ROUNDTABLE

Registration and monitoring of pregnant women in Tamil Nadu, India: a critique

Rakhal Gaitonde
Training and Research Associate, Society for Community Health Awareness, Research and Action, Chennai, Tamil Nadu, India. Correspondence: rakhal@sochara.org

Abstract: In 2008 a pregnancy registration system was introduced in rural Tamil Nadu, India, which is now being scaled up. It will collect data on antenatal, delivery and post-partum care in pregnant women and infant health. This is seen as an important public health intervention, justified for its potential to ensure efficiency in provision and use of maternity services. However, from another perspective, it can be seen as a form of control over women, reducing the experience of safe pregnancy and delivery to a few measurable variables. The burden of implementing this task falls on Village Health Nurses, who are also women, reducing their time for interacting with and educating people and visiting communities, which is their primary task and the basis on which they are evaluated. In addition, they face logistical constraints in rural settings that may affect the quality of data. In a health system with rigid internal hierarchies and power differentials, this system may become more of a supervisory and monitoring tool than a tool for a learning health system. It may also lead to a victim-blaming approach (you missed two antenatal visits) rather than health system learning to improve maternal and infant health. The paper concludes by recommending ways to use the system and the data to tackle the broader social determinants of health, with women, health workers and communities as partners in the process. 2012 Reproductive Health Matters Keywords: health information systems, registration of pregnancy, maternity services, maternal and infant health, Tamil Nadu, India Recently there have been a number of press reports in the state of Tamil Nadu in India announcing a new registration system for pregnant women and infants, called the Pregnancy and Infant Cohort Monitoring and Evaluation (PICME) system.1,2 This system was first introduced in Tamil Nadu in April 2008 in all rural areas and now in 2012 also in urban areas (Personal communication 2012).*
*In preparing this paper, I met three senior officials in Tamil Nadu and asked for their opinions, which are cited here as personal communications. They are not named to protect confidentiality. The rest of my observations were formed during regular visits to a number of rural areas as part of a project on community monitoring and planning for health, which I manage. I made specific inquiries about this new scheme on occasion, but mostly it was discussed spontaneously in meetings, for which I have field notes. Thus, my analysis and understanding of the issues was informed mostly by what I observed and heard in the field.

Tamil Nadu is held up worldwide as a model and a success story for having reduced maternal mortality from 450 deaths per 100,000 live births in 1980 to 90 per 100,000 in 2006.3 The documented reasons for this remarkable decline include consistent political commitment, not changing senior policy makers and officials too often and thus maintaining continuity in policy and programme management, innovative schemes such as maternal death audits, and institutional and managerial health system strengthening.3,4 Given the fact that the state is among the few in India with a specialized public health cadre in charge of primary and first referral level of health care, the health system has given a lot of importance to information. As early as 1998, the state implemented a DANIDA-funded Institutional Service Monitoring Report, which evolved into an effective monitoring mechanism.5 Similarly, there was a lot of emphasis on early registration of pregnancies with the Village Health Nurse, with the
Doi: 10.1016/S0968-8080(12)39619-5

118

Journal contents online: www.rhm-elsevier.com

R Gaitonde. Reproductive Health Matters 2012;20(39):118124

emphasis on registering by the third month.3 This was meant to allow the Village Health Nurse to plan her work according to the number of registered antenatal women (among other things). This system, while helping the Village Health Nurse at the local level, was described in a 2008 report by the Directorate of Public Health and Preventive Medicine as having certain inadequacies,6 including poor linkage with service delivery, due to the various data collected being held and maintained in separate registers, poor reporting of outcomes and under-reporting of infant deaths.6 To address this issue, the Directorate has designed a pregnancy and infant cohort monitoring system, designed for the Village Health Nurses to record details of the pregnant mother and the infant in a single line, starting from the date of antenatal registration till the first birthday of the baby. 6 Thus, PICME was developed by the National Informatics Centre for the states Department of Public Health and Preventive Medicine, with support from several officials who championed it. The system consists of four modules for the pregnancy cohort and five modules for the infant cohort. The main information captured includes:

Pregnant women general demographics; antenatal check-ups and scans; antenatal referral and delivery; and post-partum details. Infants general; referral; immunization; growth monitoring; infant death.7

The collection of data is intended to lead to three broad outputs:

a database that contains information on the kinds and extent of pregnancy-related risk factors among currently pregnant women; routine generation of reports with the particulars of individual women who need to be followed up, such as women with high risk factors, estimated dates of delivery, and need for immunization; use of the reports to help peripheral health workers keep track of their catchment population and potentially generate automatic SMS texts, e.g. to remind pregnant women of the need to utilize appropriate services during the antenatal period and for delivery and the post-partum period.6,7

The unique PICME identification numbers generated by this system for each registered woman will be used to link this database with another

dealing with eligibility for the Maternity Benefit Scheme. The Maternity Benefit Scheme provides pregnant women with a total of Rs 12,000 (US$228) in three tranches one after they avail of all required health services, one after delivery in a government institution, and one after a live birth after the infant has completed the full immunization schedule. No one will be able to get the maternity benefit unless they are registered in the PICME system.8 The introduction of PICME is happening at the same time as the Government of India is launching a national level MotherChild Tracking System (MCTS), developed by the Health Ministry.9 The Government has also just launched a Unique Identification scheme under the aegis of the Unique Identification Authority of India (UIDAI), in which every person has a Unique Identification card.10 The card is to be used among other things for access to public health services as one of its key uses.11 Similarly, a High Level Expert Group on Universal Health Coverage (HLEG), instituted by the Planning Commission of India, has also suggested a biometric identification card, based on a unique physical characteristic such as fingerprints or iris patterns, to ensure access to all so people can realize their right to health as well as ensure portability of health care across the country.12 It thus seems that both the use of information technology (IT) and the creation of databases with large amounts of information are seen as important public health interventions. Their use is being justified due to their potential to gather large amounts of information and ensure efficiency, and thus contribute to improving health status. By introducing an IT-based system for monitoring a huge data set at the state level, as compared to the earlier decentralized registers, the PICME system introduces a significantly different situation from the past. However, while more information and more accurate information can certainly help in better planning and delivery of services, and indeed is a crying need in many health systems, there are a number of concerns being raised. Not only are data now available in real time at the state level, but also, given that they are the door to the Maternity Benefit Scheme, the system introduces the possibility of pressure on pregnant women as well as significant pressures on peripheral health workers. The rest of this paper describes how information management systems can lead to distortions in the way the health system sees women, recognizes their concerns and responds to their lived 119

R Gaitonde. Reproductive Health Matters 2012;20(39):118124

realities. It also critically analyses the dominant discourse accompanying this form of bio-surveillance, of which the PICME system is an exemplar.

Assumptions underlying the PICME scheme

The proposed monitoring using PICME is expected to contribute to the further reduction of maternal and infant mortality in the state. The dominant themes invoked to justify such tracking and IT-based surveillance seems to include assumptions that:

Once the patient is registered with PICME, village nurses will visit and take her health status data at least five times before delivery. This ensures safe deliveries and identification of women with risks. Iron tablets and other medication will be supplied through health workers in local PHCs, said a senior official. (The Times of India, 2012)2 Thus, these newspapers seem to portray this as an intervention that, by closely following up women and monitoring their utilization of services, will fill a critical gap and contribute to furthering their welfare and a reduction of maternal mortality. There is little mention or analysis of the broader determinants of maternal mortality. Instead, it is presented as a purely techno-managerial problem, amenable to reduction through better information management.

Information gaps constitute a crucial determinant of adverse health indicators; thus, more information will translate into better outcomes;6,7 More efficient and complete data collection will lead to better and more effective decisions, as problem areas and problem individuals can be identified;6,7 IT-based programmes help in increasing the transparency and accountability of the system and thus help in reducing corruption. (Personal communication 2012) Possible sex selection can be tracked and prevented, thus helping to rebalance the adverse sex ratio (Personal communication 2012). This is in the context of an adverse male-female sex ratio in the state, with many pockets much worse off than the average and some further decline shown in the latest census, sparking off further concern. Generation of new knowledge from the data, which can be fed back into the health system and contribute to the field.13

A conceptual analysis of PICME

PICME and similar interventions can be analyzed from a number of perspectives. At one level, information that is fed into a responsive system can greatly improve the capacity of the system to provide better quality care. However, from another perspective, the intensive collection of information and its inherent acceptance of certain behavioural norms (those which can be measured) can be seen as a form of state control over people, as developed by Foucault.14 By defining what is right behaviour and measuring and monitoring it, the state sets out to homogenize (and potentially control) the behaviour of diverse individuals and populations, little recognizing the broader determinants of this diverse behaviour and the constraints under which it takes place. The focus on information to do with the availability and consumption of specific measurable health care services can also be seen as deflecting attention from the broader social and cultural determinants of health,15 and in this case, also the reality of obstetric complications and maternal and newborn deaths that are not due to individual behaviour. The woman concerned and her family and community are probably primarily interested in whether or not such an intervention improves her experience, her trust and feeling of security when accessing the system. From a feminist point of view, the critical question when interrogating such interventions remains: what is the extent to which they facilitate or obstruct feminist goals of equity and emancipation?16

Presentation of the scheme in the media

In two of the leading dailies, the PICME system was reported as follows: With the inability to closely follow-up on women during their pregnancy period impairing its ability to bring down maternal mortality rate, the State government has rolled up its sleeves to address the problem [...]The software is calculated to keep tabs on the mother throughout her ante-natal period and the child until the age of one year, RT Porkaipandian, Director of Public Health, has told The Hindu. [] For a year after the child is born, the little ones weight, height and immunization details will come up on the screen. We are sure this will make an impact on our efforts to reduce our MMR and infant mortality rate, says Dr. Porkaipandian . (The Hindu, 2011)1 120

R Gaitonde. Reproductive Health Matters 2012;20(39):118124

Moreover, by linking the PICME system to the Maternity Benefit Scheme in Tamil Nadu,5 there is a significant element of coercion introduced, putting pressure on women to register and consume a set of services as defined by the government, which it is assumed they will benefit from. Apart from these larger conceptual issues we can discuss the emerging concerns as follows.

Complexities of data collection and processing

While an IT-based system is theoretically very efficient, the practical difficulties of data entry are many. In rural areas of Tamil Nadu, with power outages routinely lasting an average of ten hours, internet connections are very poor. Though each Primary Health Centre is equipped with one computer, in the context of no power and poor internet connectivity, this is not enough. Thus, Village Health Nurses have to spend huge amounts of time and money to travel to the nearest town with an internet caf to make these entries into the PICME system, which are not compensated for by the government. While the initial high load of entries, due to the need to catch up on old cases, presently queers the system, once this load is cleared there will still remain many practical issues that need to be surmounted if the government wants to ensure complete and accurate information. In a situation where the burden of implementation of all public health interventions ultimately falls on the most peripheral of workers, who also happen to be women, additional work burdens like this one pile up, leading to huge amounts of frustration. The Village Health Nurses end up doing a lot of administrative and data entry work instead of interacting with and educating people and visiting communities, which is their primary task and the basis on which they are evaluated.

Increased pressure on health worker performance

Collection and uploading of data also need to be seen in the light of the persistent tussle between supervisors/managers and health workers, and the efforts of supervisors and managers to control health workers, especially those at lower rungs of the hierarchy.17 The requirement of real-time transmission of data allows supervisors at different levels to monitor the performance of health workers, based on what is reported (or not) through the

system, when in fact there may be a number of systemic reasons affecting their data-related performance, alongside a number of dimensions that are not measured or measurable, where they are in fact contributing. In the context of a health system with rigid internal hierarchies and power differentials, PICME may become more of a supervisory and monitoring tool rather than a tool for learning, a risk which can only be evaluated further down the line. A number of recently introduced communitybased monitoring schemes, which are part of the flagship National Rural Health Mission, where community-based committees are trained to monitor their entitlement to services from the public health system,18 may further enhance the frustration of the Village Health Nurses, who will be overburdened with administrative tasks. People at community level routinely express dissatisfaction with the performance of Village Health Nurses in terms of their not visiting communities and talking to women and their families enough. This dissatisfaction occurs without them realizing the systemic reasons for it. The Village Health Nurses are squeezed between unrealistic expectations from the Department to do a whole lot of work which they do not perceive as their core work, and the dissatisfaction of the community, who have been the source of much satisfaction and fulfillment for them. The enormous pressure to collect data may also lead to the de-humanization of how Village Health Nurses interact with people. This is best exemplified by the experience of one pregnant woman when the Village Health Nurse came to collect information for the PICME registration. Despite the fact that she was suffering from diarrhoea and asked the Village Health Nurse for advice and treatment, the nurse did not offer her anything she said she had come only to fill in the form. She did not even refer her to the nearby health centre and dismissed the diarrhoea by saying it was because the babys hair had grown and was irritating her womb.

Twisting the logic of health care towards commodification

Many of the specific aspects of pregnancy and pregnancy care that are on the form for monitoring, such as, ultrasound scans and results, monitoring of weight and blood pressure, blood sugar, height of uterus, etc. are collected and interpreted in the framework of the now rejected high-risk 121

R Gaitonde. Reproductive Health Matters 2012;20(39):118124

approach to pregnancy, While these data are certainly important for following up individuals who have or who develop high-risk factors such as anaemia and hypertension, their use at a systems level takes up a lot of resources and distracts from the comprehensive care that all pregnant women need (and which is their right). Linked to these points is the much broader definition of safety in pregnancy and childbirth by women than that of the state. At present, the states conceptualization of safe delivery focuses solely on things like delivery in an institution, the presence of a skilled birth attendant, proper referral linkages and evidence-based practices. A womancentered definition, on the other hand, while giving critical importance to the presence of such infrastructure, also focuses on the quality of services and of care, the safety aspects of the whole experience of pregnancy and childbirth, such as being able to trust the care being given, trust that medicines or essential interventions are available if a complication occurs, being treated with dignity and respect, and being given the information and support that women need.19 By reducing the whole experience to a few measurable and therefore monitorable variables, the health system reduces (or refuses to acknowledge) the complexity of the experience. It is likely to focus more on individual behaviour in terms of consumption of services. This can very easily lead to a victimblaming approach (you missed two antenatal visits; you delivered at home) rather than an approach that tries to understand and tackle existing structural and social determinants to improve maternal and infant health.

dary centres, and a lack of specialists at district level, all of which are necessary to reduce maternal morbidity and mortality. While PICME will be able to pick up and measure markers of a womans poor status in society, such as poor nutrition (anaemia) and violence against women, unless these problems are addressed structurally rather than focusing on the individual, little impact is likely to accrue.

Issues of access to data and confidentiality

While the mechanisms for collecting the data are still being worked out, little attention seems to have been given to the question whether women will have access to information collected regarding their pregnancy, such as their HIV status, for example. There are also a range of issues pertaining to confidentiality and especially safety of the data, given the potential linking of various databases in the future. What are the safeguards protecting womens privacy and confidentiality? Could the system be used to track the outcome of every pregnancy, e.g. if a woman has an abortion, and thus to restrict access to abortion? Particularly in the Indian context, the possibility of using the system to help to track, curb and even prosecute the use by women of ultrasound scans to seek abortion on the grounds of sex selection is a big concern. Other aspects, such as the private insurance industry or a pharmaceutical company seeking to purchase the database to sell insurance or medicines and health products, point to need for strict regulation, to ensure that such abuses cannot take place. While at this point there may not be clear evidence for any of this, it is worth raising these issues so that they do not become problems later.

Weaknesses in the health system that collects the data

Will increased information as collected on the intended form really lead to a reduction in maternal and newborn deaths? The answer is not clear. The collection of such information, its analysis, and the generation of reports will only have an impact on womens and infants health in a system that has the capacity to collect and to analyze the data, feed the analysis into a responsive planning system with adequate and responsive staff, as well as the commitment, resources and capacity to respond as specific issues emerge. This does exist to some extent in the state of Tamil Nadu. However, more problematic is the fact that there is poor referral linkage, poor quality of care at secon122

In conclusion
This paper does not in any way question the importance of or the critical need for information and information systems for health care. In fact, the development of such systems is a critical gap in a number of systems and one of the key ingredients of any health system that aims at providing universal access and sees health as a human right. However, it would be nave to see such systems as a value-neutral activity, with only benefits in terms of efficiency and better evidence for planning. What information is collected, the way the information is used, the framework within which it is analysed and fed back into the system, the capacity of the system to use it for planning, and the resources and the

R Gaitonde. Reproductive Health Matters 2012;20(39):118124

willingness to bring about change are all critical questions that need to be answered. In addition, answers to the larger questions whether the information collected will be used for surveillance, supervision and control, or to support a learning system with health workers and communities as partners in the learning process will be the final determinant as to whether PICME or any similar intervention will truly contribute to the empowerment of women, health workers and the community or become yet another tool of control, blame and manipulation. In the absence of active and well-informed public debate and engagement of decision-makers with a wide range of views, this system may end up further commodifying both the practice of medicine and pregnant women in the eyes of practitioners, with an adverse effect on equity and well-being. To make information systems truly contribute to womens health, the health system should:

Initiate and engage in public debate and take on board a number of diverse views during the evolution and evaluation of PICME.

Explore ways of learning about the stories behind the numbers and develop systems to respond to and capture these during the policy-making process. One such possibility is to use this data in conjunction with community-generated data, such as that from the community monitoring processes described above. Bring the perspectives of peripheral health workers (who will be given the responsibility to collect and enter data) to the table right from the beginning. The information that is produced is only as good as the data that goes into the system. Enable the capacity of the system to interpret and use this data at all levels. This would include not only technical capacity, but also transforming hierarchical systems into more democratic learning systems. Explore ways in which this data can be used to tackle the broader social determinants of maternal and infant health, seeing women and communities as partners in the process rather than as passive beneficiaries.

References
1. Kannan R. New software to help bring down maternity mortality rate in State. The Hindu 17 February 2011. At: www.thehindu.com/news/cities/chennai/ article1462702.ece. Accessed 20 April 2012. 2. Janardhanan A. Govt will keep tab on health of moms-to-be. The Times of India. 5 January 2012. At: http://articles.timesofindia.indiatimes.com/2012-01-05/ chennai/30592946_1_health-workers-gestationaldiabetes-pregnant-women. Accessed 20 April 2012. 3. WHO Regional Office for South-East Asia. Safer pregnancy In Tamilnadu. From vision to reality. New Delhi: WHO, 2009. At: http://whqlibdoc. who.int/searo/2009/9789290223566.pdf. Accessed 25 April 2012. 4. Padmanaban P, Raman PS, Mavalankar DV. Innovations and challenges in reducing maternal mortality in Tamil Nadu, India. Journal of Health Population & Nutrition 2009;27(2):20219. At: www.banglajol.info/index.php/ JHPN/article/view/3364/2811. Accessed 25 April 2012. 5. Ravindran TKS. Health system reform initiatives in Tamil Nadu: contribution to improvements in sexual and reproductive health services. Chengalpattu: Rural Womens Social Education Centre, 2009. 6. Directorate of Public Health and Preventive Medicine. PHC Online Reporting System Handbook. Chennai: Directorate of Public Health and Preventive Medicine, August 2008. Department of Public Health and Preventive Medicine. Pregnancy and Infant Cohort Monitoring System. User Manual Version 1.0. Chennai: Department of Public Health and Preventive Medicine, undated. Health and Family Welfare (EAP II-1) Department. Government Order GO. Ms (276). 3 November 2011. Mother and child tracking system. Informatics 2012;20(4): At: http://informatics.nic.in/Lead_Story1.php. Accessed 21 April 2012. Unique Identification Authority of India. At: http://uidai. gov.in. Accessed 20 April 2012. Nilekani N. Building a Foundation for Better Health: The role of the Aadhaar [Editorial]. National Medical Journal of India 2011;24(3):13335. At: http://nmji.in/archives/ Volume-24/Issue-3/Editorial-II.pdf. Accessed 20 April 2012. High Level Expert Group on Universal Health Coverage. High Level Expert Group Report on Universal Health Coverage in India. New Delhi: Public Health Foundation of India, November 2011. Automation of pregnancy monitoring system boon to village health nurses. The Hindu, 10 July 2010. At: www. thehindu.com/news/states/tamil-nadu/article504772.ece. Accessed 20 April 2012.

7.

8. 9.

10. 11.

12.

13.

123

R Gaitonde. Reproductive Health Matters 2012;20(39):118124 14. Foucault M. Birth of the Clinic. New York: Vintage Books, 1994. 15. Adams V. Medicine, science, and democracy in the developing world. In: Doctors for Democracy: Health Professionals in the Nepal Revolution. Cambridge: Cambridge University Press, 1998. p.1529. 16. Hodges S, Thorsby K. Introduction: situating technology. Womens Studies Quarterly 2009;37(1/2):1118. 17. Balasamo A. Public pregnancies and cultural narratives of surveillance. May 2002. At: www.vaccinationnews.com/ dailynews/May2002/PublicPregnancies.htm. Accessed 20 April 2012. 18. Community based monitoring of health services under the National Rural Health Mission. At: www.nrhmcommunityaction.org. Accessed 20 April 2012. 19. Centre-staging safety of delivery: moving beyond the home vs. institution debate. Womens Voices Series 3. Chengalpattu: Rural Womens Social Education Centre, undated.

Rsum Un systme denregistrement des grossesses introduit en 2008 dans le Tamil Nadu rural, en Inde, est maintenant en cours de gnralisation. Il recueillera desrenseignementssurlessoinsprnatals,obsttricaux et du postpartum chez les femmes enceintes et sur la sant infantile. Il est considr comme une importante intervention de sant publique, capable de garantir lefficacit de la prestation et de lutilisation des services de maternit. Nanmoins, dun autre ct, on peut y voir une forme de contrle des femmes, qui rduit lexprience dune grossesse et dun accouchement moindre risque quelques variables mesurables. Les infirmires de village, aussi des femmes, sont charges de mettre en uvre ce systme, ce qui rduit le temps dont elles disposent pour linformation des populations et les visites aux communauts, qui sont leurs tches principales et la base sur laquelle elles sont values. De plus, dans les zones rurales, elles se heurtent des obstacles logistiques qui peuvent influencer la qualit des donnes. Dans un systme de sant aux hirarchies internes rigides et aux carts de pouvoir, ce systme peut devenir davantage un outil de supervision et de suivi que linstrument dun systme de sant dsireux dapprendre. Il peut aussi conduire une approche blmant les victimes ( vous avez manqu deux visites prnatales ) au lieu dencourager le systme de sant apprendre amliorer la sant maternelle et infantile. Larticle conclut en recommandant des utilisations du systme et des donnes pour sattaquer aux dterminants sociaux plus larges de la sant, avec les femmes, les agents de sant et les communauts comme partenaires.

Resumen En 2008 se introdujo un sistema de registro de embarazos en el pueblo rural de Tamil Nadu, en India, el cual ahora se est ampliando. Se recolectarn datos sobre la atencin antes, durante y despus del parto en mujeres embarazadas y sobre la salud infantil. Esto es visto como una importante intervencin en salud pblica, justificada por su potencial de garantizar eficiencia en la prestacin y utilizacin de los servicios de salud materna. Sin embargo, desde otra perspectiva, se puede ver como una forma de control sobre las mujeres, que reduce la experiencia de embarazo y parto seguros a unas pocas variables mensurables. La carga de cumplir esta tarea recae sobre las Enfermeras Comunitarias, quienes tambin son mujeres, lo cual reduce su tiempo para interactuar con las personas y educarlas y para visitar a las comunidades, su principal tarea y la base para evaluarlas. Adems, se enfrentan con restricciones logsticas en zonas rurales, que podran afectar la calidad de los datos. En un sistema de salud con rgidas jerarquas internas y diferenciales de poder, el sistema de registro podra pasar a ser ms como una herramienta de supervisin y monitoreo que una herramienta para un sistema de salud en proceso de aprendizaje. Asimismo, podra provocar la tendencia de culpar a la vctima (falt a dos consultas de atencin prenatal) en vez de ayudar al sistema de salud a aprender a mejorar los servicios de salud maternoinfantil. Se concluye el artculo con recomendaciones sobre cmo utilizar el sistema y los datos para enfrentar los determinantes sociales de la salud, con las mujeres, profesionales de la salud y comunidades como entidades aliadas en el proceso.

124