at an early age were found to have IQs
and school achievement levels comparable to those of the North American popula-
tion several years later [39].
Monckeberg also reported dramatic improvement in the IQs of Chilean children
recovering from severe malnutrition who were adopted [40].
There are only a few reports on the psychological rehabilitation of severely
malnourished children [41]. In Lebanon [42] and in Mexico [43] malnourished
children were provided with increased stimulation while in hospital. The children
showed marked improvements in development compared to children who did not
receive stimulation. However, these interventions lasted only for the duration of the
children's hospital stay and the gains in development were no longer present a year
later.
The Jamaican study
The only report of the effects of long term stimulation on the mental development of
severely malnourished children is one which we have just completed in Jamaica [33].
A group of severely malnourished children who received increased stimulation while
in hospital and for three years after discharge was compared to a group of mal-
nourished children who did not receive stimulation and a group of adequately
nourished children who were in hospital for reasons other than malnutrition.
Both malnourished groups had similar low levels of development on admission
which were significantly below that of the adequately nourished group. The group
which did not receive intervention (non-intervened malnourished) continued to show
GRADE
LEVEL
o
READING SPELLING
Figure 3. Means and standard errors of reading and spelling grades six years after intervention ceased.
~ = 16 non·intervened malnourished children, []l = 17 intervened malnourished children, 0 =20
adequately nourished children.
135
GRADE
LEVEL 2
o
ARITHMETIC WRAT
Figure 4. Means and standard errors of arithmetic and WRAT grades six years after intervention ceased.
~ = 16 non-intervened malnourished children, I] = 17 intervened malnourished children, 0 =20
adequately nourished children.
a large deficit compared to the adequately nourished group throughout the study and
showed no signs of catching-up. The intervened-malnourished group showed marked
improvements in development. They actually caught up to the adequately nourished
group 2 years after leaving hospital [44]. Three years after intervention ended they
remained with significantly higher developmental levels than the non-intervened
malnourished group but had slightly lower levels than the adequately nourished
group.
We have recently re-examined these children 6 years after intervention ceased,
when they were 10 to 11 years of age. School achievement levels were assessed with
the Wide Range Achievement Test [45] (Figures 3 and 4). The reading and spelling
grades of the intervened malnourished children were significantly better than those
of the non-intervened malnourished and were not different from the adequately
nourished group. The difference between the two malnourished groups in arithmetic
did not reach significant levels and both groups had significantly lower scores than
the adequately nourished group. Although the intervened malnourished children had
lower global scores than the adequately nourished group (p < 0.05), they had
considerably better scores than the non-intervened malnourished group (p <0.002).
The maintenance of benefits up to six years after the intervention ended is
encouraging. It is important therefore that psychosocial stimulation is included in the
rehabilitation of severely malnourished children.
It is unknown at this time whether severely malnourished children can reach their
potential when exposed to enriched environments. It is clear however that impressive
improvements can occur.
136
Mild to moderate undernutrition
Mild to. mo.derate undernutritio.n affects many mo.re children than severe malnutri-
tio.n. Asso.ciatio.ns between mild to. mo.derate undernutritio.n and lo.w levels o.f IQ and
scho.o.l achievement have been repo.rted in many studies [46, 13]. Ho.wever, as in the
case o.f the studies Qf severe malnutritio.n, po.o.r so.cial backgrQund cQnfo.unds the
establishment o.f a causal relatio.nship. Unlike severe malnutritio.n, it is po.ssible to.
cQnduct experimental studies.
Several nutritio.nal supplementatiQn studies have been carried Qut in under-
nQurished PQPulatiQns. MQstly supplementatiQn Qf the mQthers was started frQm
pregnancy and cQntinued with the children after birth. In a few studies children who.
were already undernQurished have been supplemented. Rush [12] has reviewed these
studies in detail. In this paper we will review Qnly studies from develQping cQuntries.
Preventative studies
In Taiwan [47] a clinical trial o.f supplementatiQn was carried Qut with rando.m
assignment to. treatment Qr placebo. grQups. High risk mQthers were supplemented
thrQugho.ut pregnancy and lactatiQn, but the children were nQt. At eight mo.nths Qf
age the children had significantly higher SCQres Qn the mQtQr scales Qf the Bayley
Scales Qf Infant DevelQpment, than thQse whQse mQthers were nQt supplemented.
Their perfQrmance Qn the mental scales was no.t affected.
In a study in BQgQta [48,49], high risk pregnant WQmen were randQmly assigned
to. six treatment groups. FQur will be cQnsidered here; stimulatio.n alo.ne, supplementa-
tiQn alone, supplementatio.n with stimulatio.n o.r no. treatment. Supplementatio.n and
stimulatio.n had small, separate benefits o.n co.gnitive abilities in the children at 18
mo.nths and three years o.f age. The strongest effect was fQund in the supplemented
and stimulated group. There was a significant interactio.n between supplementatio.n
and stimulatio.n at 18 mo.nths but no.t at 36 mQnths. Overall the benefits were very
small.
In Guatemala, pregnant wo.men and children from fQur villages tQQk part in a
nutritio.nal supplementatio.n pro.ject. Two. villages were o.ffered a high protein
supplement and the o.ther two. a lo.w calo.rie supplement which were taken o.n a self-
selectio.n basis. Small co.ncurrent improvements in cQgnitive functiQns which were
related to. the amo.unts o.f supplement co.nsumed were fo.und while the supplement
was being given [50].
A much smaller study was co.nducted in a Mexican village [9]. Twenty mo.thers
were supplemented throughQut pregnancy and their children throughQut early
childho.o.d. The children were co.mpared with a gro.up o.f unsupplemented children
fro.m the same village who. were bo.m in the preceding year. The supplemented
children had higher sco.res o.n the Gessel Schedules between 6 and 60 mo.nths o.f age
than the un supplemented group. They alSo. sho.wed a remarkable increase in activity
levels and explo.ratio.n.
137
·5
·4
GRADE ·3
LEVEL
CHANGE
·2
o
SPELLING ARITHMETIC
Figure 5. Change in spelling and arithmetic grades after three months of intervention. 0 = 68 controls,
~ = 38 with breakfast.
Remedial studies
The effects of supplementing children who were already undernourished have only
been examined in two studies. In Bogota, Columbia [51] pairs of undernourished
children and their adequately nourished siblings were randomly assigned to either a
supplemented group or a control group. Their ages ranged from 4 to 60 months.
After one year of supplementation the undernourished children showed a small
improvement in developmental levels whereas the adequately nourished group did
not. Unfortunately only scanty details of this study are reported.
In Cali, Columbia [52] three year old children were randomly assigned to groups
of supplementation with stimulation or supplementation alone for varying periods of
time. The supplement alone had no effect on the children's mental development, but
the combined interventions had substantial benefits which were proportional to the
duration of the programme. It is not possible to determine whether supplement
improved the benefits of stimulation, because there was no stimulation alone group.
Supplementation in older children
Very few studies have been conducted in which older children were supplemented.
In Jamaica, we studied the effects of providing a school meal on the nutritional
status, school achievement and school attendance of a group of undernourished 12
year old children [53]. Breakfast was given to one class over a period of three
months. A second class was given a placebo low calorie drink. and a third class no
treatment.
There was no improvement in nutritional status. However there was a significant
138
improvement in arithmetic scores and the change in spelling approached significance

(Figure 5).
Comments on supplementation studies
There are many problems with the interpretation of these studies. Firstly, group
assignment was sometimes not random [9,50]. Secondly, in many studies there was
no attempt to control for the extra attention the supplemented children may have
received. The Taiwan study [47] was the only one to have a placebo. Further,
supplement produced relatively small increases in growth in all except the Mexican
study [9]. Beaton and Ghassemi [54] have documented the problems of obtaining a
growth spurt in children in the community. These studies were no exception.
In spite of these problems, the findings from the preventative studies are consis-
tent in that small concurrent benefits were obtained. Only the children in the
Mexican study had major developmental benefits and there were many other design
problems in this study making interpretation difficult.
There were only two studies with already undernourished children and the
findings are inconsistent. It is therefore uncertain whether supplementation produces
a benefit in these children.
It is conventionally thought that the behaviour of young children is more suscep-
tible to nutritional effects than that of older children. The finding of an improvement
in arithmetic in older Jamaican children [53] was provocative. We hypothesised that
the improvement may have been mediated through the alleviation of the transient
effects of short term food deprivation.
Long term effects
Very few reports are available on the possible long term effects of early childhood
supplementation. Supplemented children from the Bogota study [55] showed small
benefits in school readiness tests at seven years of age. Stimulated children from the
Cali study made more progress through the first three grades in school and had
higher IQ scores at nine years of age [56]. In Mexico, the supplemented children
performed better in the first grade, but it is not clear whether they were still being
supplemented [57].
Conclusions
Nutrient deficits can cause transient behavioural changes. In contrast severe malnutri-
tion in early childhood is usually associated with long term poor mental development
when the children return to poor environments. The duration of undernutrition may
be critical.
139
Nutritional supplementation in early childhood may prevent some of the poor

development found in mild to moderately undernourished children, for the duration
of supplementation.
Acknowledgement
The Jamaican studies were funded by The Overseas Development Agency and
Welcome Trust, U.K., The Ford Foundation, U.S.A. and the Nestle Foundation,
Switzerland.
References
1. Golden MH and Golden BE: In: G1adrke E, Heiman G, Lombeck I and Eckert I (eds)
Spurenelemente. Georg Thieme Verlag, Stuttgart and New York, 1985, p. 47.
2. Editorial: Lancet 2:302, 1970.
3. Stoch MB and Smythe PM: Arch Dis Child 38:546, 1963.
4. Toron B and Viteri FE: In: Toron B, Young VR and Rand WM (eds) Protein-energy requirements of
developing countries; evaluation of new data. United Nations University, Tokyo, Japan, 1981,
p.229.
5. Stewart ME, Schofield WN and Grantham-McGregor SM: In: Proceedings of the 23rd Common-
wealth Caribbean Medical Research Council, Barbados, 1978.
6. Pollitt E, Leibel R and Greenfield D: Am J Clin Nutr 34:1526, 1981.
7. Pollitt E, Lewis N, Garcia C and Shulman R: J Psychiatric Research 17:169, 1983.
8. Simeon D and Grantham-McGregor SM: Am J Clin Nutr 49:646, 1989.
9. Chavez A and Martinez C: In: Growing up in a developing community, INCAP, Guatemala City,
1982.
10. Bedi KS: In: Dobbing J (ed) Early nutrition and later achievement. Academic Press, London, 1987,
p. 1.
11. Pollitt E: In: Schurch B and Scrimshaw NS (eds) Chronic energy deficiency: consequences and
related issues. NestleS Foundation, Switzerland, 1987, p. 77.
12. Rush D: In: Galler J (ed) Nutrition and behaviour. Human nutrition: a comprehensive treatise.
Plenum Press, New York, 1984, p. 119.
13. Powell CA and Grantham-McGregor SM: Am J Clin Nutr 41: 1322, 1985.
14. Pollitt E: In: Poverty and malnutrition in Latin America. Early childhood intervention programs. A
report to the Ford Foundation. Praeger, New York, 1980.
15. Grantham-McGregor SM: In: Shepherd R (ed) Handbook of the psychophysiology of human eating.
John Wiley & Sons Ltd, England, 1989, p. 321.
16. Galler J: In: Nutrition and behaviour. Human nutrition: a comprehensive treatise, Plenum Press, New
York,N.Y., 1984,p.63.
17. Pollitt E and Thomson C: In: Wurtman R and Wurtman J (eds) Nutrition and the brain. Vol. 2.
Raven Press, New York, 1977, p. 261.
18. Birch HG, Pineiro C, Alcalde E, Toca T and Cravioto J: Pediatric Research 5:579,1971.
19. Cravioto J and Delicardie ER: In: Prescott M, Read and Coursin D (eds) Bmin function and
malnutrition, neuropsychological methods of assessment, John Wiley and Sons, England, 1975,
p.53.
20. Hertzig ME, Birch HG, Richardson SA and Tizard J: Pediatrics 49:814, 1972.
21. Richardson SA, Birch HG and Hertzig ME: Am J Ment Def 5:623, 1973.
22. McLaren DS, Yatkin US, Kanawati AA, Sabbagh Sand Kadi Z: J Ment Defic Res 17:273, 1973.
23. Nwuga UCB: Am J Clin Nutr 30:1423,1977.
24. Evans DE, Moodie AD and Hansen JD: S Afr Med J 25:1413,1971.
140
25. Pereira SM. Sundararaj R and Begum A: BrJ Nut 42:165,1979.
26. Graham GG and Adrianzen BT: In: Brozek J (ed) Behavioral effects of energy and protein deficits.
D.H.E.W. (NIH) Pub No 79-1906. Washington, 1979, p. 185.
27. Moodie AD, Bowie MD, Mann MD and Hansen JDL: S Afr Med J 58:677, 1980.
28. Bartel PR, Griesel RD, Burnett LS, Freiman I, Rosen EV and Geefuuysen J: S Afr Med J 53:360,
1978.
29. Champakan MA, Srikantia SG and Gopalan C: Am J Clin Nutr 21:344, 1968.
30. Hoorweg J and Stanfield JP: Develop Med Child NeuroI18:130, 1976.
31. Galler JR, Ramsey F, Solimano G: Pediatric Research 18:309,1984.
32. Galler JR, Ramsey F and Forde V: Nutr Behav 3:211,1986.
33. Grantham-McGregor SM, Schofield W and Powell C: Pediatrics 79:247,1987.
34. Cravioto J and Delicardie E: In: Nutrition, the nervous system and behavior. P.A.H.O. Sc Pub no
251, Washington, 1972. p. 73.
35. Richardson SA: In: Schulman I (ed) Advances in Pediatrics, vol 21. Yearbook Medical Publishers
Inc, Chicago, 1974, p. 167.
36. Waterlow JC and Rutishauser HE: In: Cravioto J, Hambreus L and Valhquist B (eds) Early
malnutrition and mental development. Stockholm, Sweden, 1974, p. 13.
37. Grantham-McGregor SM, Powell C and Fletcher P: Eur J Clin Nutr 43:403, 1989.
38. Galler JR, Ramsey F, Solimano G and Lowell WE: J Am Acad Child Psychiatry 22. 1: 16, 1983.
39. Winick M. Meyer KK and Harris R: Science 190:1173, 1975.
40. Monckeberg FB: Proceedings of the 19th Nestle Nutrition Workshop, Indonesia (in press).
41. Grantham-McGregor, SM: In: Brozek J and Schurch B (eds) Malnutrition and behavior: Critical
assessment of key issues. Nestle Foundation, Switzerland, 1984, p. 531.
42. Yatkin US and McLaren DS: J Ment Defic Res 14:25, 1970.
43. Cravioto J and Arrieta R: Lancet 2:899,1979.
44. Grantham-McGregor SM, Schofield W and Harris L: Pediatrics 72:239, 1983.
45. Jastak J and Bijou S: C L Story, Wilmington, Delaware, 1946.
46. Lasky RE, Klein RE, Yarbrough C, Engle PL, Lechtig A and Martorell R: Child Develop 52:220,
1981.
47. Joos SK, Pollitt E, Mueller WH and Albright DL: Child Develop 54:669, 1983.
48. Waber DP, Vuori-Christiansen L, Ortiz N, Clement JR, Christiansen NE, Mora JO, Reed RB and
Herrera MG: Am J Clin Nutr 34:807,1981.
49. Mora JO, Clement J, Christiansen N, Ortiz N, Vuori L and Wagner M: In: Brozek J (ed) Behavioral
effects of energy and protein deficits. DHEW NIH Pub No 79-1906, Washington. 1979, p. 255.
50. Freeman HE, Klein RE, Townsend JW and Lechtig A: Am J Pub Hlth 70:1277,1980.
51. MoraJO, Amezquita A, Castro L et al. World Rev Nut Dietet 19:205,1974.
52. McKay H, Sinisterra L, McKay A, Gomez H and Lloveda P: Science 200:270,1978.
53. Powell C, Grantham-McGregor SM and Elston M: Human Nutrition: Clinical Nutrition 37C:381,
1983.
54. Beaton GH and Ghassemi G: Am J Clin Nutr 35 (suppl), 1982.
55. Herrera M and Super C: In: Report to the World Bank. Harvard School of Public Health, Cambridge,
Mass, 1983.
56. McKay A and McKay H: In: King K and Myers R (eds) Preventing school failure: The relationship
between preschool and primary education. International Development Research Center, Ottawa.
1983, p. 32.
57. Chavez A and Martinez C: In: Liss AR (ed) Nutrition in health and disease and international
development, New York, 1981, p. 393.
Protein-energy malnutrition in Papua New Guinea:
Its functional significance
PETER F. HEYWOOD and ALISON H. HEYWOOD
Introduction
In Papua New Guinea (PNG) protein-energy malnutrition has usually been assessed
by comparing attained growth with international growth standards, the most
commonly used of which have been the Harvard Standards [1]. Within health,
emphasis is on children under 5 years of age and those under 80% weight-for-age
(W/A) are defined as malnourished.
The best estimate of the nationwide prevalence of malnutrition is derived from the
1982/83 National Nutrition Survey (NNS) [2]. Overall, approximately 40% of
children under 5 years of age were less than 80% W/A on the Harvard standards.
There was, however, considerable spatial variation in prevalence. Children in the
middle altitude zone (600 to 1200 m) had much higher rates of malnutrition than
those in the lowlands (0--600 m) or in the highlands (above 1200 m). When con-
sidered on the basis of administrative districts, high prevalence rates were found in
three clusters - the four districts on the northern side of the country between the
Sepik River and the Torricelli Range; the five districts on the south eastern fringe of
the central highlands covering the Anga language group; and four districts in Milne
Bay Province.
Consistent with the results of previous smaller surveys, the NNS also showed that
highland children are, in general, shorter than lowland children. However, at any
given height, highland children are heavier than lowland children.
There have been many studies of the growth of young children in PNG. A similar
general pattern of growth with age has been seen in both cross-sectional [3,4] and
longitudinal [5] studies (see Figure 1). In comparison with international standards,
this pattern shows an increase in mean weight-for-length (W/L) in the first three
months of life followed by a rapid decrease during the rest of the first year to reach
its lowest point at approximately fifteen months of age, before rising again in the
fourth and fifth years of life. Length-for-age (L/A) falls gradually from birth across
the first five years of life. Consequently, W/A rises in the first three months before
falling rapidly to 12 months of age, after which there is a gradual decrease until five
142
median per cent of standard

110
105
100
95
90
85
80
75 W/L
70
+ LlA
-* W/A
65+----,----,----,----,----,----,----,----,----,---
o 6 12 18 24 30 36 42 48 54
age in months
Figure 1. Median per cent Weight-for-Age (W/A), length-for-age (L/A) and weight-for-Iength (WfL) by
age.
years, by which time approximately half the children are below 80% W/A.
In the 1970's there was considerable discussion in PNG as to the significance of
these levels of apparent growth retardation. Some argued that international growth
standards may not be appropriate to PNG and may lead to incorrect estimates of the
extent of malnutrition. In addition, others have argued that small body size represents
not a pathological response but a successful adaptation to environmental stress.
As such, it is argued, it is not malnutrition at all. A final reason for uncertainty
about the implications of the apparent growth retardation derives from failure to
separate the use of growth as a general index of health from its use as an index of
nutritional status. Where it is being used as an index of general health status it is
essentially retrospective in nature and summarises what has gone before. Where it is
used as an index of nutritional status, the implications are essentially prospective in
nature - its significance is mainly as a risk factor for various outcomes.
One approach to assessing the significance of the levels and patterns of growth
seen in PNG is to determine their impact on the function of the individual. A number
of studies in the Nutrition Research Program at the Papua New Guinea Institute of
Medical Research have attempted to do this by investigating the effect of apparent
growth retardation on a number of functional outcomes - namely morbidity,
mortality, eruption of deciduous teeth, motor development and visual attention. The
results of these studies are briefly reviewed in this paper.
143
mortality rote (per 1000/24 months)

275
250
225
200
175
150
125
100
75
W/L
50
-+- UA
25 -+ W/A
11010510095 90 85 80 75 70 65 60
% standard
Figure 2. Prospective mortality rate by nutritional status.
Results
Mortality
A prospective study of the relationship between nutritional status and mortality in

young children was carried out through the Tari Research Unit in Southern High-
lands Province. The majority of the population live at an altitude of approximately
1500 m. Demographic surveillance of the population has been maintained since the
early 1970's. Ages are known to at least the month, and usually the day, for in-
dividuals born since the mid 70's. Mortality is closely monitored by reporters. We
present here the relationship between nutritional status, as assessed by
anthropometry, and mortality [6].
Children were 6-29 months of age at the beginning of the study. Weight and
length were measured at the beginning of the study and compared to the Harvard
Growth Standards [1]. These results were used to assign each child to three cohorts,
one according to the level of W/A, one using L/A and another using WfL. Of the
1232 children on whom complete information is available, 71 died within two years
of the initial measurement.
For the analysis children were grouped into 10% intervals for W/A and WfL and
5% intervals for L/A. The results are shown in Figure 2. For W/A there is a gradual
increase in mortality as the level falls below 100%. Below 80%, mortality rises
144
Mortality Rate (per 1000/24months)
180
LJA> 95
160 +- LJA91-95
-+- LJA::;90
140
120
100
80
60
40
20
O+----------------------,----------------------~
< 90 91 - 100 >100
NIL (%)
Figure 3. Prospective mortality rate by length-for-age and weight-for-length (W/L).
sharply to reach 273/1000 per 24 months in children less than 60%. There is a
smooth rise in mortality as W/L falls below 90%, with a steep increase to 235/1000
per 24 months in children less than 80% at the beginning of the study. For L/A there
is a sharp rise in mortality to 258/1000 per 24 months for those less than 85%.
Figure 3 shows the joint effect of W /L and L/A. At a given L/A survival tended to
decrease with decreases in W/L, the effect being most marked in children less than
90% L/A. Similarly, at a given W/L, survival tends to decrease with L/A and the
effect is most marked in children whose W/L is less than 90%. Although these trends
are suggestive, the interaction between W/L, L/A and survival was not statistically
significant.
The relationship between nutritional status, as assessed by anthropometry and
mortality, has also been assessed in hospitalised children in Mexico [7], and in field
studies in India [8], Bangladesh [9, 10] and Zaire [11]. With the exception of the
study in Zaire, all have shown a decrease in survival as nutritional status falls.
In comparison to the other study in which all three indices of nutritional status
were assessed, that of Chen et al. in Bangladesh [10], the level of W/A and W/L at
which risk increases sharply is much higher in Papua New Guinea than in
Bangladesh. The reasons for this difference, which raises the question of whether the
level of an index at which risk of mortality rises is the same across population
groups, is unclear. The difference is, nevertheless, of relevance to Papua New
Guinea where lowland children are, in general, taller and thinner than those in the
145
highlands, where this study was conducted. However, it does seem reasonable to
assume that lower nutritional status is a risk factor for mortality in lowland children,
even though the level at which mortality rises sharply may be different.
Morbidity
The effects of nutrition on the most important diseases of young children in Papua
New Guinea, pneumonia, diarrhoea and malaria, on nutrition have been described by
Lehmann et al. [12].
Birthweight is a very significant risk factor for pneumonia in highland infants,
those less than 2.5 kg at birth having four times the risk of dying from an acute lower
respiratory infection in the first year compared to those born heavier (D. Lehmann
and P. Heywood, unpublished). Studies at Goroka Base Hospital in the highlands
indicate that malnourished children under 2 years of age have four times the chance
of being admitted with pneumonia and, when admitted, four times the chance of
dying of the disease [13].
With respect to diarrhoea, available data suggest that the incidence of diarrhoea in
Papua New Guinea is low compared to many developing countries [12]. Despite
these lower incidence rates, one study [14] found significantly higher incidence rates
in children less than 80% W/ A as compared to those above this level.
For malaria, there is no reliable evidence at the moment to indicate that children
with protein energy malnutrition are more likely to become infected than those who
are well-nourished [15].
Motor development
Several cross-sectional studies of the relationship between nutritional status and

motor development have been carried out in Papua New Guinea. In the first, this
relationship was examined in 457 children in a population in Madang Province
characterised by mild malnutrition. Nutritional status was determined by
anthropometry and comparison with the Harvard growth standards. Motor develop-
ment was assessed using the Bayley Scales of Infant Development [16]. The median
ages at which the major developmental milestones were achieved were similar to
western standards in the first 6 months. However, after this age, Madang children
began to lag behind, with the lag at the time of walking being 2-3 months. Further, at
any given age, children of higher nutritional status were more likely to have attained
a given developmental landmark than their malnourished peers (A. Heywood,
unpublished manuscript). Further studies in the Southern Highlands and East Sepik
Provinces yielded similar findings (A. Heywood, unpublished manuscripts).
Eruption of deciduous teeth
Studies in Simbu Province in the highlands indicate that as growth improves,

deciduous teeth erupt earlier [17]. Children under 5 years of age in 1980 were
146
significantly heavier than children of the same age measured by Bailey in 1961. In
addition, at any given age between 6 and 24 months children measured in 1980 had a
greater mean number of deciduous teeth.
Visual attention
There have been many studies which indicate a negative effect of malnutrition on
intellectual competence. It has been proposed that one effect of undernutrition is to
alter specific psychological processes, including visual attention.
This effect was studied in Madang Province in mildly malnourished children
using a visual habituation paradigm (A. Heywood, unpublished data). The children
studied were 15-26 months old and were tested on a routine visit to an MeH clinic.
Well-nourished children tended to have greater total fixation times. There was a
similar effect on mean length of fixation, well-nourished children having a sig-
nificantly greater attention span than those who were undernourished.
Discussion
An important reason for conducting these studies was to contribute to the debate over
the significance of apparent growth retardation in Papua New Guinea. One important
component of this debate was the appropriateness of international growth standards,
based on the growth of children in the United States of America, to the local
situation. One approach to the resolution of this issue would be to set local standards
based on the growth of children in economically privileged groups within the
country. However, large numbers of children are required to set reliable standards
and the number of children raised in such circumstances to date is too small. An
alternate approach was to assess the extent to which departure from international
growth standards was a factor in impaired function.
The studies briefly reviewed here provide clear evidence that the apparent growth
retardation has negative effects on function in young children - it is a risk factor for
mortality and morbidity for certain infectious diseases, is a factor in delayed eruption
of deciduous teeth and delayed motor development, and has a negative effect on
visual attention.
The health significance of the effects on morbidity and mortality is clear.
The significance of the effects on eruption of deciduous teeth lies in their indirect
effect on child feeding and, ultimately, on growth. Other work at the Papua New
Guinea Institute of Medical Research by Jenkins [18] indicates that both motor
development and the eruption of deciduous teeth are important cues for the introduc-
tion of supplementary food. At tbe same time, their emergence is delayed by growth
retardation. Thus, growth retardation itself may be an indirect factor in delayed
147
introduction of supplementary foods. To the extent that breast milk alone does not
meet the requirements of the young child for energy and protein, this delay in
introduction of supplementary foods may lead to sub-optimal nutrient intakes at a
time when episodes of illness are also becoming more frequent. The combined
effects of delayed motor development and reduced visual attention may be to reduce
the opportunities to explore the environment and to attend to it, thus reducing overall
opportunities to learn. This area, and particularly the combined effects of delayed
motor development and reduced visual attention on cognitive development, requires
further study.
Thus, in general, protein-energy malnutrition (growth retardation in relation to
international growth standards) in Papua New Guinea has important negative effects
on function. At the same time there is clear evidence that the growth of young
children in some areas of the country has improved in recent decades. In 1956
Venkatachalam [19] carried out a broad nutrition study in the highland province of
Simbu. His work included a cross-sectional study of growth of young children and
anthropometric measurements of adults. The study was repeated in 1981 by Harvey
and Heywood [17]. Comparison of the two studies showed marked improvements in
nutritional status in the intervening 25 years. Children 5 years of age were ap-
proximately 2 kg heavier and 6 cm longer in 1981 than in 1956. In 1981 women
were taller than in 1956 and there had been a significant improvement in birth weight.
These very marked improvements in growth occurred in association with general
social and economic development in the area which included basic health and
education programs and widespread participation in economic development based on
smallholder production of coffee.
An important issue remaining is that of the functional significance of regional
differences in growth. There are regional differences in infant and child mortality
which are not readily explained by differences in attained growth, the distribution of
health services, or levels of cash income. These issues deserve attention in future
nutrition and health research in Papua New Guinea.
Acknowledgement
The work reported here was carried out while the authors were, respectively, Deputy
Director and Research Officer in Nutrition at the Papua New Guinea Institute of
Medical Research.
References
1. Jelliffe DB: The assessment of the nutritional status of the community. World Health Organization
Monograph Series No. 53. World Health Organization, Geneva, 1966.
2. Heywood P, Singleton N and Ross J: Papua New Guinea Med J 31:91, 1988.
3. Heywood PF: J Hum EvoI12:133, 1983.
148
4. Bailey KV: J Trop Pediat 10:3.1964.

5. Orr-Ewing AK: A longitudinal study of infant growth in relation to breast milk consumption, total
food intake and morbidity. Ph D thesis, University of London, 1985.
6. Heywood PF: In: Taylor TG and Jenkins NK (eds) Proceedings of the Thirteenth International
Congress of Nutrition. John Libbey, London, 1986, p. 103.
7. Gomez F, Ramos Galvan R, Frank S, Cravioto Munoz J, Chavez R and Vazquez J: J Trop Pediat
2:77,1956.
8. Kielmann AA and McCord C: Lancet 1:1247, 1978.
9. Sommer A and Lowenstein MS: Amer J Clin Nutr 28:287,1975.
10. Chen LC. Chondhury AK and Huffman SL: Amer J Clin Nutr 33:1836. 1980.
II. The Kasongo Project Team: J Trop Pediat 29:69, 1983.
12. Lehmann D, Howard P and Heywood P: Papua New Guinea Med J 31:109.1988.
13. Barker J, Grauen M, Riley I. Lehmann D, Phillips p. Montgomery J. et al. J Inf Dis 159:348. 1989.
14. Binns CW: J Trop Pediat 22:9, 1976.
15. Heywood P and Harvey PWJ: Papua New Guinea Med J 29:45, 1986.
16. Bayley N: Manual for the Bayley Scales of infant development. The Psychological Corporation,
New York, 1969.
17. Harvey PWJ and Heywood PF: Nutrition and growth in Simbu. Research Report of the Simbu Land
Use Project, Volume IV. Papua New Guinea Institute of Applied Social & Economic Research, Port
Moresby, 1983.
18. Jenkins C: Papua New Guinea Med J 31:125,1988.
19. Venkatachalam PS: A study of the diet, nutrition and health of the people of the Chimbu area (New
Guinea Highlands). Department of Public Health Monograph No.4, Port Moresby, 1962.
Mother and child health at village level:
The effect of the mother and child card
A. ALISJAHBANA, R. PEETERS, M. MAERTENS, S. TANUWIDJAJA,

H. TJOKROHUSODO, W. NGANTUNG and D.A. PRlMANA
Introduction
The population of Indonesia will reach 200 million by the year 2000. Efforts are
being made to decrease the birth rate, infant mortality rate and maternal mortality
rate. One of the major problems is how to increase coverage of high risk groups
(mothers and children) by safe maternal and child health services. Although there
may be some variation in the statistics in different parts of the country, the level of
maternal and child health care is generally the same.
In Indonesia there are more Traditional Birth Attendants (TBAs) than doctors or
midwives. This makes TBAs the most accessible providers of maternal health care.
Even when trained midwives or doctors are available, women still prefer TBAs
because they are familiar with the rituals of pregnancy and birth. In some parts of the
country TBAs are males and literate, but in general TBAs are female, elderly and
illiterate. More than 80% of deliveries are still attended by TBAs, and more than
90% are home deliveries.
It is generally recognized that adequate antenatal care results in a better pregnancy
outcome. However, the uneducated and the poor receive the least antenatal care
(ANC) because it costs too much time and money. As a rule women are much more
likely to make use of antenatal care facilities when they live within 3 km of a health
centre. In rural villages antenatal care attendance is low, because TBAs are cus-
tomarily only consulted at a very late stage in pregnancy or even after delivery.
Pregnant women usually go to health centres only for confirmation of pregnancy,
particularly if it is the first pregnancy. Therefore, for most pregnancies no informa-
tion is available from the time of the first ANC visit until the time of delivery.
In many developing countries formal health care and health facilities are scarce
and do not cover the groups most at risk (mothers and babies). Therefore, the main
objective (besides increasing total coverage) is to find the best match for the patient's
needs using the existing health services and skills.
Virtually every component of primary health care is based on the assumption that
women, and particularly mothers, are the most effective frontline providers of health
H.K.A. Visser and 1. G. Bindels (eds.) Child Nutrition in South East Asia. 149-158.
150
care. This ability to carry out the task assigned to them in primary health
programmes is dependent on four equally important factors, namely: health technol-
ogy; knowledge; resources and time. Transmission of new technology, such as
immunization and Oral Rehydration Solution (ORS) has improved infant survival
rates [2].
The mother and child recording and action card
A possible tool for high risk screening and for the provision of maternal and child
care throughout the pregnancy and until the child is 5 years of age is the Mother and
Child Recording and Action Card (the M&C Card). The purpose of the card is to
identify low and high risk cases and start the appropriate action. The risk criteria on
the card are the result of an epidemiological survey of perinatal mortality and
morbidity and low birthweight in Ujung-Berung, an area adjacent to the current
survey area of Tanjungsari. A field test was carried out before the implementation of
the main survey. Risk factors which were not perceived as risks were not included in
the M&C card. These factors are mostly biological factors, such as the mother's age,
parity and birth interval. The reason for not including these risk factors was because
women had different perceptions of them. Other factors that have to be considered
are the capacity and facilities of the health centre. During the pilot survey it was
found that more than 50% of women in the village had at least one risk factor. Those
at risk had an average of 1.7 risk factors per woman. In order to balance the number
of referrals with the capacity of the health centre, risk factors associated with the
highest mortality were selected for intervention [3].
The basic principles of the 'recording & action cards' were taken from Essex et
al. [4] who designed an action-oriented card for the purpose of antenatal screening.
Because most of the women in the village are less well-educated and the TBAs who
were asked to fill in the card are mostly illiterate, a pictorial card was designed. This
was also one of the reasons for using single variables to identify risk factors (instead
of using risk scoring).
In the villages of West Java more than 80% of the pregnant women use the care
provided by the TBAs. Therefore the design of the M&C card is directed at, and
tailored to, the needs of the workers who will use them most. It is expected that the
M&C card will also solicit the active participation of the mother and the family in
promoting the concept of selfcare. Therefore the M&C card can also be considered a
Home Based Mother and Infant Recording Card kept by the mother.
In the Tanjungsari Intervention study, the card was modified and adapted to local
conditions. The condition of the mother and infant are shown and recorded through
pictures. This card is divided into four sections; the pregnancy history section, the
pregnancy itself, the delivery section and the postnatal condition of the infant until
the age of 5 years. This last section puts special emphasis on the four main causes of
infant death: diarrhoea, febris, whooping cough and convulsions (Figures 1 and 2).
FAKULTAS KEDOKTERAN CATETAN KAKANDUNGAN CATETAN NGALAHIRKEUN
UNIVERSITAS PADJADJARAN BEURAT Kg. TANGGAL NGALAHIRKEUN JAM:
BAN DUNG 1986 JANGKUNG Cm 1< 145)' CICINGNA OROK
LlNGKERAN PEUPEUTEUYAN CM 1< 22)"
KARTU IBU - ANAK

Kg 67 I I I ++--l-H-+-t-+i-
:1=..--- Sa.
TANGGAL KODE DUKUN: 641-
611-~ P.rhatosan~ 1 1 1 1 1 1 1 1U
KATERANGAN UMUM : 621-
NAMI 61r- --....... Dikintun· .... j..(?
60
UMUR 5!l ., ._.
58
ALAMAT
~ ~".
KATERANGAN KAKANDUNGAN :
;; ),0"
JUMLAH KALAHIRAN 53
52
JUMLAH BURUSUT TULUY 1 ~ 2") 51
JUMLAH LAHIR - HIRUP 50
49
JUMLAH BUDAK ANU MASIH HIRUP: 49
47
46
KALURON : 1 ~ 3°) 46 ~.
~ ~
KUNGSI OPERASI, NGALAHIRKEUN ANU
43 .;y"
SAADI 42
41
40
o MUHUN" o HENTEU :B OROK:
2131415161718 9 110 KAAYAAN OROK WAKTU KAKARA LAHIR
TANGGAL NGALAHI RKEUN ANU SAADI TANGGAL LAHIR :
BEURAT LAHIR gram
BULAN KARESEBAN ANU PANUNGTUNG:
DIKINTUN : ") PUSKESMAS/POSYANDU

.. ) RUMAH SAKIT
Ji
~ ----~-
DIRANCANG KU : @ ~ :;; ~ ~.M""":' ... I
- -I@-I@-
LABORATORIUM ILMU KESEHATAN ANAK
F. K. UNPAD ObeurewdDgenne
.....
IJt
Figure J. The Mother and Child Card (pregnancy and delivery infonnation).
~
Umur 3-4 tahun Umur 4-5 tahun ....
20 VI
N
CATETAN BEURAT BUDAI< 19 .....
i-""
N A M I : 1S
IMMUNISASI : 17
Umur 2-3 tahun
k'
16 16
dasar ulangan I,...-
SCG
1- Sa. Umur 1 - 2 tahun
15
,. I,...-
-
;
OTP
I I I I
.--- Perhatosan
" f-
0--+-4 ~ 1
POLIO
I I I I 13
CAMPAK_ I --...... Dikintun • I-
12
Umur 0-1 tahun
....
Kg 11
10 I,...-
keluarga berencana
kaping 9
- I-
~ 5
4
I!IIII: 3 1/
, 8910" 12131.,51617 181920 21 ~ 2'1 25 26 'l7 211 ~g JO , 323.M •• n •••• nu«q.o •••••••• 51 54 5t1 eo
,
., , , I
I
I II ,
~ MENCIIET ORALIT I
IC.AKIRANGAIliCAI
I I
UBARBATUK I
SAtUK
SESEI(IIIAFAS 1
fit I
UBARI>AIIIAS I I
l - - e : f PANAS 31>0£ 1
1
~ KEJANG-ICEJA",G
. . '--- - 1---"--1 I JJ :
Figure 2. The Mother and Child Card (child growth and health),
153
The problem action guidelines
To be able to provide health services to all pregnant women, guidelines were

designed for TBAs to identify risk factors and to carry out appropriate action. This
action includes treatment, health education and/or referral to higher levels of health
care. Preventive measures, such as iron tablets and tetanus toxoid, were also given.
'Problem-Action Guidelines' were developed. The original idea came from The John
Snow Public Health group for the purpose of extending effective services through
auxiliary health workers.
The flowchart is based on a logical algorithm to assist health workers to improve
the services they provide by helping to
1. classify the problems patients bring to the health worker,
2. logically display important questions and examination steps to be followed by
the health worker,
3. isolate the most appropriate actions and treatments for the problems, using a
limited set of available drug supplies,
4. encourage referral of difficult problems to the appropriate worker or physician,
5. provide a basis for supportive supervision and inservice training.
Alisjahbana et al. [6] and Peeters et al. [1] studied the characteristics of TBAs in two
rural areas of West Java. Based on the typical West Java TBA profile, the John
Snow Problem Action Guidelines [5] had to be changed and adapted to the local
level of knowledge and perception. These guidelines were modified into a pictorial
handbook for the TBAs. This pictorial booklet could also be used as a manual for
identification of at risk cases and for taking appropriate actions (Figure 3 gives an
example of it).
Results
This preliminary report is part of the 'The Tanjungsari Intervention Study' currently
being conducted in Tanjungsari, a subdistrict covering more than 80,000 people in a
rural area approximately 40 km from Bandung in West Java. The aim of the research
is to study the effect of an intervention based on the risk approach (identify risk-
groups and direct care to those in greatest need).
The research area is divided into two sections, the intervention area and the
control area. In the intervention area, training was given to all TBAs on how to
identify and measure risk as well as how to manage risk factors using the available
facilities and personnel. In the intervention area TBAs were trained in the ap-
propriate management of risk; while in the control area, no activities other than the
usual vertical government programmes were carried out.
In this paper, the preliminary results of one year's intervention - infants born
from the first of January until December 31, 1988 - will be presented. During this
154
Figure 3. Example from the pictorial booklet for TBA's to identify at risk pregnancies and to take
appropriate action.
period a total of 1044 births were followed from early pregnancy up to 7 days
postnatal.
The study group consisted of all pregnant mothers, living in the intervention area
of Tanjungsari.
A comparative situation evolved in the first year of the programme because the
M&C Cards were not distributed to all mothers as had been planned. This made it
possible to compare pregnant women with the card and those without. The assumpti-
on was that pregnant women holding an M&C card had received better ANC
(Table 1).
Pregnant women using the M&C card were significantly younger and conse-
quently had a lower parity: 61 % of the cardholders had been pregnant before. For the
group that did not use the card the proportion was 68%.
Table 1. Characteristics of women from both groups (singletons only)
Characteristics Women from intervention area
M&C card (+) M&Ccard (-)
Total number 666 378

Mean age 24.6 25.5 P < 0.026
Mean education 5.7 5.9NS
Delivered by TBA (%) 92.6 86.7 P<O.OI
155
Table 2. The relationship between maternal nutritional variables and LBW (singletons only)
Birthweight Nutritional variable

(grams)
Height MUAC Weight gain
< 145 em - ~ 145 em <22em - ~22em No Yes
<2500 19.1 13.8 34.7 13.5 22.0 13.5

~2500 80.9 86.2 65.3 86.5 78.0 86.5
x,z-value 1.92 17.28" 6.59
a significant p < 0.001. MUAC =Mid-Upper-Arm-Circumference.
Risk variables presented in this paper are those related to the nutritional status of
the women and were collected only from pregnant women who could produce an
M&C card. These nutritional variables are height, Mid-Upper-Arm-Circumference
(MUAC) and weight-gain during pregnancy. The sensitivity and specificity and cut-
off point for these variables were calculated from previous baseline surveys. Special
attention was given to the perception of risk of the women in the community and to
the capacity of the health centre services in the village [7]. These variables were
cross-tabulated to birthweight to measure the association between maternal nutri-
tional variables and pregnancy outcome, in this case the percentage of low birth
weight. This information was taken only from mothers holding M&C cards having
delivered a single infant.
The relation between nutritional variables and low birthweight is significant only
for MUAC (x 2 = 17.28, df= 2, P < 0.001), although a trend towards lower
birthweight can be seen for mothers smaller than 145 cm and with no weight gain
during pregnancy (Table 2).
Effect measurements
One of the main questions that can be put forward regarding the evaluation of the
M&C card is whether there is any significant difference between the health be-
haviour of women who use the card and those who do not. The effect measurement
used here is the number of antenatal care visits indicating the health behaviour of the
women.
Women having an M&C card had more ANC visits (during the current preg-
nancy) compared to pregnant women without M&C cards (5.5 against 5.1 times)
(Table 3). This difference is statistically not significant. Of all the pregnant women
using the card, 86% attended ANC while the proportion in those without a card is
only 71% (current pregnancy). This difference is significant. (x 2 =32.55, df = 1;
p < 0.001). Women with a card tend to choose the TBA as the delivery attendant (for
current pregnancy) more often than those without a card: 93% against 87%
156
Table 3. Number of ANC visits for those with M&C card and those without
Pregnant women Number of visits (%)
0-2 3-6 >6times
M&C cards (+) 11.9 57.8 28.3

M&C cards (-) 32.2 44.9 20.2
(x2 = 38.8, d/= 1; p < 0.001). Women with the card are referred to the Health Centre
by the TBA more often than those without the card (17% against 14%, so the
difference is not significant). They are also more aware of the referral reasons (44%
against 20%, again not significant). The same conclusion can be drawn when looking
at TBA attendance rates during the antenatal period (88% vs 47%, x2 = 177.13,
d/= l;p < 0.001).
Interestingly enough, this effect is also noticeable when looking at ANC for the
last pregnancy (= the pregnancy before the current one): women who were using the
card then had a higher ANC attendance rate (75%) than those who were not using the
card (71 %). These women also had a higher tetanus vaccination rate during the last
pregnancy (63% against 56%, not significant). The differences are statistically not
significant, but are consistent enough to merit further study. Although the data are
still preliminary, this finding gives ground for the proposition that women using the
card are already different from those not using the card to begin with.
Do pregnant women holding an M&C card have better pregnancy outcomes

compared to those without?
This is a question of impact measurement, and has to be answered based on the

assumption that pregnant women holding an M&C card receive more ANC com-
pared to those without a card. No statistically significant association was found
between the two groups concerning the percentage oflow birth weight (Table 4).
Impact measurements also include mortality outcomes. Table 5 shows the
stillbirth rate, early neonatal death rate and perinatal mortality rate of both groups.
The results point in the direction of better outcomes; however, this improvement
cannot be statistically proven at this stage.
Table 4. A comparison of birthweight distribution between pregnant women with and without M&C cards
Birthweight (gr) M&C cards (+) M&C cards (-)
<2500 14.2% 13.7%

~2500 85.8% 86.3%
Mean birth weight 2841.36 2863.90

157
Table 5. A comparison ofPMR (%) between women with and without M&C cards
Availability of M&C cards SB ENR PMR
M&C cards (+) 21.8% 20.6 41.6

M&C cards (-) 26.6% 34.6 59.6
(NS) (NS) (NS)
SB = Stillbirths
ENR = Early Neonatal Mortality Rate
PMR =Perinatal Mortality Rate
Discussion
In developing countries, three major problems exist in relation to perinatal health.

First, the availability of health services including their coverage; second, the
nutritional status and daily physical workload of the mother; and third, infection
during the perinatal period. In relation to these problems, the first strategy to improve
maternal health is through increasing the coverage of health services providing the
women with minimal health care. The second approach is nutritional supplementa-
tion and prevention of infection and complication during pregnancy. Finally,
strategies should focus on safe delivery and postpartum care. Thus far, food sup-
plementation programmes have been mainly focused on children. Pregnant women
have been given supplementation largely to increase birthweight.
Nearly all food supplementation programmes have been pilot projects, because to
implement food supplementation as an ongoing effort would be quite expensive.
Besides the high costs, food supplementation presents other difficulties and in order
to economize, supplementation programmes are in the main directed at those most in
need. Many studies have taken anthropometric measurements as indicators for food
supplementation. In the Tanjungsari Intervention Study, food supplementation,
prevention of infection and improving hygienic living conditions and personal
hygiene are all integrated.
The risk approach is a health management strategy and aims to improve health
care for all. It gives special attention to those individuals or groups with greater
needs. Inherent in this approach is maximum utilization of all resources, traditional
as well as formal. It also allows for and promotes self-reliance in the community and
the family, particularly for mothers of young children. It is expected that the risk-
approach strategy aimed at mothers and children at special risk will have a far-
reaching effect on the whole organization of maternal and child health services and
may lead to improvements in both the coverage and quality of health care at all
levels, but particularly at the primary health care level ([8] Workbook on RAS,
1984). The implementation of the risk-approach may differ from one country to
another and may even have to be adapted to local communities within the same
country. The implementation is very much dependent on the socio-cultural condition
of each area and the method of improving maternal and child health care in a
158
particular area may also be different. In the Tanjungsari Intervention Study, more
than 80% of deliveries are attended by TBAs and more than 90% are home
deliveries. The overall objective of the study is to develop a practical methodology
for the assessment of local strategies by matching levels of risk to levels of resources
and to test their efficiency and effectiveness. This study also aims to improve MCR
services so as to achieve a total coverage of all risk groups (mothers and children).
The involvement of the mother in monitoring her own health and the health of her
young child and in the implementation of appropriate actions to minimize risk is in
line with her task as the primary health care provider. Integrating traditional and
modem health care systems is difficult. It calls on health providers in both systems to
develop new skills and knowledge in mothers and traditional health care providers
which can also be used by the formal health care system. The preliminary results of
the Tanjungsari Intervention Study show a tendency towards better effect and impact
measurements but there is no noticeable effect on birthweight at this stage.
Acknowledgement
The authors gratefully acknowledge the financial support provided for this study by
the Sophia Foundation Rotterdam, N.V. Nutricia, WHO-SEARO, EC Contract No.
TSD-M-361.B(B) and the Ford Foundation project no. 870--0714.
References
1. Peeters RF, Tanuwidjaja S and Meheus AZ: Attitudes and practices of the traditional birth attendants
in rural West-Java: An ethnographic survey, ESOC Publication 2, Belgium, 1987.
2. Pizurki H, Mejia A, Butter I and Ewart L: Women as providers of health care, WHO, Geneva, 1987,
p.9.
3. Alisijahbana A, Peeters RF and Meheus A: World Health Forum 7:240, 1986.
4. Essex BJ and Everett VJ: Tropical doctor, 1977, p. 134.
5. The John Snow Public Health Group Inc, Problem action guideline for basic health care: A tool for
extending effective services through auxiliary health workers, Boston.
6. Alisjahbana A, Ranadipura D and Tanuwidjaja S: Bulletin Pen Keseh 11:35, 1983.
7. Alisjahbana A, Utama D and Peeters RF: Maternal nutritional variables and pregnancy outcome,
Menzies Symposium, 57th Anzaas Congress, Townsville, 1987, p. 47.
8. A workbook on how to plan and carry out research on the risk approach in maternal and child health
including family planning, WHO, Geneva, 1984, p. 11.
Paper by C.A. Powell and S.M. Grantham-McGregor
A.M. Molla:
I have two questions on your intervention study. First, a methodological one: how did you keep your
controls in the hospital? Second, an ethical one: on what reasons was it found to be justified not to
intervene in the malnourished group since you had the knowledge of their status and the possibility of
supplementing?
C.A. Powell:
Because in a number of studies the effect of hospitalisation was not investigated, we did so in our
intervention study. The group of adequately nourished children were recruited from children admitted
to the pediatric ward for several other reasons (for example, breaking a leg). They were tested at that
time and followed up after leaving the hospital.
The second part of your question indeed refers to a difficult ethical matter that one is faced with in
doing this kind of study. The children would normally have gone back to their poor environment and
there is no nationwide programme for stimulation. Since we wanted to see whether it was possible to
demonstrate any benefits to the children's development, we judged it to be necessary to leave the
control group in the situation that they would normally have gone back to.
L. Mala:
What would you consider to lead to the most significant improvement in behavioral aspects in
malnourished children: nutrition or social stimulation?
CA. Powell:
This is not a new question and in my opinion there is not much relevance to try to answer it The fact
is that nutrition on its own can have an independent effect, but we also know about the importance of
the social background.
C.Gopalan:
I am very happy with the findings concerning the effects of the school meal programme in your study,
certainly in view of the many discussions on the relevance of such programmes. Since this kind of
supplementation can only be given during 200 days a year and may supply just about one third of the
daily requirements, it is not astonishing that miracles such as significant differences with respect to
growth performance cannot be found as a result of such school meal programmes. Although we found
in some of our studies at the Nutrition Foundation that the learning abilities, in agreement with your
study, are improved in schools with school meal programmes, these data are not considered to have
the same conviction as, for instance, growth performances. Therefore, the question has been raised
many times whether we are justified in investing millions of dollars in school meal programmes in
rural schools.
H.KA. Visser and J.G. Bindels (eds.) Child Nutrition in South East Asia. 159-165.
160
C.A. Powell:
The literature on school feeding is indeed very conflicting and to my knowledge only a few studies
have dealt with really undernourished children. Although I am very much in favor of school feeding
programmes, I would say, also based on our own study that if the resources are extremely limited, we
perhaps should target the children especially at risk, rather than feeding everybody at school.
E. Suroto-Hamzah:
Did you also explore in your study any differentiation between marasmic and kwashiorkor children
with respect to behavioral effects? I am aware of one study where it was found there were no
differences with either marasmic or kwasiorkor boys, but that the situation with kwasiorkor girls was
less unfavorable than with marasmic girls.
C.A. Powell:
We have investigated this in our study and found, in agreement with the results published from some
other studies, that the development of kwasiorkor children is not as adversely affected as that of the
marasmic child. We suggest that this may be due to the duration of malnutrition, realising that
kwasiorkor is a much more acute state less often accompanied with severe stunting than is the case
with marasmus. We found that stunting on its own showed dramatic effects, rather than actual severe
episodes. Therefore we think that the observation that kwasiorkor children show less deficits than
masasmic ones is related to the fact that they are less stunted.
ALi Ming Cheng:

Did you consider in your study the possible contribution which iron deficiency could have in the
behavioral change?
C.A. Powell:
In the present studies we did not look at iron deficiency and its effect on behavior. Several other
studies, however, of which one has been performed in Indonesia, demonstrated that iron deficient
anaemic children on receiving supplementation showed improvements in functions which were
previously depressed.
S. Drahaman:
Please allow me to make some comments. First, from our own previous studies on the effects of
nutrition supplements we were not convinced that they were very helpful in mental improvement.
Recently, however, new studies from Sri Lanka showed that stimulation did reveal definite improve-
ments in these children. Therefore, I think we should all get together and agree on a number of
common series of tests. Secondly, I would like you to comment on the theory of aflatoxins being one
of the major causes of kwashiorkor.
C.A. Powell:
We are currently investigating the effects of supplementation and/or stimulation on the development of
stunted children. The three groups in this study received supplementation alone, stimulation alone, or
supplementation and stimulation. Our preliminary results show that for both groups in which the
effects of supplementation or stimulation alone were investigated an improvement was observed. The
most interesting results should be obtained in the near future when we hope to find out whether the
children who receive both treatments will show improvements over and above those from the other
groups.
Your question to comment on the relation between aflatoxins and kwasiorkor is difficult to answer
since we are not really convinced that the published work indeed substantiates this theory to be of
general importance. More research is needed.
A. Alisjahbana:
I am curious to know your experiences on how the mothers reacted on the interventions and how much
time they gave to stimulate their child.
161
CA. Powell:
All the studies I have described were done in urban areas. We obtained a very high level of coopera-
tion from the mothers. The mothers which were not so literate sometimes gave trouble at the start,
probably because they were afraid to show up their level of literacy. Once they got used to us the
cooperation was very good. We made materials both for the mothers and the children which had a
positive effect on their compliance. We put strong emphasis on language development; we encouraged
the mothers to talk much more to their children than they were used to.
Moersintowarti Navrendra:
If we are aiming for an appropriate intervention, what would be the most adequate time to start and
what is the most sensitive period of catching up?
C.A. Powell:
In our studies we found that development in the first year is mostly very good. Many poor deprived
children start to show a decline somewhere between 12 and 18 months of age; at about 4 years of age
they are already considerably behind. Unless a preventive programme is executed, I would say the
need is less to start intervention under the age of one year; above that age the damage really starts to
become obvious.
Paper by P.F. Heywood and A.H. Heywood

C.Gopalan:
On comparing the presented data with those from the 1956 study by Venkatachalan, the most striking
observation is indeed the remarkable secular trend, the increase in height. Observations on any high
prevalence of parotid swelling or gynaecomastia in this 1956 study have not been referred to in studies
performed in recent years, including the above paper. In India and many other developing countries,
the clinical profile of malnutrition is changing: we do not observe parotid swellings or gynaecomastia
anymore. Does this also apply to the above study?
P.F. Heywood:
Indeed we could observe the same, we also see only very little gynaecomastia and parotic swelling. In
the area where this 1956 study was performed, the Zimbu province, we now hardly observe any
kwashiorkor.
M. Gracey:
The very dramatic secular trend in the growth of children in Papua New Guinea raises an important
practical point with respect to comparison of the investigated growth data with a set of standards or
reference values for growth. Is there a set of data which can be considered to be the most appropriate
one in this respect? Moreover, I wonder to which extent this is due to nutrition alone and what the
contribution of eradication of endemic and/or parasitic diseases during the past period of 25-30 years
was.
P.F. Heywood:
We used the international standards merely to have a constant point of reference that allows us to
make comparisons with other populations. We do not consider these standards to be appropriate as a
goal for the several populations in our study.
The issue about whether this secular trend in growth is due to nutrition alone might perhaps never
be answered satisfactorily. It should be reminded in this place that in the area we performed our study,
malaria was not endemic because of the altitude and that diarrhea is neither a dominant cause of
morbidity or mortality. The most dominant cause for disease or death in children here has probably
already been for a long time pneumonia. There have also been large improvements during the last 10
years regarding the education of the mothers.
D.Karyadi:
Since we found in a study in Indonesia very strong correlations between anaemic condition and child
162
behavior, I wonder whether your data might support this work. Secondly, do you also have data
available on the changes in food habits and the change in lifestyle as a result of the improvement in
income during the last 25 years?
P.F. Heywood:
On the question of iron, I would like to refer to two of our own studies on this subject. Firstly, we did a
double blind supplementation programme in infants at 2 months of age in a lull setting and then looked
at their visual attention when they were 1 year old. We found a strong and significant effect of
supplementation on visual attention. In a second study, it was aimed to study behavioral effects of iron
supplementation together with the possible effect iron supplementation might have on malaria. In this
study we neither found any effects of iron on malaria, nor did we find any effect of iron on the
attending behaviour of the school children. Therefore, one might conclude that iron supplementation is
only effective in neonates and not in school children.
The question on the food habits in the highlands is an important one. Indeed there have been
enormous changes in what the people actually are eating. In 1956 sweet potato contributed to 90% of
total calories and probably around 80% of total protein intake. In 1981 over 50% of the energy was
derived from cereals, while protein intake from cereals was also near 50%. Their total protein content
did also show a marked increase as a result of these changing food habits.
Fortunately for the highlanders of New Guinea, the attitudal changes which have come with
participation in the cash economy did not result in a cultural conflict. Survival and competition in the
cash economy was for the highlanders not so different from survival and competition in the traditional
society. The availability of western medical care, however, represents another marked change in the
peoples' attitude. Western medical care now exists alongside persistent beliefs about traditional
medicine. For adults it is quite common to first consult a traditional healer and as a second resort the
available western medical services are consulted. There are some indications that with children this
hierarchy is actually reversed and that mothers directly seek western medical care for their infants.
Paper by A. Alisjahbana, R. Peeters, M. Maertens, S. Tanuwidjaja,

H. Tjokrohusodo, W. Ngantung and D.A. Prim ana
D. Karyadi:
From your paper, I understand that the mother and child card is intended to be a tool of management, a
tool of evoking the awareness of the mothers as well as the TBA's. I am wondering whether it is
possible to demonstrate the benefits of the card by analysing the range of activities exerted as a result
of the use of the card such as immunisation, iron supplementation and pill distribution. If you could
show that the extent of these activities differ between the areas where the card is used and those where
this is not the case, you might have an extra validation of its use.
A. Alisjahbana:
In theory, you are absolutely right and we should be able to extract these data from our material. In
practice, the situation is more complex. It is not yet possible to perform all the actions indicated in the
card at all sites where the card is used, neither in any area which you would choose to be a control
area. Health service programmes like iron supplementation or pill distribution are not always
available.
l.A. Kusin:
I have two questions: firstly, I am interested in the profile of the mothers in your study which do not
have a card; how do they compare with the group having a card? Secondly, I would like to ask you
whether you have set yourself a time target in your programme in which you would hypothesise to
reach certain percentages, for instance within five years.
A. Alisjahbana:
With respect to the first question, the type of mothers in both groups is not very different. The main
difference happens to be that mothers without a card do live far apart from the main road and are on
the average older than the mothers with a card. The reason for this is trivial since it would take these
163
mothers thtee hours or longer to walk to the health centres. The intervention in both groups, however,
is the same.
About the question concerning the time target, we fully realise that it is impossible to achieve a
change in behavior in I or 2 years; probably at least 3 to 5 years are necessary. Our study will last only
until the child is 2 years of age, but we will continue to do cross-sectional studies to monitor any
behavioral changes of the women in the community.
M. Gracey:
We have tried a similar system to your child cards with aboriginal mothers in north-west Australia
which has not been very successful because many mothers lose the cards. This caused unnecessary
extra work for child health centre nurses.
W.C.Liu:
In Taiwan we have already used a similar card for about 20 years. The cards are distributed by the
public nurse and the main success of it now is its role in birth control.
A. Alisjahbana:
We monitor the availability of the cards by the mothers every 3 months and have so far found that only
a small amount of the mothers lose the card: 2-3%. Since we are only at the start of the study, it is not
possible to predict whether this figure will also be applicable after 3-5 years. For practical reasons (the
population in the research area amounted to 80,000 with only 5 nurses, I midwife and I doctor) it was
decided not to keep central records at the child health centres. The card handed out to the mother is the
only one and is used both for the grass root level as well as for the upper level.
A. Li Ming Cheng:
I am interested in the utilisation of the TBA's in your approach which appears very rational. Did you
observe difficulties of performance with these people based on their educational background? Do you
have difficulties in changing their attitude from their former practice by training?
A. Alisjahbana:
Indeed, we do have problems in training the TBA's. We hope to improve on this situation by also
involving the TBA's in antenatal care, so they have more contact with the patient. Since this also
increases the income of the TBA' s, they are better motivated to do their work and undergo training. As
a result of the ongoing study, we were able to increase the number ofTBA's from 70 to 100 by mainly
young literate women attracted by the programme. These TBA's all have potentials to develop into
primary health workers.
A.W. Qureshi:
In Pakistan, where we have a similar system in which more than 80% of the deliveries are conducted
by TBA's, we are now training the TBA's because the incidence of neonatal tetanus was rather high.
These programmes are performed with the help of UNICEF and the WHO where the TBA's are also
equipped with kits to enable them to perform deliveries.
A. Alisjahbana:
In doing the baseline survey, we were very surprised that the levels of neonatal tetanus were
essentially the same between trained and untrained TBA's. This points to the fact that some of the
training the TBA's have received is not well utilized. The incidence of neonatal tetanus in our study is
rather low. In the intervention study, there was only one case of neonatal tetanus over the whole period
of nearly one year.
DR. Karunaratne:
Please allow me to comment on this, based on the situation in Sri Lanka. We have managed to achieve
that 95% of deliveries are in hospital and that the literacy rate for women is about 85-95%. In this way
we have nearly banned neonatal tetanus. To my opinion, education of the women is indeed very
important as a contributary factor in the health of the mothers.
164
A. Alisjahbana:
That is indeed impressive, but in our country we have to face literacy rates up to 50% maximum,
which is rather indicative for our approach. However, we just cannot wait until all mothers are
educated to start our field experiment.

WC.Liu:
I would like to ask Dr. Powell whether, in the study she mentioned on the effect of intervention by
stimulation, there is a dependency of the effect at the age of the child between 6 and 24 months.
G.A. Powell:
Although in that particular study we did not look at the age of entry, we found in another study, which,
however, was not performed with severely malnourished infants, that children from just below 1 year
to 2 years of age responded equally well upon intervention. The reason why others did not find an
effect on intervention may be that they started later with intervention, say at 3 years of age. By that
time the damage may already be done.
L.Mata:
Did you do any studies on comparing the mental development between kwashiorkor infants and, let us
say, normal ones and follow them longitudinally? In a study in the highlands of Costa Rica, one of my
colleagues found that many children at 3 and 6 months perform less than the standards while passing
the tests at 10 and 12 months of age. It appeared that the deficits were corrected by the contact the
field worker had in these rural areas with the mothers and the infants. Traditionally the mothers do not
interact enough with the infants because of their busy daily work. However, after they saw the
psychologist play with the children, they started to like this and adapted to it. The problem in the
highlands of Costa Rica is not only that of malnutrition.
GA. Powell:
We did not perform studies with newborns. With infants up to approximately 6 months of age the
development is quite good, probably because the contact with the mother is very close. Later on,
however, when they are left to fend for themselves, we found that their development slopes off.
S. Drahaman:
Could you find in your studies that breast fed infants might have a better development of IQ than non-
breast fed infants because of the form of stimulation the former might receive?
GA. Powell:
This is an interesting question but I am sorry that I cannot answer you since we have not attempted to
establish any relationship here.
L.Mata:
I would like to change the subject somewhat and come back to the discussion on the possible
explanations of the secular trend in the growth of children in Papua New Guinea. I can hardly accept
that this is accomplished only by changes in nutrition. In my opinion we just cannot uncouple the
improvements in nutrition from other parameters which obviously have also been changed. In the area
between 600 and 1200 metres, where most intestinal parasites are found, I suspect that most people
now walk in shoes and perhaps take showers frequently. In doing so, infection is efficiently combated.
Therefore, I would like to ask professor Heywood whether he has data on the incidence of diseases
which determine mortality like tuberculosis, measles and whooping cough. I would expect that the
incidence of these diseases has decreased and that this would also be an explanation for the secular
trend, next to the improved nutrition, of course.
P.F. Heywood:
Let me first mention that the infant mortality for the whole country has decreased dramatically: from
165
around 130 10 aboUl 70 now. However, I do not think that shoes and showers are an acceptable
explanation for this, since they are not yet available on a larger scale in the middle altitude zones.
Moreover, perhaps unlike Professor Mata's experience, we were astonished to find out that the Papua
New Guinea people are traditionally extremely careful about water in terms of what they drink.
Unfortunately, we do not have reliable data on the incidence of diseases over the last 50 years. When
we compared different levels of economic development with respect to the incidence of acute
respiratory infections, we were astonished to learn that there was no dramatic difference in the attack
rates on comparing the different levels of environments. Although, this perhaps is not too unexpected,
I would never state that infections are unimportant in this context. However, it would be equally
wrong to stress only infections without recognising the economic factor. There are places in Papua
New Guinea where health services have been significantly improved but without any economic
development. As a result of this we see halved infant mortality, which did not make a difference in the
growth rate of the children. The point I want to make here is that improving health services alone
without paying attention to economic and social development appears to be insufficient.
M.Gracey:
Could I bring in another interesting variable, namely the rapid urbanisation in Papua New Guinea. Is it
known what effect this has on children's nutrition and growth apart from the issues Professor Mata has
raised before?
P.F. Heywood:
First of all, I think there has been a deterioriation in the environment of the people who move to the
urban areas. If anything, their health status has probably decreased overall, although their access to
services has probably increased. There is no baseline level against which we can compare the growth
of urban children, partly because urbanisation is recent. At the moment people do feel that children in
urban areas are growing better than those in rural areas, but there is really no good data to support or to
deny this.
M.Gracey:
What about the ethnic and racial diversity of the country?
P.F. Heywood:
There is clearly enormous diversity in the country in the way people grew at the moment. In general
people in the highlands are short compared to western standards but have very high percentages of
weight for height. The lowland people are taller and are also slimmer. We have to some extent tried to
avoid the question of whether they are all, growth wise, going the same way. We are making only
comparisons within a region and not across regions. It would Rot surprise me that we will find out that
people in Papua New Guinea have the same potential in terms of linear growth but that the body
proportions are rather different group by group.
D.R. Karunaratne:
Apart from nutrition, environment and economic factors, there is obviously also a genetic factor in the
size of the newborn babies. For instance, Eskimos or Cameroons give birth to relatively large babies,
while Pigmies bear small babies.
P.F. Heywood:
I think that there is clearly a genetic factor involved in birth weight. We do not really know at which
point environmental factors become more important. There are also inter-generational effects going on
in the secular trend in birth weight which means that it is even harder to sort these effects out. We see
in Papua New Guinea that it is possible to modestly increase birth weight as people change their
lifestyle; the next generation is going to see a further increment which is related to them having larger
women bearing those children and so on. These long-term inter-generational effects are the basis of
the secular trend, I think.
Session 4
Nutritional deficiencies in children
Chairman: D. Karyadi
Iron deficiency anemia and its problems
A.G. SOEMANTRI HARDJOJOEWONO
Introduction
From the public health viewpoint, the most significant common cause of nutritional
anemia in Indonesia is iron deficiency. The second most common cause is folate
deficiency, often present in pregnant women and low birth-weight babies. Other
nutritional deficiencies which playa less important role are vitamin B12 deficiency
and possibly protein deficiency. There are many factors contributing to this problem:
Deficiency anemia, especially iron deficiency anemia, may have an effect on
morbidity, growth and behavioral alterations in children; it may increase the
incidence of perinatal or neonatal morbidity and mortality in pregnant women;
and may have an impact on physical endurance and productivity significant for
national development planning in Indonesia.
The popUlation growth, which is 2.15 percent and the total population of ap-
proximately 164 million inhabitants.
Geographical situation.
Feeding pattern and caloric pattern.
Socio-economic condition.
Considering these factors, the Government of Indonesia decided that iron deficiency
anemia is one of the major deficiency problems; the others are protein caloric
malnutrition, vitamin A deficiency and iodine deficiency.
It is in this context that iron deficiency anemia and its problems in Indonesia will
be discussed.
The prevalence of iron deficiency anemia in Indonesia
Based on the 1985 Population Survey, the total population of Indonesia is 164
million. Indonesia is the fifth most populous country in the world after China, India,
the Soviet Union and the United States of America. There was a high annual growth
H.K.A. Visser and J. G. Bindels (eds.) Child Nutrition in South East Asia. 169-175.
170
Tabll! 1. Recorded prevalence of iron deficiency anemia in several Indonesian studies. References to these
studies are available from the author; most of these studies are published in the Indonesian language [16].
Prevalence (%)
Children 6 months-5 years

-low socio-economic groups:
• well-nourished 38-73
• with mild protein deficiency 83
• with severe malnutrition 85-100
- medium and higher socio-economic groups:
• well-nourished 24
Children 5 years-14 years
- low socio-economic groups:
• good nutritional status 47-64
• with mild protein-caloric malnutrition 38-67
- medium and higher socio-economic groups 20
Pregnant women 21-92

Non-pregnant women 35-85
Laborers and plantation workers 30-65
rate of 2.15 percent in the 1980-1985 period. In order to lower the level of the
fertility rate various government programmes are launched. By these programmes
the annual growth rate in the 1995-2000 period is expected to decrease to l.9
percent.
Indonesia has a young population. The percentage of people aged 0-14 years in
1980 was 40.3 percent, in 1985 it was estimated as 39.2 percent. It is estimated that
in 1990 and 2000 these figures will decline to 37.7 percent and 34.2 percent
respectively. Information from various studies in Indonesia about the prevalence of
iron deficiency anemia is summarized in Table l.
Information from these studies indicates that iron deficiency anemia is prevalent
in the most vulnerable segments of the population. This is a national public health
problem and should be solved effectively.
Etiology and influencing factors

Various in-depth studies on different aspects of the etiology of iron deficiency
anemia have been done. The results demonstrate an interaction between many factors
which predispose to develop iron deficiency anemia as shown in Figure 1. The high
prevalence is due to multiple and complex factors that abound in developing
countries. In these countries, there is a low standard of living with a high prevalence
of malnutrition, poor environmental sanitation, high morbidity among children, and
unequal distribution of wealth leading to extremes of wealth and poverty; all these
conditions favor the onset of anemia. Although the major causes of nutritional
anemia are, firstly, of a dietary origin due to low intake and low bioavailability of
UNDERLYING CAUSES IMMEDIATE CAUSES 171
Low tood supply - - - - - - ,
Erronsous fS8ding
practices
Low sociO-9COnomic Inadequate diet - - - - - - ,
status _ _ _ _- '
Lowintakeot
available iron~~
Unsuitable meal Iron
composition Poor absorption deficiency
excess of
inhibitors
Lack of enhancers
GroWlh - - - - - - ,
Pregnancy &
lactation _ _ _---l.
r- Increased requirements
Acute bleeding
Chronic blood loss ~ Blood loss
I
C Poor s~i.'ation ~
parasitism
In:~~~:~~ Infection - - - - - - - '
Figure J. The underlying and immediate causes of iron deficiency [6].
iron and, secondly, due to blood loss caused by parasitism, these two factors are
basically the result of interaction of other deep-seated factors linked to underdevelop-
ment [6].
It is imperative therefore that the causes of iron deficiency and of anemia should
be viewed beyond the immediate environment of the affected individual. The disease
can be traced to its basic social, economic, and political environment. Figure 1
summarizes the underlying and immediate causes of iron deficiency.
Not only is iron deficiency caused by a low amount of iron in the diet, but also by
its poor absorbability. The poor absorption of iron from cereal-vegetable-based diets
is well-known. Economic factors often limit the intake of meat and ascorbic-acid-
rich foods that would enhance the absorption of iron. There are also socio-cultural
practices that directly or indirectly influence dietary intake. Some religions prohibit
the intake of certain meats. Pre-school children, girls and women are often at a
disadvantage; cultural taboos and superstitious beliefs may prohibit the intake of
certain foods at certain times, especially during infancy and childhood or during
pregnancy and lactation.
Parasitic infection is the second leading cause of iron deficiency anemia, and
hookworm is the most notorious in this regard. Hookworm assumes greater sig-
nificance among older children, but fortunately not much in infants. Parasitic
infection thrives in an unsanitary environment, and transmittal is enhanced by poor
sanitary habits, inadequate health facilities, ignorance, poverty and poor agricultural
practices.
The hidden losses become significant when iron stores are depleted because of
increased physiological requirements, making infants and children particularly
vulnerable. The iron requirements of infants and children, despite their much smaller
body size, are almost as high or higher than those of the adult male. This situation is
responsible for the high prevalence of nutritional anemia during infancy and
childhood.
172
Deleterious effects of iron deficiency
Lozoff and Brittenham [9] reviewed the studies on behavioral alterations in iron-
deficient human subjects that have been published during the past 10 years.
This review has examined recent evidence from human studies that iron
deficiency adversely affects behavior by impairing cognitive function, producing
non-cognitive disturbances, and limiting activity and work capacity. The body of
research taken as a whole provides increasingly persuasive arguments for intensify-
ing efforts to prevent and treat iron deficiency anemia.
After iron administration, improvement of subnormal mental performance in iron
deficient infants and children has been reported [7,12-14].
Cognitive function
There is reasonably good evidence that mental and motor developmental test scores
are lowered among infants with iron deficiency anemia. Although the research on
cognitive function in iron-deficient older children and adults is sparse and diverse, it
suggests that there may be alterations in attentional processes associated with iron
deficiency. Iron therapy has not yet been shown effective in completely correcting
the observed disturbances.
Although some aspects of the cognitive function seem to change with therapy,
lower developmental, IQ, and achievement test scores have still been noted after
treatment.
Non-cognitive disturbances
A variety of non-cognitive alterations during infant developmental testing has also

been observed, including failure to respond to test stimuli, short attention span,
unhappiness, increased fearfulness, withdrawal from the examiner, and increased
body tension. Exploratory analyses suggest that such behavioral abnormalities may
account for poor developmental test performances in infants with iron deficiency
anemia. These studies indicate the fruitfulness of examining non-cognitive aspects of
behavior, such as affect and activity, in addition to specific cognitive processes.
Activity and work capacity
There has been a steady accumulation of evidence that iron deficiency anemia limits
maximal physical performance, submaximal endurance, and spontaneous activity in
the adult, resulting in diminished work productivity with attendant economic losses.
The mechanisms underlying these effects, the extent to which anemia or iron
deficiency separate from anemia is responsible, and the counterpart in infants and
children still remain to be established.
173
LOW PREVALENCE MEDIUM PREVALENCE HIGH PREVALENCE

Pilot therapeutic
supplementation trial
Field
fortification trial
Regional or national
fortification
program
Figure 2. Scheme for the control of nutritional anemia adapted from International Anemia Consultative
Group (lNACG) (1977).
Infections
Although the complexity of the interaction between the infectious agent and the host
makes such proof extremely difficult, there is suggestive evidence of an increased
susceptibility to infection in subjects with iron deficiency anemia.
A positive correlation between morbidity from infectious diseases and hemo-
globin levels was observed by Darwin Karyadi [4]. Basta [1] found that anemic
laborers in Indonesia were significantly more prone to infection than non-anemic
laborers, and that supplementation with iron had a beneficial effect on the frequency
of infections in anemic workers. Soemantri et al. [10] and Soemantri [11] stated that
several biochemical and immunological effects of iron deficiency anemia may be
relevant to infection as decreased phagocytic effect of leucocytes due to impairment
of myeloperoxydase activity.
Other effects
It has been documented that iron deficiency may affect many abnormalities of the
body's functions, such as decreased gastric juice secretion [5], reduced activity of
intestinal cell enzymes [2] and subcellular structural abnormalities including
mitochondrial enlargement [8, 3].
Zinc deficiency, which may accompany iron deficiency, can lead to growth
retardation in adolescents and may also contribute to the production of some of the
'classical' signs of iron deficiency, such as koilonychia and Plummer-Vinson
syndrome.
174
How to solve the problem?
The Department of Health of the Republic of Indonesia is improving the health

delivery system and the eradication of diseases through the development of planning.
Especially in relation to solving the problems of iron deficiency anemia, the
government has decided on a programme which is in agreement with the INACG
programme (see Figure 2). Each step should be successfully completed before
proceeding to the next one. If a given trial is unsuccessful, it must be redesigned and
the step repeated.
The programme comprises iron supplementation, parasite control, food and health
education, conducted in short and long terms. The short-term programme is direct
intervention with iron supplementation and the eradication of intestinal nematodes.
The long-term programme consists of educating the people concerning the effect
of anemia, education for primary health care and training medical personnel at all
levels. Iron supplementation programmes are generally limited to a selected group of
the population, where iron distribution on an individual basis is possible. It has been
carried out in most of the health centres and in almost all segments of the population,
e.g. the under five year old group, pregnant women, laborers and factory workers.
Food and education

While emphasis has been given to supplementation, dietary modifications can
increase the intake of available iron and folate and increase iron absorption from
meat and fish and vitamin C-rich fruits and vegetables. Foods that facilitate iron
absorption such as meat, pOUltry and fish are not freely available to low socio-
economic classes due to scarcity, poverty or local dietary practices. In medium
socio-economic classes, iron-rich foods are frequently not consumed in adequate
amounts. In either low and medium socio-economic strata adequate educational
programmes can do much to alleviate iron deficiency. To be effective, such efforts
must be addressed not only to the nutritional properties of foods and nutrient needs,
but also the cultural role of foods and attitudes towards them. Simple principles of
hygiene and health must be presented and basic understanding of cost values of
foodstuffs given. Iron supplementation and parasitic control programmes often serve
as motivational stimuli for interest in nutrition and health education. Among low
socio-economic classes, economic improvements should be a prerequisite to the
eradication of malnutrition. Changes in socio-cultural attitudes although desirable, do
not occur in a short time; nevertheless, a national programme to combat iron
deficiency anemia must include such objectives.
Educational efforts concerning selection, production and consumption of foods
may have to be modified to include information on other relevant factors such as
environmental sanitation and personal hygiene. The education of parents on the
benefits of birth spacing and the special needs of young mothers can help in reducing
the problems of anemia. These efforts should be included in the total educational
175
programme administered by agencies responsible for health, agriculture or education.

A well-nourished and healthy individual is the ultimate aim of any nutritional
programme. The solution to iron deficiency anemia, or any other nutritional problem,
depends upon an adequate intake of affordable and acceptable nutritious foods.
Foods must be available and the economic level of the population adequate to
purchase them. Educational efforts must be broadly conceived, long-term in nature,
and must involve the total educational, health and agricultural organizations of the
country. The solution depends upon an adequate food supply, either from domestic
or foreign agriculture production, appropriately processed, distributed, and consumed
by the population. Education concerning food and nutrition must be an integral part
of any programme.
International cooperation
Cooperation among research workers and supporting agencies should be encouraged.
Close international cooperation in solving the global problems of nutritional anemia
is highly desirable because it facilitates the exchange of ideas and experiences and
stimulates all concerned. In addition, institutions experienced in developing and
applying the more complex laboratory methods should be encouraged to assist
research groups with more limited resources and capabilities.
References
1. Basta SS: Iron deficiency anemia in adult males and work capacity. Sc D Thesis, M.I.T., February,
1974.
2. Dallman PR, Sunshine P and Leonard Y: Pediatrics 39:863, 1967.
3. Dallman PR and Goodman JR: Blood 35:496, 1970.
4. Karyadi D: Hubungan ketahanan fisik dengan keadaan gi zi dan anemi gizi besi. Thesis. Fakultas
Kedokteran Universitas Indonesia, Jakarta, 1973.
5. Davidson WMB and Markson JL: Lancet 11:639, 1955.
6. Florentino RF and Guirriec RM: In: Stekel A (ed) Prevalence of nutritional anemia in infancy and
childhood with emphasis on developing countries. Iron nutrition in infancy and childhood. Nestle,
Vevey/Raven Press, New York, 1984.
7. Gam SM and Smith NJ: J Pediatr 8:346, 1973.
8. Jacobs A: Brit J Haemat 16:1, 1969.
9. LozoffB and Brittenham GM: Hemat Oncol Clin N Am 1:449, 1987.
10. Soemantri AG, Taty Hendarto, Trastenojo Moeljono S, Kusriyati, Kuspandji and Bambang S:
Myeloperoxidase in children with iron deficiency anemia. International Pediatric Congress, New
Delhi, 1977.
11. Soemantri AG: Hubungan anemi Kekurangan zat besi dengan konsentrasi dan prestasi belajar.
Thesis, Fakultas kedokteran Universitas Diponegoro, Semarang, 1978.
12. Webb TE and Oski FA: Pediatrics 7:294, 1973a.
13. Webb TE and Oski FA: J Pediatr 82:827, 1973b.
14. Webb TE and Oski FA: J Special Ed 8:153,1974.
15. Statistik Indonesia (Statistical Year Book ofIndonesia) 1986. Biro Pusat Statistik Jakarta-Indonesia,
Hal 37-43, 1987.
16. Soemantri AG, Soenarto Y, and Sudigbia I: Iron deficiency anemia in Indonesia. Proceedings of the
4th Meeting of Asian-Pacific Division of the International Society of Technology, Seoul, 1979.
Iron deficiency and infant development
T.SADflMIN
Introduction
Iron deficiency is one of the five deficiency problems in many developing countries
as well as in developed countries, [1] alongside deficiencies in protein and energy,
iodine, vitamin A and water (dehydration in diarrhea cases). Ignorance and poverty
are the most common important factors challenging the health professionals. Studies
in many countries indicate that from the public health point of view, there are three
broad categories of anemia:
a) Nutritional anemia,
b) Anemia associated with chronic infections or infestations, and
c) Anemia associated with hereditary abnormalities of the hemoglobin molecule
(including thalassaemia).
These categories are by no means mutually exclusive, and overlapping is frequently
found among them. Some infestations, e.g. hookworm, increase the requirement for
hemopoietic nutrients and increase the occurrence of protein caloric deficiency.
Nutritional anemia may be defined as a reduction in hemoglobin concentration
below that which is normal for the individual, due to an inadequate supply of
hemopoietic nutrients.
Although many nutrients are involved in the production of red cells and hemo-
globin, iron deficiency is the most common cause of nutritional anemia. As a single
micronutrient, iron is of major clinical importance. Iron deficiency is the most likely
form of single micronutrient deficiency to occur in the absence of any other accom-
panying form of malnutrition. In children with severe protein energy malnutrition,
protein deficiency may play a role in the accompanying anemia, but less marked
degrees of protein deprivation have not been shown to cause anemia [2].
A World Health Organization Scientific Group has suggested arbitrary cut-off
points below which anemia is likely to be present in individuals [3]. Persons with
hemoglobin concentrations below these cut-off points have a high probability of
demonstrating a rise in hemoglobin concentration following appropriate intervention.
H.K.A. Visser and J.G. Binde/s (eds.) Child Nutrition in South East Asia, 177-183.
© 1990 K/uwer Academic Publishers.
178
However, there will be a small proportion of subjects with a hemoglobin concentra-

tion below these cut-off points who will not demonstrate a rise in hemo-
globin concentration when supplied with appropriate hemopoietic nutrients. These
values should be used as guidelines and the greater the percentage of the population
with hemoglobin concentrations below the cut-off point, the greater is the problem of
anemia in the community.
By using a standard hemoglobin concentration of less than 10 gldl, Soewandono
[4] found that the anemia status of children aged less than five years has an associa-
tion with the family's economic, maternal, educational, and antropometric nutritional
status and the routine health maintenance examination at the health centre.
The normal well-nourished individual is in a state of iron balance in which the
amount of given iron absorbed from the diet is equal to the amount used for growth,
plus the amount lost from the body.
The iron status of any individual is affected by the amount of iron absorbed from
the intestine. The lower the body iron store, the greater the absorption of iron. During
periods of child growth extra iron is needed for the increased blood volume. The
more rapid the growth, the greater the relative iron requirement and the greater the
risk of developing iron deficiency. Iron is lost from the body through the urine, the
skin and gastrointestinal tract. These losses occur in all individuals and are relatively
constant at about 141..lg per kg of body weight daily [5].
Subjects with chronic iron deficiency might find themselves in a vicious circle
situation of iron deficiency and infections [6]. In this study there was a relationship
between the history of illness in the previous months with the iron status and
hemoglobin response to intervention. The double blind randomized study design that
was implemented in the study unfortunately was not used to prove any causal
relationship between iron deficiency and infection. The relationship between iron
deficiency and infections, however, is still controversial [7]. From a re-examination
of relevant literature Dallman concludes that abnormalities in cell-mediated im-
munity and the ability of neutrophils to kill several types of bacteria are well
established under experimental conditions in iron deficient patients [8]. In this article
the evidence for a causal relationship between iron deficiency and child development
is reviewed from the literature. Due to limited resources, however, we were unable to
consult all available literature.
Conceptual framework
The term 'developmental pediatrics' encompasses the study and promotion of

optimum physical and mental health in childhood. It covers the changes of motor,
mental, sensory and emotional functions together with the detection, investigation
and management of children in whom the development of these functions is suspect
or abnormal.
179
To study the evidence for a causal relationship between iron deficiency and infant
development the following issues will be discussed. Firstly, we will try to review
studies that report on the association of iron deficiency and infant development.
Secondly, we will try to review studies that report on intervention in iron deficiency
and the improvement of infant development.
It is well-established that acute and chronic protein-energy malnutrition can affect
mental development in children. The age of malnourished children and the severity
of the malnutrition are important factors in relation to the question of whether
development can be permanently retarded [9]. Chase and Martin [10] observed 19
children who had been hospitalized with under-nutrition in the first year of life and
compared them with a control group 3-4 years later. It was found that impairment of
mental development appeared to correlate with the duration of malnutrition.
The causes of iron deficiency have been listed as an inadequate iron intake,
malabsorption, excessive losses of iron and an increase in iron requirements. The
iron deficiency status of the 6 to 12 month-old infant depends mainly on whether
most of the infant's calories come from human milk [11] or whole cows milk, and
whether this is consumed on a regular basis [12-19].
Iron deficiency and infant development
Some methodological problems will be discussed in this review to analyse the

evidence of a causal relationship between iron deficiency and infant development.
It is well-known that a randomized controlled trial study design will provide a
valid study result. Other study designs, such as analytic observational and descriptive
methods, will be less conclusive. From eight studies under review, five studies
implemented a non-randomized control trial [20-24], two studies used cross
sectional study designs [25,26] and only one study conducted a randomized control
trial [27].
Oski and Honig [20] allocated the iron deficient infants to an experimental and
control group randomly, which is accepted as a strong study design. However, in the
analysis of the study the authors compared the Bayley Scale of Infant Development
(BSID) scores for each group, pre- and post-intervention, without an attempt to
compare the two groups to which the random assignment was implemented. This
method of analysis was applied in other studies [21, 22]. Lozoff et al. [27] compared
the performance with the Bayley Infant Scale of Mental and Motor Development of
iron deficient children with various degrees of anemia who received one of three
therapeutic interventions (iron injection, one week of oral iron drops, or three months
of oral iron drops) or placebo.
Most of the studies evaluated the relationship among infants aged less than 24
months, one study included children aged 18-60 months [24] and one study did not
describe the age group of the subjects. Lozoff et al. [27] mentioned that Bayley
Scales of Infant Development scores in the second year of life are only moderately
180
correlated with measures of cognitive function in childhood and this battery of infant
developmental tests is designed to assess a child's developmental status from 3 to 30
months of age. If this is so, an adjustment to age discrepancy should be taken into
account.
The representative populations from where the subjects were drawn are well
distributed from hospital, community and primary clinics. None of the studies
reviewed calculated a predesign sample size, one study calculated the statistical
power of significance. Thus, although the aim of the studies was the same, the
number of subjects included in the studies was different. Oski and Honig [20]
divided 24 iron deficient anemic subjects randomly into 2 groups, 12 subjects in each
group. Oski et al. [21] recruited infants without signs of anemia, 10 subjects in each
group with sufficient iron, iron depletion and iron deficiency as reflected by an
increase in the erythrocyte porphyrin concentration and 8 subjects with iron
deficiency with increased erythrocyte porphyrin accompanied by a decrease in the
mean cell volume (MCV) of the red cells.
Lozoff et al. [22] included 68 infants, 28 iron deficient anemic infants and 40 non-
anemic controls. Each group was divided into those who received iron supplementa-
tion and those who received a placebo. This particular study was conducted in a
double-blind manner. Walter at al. [23] started their study with 51 infants, but only
37 (74%) infants completed the study'S procedure. Deinard et al. [24] recruited 25
iron deficient anemic infants, 45 iron deficient non-anemic infants and a control
group of which the number was not mentioned. Lozoff et al. [25] as part of a main
study analyzed the association between developmental test scores and the degree of
iron depletion with a total sample similar to that reported elsewhere [22].
The diagnostic criteria mostly used for anemia were Hb of less than 10.5 g/dl,
hematocrit < 33% and MCV of 73 J.L3 or less, whereas the criteria used for iron
deficiency were the combinations of ferritin < 12J.Lg/l, transferrin saturation < 10%,
and free erythrocyte protoporphyrin > 100 J.Lg/dl. Some individual studies had a
variety of criteria for anemic and iron deficient children and those who were labelled
as normal.
All studies reviewed used Bayley Scales of Infant Development, Uzgiris and Hunt
Ordinal Scales of Psychological Development {UIHSPD) and others to assess the
developmental status of subjects. Four issues should be considered when assessing
the developmental status of infants [28]: (1) the tests should be correctly adminis-
tered and scored in the prescribed manner (2) intertester and interobserver variability
should be controlled to ensure consistent measurements (3) the sensitivity and
specificity of the tests should be specified when one attempts to predict the later
development status from an earlier developmental screening test (4) one should
avoid making inappropriate generalizations or drawing inappropriate conclusions
from study results. The first two issues are the most important ones to consider when
evaluating the developmental status of infants. These issues were not mentioned in
any studies under review.
When the causal factor is removed and the clinical manifestation will disappear, a
181
causal relationship may be concluded. In this case it is necessary to overcome the

problem of iron deficiency by giving iron to the subjects, and expect that the
impaired infant development will be improved.
Lozoff et al. (25] and Deinard et al. (26] attempted to show the association
between iron deficiency and infant development, but not a causal relationship.
Lozoff et al. (25] found that the correlation coefficient between the degree of iron
depletion and Mental Development Index (MDI) was 0.73 and the correlation
between iron status and Psychomotor Development Index (PDI) was not statistically
significant at the 0.05 level with r =0.29. Deinard et al. (26] compared iron depleted
children without signs of anemia (hematocrit> 34%) with those who had no proof of
iron depletion. The results from an analysis of variance indicated that there were no
statistically significant differences between groups under comparison on the motor
and mental scales of the BSID.
Iron supplementation and the improvement of infant development
There were six studies which investigated the impact of iron supplementation on the
improvement of infant development scores (20-24]. As mentioned earlier, the
association between iron deficiency and infant development is a multifactoral
process. Soemantri et al. (29] carried out a study to measure the benefits of iron
supplementation for anemic children in school achievement test scores. Oski and
Honig (20] and Oski et al. (21] observed the changes of MDI scores before and after
iron supplementation. In the first study the investigated groups were shown to be
similar as far as sex, race, parents, years of education, weight and age were con-
cerned. The retest took place on average after 7 and 10 days both for the experimen-
tal and control group (20]. In the second study (21] factors that could influence the
study results were controlled by exclusion criteria which were premature birth,
neonatal distress, chronic illness or congenital anomalies. The other factors were
shown to be comparable since sex and colour were matched. All subjects were tested
within 2 weeks of the initial screening procedure.
Lozoff et al. (22] provided a careful study in controlling the potential confounding
factors. The change in MDI did not vary as a function of the anemic or treatment
group. Walter et al. (23] who controlled the potential confounding factors with
inclusion criteria and matched control selection found interesting results. In this
study there were 3 groups of infants: iron sufficient, iron deficient anemia and iron
deficient without anemia. It turned out that only the iron deficiency anemia group
had an MDI mean score which was significantly different from the control group at
the first screening procedure, but not the group with iron deficiency without anemia.
Iron therapy gave a significant improvement of the MDI mean score in the anemic
group only. Deinard et al. (24] included infants with a wide age range, 18-60
months. Those who were iron deficient with anemia and a control group matched for
the mother's educational level showed no significant difference in the mean mental
182
development score at the baseline. After 3 and 6 months of iron treatment, iron
deficient infants and pre-school age children did not increase their developmental
test score, but the controls did. In a clear and carefully conducted study, Lozoff et al.
[27] observed no significant difference in MDI score improvement between the
groups under comparison.
Pollitt et al. [30] included children aged 30 to 70 months in their study. The
performance of 25 iron deficient anemic and 25 iron depleted children in a series of
attention, memory and learning tasks before and after 11-12 weeks of oral therapy
was studied. The delta of test scores in the control group was larger than that of the
experimental group.
Conclusion
All studies have observed that infants with iron deficient anemia have lower MDI
scores than infants who were iron sufficient without anemia. These results
demonstrated the evidence of a relationship between iron deficiency and infant
development. The time of a critical iron level at any age that is related to the intake
and utilization of iron was never documented. Most of the studies reviewed have
problems in controlling the potential confounding variables that might influence the
conclusion of a causal relationship. Study designs and the analysis to answer the
questions in some studies could be improved. Some issues of the causal relationship
problems, such as case response, were not analyzed in the studies reviewed. The role
of iron supplementation to improve the mental developmental score of iron deficient
infants is still controversial. Several authors made suggestions for investigations with
a stronger study design. All agreed that physical developmental scores are not
affected by iron deficiency.
References
1. Expert Scientific Working Group: Am J Clin Nutr42:1318, 1985.

2. Beisel WR: Am J Clin Nutr 35:417, 1982.
3. World Health Organization WHO: Techn Rep Ser 405, 1968.
4. Soewandono A: In: Simposium Anemia Gizi. Fakultas Kedokteran Universitas Diponnegoro / RS Dr
Kariadi, Semarang, 1977.
5. Baker SJ: Bull WHO 56:659, 1978.
6. Oski FA: Am J Dis Child 133:315,1979.
7. Hershko C, Peto TEA and Weatherall OJ: Br Med J 296:660, 1988.
8. Dallman PR: Am J Clin Nutr 46:329, 1987.
9. Frisch RE: Am J Clin Nutr 23: 189, 1970.
10. Chase HP and Martin HP: New Engl J Med 282:933, 1970.
11. Saarinen UM: J Pediatr 93:177,1978.
12. Committee on Nutrition: Pediatrics 72:253, 1983.
13. Fomon SJ, Ziegler EE. Nelson SE and Edwards BB: J Pediatr 98:540,1981.
14. Sadowitz PD and Oski FA: Pediatrics 72:33,1983.
15. Birch HG: Am J Dis Child 120:395, 1970.
183
16. Martin HP: Am J Clin Nutr 26:766. 1973.

17. Soerats S: Thesis School of Physiotherapy. Western Australian Institute of Technology. 1982.
18. Escalona SK: Pediatrics 70:670, 1982.
19. Meisels SJ, Plunkett JW, Roloff DW, Pasick PL and Stiefel GS: Pediatrics 77:345, 1986.
20. Oski FA and Honig AS: J Pediatr 92:21,1978.
21. Oski FA, Honig AS, Helu B and Howanitz P: Pediatrics 71:877,1983.
22. Lozoff B, Brittenham GM, Viteri FE, Wolf AW and Urrutia 11: J Pediatr 100:351, 1982.
23. Walter T, Kovalskys J and Stekel A: J Pediatr 102:519. 1983.
24. Deinard AS. List A. Lindgren B. Hunt N and Chang PN: J Pediatr 108:681, 1986.
25. LozoffB, Brittenham GM, Viteri FE, Wolf AW and Urrutia JJ: J Pediatr 101:948,1982.
26. Deinard A, Gilbert A, Dodds M and Egeland B: Pediatrics 68:828, 1981.
27. LozoffB, Brittenham GM, Wolf AW et al.: Pediatrics 79:981,1987.
28. Frankenburg WK, Chen J and Thornton SM: J Pediatr 113: 111 O. 1988.
29. Soemantri AG, Pollitt E and Kim I: Am J Clin Nutr 42:1221, 1985.
30. Pollitt E, Saco-Pollitt C, Leibel RL and Viteri FE: Am J Clin Nutr 43:555. 1986.
Infection and immunity
of vitamin A and iron deficient children
P. BHASKARAM
Introduction
Vitamin A is one of the vitamins with widespread physiological functions in the

body. Apart from its role in dark adaptation, the functions of vitamin A in maintain-
ing the structural and functional integrity of mucosal epithelium, cellular prolifera-
tion and its adjuvant actions on the immune system are important.
Vitamin A deficiency is one of the important nutritional deficiency disorders in
young children in the developing countries. Besides the xerophthalmic effects and
the consequences thereof, extra ocular manifestations of vitamin A deficiency have
attracted the attention of several researchers in recent times. Of these, the effects of
its deficiency on the immune system are of practical relevance.
As early as 1917, McCollum described the frequent occurrence of naturally
acquired infections in vitamin A deficient rats and described the interactions between
vitamin A deficiency and infection as synergistic [1]. In 1928, Green and Mellanby
named vitamin A as an anti-infective vitamin with the help of the literature available
in the previous decade, which was supported by their own studies wherein a high
frequency of histopathological evidence of infection was demonstrated in a large
number of rats fed on vitamin A deficient diets over a long period [2]. Subsequently,
several studies have been conducted in experimental animals to establish the causal
relationship between vitamin A deficiency and infection. While several studies
supported the anti-infective nature of the vitamin [3-5], certain others found no
effect of either vitamin A deficiency or vitamin A supplementation on the course and
severity of infections [6,7]. The conflicting nature of the results was mainly due to
the number of variables that could not be avoided in these experiments. In most of
the studies, the deficient animals were compared with controls fed ad libitum, thus
making it impossible to exclude the effects of other co-existent nutritional
deficiencies.
The introduction of pairfed controls into the experimental designs reduced the
variability of results due to associated protein energy malnutrition to a great extent.
Using this model, Krishnan et al. [8], Nauss et al. [9] and Chandra and Au [10]
H.K.A. Visser and J.C. Bindels (eds.) Child Nutrition in South East Asia, 185-197.
186
reported impainnent of immune mechanisms in vitamin A deficient animals,

explaining the role of vitamin A deficiency in increasing the susceptibility to
infection. Recent advances in technology have made it possible to induce single
vitamin A deficiency by retinoic acid cycling methods in experimental animals.
Using this technique, Darip et al. [II] demonstrated an increase in overall mortality
rates in rats by subcutaneous challenge with sublethal doses of rat lung worms. An
impaired function of the reticuloendothelial system (RES) and polymorphonuclear
leukocytes were reported by Ongsakul et al. in rats made vitamin A deficient by the
same method [12]. In all these experiments, the degree of vitamin A deficiency
induced is severe. Further, supplementation studies showed an enhancement of
immunological responses [13-15] and changes in mortality rates and severity of
clinical manifestations [16, 17] due to infections, but the dosage used was very high
and far-removed from the actual physiological situation, thus barring the extrapola-
tion of the results to humans. However, the situation in humans is more complex. It
has been observed that in poor communities where vitamin A deficiency is
widespread, infection rates are also high, one being the cause or consequence of the
other.
In children with severe PEM, mortality was found to be four times higher when
they had associated xerophthalmia [18]. Brown et al. found similar infection rates in
malnourished children with or without xerophthalmia, but observed a higher
incidence of bacteruria in the former group [19]. Thus, it is difficult to establish a
causal relationship between vitamin A deficiency and infection because of the
concomitant multiple nutritional deficiencies. For similar reasons, there is little
information available on the effects of severe vitamin A deficiency on the immune
function in humans.
While severe nutritional deficiencies occur in a relatively small percentage of
young children, marginal deficiencies of several nutrients exist in a majority of the
children in developing countries and could have several functional implications.
Keratomalacia due to severe vitamin A deficiency occurs in about 0.1 % of the pre-
school children and represents merely the tip of the iceberg of the problem of
vitamin A deficiency. On the other hand, a majority of the children in the community
suffer from a subclinical deficiency of vitamin A with or without manifesting clinical
signs of xerophthalmia and have serum retinol levels below 20 ).lg/dl. In recent years
research on the functional significance of marginal nutritional deficiencies has, in
general, received priority, while the effects of marginal vitamin A deficiency and the
administration of a single large dose of vitamin A on infection have in particular
drawn global attention. Recent epidemiological studies from Indonesia suggest a
significant association between mild vitamin A deficiency in children and morbidity
due to respiratory and gastro-intestinal infections and mortality [20]. Further, a
significant reduction in morbidity and mortality have also been demonstrated
following the administration of a single large dose of vitamin A [21]. A recently
reported Indian experience from retrospective analysis of morbidity data collected
from children residing in urban slums does not, however, totally agree with these
187
observations [22]. Therefore it is essential to carry out controlled double-blind

prospective studies in different populations living in different environmental
conditions, to identify the relationship between mild vitamin A deficiency and
infection in children.
Though changes in the mucosal epithelium and several of the immune
mechanisms were demonstrated in severe vitamin A deficient animals, the changes
were different in the animals subjected to milder degrees of vitamin A defiCiency.
Nauss et al. observed impaired transformation of splenic lymphocytes to mitogens
and antigens in the early stages of vitamin A deficiency in rats [9]. Chandra and Au
[10] later confirmed these observations. However, Hof and Wirsing [23] observed no
change in the activity of RES and antibody response in mice subjected to milder
degrees of vitamin A deficiency. It is only in recent years that attempts have been
made to investigate the effects of mild or subclinical vitamin A deficiency on
immune mechanisms in humans. Although it is not possible to totally eliminate the
effects of subclinical deficiencies of several other nutrients and effects of latent
infections, care was taken to exclude gross PEM, anemia and overt infection in all
these studies.
Immunological studies in children with vitamin A deficiency
Mohan Ram et al. [24] demonstrated reduced leukocyte and plasma lysozyme
activity in apparently normal children with ocular signs of vitamin A deficiency. The
leukocyte lysozyme levels returned to normal following therapy while there was no
change in plasma lysozyme activity. Bhaskaram and Reddy [25] observed a sig-
nificant decrease in the percentage of circulating T lymphocytes but with a normal in
vitro response to PHA stimulation in children having mild signs of xerophthalmia.
The T cell percentage returned to normal following therapy with vitamin A. In a
different study the decrease in T cell numbers in vitamin A deficient children was
confirmed. Further, a proportionate decrease in T4 (helper) and Ts (suppressor)
subsets of T cells was observed without, however, a change in the T/fs ratio [26].
Evaluation of humoral immune response by enumeration of circulating B cells,
estimation of immunoglobulins and antibody response to diphtheria and tetanus
toxoid revealed no changes in the deficient children compared to the controls [27].
Bhaskaram et al. investigated the macrophage function in apparently normal children
having low serum retinol levels, with or without associated clinical signs of xeroph-
thalmia. In vitro macrophage cytotoxicity and Interleukin 1 (ILl) production were
found to be similar between the deficient children and children with normal serum
retinol levels [28]. In a subsequent investigation, the effects of subclinical vitamin A
deficiency in children and the administration of a single large dose of vitamin A on
various facets of host resistance mechanisms were studied [29]. It was observed that
the neutrophil function, as assessed by hydrogen peroxide (H20 2) and superoxide
(02") production (Figure 1) and the antibody response to diphtheria and tetanus
188
8 I'm S. Retinol >20 !'g/dl.

liiWiajlil S. Retinol < 20 tg/dl.
Figure 1. H20 2 and 02" generation by neutrophils in vitamin A deficient children [28].
I = INITIAL F = FINAL
1M)
{ ••••••
xxx •
",,1
1..... }
620
)20
••• xxx • • •
lOO
'"'c.:"'
160 - ••• )f)()(X )(Xxx • • • • xxx X >-
....
j::
>- 120
0
2j::
z
<
&0 •••• l( XlOC xxx • • • x x x )( r-
.a - .... xxxx • • • )( ~
)( x
F F F F
DIPHTHERIA TETANUS DIPHTHERIA TETANUS
SERUM RETINOL <. 20 f&/dJ ; > 20 f&/dJ
Figure 2. Antibody response to D&T toxoid in vitamin A deficient children [28].
toxoid (Figure 2) and salivary secretory IgA (SIgA) were comparable between
children having serum retinol levels less than 20 Ilg/dl and those with more than 20
Ilg/dl. The T cell percentage was, however, significantly lower in the deficient
children in line with the earlier reports (Figure 3).
189
f.l'LJ S. Retinol :> 20 rg/dl•

"~ri<;?3 S. Retinol <: 20 fg/dl.
60
.;
1f
E
QI
40
~
c..
a
l-
20
Figure 3. Salivary SIgA and T cell percentage in vitamin A deficient children [28].
32 mill INITIAL
~FINAL
24
] 16
~
E
2
~ 8
Vitamin A 100,000 200,000 100,000 200,000

Dose IU DEFICIENT NORMAL
Figure 4. Serum Retinol levels following a single large dose of vitamin A [29].
These observations point out that except for the lysozyme activity and T cell
percentage, several other major host resistance mechanisms are not altered in
children with mild/subclinical vitamin A deficiency. Therefore, it is possible that
factors other than immune mechanisms might be contributing to a greater extent to
the reported increase in susceptibility of such children to infections, in certain
communities.
Recent investigators are focusing their attention more on the action of vitamin A
in regulating the structural and functional integrity of epithelial cells. In a recent
study Chandra observed an increase in the bacterial adherence to the respiratory
mucosal cells of vitamin A deficient children, suggesting a potentially increased risk
of mucosal penetration and systemic invasion by the bacteria [30].
190
g
f'.7Ll S. Retinol;:> 20 l'g/dl.
I~Sj!tllJ S. Retinol<=; 20 fg/dl.
Iillm Initial
C 6
's ~ Final
-1l'"
0
'"
U 4
-.!l
'@,
~ 2
c
Vitamin A dose
IU 100,000 200,000
Figure 5, Hz02 production by neutrophils following vitamin A therapy [29].
~ S. Retinol >20 fg/dl.

w(i@'{iJ S. Retinol'('20 fg/dl.
mill Initial
~ Final
Vitamin A dose.
JU 100,000 200,000
Figure 6, 02" generation by neutrophils after vitamin A therapy [29].
The absence of any major impairment in the specific immune mechanisms in

children with subclinical vitamin A deficiency is of practical relevance. Immunisa-
tion programmes which are powerful tools for the prevention of infections and
malnutrition therefrom would achieve the expected results even in communities with
widespread vitamin A deficiency.
Effects of the administration of a single large oral dose of vitamin A [29]
The administration of a single large dose of vitamin A (100,000 or 200,000 units)

improved the serum retinol levels in deficient children with no significant changes in
191
I - INITIAL F - FINAL
"0
{ ••••• lC~X
••• x
1
{
620
320
300
••••••
••••••
••••••
')(XXl(XXl(
XX ""'lIlIX ...,
•••••••
::- """
,. X)<)OI)(
}
:a
~
160
••• ••• xxx )t )O( X )(
xxx
II
••• • ••••• xxx lIXlIr-
)(XXX
I-
i=
>-
Q
g
120 - I-
....-
i=
~
80 - X lC 'JI •••••• J(J(XXXXlC
- ....- ....- -
- .....
.0 XXXXXX\( X'JIX'IIXXlC
XXXX
20 " )tlllC)(lIJ • ••• X xxx -
••• xx ••• X )(XJ(X
I F I F I F
DIPHTHERIA TETANUS DIPHTHERIA TETANUS
VITAMIN A DOSE 100,000 IU 200,000 IU
Figure 7. Antibody titres to D&T toxoid following vitamin A therapy [29].
the nonnals (Figure 4). It had different effects on different immune parameters.
However, the effects observed on a given immune response were independent of the
dosage used. It was observed that there was a significant increase in HzOz concentra-
tion (Figure 5) with no change in 02' generation (Figure 6), enhancement in antibody
titres to D&T toxoid (Figure 7) and an increase in ILl elaboration by macrophages
following the administration of vitamin A, and these were observed in all the
children irrespective of their initial serum retinol levels. These effects could be due
to the adjuvant action of vitamin A. However, the functional significance of the
enhancement of these responses as contributors to decrease the susceptibility to
infections needs to be evaluated. The T cell percentage which was lower in vitamin
A deficient children was restored to nonnal but there was no change in the T cell
percentage of children who had nonnal serum retinol levels (Figure 8). This
observation points out the role of vitamin A in maintaining the circulating T cell
number but fails to attribute a non-specific immuno potentiating effect to a single
large dose of vitamin A administration. These observations are, however, not
comparable with the results obtained in several other studies in animals as well as
humans where large doses of vitamin A led to the enhancement of cell mediated
192
Ell Healthy Children

so rm S. Retinol> 20 ffJdl.
E'illi.$:i S. Retinol L. 20 fF.'dl.
~ Initial
~ Final
20
Vitamin A dose 100,000 200,000

IU
Figure 8. Circulating T cell percenJage following vitamin A therapy [28].
If'22l S. Retinol > 20 pgIdl.

~ S. Retinol < 20 !"d!.
60 ~ Initial
~ Final
C
'ij
ItO
0
~
DO
E
i. 20
Vitamin A dose 100,000

IU
Figure 9. Salivary IgA following viJamin A therapy [29].
immunity. However, the dosage, frequency of administration and the nature of the
vitamin A preparation which are important factors in deciding the type of immune
response, are different in different studies.
A consistent and significant decrease in salivary SlgA was observed following
vitamin A therapy in the deficient as well as in the normal children (Figure 9).
Karnm et al. reported dryness of mucus membranes due to decreased mucus
production, cheilosis etc. [31], suggesting the adverse effects of administration of
large doses of vitamin A on mucosal epithelium. It is possible that similar distur-
bances could be caused by a single large dose of vitamin A, also leading to an
impairment of the generation of SlgA. However, this observation needs to be
confirmed in other secretions for a better understanding of its functional significance.
193
Summarising the results of these two studies, it was observed that a single large
dose of vitamin A had nonspecific potentiating effects on phagocytic function and
antibody response in deficient as well as control children; there was a normalisation
of T cell numbers of the deficient children with no effect on that of normal children
and a decrease in salivary SIgA in both the groups. The practical relevance of
relating these effects on the immune system to the morbidity of children which is the
result of a complex interplay of several host and environmental factors in the
community should be carefully evaluated.
Further, the effects observed in both groups of children were comparable between
the two dosage schedules and thus were dose independent. In view of the observa-
tions that the changes in serum retinol levels following the two different dosage
schedules were also comparable in the deficient and normal children respectively,
the possibility of achieving more uniform immunological effects with smaller doses
of vitamin A should be considered.
Immune responses in iron deficient children
Nutritional anemia is one of the public health problems in the developing countries.
Although no age group is exempt, women and pre-school children constitute the
most vulnerable segments of the population suffering from nutritional anemia, the
important cause for which is identified as iron deficiency [32].
Optimal iron status is essential for several body functions. Besides being a
hemopoietic factor, iron plays a crucial role in modulating the function of many cells
in the body. Iron is essential for the function of several iron containing and iron
dependent enzymes, thus regulating oxidative metabolism oftissues [33]. Robins and
Pederson [34] suggested a crucial role for iron in mitosis. A reduction in DNA and
RNA content as well as protein synthesis was observed in bone marrow cells [35].
As a result of these varied and widespread functions of iron, the functional distur-
bances due to iron deficiency are also widespread and include the immune system.
Anemia and infection are frequently associated with each other. It is generally
believed that susceptibility to bacterial infections increases in iron deficient subjects.
There have been several epidemiological, clinical and experimental studies trying to
link: the susceptibility of iron deficient hosts to various infections. These studies have
been discussed extensively in a recent review [36]. Several variables in the study
design, parameters selected, varying environmental factors and a number of other
variables like co-existent nutritional deficiencies are perhaps responsible for the
varied results obtained by different investigators in different populations, under
different environmental conditions. However, recent clinical studies demonstrating
the occurrence of opportunistic infections among iron deficient subjects support the
hypothesis of altered immuno competence in these subjects, leading to an increase in
susceptibility to infections. Reversal of the clinical and immunological defects with
adequate iron therapy further strengthens the hypothesis [37].
194
Several investigators have demonstrated impaired leukocyte function in terms of

decreased intracellular bacterial killing by neutrophils which improved on the
restoration of the iron status to normal [38, 39]. Srikantia et al. [40] observed a
significant decrease in the bactericidal activity of leukocytes in children having Hb
levels less than 10 gldl and in children having latent iron deficiency without change
in Hb levels. Prasad [41] reported a decrease in the MPO content of neutrophils
obtained from children with iron deficiency which could to some extent explain the
decreased intracellular bacterial killing capability. Circulating B cell numbers and
immunoglobulin synthesis were observed to be normal in iron deficient children [38,
39]. Bagchi et al. demonstrated an adequate antibody response by anemic children on
challenge with diphtheria and tetanus toxoid [42]. These observations indicate
normal synthesis of immunoglobulin in response to infection or antigenic challenge
in iron deficiency. Several investigators reported varying results on complement
levels in iron deficient children [38, 39]. The majority of the studies indicate normal
levels of C3, C4 and total hemolytic complement levels in anemic children. However,
Jagadeesan and Reddy [43] observed a decrease in total hemolytic complement
levels as well as C3 levels in iron deficient children which returned to normal
following iron therapy. Joynson et al. [44] were the first to describe the effects of
iron deficiency on cell mediated immune responses in human volunteers. Subse-
quently, similar observations were reported by several investigators in severely iron
deficient children [25, 38]. Srikantia et al. [40] reported significant decrease in T cell
numbers and response to in vitro stimulation with PHA in children having Hb less
than 10 gldl and in children having latent iron deficiency but with no evidence of
anemia. Iron therapy restored these functions to normal levels [45].
To summarise, despite the controversies, several of the laboratory studies indicate
that iron deficiency adversely affects immuno competence. A direct correlation of
these results obtained from laboratories and morbidity from infections in the
community may not be possible because of the often associated multiple nutrient
deficiencies and unsatisfactory environmental conditions. However, the potential
dangers of the immuno suppression induced by iron deficiency acquire greater
significance in the presence of other factors that are conducive to promoting
infective diseases. The adverse effects of latent iron deficiency on the immune
function warrant effective implementation of intervention programs in poor com-
munities.
Summary and conclusions
Deficiencies of vitamin A and iron are common among children in developing

countries. Several epidemiological and clinical studies have shown an association
between the deficiencies of these nutrients and susceptibility to infections in
children. Single nutrient deficiencies produced in experimental animals revealed
their immuno suppressive effects. Apart from the effects of severe deficiencies, the
195
functional significance of marginal deficiency of nutrients in humans is important

and is of practical relevance.
Vitamin A deficiency and the immune function
Children suffering from severe vitamin A deficiency manifesting as keratomalacia

constitute 0.1 percent of the pre-school population of children and these children in
India invariably have an association with severe protein energy malnutrition. A
majority of the children in the community suffer from marginal or subclinical
vitamin A deficiency with or without clinical manifestations of xerophthalmia. In
view of the recent reports identifying a close association between marginal vitamin
A deficiency and increased incidence of infections, the immunological basis for the
increased susceptibility to infections in such children was investigated. The func-
tional status of neutrophils, humoral and cell mediated immune mechanisms,
macrophage functions and secretory IgA content of saliva were determined in
children having serum retinol levels of less than 20 )lg/dl and compared with those
having levels of more than 20 )lg/dl. It was observed that the elaboration of hydrogen
peroxide (~02) and superoxide anion (02") by neutrophils, the specific antibody
response to antigenic challenge, the cytotoxic function and Interleukin-l (ILl)
production by in vitro activated peripheral blood monocytes and SIgA content of
saliva were all comparable between the two groups of children. Circulating T
lymphocyte percentage was significantly decreased in deficient children as compared
to the controls. However, the in vitro response of the T cells to stimulation with PHA
was found to be unaltered. The subsets of T lymphocyte population T4 (helper) and
Tg (suppressor) cells showed a proportionate decrease but the T4/Tg ratio was not
changed. These observations indicate that marginal vitamin A deficiency in children
does not alter several of the host resistance mechanisms to any significant extent
except T cell numbers and suggest that mechanisms other than immuno-suppression
might be contributing to the observed increase of morbidity in such children. Further,
these observations point out that even in communities where vitamin A deficiency is
widespread, massive immunisation programs can achieve their expected benefits in
the prevention of malnutrition.
The administration of a single large dose of vitamin A either as 100,000 or
200,000 IV showed dose independent, but varied responses on the various immune
parameters. There was potentiation of phagocytic function and antibody response
with no accentuation of T cell number. The SIgA content showed a marked decrease.
In view of these results, the usefulness of a single large dose of vitamin A to reduce
morbidity due to infection among children in the community should be carefully
evaluated. Further, in view of the independent nature of the results, the possibility of
achieving more uniform immunological effects with smaller doses of vitamin A
should be considered.
196
Iron deficiency and immune function
The available epidemiological clinical and experimental evidence describing the

association between iron deficiency and increased risk of infections in children is
conflicting. However, studies demonstrating significant impairment of various facets
of host resistance mechanisms in iron deficient children which are corrected with a
replenishment of the iron status suggest the role of iron in maintaining optimal
immune status. Children suffering from severe iron deficiency anemia had decreased
bactericidal activity of neutrophils and impairment of cell mediated immune
mechanisms. Antibody responses to challenge diphtheria with and tetanus antigens
were found to be adequate but complement levels were low. The impairment of these
parameters was observed in children having Hb levels less than 10 g/dl and all the
responses were restored to normal after adequate therapy with iron. Latent iron
deficiency also had similar effects. These studies suggest the need to strengthen the
iron supplementation programs in developing countries.
References
1. McCollumEV: 1 AmerMed Assoc 68:1379,1917.
2. Green HN and Mellanby E: Brit Med 1 2:691, 1928.
3. Turner RG, Anderson DE and Lowe ER: 1 Infect Dis 46:328, 1930.
4. Erasmus 1, Scott ML and Levine PP: Poultry Sci 39:565, 1960.
5. Lawler HI: Amer 1 Hyg 34D:65, 1941.
6. Roos A, Hegsted DM and Stare Fl: 1 Nutr 32:473, 1946.
7. Waldrong PN, Simpson CF, Cox DD and Hanns RH: Poultry Sci 42:274, 1963.
8. Krishnan S, Bhuyan UN and Talwar GP: Immunology 27:383, 1976.
9. Nauss KM, Mark DA and Suskind RM: 1 Nutr 109:1815, 1979.
10. Chandra RK and Au B: Nutr Res 1:181, 1981.
11. Darip MD, Sirisinha S and Lamb A: Proc Soc Exp Bioi Med 161:600,1979.
12. Ongsakul M, Sirisinha S and Lamb AI: Proc Soc Expt Bioi & Med 178:204, 1985.
13. Medawar PB, Hunt R and Mertin 1: Proc R Soc London. Ser 206:265, 1979.
14. Seifter E and Rettura G: 1 Nat! Cancer Inst 67:467,1981.
15. Moriguchi S, Werner L and Watson RR: Immunology 56:169, 1985.
16. Cohen BE and Elin RJ: 1 Infect Dis 129:597, 1974.
17. Chew BP, Luedecke LO and HolpuchDM: 1 Dairy Sci 67:2566,1984.
18. McLarenDS, Shirajian E, Tchalian H and Khoury G: Am 1 Clin Nutr 17:117,1965.
19. Brown KH, Gaffer A and Alamgir SM: 1 Pediatr 95:651,1979.
20. Sommer A, Katz 1 and Tarwotjo I: Am 1 Clin Nutr 40: 1090, 1984.
21. Sommer A, Tarwotjo I, Djunaedi E, West lr KP, Loeden AA, Tilden R, Mele L and Aceh Study
Group: Lancet 1:1169,1986.
22. Milton RC, Reddy V and Naidu NA: Am 1 Clin Nutr 46:827, 1987.
23. HofH and Wirsing CH: 1 Nutr Sci 18:221, 1979.
24. Mohanram M, Reddy V and Mishra S: Br 1 Nutr 32:313, 1974.
25. Bhaskaram C and Reddy V: Brit Med 1 3:522, 1975.
26. Bhaskaram P: In: Taylor TG, lenkins NK (eds) Proc XIII International Congress of Nutrition. lohn
Llbbey, London, 1985,p. 132.
27. Kutty PM, Mohanram M and Reddy V: Acta Vitaminol enzymo13:231, 1981.
28. Bhaskaram P. Sharada K. Sivakumar B. Visweswara Rao K and Nair M: Nutr Res 9:35, 1989.
29. Bhaskaram p, lyothi SA. Visweswara Rao K and Narasinga Rao BS: Nutr Res 9:1017, 1989.
30. Chandra RK: Br Med 1 297:834, 1988.
197
31. Kamm JJ, Ehmann CW and Ko AS: In: Spenn MB, Robberts AB and Goodman DS (eds) Retinoids.
Academic Press Inc, New York, 1984, p. 287.
32. World Health Organization. Scientific Group: Tech Rep Ser 405, 1968.
33. Cohen E and Elvehjin CA: J Bioi Chern 107:97, 1934.
34. Robbins E and Pederson T: Proc Natl Acad Sci USA 66:1244,1970.
35. Hershko CH, Karsai A, Eylon Land Izak G: Blood 36:321, 1970.
36. Bhaskaram P: In: Chandra RK (ed) Contemporary issues in clinical nutrition; nutrition and
immunology, Alan R Lirs Inc, New York, 1988, 11:149
37. Chandra RK and Vyas D: In: Chandra RK (ed) "Critical reviews in tropical medicine" Plenum New
York, 1984,2:93.
38. Chandra RK and Saraya AK: J Pediatr 86:899, 1975.
39. MacDougall LG, Anderson R and Menab GM: J Pediatr 86:833, 1975.
40. Srikantia SG, Prasad JS, Bhaskaram C and Krishnamachari KA VR: Lancet 1:1307, 1976.
41. Prasad JS: Am J Clin Nutr 32:550,1979.
42. Bagchi K, Mohanram M and Reddy V: Br Med J 280:1249,1980.
43. Jagadeesan V and Reddy V: Ind J Med Res 70:745,1979.
44. Joynson DHM, Walker MD, Jacobs A and Dolby AE: Lancet 2: 1058, 1972.
45. Bhaskaram P, Prasad JS and Krishnamachari KA VR: Lancet 1:1000, 1977.
Supplementary feeding in
childhood diarrhea
IGNAnus SUDIGBIA
Introduction
Diarrheal disease is still a major problem in developing countries such as Indonesia

which has a high incidence and mortality rate. The National Program on Oral
Rehydration was started in order to decrease the incidence and the mortality of
childhood diarrhea in Indonesia. In general, acute diarrhea is 'self-limited', but when
it is followed by complications, the management (feeding management) is not
normal [12].
Harries [9] stated that the complications of diarrheal disease are influenced by
mucosal degradation, bacterial overgrowth in the intestine, intolerance and malab-
sorption and intraluminal metabolic substances.
The interaction between diarrhea, infection and nutrition is an important factor
with regard to the use of supplementary feeding in the management of childhood
diarrhea. In developing countries, the five D's rule [18] in managing diarrheal cases
is still useful:
dehydration treatment
diagnostic procedure
dietetic procedure
- drugs
disacharidase deficiency
The dietetic procedure as one of the important factors in the five D's rule should
focus on:
feeding intake
intolerance and malabsorption
the protein loss problem and
catabolism
200
Nutritional consequences of diarrhea
According to Chen [3] the nutritional consequences of diarrhea are thought to be due
to several factors:
1. decreased intake of food, resulting from loss of appetite and the withholding of
food as practised in communities as a measure to control diarrhea,
2. loss of major nutrients as well as vitamins and minerals in the stool because of
rapid transits or malabsorption, and
3. increased catabolism in response to infection.
Decreased intake offood
Reduction of food intake during diarrhea may be due to child anorexia or maternal
(and health personnel) withholding food or both. Anorexia could be a consequence
of clinical disturbances, including dehydration, fever, vomiting and abdominal
discomfort. Food withholding by mothers could be a response to the child's anorexia
or culturally defined changes of dietary practices either in response to illness or as a
component of the control of diarrhea. In many situations, the net outcomes are the
cessation of breast-feeding, a reduction of the quantity and the quality of weaning
foods and sometimes a restriction in the lactating mother's food.
Effect of diarrhea on absorption and direct nutrient loss
Malabsorption of sugar (lactose and glucose) and fats is well substantiated and solid
evidence exists for malabsorption of nitrogen (amino acids and protein).
Some vitamins (water and fat soluble vitamins) and trace minerals are also
malabsorbed. The mechanism by which diarrhea causes malabsorption has not been
clearly elucidated; bacterial overgrowth, bacterial fermentation of sugar and
disturbed bile metabolism could be invovled. Certain diarrheal diseases may also be
associated with the reduction of the absorbing surface due to mucosal damage and
increased transit time of the chyme.
Chung [4] stated that absolute absorption was improved during a higher intake
even when stool loss increased.
Molla [17] showed that the absorption of carbohydrates, fats and nitrogen was
reduced during acute periods in comparison to the convalescent period of childhood
diarrhea caused by cholera, E. Coli and others.
Dossetor [6] reported protein loss during acute measles enteritis; morphological
damage could be expected to be associated with protein and other nutrient loss.
Rahman et al. [21] showed more subtle protein losses in patients with diarrhea.
201
Increased catabolism
In general, a systemic infection disturbs all normal metabolic and endocrine factors.
The response is both anabolism and catabolism. Catabolism overshadows anabolism
during clinical disease and is related to the severity and duration of fever.
Briscoe [2] estimated that each degree of fever causes a basal metabolic increase
of 5.0-8.2%.
Beisel et al. [1] showed schematically the sequence of nutritional responses that
evolve during the course of a 'typical' general febrile infectious illness. The
conclusion was a negative balance of nitrogen and other nutrients.
Powanda [20] noted that diarrhea of infectious origin induced an average daily
negative nitrogen balance of 0.9 gr/kg bodyweight per day.
Supplementary food
The feeding pattern and food habits amongst nations, even amongst persons, is quite
different, due to several factors:
the geographical factor:
a city or country with different feeding patterns
the availability offoods factor:
agricultural area, industrial area or trading area
- the condition of the baby:
the appetite of the baby, the condition of the illness
the family condition:
socio-economic factors
- socio-cultural factors
A statement on infant feeding has been recommended by the Indonesian Working

Unit for the Promotion of Breast-Feeding (Semarang 1977):
- breast-feeding is best for healthy and sick babies and should be given until 2 years
of age
in case breast-feeding is unavailable, a proper formula should be introduced
the ideal infant feeding formula, especially in diarrheal disease should be easily
digested, easily absorbed, locally available, cheap in price and easily served
The purpose of 'supplementary food' for a baby is to provide:

energy resources
protein, vitamins and minerals required for growth
a supplement for the bulk and the taste of the food
Supplementary food should be adjusted to the biological needs of the children and to
the infant's acceptability and appetite.
202
In general, solid food should be started at the age of 3-4 months or at least at the
age of 6 months.
According to Jelliffe [11] the Indonesian infant's feeding pattern is the following:
breast-feeding is generally used for young infants, and is commonly prolonged for
one to two years or until the next pregnancy
sometimes a little honey is given immediately after birth
animal milk appears to play but a small part in infant feeding, as little is available.
As in other developing countries, it is used in low dilution, rather as a flavouring
than as a food
the introduction of semi-solid foods often begins early in life. Rape mashed
banana or mashed rice and banana are most commonly used. Milk cereals are
often introduced at 3-4 months of age. Later the child is given some adult food
such as various types of 'nasi tim', composed of steamed rice to which may be
added a variety of green vegetables, tomato, meat (chicken liver) and soya bean
products such as 'tempe' (fermented soya cake) or 'tahu' (soya bean curd). Some
people, mostly uneducated, believe that eating meat or fish results in worms.
Gribosky [7] reported that a mixed diet ('BRAT' diet) of bananas, rice cereal, apple
sauce and toast was recommended.
Hariyono [8] conducted a nutrition survey in 3 rural villages of central Java and
the results were:
breast-feeding is very common in the villages and at the age of less than one year
only 2.6% received milk formula
the percentage of children weaned at the age of 1-2 years was 27%; 2-3 years
51 % and more than 3 years 22%
the percentage of children that received solid food at the age of less than one
month was 16.5%; 1-5 months 29%; 6-11 months 29% and more than 1 year
25.5%. The solid food consisted of bananas or bananas pounded with rice.
In 1983, Sukirman conducted a nutritional survey in Central Java, with the following
results: The prevalence of breast-feeding was 84.6% at one month of age, 53.7% at
three months and 34.7% at six months. Extra fluid (except milk formula) was
introduced at the age of 2-3 months. Solid food was given to 60% of infants before
three months of age. Infants older than 3 months had a varied feeding pattern. There
was a significant correlation between the use of weaning food and the nutritional
status (at 3-6 months of age). There was a highly significant correlation between
breast-feeding and the incidence of diarrhea at the age of 3-6 months.
According to Lizal et al. [13] the introduction of fruit and solid food at a proper
time was found in 32.8% (at the age of 1-3 months) and 45.3% (at the age of 4-5
months) of cases, respectively. Better results were noted among the higher socio-
economic groups with regard to the time of introduction as well as its variation and
combination of fruit and solid food.
203
In general, several studies on the pattern of infant feeding in Indonesia can be

summarized as follows:
breast-feeding is still common practice in the rural community, while there is a
decrease in the urban areas
almost all (more than 90%) of the infants are given supplementary solid foods,
particularly mashed bananas and rice at an early age, from several days after birth
to one month. As in other developing countries, the popularity of infant formulas
of various kinds of milk products is growing in Indonesia, especially in urban
areas.
Feeding practice in childhood diarrhea
Martorell [15] showed that recurrent attacks of diarrhea lead to significantly more
growth retardation compared to that associated with other illnesses.
Rowland [22] reported that diarrhea is the most important infectious disease
causing growth retardation in children. Several investigations demonstrated that
appropriate child care practices in terms of nutrition during episodes of diarrhea are
very important.
Mahalanabis [14] reported significantly reduced mortality rates in breast-fed
infants with diarrhea in eight Latin American countries. There is good evidence that
the nutritional status in breast-fed children and diarrhea-related mortality are closely
related.
Mahalanabis showed that oral rehydration treatment for acute diarrhea in children
followed by early feeding and unrestricted breast-feeding was highly successful in
terms of cessation of diarrhea and weight gain. He also demonstrated excellent
weight gain in the course of cholera with oral rehydration therapy followed by a
more liberal intake of food as soon as the patient was able to eat.
Chung and Viscorova [5] highlighted the beneficial effect of early feeding during
acute episodes of diarrhea in infants; mean duration of diarrhea was significantly
lower and mean weight gain was significantly higher.
Natural history of acute diarrheal disease
Lebenthal [12] stated that acute diarrheal disease is generally a 'self-limited' disease,
but many factors are involved in the persistence of diarrhea; the latter in tum may
lead to serious complications such as severe malnutrition, repeated episodes of
infections and even the death of the child.
The main underlying cause of the continuation of diarrhea appears to be a
prolonged mucosal injury of the small intestine. The factors responsible for the
prolonged mucosal injury are the ineffective repair of the small intestinal villi and
disarray of the brush border of mature enterocytes. Brush border activities are
204
diminished, active transport of nutrients is disturbed and malabsorption follows.

Further results of prolonged mucosal injury of the intestine are maldigestion and
malabsorption of foods, bacterial overgrowth, deconjugation of bile acids, an
increase in macromolecular absorption, less gut hormones synthesized in the
epithelium and decreased SIgA production.
When nutritional treatment during acute diarrhea is considered, special attention
should be given to the following factors:
the food should be easily digested and absorbed

there should be a minimal risk of sugar intolerance, fat malabsorption and protein
allergy
the food should be sufficient in energy, protein, vitamins and minerals
the food should be cheap, locally available and easily accepted.
Special milk formulas such as low lactose or free of lactose formulas (MeT for-
mulas) are available on the market in metropolitan cities in Indonesia, but the prices
are too high for the common people. When facing the problem of supplementary
feeding in diarrheal disease traditional foods should be used.
Supplementary feeding in childhood diarrhea
Within the community, supplementary feeding in childhood diarrhea comprises of:
'Nasi tim'
'Nasi tim' is usually served as steamed rice combined with meat (beef or chicken)
and without vegetables to avoid fibre. Generally the reduction of food intake
(including 'nasi tim') during diarrhea episodes may be due to child anorexia or
maternal (and health personnel) withholding of food.
special food formula in trade-packing
* rice porridge with chicken
* chicken rice with milk
* cereal formula; containing rice, soya and maize and a small amount of milk.
Milk cereal formulas are very popular in the community but there are problems
due to the lactose contents of the milk.
- food formula with tempe
Tempe is a soybean fermented product first developed in Indonesia centuries ago,
in which soybeans are soaked, dehulled, partially cooked and inoculated with
moulds belonging to the genus Rhizopus. During I to 2 days of incubation in a
warm place, the soybeans are knitted into a tight compact cake by the fibrous
mould mycellium. The protein content is about 49% of dried tempe [16]. Due to
its high nutritive value, Indonesian people have had this product in their diet for
centuries. Tempe is also becoming an important food in the USA, the Nether-
lands, Malaysia and Sri Lanka. Tempe is one of the first vegetable foods shown to
205
contain nutritionally important amounts of vitamin BI2 (30/-lg). It has been

reported that tempe has a beneficial effect in patients with dysenteria due to
antibacterial activity of growth inhibitors produced by Rhizopus oligosporus
during tempe fermentation. Although antibacterial tests indicated that the
compounds produced by Rhizopus oligosporus do not exhibit a broad spectrum
activity, they are very active against some gram positive bacteria.
Hermana [10] reported that a food supplement consisting of 70% rice and 30%
tempe gave better results on weight gain in school children with protein-energy
malnutrition than a mixture of 70% rice and 30% soybean.
Due to its low cost, unique characteristics and familiarity, tempe has great
potential as a key protein source and as a basic ingredient for food supplements.
The effect of tempe as a food supplement in the treatment of diarrhea was

demonstrated in two studies.
Mien [16] studied a food formula with tempe (wheat, sugar, comoil and tempe) in
the nutritional treatment of chronic diarrhea. There was a significant effect on the
cessation of diarrhea, even in malnourished children. The acceptability of the tempe
food was better compared with milk cereal formula.
Sudigbia et al. [23] conducted a study with a tempe food formula in 17 mal-
nourished children with chronic diarrhea and malabsorption problems. In the second
week of treatment 47.1 % of the children showed a significant weight increase;
another 11.8% showed a good response after 4 weeks of treatment, while 17.6% had
no response after 4 weeks.
References
1. Beisel: Am J Clin Nutr 30:1236,1977.

2. Briscoe J: Am J Clin Nutr 32:648,1979.
3. Chen L and Scrimshaw NS: Diarrhea and malnutrition, Plenum Press, New York, 1983.
4. Chung A: J Pediatr 33:1,1948.
5. Chung A and Viscorova B: J Pediatr 33:14,1948.
6. Dossetor JFB: Br Med J 2:592, 1975.
7. Gribosky, J: Gastrointestinal problems in the infant, 2nd Ed WB Saunders Co, Philadelphia, USA,
1983.
8. Hariyono R: MCH Services and breast-feeding practices. Proceedings of traveling seminar on recent
developments in breast-feeding, 1976.
9. Harries JT: Essentials of paediatric gastroenterology, Churchill Livingstone, Edinburgh, London,
New York, 1977.
10. Hermana: The influences of the consumption of food formulas with tempe or soybean on the
malnourished under five children in a rural area of Bogor. Bogor Institute of Agriculture, 1983.
11. Jelliffe: Infant nutrition in the subtropics and tropics. World Health Organization, Geneva, 1968.
12. Lebenthal E: Prolonged small intestinal mucosal injury as a primary cause of intractable diarrhea. In:
Lebenthal E (ed) Chronic diarrhea in children, Nestle Nutrition Workshop, Raven Press, New York,
1984.
13. Lizai: Breast-feeding and infant feeding practices in Ujung Pandang. Proceedings of traveling
seminar on recent developments in breast-feeding, Semarang, 1987.
14. Mahalanabis D: In: Chen Land Schrimshaw NS (eds) Diarrhea and malnutrition, Plenum Press,
206
New Vork. 1983.
IS. Manorell R: Am J Dis Child 129:1296, 1975.
16. Mien KM: The role of infant feeding formula with tempe in the management of under five of age
diarrheal children. Thesis in the Post Graduate Programme of Bogor Institute of Agriculture, 1987.
17. Molla: Food intake during and after recovery from diarrhea in children. Plenum Press, New York
and London, 1974.
18. Morley D: Pediatric priorities in the developing world, Butterwortbs, London, 1974.
19. National Seminar on Rehydration III. Conducted by Department of Health of the Republic of
Indonesia in collaboration with Coordination Board of Indonesian Pediatric Gastroenterology,
Semarang, Recommendation, December 1982.
20. Powanda MC: Am J Clin Nutr 30:1254,1977.
21. Rahman MM et al: In: Chen L and Scrimshaw NS (eds) Diarrhea and malnutrition, Plenum Press,
New York, 1983.
22. Rowland MGM: Br J Nutr 37:441,1977.
23. Sudigbia I et al: Chronic diarrhea due to the persistence of acute diarrhea, Scientific Meeting of
Coordination Board of Indonesian Pediatric Gastroenterology, Surakarta, November, 1986.
Paper by A.G. Soemantri Hardjojoewono
D. Sinniah:
I would like to know whether there is any information on the type of possible inhibitors of iron
absorption which are present, especially in Indonesia, for instance due to excessive tea drinking. I
would like to hear your comments on the possible role of infection on iron absorption. I think that it is
important to promote breast-feeding to reduce the incidence of iron deficiency in infancy.
A.G. Soemantri Hardjojoewono:

Thank you for your question. We have no data on iron absorption. The role of infection on iron
absorption is not clear; there is always the effect of anorexia in children with infections. I fully agree
with your comment on breast-feeding.
C. Gopalan:
Considering that the Asian diet is largely cereal based and that the bioavailability of iron from such a
diet is low, we understand that practically all groups are affected, not just women. What are your
views about the current strategy of fortification of salt with iron?

I have no personal experience, but I know that there are some trials in Indonesia. Perhaps Prof.
Karyadi will tell us about it.
D.Karyadi:
There were a few limited food trials in Indonesia organized by the Ministry of Labour in the period
1979-198 I. There is the technical problem of the compatibility of iodine and iron, both added to salt.
The shelf life cannot be long. There is also the problem that salt is produced locally. I know that there
is new information from India.
C. Gopalan:
You can add both iron and iodine to salt, the bioavailability is not compromised. It is still at the
research stage. In those areas where there is no goitre problem, you can add only iron. The results are
good, you get an increase in hemoglobin concentration. I agree that there are practical problems when
you go from laboratory-scale to large-scale application.
P. Bhaskaram:
The iron fortified salt has been tried in some villages in India and the results were quite acceptable.
Preliminary results show increased hemoglobin levels and now further studies are underway.
H.KA. Visser and J.G. Bindels (eds.) Child Nutrition in South East Asia. 207-213.
208
s. Drahaman:
It is interesting that in your country the basic factors in the etiology of iron deficiency anemia are
socio-economic factors and the political environment. What do you mean by political environment?
My second question is how prevalent is hookworm infection in your country?

The prevalence of hookworm infection is very high, especially in the rural areas because of the
sanitary conditions. There is chronic blood loss because of the hookworm infestation. In the local
health centers iron is not always available. The local government should give it a high priority, that is
why I stressed the political environment.
L.Mata:
We should not forget that iron is a growthfactor for bacteria. In human milk the iron is bound to a
specific protein, and there is no free iron available for bacterial growth. Iron deficiency anemia is
particularly seen during the weaning period. In Guatemala I found iron deficiency anemia in the age
period between 9 months and 2.5 to 3 years when the children are heavily infected by bacteria, viruses
and parasites. At about 3-3.5 years of age the children recovered and at 5 years of age they no longer
had any anemia. In my opinion interleukines play an important role in the defense mechanisms against
infection.
Paper by T. Sadjimin
L. Mata:
I am very happy with the critical approach you used for this excellent review, since one always should
be rather sceptic in dealing with these issues. In spite of the fact that you consider Lozoffs studies
having the most appropriate design, I would like to know what could be improved if you were to
repeat these studies.
T. Sadjimin:
Firstly, there is a lack of clear dose-response investigations. Secondly, it should be aimed for that the
level of iron deficiency is measured before any infant will enroll in the study. In this way we would be
able to differentiate between changes in infant development due to severe or to moderate iron
deficiency. Also the changes in baseline levels should be assessed for which is necessary to take into
account on evaluating the effects of iron supplementation. Finally, there is the point of doing a
randomized controlled trial or not. If one would again find significant differences in a non-randomized
contrOlled, "pre-post" designed trial, there are still lots of confounding variables which weaken the
scientific value of the study.
W.C.Liu:
In Taiwan, nutrition is no longer a problem, but iron deficiency in infancy is still noticed to some
amount. Therefore, we expect that some genetic factors are also involved.
T. Sadjimin:
I fully agree because I am sure that a substantial proportion of the prevalence figures for anemia
especially in South Asia is caused by genetic factors. With respect to Indonesia, I am aware of
investigations on hemoglobinopathy done some 30 years ago.
A. Li Ming Cheng:
We should not forget in this respect that the terms iron deficiency and anemia are only partly
synonimous. You can find iron deficient infants without being anemic and also because of a
considerable incidence of hemoglobinopathy in this area, many children are anemic without being iron
deficient; they may of course also have both. I just would like to mention a study we did in Hong
Kong with about 600 adolescents and found that the incidence of iron deficiency was below 3% if
based on serum levels. Based on hemoglobin levels, the incidence for anemia was higher mainly due
to hemoglobinopathy.
209
A.M. Malia:
It is not easy to point out which has the highest incidence in Pakistan: iron deficiency anemia or
hemoglobinopathy because of the constant unity of both. Usually we do a preliminary test serum
ferritin and find that it is low. If the serum ferritin level is high, we exclude iron deficiency anemia and
perform hemoglobulin electrophoresis.
D. Sinniah:
To my opinion, it is important at this stage to clearly differentiate hemoglobulinopathy from
thalassemia. The former refers to the presence of abnormal hemoglobulins, while the latter results in
the synthesis of abnormal proportions of the normal hemoglobulin subunits. The most common
hemoglobinopathy in our region is hemoglobulin E. In Malaysia we find this problem mostly with the
Malay popUlation and Thailand, up to 67% of the popUlation is affected by hemoglobulin E.
W.C.Liu:
In Taiwan, the most prevalent thalassemia is the beta-thalassemia trait with an incidence as high as
about 4-5%. Blood samples from cases with anemia are now routinely investigated by hemoglobulin
electrophoresis to differentiate between beta- and alpha thalassemia and to look for hemoglobulin H. If
we have these results at our disposal, we can clearly diagnose whether an anemia is caused by iron
deficiency or not.
M.A. Hussaini:
There now is growing consensus about the classification of iron deficiency based on hematological
criteria. At least 3 parameters should be determined: transferrin saturation, free erythrocyte porphyrins
(FEP) and serum ferritin. If two of these indicators are abnormal and the hemoglobulin levels is below
the diagnostic criterium, we are dealing with iron deficiency anemia. If the hemoglobulin is normal
and two of the three indicators for the iron status are abnormal, we have to do with iron deficiency
without anemia
T. Sadjirnin:
Thank you very much for these contributions. I can only once more emphasize the importance of
clearly defining anemia, iron deficiency and any combination. If we want to have a proper diagnosis of
anemic iron deficiency and want to assess the efficacy of any supplementation program, we really
should measure the iron status with sensitive and specific tests, and should exclude anemia caused by
other factors than iron deficiency.
Paper by P. Bhaskaram
W.C.Liu:
Three questions. Did you measure the lysozyme activity? Did you estimate the interleukin-l
production, both intracellular and in the extracellular supernatant, also after stimulation? Did you
perform the opsonization test?
P. Bhaskilrarn:
In this particular assay we have not carried out the lysozyme test. We did measure interleukines, both
intra- and extracellular. We have done the opsonization test in malnourished children and found
decreased values.
M A. Husaini:
What is the mechanism of action of vitamin A on the immune system?
P. Bhaskilrarn:
Vitamin A has an adherent function, it sticks to the cell membrane, and by this way it may activate
macrophages. This leads to stimulation of immune key-cells and enhances interleukin-l production.
This is how I explain our observations.
210
S. Drahaman:
Your studies are very important. Vitamin A deficiency results in higher rates of infection and higher
mortality. We heard from Prof. Mata about the prevalence of infection in developing countries.
Malnourished children often have a decrease in appetite and there is evidence that iron supplementa-
tion may stimulate appetite in these children. Do you recommend vitamin A and iron supplementation
for our children?
P. Bhaskaram:
We have not studied the effect of vitamin A deficiency on morbidity and mortality. I referred to the
data of Sommer et al. in Indonesia. We found that children with mild vitamin A deficiency do not
manifest major immunological imbalances. Administration of vitamin A produced some non-specific
immune activation, but also the large dose had some negative effects. So far administration of large
doses of vitamin A has been done as a preventive measure for blindness and only acute and chronic
toxicity has been taken into account. Because children with acute or chronic toxicity symptoms are
seen in negligible numbers, it has been assumed as a safe measure to administer such doses. This is the
first time that at doses of 100,000 or 200,000 units at the cellular level a negative effect has perhaps
been found. It has to be confirmed in other laboratories. When morbidity in a community consists of
an interaction of a number of factors, including the host factors and environmental factors, we really
have to think seriously about a single nutrient supplement and whether it will achieve the expected
results. We are talking at this conference about an integrated and comprehensive approach and I do not
know whether on the basis of a few immunological observations we should go ahead and recommend
a massive supplementation in order to obtain this particular benefit. We have to evaluate it very
carefully.
S. Drahaman:
And what is your comment about iron supplementation?
P. Bhaskaram:
I had no time to discuss the literature about iron and infections in some detail. Some people have
argued that administration of iron has given an increased incidence of infection. The consensus of
opinion, however, is that iron deficiency leads to immuno suppression and administration of iron
corrects this. But we cannot jump to the conclusion that administration of iron in the community will
lower morbidity. In fact, we did a study where we selected two communities and supplemented iron in
one and did not supplement in the other, and measured the morbidity over a period of one year. We
found an increase of hemoglobin and iron status in the supplemented children but to our surprise there
were no differences in morbidity in the two groups. This again suggested that in the communities in
the developing countries there are a number of environmental factors that should be taken care of.
R.E. Eeckels:
I fully agree with Dr. Bhaskaram. There are no short-cuts, there are no magic bullets. There is only the
holistic approach that might bring the solution. We cannot hope that with some iron and vitamin A we
will solve the problems that are fundamentally based on illiteracy, poverty and living conditions that
are just not what they should be.
L.Mata:
I would like to congratulate Dr. Bhaskaram on her studies. Are you now going to study subsets of
lymphocytes and, furthermore, specific receptors on the macrophages?
P. Bhaskaram
Such studies are now in progress.
Paper by I. Sudigbia
A.M. Molla:
Acute and chronic diarrhea are two completely different conditions. We should define carefully what
we mean by acute and chronic diarrhea.
211
I. Sudigbia:
We defined chronic diarrhea as more than 2 weeks of persistent diarrhea.
A.W. Qureshi:
You have demonstrated - as others have done - that food has to be continued during diarrhea. Do you
find difficulties in your country to implement this strategy? We have such difficulties in Pakistan. It is
difficult to convince the doctors that feeding should be continued.
I. Sudigbia:
We not only try to motivate our doctors, but also the community itself to continue feeding during
diarrhea, particularly to give tempe. Also breast-feeding should be continued.
M.C.M. Rowland:
It would be interesting to know the energy density (calories per 100 grams) of preparations like nasi
tim, tempe and others.

R.E. Eeckels:
I would stress the importance of giving supplementary food in chronic diarrhea. Supplementary food
might well be weaning food, good, cheap and safe weaning food. Such food must be found in each
country according to its own traditions and acceptability. Chronic or prolonged diarrhea is very
common in this part of the world. According to WHO acute diarrhea that goes on for more than 14
days is considered prolonged diarrhea. Prolonged diarrhea may be the major cause of death in certain
areas in Bangladesh. It can be estimated as 30-40% of mortality between 1 and 4 years. An
appropriate diet to give to children with prolonged diarrhea is of the utmost importance.
A.W. Qureshi:
In Pakistan around 15-20% of the children treated for diarrhea have prolonged diarrhea according to
the WHO classification.
D. Karyadi:
The tempe preparation as used by Dr. Sudigbia is a fermented soy bean preparation developed in our
National Institute of Nutrition. Studies have shown that this preparation has unique properties, also
antibacterial. It is important to develop production methods and to produce it on an industrial scale as
a low-cost weaning food.
M.C.M. Rowland:
There has been some thought that vitamin A might have a therapeutic role in acute watery and
invasive diarrheas. I know that there are some studies going on. Are the results available?
T. Sadjimin:
We have a collaborative study with Johns Hopkins University - a double blind randomized study - to
answer the question whether supplementation with vitamin A might reduce the incidence of diarrhea
and other infections.
M.G.M. Rowland:
My actual question was the possible therapeutic role in the established attack of diarrhea.
R.E. Eeckels:
We have been studying this in Bangladesh, and there were no effects whatsoever. In my opinion, the
data of the studies by Sommer and co-workers in Indonesia, are still preliminary. The possible
influence of marginal vitamin A deficiency on mortality and the incidence of infectious diseases seems
to me uncertain. As I have said earlier, I do not believe that vitamin A is a magic bullet.
212
D.Karyadi;
In a recent WHO bulletin high doses of vitamin A are recommended in areas where measles, acute
diarrheal disease and protein malnutrition are commonly seen.
R.E. Eeckels:
I still think we need more data to justify such recommendations. I do not deny it is true, I just say that
it is not proven in such a way that we may consider vitamin A as a treatment.
P. Bhaskararn:
We carried out studies in children with measles, acute respiratory infections and acute diarrhea and
found decreased serum retinal levels in many patients. But these levels came back to normal
spontaneously without intervention about 8 weeks after the episode. During infections also the serum
zinc and iron levels may go down. Perhaps there is a mechanism where the nutrients are redistributed
in the body. A low level may not reflect a deficiency status but it could be just a metabolic transient
disturbance which spontaneously becomes corrected. So I do not know if we should interfere by
administering vitamin A. Perhaps in communities where malnutrition is common, where already in
children the Iiverstores of vitamin A are reduced, maybe infections like measles have a potential
chance of making the vitamin A deficiency worse. But we also observed that well-nourished and
malnourished children during measles showed the same degree of immunosuppression. For almost a 6
month period following measles both groups of children had recurrent infections, so the better
nourished children did not respond in a different way. As I said before, we should discuss the pros and
cons before we rush for a recommendation regarding the administration of larger doses of vitamin A in
acute infections.
A. Li Ming Cheng:
Dr. Bhaskaram mentioned in her lecture that even with latent iron deficiency the immune function
would be altered. I wonder at what stage of latent iron deficiency that would take place.
P. Bhaskararn:
We have defined latent iron deficiency when the hemoglobin level was about 11 grams per deciliter,
but the serum iron level was low, as well as the saturation, indicating a depletion of the tissue. Such
children had a reduced neutrophil function and other immune functions which were reversed with
adequate iron therapy.
S. Draharnan:
We heard Prof. Mata discuss the role of infections on growth and child health in general. We also
heard about the role of vitamin A and iron deficiency in relation to infections. We should take this
message back to our countries: control infections and use supplements of vitamin A and iron. Why
should we not start with these measures, rather than wait for the promised day and hold on as Prof.
Eeckels says.
P. Bhaskararn:
The effects of infections on the nutritional status are clear. In our community studies we have found
that measles was closely associated with precipitation of severe malnutrition. Both in the acute stage
of measles and in the following period due to recurrent infections the child's nutritional status went
down. Then the question arises, should you prevent measles or supplement the children to prevent the
precipitation of malnutrition? The various nutritional supplementation programmes which have been
carried out in several developing countries, if you take a critical look at them, have perhaps not
attained the results that were expected. Perhaps the reason is that we are pouring our nutritional inputs
into leaky pots; when there is recurrent infection how much nutrition supplements do you give,
perhaps they are not utilized properly. However, with this statement we cannot also dismiss the role of
nutritional supplements. The precipitation of malnutrition in a child with measles is dependent on the
nutritional status of the child at the beginning of the episode. In a well-nourished child when the
measles occurs, the child loses weight, goes into a milder grade of malnutrition and maybe with
intervention will recover to normal. But the child who is already starting with depleted reserves will go
into severe malnutrition. So we cannot dissociate the control of infection and nutritional supplementa-
213
tion. The message that we have to carry home is again the old message of integrated and comprehen-
sive approach: you have to control infections, you have to provide primary health care, cut short the
duration of infections and also give nutritional supplements.
R.E. Eeckels:
We have major problems to fight: our four main targets are low birthweight, malnutrition, diarrhea and
acute respiratory infection. We need doctors - not too many - but most of all we need health workers,
teachers, social scientists, economists and politicians. Most important of all is maternal education. We
have to tackle the problem as Dr. Gopalan has said, at the level of the young girl, the mother of the
next generation. We need desperately better obstetrical care. Maternal mortality in too many countries
is unacceptably high, also infant mortality due to obstetrical causes is very high. We must fight
infections, we must promote breast-feeding. We must push vaccination, most of all measles
vaccination. The message indeed is an integrated and comprehensive approach. I am not pessimistic.
Session 5
Diarrhea and malabsorption in children
Chairman: Yati Soenarto

Diarrhea, growth and feeding practices
M.G.M. ROWLAND
Introduction
The high mortality rates in childhood in developing countries are typically associated
with malnutrition and a heavy burden of infectious diseases [1]. This relationship has
been demonstrated in rural Bangladeshi children [2].
The synergism between nutrition and infection is a well-described phenomenon
[3]. In many developing countries diarrheal illness is a major direct cause of
morbidity and mortality [4], as is the case in our host country, Indonesia [5].
Diarrhea is also the most important non-dietary cause of growth impairment in
children [6-8]. Work in Bangladesh has confirmed its importance in this context [9].
Thus it is appropriate in a symposium on the nutrition of children in South East Asia
that we examine in some detail the relationship between diarrhea and growth and the
influence of feeding practices. This paper will refer to studies carried out in diverse
geographical areas though it should be acknowledged, and this symposium testifies
to the fact, that much of the definitive work has been carried out in this part of the
world.
Diarrhea and feeding practices
There is considerable evidence that, particularly in underprivileged communities in

developing countries, exclusively breastfed infants may benefit from reduced
incidence, morbidity and mortality from diarrheal disease as compared with non-
breastfed infants [10]. Partially breastfed, ie weaning, children suffer an intermediate
degree of mortality [11]. The protective effect of breastfeeding against the severity of
shigellosis has been detected even in the third year of life in Bangladeshi children
[12].
Anorexia commonly accompanies infections, particularly when the subject is
febrile. Profound reductions in food intake have been described in association with
infectious episodes, including diarrheal illness, in Guatemalan children [13,14]. In
H.K.A. Visser and J. G. Bindels (eds.) Child Nutrition in South East Asia, 217-225.
218
rural Bangladesh Brown found only minor decreases in dietary intake in young,
largely breastfed children with diarrhea not requiring hospital treatment [15]. In
children from the same area whose diarrhea warranted hospitalisation anorexia was
responsible for a substantial reduction in intake of supplemental foods but the intake
of breast milk was apparently unaffected [16]. Similar findings of a 'sparing' of
breast milk intake despite anorexia for supplementary foods, caused by mild
intercurrent infections, were also obtained in rural Gambian infants (unpublished
results). Studies in East Java provide additional evidence for such relationships [17].
Thus it is noteworthy that breastfeeding, both exclusive and partial, appears to
protect children in terms of the severity of diarrhea experienced and in terms of the
reduction in nutrient intake which commonly accompanies an attack. Breastfeeding
might therefore be expected to lessen the impact of diarrhea on growth, a point
which will be addressed below.
Diarrhea and growth
Malnutrition measured by various indices has sometimes been found to be associated

with increased duration of diarrheal symptoms, for example in Nigeria [18] and in
MatIab, Bangladesh [19]. In Teknaf, in the south-east of Bangladesh, a relationship
was found between stunting and dysenteric illness but not watery diarrhea [20].
Malnutrition in Asian children has also been associated with more severe illness
[21-23]. Evidence for a relationship between nutritional status and diarrhea attack
rates is less consistent. Nevertheless supplementary feeding programmes aimed at
improving nutritional status in preschool children have been considered relevant in
the context of control of diarrheal diseases [24]. It is not proposed to discuss these
relationships further here but to concentrate on the effect, in the reverse direction, of
diarrhea on growth. A major concern should be to prevent, reduce or aid recovery
from the impact of diarrhea on the growth of young children. In order to do this, a
more precise understanding of the relationship is needed, starting with the concept of
diarrhea-induced growth faltering.
An individual's growth performance, or growth trajectory, may be compared with
international norms at the same age. One problem is that in almost all growth charts
in current use, the shape of the curve, particularly in the first half of infancy is almost
certainly inappropriate for the young breastfed child [25]. Another approach is to
compare the growth of the individual with that of a contemporary cohort mean.
However there is marked variation in 'normal' growth patterns amongst individual
infants, particularly where feeding practices and morbidity experience vary; these
variables can of course be controlled for. Another way of detecting variations in
growth velocity is in relation to the individual subject's own pattern of growth. Since
weight velocity declines rapidly in early life, it is important to make a suitable
correction for age. Static weight over a period of a week during the first month of
infancy in a Gambian child with a mean expected weight gain of around 300g would
219
be more worrying than slight weight loss in the last month of infancy when mean
growth in one week is around 25 g [26]. Weight varies more dynamically than
height, and growth in weight is not smoothly progressive, even in normal children
[27]; too frequent measurements over a short period of time, even made with care,
may defy interpretation. Moreover, all secretory diarrheas are, by definition,
associated with some degree of fluid loss and probably dehydration also, though this
may not be clinically detectable. Thus unqualified statements that some diarrheal
attacks are not associated with weight loss must be interpreted with caution; some
degree of weight faltering, however transient, is probably almost inevitable.
Black et al. [28] have shown that in Bangladeshi children whose growth was
unaffected by a diarrheal episode in the long term, initial pre-attack weight was
regained within two weeks and weight predicted by their pre-morbid growth
trajectory was achieved after a further four weeks. These findings imply that the
customary diet of such children was sufficient to sustain a degree of catch-up growth
for a period lasting for at least a month. More recently, Briend and colleagues,
working with another rural population in Mirzapur, Bangladesh have obtained
similar results [29]. They found that growth in weight over three month intervals Was
reduced only when diarrheal episodes occurred in the second half of the interval, as
compared with those in which watery diarrhea was absent or confined to the first half
of the interval. Dysenteric illness (diarrhea with blood) had an effect on weight gain
which was detectable over longer periods and on linear growth also.
Confining our attention only to diarrheal episodes associated with deficits in
weight which are still detectable two months or more later, or with deficits in height,
avoids many of the methodological problems mentioned above; in addition we have
an indication of severity suggesting a problem meriting special attention.
Even then there are marked variations in the extent of the impact of diarrheal
diseases on growth in the same community in different seasons [8,30], on the growth
of children in different communities in the same country [26] and in different
countries [9,31,32]. Some of these differences have methodological origins but it is
appropriate to consider other factors that might be responsible for variations in
impact, to gain insights which might lead to better preventive or therapeutic
strategies.
Are different pathogens associated with differences in degrees of

growth impairment?
The limited information available suggests that this may be so. Black and colleagues
found that diarrheal episodes associated with a significant negative impact on the
bimonthly weight growth of children were due either to enterotoxigenic Escherichia
coli or to Shigella spp [9]. In both cases a substantial number of episodes were
prolonged beyond ten or even twenty days, and it was these episodes that had the
most profound effect on growth [28]. The authors speculated that prolongation of the
220
normal short, self-limiting illness caused by E. coli might be associated with small
bowel colonisation and malabsorption. Certainly there are a number of mechanisms
whereby this might be mediated [33]. The work of Henry et al. [20] tends to confirm
the importance of shigellosis, presumably the cause of dysentery in their study, as
does Briend's work (see above).
Black et al. [28] also found that in one half of all diarrheal episodes studied in the
community stool pathogens could not be isolated, and these episodes were not
associated with growth faltering. The findings of Rowland et at, in their urban
Gambian study [26], of a lack of association between growth and diarrhea in the first
half of infancy, when enteropathogens were rarely detected in cases of diarrhea [34],
were compatible with this.
Is severe growth impairment associated with malabsorption?
Molla and colleagues have studied absorption of nutrients during episodes of acute
diarrhea and during the recovery period, in children whose initial illness was
sufficiently serious to warrant hospitalisation [35] and the subject is reviewed in the
following paper [36]. It will not be discussed further here except to say that malab-
sorption is likely to be more of a problem in chronic than in acute diarrhea, yet the
latter also contributes to growth faltering. Indeed in a study of infants in Taiwan
[37], it was frequency rather than duration or severity of diarrheal episodes which
correlated with growth impairment.
Is diarrhea-induced weight faltering associated with direct loss from the gut?
Gut losses might help to explain the particularly severe effect of shigellosis on
growth, expressed as medium term weight faltering and as linear growth faltering in
the longer term. In Bangladesh persistent protein-losing enteropathy in post-measles
diarrhea was most severe and prolonged when Shigella was the infecting organism
[38]. More surprisingly, a protein-losing enteropathy has been described in Finnish
children with acute diarrhea, some of which was due to rotavirus [39]. Substantial
endogenous nitrogen and fat losses in the stool have been described in South African
infants with severe, prolonged (more than eight days) diarrhea [40] and gut loss was
suspected to be the cause of the sudden onset of hypoalbuminaemia and general
deterioration in a group of Gambian children with diarrhea [41].
Does case management affect diarrhea-induced weight faltering?
It is well over a decade since oral rehydration therapy using a glucose electrolyte
mixture was shown to improve the outcome of acute dehydrating diarrheal episodes
221
in tenus of growth of hospitalised children [42]. This relationship has been less
clearly demonstrated in respect of the domiciliary treatment of a presumably milder
spectrum of diarrheal illness. The use of cereal-based and other fonus of oral
rehydration solutions, both in the hospital treatment of severe diarrhea and in the
domiciliary management of milder attacks, appears to substantially mitigate the
impact of diarrhea on growth [43,44]. Such preparations appear to directly affect the
duration and severity of symptoms and it seems unlikely that their effect on growth
is simply one of extra calories provided.
It is more than 40 years since the absorption of nutrients was shown to be
proportional to the intake in acute infantile diarrhea [45] and that the duration of
diarrhea was shorter in children who were fed as compared with those who were
starved [46].
Do infant feeding practices influence the outcome of diarrheal episodes in relation to

growth?
In urban Gambian children Rowland et al. found a uniformly high mean attack rate
of diarrhea through infancy, of the order of 9 per year and a prevalence rate of 15%
(expressed as percentage of time ill) [26]. Despite a substantial effect on the growth
of mixed-fed (weaning children) discernible over a two month period, there was no
significant effect on the growth of exclusively breastfed children. A similar relation-
ship has been described by Zumrawi et al. in Sudanese infants in whom the
prevalence of diarrhea was substantially lower and the effect on weight of one day of
diarrhea was much higher [32]. Though one may speculate on the reasons for the
sparing of growth in exclusively breast fed children, in neither study is data provided
on nutritional intake.
Launer and colleagues have studied the effect of varying levels of breastfeeding
on the impact of infections on growth in Indonesian infants [47]. They found that
respiratory infection had a significant negative effect on growth and that the severity
of the effect varied inversely with the amount of breastfeeding. Unfortunately an
unusually low incidence of diarrhea in their study population prevented them from
ascertaining whether or not a similar relationship existed with respect to diarrhea,
growth and breastfeeding.
How important is the contribution of diarrhea to growth impairment where

childhood 'malnutrition' is endemic?
Mata and co-workers [13] have taken the view that infectious diseases, particularly
diarrheal disease, are more important than food in the causation of malnutrition in
childhood, and that control of these infections would substantially solve the problem.
Rowland et al. studied this in more quantitative tenus. They estimated that
222
diarrhea-induced weight faltering could account for almost all of the growth deficit
observed in rural children under the age of three years [8]. In a more satisfactorily
age-stratified study of urban children to age two years they estimated that diarrhea
was responsible for 50% of the weight deficit in infancy, and lower respiratory tract
infection for a further 25% [26]. Infection in the second year of life had no
demonstrable lasting impact on growth.
Martorell and co-workers [14] found that the reduction in nutrient intake due to
diarrhea in Guatemalan children was modest in relation to their overall dietary
shortfall. He has also drawn attention to the fact that in the Taiwanese study [37],
whilst diarrheal diseases were negatively correlated with growth, only 3% of the
variance in growth could be explained in this way [48].
More recently Briend and colleagues, in the Bangladesh study already mentioned,
were not sanguine that diarrhea control would have a noticeable effect on the
nutritional status of their study children [29].
Black (personal communication) and colleagues have been reappraising
Bangladesh data to look simultaneously at dietary intake and determinants of growth.
Their preliminary results suggest that whilst elimination of diarrhea can improve
growth the effect would not be sufficient to achieve the weight gain expected by
international standards. By contrast they predicted that increased energy intakes to
recommended levels would achieve this even in the presence of existing levels of
diarrhea.
Conclusions
The most reasonable and constructive interpretation of the evidence has probably
come from Martorell who concluded that infections are as important a cause of
malnutrition as is the limited availability of food [48].
We must therefore try to limit diarrhea-induced growth faltering in the home and
to correct it in more seriously ill hospitalised cases. Unless we can achieve this a
succession of infections may in some cases lead to a progressive deterioration in
nutritional status and thus result in a measurably reduced likelihood of survival. This
sequence is particularly likely to occur where there are dietary constraints, possibly
permitting maintenance but not catch-up growth and/or persistent gut losses or
impaired absorption, and where a further attack of diarrhea occurs before complete
recovery, including growth catch-up, has occurred.
Clearly one way to prevent diarrhea-induced growth faltering would be to prevent
diarrhea. A reduction in the incidence of diarrheal disease and, particularly in this
part of the world, shigellosis, should help prevent progressive deterioration caused
by a succession of infections which interrupt the normal process of recovery and
catch-up. Water supplies and sanitation are of fundamental importance in diarrhea
control though the minimum requirements remain undefined. Greenough has
reviewed low cost public health measures [49], drawing attention to the value of
223
soap [50] and potassium alum sulphate or fitkari [51], both demonstrated in Asia.
Others have shown that epidemiological studies aimed at identifying behavioural risk
factors for diarrhea, followed by targeted interventions to modify such behaviour can
have an impact on diarrheal morbidity [52]. Thus prevention of diarrheal disease
should remain one of our aims even if we can hope only for partial success in the
environmental conditions in which the majority of developing countries peoples
continue to live.
Breastfeeding appears to be an effective strategy, reducing the incidence of
diarrhea in some communities and substantially diminishing the impact of diarrhea
on growth in others [26,32]. Whilst most obvious in the exclusively breastfed infant
it is likely to be partly effective in the mixed-fed, or weaning, infant [47]. To obtain
maximum benefit we need to consider strategies for optimising lactation, two of the
most obvious being measures to improve birthweight and avoiding unnecessarily
early initiation of weaning [53]. In a small study in Sri Lanka evidence was found of
a prolonged association between low birthweight and diarrheal morbidity [54]. In a
study of rural Gambian infants, Watkinson found that those with above average
intakes of breast milk were given supplementary feeds at a later age than others, that
diarrhea-induced weight faltering occurred later, and they were substantially heavier
at the end of the first year than those receiving less breast milk [55]. The very early
initiation of weaning, as occurs for example in some parts of Indonesia [56], must be
regarded with suspicion, even though adverse effects on infant health have not been
demonstrated.
Feeding during diarrhea is normally considered desirable though there are caveats
with respect to children receiving cows' milk products, and in the case of the
severely malnourished [57]. Unfortunately, attempts to feed children during the acute
phase of illness may be thwarted by profound anorexia. We should pay more
attention to strategies for enhancing nutrient intake, aiming at rapid, complete catch-
up growth in the immediate post-recovery phase, when subjects are usually in their
own homes. This implies the need for community-based initiatives as well as
improved practices on hospital discharge.
Simple education regarding diarrhea prevention [49,52], breast and supplemental
feeding, and oral rehydration should be attempted whenever mothers and children
are in contact with medical or paramedical personnel for the treatment of diarrhea.
Lastly, treatment of invasive diarrheas may involve the use of antimicrobials in
selected cases. At the same time there is enough evidence of the adverse effect of
giardiasis and other intestinal parasitoses on absorption and gut loss [58,59] to justify
concomitant treatment, whether or not this is believed to be part of the presenting
diarrheal illness.
References
1. Puffer RR and Serrano CV: P.A.H.O Scientific Publication No 262, Pan Am Hlth Org, Washington
DC, 1973.
224
2. Chen LC, Rahman M and Sarder AM: Internal I Epidemiol 9:25, 1980.
3. Scrimshaw NS. Taylor CE and Gordon JE: WId Hllh Org Monogr Ser No 57. 1968.
4. Mata U: In: Bellanti JA (ed) Acute diarrhea: Its nutritional consequences in children, Nestle
Nutrition Workshop Series 2, Raven Press, New York, 1983, p. 3.
5. Suharyono: In: Eeckels RE, Ransome-Kuti 0 and Kroonenberg CC (eds) Child health in the tropics,
Sixth Nutricia - Cow & Gate Symposium, Martinus Nijhoff Publishers, Dordrecht, 1985, p. 95.
6. Mata U, Urrutia JI, Albertazzi C, Pellecer 0 and Arellano E: Am J Clin Nutr 25:2167, 1972.
7. Martorell R, Habicht I-P, Yarbrough C, Lechtig A, Klein RE and Western KA: Am I Dis Child
129:1296,1975.
8. Rowland MOM, Cole TJ and Whitehead RO: Br J Nutr 37:441,1977.
9. Black RE, Brown KH and Becker S: Pediatr 73:799, 1984.
10. Rowland MGM: In: Gracey M (ed) Diarrheal disease and malnutrition: A clinical update, Churchill
Livingstone, London, 1985, p. 119.
II. Feachem RG, Koblinsky MA: Bull Wid Hlth Org 62:271, 1984.
12. Clemens ID, Stanton B, Stoll B, Shahid NS, Banu H and Chowdhury AKMA: Am I Epidemiol
123:710,1986.
13. Mata U, Kronmal RA, Urrutia II and Oaricia B: Am I Clin Nutr 30: 1215, 1977.
14. Martorell R, Yarbrough C and Klein RE: Am J Clin Nutr 33:345, 1980.
15. Brown KH, Black RE, Robertson AD and Becker S: Am J Clin Nutr 41 :343, 1985.
16. Hoyle B, Yunus Md and Chen LC: Am J Clin Nutr 32:2365, 1980.
17. Kusintja and Kardjati: personal communication.
18. Tomkins A: Lancet 1:860, 1981.
19. Black RE, Brown KH and Becker S: Am J Clin Nutr 37:87, 1984.
20. Henry FJ, Alam N, Aziz KMS and Rahaman MM: Hum Nutr: Clin Nutr 41C:243, 1987.
21. Palmer DL, Koster FT, Alam AKMJ and Islam MR: I Infect Dis 134:8.
22. Mathur R, Reddy V, Naidu AN, Krishnamachari R and Krishnamachari KAVR: Hum Nutr: Clin
Nutr 39C:447, 1985.
23. Bhan MK, Arora NK, Ohai OP, Ramachandran K, Khoshoo V and Bhandari N: Indian 1 Med Res
83:9, 1986.
24. Feachem RO: Bull Wid Hlth Org 61:967, 1983.
25. Rowland MOM, Paul AA and Whitehead RG: Br Med Bull 37:77, 1981.
26. Rowland MGM, Goh Rowland SGJ and Cole TJ: Am J Clin Nutr 47:134,1988.
27. Whitehead RG: In: Neuberger A and Jukes TH (eds) International Reviews of Biochemistry:
Reviews of Nutrition, University Park Press, Baltimore, 1979, p. 281.
28. Black RE, Brown KH and Becker S: In: Bellanti IA (ed) Acute diarrhea: Its nutritional consequences
in children, Nestle Nutrition Workshop Series 2, Raven Press, New York, 1983, p. 75.
29. Briend A, Hasan K, Aziz KMA and Hoque BA: Lancet 2:319,1989.
30. Rowland MOM: In: Holme T, Holmgren I, Merson MH and MOiby R (eds) Acute enteric infections
in children. New prospects for treatment and prevention, Elsevier/North Holland Biomedical Press,
1981, p. 253.
31. Cole TJ and Parkin 1M: Trans R Soc Trop Med Hyg 71:196,1977.
32. Zumrawi FY, Dimond H and Waterlow IC: Hum Nutr: Clin Nutr 41C:453, 1987.
33. Rowland MOM: In: Manuel PO, Walker-Smith IA and Tomkins AM (eds) Infections of the gastro-
intestinal tract, Churchill Livingstone, Edinburgh, 1986, p. 202.
34. Ooh Rowland SOl, Lloyd-Evans N, Williams K and Rowland MOM: J Diar Dis Res 3:7, 1985.
35. Molla AM, Molla A, Sarker SA and Rahaman MM: In: Chen LC and Scrimshaw NS (eds) Diarrhea
and malnutrition, Plenum Publishing Corporation, New York, 1982, p. 113.
36. Molla AM and Molla A: In: Visser HKA and Bindels JC (eds) Child nutrition in South East Asia,
8th Nutricia Symposium, Kluwer Academic Publishers, Dordrecht, 1990.
37. Baumgartner RN and Pollitt E: Nutr Res 3:9, 1983.
38. Sarker SA, Wahed MA, Alam AN, Islam A and Iahan F: Arch Dis Child 61:739, 1986.
39. Miild M, Harmoinen A, Vesikari T and Visakorpi IK: Arch Dis Child 57:154,1982.
40. Mann MD, Hill ID, Peat OM and Bowie MD; Arch Dis Child 57;268,1982.
41. Lunn PO, Whitehead RO and Coward WA: Trans R Soc Trop Med Hyg73:438, 1979.
42. Hirschhorn N and Denny KM: Am J Clin Nutr 28; 189, 1975.
43. Patra FC, Mahalanabis 0, Ialan KN and Banerjee P: Arch Dis Child 57:910, 1982.
44. Ahmed HS and Molla AM: J Diar Dis Res 5:1,1987.
225
45. ChungAW:JPedilltr33:1,1948.
46. Chung AWand Viscorova B: ] Pediatr 33: 14, 1948,
47. Launer U, Habicht I-P and Kardjati S: Am J Epidemiol 131(2):322, 1990.
48. Martorell R: Nutr Res 3: 1, 1983.
49. Greenough W: In: Halstead SB, Walsh lA and Warren KS (eds) Good health at low cost, Rockefel-
ler Foundation, New York, 1985, p. 215.
50. Khan MU: Tarns R Soc Trop Med Hyg 76:164, 1982.
51. Khan MU, Khan MR, Hossain B and Ahmed QS: Lancet 2:615,1984.
52. Stanton BF and Clemens 10: Am J EpidemioI125:292, 1987.
53. Prentice AM, Whitehead RG, Prentice A and Cole TJ: In: Taylor TG and Jenkins NK (eds)
Proceedings of the XIII International Congress of Nutrition, John Libbey, London, 1985, p. 619.
54. Mertens TE, Cousens SN and Feachem RG: Trans R Soc Trop Med Hyg 81: 196, 1987.
55. Watkinson M: Trans R Soc Trop Med Hyg 75:432,1981.
56. Kusin JA and Kardjati S: In: Eeckels RE, Ransome-Kuti 0 and Kroonenberg CC (eds) Child health
in the tropics, Sixth Nutricia - Cow & Gate Symposium, Martinus Nijhoff Publishers, Dordrecht,
1985, p. 269.
57. Manuel PD: In: Manuel PD, Walter-Smith JA and Tomkins A (eds) Infections of the gastrointestinal
tract. Churchill Livingstone, Edinburgh, 1986, p. 112.
58. Farthing MJO, Mata L, Urrutia JJ and Kronmal RA: Am J Clin Nutr 43:395, 1986.
59. Crompton DWT: Trans R Soc Trop Med Hyg 80:697,1986.
Dietary management of acute diarrhea: A scientific basis
A MAJID MOLLA and A YESHA MOLLA
Introduction
With the discovery of oral rehydration therapy, mortality in acute diarrhea due to
dehydration has decreased. But the nutritional effect of diarrhea in tum makes the
child more vulnerable to other infectious diseases, including severe episodes of
diarrhea. This remains an important hindrance towards improved child health
especially in the developing countries. It is now essential to focus more attention on
the nutritional aspect of diarrhea rather than rehydration alone. In 1924 Park
advocated the nutritional benefit of the child rather than the frequency of the stool as
being more important [1]. The same was elegantly shown by Chung and his co-
workers in 1948 [2]. Recently, the value of this concept has been amply rediscovered
both in the developed and the developing countries [3-5]. However, the main
principles for the treatment of diarrhea in the west as well as in many of the develop-
ing countries are still rehydration and the modification or withholding of the pre-
diarrheal diet along with a variable period of starvation. All these are aimed at
reducing the malabsorption of nutrients and thereby inducing early recovery from
diarrhea.
This manuscript reviews the nutritional effect of diarrhea, from the available
scientific evidence in children, as well as providing scientific justification for feeding
and suggesting an acceptable diet for feeding during diarrhea.
How diarrhea causes malnutrition
The nutritional effect of diarrhea is a complex process involving host resistance,

invasiveness of the organisms and a number of socio-environmental factors. Many of
these are not quantifiable but some of them can be measured. From the available
information it seems clear that the following are the major contributory factors:
* Anorexia due to loss of appetite
* Withholding or modification of pre-diarrheal diet as a measure to control the
diarrhea
H.K.A. Visser and J.G. Bindels (eds.) Child Nutrition in South East Asia. 227-236.
228
* Loss or malabsorption of nutrients during the acute phase of diarrhea due to

decreased enzymatic activity
* Increased catabolism in response to infection
The role of these factors will be elaborated from the literature data followed by our
personal data.
A review of the literature
It has been well-documented that during diarrhea there is a substantial reduction in

food consumption. Leonardo Mata demonstrated a strong inverse correlation
between infectious disease and calorie intake among a cohort of Guatemalan children
[6].
Martorell et al. [7] showed that during the acute phase of diarrhea food consump-
tion is reduced by 20%, accounting for a reduced daily intake of 175 Kcal and 4.8 g
of protein. A comparison of three groups of age-matched hospitalised children in a
rural treatment centre in Bangladesh (a control group, a group with acute diarrhea
and a group with acute diarrhea encouraged to eat) clearly demonstrated that while
the healthy control group consumed 130 Kcal/kg/day, the children with diarrhea
consumed only 77 Kcal/kg/day [8]. Encouragement or intensive effort to promote
food intake in the third group did not substantially increase the total calorie intake,
suggesting that anorexia is a major barrier to the improvement of feeding [8].
Several studies have documented malabsorption of macro and micro nutrients
during acute diarrhea [9]. The mechanisms of malabsorption in acute diarrhea have
not yet been clearly established. Bacterial overgrowth, disturbed bile metabolism
leading to abnormal conjugation of bile salt and fat malabsorption have been
suggested [10]. Disacharidase enzymes, the most sensitive of which is lactase, have
been shown to be reduced during acute diarrhea [11, 12].
A rapid transit time allows a shorter contact period with the mucosa for ap-
propriate digestion and absorption of nutrients. But recent studies have shown that as
little as 2 hours transit time is adequate for nutrient absorption [13]. It is important to
note that if carbohydrate is withheld from the diet, it will reduce absorption capacity
because of a lack of inductive effects of certain enzymes and a reduced rate of cell
renewal [14, 15].
Chung et al. studied two groups of children with diarrhea, one group was fed a
normal quantity of food and the second group a reduced amount of food [15]. This
study elegantly showed that nitrogen absorption was 67% and fat absorption 52%
among the two groups of children. But those eating a normal diet were associated
with a substantial extra weight gain compared to those eating at a reduced rate.
Studies related to metabolic losses due to diarrhea are scanty. The catabolic effects
of infection are due to its impact during the acute phase as well as the delay in
recovery. Twenty days with typhoid fever may cause a loss of 130 g of body
229
nitrogen which is equivalent to a loss of 4 kg of body weight [16]. Similarly, each

degree of fever implies an increase in the base metabolic rate of 5.0 to 8.2 percent
[17]. Powanda noted that diarrhea of infectious origin induced an average negative
nitrogen balance of 0.9 g/kg per day [18].
Direct protein loss in diarrheal diseases and its magnitude has not been adequately
studied. Extensive protein loss has been documented in children suffering from
coeliac disease [19], measles [20] and kwashiorkor [21]. Shigella dysenterie and
Shigella Flexneri have been associated with extreme hypoproteinaemia [22]. The
mechanism of protein loss in invasive diarrhea may be be explained by the ulceration
of the large bowel. But the loss in some of the less invasive or non-invasive agents
causing diarrhea is difficult to explain. There have been very few studies document-
ing the protein loss in specific aetiological diarrhea.
Present research
In order to persuade the professionals about the continuation of feeding during

diarrhea, several of the genuine questions have to be answered using scientific data.
Though these questions have been indirectly answered by many of the workers in
this field, more direct answers may be expected by those involved in the manage-
ment of diarrhea in the developing countries. With these objectives in mind, the
present research was carried out in the metabolic research unit of the International
Centre for Diarrhoeal Disease Research in Bangladesh during the years 1981 to
1985. The research aimed at the following:
1. Measuring the quantitative intake and absorption of nutrients during the acute
stages of diarrhea and after recovery, due to specific aetiologies.
2. Measuring nutrient loss, specifically protein loss in the acute phase of diarrhea.
3. Discovering the pattern of dietary intake in different aetiologies of diarrhea.
4. Assessing the impact of feeding on the stooling rate or outcome of diarrhea.
Methods and materials
A total of 74 children between 1 and 5 years of age suffering from acute diarrhea due
to V. Cholerae [29] Enterotoxigenic E. Coli [21], rotavirus [13] and Shigella [11]
were included in the study. All the children admitted had diarrhea of 3 days or less
duration, no prior treatment with antibiotics and a moderate to severe degree of
dehydration. After admission all patients went through a thorough physical examina-
tion by one of the investigators. Children with signs of systemic illness or severe
malnutrition (weight for age < 60%) were excluded from the study. Accurate weight
and height were taken and initial hydration was carried out by using glucose base
oral rehydration therapy or an intravenous rehydration solution according to clinical
230
indications. Blood was withdrawn for electrolytes and routine haematological

investigations. The stool was examined for ova, parasites, leucocytes and RBC. After
the initial hydration, patients were placed in the metabolic beds. The stool, urine and
vomit were accurately collected by the nursing staff who were specially trained for
metabolic research under the constant supervision of two of the research tech-
nologists.
Stools were sent for culture for Salmonella, Shigella and E. Coli by using
standard methods [22]. E. Coli isolates were tested for toxigenecity [23]. Rotavirus
was diagnosed by the ELISA technique [24]. After an overnight fast, one hour serum
xylose was estimated by feeding with a 5 g dose of D-Xylose [25]. Non-absorbable
charcoal marker was fed followed by a meal of known composition. A measured
quantity of a diet prepared from rice, chicken, lentils and soya oil [13] was offered ad
libitum and fed under supervision by the mothers.
Sample collection
The appearance of the first marker in the stool was noted as '0' hour. The intake
study was continued from this time up to 72 hours, when a second marker was fed.
All food was weighed before and after feeding. Any left-over was weighed and
included in the calculation to find out the actual amount consumed. Test weighing
was done before and after breast-feeding. Samples from all food were saved for
estimating fat, protein and calories. Stools were collected between the two markers,
weighed and frozen. Before analysis, stools collected from 0 to 72 hours were pooled
and homogenized in a blender. Aliquotes from the homogenized stools were
analysed for fat, protein and calories [26, 27]. The balance study was repeated at two
and eight weeks of nutrients and a co-efficient of absorption of nutrients was
calculated by using the following formula:
intake - output + intake x 1000
Results
Characteristics of the study patients on admission are presented in Table 1. Cholera

patients were the oldest and also the most malnourished as is shown from their mean
weight. Rotavirus patients were younger and better nourished. The composition of
the diet is shown in Table 2: the recipe was based on the local traditional weaning
diet for 1 to 5 year old children. The calorie intake among the children is presented in
Table 3. It is clear that the mean calorie intake in the acute stage of diarrhea is about
80 Kcal/kg per day. After recovery (2 weeks) the mean caloric consumption
increased to about 115 Kcal/kg per day (except in rotavirus). Absorption of car-
bohydrates is presented in Table 4 and it is obvious from the results that the absorp-
tion of carbohydrate is quite high (beween 75-90%) in the acute stages of any kind
231
Table J. Characteristics of the study patients. mean ± SD
Variables Cholera Rotavirus ETEC Shigella
Number 29 13 21 11
Age (month) 43±17.8 23 ± 14.2 36 ± 14.2 33 ± 12.2
Body wt(kg). Adm 1O.5±2.5 9±2 9.8±2 9.7±2.1
Body wt(kg). Rec. 11.2 ± 3.6 10.2 ± 2.0 1O.0± 1.8 10.4 ± 1.8
Acute Stage
Stool output 1700 ± 750 600±200 750±300 250 ± 150
24 hrs (mls)
Haematocrit 38.5±4 32±4 32.6±2.5 33±2.5
Table 2. Composition of the study diet (approximately 1 Kcall1 gm)
Food items Kcal/gm Fat (gm%) Protein (gm%)
Boiled rice 1.2 0 2.1

Curry" 1.3 5.2 7.5
Banana 1.24 0.2 1.8
Khitchri b 0.9 1.7 2.6
Whole milk 0.7 0.4 3.4
HalwaC 1.5 3.6 2.7
Bread 3.0 2.5 8.0
" Containing: Potatoes. oil. chicken. pumpkin

b Containing: Rice. dal. oil. chicken, potatoes
C Containing: Milk, semolina, oil and sugar
Table 3. Calorie consumption (Kcal/kg/d) in the acute stage and after recovery. in different aetiology of
diarrhea. mean ± SD
Aetiology Acute stage Recovery % improved
Cholera 82 ± 8.5 118 ± 8.7 35

Rotavirus 87± 10.4 103 ± 11 16
ETEC 82±8 116±6 29
Shigella 73 ± 11 114±8 31
Table 4. Absorption of carbohydrates during the acute stage, and 2 and 8 weeks after recovery, mean
±SD
Co-efficient of absorption (%)
Aetiology N A RJ R2
Cholera (29) 88± 19 93±7 92±5

Rotavirus (13) 78±23 90±4 86± 12
ETEC (21) 91 ±6 89±9 91 ±6
Shigella (11) 77±27 84± 18 92±3
A =Acute stage. R J ;0 2 weeks after recovery. R2 =8 weeks after recovery.

232
Table 5. Absorption of fat during the acute stage, and 2 and 8 weeks after recovery, mean ± SD
Aetiology N A RJ R2
Cholera (29) 55±8 78±9 82±8

Rotavirus (13) 57±6 68±7 71 ±8
ETEC (21) 68±4 62±6 61 ±5
Shigella (11) 51 ±4 69±8 70±4
A = Acute stage. RJ =2 weeks after recovery. R2 = 8 weeks after recovery.
Table 6. Absorption of protein during the acute stage, and 2 and 8 weeks after recovery, mean ± SD
Aetiology N A RJ ~
Cholera (29) 55±6 63±6 62±7
Rotavirus (13) 46±9 57±7 59±6
ETEC (21) 57±4 52±2 46±6
Shigella (11) 32±8.8 66±6 68±5
A = Acute stage. R J =2 weeks after recovery. R2 = 8 weeks after recovery.
of diarrhea and it improves further after recovery. Tables 5 and 6 present the
absorption of fat and protein, respectively, in different aetiology of diarrhea. The
average fat absorption in the acute stage varies between 55 to 65 percent, which
further improves within two weeks after recovery .
The absorption of protein is lower than 50% in the acute stage in most of the
aetiologies, and within two weeks after recovery it further improves (60%) in most
of the aetiologies, except in rotavirus and ETEC. Absorption of carbohydrates, fat,
protein and calories in cholera are demonstrated in Figure 1. Improvements in caloric
intake and its relationship with the different stages of diarrhea are presented in
Figure 2. In all aetiologies, diarrhea stops within 3 to 5 days and immediately after
that there is an improvement in the intakes of calories. Figure 3 demonstrates the
relationship of daily calorie intake and stool output in different aetiologies of
diarrhea. It is clear that while the calorie intakes improve daily, there is a simul-
taneous decrease in the stooling rate in all the aetiologies of diarrhea.
Discussion
The purpose of this presentation is to review the data from the literature and to
provide scientific evidence of the beneficial effect of feeding during acute diarrhea.
Along with this, the authors' personal work in this important area has been
presented. It seems obvious that substantial research has been done by previous
workers in this area, although the opponents of feeding in diarrhea have been able
233
120
•
CARBOHYDRATE
c 100
.2
a.... CALORIE
g
,g 80 FAT
"..0- -- - - - -- - -- -- --------1)
a"' .-'.-' PROTE~~._._ • ..e.
- .-
._.-"
Q) 60 , ... " ........ "..-.-.~.-.-.
01
IV .... '"
.... .......
C
Q)
!:s.-
~ 40
If
20
Acute Stage 2 weeks after 8 weeks after

recovery recovery
Figure I. Absorption of nutrients in the acute stage. and 2 weeks and 8 weeks after recovery from
diarrhoea due to vibrio cholera.
150
100
I CESSAT ION OF
t DIARRHOEA
L-____--~----~--~--~~--__--~~----~~----~~
2 3 4 5 6 7 14 22
DAYS
Figure 2. Consumption of calories in different aetiology of diarrhoea. The arrow indicates the cessation of
diarrhoea.
234
CHOLERA ETEC
300
~ CALORIES
!!!Ill STOOL
SHIGELLA
1 234 1 234 1 234 12 34
DAYS
Figure 3. Relationship of calorie consumption and stool output, and the outcome of diarrhoea in different
aetiology.
to overcome the scientific evidence and suppress the idea of feeding during diarrhea.
Our present work has been carried out following the elegant principles of previous
workers, but more precise methods have been adopted to study the effect of specific
aetiology.
The diet used in this study is more varied and has been made up based on the use
and preference of the mothers. To carry out a study with this diet was difficult but it
provided an opportunity to measure the actual dietary intakes of the children. A
liquid homogenized diet is convenient for the sake of study, but it does not allow the
measurement of the real appetite of a child, because repetition of the same diet
makes it boring and therefore it artificially reduces the food intake. The present study
also confirms several other studies that diarrhea reduces the appetite by about 30%.
In severe invasive diarrhea, of course, appetite loss is more drastic. But the degree of
anorexia can be improved substantially by introducing appropriate antibiotics against
the invasive diarrhea.
The important observations which clearly emerge from this study are the follow-
ing:
* Despite anorexia, a substantial quantity of food is consumed in acute diarrhea due

to the most common aetiology of diarrhea.
* The appetite improves significantly with recovery from diarrhea as reflected from
the quantity of food consumed.
235
I< Of the food eaten, carbohydrate absorption from the cereal and legume based diet
is almost unimpaired, even in the acute stage of severe diarrhea.
I< Absorption of fat and nitrogen is low in the acute stage of diarrhea, but fat
absorption improves to more than 60% after recovery.

'" The net absorption of nitrogen remains low for a longer period in rotavirus and E.
Coli diarrhea.
'" Along with a daily improvement in food consumption, stool output reduces in all
aetiologies of diarrhea.
* There is an extra intake of food following recovery and this may be due to a
compensatory hyperappetite.
Diarrhoea is not a disease but is the symptom of infection by a variety of enteric
organisms. It is not always possible in most of the developing countries to isolate the
organisms causing diarrhea, but the effects of diarrhea in children are common.
These are:
(a) Dehydration in varying degrees.
(b) Malnutrition in varying degrees and/or duration.
Dehydration can be best managed by oral rehydration therapy both at home and in
hospital. Recently introduced food based oral rehydration therapy seems to have
introduced a major breakthrough in the management of dehydration and diarrhea [28,
29]. But nutritional therapy needs definite adaptation of firm principles and policies.
It is very satisfying to see that several recent works have provided excellent scientific
evidence for including dietary therapy as an integral part of the treatment for diarrhea
[30-32].
References
I. Park EA: Proc Conn State Med Soc 20:190,1924.

2. Cung AW: J Pediatr 33:1, 1948.
3. Hirschhorn N: Am J Clin Nutr 33:637, 1980.
4. Soeparla, Soenartoy, Nelwan et al.: Tropical Pediatr and Environ Child Health 25:97, 1979.
5. Sarker SA, Molla AM, Karim AKMM et al.: Nutr Rep Ind 26:581, 1982.
6. Mala L, Kromal RA, Urrutia JJ and Garcia B: Am J Clin Nutr 30: 1215, 1977.
7. Martorell R, Yarbrough C, Yarbrough S and Klein RE: Am J Clin Nutr 33:345, 1980.
8. Hoyle B, Yunus M and Chen LC: Am J Clin Nutr 33:2365, 1980.
9. Rosenberg IH and Scrimshaw NS: Am J Clin Nutr 25:1046,1972.
10. Rosenberg IH, Solomons NW and Schneider RE: Am J Clin Nutr 30:1248,1977.
I I. Brown KH, Parry L, Khatun M and Ahmed MG: Am J Clin Nutr 32: 1962, 1979.
12. Hirschhorn N, Molla A and Molla AM: Johns Hopkins Med J 125:291, 1969.
13. Molla A, Molla AM, Sarker SA and Khatun M: Scand J Gastroenterol 18:537, 1983.
14. Rosenberg NS, Am J Clin Nutr28:648, 1975.
15. Chung A and Viscorova B: J Pediatr 33:14,1948.
16. Pollack H and Sheldon DR: J Am Med Assoc 212:598,1970.
17. Beisel WR: In: Proceedings of the Ninth International Congress of Nutrition 2:160,1970.
18. Powanda MC: Am J Clin Nutr 30:1254,1977.
19. Rotem Y and Ezerniak P: Am J Dis Child 107:58, 1964.
236
20. Axton JHM: Bf Med J 3;79, 1975.

21. Cohen H, Metz J and Hart D: Lancet 1:52, 1962.
22. Edwards PR and Edwin WH: In: Identification of Enterobacteriacae, Burgess, Minneapolis, 1972.
23. Merson MH, Sack RB, Kibria AKMG et a1.: J Clin Micro Bioi 9:493, 1979.
24. Yolken RH, Kim HW, Clem T, Wyat RG et a1.: Lancet 2:263, 1977.
25. Van de Kamer JH, Ten Bokkel Huinink H and Weyers HA: J Bioi Chern 177:347, 1949.
26. Roe JH and Rice EW: J Bioi Chern 173:507, 1948.
27. Henry RJ: In: Clinical Chemistry; Principles and techniques, Harper and Row, New York, 1964.
28. Mona AM, Ahmed SM and Greenough WB: In: Rice based ORS decreases stool volume in acute
diarrhea, Bun WHO 63:751,1986.
29. Mona AM, Mona A, Nath SK and Khatun M: Lancet 2:429, 1989.
30. Santosham M, Foster S, Reid R, Bertrando R, Barbara B and Sack B: Pediatrics 76:292,1985.
31. Brown KH and Maclean WC: Pediatrics 73: 119, 1984.
32. Mona AM, Mona A, Rohde J and Greenough WB: J Ped Gastroentero1 Nutr 8:81, 1989.
Chronic diarrhea and its management:
Breast-feeding and short chain peptide
SUHARYONO
Introduction
Indonesia, as a developing country in the South East Asia region, is trying hard to
improve the level of child health and welfare. Government and non-government
organizations (NGOs) including the Indonesian Paediatric Association are engaged
in a collaborative endeavour to pursue that goal. The current child health situation of
Indonesia indeed poses a great challenge and calls for hard work and dedication.
The total population of Indonesia was 160 million in 1983; at present it is
assumed to be around 170 million.
The Infant Mortality Rate (IMR) in 1983 was 90.3 per thousand live births, it was
75.0 in 1985 and at present it is about 70.0; this figure is still the highest among the
Asian countries [1].
In spite of the socio-economic improvements, the three leading causes of
morbidity and mortality in children below 5 years of age in Indonesia are still acute
respiratory infection, diarrheal diseases and malnutrition [2].
As stated by the World Health Organization (WHO), children under 3 years of
age in the developing countries may have an average diarrhea incidence rate as high
as 10 episodes per year; although a rate of 3-4 episodes per year is more frequently
reported. The majority of episodes are of a relatively short duration (i.e. less than 7
days) and can be treated easily and effectively with oral rehydration therapy and
continued feeding with an appropriate diet; antibiotics are necessary only when
dysentery is recognized or cholera is likely. This simple treatment regime has
reduced child mortality due to diarrhea, worldwide.
The number of episodes of diarrhea in Indonesia is at present still about 60
million (cases) per year.
Morbidity and mortality studies at the end of 1983 revealed that the number of
episodes of diarrhea in infants and under-fives was 2.8 and 2.0 respectively. At
present the trend is still the same [1]. The lowest number of diarrhea episodes per
child per year, that is less than one episode, is found in the provinces of East Java,
West Nusa Tenggara and South Borneo. The highest (more than 3 episodes per child
238
per year) is in West Java, South Celebes and East Nusa Tenggara.
Concerning the CFR (Case Fatality Rate), in our situation acute diarrhea is no
longer a major problem, since the CPR is 0% since 1980, but the CFR of chronic
diarrhea is still high, i.e. 20.3% [2]. In some instances, however, diarrhea lasts
longer than usual. Such episodes of 'persistent diarrhea' are often associated with a
deterioration in nutritional status and there is a substantial risk of death. Relatively
few studies have been directed towards the description, treatment and prevention of
persistent diarrhea. There is very limited knowledge on the epidemiology, etiology
and pathophysiology of persistent diarhroea and also about current approaches to its
management. A number of research topics are proposed which focus especially on
improving the understanding of the causes of persistent diarrhea and developing
more effective methods for its treatment and prevention.
Clinical studies in Indonesia reveal that 5-9% of cases with acute diarrhea will
develop chronic diarrhea, whereas the figure in the community in general is around
1%. Chen in his study in Bangladesh found a figure of 0.8% [8].
Even if it is assumed that only 1% of children with acute diarrhea will develop
chronic diarrhea, it is estimated that at least 600,000 cases of chronic diarrhea per
year will occur in children below 5 years of age in Indonesia. The CPR of CD was
20.3%, so that approximately 120,000 children may be expected to die annually due
to chronic diarrhea.
In most studies, 'persistent diarrhea' as stated by the WHO, is operationally
defined as an episode which lasts for at least 14 days. Using this definition, studies in
several developing countries have shown that 3-20% of acute diarrheal episodes in
children under 5 years of age become persistent and such children have prolonged
diarrhea. Persistent diarrhea seems to be uncommon in developed countries, but it is
relatively common in developing countries, however. Many children with chronic
diarrhea, the world over, have had intestinal biopsies. All, regardless of suspected
etiology, have essentially shown the same histological changes, which are varying
degrees of inflammation and/or villous atrophy. Intestinal biopsies of patients with
chronic diarrhea and malnutrition have revealed a severe degree of intestinal mucosal
atrophy which causes maldigestion and malabsorption of the nutrients and finally
leads to protein energy malnutrition (PEM). Our study in 1967 in 31 PEM children
showed intestinal villous atrophy, to a degree of two to five [3,4]. A deficiency of all
enzymes in these PEM children will lead to malabsorption of all nutrients.
The prolonged small intestine mucosal injury is caused, intensified and per-
petuated by associated factors, namely protein-energy malnutrition, deficient enteric
hormones, increased absorption of native foreign protein, ineffective villous repair,
bacterial overgrowth, infection and malabsorption of nutrients [4,5].
Deficits of enzymes, such as pancreatic amylase, lipase or trypsin, could lead to
malabsorption and diarrhea.
Continuing damage, impaired renewal or alteration in function because of
persisting infection or post-infectious processes could result in an illness of long
duration.
239
Starting from this point, the important task for us is to develop more effective
methods for treatment and prevention. In breast-fed infants chronic diarrhea is rarely
seen. Promotion of breast-feeding is 'a must' for every country in the developing
world.
Short-chain peptide (SCP) in chronic diarrhea
Apart from the breast-feeding promotion efforts, a new concept on the 'short-chain
peptide' has to be taken into consideration.
In recent years, it has become increasingly apparent that there is another mode of
protein absorption involving the mucosal uptake of intact peptides followed by
hydrolysis within the absorptive cells [6]. Patients with pancreatic insufficiency may
have serious impairment of protein digestion and increased loss of protein in the
faeces [7]. The brush border membrane of intestinal mucosa is quite efficient in the
hydrolysis of peptides with three to six acid residues [8]. For di- and tripeptides there
is a specific transport system and an intestinal cellular hydrolysis since there is a
tripeptidase and a dipeptidase.
Over the past few years, a number of commercially available elemental diets for
the treatment of patients with absorptive and digestive disorders have become
available.
The protein component of elemental diets is provided in the form of free amino
acids which render these diets highly hypertonic. A peptide diet would be more
advantageous than an elemental diet for the treatment of patients with chronic
diarrhea, protein energy malnutrition or short bowel syndrome, all of whom have
reduced rates of amino acid absorption [9,10]. The rate of amino acid uptake is
considerably greater from solutions consisting of individual dipeptides, individual
tripeptides, or partly hydrolized proteins than from solutions composed solely of free
amino acids. Furthermore, hypertonicity is poorly tolerated by patients with reduced
absorptive capacity due to one of the above conditions; the tonicity of peptide
solutions is very much less than that of equivalent free amino acid solutions.
The present investigation is a clinical study and was performed with the aim of
studying the effectiveness of the 'short chain peptide' formula in protein maldiges-
tion/malabsorption diarrheal children.
The following were examined during the 3 month follow-up study: blood, urine,
faeces (once within a 2 week interval, whereas the tryptic activity test of the
intestinal juice was done within a 4 week interval for each patient).
For each patient the gradual acceptance of the SCP and the non-SCP formula, the
amount of formula given and all important symptoms and signs were noted.
Special attention was given to the gaining of body weight, the improvement of the
protein maldigestion/malabsorption and the discontinuance of the diarrhea.
The effect of the SCP formula on body weight was evaluated as 'excellent' if the
increase in body weight within the 3 month study divided by the normal increase in
240
Table 1. The effect of the SCP formula on the gaining of body weight
Formula group Excellent Good Poor Total
SCP 34 I o 35
non-SCP 9 14 7 30
Total 43 15 7 65
KS ---t K =2.6986, P < om (highly significant)
Table 2. The effect of the SCP formula on protein maldigestion/malabsorption
Formula group Good Poor Total
SCP 32 3 35
non-SCP 16 14 30
Total 48 17 65
X 2 = 10.2459, p < om (highly significant)
Table 3. The effect of SCP formula on the discontinuance of diarrhoea
Formula group Excellent Good Poor Total
SCP 2 32 I 35
non-SCP 0 22 8 30
Total 2 54 9 65
KS ---t K = 0.9569, p > 0.05 (not significant)
Table 4. The patients' acceptance of the SCP formula
Formula group Good Poor Total
SCP 28 7 35
non-SCP 29 30
Total 57 8 65
Fisher (P) =0.0446, p < 0.05 (significant)
body weight for age was more than 120%; 'good' if between 80-120%; 'poor' ifless
than 80% of the body weight. The results can be seen in Tables 1 to 4.
Table 1 reveals that the body weight gain of the SCP group was excellent in
97.1 % (34 out of 35), whereas the non-SCP group was excellent in 30%, good in
46.7 and poor in 23.3% (K = 2.6986; p < 0.01).
241
In the SCP group there were 28 patients with under-nutrition and 7 with maras-
nus, and in the non-SCP group there were 25 with under-nutrition and 5 with
narasmus.
Criteria for the evaluation of improvement of protein maldigestion/malabsorption
:using the TAT test or tryptic activity test, recommended by Schwachman) [11] were
'excellent' if the maldigestion disappeared after 1 month SCP formula; 'good' when
the maldigestion disappeared between 1 or 2 months; 'poor' after more than 2
months.
The data in Table 2 show that in the SCP group vs non-SCP protein maldiges-
tion/malabsorption improved in 91.4% vs 53.3% (p < 0,01) (good) and the others
gave poor results. No one was excellent.
From Table 3 can be seen that there is no significant difference between the SCP
group and the non-SCP group in the effect of stopping the diarrhea, although the first
group is slightly better than the second (a good result in 91.4% vs 73.3% and poor in
2.9% vs 26.1 %, whereas an excellent result was seen in only 5.7% vs none
(K = 0.9569; p > 0.05».
The criteria used for the evaluation of the cessation of the diarrhea were
'excellent' if it stopped in less than 2 days, 'good' between 2-4 days and 'poor' after
more than 4 days.
It seems that the problem of malabsorption of fat is more important in PEM
children. The SCP formula contains more medium-chain triglycerides (MCT) than
the non-SCP formula (50.0% vs 31.5% of the total fat). However, the lactose content
of the faeces in the SCP group is rather lower than the non-SCP group, i.e. 0.19% vs
0.8%. The fecal micro-organisms found in those 2 groups showed no significant
difference.
The last table (Table 4) shows that the acceptance of the non-SCP group is
significantly better (p < 0.05) than the SCP (good in 96.7% vs 80.0%). The criteria
used for evaluating acceptance were 'excellent' if the patient accepted all the food;
'good' if 80-100% of the total food was accepted; 'poor' if less than 80% of the total
food was accepted. The SCP formula may not be so readily accepted because of the
MCT. In this study, the acceptance of the formula could be improved by adding
small amounts of sucrose or glucose. It can be concluded that SCP is very beneficial
for treating protein malabsorption in diarrheal children.
Breast-feeding
We have shown in other studies that non-breast-fed children have more than three
times the chronic diarrhea of breast-fed children. Whether the high incidence of
chronic diarrhea in non-breast-fed children is due to a reduced intake of protective
factors in breast-milk, contamination of formula milk with enteric pathogens,
damage to the gut mucosa due to hypersensitivity to animal milk protein or in-
tolerance to lactose or some other mechanism, is not known. In low birthweight
242
children with chronic diarrhea breast-feeding gave excellent results. We found that
the growth of breast-fed low birthweight infants with chronic diarrhea was better
compared with LBW infants fed with pregestimil, a protein hydrolysate, MCT,
lactose-free formula. Many studies have shown that diarrhea contributes to malnutri-
tion; however, malnutrition is a risk factor in prolonged diarrhea. In malnourished
children, the mean duration of diarrheal episodes is longer and there is a higher
incidence of persistent diarrhea. For children with protein energy malnutrition breast-
feeding is the best answer to chronic diarrhea, but we should not forget appropriate
supplement feeding, which is highly nutritious and absorbable.
Conclusions
Chronic diarrhea is a major problem, especially persistent diarrhea with a high

fatality rate.
Breast-feeding is the best way of treating chronic diarrhea, and after the age of 4
months this should be complemented with appropriate supplementary feeding, which
is very nutritious and absorbable.
Short-chain peptide formula is useful for treating chronic diarrhea, if breast-milk
is unavailable.
References
1. National Population Survey between Two Census, Republic of Indonesia, 1985.

2. Suharyono: Diare di Jakarta dan masalahnya, Disertasi 1985.
3. Suharyono and Cotton DG: Coeliac disease in the hospital for sick children, London, 1970.
4. Suharyono, Sunoto, Aswitha Darnayanti, Sadikin Darmawan and Cotton DG: Paediatr Indones 11:7,
1971.
5. Suharyono, Sunoto, Aswitha Boediarso and Hendardji H: Malabsorption syndrome in Indonesian
children. The 5th World Congress of Gastroenterology, Ciudad de Mexico, 1974.
6. Matthews, DM and Adibi SA: Gastroenterology 71:151, 1976.
7. Davenport HW: In: Physiology of Digestive Tract, third edition, Chicago Year Book Medical Publ
Inc 1971, p. 191.
8. Chen LC: Am J Clin Nutr 31:2204, 1978.
9. Sadikali F: Gut 12:276, 1971.
10. Adibi SA and Suleimanpour AN: J Clin Invest 53:1368, 1974.
11. Schwachman H, Patterson PR and Laguna J: Pediatrics 4:222, 1949.
Breath hydrogen test for lactose malabsorption
in Chinese children and infants
c.-y. YEUNG, F.H.W. WONG and A.Y.C. TAM
Introduction
The breath Hz test is a non-invasive means of detecting sugar malabsorption in

children. The end-expiratory portion of the breath is usually preferred to ensure
constancy of the gaseous composition to obtain more reproducible results [1]. To
overcome the technical difficulties of end-expiratory sampling from small infants
and children, we have developed an electronic automatic end-expiratory air sampling
device for infants [2] and a simple T-adaptor to 'entrap' the end-expiratory air for
older children [3]. We tested the efficacy of these devices we have developed and
performed a lactose challenge test to study the children's response in producing
Hydrogen (Hz) in their breaths.
Materials and methods
Chinese newborn infants admitted to the nurseries of Queen Mary Hospital and
school children between 3 and 12 years of age from a primary school and a kindergar-
ten were selected for this study. Informed consent was obtained from the parents for
all tests. End-expiratory breath samples were collected for gases analysis with the
Shimadzu (GC8 APT) Gas Chromatograph. For newborn infants, the automatic
electronic sampling device [2] was used. For older children the T-adaptor 'end-
expiratory sample trapping device' [3] was used. The gas-analysis was by thermal
conductivity detector, molecular sieve chromatograph [4] as reported elsewhere [5].
Eighty-four Chinese newborn infants between 25 and 43 weeks gestation and 1.17
to 3.95 kg weight were studied. 280 older Chinese children of 3-12 years old were
also investigated. Breath-Hz production following a standard lactose load of 1 gm/kg
was studied by half-hourly interval sampling for 3-4 hours and the profile charted.
244
Table 1. Efficacy of the end-expiratory sampling devices
Device CO2 concentration Percentage end-tidalness
Infant automatic sampler 4.35% (3.9-5.1) 87% (78-100)

T -adaptor for children 4.08% (3.8-4.75) 78% (72-86)
Figures are expressed as mean and range.
Histogram Frequency
I .... + .... r .... + .... I .... + ..•. I ...
o 4 B 12
3- i •••••••• ··.####
5 '.## •••• #.#.######.####
7 :#••••• ## •• # •• #.
...
9 :# ••••••••••••••• ### •• #.####
11 :.#.# •••••• # ••••
13 : •• # •••••••• # ••••••••••
15 : •••• #
17 :
:#
, .........
...........
19
i...
... 21
:.#
.... .
23
.! 25 :.#•••
;; 27 :
.......
:
== .
29
= 31 :
:
.....
IIC
:z: 33
..
35 :
37 :
39 :
41 :
...#...
43 : . . .##
45
47
49
51
53
55
57
...#
59
61
63
65
67
69
71
73
75 ••
77
Figure 1. Breath H2 response in 84 newborn infants (NRBH ppm).
Results
Table 1 shows the degree of end-tidalness (or approximation to alveolar concentra-

tion) of the breath samples as obtained in the newborn and older children. The infant
automatic electronic sampler was highly efficient in obtaining 87% end-tidalness [2]
as judged by the arterial paC0 2 which was used as an alveolar reference. The T-
245
100..
I •
E i y =O.99x "14.45
c.
I r =0.1386
.
D-
80+ peO.05
•
.,,,
.£
-L
•
.
.c
1;;
• • •
.... ..••
80..
1i •
-- .--- ....-
I
• •
.
-
"D
Ql
,, • •• ••
--
~ •
.. -....• .-• ......•.. --. .-...........
E
0
40 ..
•• •
.• ... -• - ...,..• .....•• -...- "'t

c
.£ ,,, •
-• - • • •••
,
-
GO
a:'" 20"
-• -
R
,, •• ••
,, .,. •
0+ •
•
._-------+---------+---------+---------+---------+--
4 6 8 10 12
Age In years
Figure 2. Breath H2 response to lactose in 280 children.
adaptor was obtaining 78% (72-86%) end-tidalness from older children [3]. Figure 1
shows the 84 lactose-breath-Hz test results obtained in the newborn infants [6]. It can
be seen that two overlapping populations appear with a dividing point near 20 ppm
which corresponds to the conventional cut-off point [7] for diagnosing lactose-
malabsorption. One population had a normal distribution of breath Hz response
below 20 ppm and the other population was more widely dispersed above 20 ppm.
The latter is apparently the group of malabsorbers. Similar findings were also
obtained for the older children [3] who also showed an increasing frequency of high
Hz response with advancing age (Figure 2). Figure 3 shows a typical study result of
breath-Hz response to lactose and simultaneous blood glucose profile in a lactose
absorber and Figure 4 that of a lactose malabsorber.
Discussion
We have confirmed earlier reports [8-10] that the breath-Hz response to lactose is a
useful test for lactose malabsorption. It has the distinct advantage of being non-
invasive and apparently equally diagnostic as the standard lactose tolerance test.
We have also overcome some of the technical difficulties in obtaining end-
expiratory or alveolar samples for tests from small infants. Our newly developed
electronic device can collect cumulative end-expiratory samples from rapidly
breathing small infants for analysis of gases [2]. A mean of 87% and range from 72
to 100% of end-tidalness were attainable [2]. The equipment depends on a sensor to
pick up expiratory signals from the infant and to automatically collect a sample of
the expired air to test. To date, the device is the first of its kind to perform such a
function automatically with such a high degree of efficiency.
246
2!r~~·~~~·~iD~pp_m~)__________________________--,
Lactose.....................................................................................
+
5 ...........
o~----~-------L ______ ~ ______ ~ ____ ~
-1 o 2 3
No. of hours
8rOooe==~(~mma~I~~) ______________________________,
e ...................... '.. ~ ............................................................,
It · . . . . . . . . . . . ./. . . . . . . . . . . . . . . . . . ~
. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ..
2 ...... ·.... ·Lactose ..............·............................................·..................·..........
oL-----~--~~--
~ ____ ~ ____ ~ ____ ~
-1 o 1 2
No. of hours
Figure 3. Breath H2 response to lactose and simultaneous blood glucose profile in a lactose absorber.
TOOYc~=·~(~~m~w=.~)____________________________,
.!
~
30
20 ................ +. . . . . . . . ..
......·Uictbse ..·····
10
.,
o~----~----~----~----~----~--~
o 2 3
No. Of hours
erOooe==~l~~~~~) ____________________________--,
I· . . . ·. . . . . . . ~ . . . . ·. . . . . . .
] .............................................................................................................................
III
z ....·..·....Lactose·......·..·····.. ······..····..····....··......··........................................
t
-,
o~----~------~------~----~------~
o 1 2 3
No. of hours
Figure 4. Breath H2 response to lactose and simultaneous blood glucose profile in a lactose malabsorber.
247
The T-adaptor which we have developed is a very simple piece of equipment [3]
which can be readily used in field studies and many clinical situations. It is ex-
tremely simple and easy to operate. A big breath is all that is needed from the child.
Little intelligence or teaching is required, and absolutely no discomfort is elicited.
We have shown in our survey that many Chinese newborn infants demonstrate
lactose malabsorption. This is particularly prevalent in the low birthweight and the
prematurely-born. In the older children, we have shown that there is an increasing
trend of excessive breath-H2 production with advancing ages from 3 to 5 years to 12
years old (Figure 2). Our findings differ significantly from those in Japan [11],
Finland [12], or other Caucasian countries. Our data [2,3,6] may be a useful
reference for future studies on Chinese children or in other anthropological studies.
Acknowledgement
This work was accomplished by the joint effort of two other departments of the
University of Hong Kong viz the Department of Chemistry (Dr. K.W. Fung) and the
Electronic Services Unit (Mr. Y.P. Ma, Mr. H.C. Kwan and Mr. E.S.H. Ng).
References
1. Wilson HK: Scand J World Environ Health 12:174, 1986.

2. Yeung CY, Ma YP, Wong FH. Kwan HC. Fung KW, Tam YC and Ng PY: An automatic electronic
end-expiratory sampling device for breath Hz test in small infants (submitted).
3. Wong FHW. Yeung CY. Tam YC and Fung KW: Breath Hz response to lactose in Chinese children.
(submitted).
4. Douwes AC, Fernandes J and Rietveld W: Clin Chim Acta 82:293, 1978.
5. Wong FHW: Breath Hz tests to study lactose malabsorption in children. M Phil thesis, Department
of Paediatrics. University of Hong Kong. 1989.
6. Yeung CH, Wong FHW, Tam YC, Fung KW. Ma YP, Kwan HC: Breath H2 response to lactose in
newborn infants. (submitted).
7. Levitt MD, Donaldson RM: J Lab Clin Med 75:937, 1970.
8. Calloway DH, Murphy EL, Bauer D: Am J Dig Dis 14:811, 1969.
9. Ford RPK: Arch Dis Child 58:595, 1983.
10. Newcomer AD. McGill DB. Thomas PJ. Hoffmann AF: N Engl J Med 293:1232.1975.
11. Nose 0, Lida Y, Kai H, Harada T, Ogawa M. Yabuuchi H: Arch Dis Child 54:536.1979.
12. Sahi T, Isokoski M, Jassila J, Launiala K: Acta paediatr Scand 61:11.1972.
13. Lebenthal E, Antonowicz I. Schwachman H: Am J Clin Nutr 28:595, 1975.
Papers by M.C.M. Rowland, A.M. Molla and A. Molla
A.W. Qureshi:
I would like to ask Dr. Rowland about the growth pattern when children with diarrhea are followed up
to 3 or 4 years of age. Do they catch up their growth when the diarrhea is over?
M.G.M. Rowland:
In such a situation I would not distinguish between growth faltering with respect to diarrhea and all the
other things that are going on. Children falter for a number of reasons, and almost all children falter.
But it seems to me that by the age of I or 2, children are more or less on a track and that track parallels
very often the international standards. The fIrst year of life is the most important. Anything we can do
to modify transient setbacks in growth at that time may have a benefIcial effect, but if we fail to do so
then the evidence that these children substantially improve in their nutritional status with respect to the
standards, do not look optimistic.
D.Karyadi:
Did you study the effect of vitamins and other micronutrients in children with diarrhea?
A.M. Molla:
We only studied in a limited way the effect of zinc supplementation. We found in malnourished
children a profound improvement in appetite and intake within 24 hours; the absorption rate improved
after 5 days.
l.A. Kusin:
I would like to stress that during sickness the breast-milk intake may be substantial. In our studies in
the fIrst year of life it was 600 ml per day and in the second year 450 ml per day. We have followed
about 300 children at weekly intervals. The anorexia is profound at the age of 6-18 months,
particularly between 8-12 months. It seems to be more related to age than the impact of disease.
Analyzing our data, the median number of disease periods is 10 per 12 months, with a peak at the age
of 9-12 months. In these 300 children there are about 10% who were never sick, although they were in
the same environment, and about 30% had 18 or more episodes in a year. The question is what is an
episode? When there are less than 3 days between 2 episodes, are we dealing with a new episode, and
what is acute or chronic? How do you analyze these successive number of episodes?
M.G.M. Rowland:
I agree with you. In some populations or groups in populations who are having a lot of diarrhea it
becomes almost academic where you could say one episode finishes and the other starts, and those are
very likely the ones which are contributing to growth faltering. It is important to refer to episodes of
diarrhea when one studies the etiology or pathogenesis. For the kind of epidemiologic studies I refer
HKA. Visser and l.G. Bindels (eds.) Child Nutrition in South East Asia, 249-253.
250
to, it seems reasonable to talk about the percentage of time a child is ill. You have said that some
children are almost never sick. It seems that the children who start off with the higher breast-milk
intakes are in general those who have weaning foods introduced rather later in life and are those who
suffer less from growth faltering associated with diarrheal disease. I think there is a public health
message in that for us.
C. Gopalan:
I would like to congratulate the two speakers on their excellent presentation. My problem is that the
great attention that has been devoted to the diarrhea problem is very largely clinically orientated.
Diarrhea is a public health problem. In my opinion pediatricians can play an important role in teaching
the mother how to prevent diarrhea, and this is not being done in the several diarrhea clinics attached
to Pediatric Departments in my country. I think it is the responsibility of the pediatricians to use the
unique opportunity they have for a programme of nutritional education around their diarrhea clinics.
M.G.M. Rowland:
I would echo that. If we want to prevent diarrhea induced weight faltering as an important contribution
to childhood malnutrition we need to prevent diarrhea. We need particularly to concentrate on
trigolosis, we need to think in terms of things like personal hygiene, sanitation and so on. Then it is
our aim where diarrhea occurs to see that dehydration deaths are prevented by effective treatment in
the home as far as possible. But when children unfortunately do come to hospital and there is contact
with medical or paramedical personnel they should take every opportunity to see first of all that the
child does not have an attack of diarrhea that requires hospital admission again. they should take every
opportunity to advise the mother on how she may best achieve catch-up growth when her child is
discharged from hospital and they might take the opportunity to also treat some of the parasites. which
have been shown to contribute to poor nutritional status.
P. Bhaskaram:
I would like to respond to Prof. Karyadi' s question regarding the micronutrient absorption in diarrhea.
We have carried out a few studies on the absorption of vitamin A in children with infection and we
found that in diarrhea, or any other respiratory infection or other systemic infection, there was a
considerable decrease in vitamin A absorption from the gastrointestinal tract. However, though the
absorption was decreased it was still substantial enough to protect the child. In a subsequent study we
then administered oral rehydration solution with vitamin A to the child during diarrhea and we found
that the absorption was quite satisfactory. The question to Dr. Rowland is, did you observe any
differences in growth faltering in well-nourished children compared to malnourished children with
diarrhea?
M.G.M. Rowland:
The kind of studies that we were doing were not designed to answer this question. However. given the
fact that severe malnutrition in its various forms, wasting and stunting, is in general associated with a
greater severity of diarrhea, one might well expect that one of the ways in which that greater severity
is expressed is in its impact on growth. There is very little data I think in the literature that specifically
addresses that question, but what little there is would suggest that at least in early childhood more
severely malnourished children suffer a greater degree of weight faltering in association with diarrhea.
P. Bhaskaram:
The other two questions are to Prof. Molla. One is, what is the exact role of malnutrition in increasing
the severity of diarrhea and. secondly. because your studies are mostly from Bangladesh, I would like
to know what is the contributory role of measles in the recurrent diarrheas in this particular area?
A.M. Molla:
Why diarrhea becomes more severe in malnourished children, why probably all infections are more
severe in those children, is not really known. The answer should come from studies such as those you
have shown at this symposium. In Bangladesh we have found more severe and chronic diarrhea after
measles.
251
L.Mata:
Home surveillance studies are very important. You have to go to the homes, once or twice a week, in
order to find more accurate incidence rates. I would like to stress the importance of parasitic infections
on growth faltering. There are many parasites and we need careful laboratory investigations to find
them.
M.G.M. Rowland:
I fully agree. Active surveillance to the community is important and I simply do not believe it when I
hear about low incidence rates of diarrhea. I am unimpressed how unsuccessful routine hospital
laboratories are in isolating common parasites. We have not talked about small bowel colonisation and
the effects of anaerobes. In the malnourished child the gastrointestinal flora is totally disturbed. Our
treatment should address that, but our laboratory investigations are inadequate.
Paper by Suharyono
P. Bhaskaram
What is the composition of the short-chain peptide formula?
Suharyono:
The composition is as follows (per 100 ml):
protein 2.0 g (12 energy %)

(whey protein
hydrolysate)
fat 3.7 g (47 energy %)
vegetable oils 1.85 g
MCT 1.85 g
linoleic acid 0.9 g
carbohydrates 6.7 g (41 energy %)
glucose 0.4 g
lactose 0.1 g
maltose 0.9 g
polysaccharides 5.3 g
minerals 0.3 g
vitamins 0.02 g
energy 66 kcal
osmolarity 190 mosm/l
The renal solute load is I IO mosm/l
C. Gopalan:
There is a discussion about the validity of the figures with respect to prevalence and you mention that
you went by the public health statistics that are provided by the Ministry of Health. The remarks that I
have to offer have nothing to do with the Ministry of Health of Indonesia, but I am making a general
remark. I reviewed the public health statistics data from many Asian countries for a meeting I attended
recently in Malaysia and I think the greatest hurdle in the field of public health work is the un-
reliability of all public health statistics in many Asian countries. Even infant mortality data have
become suspect, have become classified information because they carry some political overtones, and
I think that unless we improve the database and the public health statistics collection then many of our
claims may not carry conviction.
M.G.M. Rowland:
I should like to make a very simple point about the incidence of chronic diarrhea. It is customary to
define chronic diarrhea in terms of duration of episode, but if you are in the situation as Prof. Kusin
was describing where children are having frequent diarrhea on and off, it is equally important then to
252
define the duration between symptoms which determines the beginning and end of an attack. If you
have a child that has diarrhea for 7 days, no diarrhea for 48 hours and diarrhea again for 7 days, if you
define attacks as being separated by 24 hours or more, then you have a situation where you have had 2
attacks of acute diarrhea. If you define attacks as separated by 3 days or more it is an attack of chronic
diarrhea and this is just part of the reason why our actual epidemiological data on chronic diarrhea in
the community are very poor.
Suharyono:
I follow the definition of the WHO, which is very commonly used in developing countries: chronic
diarrhea is a persistent diarrhea that is prolonged for longer than 14 days.
A.M. Molla:
I want to stress the point that the problems are very complicated. We cannot get a very quick answer.
For instance, shigellosis is a small bowel disease, but we have observed that in such infections the
absorption of nutrients is affected. When you treat with antibiotics, there is a great improvement of
nutrient absorption within 24 hours, also when you feed with ordinary diets.
Dr. Suharyono compared a short-chain peptide diet with a non-short-chain peptide diet. Is that just
a locally available diet? We have just completed a study on patients with chronic diarrhea and
observed good effects from yoghurt.
Suharyono:
When the acidity of the gastric juice is reduced the defence mechanism against micro-organisms from
outside is decreased. We found bacterial overgrowth in the intestinal flora of children with chronic
diarrhea. In the past, immediately after the Dutch occupation, we used acidified milk, but this is no
longer so.
A. Li Ming Cheng
Was there also evidence of disaccharidase deficiency in the children you studied?
SUharyono:
Yes, we found a prevalence of 86.5% and in the low birth weight infants 72.3%. So in children with
chronic diarrhea there is not only protein malabsorption, but also lactose intolerance and malabsorp-
tion of other nutrients.
Paper by C.-Y. Yeung, F.H.W. Wong and A.Y.C. Tam

Purnomo Suryantoro:
I wonder if we can use the hydrogen breath test in infants. The sensitivity and specificity of the test is
not high. We found 80 ppm in breast-fed babies without any sign of diarrhea.
G.-Y. Yeung:
We would also seriously question those people who use the test to indicate a low lactose diet in
infants. We prefer to use a physiological lactose challenge test, 1.1-1.4 grams lactose in the milk,
rather than just 1 gram lactose in an aqueous solution.
M.G.M. Rowland:
Prof. Yeung is drawing a clear distinction between lactose malabsorption and lactose intolerance. The
observations that lactose is better tolerated when milk is the base was made earlier in the 70s and
given the fact that the hydrogen breath test is apparently an over-sensitive test, I wonder whether
pediatricians should not make it a standard procedure to use milk as the challenge rather than aqueous
lactose.
A. Li Ming Cheng:
In Hong Kong some of the recurrent diarrhea is not due to infection but probably the result of a
disaccharide problem. In some babies who had acute diarrhea and then after everything subsided had
recurrent diarrhea there was usually a history of the family giving a lactose containing formula. When
lactose was removed from the feeding, the diarrhea rapidly subsided again.
253
A.M. Molla:
I think the lactose intolerance after diarrhea is a fact, but it should not be over-emphasized. Although
lactose is an enzyme which in patients with diarrhea is the first one to go and the last one to return,
there is usually a substantial amount of enzyme left. It is important to give the child milk along with
other semi-solid food as weaning food. Then the lactose load is less and the symptoms will be less.

T. Sadjimin:
What is the biological background of the relationship between the improvement of appetite and the
cessation of diarrhea. How do you measure anorexia?
A.M. Molla:
In children with cholera we have seen that appetite reached the optimal level when the stool became
soft. We do not know about the mechanism involved. Much research has to be done.
It is difficult to measure anorexia. We think that the mother knows better than anybody else about
the demand and appetite of her child. We have to give the message that if the child is willing to eat she
should give more, because if the mother believes that food is bad for the child, the child might
improve his appetite but might go into starvation.
AA.Khan:
For how long would you advise giving the extra food supplement in the post-diarrheal period?
A.M. Molla:
We do not have the answer. Prospective studies have to be done. It seems to me that periods of 3 to 4
weeks would be enough.
D.R. Karunaratne:
I admire the message of Prof. Molla about the management of diarrhea and the continuation of normal
feeding in these children. In Sri Lanka we have great problems with shigellosis infections. Even with
antibiotic treatment the diarrhea may persist. Dr. Rowland has said that you can control growth
problems in children if you control infections, like diarrhea, respiratory disease and malaria. But how
can you effectively control respiratory disease and malaria?
A.M. Molla:
Shigella is a problem, it is becoming a more severe problem. In several countries such as Sri Lanka,
Bangladesh, India and Pakistan shigellosis is becoming resistant to most antibiotics. We have to use
antibiotics effectively and warn against misuse of antibiotics.
M.G.M. Rowland:
If I can just quickly respond to the question about control of infection. One thing that I think we never
adequately stressed in our publications is that when we show the effect of an infection on growth, we
are talking about the effect of an effectively rapidly treated infection. which is not allowed to run its
natural course. What we are saying here is not go out and control all infections and forget about the
diet, what we are saying is that whatever your natural diet programme, your supplementary feeding
programmes, unless you take measures to improve the amount of infection in the community as far as
you can you must expect the result to be severely compromised in many situations.
L.Mata:
The problem of resistance to antibiotics is extremely important. It is a more serious problem in
developing countries than in the industrial nations. We have to ban the counter sales of antibiotics and
- very important - to promote the use of 2 or 3 antibiotics in hospitals. This has been done with good
success in South America. the micro-organisms again become sensitive to antibiotics. We have to rely
on antibiotics to treat shigellosis, because it is an invasive killing disease and one of the few cases in
which diarrhea has to be treated with antibiotics. As far as bacterial overgrowth is concerned,
antibiotics should not be used. When you feed milk or other adequate feeding the intestinal flora
becomes normal.
Session 6
Medical teaching on child nutrition
Chairman: A.S. Ongkie

[ntegrated medical teaching curriculum
em child nutrition
PITONO SOEPARTO
Health problems in Indonesia
Indonesia is an archipelago in South East Asia consisting of over 13,000 islands in

six island groups, spread over an area of nearly 2 million sq km.
The climate is tropical, with high rainfall. The demographic data as issued by the
Indonesian Central Bureau of Statistics ("Statistical Profile of Mothers and Children
in Indonesia 1985") are as follows:
Population 165.1 millions

Population 0-4 years 23.6 millions
Live births per annum 5.2 millions
Average annual population growth rate 2.2 percent
Crude birth rate 31 per 1000
Crude death rate 11 per 1000
Infant mortality rate 93 per 1000
Life expectancy at birth 53.3 years.
In addition to its impact on nutritional status and growth of infants and children,
diarrhea along with tetanus, perinatal cases, and respiratory tract infection are the
leading diseases which cause a high infant mortality rate [1] (Table 1). Diarrhea,
respiratory tract infection and skin disease also have a high morbidity rate (Table 2).
It is estimated that diarrheal disease causes at least 250,000 annual deaths in
children under 5 years.
Facing these problems the Indonesian Government through its national health
system has set up 2 programs to control diarrheal disease:
1. Long-term Diarrheal Disease Control Program: Community health & clean

water supply to reduce the morbidity rate.
2. Short-term Diarrheal Disease Control Program which deals with case
/I.K.A. Visser and J. G. Bindels (eds.) Child Nutrition in South East Asia, 257-266.
~ 1990 Kluwer Academic Publishers.
258
Table J. Mortality pattern in the paediatric group [1]
Rate per 1000 population
< I Year 1-4 Years 5-14 Years
1. Tetanus 13.84 0.16 0.05

2. Perinatal cases 13.21 0.00 0.00
3. Diarrhoea 11.19 2.78 0.32
4. Acute resp. tract info 8.93 0.88 0.14
5. Diphtheria, pertussis, measles 6.16 2.92 0.19
6. C.N.S. diseases 4.03 0.49 0.11
7. Congenital anomalies 3.02 0.66 0.0
8. Accidents 1.76 0.52 0.20
9 Dental, oral & digestive syst. 1.51 0.66 0.04
10. Bronchitis, asthma 1.39 0.26 0.00
11. Malaria 1.01 0.08 0.06
12. Avitaminosis, malnutrition, anemia 0.76 0.26 0.00
Table 2. Morbidity pattern (based on the complaint) in the paediatric age group [I]
Rate per 1000 population
< I Year 1-4 Years 5-14 Years
1. Acute resp. tract inf 70.2 73.8 18.6

2. Diarrhoea 25 3 1.7
3. Skin infection 20.1 26.3 6.6
4. Bronchitis/asthma 13.3 12.9 3.8
5. C.N.S. diseases 7.5 1l.5 5.2
6. Diphtheria, pertussis, measles 6.7 4.4 1.2
7. Malaria 2.6 7.9 7.0
8. Dental, oral & digestive syst. 2.1 3.3 2.1
9. Avitaminosis, malnutrition, anemia 0.8 1.3 0.3
10. Accidents 0.8 1.9 1.0
management of diarrheal disease and its impact on the nutritional status of the
patients and early detection and control of outbreaks of diarrheal disease: the
main goal of this program is to reduce the mortality rate.
To achieve the short-term program, skill and knowledge on case management is
consequently required. There exist two possibilities in meeting these requirements:
through upgrading/courses for medical personnel, or through the improvement of the
medical school's curriculum.
In conjunction, diarrhea training units (DTU) have also been set up in teaching
hospitals where various programs of diarrhea (oral rehydration solution), mother-
child welfare (proper nutrition, breast-feeding and personal hygiene). immunisation,
have been integrated in the system.
The problems of undernutrition in Indonesia are shown in Tables 3 and 4.
259
Table 3. The distribution of the nutritional status of Indonesian children in 2-4 year age group. based on
weight to height index [1]
Nutritional status (%)
Age group N >90% 80-90% 70-79% <70%

Standard Standard Standard Standard
2 year 6249 54.2 32.7 11.0 2.1

3 year 7041 58.1 32.4 8.3 1.2
4 year 7204 62.2 29.9 7.0 1.3
Total 20494 58.3 31.6 8.7 1.3
Table 4. Distribution of the nutritional status of Indonesian children aged 2-4 years based on height for
age index [1]
Nutritional status (%)
Age group N >90% 90-94% 85-89% <85%

Standard Standard Standard Standard
2 year 6495 41.4 26.7 21.8 10.1

3 year 7278 32.6 32.6 22.8 12.1
4 year 7391 30.7 29.8 25.3 14.2
Total 21164 34.6 29.8 23.3 12.2
It is estimated that about 10% of the population of 2-4 years of age suffer from
acute undernutrition (Table 3), and 35.5% from chronic undernutrition (Table 4).
The acute undernutrition is confined more to the 2 year olds group, whereas the
chronic undernutrition is confined to the 4 year olds. These data clearly show that the
nutritional status of children in Indonesia is still a health problem.
Infections (respiratory tract infection, gastrointestinal tract infection, tetanus,
diphtheria, pertussis, measles) are also still a health problem in Indonesia (see Tables
I and 2).
Over 60 percent of children under 15 months of age in Indonesia have received
one or more doses of vaccine (BCG, DPT, OPV, measles). Immunization of pregnant
women for the prevention of neonatal tetanus (a disease which kills 80,000 neonates
each year, 10 every hour) has doubled since the previous review in 1982.
It is recommended by the Government to set up an integration of the Expanded
Programme on Immunization (EPI) and the Programme for the Control of Diarrheal
Diseases (CDD) in curative services [2].
The opportunity provided by each presentation for clinical care should be taken to
check the immunisation status of each mother and child and to administer vaccine as
required.
Each assessment of a child under the age of 5 years should also include a review
260
of the child's growth chart and the mother's knowledge of the management of
diarrhea. These activities can usually be done while waiting to see the doctor or the
nurse.
Health problems in Indonesia thus need an integrated managerial approach. The
integrated program should not merely continue towards curative services but should
also extend to the medical curriculum as well.
The aspects of disease inter-relationships
Food digestion/absorption and the nutrtional status
Gastrointestinal alterations frequently accompany severe protein energy malnutri-

tion. Abnormalities include pancreatic exocrine insufficiencies, reduced gastric
secretion, morphological alterations of the intestinal mucosa, reduced proliferative
activity and renewal of mucosal epithelial cells, deficiencies in brush border
enzymes and altered intestinal permeability to macromolecules. The binding
properties of enteric bacteria at the brush border of the intestinal epithelium may also
be affected by malnutrition. All of these findings may be a direct response to isolated
or generalized macro-micronutrient deficiencies, to specific enteric infections
complicating malnutrition, to small intestinal bacterial proliferation in the mal-
nourished host, or to combinations of these factors.
Diarrhea and the nutritional status
Recent evidence has convincingly demonstrated an inverse relationship between the

prevalence of diarrheal disease and the growth of children in less developed
countries [3].
Growth failure secondary to diarrhea has been attributed to the negative impact of
diarrhea on dietary intake and to impaired intestinal absorption during and after
enteric infection. Catabolic responses to infection also exact a nutritional toll,
especially in febrile illnesses.
Diarrheal disease is probably the most important non-dietary cause of growth
impairment in third world children [4].
It is apparent from Table 5 that the nutritional status among diarrheal patients
after 10 years remains unchanged (1978-1988/1989).
The number of cases of undernutrition among diarrheal patients, however, is
higher than those found in the general population (see also Table 3). These data
support the impression that there is an interaction between diarrhea and the nutri-
tional status of the patients. One may therefore regard diarrhea as a nutritional
disease.
261
Table 5. Nutritional status of infants under 2 years of age with gastroenteritis (Dr. Soetomo Hospilal
Surabaya).
Year >90% 70-80% 60%

Standard Standard Standard
1978 33 (39.76%) 42 (50.60%) 8 (9.64%)

(Irwantono, weight/age)
1985 125 (37.42%) 20.5 (61.38%) 4 (1.20%)
(Soeparto P.• weight/height)
1988/1989 392 (38.70%) 474 (46.79%) 147 (14.51%)
(Soeparto P., weight/age)
Table 6. Clinical response to treatment in acute diarrhoeal patients related to their nutritional status (7)
Number of patients
Nutritional status Poor response Good response Tolal
100-80% 17 237 254

Standard
70-60% 16 64 80
Standard
Tolal 33 301 334
X2 = 12.l1,p < 0.001.
Nutrition may play an important role in the aggravation of the clinical outcome of
diarrhea by the development of chronic diarrhea following acute ga:;troenteritis [5].
It has also been known for many years that the intestinal function is impaired in
malnutrition, although an understanding of the basic underlying mechanisms remains
inadequate [6].
Nutrition may also play an important role in the manifestations and complications
of diarrhea [7] as is shown in Table 6.
Malabsorption and nutritional status
Fat absorption is impaired in malnutrition. Steatorrhoea in malnutrition may be

related to impaired micelle formation due to depletion of the bile salt pool but may
have other underlying causes, such as decreased intraluminal lipolysis, from
depression of exocrine pancreatic secretion or defective betalipoprotein production
before transport of fat into lymph.
Carbohydrate intolerance and malabsorption is also of major importance in
undernourished children (see Table 7) as demonstrated in Table 7 [8]. Malabsorption
of sugar may be related to bacterial overgrowth of the upper intestinal contents
262
Table 7. Clinical intolerance to fat and lactose in acute diarrhoeal patients related to their nutritional status
[8]
Number of patients
Nutritional status Intolerance (+) Intolerance (-) Total
100-80% 22 232 254

<80% 35 45 80
Total 57 277 334
X2 =52.93, P < 0.001.
leading to bile salt deconjugation which damages the energy dependent transmural
active transport mechanism and the ultra structural components of the enterocytes
[6].
Intestinal mucosa and nutritional status
Several studies have shown that the gastric mucosa is abnormal in a small group of
malnourished Indonesian children. Chronic gastritis was present in most of them and
was associated with reduced levels of resting gastric acid secretion and very marked
impairment of the responses of the gastric mucosa to stimulation by the hormone
pentagastrin. Secretion of hydrochloric acid is one of the normal factors controlling
the upper intestinal micro flora and the loss of this mechanism in malnutrition could
contribute to the heavy bacterial populations found in the upper gut in malnourished
children [9].
Soeparto [8] showed that patients with clinical malnutrition and diarrhea had
abnormalities of the intestinal mucosa varying from shortened, broadened and
irregular cells of the villi (grade I) to a complete absence of the villi with cuboidal
epithelial cells and dense mononuclear infiltration of the lamina propria (grade III)
(see Table 8). In Table 9 it is also shown that an abnormal mucosal pattern was more
common in prolonged diarrheal patients with nutritional disturbances.
A study by Gracey [10] in malnourished Indonesian children has shown that the
intestinal secretions obtained in vitro by peroral aspiration have significant con-
tamination by a large number of bacteria of varying types which do not belong to the
enteropathogen groups.
It was further shown that sugar absorption was significantly inhibited by these
bacteriae including Staphylococcus pyogenes, Streptococcus faecalis, Proteus sp,
Pseudomonas sp, Klebsiella sp, and Candida Albicans [11].
Bacterial overgrowth in the proximal gut of malnourished children can indeed
produce a wide range of clinical effects including malabsorption, hyponatremia, and
anemia. These effects are related to the bacterial degradation of bile salts affecting
263
Table 8. Intestinal mucosal pattern of patients with malnutrition [8]
Kwashiorkor Mixed type Marasmus
Grade I 6 7 4
Grade I-II 1
Grade II 8
Grade III 5
Table 9. Mucosal pattern of patients with prolonged diarrhoea related to their nutritional status [8]
<80% >80%
Standard Standard
Normal mucosa 12.2% 17.1%

Abnormal mucosa 43.4% 26.3%
micellar solubilisation of fat, damage to the function of mucosal enterocytes and

binding of vitamin B 12 [6].
H epato-pancreatic function and nutritional status
To some extent the clinical effects of malnutrition can also be influenced by the
degree of insufficiency of the liver and exocrine pancreas which in tum can be
aggravated by nutritional disturbances [12].
The disturbed pancreatic function in malnourished Indonesian children has been
shown by Suharyono et al' [13] using the 'Gelatin film Test'. The low trypsin
activity found in this study may explain the maldigestion of protein in malnourished
children.
Infection and nutritional status
The important synergism between malnutrition and infection (including infectious

diarrhea) has been documented in the last two decades in many parts of the world.
In malnutrition, other mechanisms may also operate, particularly with recurrent
and chronic gastrointestinal infection and infestation damaging the intestinal mucosa.
The interaction between infection and nutrition is complex [14]. The pathogens
recovered from stools of prolonged and chronic diarrheal patients with some degree
of undernutrition are shown in Table 10 [15].
Breast-feeding and the nutritional status
That breast-fed infants experience less diarrheal illness or death than other infants,
may be due to the following factors:
264
Table 10. Pathogens recovered from stools of patients with chronic and prolonged diarrhoea [15]
Pathogens Number of patients
E.Coli 27
Salmonella non typhoid sp. 15
Staphylococcus aureus 9
Negative stool cultures 33
Total 84
* The immunological and antimicrobial properties of breast milk.

* The "bifidus factor".
* Less risk from contamination.
* Breast-fed infants may have a better nutritional status, and thus a lesser risk of
death from diarrhea.
A study involving Indonesian mothers by Sunoto et al. [16] revealed that breast-
milk, produced during the first week of delivery, contained a low concentration of
lactoferrin but a substantial concentration of protective substances against infection
except to E. Coli.
Immune system and the nutritional status
Two separate immune systems appear to be involved in gastrointestinal infection: the

intestinal mucosal secretory immune system which seems to determine the degree of
resistance to tissue infection), and the generalized immune system (which is capable
of mobilizing a variety of specific effector mechanisms operating in conjunction with
the inflammatory response).
The study by Sunoto et al. [16] involving 32 infants with chronic diarrhea and
varying degrees of malnutrition revealed a high concentration of Ig G, Ig A, Ig M,
and Ig E in serum and gastrointestinal secretion. It was concluded from the study that
the high concentration of immunoglobulins may be due to repeated gastrointestinal
infections.
Cow's milk protein sensitive enteropathy (CPSE) in Indonesian children has been
investigated by Soeparto et al. [8] involving infants under 2 years of age suffering
from post enteritis syndrome with variable degree of undernutrition (see Table 11).
CMPSE was found in 27 out of 37 patients investigated (73%). Severe mucosal
abnormalities were found more commonly in patients with CMPSE compared to
those without CMPSE.
The aspects of integrated medical teaching

Teaching methods used in many medical schools interfere with good learning.
Indonesia has worked hard under the leadership of the Consortium on Health
265
Table 11. Degree of mucosal abnonnality in 37 patients who underwent a complete serial investigation for
cows' milk protein sensitive enteropathy (MPSE, [8])
Grade Total
1 to 2 3
CMPSE(-) 6/1 3 10
CMPSE(+) 9 18 27
Total 37
x2 = 4.1,0.05> P > 0.02.
Sciences (CHS) to improve teaching methodology in medical schools.

Nevertheless lectures still predominate over more effective participatory student-
oriented approaches. Practical work under supervision and guided simulation is not
common.
Integration, either horizontally based on organ systems, or vertically using clinical
teachers to teach basic science or vice versa, is rare.
On the clinical service, students may be given little actual responsibility and work
primarily as observers. The opportunity to gain the necessary "hands on" clinical
experience and practice may not be given [17].
Most of the teachers may be very busy, particularly the clinicians, and may not
have the time or inclination to develop new or improved learning materials or to
develop organized lesson plans.
The basic science given during the preclinical term is often too extensive and too
crowded.
Clinical teachers are often dissatisfied with the basic knowledge of the students
when they enter the clinical term, which often leads to repetitive lectures on the same
subject. This is not because the basic knowledge during the preclinical term was not
given in a complete and comprehensive fashion, but merely because the subjects
given were not formulated to be oriented to the clinical/community settings. The
students simply do not have any orientation of what to do in the clinical internship.
Realizing the aspects of various disease interrelationships, one may anticipate that
integrated medical teaching may be one of the best approaches in overcoming the
above-mentioned medical teaching problems.
Based on the knowledge of the disease interrelationship, integrated models for
teaching proposals can be developed. In such a model, hepato-biliary, pancreatic and
gastrointestinal systems are simultaneously integrated into one system of digestion
and absorption, a system which influences the nutritional status of the child.
Infection, diet (nutrition) and the developmental process may also influence the
integrated system. These components of the model constitute the basic knowledge in
the medical curriculum which can be given during the preclinical term.
Departments like Physiology, Biochemistry, Anatomy-histology, Microbiology,
266
Nutrition, etc, can be engaged as an integrated team in teaching the basic knowledge
in such a model.
The clinical results of the model, on the other hand, can be given during the
clinical internship of the medical curriculum.
After the students have been given the integrated basic knowledge during their
preclinical term, they can have more relevant knowledge when entering the clinical
internship.
The teaching approach during the clinical internship may be subsequently directed
more to the tutorial/problem based learning activities.
Another approach in clinical education should be the combining of individual and
team learning activities. While students are responsible as individuals for their
learning during a clinical internship, this learning activity should be in the context of
the work team. Clinical education is a challenging experience for most students
because it allows them to participate actively in the health care team, to seek
solutions to real problems and to learn by practice while caring for the patients.
Clinical internships should therefore ideally be directed to (l st) a problem
centered approach in the context of professional practice, (2nd) an integrated basic
knowledge based learning model, and (3rd) a combination of individual and team
learning.
References
1. Sidharta Y and Budiarso RL: House Hold Health Survey Seminar, Jakarta, 1986.
2. The Joint Government of Indonesia/UNICEF/US.AID/WHO Mission: Review of the programme for
the control of diarrheal diseases and the expanded programme on immunisation, Indonesia, 1986.
3. Brown KH and MacLean WC: Pediatr 72:119, 1984.
4. Rowland MR: In: Walker Smith JA, Hamilton JR and Walker WA (eds) Practical paediatric
gastroenterology, Butterworths, London, 1983, p. 148.
5. Lifshitz F: In: Lebenthal E (ed) Advances in pediatric gastroenterology and nutrition, Mead Johnson
Symposium Series No 1 Excerpta Medica 108, 1984.
6. Gracey M: Paediatr Indon 21:235,1981.
7. Soeparto P: Thesis, 1987.
8. Soeparto P, Noerasid H, Martosoedarma S, Djupri Land Giri IW: XIIth International congress of
pediatrics, Manila, 1984.
9. Gracey M, Houghton M, Cullity GJ, Suharyono and Sunoto: 5th Annual meeting ofthe Coordinating
Board of the Indonesian Pediatric Gastroenterology, Medan, 1977.
10. Gracey M, Suharyono, Sunoto and Stave DE: Am J Clin Nutr 25: 1170, 1973.
II. Houghton M, Gracey M and Suharyono: 5th Annual meeting of the Coordinating Board of the
Indonesian Pediatric Gastroenterology, Medan, 1977.
12. Heim T: In: Lebenthal E (ed) Advances in pediatric gastroenterology and nutrition, Mead Johnson
Symposium Series No:l Excerpta Medica 72, 1984.
13. Suharyono, Adnan SW, Ibnu S, and Boediarso A: 5th Annual meeting of the Coordinating Board of
the Indonesian Pediatric Gastroenterology, Medan, 1977.
14. Morley D and Woodland: In: See how they grow, Oxford University Press, New York, 1979 p. 89.
15. Soeparto P, Soebijanto MS, and Noerasid H: Paediatr Indon 22:83,1982.
16. Sunoto, Bell R, Gracey M, Sumarmo and Suharyono: 2nd Annual meeting of the Coordinating
Board of the Indonesian Pediatric Gastroenterology, Medan, 1977.
17. Northrup RS: National Rehydration Seminar, Surakarta.
Panel and group discussion: integrated medical teaching on child

nutrition
A.M. Molla:
We have to be realistic and see how we can integrate the practical approach of teaching nutrition to
our different groups of people. The medical curriculum in the developing countries is not geared to
nutrition, because many of our teachers - I am sorry to say - have no concept about what is happening
in the community, with the mother and child in that environment. Unless we try to take our students to
the community and to spend part of their time there and to leam from the basic problems of the
community, we will not achieve much. Secondly, I think we should not only teach the medical
students and doctors, but we have to step down and simplify the message, to bring it to the village, to
the health workers at the primary health care level. Lastly, the message has to go to the mother and
child. The solution of the problem needs a holistic approach. By giving a 2 year diploma in nutrition at
our medical school, I do not think we shall touch upon the problem.
D. Sinniah:
We must define what is the type of doctor that we intend to turn out and what we expect of them in
tenns of the needs of the country. There is always a need to revise our medical curriculum. We have
not revised our curriculum for the past 25 years and are now in the process of doing so. We have
always paid very great importance to social and preventive medicine in our medical curriculum, this
has been carried out throughout the clinical years and any student who does not pass this important
complement is not allowed to sit his final medical examination. The incidence of diarrheal diseases
has gone down greatly and we only see very mild cases of malnutrition in Malaysia. The emphasis on
nutrition is still important, this is carried out throughout all the years. We have integrated teaching
with the Departments of Physiology, Bacteriology, Immunology, Social and Preventive Medicine and
so this is emphasized throughout our medical course. As to what should be the actual content of the
teaching program, I am sure that this can be ironed out by each medical school when they are looking
into this.
R.E. Eeckels:
Regarding integrated or non-integrated teaching, I would think personally that if you call that
techniques of teaching, then they are not important, but if you have good teachers and students,
whatever technique you use, you will probably have a good end product. But what the end product
should be, we do not know. Because we do not know what the future is, we do not know the future
structure of our medical services and they of course detennine to a very large extent what type of
doctor we need. Frequently we do not base our teaching on the real problems as they exist in the
community and that is what we have to do. We are very bad at teaching the real problems, both in the
hospital and in the community and in the field. I think we have to teach students to teach themselves,
that again is something where most medical schools are not so terribly good at, to teach themselves for
H.KA. Visser and J.G. Bindels (eds.) Child Nutrition in South East Asia, 267-271.
268
the rest of their lives, which means to read, which also means that they need good things to read and it
is not easy to get medical journals in many Third World countries. And then we have to teach students
to think for themselves, to challenge their professors and we have to ask for their criticisms, again
something that world-wide is very rarely done. I think that teachers have to inspire their students by
their own enthusiasm and commitment and teachers should not try too hard to impress with their
knowledge. Finally, I think we have to introduce, far more than we have been doing in the past, non-
medical persons in the teaching of our students, we are far too much doctor orientated.
M.Gracey:
I live and work in a country that thinks itself industrialized and western and to a large extent this has
altered our thinking in terms of teaching our students, whether they are medical students or students in
other health professions. Because malnutrition is really very rare in our country, most of our students
and I think many of our teachers have become mesmerised by the technology of medicine and medical
research and the importance of teaching, particularly in problems that are community based, I am sorry
to say, has been to a large extent overlooked. But I think we have to review this situation because there
are many problems in our westernised countries that are community based and we do have pockets of
poverty and undernutrition and disease in our community. We also have problems in particular with
minority groups, one that I spoke about is the Aboriginies but we have the elderly, we have drug
addicts, we have recently arrived migrants, we have other people that have particular problems and our
students I think are ill-equipped to deal with many of these problems. Of course in the western
countries we have to think about the other nutritional problems, the problems of obesity, hypertension,
adult onset diabetes, for example, and with the rapid changes that are occurring in many of the now so-
called developing countries these nutritional problems of affluence are going to become even more
important. Now we have JO medical schools I believe in Australia, we have only 2 professors, or 2
departments of human nutrition and I think that is really a reflection of our interest or lack of interest
in nutrition in my country. I am pleased that in the audience today we have a professor of community
nutrition (Prof. Heywood) and that is a recent development, and I am looking forward to seeing how
he can show the medical teachers in our country the importance of community based diseases that we
have come to overlook in the western countries. I agree with Prof. Eeckels that the methodology is
really not terribly important whether it is integrated or not, but the end product is particularly
important and we have to point the end product I believe much more towards the community where
the origins of these problems are.
C.Gopalan:
Mr Chairman, there was a time when in my enthusiasm for promoting the teaching of nutrition I
thought that we should campaign for the creation of chairs in nutrition at medical colleges. I later
discovered that they may be a disservice to nutrition because we may end up by putting nutrition in a
cul-de-sac. I realize that while there is still a case for chairs in nutrition in special institutions such as
universities, in the medical colleges our strategy must be to capture those of us who are interested in
the promotion of nutrition, to capture the commanding heights of medical education and this is the
policy which we tried to do through the National Institute in India. That strategy was that we invited
people who were already teachers, lecturers, associate professors - we are talking about people who
are still associates and who are going to become the professors of tomorrow, professors of pediatrics,
preventive medicine, medicine and biochemistry - to our institution where they underwent a 3 month
certificate course. Because these are very senior people, they do not require months and months of
teaching. All the courses were supported by WHO and UNICEF. I am glad to say that there are at least
300 people in this South East Asian region who have had exposure to that type of program, apart from
200 pediatricians or more in my own country. I still think that creation of chairs in nutrition in medical
colleges is not going to do much, we have to capture specialists in key disciplines. I think pediatricians
in particular should receive the nutrition message, as should also the professors of preventive medicine
and the internists. More and more pediatricians should be offered opportunities for training in
institutions where there is a high level of nutrition expertise available. I think that that is the way by
which we will create the necessary climate in the medical college and by virtue of the fact that they
have such people who are exposed to nutrition. When they develop an integrated program they could
themselves become models. I am only talking of my own personal experience with the National
Institute of Nutrition's training programs entirely directed to teachers of medical colleges.
269
S. Drahaman:
In our medical college in Sri Lanka we take the students to the field once a week, two of them are
given one home to look after, they have to make a nutrition survey of the family, a nutritional status of
the whole family and then once they get that data they are to correct the defects on their own. They
have to advise, educate the family, even physically bring them some food, supplementary food, and to
teach them to plant nutritious plants in their gardens and even teach the mothers how to generate
income. These families live in homes just near the medical college. The most important problem is
poverty in our families.
E. Suroto-Harnzah:
In my opinion we should also use the infonnal channels for teaching nutrition to the community. In
Indonesia we have songs that teach about nutrition and breast-feeding and I think it has some effect.
Also we should use key persons in the community, such as religious leaders. They have been
influential in the family planning program in Indonesia.
A.A. Khan:
With our best efforts we could not achieve the desired goal of integrated teaching and community
teaching. We have 300 students in each class and things are not as simple as they appear on paper.
A.M. Molla:
I agree. But we have to change our old outdated curriculum which we inherited from the Western
countries. We have to replace lectures by more tutorial teaching. One fifth of our curriculum is for
teaching in the community where our teachers go with the students. Each student works with a family.
A. Li Ming Cheng:
I would like to bring up one thing that is possible in the Hong Kong community in which each family
now has a television. Some years ago when we had some problems with the weaning diet for our local
children we thought that for public education the best thing is to go through the media. In the last 5-6
years I think the radio and television people have become increasingly interested in broadcasting
nutritional programs. It involved considerable time on behalf of doctors who became involved with
this, but it can be very effective, provided the program is properly organized and made very attractive
and at a level which the public can understand. Whether this is applicable to all developing countries
depends on the situation in the community.
P.F. Heywood:
I want to follow up a couple of comments that have just been made and perhaps ask the panel to
comment on another one. I think that the need is certainly there to involve other health professionals in
teaching medical students. The medical profession has been a little reluctant to accept this I think. I
would like to ask about the involvement of doctors who are not working in the medical school but who
are working in the community in teaching medical students in the community.
R.E. Eeckels:
I fully agree that indeed medical practitioners and other health professionals or people who are not
really called health professionals, but are doing active work in the community, should be invited to
teach medical students. I would also like to say that when we are talking about professors of
community medicine and professors of medicine, the professor of medicine is a man with a high
prestige and a big car and a big private practice and the professor of community medicine is the man
who uses a bicycle, and who has no prestige whatsoever amongst the students. Why would we need to
make the distinction between a professor of medicine and a professor of community medicine, there is
just no difference. Medicine first of all should be community medicine and then possibly also hospital
based medicine. We have to make a proposal to principals or Vice-Chancellors of universities to
change the titles and to make the professor of community medicine the professor of medicine and the
other person will be called the professor of clinical medicine. We should also give the new professor
of medicine (formerly community medicine) a big car and special advantages.
270
L.Mata:
I think that the main problem that I see in Latin America is that we are trapped in a very orthodox,
very rigid scheme of teaching medicine. Some time ago we tried to compare the curriculum of our
schools with, for instance, Harvard and Johns Hopkins and we found that there was more teaching of
tropical medicine, public health and nutrition in Harvard and Johns Hopkins than in Brazil or in Costa
Rica. I think that we have to elevate the prestige of the people engaged in teaching in pediatrics,
nutrition, epidemiology and public health. We need more emphasis on epidemiology and statistics. We
need more emphasis on laboratory work applied to the community. I think that will produce the
prestige that these teachers need. When they are involved in good experimental field work with the
students and they will be doing research that will involve epidemiology as well as biology they will be
doing good internal medicine, but applied to the field. Then they will have prestige because they will
publish and they will have grants. The lack of prestige is due to the fact that the very smart fellow will
tend to be in surgery or internal medicine, the fellow that is not so smart ends up trapped in public
health or in teaching nutrition. But when we improve the teaching of epidemiology, statistics and good
laboratory science then we will attract intelligent people to these areas and then the people will have
pnestige, as an infectious disease man or a public health man has a lot of prestige in Boston or in Johns
Hopkins or in California.
AA.Khan:
I would like to point out that pediatrics as a discipline does not have a major status in countries such as
India, Bangladesh, Sri Lanka and Pakistan. It is a small complement of general medicine, it is not
independently examined. This meeting should make a recommendation to consider pediatrics as a
speciality just like internal medicine.
D. Hull:
I hesitate to make a comment as to what ought to happen in this part of the world. In Nottingham when
we were concerned with introducing a teaching program there 15 years ago and started with the
concept of integrated teaching and problem solving, we were aware of the McMaster experiment and
other people trying out similar experiments. And whilst I think that a lot has to be said for it, it seems
to me that what will determine what the students want to learn and what they will choose is what they
require as they see it and therefore I would echo what Prof. Eeckels said, it is the quality of the
teachers and the attitudes of the teachers that are the main determinants of what students acquire. So
when I hear people saying it is because they have inherited a curriculum which they have been unable
to change, I would tum around to them and say gently that it is the quality of the people that teach and
their attitudes that are going to determine what the students acquire and what they aim for. Now we
may be a bit cynical about the aspirations of the young, they want a big car and they want to be
successful and that is a little bit sad, because there are other values, but I think the young are looking
for those other values and we have got the responsibility of inspiring them to look for those other
values. Could I say when people are talking about integrated teaching and are identifying the common
problems, whilst I think this is very proper, it is important that we should recognize that the students
require a basic scientific training and need to acquire the clinical skills. These are the basic require-
ments for them to be able to adapt with the passage of time to meet the problems as they come. So I
think whilst we might identify the common problems and direct the students there, we must be very
careful to ensure that they have the full scientific understanding and the clinical skills to face patients
and to meet their needs. Finally, just to comment on these words "hospital" and "community". I worry
that people then are talking about the community as something being outside the hospital. A hospital is
a community resource and it must be recognized as such. I think if the doctors who work in hospitals
also worked in communities there would be no difficulty. I see no problems at all about teaching
medical students within the community the basic skills. I see no difficulty involving non-medical
people in teaching students. So dieticians and health visitors teach my medical students and my
medical students spend time in the community as much as they spend in the hospital. I would suggest
that the attitude of the teachers and their skills and knowledge are the most important if we are going
to improve the situation rather than hope that new teaching teChniques are going to solve our
problems.
271
Soenarlo:
I just want to share our experience involving the local health centre doctors and their staff in teaching
our students. They are very proud of being a teacher. It also changes their own attitude. We found for
instance that when they were involved in the teaching program, the use of antibiotics in the treatment
of diarrhea dropped and the use of oral rehydration solution increased.
R.E. Eeckels:
I fully agree with Prof. Hull that the hospital should be part of the community, it should have outreach
functions to the community. We all should agree with Prof. Khan, that pediatrics should be a distinct
academic discipline. Pediatrics can be defined as the total care of the child; all aspects of child health
are involved, including nutrition. Pediatrics should get more prestige, so that students feel attracted
to it
Closing remarks
H.K.A. VISSER
I have been asked to make some closing remarks on behalf of the orgamzmg
committee and to thank the speakers and the committee members. It is not my
intention to summarize the meeting; the time is not available and moreover I am not
in a position to do so, but I will try to review some of the important problems we
have discussed.
I am a pediatrician, interested in growth and development and in the nutrition of
children. I work in a highly industrialized, rich country and in a university hospital
that spends most of its money in advanced technological care. Most of the problems
that were discussed here these last couple of days, we do not have in our industrial-
ized northern countries. But I would like to remind you that we too had these same
problems not very long ago. We are interested in your problems, we like to listen to
you, to learn from you and to help you.
When Nutricia came to me some time ago and said "you may organize another
symposium", I proposed not to organize this in the Netherlands, but in Indonesia. It
should be a workshop for pediatricians and nutritionists working in this part of the
world. I am very happy that we have organized it in this way. This has been a
meeting of pediatricians and nutritionists, and we are particularly grateful to the
nutritionists who came to this meeting.
The pediatrician is, more than other medical colleagues, a generalist, interested in
total integrated care of children. The pediatrician's approach has traditionally been a
holistic one. Indeed, as Dr. Gopalan has pointed out, the pediatrician is the
nutritionist's best friend. A pediatrician is interested in growth and development, in
the interaction of genetic backgrounds and the environment. Several speakers at this
symposium reported on growth and development studies concerning children in
different areas in this part of the world.
Looking at the growth curves, and comparing these with standards such as the
Harvard or HeNS curves, it is remarkable to observe the secular trend changes in
South East Asia.
H.K.A. Visser and J. G. Bindels (eds.) Child Nutrition in South East Asia, 273-277.
274
Changes in height and weight over a period of time (the so-called secular trend)
are important parameters for the socio-economic and health background of a
population, just like parameters of low birth weight percentage, infant mortality and
age of menarche in girls.
Professors Gracey and Heywood described incredibly rapid changes in height and
weight in Australian aborigines and Papua New Guinea children. Changes can be
seen in higher socio-economic groups in India. It is a fascinating hypothesis to
propose that all human people in this world will at the end have the same growth
curves, and that genetic differences in height will not be more than the difference
between the 3rd and 97th percentile. I was also very much impressed by the data
presented by Professor Yeung about growth in Chinese children in Hong Kong and
in Canada.
I would like to mention that even in my country, the Netherlands, the growth
curves have to be changed every ten years. The secular trend in Holland and the
Scandinavian countries is not yet finished. The Harvard standards that were shown
here are already somewhat out of date.
For some time the weight for height (W/H) ratio has been used as a parameter, but
both Dr. Gopalan and Professor Heywood have stressed the point that this is a wrong
concept. Heywood showed convincing data on different mortality rates with a height
for age ratio (H/A) lower and higher than 90 and the same W/H ratio in Papua New
Guinea children.
The relation of the height of the mother with the birth weight of her child, as well
as the obstetric problems around delivery was established many years ago. There are
now also data on low birth weight percentage and child mortality. Indeed, preventive
care of pregnancy starts with the birth of the baby girls who will be pregnant 18
years later.
A most important problem was very well discussed by Dr. Powell from Jamaica.
What are the short-term and long-term effects of under-nutrition on mental develop-
ment? Definite conclusions still cannot be made, but there is some progress. There is
now strong evidence for short-time effects, but the evidence for long-term effects is
still incomplete. Severe protein-energy malnutrition (PEM) superimposed on chronic
malnutrition and poor socio-economic background may have long-term effects. The
effect of intervention studies is most fascinating: the importance of stimulation of the
infants and children is clear. It takes years of continuous study before we can answer
the question if intervention will have permanent effects.
The critical period for under-nutrition is probably 12-18 months after birth, and
thereafter up to circa 4 years. Another important - and still not solved - problem was
discussed by Professor Kusin. Her pregnancy-study - with supplementation in the
3rd trimester - demonstrates a small effect on birthweight (50 gr per 10,000 Kcal
supplied). An effect on infant growth after birth can be seen, and it will be most
interesting to wait for the follow-up studies up to adolescence. Further studies on
energy metabolism of the pregnant mother - particularly on her activity pattern -
275
will be necessary; also studies on changes in body composition during pregnancy in

different areas of the world. I know that such studies are underway.
The pattern of infant feeding was discussed in several papers. Dr. Husaini and co-
workers reported on data concerning a national survey in Indonesia, Dr. Counsilman
about ethnic differences in breast-feeding in Singapore. The picture is very complex.
Both in Indonesia and Singapore, women with higher education seem to have a
lower prevalence of breast-feeding. Nowadays in Western-Europe and the U.S.A. it
is just the opposite.
There is no question that breast-feeding should be stimulated allover the world
and especially in this part of the world. It is a natural way of feeding your child. Also
in countries such as Great Britain and the Netherlands and the Scandinavian
countries the incidence of breast-feeding has increased during the last couple of
decades, but it has now stabilized at around 25-35% at 3 months after birth. In our
Western countries those women who for some reason cannot or will not feed their
babies, should not feel guilty. There are no good scientific data which indicate that in
our Western countries breast-fed infants are better off compared with formula fed
infants. This is also the case when behavioural and psychomotor development is
studied. The physiological effect of hormones and growth-factors recently found in
human milk is so far unknown.
It is very important to establish lactation in the first weeks after birth. Once this
succeeds it will continue and there is no interference by the introduction of solid
foods or by the work of the mother.
The most important nutritional deficiencies in this part of the world are vitamin A,
iron, and iodine deficiency and protein caloric deficiency. The importance of vitamin
A deficiency was discussed in papers by Professor Karyadi and Dr. Bhaskaram. The
causes of vitamin A deficiency are clearly multifactoral. Dr. Bhaskaram showed that
it is now possible to study the effects of vitamin A deficiency on the immune system
with sophisticated modem laboratory techniques. Further work is needed, and at this
moment the role of vitamin A in relation to infectious diseases such as measles,
upper respiratory infections and diarrhoea is unclear. The possible effects of vitamin
A deficiency on child survival, as reported in Indonesia by Sommer et al., are
surprising and further studies are needed.
The problems of iron deficiency anemia were discussed by Dr. Soemantri and Dr.
Sadjiman. It is a common nutritional deficiency, also in Western countries. Recently
there is renewed interest in the problem, as there is increasing evidence that iron
deficiency anemia has at least short-term effects on behaviour and development in
children. Further studies are needed, and some large studies in the U.S.A. are
underway.
The importance of recurrent infections in malnutrition was stressed by Professor
Mata. His data are impressive, and it seems to me that the message is loud and clear:
what is needed is improvement of general hygiene, good sanitation, water control,
preparation of food, but also treatment of acute illness, to make the disease periods
shorter. I was very interested - as is Professor Eeckels - in the loss of appetite -
276
anorexia - in undernourished children. Professor Mata's theory about the role of

interleukines needs careful investigation.
Diarrhoea - both acute and chronic - is still a major problem in greater parts of
this world. Two papers from investigators who worked earlier in the International
Centre for Diarrhoea Research in Bangladesh (Dr. Rowland and Professor Molla),
illustrated the important progress that has been made recently. The effect of diar-
rhoea on growth is impressive. Professor Molla's message on the nutritional
management of diarrhoea was very clear. We know that such a message has to be
repeated again and again to convince the practising doctors, including pediatricians.
Learning is a process of repetition. Dr. Suharyono demonstrated that in severe cases
of chronic diarrhoea special dietary treatment in the hospital can be important, even
life-saving. The role of supplementary feeding in chronic diarrhoea was also
discussed by Dr. Sudigbia. Supplementary feeding as he defines it, is basically
weaning food, home-made, cheap and nutritionally - maybe even therapeutically -
of value. There is a basic need for this kind of weaning food in South East Asia and
it seems to me that the tempe-food as developed in the Nutrition Centre in Bogor and
studied by Dr. Sudigbia should be further evaluated.
What then is the general message for the future? What is the best strategy? Dr.
Gopalan already discussed it in the first paper of this symposium, and it was a good
idea indeed to have this paper at the beginning.
It seems to me that we had at least a consensus that we should aim at integrated

programs, not isolated measures. There are no magic bullets, as Professor Eeckels
pointed out. A holistic approach: we have to combine health programs with
economic programs. We need more and better education of the population by
teachers and health-workers. Dr. Gopalan and others stressed the importance of
education for the pre-marital girl, the adolescent girl. In the East-Java study and
other studies there is a correlation between pre-pregnancy weight and birth weight.
We have to fight infections with good water supplies, sanitation, improvement of
housing conditions, and by immunization programs. We have to promote breast-
feeding, we have to provide better obstetric care.
We may have to compromise between short-term measures and long-term
strategies, but we have to convince the politicians and the funding-agencies that all
of this needs time. The results will be seen after 5, 10, 15, 25 years. Let us be
optimistic. Some of the data presented here show remarkable change and progress
during the last 25 years. Let us not forget how the situation in Western Europe was
25-35 years ago. The countries in Western Europe have experienced dramatic
changes in their health systems since the Second World War. The Netherlands still
had malaria and tuberculosis in 1950, and a severe epidemic of polio in 1955-1957.
Most important of all may be education, if possible at the basic level. The
cooperation of individual people is absolutely necessary. A good example of such an
approach was discussed by Dr. Anna Alisjahbana from Bandung. Her risk-approach
strategy studies by teaching traditional birth assistants (TBA's) are very important
277
indeed. It is a long way to go, and the result will not be visible soon, but the con-
tinuous efforts by her and her co-workers have to be recognized.
We should not forget to teach our medical students and doctors. It was a good
idea to include in this symposium a session on the teaching of nutrition in the
medical curriculum. This is just as important for the medical schools in the Western
world. We have badly neglected the teaching of nutrition in our medical schools. The
same is true for postgraduate teaching.
There is much work still to do. The problems can only partly be solved by doctors
and nutritionists. But we can do something. It is our job, our obligation to do
research, to extend our knowledge, and to teach, to educate the next generation, to
stimulate our young people, who will soon take over our responsibilities. It seems to
me necessary that in South East Asia - as we do in the Western countries - you have
to concentrate your efforts. Not all centers, all medical schools, all departments can
do everything. The concentration of workers from different disciplines in institutes
for teaching and research is one way to solve this problem, and we have seen at this
meeting how important results can be achieved.
Ladies and gentlemen, on behalf of the organizing committee I would like to
thank all of you; the distinguished speakers and guests; our hosts in Yogjakarta; our
committee members; our sponsors who made it generously possible to organize this
workshop.
I think it was a good meeting. The advantage of such a workshop with a limited
number of people is evident. We had very good discussions, sometimes we had
different opinions, but at the same time we renewed our friendships and made new
friends. And all this in such a beautiful place in such a beautiful country, and above
all, with such wonderful people.
Let us be optimistic for the future. Life is not always easy. But the progress of
today is the experience of tomorrow. I wish all of you a good journey home and
much success with your work. Thank you again.
Index of subjects
Aboriginal children family's expenditure 46

growth and nutrition 107 incidence of diarrhea 223
Adaptation misconcept 15 level of education, duration 49
Adolescent girls milk insufficiency syndrome 118
nutritional upliftment 10 nutritional status 263
Anemia promotion 14
control 173 situation, Hong Kong 60
nutritional 177 social freedom 114
heridatory 177 socio-economic groups 4
genetic factors 208 superiority 60
iron deficiency 169, 208 supplementation 14
Anorexia 39 weaning, relative to 89
acute diarrhea 217,234 Breast-feeding patterns
Antenatal care urban and rural communities 52
pregnancy outcome 149 Breath hydrogen test
Antenatal screening 150 lactose malabsorption 243
Appetite improvement
cessation of diarrhea 253 Calorie consumption
Appetite loss diarrhea 231
severe invasive diarrhea 234 Carbohydrate absorption
diarrhea 231
Behavioral effects Catch-up growth 121
marasmus vs. kwashiorkor 160 Child behavior
Beta-carotene 18 activity ratings 126
Birth weight exploration ratings 126
access to food supplies 113 Child mortality 12
appropriate for gestational age 5 Child survival
small for gestationale age 5 female literacy 12
prenatal energy supplementation 72 health care facilities 12
financial management 113 Childhood malnutrition
maternal characteristics 113 behavioral effects 125
personal and family hygiene 113 Cholera
Breast milk banks 57 diarrhea 233
Breast milk intake Cognitive functions
maternal nutritional status 74 missing breakfast 127
breastmilk substitutes 46 Corneal lesions 56
defense against infection 40 Cultural habits
ethnic differences 95 mother's choice offood 58
280
Diarrhea salt with iron 207

acute, dietary management 227
bacterial colonization 3 Gastroenteritis
calorie consumption 231 infant nutritional status 261
carbohydrate absorption 231 Giardia
cessation, appetite 253 infection 40
cholera 232 GOBI 11
chronic, breastfeeding 241 GOBI-FFF 55
chronic, incidence 251 Green leafy vegetables 57
chronic, management 237 Growth 6
chronic, supplementary food 211 excess to food supplies 113
chronic, WHO definition 238, 252 financial management 113
dehydration 235 hygiene 113
digestion, interference 39 maternal characteristics 113
E coli 233 Aboriginal children 107
fat absorption 232 secular trend 161
feeding 203, 232 Growth curves
food consumption 228 standard 8,38,57,77,109,121
growth 218 Growth faltering 121
malnutrition 235 diarrhea, nutritional status 251
mucosal lesions 39 Growth monitoring 13
natural history 203 Growth standards
nutrient absorption 233 application to Asian children 16
nutritional consequences 200 international, appropriateness 146
prevention 223
protein absorption 232 Health knowledge 120
rotavirus 40, 233 Health problems in Indonesia 257
sanitation 39 Health services
shigella 40,219,233 coverage 157
short-chain peptide formula 239 Height for age index
supplementary feeding 199 nutritional status 259
Diarrheal patients Hemoglobulin
clinical intolerance 262 anemia 180
clinical response, treatment 261 electrophoresis 209
fat intolerance 262 Hemoglobinopathy 208
mucosal pattern 263 Hepato-pancreatic function
Digestive absorptive processes 39 nutritional status 263
Dysentry 219 Hookworm infection
iron loss 171, 208
E Coli (escherichia coli) 219 Hospital admittance
diarrhea 233 nutrition and infections 113
Energy supplementation Hospital based medicine 269
prenatal 63 Human milk 89
ETEC (enterotoxigenic E Coli) 219,229 iron availability, bacteria 208
Fetal growth Immune function

maternal dietary uptake 85 iron deficiency 196
Food consumption vitamin A deficiency 195
reduced 38 Immune response 39
Food deprivation Immune system
suppressed infection 41 nutritional status 264
Food habits 164 Immunisation 14
Food supplement Immunity and infection
post diarrheal period 253 vitamin A and iron deficiency 185
Food withdrawal 38 Infant development
Fortification Bayley scales 180
281
iron deficiency 179 infant development 181

Infant feeding
cultural effects 102 Jamu 65,86,118
Ramadan 102
yang and yin 104 Kwhashiorkor 4,229
Infant mortality 12 aflatoxins 160
Infant mortality rate
epidemiological aspects 85 Lactation
organizational aspects 86 initiation, stimulation 120
political aspects 86 Lactose
technological aspects 86 low lactose diet 252
Infection Lactose intolerance
early neonatal 37 after diarrhea 253
nutrition and growth 36 diarrheal patients 262
nutritional status 263 Lactose malabsorption
malnutrition 35 breath hydrogen test 243
metabolic response 41 Chinese newborn infants 247
pregnancy 37 Length for age
prevention, pregnancy 157 median percentage 142
weaning 37 intervention studies 85
Infection and immunity Low binh weight infants
vitamin A and iron deficiency 185 definition cut-off point 122
Infectious disease full term, SGA 5
growth 38 maternal nutritional status 155
Intestinal mucosa
nutritional status 262 Malabsorption
Iodine deficiency 18 nutritional status 261
Iron absorption Malnourished children
foods which increase 174 psycho-social stimulation 135
Iron deficiency reading and spelling grades 134
activity and work capacity 172 Malnutrition 3
anemia 208 caused by diarrhea 227
causes 179 changing clinical profile 161
chronic, causes 178 concepts 4
cognitive function 172 infection 35
deleterious effects 172 intelligence tests 132
development 171 long term effects 131
hematological criteria 209 school grades 132
immune function 196 severe 40
infant development 177 studies with siblings 130
infections 173, 210 studies with matched controls 129
latent, definition 212 visual attention 146
mental development index 181 Marasmic kwashiorkor 4
non-cognitive disturbances 172 Marasmus 4
other effects 173 Maternal health
psychomotor development index 18 I nutritional status 5
Iron deficiency anemia Maternal nutrition
prevalence 169, 208 childhood malnutrition 5
diagnosis 209 Maternal nutritional status
Iron deficient children relation with low birth weight 155
immune responses 193 Medical teaching
infection and immunity 185 integrated 264
Iron fonification 207 Micronutrients 249
Iron supplementation 162 Morbidity
bioavailability 207 pattern 118,121,258
effectiveness in neonates 162 Mortality
282
impact mother & child card 156 weight gain 10

pattern 118,258 Pregnancy outcome
Mother & child card antenatal care 149
effect measurements 155 mother & child card 156
mother & child health 149 Pregnancy study
pregnancy outcome 156 birthweight 69
Mucosal pattern breast milk intake 74
diarrheal patients 263 changes in skinfold thickness 72
home dietary intake 69
Nasi tim 45, 204 maternal anthropometry 70
Neoplasms 18 mother and infant 63
Nutrient absorption offspring and mother 80
diarrhea 233 postnatal growth 76
Nutrient intakes weight gain 72
suboptimal 147 Pregnant women
Nutrition supplementation 55
Aboriginal children 107 Preterm infants
diarrhea 145 AGA 5
malaria 145 Protein absorption
pneumonia 145 diarrhea 232
Nutritional blindness 18 Protein energy malnutrition
Nutritional problems growth standards 141
technological solutions to 18 kwashiorkor 130
Nutritional status marasmus 130
adolescent girls, upliftment 10 mixed 130
breast-feeding 59, 261 Protein losing enteropathy 40, 220
gastroenteritis in infants 261 Protein maldigestion/malabsorption
height for age index 259 peptide formula 240
hepato-pancreatic function 263 Psycho-social stimulation
immune system 264 rehabilitation 135
infection 263
intestinal mucosa 262 Ramadan 102
malabsorption 261 Reading and spelling grades
maternal health 5 child nutritional status 134
mortality 143 Red palm oil 18
motor development 145 Reproductive performance
pattern of feeding 45 maternal nutrition 63
weight/height, Indonesian children 259 Rice/plant protein formula
Nutritional supplementation 157 weaning food 92
Rotavirus 220
Oral rehydration 14 diarrhea 40, 233
Oral rehydration therapy (ORT) 42,218
food based ORT 233 School meal 137, 161
Shigella spp 219
Peptide formula diarrhea 233
acceptance 240 Shigellosis
body weight 240 absorption of nutrients 253
chronic diarrhea 239 antibiotic treatment 253
composition 251 Socio-economic class
diarrhea discontinuance 240 introduction of weaning food 91
protein 240 Socio-economic groups 5
Post partum care 157 Stunting 10, 39
Postnatal growth weight for height 16
prenatal energy supplementation 79 Supplementary feeding programs 19
Pregnancy Supplementary food 201
body weight 10 adult foods 47
283
nasi tim 204 night blindness 22

soft foods 47 protein energy malnutrition 25
solid foods 47 research priorities 25
tempe 204 xerophthalmia 22
weaning food 90 Vitamin A deficient children
Supplementation infection and immunity 185
pregnancy and lactation 136 salivary SIgA and T cells 189
pregnant women 55 Vitamin A fortification 30
solid food 45 Vitamin A therapy
large dose 190
Teeth H 20 2, 02 production, neutrophils 188
eruption of deciduous 145 VitaminE 56
Thalassemia
anemia 178, 209 Wasting effect 39
Traditional Birth Attendants (TBA) Weaning food 89
Indonesia 149 early introduction, hazards 90, 118
milk cereal fonnula 92
Undernutrition nasi tim 45
long tenn effect 125 recipies 94
transient effect 125 rice/plant protein fonnula 92
soft food 45
Vitamin A solid food 45
child health 21 transmission of fecal pathogens 53
diarrhea, therapeutic role 211 Weaning food introduction
green leafy vegetables 33 socio-economic class 91
immune system, mechanism 190, 209 Weight for age
immunoregulatory agent 22 median percentage 142
potential toxicity 32 Weight for height
pregnancy and lactation 31 stunting 16
prevention programs 31 nutritional status, Indonesia 259
teratogenic effects 32 Weight for length 141
single large oral dose 190 joined effect of length for age 144
subclinical deficiency 190 median percentage 142
synthetic vitamin A 18 Weight gain
Vitamin deficiency 18, 21 pregnancy 10
acute respiratory infection 25
antibody response, D&T toxoid 188 Xerophthalmia
blindness 22 mortality, PEM 186
clinical aspects 23 vitamin A deficiency 22
diarrhea 25
functional aspects 22 Yang and Yin 104
immune function 195
immunological studies, children 187 Zinc deficiency
keratinisation, epithelium 22, 186 iron deficiency 173
morbidity and mortality 210 Zinc supplement 249

C. Gopalan Auth., Prof. H. K. A. Visser, Dr. J. G. Bindels Eds.

Încărcat de

Informații document

Titlu original

Drepturi de autor

Formate disponibile

Partajați acest document

Partajați sau inserați document

Opțiuni de partajare

Vi se pare util acest document?

Este necorespunzător acest conținut?

Drepturi de autor:

Formate disponibile

C. Gopalan Auth., Prof. H. K. A. Visser, Dr. J. G. Bindels Eds.

Încărcat de

Drepturi de autor:

Formate disponibile

cmLD NUTRITION IN SOUTH EAST ASIA

Eighth Nutricia Symposium

H.K.A. VISSER and J.G. BINDELS

Kluwer Academic Publishers

Nutricia Symposium (8th: 1989: Yogyakarta, Indonesia)

ISBN-13: 978-94-010-7393-6 e-ISBN-13: 978-94-009-1996-9

Published by Kluwer Academic Publishers,

Kluwer Academic Publishers incorporates

Sold and distributed in the U.S.A. and Canada

In all other countries, sold and distributed

Printed on acid-free paper

All Rights Reserved

March 1990 THE EDITORS

OPENING OF THE SYMPOSIUM xv

C. Gopalan, Childhood malnutrition in South East Asia: Concepts and

C.A Powell and S.M. Grantham-McGregor, Selective review of studies on

M.G.M. Rowland, Diarrhea, growth and feeding practices 217

Pitono Soeparto, Integrated medical teaching curriculum on child nutrition 257

CLOSING REMARKS 273

INDEX OF SUBJECTS 279

Prof. S.H. Pudjiadi, Child Health Department, Medical Faculty University of

Ladies and gentlemen,

General aspects of child nutrition in South East Asia

Chainnan: H. Iskandar Wahidiyat

Table 1. Some examples of changing concepts regarding childhood malnutrition

being extolled. The futility of strategies based on such misconceptions came to be

Improvement of maternal nutrition - an essential part of a

Age PretermAGA Full-term SGA Full-term AGA

Birth 44.6 43.4 49.3

Source: S. Ghosh, S.K. Bhargava and V. Kapani (unpublished observations).

Age PretermAGA Full-term SGA Full-term AGA

Birth 1.8 1.7 3.0

Source: S. Ghosh. S.K. Bhargava and V. Kapani (unpublished observations).

65 income p.cap. >200Rs

55 income p.cop. :s;SORs

50 -*- birth weight ~2500 g

45 income p.cap. ~50Rs

8 income p.cap. >200Rs

2:birth weight .;2500

3:birth weight >2500

4:birth weight ,;;2500

Maternal height (cm) Income group Incidence ofLBW

Source: S. Ghosh, S.K. Bhargava and V. Kapani (unpublished observations).

Recent Studies of the Nutrition Foundation of India. The Nutrition Foundation of

--B- < 2.5 kg

Wt;>38kg & Ht>145cm

-+- Wt;::38kg & Ht< 145cm

-+ Wt<38kg & Ht<145cm

Nutritional upliftment of the mothers-to-be - adolescent girls. In most Asian

"Child Survival" and "GOBI"

"Child Survival". "Child Survival" has now almost displaced old-fashioned

Table 5. Birth rates and death rates in U.P. and KeraIa

State Birth Death Infant Child

KeraIa 23.3 6.5 31.3 10.2

State Normal Grade of malnutrition

Kerala 16.4 45.3 33.5 4.8

Growth-monitoring, Growth-monitoring has unfortunately been pushed in many

Immunisation. The drive for universal immunisation is welcome. However, it must

Breast-feeding. Promotion of breast-feeding as the mass-anchor of infant nutrition is

"Small is healthy for the poor?" - The "adaptation" misconcept

Are international growth standards applicable to Asian children?

NCHS (50th peretile)

Technological solutions to specific nutrition problems