"

I
(
,I
'"
I
j
1
J

i
J
I
1
I


I
I
\
f
I
&earnceO. Wau!>. 1.1("
Occupational
Therapy lor
Physical
Dysfunction
Fourth Edition
Editor
A. TroDlhJy
O.T.R.,F.A.O.T.A.
Professor, Department of Occupational Therapy
Sargent College of Allied Health Professions
Boston University
Boston, Massachusetts
Williams &
HO,NG KONG
teNDON. MUN!CH SYDNEY" TOKYO
A W",VI::RlY COMPAN Y
NCSStudyGroup
ScientificInquiry
8/05
DonStraube,PT,MS,NCS
Topicscovered:
Theorydevelopment
PrincipalsofMeasurement
Sensitivityandspecificity
Reliability
Validity
Researchdesigns
Experimental
Quasi-experimental
Single-subject
Parametricandnonparametricdata
Descriptivestatistics
Statisticalinference
Analysesofvariance
Analysesof frequencies
Correlation
Regression
Epidemology
1
Theory Development
Theory: an abstract idea or collection of ideas used to explain physical or social
phenomenon. See Figure.
Theories are not directly testable, but hypotheses are. Researchers set up
hypotheses based on the theory> collect data > statistically analyze / test the data
> interpret the results and either support our hypothesis (and indirectly the
theory), or don't support our hypothesis and theory (positivism approach).
We use theoretical frameworks I paradigms to help describe theory and
influencing factors and underlying assumptions.
Basic tenets of a theory:
1. Evolves from experience / research
2. Dynamic-
Newtonian Physics to Theory ofRelativity
3. Not directly testable
4. Requires scope conditions - conditions or situations under which the theory will work.
5. Requires operational definitions - ofthe major constructs ofthe theory (e.g., tone,
normal movement). When operational definitions are absent, then there is disagreement
among researchers (lack of consensus related to phenomenon).
Principals of Measurement
Measurement: the process ofassignin numer to objects to represent quantities of
to certain rules. ere is a difference between numerals and
numbers!!! umb ve conjoiqt additivity (can addlsubtract/multiply/divide and
maintain meaning of numbers). Numerals do not have this!!! This is important to
consider when applying statistics ... for example the FIM. This limitation is overcome by
models associated with Item Response Theory (Rasch Measurement Model, 2 & 3
parameter models).
Scales
Nominal: objects or people are assigned to categories based on some cri,erion.
(e.g., yeslno, OIl, present/absent). Uses vs numbers!!! I (l.Au:.:
Ordinal: categories are rank-ordered on the basis of an operationally defined
characteristic. Higher levels usually respond to "more" of the construct of
interest. Uses - not numbers! !
(e.g. 0-10 pain scale, FIM, Berg Balance Scale, MMT)
Interval: has the rank-order characteristic ofan ordinal scale, but also equal
intervals between response categories. These are not related to a true zero, so not
representing an absolute quantity. (e.g., temperature, IQ, ROM). Involves
numbers .
...----
2
Ratio: numbersrepresentingunitswithequalintervalsandhave~ (e.g.,
age,bloodpressure,dynamometer). Involvesnumbers.
Thetypeofdatayouhavewilldictatethetypeof statisticsused. Non-parametric
statisticsareusedfornominal!ordinaldatainwhichthedataarecomposedof numerals.
Parametricstatisticsareusedforintervalandratiodatainwhichthedataarecomposedof
numbers.
Sampling
Samplinginvolvesselectingsubjectsthatarerepresentativeof thepopulationof
interest. Inclusion!exclusioncriteriaareoftenusedtopickthestudysUbjects. Random
samplinghelpstoincreasetheabilitytogeneralizefromthestudypopulationtothe
targetpopulation,asthesamplesarethoughttobemorerepresentativeofthegeneral
popUlation.
Probabilitysampling
Simplerandomsampling
Systematicsampling
Samplingintervalsused
Non-probabilitysampling
Conveniencesampling
Quotasampling
3
Sensitivity I Specificity
When a measurement tool is intended to be used to screen patients for the presence or
absence of a condition (e.g., risk for falling), then the understanding ofthe test's
sensitivity I specificity is important.
Sensitivity: ability of a test to obtain a positive result when the condition is present (true
positive). Sensitivity is more important when the risk associated with missing a condition
is high.
Specificity: ability of a test to obtain a negative result when the condition is absent (true
negative). Specijicity is more important when the risk associated with further
intervention is substantial.
Tests are never both high in sensitivity and specificity, but rather a trade offoccurs.
When possible, using tests that compliment each other is helpful.
Diagnosis
Dx+ Dx-
Test Results
Sensitivity =a I a + c Specificity = d / b +d
Reliability: the extent to which a measurement is consistent and free from error
.;.NA 0{
,:p Types of reliability
; AJI
Observed score =true score + error
Sources oferror: individual! instrument! variable being measured
Intrarater reliability: same person perform measure over time
'pr Assesses error from the individual and variable being measured
Interrater reliability: different people measuring same thing
Assesses error from individual
Test-retest reliability: repeat measures on same sample on two different occasions
Assesses if error from the instrument / variable being measured / rater
Internal consistency: concerned with the extent to which the items of an
instrument measure the same characteristic. Different from validity!
Split half reliability
Cronbach's alpha
4
Statisticsforreliability
Kappa K ,- }ttnrlilJ ctif
fPfh6 10
WeightedKappa h.-; ::;.
::t:ec, Lj, 0
_I..ll

ICCvs .
c

/0
Howwouldyoudesignaclinicalstudytoassessthevariousreliabilitiesofa er
I
commonlyusedtest(eg.,ROM,MMT,BergBalanceScale,etc)?
q
r; '''l--
Inter-raterreliability "1
3 Ct
Q &.uz..,
tf
"2>
/f
,.
"1-
J
Intra-raterreliability
Considerissuestopopulationspecificity..
5
Validity
Face:weakestform ofvalidity- doestheinstrumentmeasurewhatitissupposed
tomeasure? Basedon"expert"opinion- isthemethodplaUsible?
Construct:establishestheabilityof aninstrumenttomeasureanabstract
constructandthedegreetowhichtheinstrumentreflectsthetheoretical
componentsoftheconstruct. (Correlationof motorskilldevelopmentwithage).
Striveforinstrumentsthatareunidimensional- measureoneconstruct(FIM).
Supportedby:knowngroupsmethod,discriminantvalidity(CPl),factoranalysis,
IR Tmethods (fitstatistics)
Content:Indicatesthattheitemsthatmakeupaninstrumentadequatelysample
theuniverseof contentthatdefinesthevariablebeingmeasured. MostusefulforI
duringthedevelopmentofquestionnairesandinventories. (e.g.Testofmath
ability- butwordbasedproblems). Usuallyagreeduponbyexpertsinthearea.
Criterionrelatedvalidity:indicatestheoutcomeof oneinstrument(targettest)
canbeusedtosubstitutemeasurefora"goldstandard"criteriontest. (ROMI
radiographs) Canbeconcurrentorpredictive.
Concurrent:estabilishesvaliditywhentwomeasuresaretakenatthesame
time. (e.g.RIC-FASandFIM,FunctionalReachTestandposturalsway
measuresfrom aforceplate)
Predictive: establishesthattheoutcomeofthetargettestcanbeusedto
predictafuturecriterionI score. (e.g.BergBalanceTest,Functional
ReachTest)
ConsiderI discussthevalidityofthesetests: MMTI AshworthScaleI ROMI
FIMI anyothers?
Considerissuesrelatedtopopulationspecificity.
~
6
~ Research Design
Characteristics may include:
Independent variable I dependent variable
Random assignment
Manipulation of variable (IV)
Control groups
Research protocol
Blinding of investigator and subject (double-blind) or just one group
(single-blind) to eliminate biases
Issues:
Internal validity: potential for confounding factors to interfere wl
The relationship between IV /DV. (maturation I testing I attrition)
External validity: extent to which results can be generalized
outside of experimental situation. (random sampling I assignment
help).
Experimental Design: uses random assignment to at least two comparison groups
and controls for threats to internal validity. Strongest evidence for casual
relationship.
Think of a study that would represent I include an experimental design .....
Quasi-experimental Design: lacks random assignment and I or 2 comparison
groups.
Think of a study that would represent I include a quasi-experimental design ..... .
Single subject design (see handout I Figure)
Repeated baseline measures: subjects serve as their own controls.
7
Descriptive Statistics
~ f central tendency
/ ean - total score! # in sample
~ ~ if' Mode - most commonly occurring score
~ Median - value that represents the 50% in ranked distribution
~ Range - dispersion equal to difference between highest & lowest scores
Standard deviation - value used to describe the variance in the data
= square root of sum (X - mean)! N))
X - mean =deviation
Sum ofdeviations = sum (X-mean)/N
8
ParametricI NonparametricStatistics (seehandouts/Figures)
SampleDistribution
z-Scores:usedtodescribethelocationof anindividualscoreinadistributionandallows
forcomparisontootherdistributionsthathavebeentransformed
z=X-mean
SD
Allowsustosayindividualwas+/- oneortwostandarddevabovelbelowmean.
Confidenceintervals/significancelevels(alphalevel). Thesearesetapriori-
aheadoftimefortestof significance.
TypeIerror- falselyrejectthenullhypothesis.
TypeIIerror- falselyendorse/retainthenullhypothesis.
Parametric ... AO g..
Assumptions"
>( iYJ. Nomaldistribution,equaldistribution,interval/rationdata
T-test,ANOV A, etc


V\ Man-WhitneyUTest,SignTest,Wi1coxonMatch-PairsSigned-Rank 'r
Test,Kruskal-WallisTest
OAltJ 1.10 P ell sIT-l
I V---..
-
9
StatisticalInference
Whatcanbeinterpretedfromtheresults?
Haveassumptionsbeensupportedorviolated?
Nonnaldistribution
Intervallevelscore
Appropriatesamplesize
AnalysesofVariance
Comparisonof>2means(vsTtestisfor2means)
One-wayANOVA
Two-wayANOVA
Analysesoffrequencies
Frequencydistributiontable
Frequencydistributiongraph
Histogram
Bargraph
Polygon
Symmetricaldistribution
Positivelyskeweddistribution
Negativelyskeweddistribution
Quartiles
10
Correlation
Mathematicalmethodforassessingrelationshipamong2ormorevariables(DV
totheIV). Correlationcoefficientsrangefrom-1.0to 1.0 Shouldnotbeinterpretedas
cause& effect. Onlyassessmentofrelationshipamongvariablesundervarious
conditions.
Positivecorrelation
Negativecorrelation
Regression
Multipleregression
Assessingtherelationshipof>2independentvariablestothedependent
variable. AbletosayindependentvariableXlexplainsX%oftheDV,X2. .. "
explainsX%oftheDV,etc. TheDVisacontinousvariable. J
Logisticregression
Sameasmultipleregression,butwithlogisticregression,theDVisa
discretevariable.
Epidemology studyofthedistributionanddeterminantsof disease,injuryordysfunction
inhumanpopulations.(egcausalfactors,riskfactors). Helpfulwhencharacterizinga
diseaseI epidemic.
Incidence- quantifiesthenumberofnewcasesofadisorderordiseaseinthe
populationduringaspecifiedtimeperiod.
Prevalance- proportionreflectingthenumberof existingcasesofadisorder
relativetothetotalpopulationatagivenpointintime.
11
CHAPTER 2. THE ROLE OF THEORY IN CLINICAL RESEARCH 19
ting on how
aning to iso-
bservations. .
my separate
otorcompo-
Il'e also used
tance, a the--
U\d feedfor-
skill.
cumstances.
)lace during
ingleaming
II\ isoldnetic
rqueoutput
tostrength
ltcannotbe
'Ories,New-
nologywas
measure of
:plains how
)re( . how
n,
tohygiene.
idingmoti-
: a theoreti-
hypothesis
the theory.
a therapist
1 patient to
,hodoand
encetothe
Il\e theory.

! results of
monstrate
yexamin-
scientific
Figure 2.1 A model of scientific thought, showing the circular relationship between facts and
theory and the integration of inductive and deductive reasoning.
The basic building blocks of a theory are concepts. Concepts are abstractions that
allow us to classify natural phenomena and empirical observations. From birth we
begin to structure empirical impressions of the world around us in the form ofconcepts,
such as "mother," "father," "play," or "food," each of which implies a complex set of
recognitions and expectations. We develop these concepts within the context of
ence and feelings, so that they meet with our perception of reality. We supply labels to
sets ofbehaviors, objects, or processes that allow us to identify them and.discuss them.
We use concepts in professional communication in the same way. Even something
as basic as a "wheelchair" is a concept from which we distinguish chairs of different
types, styles, and functions. Almost every term we incorporate into our understanding
of human and environmental characteristics and behaviors is a conceptual entity. When
concepts can be assigned values, they can be manipulated as variables, so that their
tionships can be examined. In this context, variables become the concepts used for
building theories and planning research. Variables must be operationally defined, that
is, the methods for measuring or evaluating them must be dearly delineated.
Some concepts are observable and easily distinguishable from others. For instance,
a wheelchair will not be confused with an office chair. But other concepts are less tangi-
ble, and can be defined only by inference. Concepts that represent nonobservable
behaviors or events are called constructs. Constructs are invented names for variables
that cannot be seen directly, but are inferred by measuring relevant or correlated
iors that are observable. The construct of intelligence, for example, is one that we cannot
see, and yet we give it very dear meaning. We evaluate a person's intelligence by
observing his behavior, the things he says, what he "knows." We can also measure a
person's intelligence using standardized tests and use a number to signify intelligence.
I
A B A B
180
160
Z 140
20::
~ ~ 120
...1...1
LL.:=I 100
LL.0
0%
tn en 80
wl-
~ : : i 60
(,!)..J
W 40
o
20
1 I I I
i i i , i I i lii i I1 i i lii i I i i I i
2 4 6 8 10 12 14 16 18 20 22 24
DAYS
FIGURE6-1.ThebasicARABdesign.
8
A
3.0
Cl
w
~ 2.5
...Itn
~ w 2.0
...I
wi 1.5
(.)
zz
.c- 1.0
l-
tn
0.5
Q
2
4 6 8
10 12 14
DAYS
FIGURE6-2.TheABdesign.
(
,
MEDICALTRWSBF.SEd
A B A
FIGURE6-3.TheABAdesignandoverlap.
I
would be very difficult to defend against the possibility thatsomeunC(
variableaccountedforanychangeobservedinthedependentvariable.Ft
pie,a changeintheweathercoincidenttotheinitiationoftreatmentmil
causedtheobserved (orself.reported) changeinwalking.
TheABAdesign (Figure &-3) is strongerbecausethedependent\la
clearlyassociatedwiththeinitiationandwithdrawalof trealment.TheABA;
(Figure6-1),however,isstillstronger;asHersenandBarlowpointout,"Ul
naturalhistoryofthebehaviorunderstudyweretofollowidenticalfluctw
trends,itismostimprobablethatobservedchangesareduetoanyinfluen
somecorrelatedoruncontrolledvariable) otherthanthetreatmentvariab
systematicallychanged"\! (p.176).
There are two problems associated with withdrawal designs (ABA.
Thefirstpotentialconcernisthatsomebehaviors,bytheirnatureorbecau
subject'sresponse,donotreverttotheinitialbaselineoncegainshavebee
For example, in Figure 61 the second Phase A did notrevert to the
measurementbutthe trend did level off. Whether this is a realproblen
mustbeanaIyzedrationallyineachexperiment.Learning,especiallymob
ing,isanexampleofavariablethatisnotlikelytoreverttoanoriginalha
ashortperiodoftime.
Thesecondconce.rnis theethicsofwithdrawing3C "tmentthat81
beeffective.Onemayarguethataperiodofwithdrawr' legitimately
inthecauseof avoidingfalsepositiveinitialresultsthat..." .Jleadtothec
.. '!
2.3 I FREQUENCY DlS1RfBUTION GRAPHS
~
DEFINITION Fora bar graph, a verticalbarisdrawnaboveeachscore(orcategory)so
that
1. Theheightofthebarcorresponds tothefrequency.
2. Thereis a spaceseparatingeachbarfrom thenext
Abargntphisusedwhenthedataaremeasuredonanominaloranordinalscale.
FREQUENCY DISTRIBUTION
POLYGONS
Insteadofa histogram,manyresearchersprefertodisplayafrequencyd i ~ o n
usinga polygon.
DEFINITION Inafrequency distribution polygon, a singledotis drawnaboveeachscore
sothat
1. Thedotiscenteredabovethescore.
2. Theverticallocation (height) ofthedotcorrespondstothefrequency.
A continuouslineisthendrawnconnectingthese dots.Thegraphiscom-
pletedbydrawing a linedowntotheX-axis (zerofrequency) ata pointjust
beyondeachendoftherangeofscores.
FIGURE 2.3
Anexampleofafrequency distribution
histogram forgroupeddata.Thesameset
ofdataispresentedinagrouped fre...
quencydistributiontable andin ahisto-
gram.
X f
5
12-13 4
10-11 5
8-9 3
6-7 3
4-5 2
i:
4
0'----
2-3 4-5 6-7 8-9 10-11 12-1314-15
Scores
FIGURE 2A
Abargraph showing the distribution of
personalitytypesinasampleofcollege
students. Becausepersonality type is a
discrete variablemeasuredonanominal
scale. thegraphisdrawn with spacebe-
tween thebars.
20
~ 15
~
0'" 10
~
u.
A B c
Personall1ytype
/-""
\ f
\.,. ~ /
X
6.6
Followingaz-scoretransformation,the
X-axis is.relabeledinz-scoreunits.The
distancethatisequivalentto 1standard
deviationontheX-axis (a == 10pointsin
thisexample)conespands to 1pointon
thez-scorescale.
80 90 100 110 120
t..d I
z
-2
-1 0 +1 +2
JI.
6.2 I PROBABIUTVAND THE NORMALDlS1RIBUDON
FIGURE6.4
Thenormaldistributionfollowing a
,-scoretransformation.
2.28%

-2 -1 0 +1 +2
J1
198 CHAPTER 8 I INTRODUCTION TO HYPOtHESIS1'ES11NG
FIGURE8.4
The locationsofthecriticalregion
boundariesfor three differentlevelsof
.significance: Cl = .05, Cl = .01, and
Cl = .001.
-1.96 0 1.96
-2.68 \ ) 2.68

----a=.001----
1
/\
IATION OF CLINICAL PRACTICE DESCRIPTIVE RESEARCH
TABLE 5-2 DESCRIPTIVE STATISTICS
101
research. As indicated in
p in the research process.
with the literature to find
problem, it takes on some
etative t'eview of literature
.luating the quality, or the
her colleagues
60
"decided
udies. In this way, conclu-
drawn from the best meth-
.1 anciskin disorders has been
:linical trials (RCTs) involving
ne were generally of a better
:0 clear relationship could be
l the efficacy of laser therapy,
In general, the methodological
:onsequently, no definite con-
lerapy for skin disorders. The
1&5 seems, on average, to be
re specifically, for rheumatoid
cial pain, laser therapy seems
rther RCTs,;( , "ling the most
rto enable tJt.... .;nefits of laser
Level of Central Spread or
Measnrement . Tendency Variability Other
Nominal Mode Range Frequency counts and
percentages in
categories
Ordinal Median Range Frequency counts and
percentages at levels
(percentile)
Range; variance; Frequency counts and
~ Mean
standard . percentages at levels
deviation
......\
.. . ~
list, the range would be 0 to 3. There are two Os, seven Is, two 2s, and five
3s; therefore, the mode would be 1 because there are more Is than any
other number. The mode is a measure of central tendency for nominal
data. A sample may be bimodal if two frequencies are equal in number.
The frequency counts can be transformed into percentages. For example,
the Os make up 12.5 percent (1/8) of the total list; percentages should,
however, be used with caution when the sample (n) is small.
The median is the statistic of central tendency of a set of ordinal data;
it is the middle of the range of recorded measurements, listed in order. For
example, take all of th( '-jngs of spontaneous activity in the supine posi-
tion from the table in -:.. .:! article by Carter and Campbell. Sixteen ratinszs
~
!
--- - -- -. .... _.. ....-_............,.... ...""
he practicingclinicians.
)anything done for the
ts physiological orpsy-
)gical effectsonpatients;
cal effectsofthemodal-
underwhatcircum-
ithologydoes it change
age, the frequency, the
Iftreatment?Whatisthe
lrticular case, and what
Whatarethesideeffects
1, orintensity?Giventhe
do the answers change
eyinteractandinfiuence
youwereworkingand
latyouwereapplyingto
Ilyasanswers?Thepoint
Iberofclinicalquestions
)esigns using sequential
answer these questions
dgn has been little used
ture.ReadLightandcol-
Jnentsof thedesign dis-
llt.
23
"
ldation and the manner of
datedwitheverystatistical
test is valid under certain
(
tests.
25
Table8-3outlinesaselectnumberotClasSIC statiStiCal
defined ,above. Afrequently used statistical test is listed for each mea-
surement level and for eachsampling model given in Table 8-1.As dis-
cussedin Chapter3,parametri<:statistics,r.efeitotl?-ose'testsappropriate
fo-:jnettIcda. tests,espeCially andFtest,rnat<e

normallydistributed
asonthe'farnlliarbell testsmakefewer
andsoaremotewidelyapplicabl*26Nonparametrictestsmaybeapplied
tometric data,iftheassu,mptions'of the parametric testsarein question,
butit is nofappropriatetoapplymetricteststonominalorordinaldata.
More complexdesigns for k datasets,factorial designs, andmultivariate
designs will beillustratedinChapter9.
TABLE 8-3 STATISTICALTOOLSAPPROPRIATETO EACH
BASIC'RESEARCH DESIGNAND LEVEL OF MEASUREMENT
Level ofMeasurement
TypeofDesign Nominal Ordinal Metric
1. Onesample 'Ohi"Sqriarelx2) ,i One-sampleruns Hest,related
2. Two related,own McNemar Sigri test Hest,related
control Wilcoxop.
3. k related,own CochranQ FriedMaa Ftest,two-way
control AOV
4. Two independent X2, Median test t-test,indepen-
Mann-Whitney dent
5. Two related, McNemar Sign test t-test,related
matched WilkOXGD
6. k independent X2, Kruskal-Wallis Ftest,one-way
AOV
7. k related,matched CochranQ FriedmaI1!$! Ftest,two-way
AOV
AOV =analysisofvariance.
Sources:Datafrom Campbell.DT andStanley,JC: Experimental andQuasi-Experimental
Designs for Research,Rand-McNally, Chicago, Daniel,WW: AppliedNonpara-
metricStatistics.HoughtonMifffin, Boston, 1978.

I
1rO-I : ( ..,Scri"....
I
PubMed
PubMed, created by the National Library of Medicine, provides access to: MEDLINE and PreMEDLINE.
Compose Your Search: Define your search clearly; consider all important words, concepts, and synonyms.
Su bj ect Searching: PubMed may be searched by entering keywords or phrases into the text box. You may
enter one or more terms and press the enter key or click Go. PubMed will search multiple words as a phrase if it
recognizes the terms. Otherwise, PubMed will search the words separately and combine with AND. PubMed will
also automatically try to map your term to a MeSH heading.
Use "OR" or "AND"to combine your concepts. OR broadens a search (e.g. Breast Neoplasms OR
Breast Diseases) AND narrows a search (e.g. Vitamin C AND common cold)
Database: MeSH terminology provides a consistent way to retrieve information that may use different
terminology for the same concepts. One way to select MeSH terms is to use the MeSH Database link, found on the
blue sidebar, to browse for appropriate terms. Enter the term in the text box and click GO.
View the list ofterms that appears and click on the most appropriate term.
Scroll down the page to see everywhere the term appears in the MeSH "Trees," a hierarchial subject category display
ofMesh terms. For example, ifyou browse this listfor eating disorders, you can select.from a host ofterms including
Anorexia Nervosa, Bulimia and Pica.
To further refine your search, limit your search by using Subheadings, Major Topic or Explode.
Do Not Explode. To decrease retrieval, click on the Do Not Explode checkbox to not include MeSH terms
found below your term in the MeSH tree.
Major Topic. Click on Major Topic to limit search to citations in which your term is the main topic.
SubHeadings. Click in any of the subheading checkboxes to further narrow your search term.
Chemically Induced Complications Diagnosis
Drug Therapy Epidemiology
Click "Send To" when done narrowing the term. Continue to add terms to your search using the MeSH
database. When complete, click Search PubMed
Title Word Search: Limiting a search to a word found in the title of articles is often a good way to reduce large search
results. You can do this by entering the term in the text box. Then click on Limits, pull down the All Fields menu, and
select Title Word. You may also limit to title by entering the term followed by [tiJ.
Author Search: The Index provides a list of author names and variations. To use, click on Preview/Index found
below the textbos. Scroll down the page to Add Term(s) to Query or View Index. Using the All Fields pull-down
menu select Author Name. In the text box, enter the author's last name and initials ifknown. Click on INDEX.
Selec:t the name from the list of options that appears. Multiples names may be selected by holding down the CTRL key.
Multtple names will be ORed together. Names can be added to tlle search box by clicking AND (to combine) OR (to
find either/or), or NOT (to exclude). '
Journal Name: To retrieve articles in specific journals, use the Index. Follow the Index instructions for searching by
author (above), and select Journal Name from the All Fields pull down menu
your Terms. PubMed will provide the results of your search in the History Table. PubMed will also
proVIde a nUl11ber for each ofyour search terms or sets. You must use the # symbol before the search numbers and
search operators (AND, OR, NOT) must be capitalized
Sgarch
#3 E:carch#l AND #2
#2 St:i:lIl;l! u.n:rl aglmLs
#1 :search diabetes mellitus llon-insuJin dependent/droll therapy
Limit: searchfurtherbyclickingLimits. ThiswillallowyoutolimityoursearchbyJournalSubsets,Age
Groups,PublicahonTypes,andotherfeatures specifictothedatabases.
DisplayingSearchResults:PubMeddisplayssearchresultsinbatches;thedefaultis 20records perpage. To
change .default, theShowpull-down to thenumberofrecordsdisplayedonasinglepageupto
500. TIns IS aconvementwaytobrowsetheenhredisplaytochooseselectedarticlesto saveorprint.
PubMedcitationsinitiallydisplayin Summary (Author,Title,Journal,PubMedID)format. Documentscanbeviewed
inotherformatsincludingBrief(author,PubMedID),Abstract(Author, Title,Journal,PubMedID,Abstract),
Citation(Author,Title,Journal, PubMedID,Abstract,MeSHHeadings), and MEDLINE(forimportingintocitation
software Author,Title,Journal,PubMedID,Abstract,MeSHHeadings).
SavingResultstoDisk: Resultsmaybesavedtoadiskorharddrive. OncesearchresultshavebeenDisplayedin
thedesiredformat(Summary,Brief:Abstract,Citation,orMEDLINE),select"File"fromthepulldownmenunext
to "SendTo". NextclickSendTo. A SaveAsDialogBoxwillappear. TitlethenameofyoursearchintheFile
NameBox. ClickSave. ThesefilescanbeopenedinWord,WordPerfect,orNotepad.
SavingtoCitationSoftware (ReferenceManager,Endnote)
ToimportresultsintoCitationSoftware,youmustdisplayyourresults inMEDLINEfOID1at Select"File"fromthe
pulldownmenunextto"SendTo". NextclickSendTo. TitlethenameofyoursearchintheFileNameBox. Click
Save.
EmailResults: OncesearchresultshavebeenDisplayedinthedesiredformat(Summary,Brief:Abstract,
Citatiou,orMEDLINE),clickonthepull-downmenunexttoSendTo. Select''E-Mail''andclickonSendTo. In
thetextboxnextto"E-mail"enteryouremailaddress. ClickMail.
ClipBoard:TheClipboardallowsyoutogroupselectedcitati,onsfromoneormoresearches.Amaximumof500
citationscanbeplacedintheClipboard.Clipboarditems.willbe.1ostafteronehourof inactivity.Toadditemstothe
clipboard,clickonthecheckboxtotheleftofthecitation'andSelect"Clipboard"fromthepulldownmenunextto
"SendTo". NextclickSendTo.OnceacitationhasbeenaddedtotlleClipboard,therecordnumbercolorwill
changetogreen. OnceyouhaveaddeditemstotheClipboard,youcanclickonClipboardfrom theFeaturesbarto
viewyourselections. .
SavingSearchStrategies& TheCubby:TheCubbyisusedtostoresearchstrategies. To register,clickonCubby
fromthePubMedsidebar. Nextclick"IWanttoRegisterforCubby." FillintheformprovidedandclickRegister.
ToStoreaSearchintheCubby: Run thesearch. ClickonCubbyfromthesidebar. (Loginifyouhaven'talready
doneso.) Inthemiddleofthescreenatextboxunder"CubbySearchName" willappearWitllyoursearchstrategy.
Youcaneditthematerialinthetextboxtoameaningfulnameforthesearch.
ToRunaSavedSearchedStrategyintheCubby:ClickonCubbyfromthePubMedside. Clickonthecheckbox
nexttothesearchyouwanttorunandclick"What'snewforselected."
LinkOuts:UICsubscribestonumerousonlinejournalswhichmaybeaccessedthroughPubMed'sLinkOuts. Once
yoursearchiscomplete,clickonAbstractsintheDisplaypulldownmenu. Thiswillshowtlleabstractsof thearticles
andavailablelinkouts. IfyouhaveenteredPubMedthroughaUICLibrarywebpage,itemsownedbyUICwillbe
accompaniedbyanIconindicatingUICholdingsinformation.
Indicatesthatyoumaylink directlyto thejournalarticlefromPub Med.
.Qf___:lW: JJS
. Print holdings' IndicatesthatUICholdsthisj0U111a1 inPrint. ConsultlITCCATformorein-depthholdings
for UIC users infunnation. .
Note: NotallofUIC'sonlinejoumalsareaccessibledirectlythroughPubMed. In addition,notall ofthelinkoutsin
PubMedareaccessibletoUICusers.
ForFutherAssistance,pleasecontactReferenceStaff.
SDG09/03
o V I D CINAHL WEBONLINESEARCHING
I
I
Compose Your Search
Defineyoursearchclearlyand concisely; considerall impOltantwords, concepts,andsynonyms.
Enteryourkeywordorphrase(e.g. eatingdisorders).
CINAHLwill automaticallytryto mapyoursearchtennto asimilarorrelatedtenn(s). Ensure the MapTerm
Checkboxis checked.
Ifanyof thetermsCINAHLlistsappearappropriate,you canclickontheSelectcheckboxforanyor
all oftheterms.
Ifyou choose morethanonetenD, usethepull-downmenuatlhetopofthepageto
combinethemwith"OR",or"AND".
ORbroadensasearch(e.g. friendships orpeerrelationships)
AND nalTOWS asearch(e.g. EatingHabitsandBodyhnage)
" ,
Explode. Toincreaseretrieval, OvidautomaticallyExplodesselectedtennstoincludeanynalTowertermsor
concepts. (e.g.Exploding"EatingDisorders"willincludethenalTowertenns: Anorexia,Bulimia,andPica.)
Focus. ClickontheFocuscheckboxto limityoursearchtocitationswhereyourtermisthemaintopic.
IfyouonlyselectedonetennfromthelistofSubjectHeadingsprovided,CINAHLwillpromptyouto
choosesubheadingsfor yourtenn.

Subheadingsfor: Eatingdisorders
OChemicallyInduced(3) oDiagnosis(46) OEducation(7)
OComplications (23) ODmgTherapy(8) OEpidemiology(32)
Youcanclickinanyofthesubheadingcheckboxesto :fi.uthernarrowyoursearchtermORclick
Continue(nosubheadingswillbeselected).
If CINAHLwasunabletoprovidealistof relatedtennsforyoursubject,trytothinkofotherrelated
tennsandrepeatthe abovestepsORclickofftheMapTerm Checkbox.
CINAHLwillprovidetheresults ofyoursearchinthetableentitledSearchHistOJ)I. CINAHLwill also
provideamunberforeachofyoursearchtenTIS orsets.
EntertllenextkeywordorphrasethatyouwouldlikeCINAHLtosearch.Thenamesofauthors,journals,or
titlesmaybesearchedbyclickingontheappropriate icon.
Enterthefirstfewlettersofafulljournalname; donotuseabbreviations(e.g. BehforBehavior
Journ"J Modification). ClickPerformSearch.CINAHLwillprovidealistofJournalTitles. Clickonthebox
nexttothetitleyouaresearching.
Enterawordorphrasetobesearchedinthetitle (e.g. reportingof medication).
Entertlle Author'slastname,aspace,and firstinitialifknown (e.g. WinslerA). ClickonPerform
Search.CINAHLwillprovidealistof Authors., Clickontheboxesnexttotheauthors'namesthat
AuttJor appeartobetheonesyouaresearching.

Combine your Terms.

Oombine
CombineyoursearchtermsusingtheconnectorsANDand OR. Usethenumbers that CINAHL
providedintheSearchHistorytable.
# Seal1:hHistory ReSw.ts Display
Demetlual 1046 Display
2 Conceptanalysis/ 287
3 3 Display
_ a Limit
r ~ LimityoursearchfurtherbyclickingontheLimiticon. Thiswill allowyou tolimityoursearchby
Limit Language,PublicationYear, PublicationTypeandotherfeaturesspecifictothe databases.
Examine Your Search Results
CINAHLautomaticallydisplays theresults ofthelastsearchatthebottomof thescreen. Usethe scrollbarto
examinetheresultsof yoursearch. Youcan alsoclickonDisplaytoviewtheresults.
ClickonAbstractorCompleteReferencetoviewtherecordindetail. SomecitationsinCINAHLalso
contain thefulltextarticle. Ifavailable,you can clickonFullTexttoseetheentiremticle.
Print, Download or E-mail Full Text
Clickon"OvidFullText"whereavailabletolinktoarticlesavailablefull-textthroughOvid. Scrolltothe
OutputBoxfoundnearthe topontherightsideof thepage.
ToPrinttheFullTextof anarticle,clickonthePrintPreviewIcon. SelectaGraphicSizeandclickContinue.
ThenclickonthePrinticonfoundonyourWebbrowser. ClickFullTextPDFtoobtainarticleinPDFformat
whenavailable.
ToE-mailtheFullTextarticle,clickontheE-mailArticleTexticon.TypeintheE-mail addressandclickon
SendE-mail. (This willnotincludegraphics).
".;
ToDownloadtheFullTextarticle,clickon SaveArticleText. ClickContinue.IntheSaveIn: pull-downmenu
clickonthedriveinwhichyouwanttosaveyourfile(AorC). TypeatitleforyoursearchresultsintheFile
Namebox. ClickSave.(This will notincludegraphics).
Print, Download or E-mail Your Citation Results
Markanycitationsthatyouwishtoprint,downloadorE-mail. Ifyouarelookingatthecitationsthroughthe
TitlesDisplay,clickontheboxtothe leftof thetitle. Whenyoul ~ a v e completedviewingyourcitations, scroll
to theCitationManageratthebottomofthepage..(ITviewingthecitationsin detail, clickon theboxatthetop
of thepage.)
ToPrintthecitationsyouhavemarked,c1ick.'c)ll'theDisplayiconfoundundertheActioncolumn. Thenclick
onthePlinticonfoundonyourWebbrowser. .
ToE-mailthecitationsyouhavemarked, clickontheE-mailiconfound undertheActioncolumn, Typeinthe
E-mailaddressand clickon SendE-mail.
ToDownloadthemarkedcitations,clickontheSaveicon. Inthe"SaveIn:"pull-dowlll11enuclickonthedlive
in whichyouwanttosaveyourfile (AorC). TypeatitlefortheresultsintheFileNamebox. ClickSave.
Note:only200 ofUICsonlinejournalsareavailabledirectlythroughOvid. ConsultUICCATforUIC'sprint
journalholdingsandonlinejoumalholdings(5000+onlinejournals).
ForFurtherAssistance,PleaseSeeReferenceStaff. SDG(07/02)
ComparisonofFunctionalStatusToolsUsedin
Post-AcuteCare
AlanM.Jette,Ph.D.,StephenM.Haley,PhD.,andPengshengNi, M.P.H, M.D.
There is agrowing health policy mandate
for comprehensive monitoring offunctional
outcomes across post-acute care (PAC) set-
tings.This article presents an empirical
comparison of four functional outcome
instruments used in PAC with respect to
their content, breadth of coverage, and
measurement precision. Results illustrate
limitations in the range ofcontent, breadth
ofcoverage, and measurement precision in
each outcome instrument. None appears
well-equipped to meet the challenge ofmon-
itoring quality and functional outcomes
across settings where PAC is provided.
Limitations in existing assessment method-
ology has stimulated the development of
more comprehensive outcome assessment
systems specifically for monitoring the qual-
ity ofservices provided to PAC patients.
INIRODUCl10N
A fundamental barrier to fulfilling the
emerging health policy mandate in the
United States for monitoring the quality
and outcomes of PAC is the absence of
standardized, patient-centered outcome
data that can provide policy officials and
managerswithoutcomedataacrossdiffer-
ent diagnostic categories, over time, and
across different settings where PAC ser-
vicesareprovided(WtIkersonandJohnston,
Theauthorsarewith Boston University. Theresearch inthis
articlewassupporlEdbytheNationa1lnstituteonAgingunder
Grant Number 5P50AGl1669 and the National Institute on
Disability and Rehabilitation Research under Grant Number
HI338990005. Theviewsexpressedinthisarticlearethoseof
theauthorsanddonotnecessarilyreflecttheviews ofBoston
University,tbeNationa1InstituteonAging,theNatiomillnstitute
onDisabilityand RehabHitation,ortheCentersforMedicare&
MedicaidServices.
1997). Recently, the National Committee
on Vital and Health Statistics (NeVHS)
(2002) made recommendations on the
potentialfor standardizing data collection
and reportingfor the purposes of quality
assurance as well as for setting future
researchandhealthpolicyprioritiesinthe
U.S. TheNCVHS (2002)wasunanimousin
stressingtwo majorgoals: "... to putfunc-
tionalstatussolidlyontheradarscreensof
those responsible for health information
policy,andtobeginlaying.thegroundwork
for greater use offunctional status infor-
mationin andbeyondclinicalcare... "The
NCVHS project used the term functional
statusverybroadlytocoverboththeindi-
vidual's ability to carry out activities of
daily living (ADLs) and the individual's
participation in various life situations and
society.
Within PAC, functional outcome instru-
mentshavebeendevelopedandarewidely
usedforvariousapplicationsandforusein
specific settings. Examples include the
functionalindependencemeasure (FJMTM)
foracutemedicalrehabilitation (Guidefor
the Uniform Data Set for Medical
Rehabilitation, 1997; Hamilton, Granger,
andSherwin,1987),theminimumdataset
(MDS) for skilled nursing and subacute
rehabilitation programs (Morris, Murphy,
and Nonemaker, 1995), the Outcome and
Assessment Information Set for Home
Health Care (OASIS) (Shaughnessy,
Crisler,andSchlenker,1997)andtheShort
Form-36 (SF-36) for ambulatory care pro-
grams(WareandKosiniski,2oo1). If one
looks carefully at the content of these
HEALmCARE FlNANCINGREVIEW/Spring 13
instruments, it becomes apparent that sub-
stantial variations exist in item definitions,
scoring, metrics, and content coverage,
resulting in fragmentation in outcome data
available for use across different PAC set-
tings (Haley and Langmuir, 2000).
Differences in conceptual frameworks
used to construct each instrument, the
inability to translate scores from one
instrument to another, and the lack of out-
come coverage and precision to detect
meaningful functional Ghanges across set-
tings, severely limit the field's ability to
measure and analyze recovery through the
period of PAC service provision. If the
PAC field is to achieve the goal of compre-
hensive functional outcome assessment
and quality monitoring for different patient
diagnostic groups across different PAC
settings, efforts are needed to develop
functional outcome assessments that are
applicable across a continuum of post-
acute services and settings.
To our knowledge, no studies exist that
have directly compared the content and
operating characteristics of functional out-
come instruments commonly used in PAC
to examine their relative merits for moni-
toring outcomes across care settings. In
this article, therefore, we report the results
of a direct empirical comparison of the
FIMTM, OASIS, MDS, and the physical
function scale (PF-10) of the SF-36, focus-
ing on three aspects of each: instrument
content, range of coverage, and measure-
ment precision. The objective of this com-
parative analysis is to evaluate the com-
monly held assumption that there exist
fundamental deficiencies in the current
armamentarium of setting-specific out-
come instruments that prevents their
applicability for more comprehensive patient-
centered functional outcome assessment
across diagnoses, over time, and across dif-
ferent settings where PAC is provided. In
response to identified deficiencies in exist-
ing instruments, we also discuss the poten-
tial utility of contemporary measurement
techniques, such as item response theory
(IR1) methods and computerized adaptive
technology (CAl), to yield functional out-
come instruments better suited for out-
come monitoring across PAC settings.
METIlODS
Subjects
These analyses use data from a sample
of 485PAC patients drawn from six health
provider networks in the greater Boston
area. All patients enrolled in the study
were recruited by study coordinators with-
.in their own health care facility and com-
pleted infonned consent prior to participat-
ing in the study. Patients were recruited
from inpatient (199 from acute inpatient
rehabilitation and 90 from transitional care
units); and community settings (90 from
ambulatory services; and 106 from home
care). Eligibility criteria included: (1)
adults, age 18 or over, (2) recipients of
skilled rehabilitation services (physical,
occupational, or speech therapy), and (3)
English speaking. The sample was strati-
fied by impairment group to include
approximately an equal number of subjects
within three major patient groups: (1) 33.2
percent neurological (e.g., stroke, multiple
sclerosis, Parkinson's disease, brain injury,
spinal cord injury, neuropathy); (2) 28.4
percent musculoskeletal (e.g., fractures,
joint replacements, orthopedic surgery,
joint or muscular pain); and (3) 38.4 per-
cent medically complex (e.g., debility
resulting from illness, cardiopulmonary
conditions, or post-surgical recovery). To
assure good representation of levels of
functional severity, the sampling design
was stratified to yield a wide distribution of
subjects representing three distinct severi-
ty levels: slight (35.9 percent), moderate

, ,
HEALTII CARE FINANCING REVIEW/Spring 2003/VOfume24,Nmnber3 14
I
(44.1 percent), and severe (20.0 percent)
based on scores from an adapted modified
Rankin scale (vanSwieten et al., 1988).
The sample reflects the racial and ethnic
distribution of the greater Boston metro-
politan population. The sample contained a
wide age range (19-100 years; mean age =
62.7 years.) The majority of the subjects
were female (58.8 percent), white (81.6
percent), and non-married (61.1 percent).
More than one-half (51.3 percent) had
beyond a high school education. The wide
range of onset from initial injuryI illness
(2.0-3.9 years) characterizes different stages
of recovery within both inpatient and com-
munity settings. The SF-8 Health Survey
(Ware et al., 1999) data indicated that phys-
ical functioning of the overall sample
(mean=40.3, standard deviation (s.d.)=9.9)
was below the D.S. population norms
(mean=50, s.d.=10), although mental func-
tioning (mean=50.2, s.d.=:10.3) was consis-
tent with D.S. population norms (mean=50,
s.d.=10).
Data Collection Procedures
Our overall data collection strategy was
to assess items from existing functional
outcome tools used in PAC so that they
could be combined for analysis into one
common scale. Due to practical data col-
lection considerations and potentially high
patient response burden, we did not admin-
ister items from all four instruments to all
485 subjects. Rather, we linked items
together by administering to the entire
sample a core set of 58 activity items applic-
able for patients in both home and commu-
nity service settings. This method has
been referred to as a common-item test
equating design, in which a core set of
items serve as a scale anchor from which
unbiased parameters can be estimated on
items with missing data (McHorney and
Cohen, 2000).
We collected activity items, when avail-
able, from standardized instruments admin-
istered and recorded in the medical record.
These included the 1B-item FJMTM for per-
sons in inpatient rehabilitation settings
(N=108), 19 MDS items (physical function-
ing and selected cognitive items) for per-
sons in skilled nursing home settings
(N=91), and 19 ADL/individualADL items
from the OASIS or persons receiving home
care (N=103). Since there were no consis-
tent data available from charts in the out-
patient setting, we administered the 10
physical functioning items of the SF-36 to
individuals receiving outpatient services
(N=82).
We applied specific rules for handling
missing item data within two of the stan-
dardized instruments. In accordance with
FIMTM scoring rules, items that were not
administered are scored as "total depen-
dence." The MDS items have a response
option of "activity did not occur." 'We con-
verted these codes to the lowest score for
that item, namely "total dependence." We
reasoned that the most likely explanation
for the activity not occurring was that the
item could not be performed, an assump-
tion that others have made when compar-
ing functional instruments. (Buchanan et
al., 2002). Missing data for the PF-I0 and
OASIS were not systematically recoded.
The 58 core activity items collected on all
485 subjects were used as scale anchors. In
order to compare items from instruments ,
across PAC settings, a common link was
needed to provide a stable functional base
for comparison purposes. The common-
item test equating design was used so that
every subject had a similar core set of
items. Rasch (1980) models (Wright and
Masters, 1982) can be conducted with miss-
ing data if a core set of items is used to link
the setting-specific instruments into a
common scale. Core items included: 15
physical functioning, 14 self-care, 12 daily
HEALm CARE FINANCING REVIEW/Spring 2003/"'lume 24, Number 3 15
routine, 11 communication, and 6 interper-
sonal interaction items. Activity questions
asked, "How much difficulty do you cur-
rently have (without help from another per-
son or device) with the following activi-
ties...?" A polytomous response choice
included "not at all," "a little," "somewhat,"
"a 101," and "can not do." We framed the
activity questions in a general fashion with-
out specific attribution to health, medical
conditions, or disabling factors. For some
individuals in the community settings, we
also collected 52 additional functional items
(N==I96) and added those to the common
linking solution, however the results of
these items are not reported here. None of
the subjects had difficulty in completing the
core set of 58 items.
Data on items from three of the existing
instruments (e.g., FIMTM. OASIS. and
MDS) were recorded from retrospective
chart review. Data on the PF-I0 items (out-
patient programs) and core and communi-
ty items were collected via subject or proxy
interviews. Each interview (approximate-
ly 45-60 minutes) was conducted in a quiet
atmosphere in an inpatient setting. an out-
patient facility. or participanfs home. The
ability of a subject to take part in the inter-
view self-report was assessed by a treating
clinician who determined if the respondent
could: (1) understand the interview ques-
. tions, (2) sustain attention during an inter-
view, and (3) reliably respond to the ques-
tions. If the answer to any of these screen-
ing questions was no, then the interview
was completed by either a clinician or fam-
ily member proxy participant A proxy par-
ticipant completed less than 3 percent of
the interviews.
Interviews that included information on
core items were timed to coincide with
administration of standard outcome instru-
ments. For example. FIMTM is adminis-
tered in most facilities within 3 days of
admission and prior to discharge in the
inpatient rehabilitation setting. Thus, the
subject interview was arranged to take
place within 3 days of admission or dis-
charge such that FIMTM information was
collected close to the time of the interview.
likewise, interviews of subjects in transi-
tional care units were scheduled to coin-
cide with the MDS assessment Subjects
receiving home care therapy were inter-
, viewed either near to admission or dis-
charge such that the OASIS assessment
was performed in close proximity to the
subject interview. The interview was fully
scripted. with standard instructions and an
answer card to help subjects identify the
desired response choice. The order of pre-
sentation of test sections was randomly
assigned to mitigate the possibility of large
portions of missing data in anyone section
due to respondent fatigue.
Interviewers. who were experienced
rehabilitation clinicians (mean years of
clinical experience==5.7 years). received
training and quality assurance from: (1) an
initial 3-hour training session. (2) a proto-
col manual, (3) supervision by the research
project staff on all first interviews. and (4)
acceptable completion of a procedural
checklist and inter-rater reliability on a
subsequent interview. A project staff mem-
ber accompanied interviewers on approxi-
mately every 10th interview to check on
adherence to study protocol and to assure
overall quality control. All subjects.
regardless of setting, were administered
the SF-8 items to establish a normative
description of the sample.
Analyses
Analyses were conducted in three
stages. First, we conducted one-parameter
Rasch (1980) partial credit analyses for the
entire item pool (instruments and
core!community items) to develop an
overall functional ability scale (Wright and
HEAL1H CARE FINANCING ~ p r l n g 2003/VoIume24,Number3 16
Masters, 1982). The Rasch partial credit
model allows for different number of
response categories across items. For
example, the FIMTM scale has a seven-point
rating scale, while the OASIS incorporates
multiple response categories that differ per
item. This overall scale was used to esti-
mate parameters for all items in each of the
four comparison instruments. We used a
method of concurrent calibration, which
involves estimating item and ability para-
meters in the, overall subject and item pool
simultaneously. This treats items not
taken by subjects as missing, as the Rasch
program uses information available to esti-
mate person and item parameters. Our
goal was to develop a general scale for
functional abilities, and specifically to see
the location of item content across the four
existing instruments. We calculated inter-
nal consistency estimates-Cronbach
alpha (1951)-for each of the individual
instruments and for the overall functional
scale to determine the levels of consisten-
cy of the items within the overall function-
al scale. We also calculated goodness of fit
tests for all items with the overall function-
al scale using the standardized infit statis-
tic (+/-2.0) to determine the number of
items with poor fit within the overall scale.
Second, we examined the relative
amount and range of activity content cov-
ered by each of the four existing instru-
ments and made comparisons across
instruments. To do this, we used the
extreme (highest and lowest) item thresh-
old estimate for each instrument A
threshold represents the point of separa-
tion between adjacent item response cate-
gories for each item. The Rasch partial
credit model estimates thresholds for each
item, and the minimum and maximum
threshold values per instrument were used
to establish the range of content within the
overall functional ability scale.
Third, we calculated item and test infor-
mation functions (Lord, 1980; Dodd and
Koch, 1987; Murnki, 1993) to estimate the
location of optimal measurement precision
of each instrument The item information
function is an index of the degree and loca-
tion of information that a particular func-
tional item provides for estimating a score
along a functional ability scale. Item infor-
mation functions are related to the location
and shape of the item characteristic curve
(lCC), which describe the probabilities of
responding to particular response options
of an item. The steeper the slope of the
ICC, the more information about function-
al ability provided by an item in the scale,
thus the greater level of item discrimina-
tion and precision associated with esti-
mates of the score at that point along the
scale. ICCs that have a broad range of cov-
erage along the scale provide information
functions across a wide range of the scale.
Therefore, the information function is the
relationship of the amount of information
of an item at a particular scale level and is
described by the ratio of the slope of the
ICC and the expected measurement error.
Test information functions were calculated
by summing the item information func-
tions to obtain an estimate of the measure-
ment precision of the entire test at differ-
ent levels of the functional ability continu-
um. We compared the test information
function with the ability levels of the sam-
ple in each major setting (inpatient, com-
munity). We calculated a summary score
converted to 0-100 metric for each person
based on the overall item pool.
RESULTS
The internal consistency values of the
four functional ability instruments were as
follows: MDS=O.97, OASIS=0.99, FJMfM-
0.99. and the PF-10=0.99. The internal
HEALm CARE FINANCING REVIEW/Spring 2003/VoIutne 24. Number 3 17
Figure1
RaschModelComparisonof the FourPost-AcuteCareInstruments
. FIM 1-1--/------+---/------+---/-----1----1---+----1,------1
PF-10I
I 1.#
e
::11
i
.. /' ~
~
/' V /'\
MOO I
I I
/0
it'/
qf
OASISI
I
20 25 30 35 40 45 50 55 60 65 70
,.
ItemScale
Less Ability MoreAbility
Functional Ability
NOTES:LocationsarederMI<fasestimatesofthe 8'Yerageitemdilficullyparametergeneratedfor eachinsIntment. FIM1M is
functionalindependentmeasum.PF-10isphysicalfunctionindex(10items). !lADSis minimumdataset. OASISis
StandardizedOutcomeandAssessmentInfDrmalionSetforHomeHeahh.UEisupperexfmmily.LEislowerextremity.
SOURCES:(Hamilton,Granger,andSherwin,1987;Morris, Murphy,andNonemaker,1995;wamandKosinski,2001;
Shaughnessy,Ctisler,andSchIenker,1997.)
consistency of the items within the full
functional ability scale was O ~ 8 5 Only five
of the items within the existing instru-
ments (7.2 percent) exceeded the good-
ness of :6t values. Thus. we felt it was
acceptable to combine the items from each
of the four functional outcome instruments
into an overall functional ability scale for
the purposes of directly comparing their
range of functional content, breadth of cov-
erage, and measurement precision.
FIgUre 1 illustrates and compares the
location of items in each of the four func-
tional outcome instruments along the
broad content dimension of functional abil-
ity. These locations are derived as esti-
mates of the average functional ability para-
meter generated for items in each instru-
ment included inthe analysis. Across all
four instruments it can be seen that cogni-
tive, communication, and bowel and blad-
der continence function items achieved the
lowest functional ability estimates, indicat-
ing that those items were usually less diffi-
cult for persons inthe sample to perform
compared with other items contained in
these instruments. Ingeneral, the PF-IO
scale contained items with the highest item
REALmCAREFtNANCINGREVIEW/SpriDg2003/VolumeU. Nlllllher3 18
Figure2
ComparisonofActualRangesCoveredbytheFourPost-AcuteCareInstruments
100
99
MDS OASIS
FIMTM PF-10
Instrument
NOTES: Content coverage is calculated by the high and low step estimates for each item's response cate-
gory. MDS is minimum data set. OASIS is Standardized Outcome and Assessment Information Set for
Home Health. FIMTM is functional independent measure. PF-10 is physical function index (10 items).
SOURCES: (Hamilton, Granger, and Sherwin, 1987; Morris, Murphy, and Nonemaker, 1995; Ware and
Kosinski, 2001; Shaughnessy, Crisler, and Schlenker, 1997.)
100
90
80
70
60
C
III
50 u
..
l.
40
30
20
10
0
0 0.5
86
25
50
functional ability calibrations compared
with the other three. A substantial number
of items in the FIMTM, OASIS, and MDS
instruments achieved average functional
ability calibrations in the mid-point of the
functional ability continuum. Figure 1 also
reveals the substantial overlap of item con-
tent across these four instruments.
The actual range of functional ability that
is covered by the items contained in each of
the four functional outcome instruments is
presented in Figure 2. The content cover-
age is calculated by the high and low step
estimates for each item's response cate-
gories instead of average estimates, which
were the basis of the information presented
in Figure 1. Consistent with the general
spread of the functional ability parameters
illustrated in Figure 1, the range of cover-
age shown in Figure 2 appears greatest for
both the MDS and the OASIS instruments
as compared with either the FIMTM or PF-10
scales. Because of the high step estimate
for one of the transportation items in the
OASIS, the actual range of coverage of the
OASIS is nearly the entire range of all four
instruments.
Figure 3 displays the measurement pre-
cision of each instrument as depicted by
the test information function curves.
These curves are superimposed with the
average score on the functional ability con-
tinuum and 2 standard deviation for the
inpatients and community samples. Note
the location of the peak amount of preci-
sion for each instrument in relationship to
HEALTH CARE FINANCING REVIEW/Spring 2003/Volume 24, Number 3 19

Figure3
MeasurementPrecisionofPost-AcuteCareInstruments,byTest
27
22
c
0
17
:a
0
c
:::s
IL
c
12
0
:;
E
..
i
7
2
-3
,
,
,
.
,
.
,
.
...... \"
-- PF10items
. FIMTM items
--- OASISitems
-.MOSitems
inpatients
- ...- Outpatients
..'
,...
..
.......
, .
'\ .......
-..
...
Inpatients
Outpatients
60 70 80 20 30 40 50
FunctionalAbility
NOTES:CurvesaresuperimposedwiththeaveragescoreonthefunctionalcontimJum and 2standarddeviation
for inpatientsandcommunitysamples. MDSIsminimumdataset.OASISisStandaldi2:edOutcomeand
AssessmentInformationSetforHomeHealIh. FIMT"isfundionallndependentmeasure. PF-10isphysicalfunction
index(10 items).
SOURCES:(Hamilton,Granger,andSherwln,1987;Morris, Murphy,andNonemaker,1995;WareandKosinskl,
2001;Shaughnessy,CrisIsr,andSchIenker.1997.)
each sample. Although the OASIS items
contain a broad range ofcontent, as was
seeninFigure2, theOASISitemsprovide
ahighdegreeofmeasurementprecisionat
onlytheverylowendofthefunctionalabil-
ity dimension for both the inpatient and
community samples. The precision of
MDS items is also greatest at the lower
functional ability dimension levels,
although the MDS items have a greater
spanoffunctionalabilityinwhichtheypro-
vide some levels ofprecision. The FIMTM
itemspeakatthelowto moderateendof
theinpatientsample. Incontrast,theinfor-
mationfunctionofthePF-IOitemspeaksat
thehighendofthecommunityoutpatient
sample with very poor precision for the
inpatientsample.
DISCUSSION
Theresults oftheseanalyses ofinstru-
mentcontent,coverage,andmeasurement
precision provide directevidence ofwhat
many have argued are the major limita-
tionsofexistingfunctionaloutcomeinstru-
mentscurrentlyinusewithinPAC. While
eachofthefourinstrumentscomparedin
thisanalysisappearswellsuitedforitspri-
mary application, none of them appears
well-equipped for thecurrentpolicy man-
date for monitoring the quality and out-
come of PAC provided over time and
acrossdifferentPACsettings.
IfonelooksattheresultsfortheFJM1'M,
themostwidely used outcome instrument
in PAC, one cansee thattheFIMTM items
HEALmCAREFINANCINGREVIEW/Spring2003/Vo1ume24,Number3 20
cover a very small portion of the functional
ability continuum within a narrow range of
functional content The flMTM is most pre-
cise and relevant for PAC inpatients whose
function is at the low end of the continuum.
All of these characteristics of the FIMTM are
acceptable when one considers its primary
application is for evaluation of inpatient
rehabilitation services. These FfMTM char-
acteristics become severe limitations, how-
ever, if the intended application is assess-
ment of outcomes or quality across PAC set-
tings. An instrument such as the PF-10 suf-
fers from a similar type of limitation, as does
the FIMTM. The PF-10 covers a narrow
range of functional content, although, unlike
the FIMTM, the content covered by the PF-10
is at the higher end of the functional activity
continuum. The PF-10 appears most pre-
cise for community dwelling outpatients,
but much less so for inpatients such as
those seen in many rehabilitation facilities.
When used with high functioning communi-
ty patients, the PF-10 covers appropriate
content; its content and range is severely
limited in application to patients within insti-
tutions. The MDS and OASIS instruments,
in comparison with the FIMTM and PF-lO,
cover content from the mid portion of the
functional ability continuum with less con-
tent covering the low or high ends. Patients
functioning at the very high or very low.end
of the functional continuum would not be as
well served by the OASIS and MDS.
Concerns about existing instruments
used in PAC, such as the four examined in
this analysis, have stimulated the develop-
ment of more comprehensive functional
outcome instruments developed specifical-
ly for application across diagnostic groups,
and across PAC settings. An example of
this type of work is found in the activity
measure for PAC (AM-PAC), developed by
the Rehabilitation Research & Training
Center for Outcomes based at Boston
University (Haley and Jette, 2000). In con-
structing the AM-PAC, its developers used
the strategy of combining functional ability
items found in existing instruments and
from a variety of other sources into quanti-
tative scales that can be employed to
assess a wide range of functional content
needed to assess quality and outcomes of
patients seen across PAC settings.
Although instruments such as the AM-
PAC are promising, the continued use of
traditional, fixed-form measurement method-
ology for constructing' functional outcome
instruments presents the researcher with
two common problems that limit their utility
in clinical outcomes assessment One fun-
damental problem encountered with fixed-
form instruments of modest length is floor
and ceiling effects where large numbers of
individuals who complete these outcome
instruments score at either the top or the
bottom of the range of possible scores.
These ceiling and floor effects severely
reduce measurement precision and thus,
restrict the utility of the instruments
(Andresen et al., 1999; Brunet et al., 1996;
DiFabio et al., 1997; Rubenstein et al.,
1998). In response to concerns over inade-
quate measurement precision and inade-
quate coverage of important outcome
domains, some researchers develop more
comprehensive and lengthy outcome instru-
ments. Lengthy instruments lead to the
frustration and fatigue faced by many sub-
jects and busy clinicians overwhelmed by
large and burdensome batteries of instru-
ments (Meyers, 1999).
One promising solution offered to mea-
surement problems faced by traditional
fixed form instruments is offered through
the combined application of IRT methodol-
ogy and CAT techniques (Ware et al., 1999;
Ware, Bjorner, and Kosinski, 2000; Weiss"
1982; Hambleton, 2000). These techniques
of test construction are currently being
applied to the development of a new gener-
ation of functional assessment instruments
REALm CARE FINANCING REVIEW/Spring 2003/V01ume24, Number" 3 21
designed for use in PAC settings (Haley
and Jette, 2000). CAT methodology uses a
computer interface for the patient (or a
computerized interview/clinician report)
that is tailored to the unique functional abiJ...
ity level of the patient. The basic notion of
an adapted test is to mimic what an experi-
enced clinician would do. A clinician
learns most when he/she directs ques-
tions at the patienfs approximate level of
proficiency. Administering outcome items
that are either too easy or too hard pro-
vides little information. An adaptive test
first asks questions in the middle of the
ability range, and then directs questions to
the level based on the patienfs responses,
without asking unnecessary questions.
This allows for fewer items to be adminis-
tered while gaining precise information
regarding an individual's placement along
a continuum of functional ability. CAT
applications require a large set of items in
anyone functional area (item pools), items
that consistently scale along a dimension of
low to high functional ability, and rules
guiding starting, stopping, and scoring
procedures. Large item sets such as the
AM-PAC can be readily adapted to a CAT
format in future work.
Methods like IRT that make it possible
to calibrate questionnaire items on a stan-
dard metric (ruler) also yield the algo-
rithms necessary to run the engine that
powers CAT assessments. These statisti-
'cal models estimate how likely a person at
each level of function is to choose each
response to each survey question. This
logic is reversed to estimate the probabili-
ty of each health score from a particular
pattern of item responses. The resulting
likelihood function makes it possible to
estimate each person's score, along with a
person-specific confidence interval. In
principle, one can derive an unbiased esti-
mate of an outcome, i.e., an estimate with-
out systematic error, from any subset of
items that fits the model. The number of
items administered can be increased to
achieve the desired level of precision. Most
statistical models for estimating such item
parameters can be traced to the work of
Rasch (1980) or on a second tradition-
IRT (Hambleton and Swaminathan, 1985).
Both models assume unidimensionality;
i.e., that the items included on a particular
scale measure only one concept. Whether
these techniques, if applied to functional
outcome assessment, will solve the prob-
lems presented by traditional, fixed form
methodology, needs to be carefully evalu-
ated in future research.
There are several limitations in the
analyses reprinted in this article. To
achieve a direct comparison of these four
instruments, we used a liberal interpreta-
tion of unidimensionality to combine items
from all of the four comparison instru-
mentS into a single scale. This simplified
the presentation so that an instrument-
based rather than a content-based compar-
ison could be made. A more detailed
examination of common functional dimen-
sions that underlie the item set was beyond
the scope of this article. We also point out
that, for practical reasons, we combined
data from medical records and from inter-
views to develop the item calibrations for
the instrument comparisons. Although not
ideal, we do find only small amounts of
error from clinician and self-report inter-
view modes of testing within these func-
tional items (Andres, Haley, and Ni,
Forthcoming). More work in this area of
combining data across respondents and
modes of data collection will be needed as
the field advances in assessing functional
ability across post-acute core settings.
HEALm CARE FINANCING REVIEW/Spring 2003h>1mne24, Number 3 22
CONCWSION
Results of this analysis illustrate some of
the inherent limitations in the range of con-
tent, breadth of coverage, and measure-
ment precision found in functional outcome
instruments currently in use within PAC.
Umitations in existing functional outcome
methodology has stimulated the ongoing
development of more comprehensive out-
come assessment systems specifically for
monitoring the quality of services provided
for patients with dif:ferent across diagnoses
and across PAC settings. The careful use of
IRT-based measurement methods coupled
with CAT outcome assessment techniques
may hold future promise for making out-
comes assessment brief er and less burden-
some to patients, and thus more acceptable
for use in busy clinical settings. What is
needed is functional outcome data that is
applicable to patients treated across differ-
ent clinical settings and applications, more
efficient and less costly to administer, and,
sufficiently precise to detect clinically
meaningful changes in functional outcomes.
Contemporary measurement methodology
may hold considerable promise as a vehicle
for advancing PAC outcomes evaluation,
thus avoiding the pitfalls of traditional
assessment methodology.
REFERENCES
Andres, P.L, Haley, S.M., and Ni, P.S.: Is Patient-
Reported Function Reliable for Monitoring Post-
Acute Outcomes? American Journal 0/ Physical
Medicine and Rehabilitation. Forthcoming.
Andresen, E.M., Gravitt, G.WJ., Aydelotte MR, et
al.: limitations of the SF-36 in a Sample of Nursing
Home Residents. Age and Aging 28:562-566, 1999.
Brunet, D.G., Hopman, W.M., Singer, M.A, et al.:
Measurement of Health-Related Quality of Life in
Mu1tiple Sclerosis Patients. Canadian Journal 0/
Neurological Sciences 23:99-103, 1996.
Buchanan, ]., Andres, P., Haley, S., et al.: Final
Report on the Assessment Instrument /br PPS, MR-
1501-CMS. RAND. Santa Monica. CA 2002.
Cronbach, L: Coefficient Alpha and the Internal
Structure of Tests. Psychometrika 16(3):297-334.
1951.
DiFabio, R.P., Cboi, T., Soderberg, J., et al.: Health-
Related Quality ofIife for Patients with Progressive
Multiple Sclerosis: Influence of Rehabilitation.
Physical Therapy 77:1704-1716, 1997.
Dodd, B., and Koch, w.: Effects of Variations in
Item Step Values on Item and Test Information in
the Partial Credit Model. Applied Psychological
Measurement 11:371-384, 1987.
Guide for the Uniform Data Set for Medical
Rehabilitation, (lncluding the FJMIM Instrument,
Vemon 5.1.) State University of New York at
Buffalo. Buffalo, NY. 1997.
Haley, S., and Jette, A: Extending the Frontier of
Rehabilitation Outcome Measurement and
Research. Journal 0/ Rehabilitation Outcome
Measurement 4(4): 31-41, 2000.
Haley, S.M., and Langmuir, L: How Do Current
Post-Acute Functional Assessments Compare With
the Activity Dimensions of the International Oassi-
fication of Functioning and Disability aCIDH-2)?
Journal 0/ Rehabilitation Outcome Measurement
4(4):51-56,2000.
Hambleton, R.: Emergence of Item Response
Modeling in Instrument Development and Data
Analysis. Medical Care 38(9 Supplement m; II60-
1165,2000.
Hambleton, R and Swaminathan, H.: Item Response
Theory: Principles and Applications. Kluwer Nijhoff.
. Boston, MA 1985.
Hamilton, B., Granger, C., and Sherwin, F.: A
Uniform National Data System for Medical
Rehabilitation. In Fuhrer, M. (Ed.) Rehabilitation
Outcomes: Analysis and Measurement. Paul H.
Brooks. Baltimore, MD. 1987.
Lord, F.: Applications 0/ Item Response Theory to
Practical Testing Problems. Erlbaum Associates.
Hillsdale, NJ. 1980.
McHorney, C., and Cohen, A: Equating Health
Status Measures with Item Response Theory.
Medical Care 38(9 Supplement II); I I ~ I I 5 9 2000.
Meyers, A: Enabling Our Instruments: Assuring
Access to Health Care Research. Proceedings of the
National Conference on Health Statistics.
Washington, DC. 1999.
Morris, IN., Murphy, K., and Nonemaker, S.: Long-
Term Resident Care Assessment User's Manual.
Jle1Sion 2.0. American Health Care Association.
Washington, DC. 1995.
Murnki, E.: Information Function of the
Generalized Partial Credit Model. Applied
Psychological Measurement 17:351-363, 1993.
HEALTII CARE FINANCING REVIEW/Spring 2003/VOIuIlle24,Number3 23
Natioilal Committee on Vital and Health Statistics.
Classifying and Reporting Functional Status.
Internet address: http://ncvbs.hhs.gov. (Accessed
2(02).
Rasch,. G.: Probabilistie Modelsft' Some InteUigence
and Attainment Tests. University of Chicago Press.
Chicago, IL. 1980.
Rubenstein, LM., Voelker, M.D., Chrischilles, RA,
et al.: The Usefulness of the Functional Status
Questionnaire and Medical Outcomes Study Short
Form in Parkinson's Disease Research. Quality of
lift Research 7:279-290, 1998.
Slmughnessy, E, Crisler, K., and Schlenker, R.:
Medicare's Oasis: Standardized Outcome and
Assessment I,qormation Set for Home Health Care:
Oasis-B. Center for Health Services and Policy
Research. Denver, CO. 1997.
van Swieten, J., Koudstaal, P., Visser, M., et al.:
Interobserver Agreement for the Assessment of
Handicap in Stroke Patients. Stroke 19:6Q4.607,
1988.
Ware, J., Bjorner, J., and Kosinski, M.: Practical
Implications of Item Response Theory and
Computerized Adaptive Testing. Medical Care
38(Supplement mII73-n82, 2000.
Ware, J.E., and Kosinski, M.: The SF-3lJj) Physical
and Mental Health Summary Scales: A Manual for
User's olVersion 1, 2nd Edition. Boston, MA. 2001.
Ware, I.E., Kosinski, M., Dewey, J., et al.: How to
Score and Interpret Single-Item Health Status
Measures: A Manual for Users olthe SF..gtM Health
Survey. Quality Metric. Iinco1n, RI. 1999.
Weiss. D.: Improving Measurement Quality and
Efficiency with Adaptive Testing. Applied
Psychological Testing 6:473492, 1982.
Wilkerson, D., and Johnston, M.: Clinical Program
Monitoring Systems: Current Capability and Future
directions. In: AsslIssing Medical Rehabilitation
Practices: ThePromise ofOutcomes Research. Paul H.
Brookes Publishing Company. Baltimore, MD. 1997.
Wright, B.O., and Masters, G.N.: Rating Scale
Analysis: Rasch Measurement. MESA Press.
Chicago, IL. 1982.
Reprint Requests: AIan M. Jette, Ph.D., Sargent College of
Health and Rehabilitation Sciences, Boston University, 635
Commonwealth Avenue, Boston, MA02215. EmaiI: ajette@bu.edu
~
\
~
HEALnI CARE FINANCING REVIEW/Spring 2003/Volume 24, Number 3
i
/
/'
"
t
-e-
o
0...
(1)

...c
u
s-
e
Reliability, Internal Consistency, and
(1)
Cl)
(1)
Validity of Data Obtained With the
c:t::
Functional Gait Assessment
Background and Purpose. TheFunctional GaitAssessment (FGA) is a
lO-itemgaitassessmentbasedontheDynamicGaitIndex.Thepurpose
ofthis study was to evaluate the reliability, internal consistency, and
validity ofdata obtained with the FGA when used with people with
vestibular disorders. Subjects. Seven physical therapists from various
practice settings, 3 physical therapist students, and 6 patients with
vestibular disorders volunteered to participate. Methods. All raters
weregiven10minutestoreviewtheinstructions,thetestitems,andthe
grading criteria for the FGA. The 10 raters concurrently rated the
performanceofthe6patientsontheFGA.PatientscompletedtheFGA
twice, with an hour's rest between sessions. Reliability oftotal FGA
scores was assessed using intraclass correlation coefficients (2,1).
InternalconsistencyoftheFGAwasassessedusingtheCronbachalpha
andconfirmatoryfactoranalysis.Concurrentvaliditywasassessedusing
thecorrelationoftheFGAscoreswithbalanceandgaitmeasurements.
Results. Intraclasscorrelationcoefficientsof.86and.74werefoundfor
interrater and intrarater reliability ofthe total FGA scores. Internal
consistency ofthe FGA scores was .79. Spearman rank ordercorrela-
tioncoefficientsoftheFGAscoreswith balancemeasurementsranged
from .11 to .67. Discussion and Conclusion. The FGA demonstrates
what we believe is acceptable reliability, internal consistency, and
concurrentvaliditywithotherbalancemeasuresusedforpatientswith
vestibular disorders. [WrisleyDM, Marchetti GF, KuharskyDK, Whit-
neySL. Reliability, internal consistency, andvalidityofdataobtained
with theFunctionalGaitAssessment. Phys Ther. 2004;84:906-918.]
KeyWords: Balance; Gait disorders, neurologic; Measurement, applied; Reliability; Validity; Vestibular
system.
Diane M Wrisley, Gregory F Marchetti, Diane K Kuharsky, Swan L Whitney ."fiJ
Cfm&!-w.J rr,') f-,t?
.- . Y f'vv .0-\,.
I I - L- k _1.. ,,t')l.J rNJ' \}J
-'. /"J\ I 1 (\\\0/'
(lit'-V\,4. i '1
'--'
:
:
!
:

:
!
:
906 Physical Therapy. Volume 84.Number 10.October2004

!
:
Clinical Balance"Measures:UteratureResources
Geer Russo, MS, PT
\V"oodrowWdsonRehabilitationCenter
FishersviUe,VA
Introduction
.Physicalthempistswhowork withpatientswithneuro-
logic diagnoses routioel:y encounter in4Mdualswhosedis-
abilityis inpartattributedtobalance Theconcept
ofbalanceis onethathas received muchattention"in the
physicalthempyprofessionoverthepastdecade. A Utera-
toresean:honthe topic ofbalancereveals aplethoraofquan-
titative and quaJitativestudies investigatingthe manydomains
ofbalancefunction.factorsrelatedtoInstability,laboratory
andclinicalassessmentmeasures, aswell aseffectivenessof
various balanceretrainingandfall preventioninterventions.
Continuingeducationprogramsandprofessionalresearch
forumshavefrequentlyaddressedtherapidlygrowing body
ofknowledgeinthis area.
The clinically-relevant questionof-Is thereonesimple
measureforbalance?"was masterfullydiscussedbysixclini-
cian/researchersinthe]anuary1993issue ofPTMagazine.I
Theseexpertsinthefield ofbalanceresearchandclinical
studyquicklysurmisedthat"balanceisaverycomplexen-
titythatmustbe measuredacrossdifferentdomains"and"can-
bereflectedbyanyonesinglemeasure."Thisroundtable
"alogueconcludedthatcliniciansmustidentifywhatinfor
..nation is beingsought whenchoosing a clinical balance
measure. Further,withthegrowinginfluenceofmanaged
careonthetimeavailableto physicaltherapistsforevalua-
tion andtreatment,clinicians mustquickly focuson what
systemsareinvolvedincreatingthepatient'sfunctionalbal-
ance deficit.2 Relevant,tests are thenchosen toserve as
baselineandoutcomemt!asUreSofthetrainingprogramspe-
cificallytailored toaddress thosedeficits identified at the
impairmentandfunctlonallevels.
Objectives
Theprimaryobjectiveofthistableistoprovidearefer-
encetooltobe used bycliniciansin makingclinicaldeci-
sionsaboutwhichbalanceassessmenttouseforparticular
patientsandsettings. Using a formatsuggestedbythepar-
ticipantsinthe1993roundtableonbalancemeasures,clini-
cally familiarbalancemeasuresarereviewedinlightofpub-
lisheddataonreliability,validity,andsensitivityto change.
This table,therefore.offersmorethanabibliographyonbal-
ance.Thisresourcecanbe used forpreliminaryselectionof
ameasurementtoolwithdatatosupportthisselection.
Asecondaryobjectiveof thistableis toencourageclini-
cians toevaluatetheusefulnessofthisfonnatasa mecha-
nismforbringinginformationfrompublished resean::h into
realityof clinicalpractice.
ScopeofResourceUst
TheintentofthistableIs tohighUght, ina singlerefer-
encesoun::e,infunnation tbatshouldbe con.slderedinchoos-
ingameasw:emcnt appropriatefortheclinicalsituation
encountered. Attentionshouldbe given to theParticular
domain,ortheoreticalconstruct, uponwhicha measure Is
based. DuncanIdentifiestourdomains orcomponentsof
balaoce:physlologlcal,blomechaoical.functionaI,2O.dmecha:-
n1stic(feed.back andfeed-forwan1).each ofwblchrequlresa
differentassessmentapproachto accuratelyrepresentIts
particular contributiontothepatient's problem. 1 Payers are
increasinglydemanding thatclinicians demonstratethat
evaluationinstrumentshave demonstrated sen$itlvityto
chaoge(intervention), reliability, andvalidityforthe popula-
tionbeingtreated.:z This resource cba.rt.aloogwiththeass0-
ciatedreferenceIist.is intendedto stunInarize publishedpsy_
chometricdataforclinicianstouse as aresow:ceandtoen-
courageclinicianstofurther analyze thestrengthofthe
publisheddatawithinthecontext of hisor herpracticeset-
ting.
ClinicalImplications
AnyclinicallyrelevanttopiCcouldbereviewedusingthis
formatresultingina platformfordevelopingpracticestan-
dards,documentation guidelines,newclinician mentoring
programs,orclinicalaffiliateinstructionalmodules. Asthe
clinicianreads balanceormObilityresearch articlesthatin-
cludeanyof thereviewedbalancemeasuresintheresearch
methodology,thereadermayusethesummaryinformation
providedin thisformat to evaluate thestrengthofthere-
search designandtheappropriatenessofthechosenmea-
sures. Conversely,thoseinitiatingclinicalbalanceresearch
or designingoutcomemeasures shouldfindthisinformation
helpfulinensuringreliableand validmeasuresarechosen
thatwillalso besensitivetomeasuringchangeinresponse
totheplannedinterventionwithinthepopulationandset-
tingtobestudied.
Someofthemeasurementtools in this tablearebetter
documentedthanothers. Readersmustconsiderthenum-
bers ofstudiessupportingeachmeasurementtoolwhen se-
lectingoneforclinicaluse. ThelessweU-documentedtools
have beenincludedinthetabletoalertcliniciansto their
existenceandtoencouragefurtherdatacollectionto increase
knowledgeofreliabilityandvalidityofthesemeasures.
References
I DunemPW;Shumway-CookA.TmettiME,WblppleRH,WolfS,WooUacott
M.1slheR: onesimplemcasureCorbalmcelPTMagarllU!1993;1:7+81.
GIJI.Body KM,SbumWlJ"COOkA.StIber MU. H2Jf the Visitsand half the
time. PTMftlarlIU!1996;4::6675.
Func:dooaIkach
J WdncrDK, DunaoPW; OwldlerI, et Il. Funct10IWreach: Amarker of
phpiaJ &2flty.JAGS1992;40:203-207.
I
il
C J)ua.<:aO.Pw.WdnetOK.OrmdlerJ,et alJ:uoalooahacb:Ancwc:Uolcal
m.casun::ofbabo.c:c.J 199O;45:M192-1f7.
, '\Vdaer DK,80agiomJ.OR,St.udeasIdSA, et al DoesCwlc:dooalrcacbIm-
pnm:wftbl.dJabII1cadon?APMR 1993:74:196800.
" Duo.caI1PW;StudcDsklSA.awtdIc:rJ,etalPuacdoaaIrc:adl: Prc:d1cdve
'ftUdityIna sampleofclded:y mdevcteAIIS.JGfmmtollm:47:M93-8.
'I 'Ib3paPB. Gldeoo1\ FousbtBI.et1l.00mpadsoaof dio1c:alaodblomC-
c:baoJ.al m.easures andmobilityIIIddedynursinghomeresi-
denIs..JAGS 1994:42:49J-SOO.
OraadkrJM,Dwlc:aa.PW;Studensti SA,etalDocsJc:nverc:x.ttcmltystR:o&th
InfIuco.ccf1IIlc:doaIlteac:bPPT 1991:71:574(Absu2ct).
1imedUp&:Go
t PodsIadloD,BicbanSsoo.S.Tb.etimedUp&:Go": AtA:St ofbaslcfuno.
dooalmobilityfOrfnIlddcd.ypctS08S. JAGS 1991;9:142-148.
IfMatblasS.Nayak.lsucsB.BaJ.ancelllcldcdypaticnts:Thc-gct-upandgo-
teSt.APJUt 1986:67:387-389.
FxqIeyAtuiaBattcryISbarpenedI.ombcrg/OI.ST
11 GnI.JbldA..Frc:glyAltAnewqu:mtitativeatzb.tt:stbattcry.Ada 0t0I4tyng0l
66;61:292-512.
11 BobannonNoJ.addnPA,CookAC,etIf: DeaoseIntimedbaJancelest
$COR:$withagl.Dg.PT84;64:1067-1070.
IlBdggsllC.Gossaw:l MR,BirchR,etalBabnceperformanceamong
IlODio.stitut1oelderlywom.en.PT89; $:748-S6.
I. FIegleyAI. GraybieIA, SmithMJ.walkonfloorc:yesd,osed(WOFEC): A
new:additiDocoanataxialestbattery.Aerospat:e Med72;43:39S-399.
" IIciUDannOK.GQssmanMR,ShaddeausSR,etalBalanceperlbrmance
andstepwidthInnoninstitutiooalizedelderlyCemalef:dlcts&: non-fall-
as.PTI989-,69:923-931.
1<1 BlackFO.WaJlC,RocketterHE,etal.Normalsubjectspostur.llswaydur-
1nltheRombergtest. Am!Oto/aryngoll982;3:309-31S.
PostunlStres5Test
,- CIwldlerJM,DunCU1pw,StudenskiSA..BaI:ma:performanceonthepas-
tur.Ilstresstest:Compllrisollofyoungadults,healthyelderlyandfalIers.
PT 1990;70:410415. '
:tWolfsonU.WhippleR,Amet'llW1p,eta1.Strc:ssi.ngthepostur.Ilresponse::
Aqumtitativemethodrortcstingbalance.}AmGerlatrSac 1986;34:845
SO.
fukadaSteppingTc:st
If NewtonR.lkviewoftestsoC standingb:al2nce abilities.Brain Injury
1989;3:335-43
Clinical TestoC SensoryInte1'2CtionandBaI:ma:
J!l CohenH. BIatcbIyCA,Gombashu..A.studyofCTSlB.PT 1993; 73:346-
54.
:; Shumway..(;ookA,HocakFB..AssessIngtheInfluenceoC sensoryinterac-
tiononbalance.PT 1986;66:1S4&-50.
JZ FB.Oinicalmeasurementofpo.sturaI.controllinadults..PT 1987;
07: 1881-1885.
n Hor.akFB,llyceqiczC.Effectsof vestibul.arrebabilltationonvertigoand
postu.ra1dyscontrol inpatiCDl$withpcripbcral vestibulardisorders.lbun-
dation Foals Spring 94:18-25.
:.Honk:FBJooe$-Rycewl.czC.BlackFO.etaLFifcaoCvc:stibuIarrehabon
dizzinessandImbalaoce..J Otol Head Neck Surg106:17;'18
:.IS DIFabIoBP,BadkeMD.Relation.shi.pof sensory0fg211lzatl0nandbalance
Cwlc:donInpatientswithhemlpJesia.PT1990;70:542-548.
DukeNobilitySkIDs
t WeincrDK.DuncanPW;ChandlerJ.etal Functionalteach:Amarkerof
physicalfrailty.JAGS 1992;40:203-107.
lIoI HogueCC,StudeDsIdS.DuncanP.Asscssingmobility: Thetirststepinfalls
pn:vention.In: Funk:SG.1OmqulstEM.'Champagne MP etal,eds.Key
Aspects0/Re)ttery: Improving NutritIon.,Rest andMoblllty. NewYodc.:
SprlogerPubIfsblngCo.1990,p.275.
r- Studen.sldS.Duncanpw,etal PredictingfaDs: Theroleofmobilityand
non:-pbyslcalfactors.JAGS 1994;42:297-302.
Iflcs Kw.GoldDT.SblppKM,eta1.Assodati.onof osteoporoticvenebral
comprc:sslonfr2Ctvn::swithImpam:dfUDct10naIstatus.AmJMed 1993;
94:595..6()1: ,
OlofideooeInHobBIty/lWlsI!.fDcIq
SI11netdME,lticIuDm.D.FowdlL FaIJs EtIieaqas ameasun:ofbr of
.f.dIias.JGetonIoII99O;4S:239-243. " ,
1JncaIMII,Powd"LFeuofPaIIIIlgandIDw AOmseof
DependeoceInEr4att Fusons.JGfmmtol 1993;48:3s.58.
M MJCSAM.,l'owdllS, ldD:.IBE. et Il.l'IydJ.oJoslcaIlildlcatol:s ofbalanee
coaftdenee:ldadoashIptoIctuaIandpc:n:dmhbUWcs.JGerontol
I!)96;SI(1):M37..(3. .'
ss fowdILB,MyasAM. BaWleeCoofkIence(&BC)
Scale.J
" 11a.ettiMIl',MendesdeLeoa.a. Doucettej'l; etaLFearoff.IlUoaand&JI..
relateddIkacyIIlldatlonsb.lptoCwlc:doa1nsamong
ciders..JGeronlollS194;49(3):M140-7.
FugI-MeycrBaJinceSubsale
SI DIPabIoBP,BadkeMD.Jlclatlonsbipofsc:osoryorganiutionandbalance
functionInpatientswithbemlplegi;PT 199O;70:542-548.
,.FugI-MqI::rAllJaaskoL.IqmannI.etalThepost-fiU'Okeb.emIplegIapa_
rienCAmethodfOrevaluationofpbysk:alpaforJJunce.ScandJRebab
Med 1975;7:13-31.
,.DuncanPW;PropstM,NelsonSG.RdiabilityofFugt-Meyuassessmentof
sensorimotorttCOVerfaftctCVA. PT 1983;63:1606-1610.
..BadkeMD,Duncanpw.Pmemsof!lipidmotOrresponsesduringpas-
turaIadjustmentwhenstandinginhealthysubjectsandhcmJp1egicpa-
tients. PT 1983;63:13-20.
., SantOrd].Mordand]. Sw2nsonLR.eta1.Re1i3bilityoftheFug1-Meyuas-
sessment fortestingmotorperformanceinpatientsfoUowingstmk.e.
1993;73:4474.54.
TmettiBalanceScale
C TUletti ME.Perlonnmce-orientcduscssment of mobilityprobk:msineld-
erlypatients.JAGS 1986;34:119-126.
U TUlettiME,SpeechIyM.Ginta'SERIsk&ctot:sfOrCallsamongddedyper-
sonslivinginthecommunity. N Eng!Med 1988;319;1701-1707.
" TmettiME,W'dliaolsTEMayewskiItFallriskindexforelded:ypadentS
basedODanumberoCchronicdisabilities.JGerontoll990;45:M49-M54."
G ltobbinsAS,RubensteinLZ,.JosepbsonKR,et a1.Predidorsof Callsamong
ddcdypeople:Rc:sulrsoCtwOpopuiatlon-based.5tudl.es ...A.rcbIntemMed
1989;149;1628-1633-
" HaadaN,Qiu Damron-I.odrIguezJ,eta1.ScreeningfOr balanceand
m.obiIityImpairmentinelcIedyindividualslivinginresidential arc.&ciJi..
ties. PTI995;75:462A$.
DynamicPo.stur.agtaphy
- EkdablC,]amleGB.Au.dasonSI.StandingbalanceInbcalchysubjects.
SamdJRebabMed 1989",21:187-195.
(7 Nichols OS.GleanJM,HutchlnsonIQ'.Olangcsinthemeancenterof
b:aI2nceduringbabncetestiDginyoungadults. PT 1995;75:699-706.
..ByINN. Sinnott PI.'Y.IrlatioDsInbalanceandbodyswayIIImiddJe.ged
adults..SP'ffIll9.91;16;32S?30.
DiFablollP.Sc:nsiti'v1c:randspedfidtyofpIationuposturogI:aphyfOrlden-
tifyingpadentswithvc:stibulardysfuncdon. PT 1995;75:290-305.
FunctionalEIlYiroamentalAssessmeot
,.OIaDdIcrjM.Prc:scottB,DuncanPW;etalRdlabilityofanewlnstnlme.nt:
'lbefunclionalenvironmcnl21 PT 1991;71:586(absUad).
)
1','" .. .....
1'mct1Clll1I1)' CommcfIII
.
'nIheUlerature
I
MeasurementTool Levdof Theoretical
I
iSensltlvtto
jlqulpment
Strengths Wc2lcnesftS
Constntet
Rell:lblllly Validity
Chllnge
TestPopulation
Needed
--

, 1'Im<:tIonaJRcac:hlf(PR)
SlmdlnSb'ward usareach
It90"shoulderfla.Ion,feet
stIII,madtfrom3rd
metlcatpalhead,return
uprI&ht.
2pm.11ctI3 teCOl'dedIrlIIs
TImedUp a60"11
SlltldIromarm chall;walk
lOm,hlnl,retIImtorun
sIttInJ back Inchair
1pnctk:c13ublsm:l'OIgc
!ICOI'I:
Yardstldc JJmltedto ItJlICIIn& Mustbeableto: !lm:mt
Level
Yes: InlcrlCC'(1.3):,98' Functlon2l I'osturalcontrolof
popuIJtIon Age j1;().(O fR 14-1,. Standunassisted60sce
Tape
r) In!rd\cstICC:.924 trunk/lCJilfor1'IIIIIkmal
41.$ 13-15" Follow2-$lepcommands
W2II
COPEr=.71 Inda",965
UBmovemen!
P.n::dk:ttfcIIOI1l'IS 7Q.87 JO-I!"
Jstaffperson
Raise ann90' Internalconsistency
fOr fJII dst0IlIJ' In !Jl[lgjC[tlll) Contlnuous coefficientof Test feed-forward
c:kfaIJIlIIles f.gm;b:tlSI.IllI' J"5communltydwelling (partialr'sage- data (lnterva\) variencc=2.5% mechanismfur
correeted) malelfcmlle66-104yr with fR<7"
UDableWIth IIIMR
prepat2!OrypoItUral
rangeot impairments FIIUItymcuures: TcsIretesr.InNhc GIlt1.65Mc:c:
patientICo..577
Kt
spIcaIdc:ftxmllror DukemobUlty AsIf3tto Ic:m:
1'1.65(.52) VBJImItltIoaI
GaJtspced
nelghbodIood bc:llthyJll2le/fem2le
Dukem0b<7ltcms
1'1.71(.63)
lO1untecnage 21-87yr
PR>,. SlDaIt tastcatr
LIfe space(soda! ODeCCIIIIpOIItIIt '41maleextendedcare: GIlt 2.72Mc:c:
, ofd1JIIIIIIc ba!mtc
r=.71(.63)
mobilitymeasure) inpatients
(28 treatment13controls)
mm:ntOO"stdIlSCd)
'21-'communitydwcUing UnabletotI'f8x rIsIt
Odds males(age7().tlM) fR<6' 4x
ItatioprcdiclS 6<PR>IO' 2x
recurrentfalls In '303mn NBmale/ftm21c
('I,krly m:II,' (Rlyr) Lea InfIuenI:oedbrmengtbI
endll.rlncccbanpItor
mobilitymCllSUl'e,
ISIIOdlted withItIlcIc pp
1tI'CI!gth'
. UfCfullbrdIIlIeal
usessmcntorresearch
JJmltcd _wIIh Functlonal !JlOgm:ol' Must beableto; Notreported IntemterICC=.99' Dynamicbalance Is Annd!aIr
Pearsoncoefficients Amblllatewithorwithout componentof basic adJ'JlOMlIIbutaIotr Stopwatch
ofJog-transfonned Smn! C2tt.1!O!I5I rdIIbpatIems
Identifysystem
Tesr.-RetestICC=.9B' Continuous de9lce mobility;does not 3meter'Mllt.way
scores; strongerr's Follow3+ttepcommandor Intcmldall mobile
ImpIIlrment(s)
Ilttfrperson
Indicatecurvilinear No ImpIItmcntInd
urIClIlI hetweenICSIS
Variedonmrage5:teCS goodYlsualmemory <2OJec-
n:l:lllnn)
cllsa!mlnatlon of
&tance,plt,lIp!cd.
IndependentbuIc
'Medicallystable6().90)'l' CI\IICIbrbIImcc
and
trWfi:n
Hers = -.81 (79.5)male/ftm21etestedwith deIcIt
.Most and Independent
Gaitspeed =.61 l'2ngeofImpalnnents functionalcapacitytor toUct trms!tt
householdmobility llarthel = .78
GIlt.slllhec orbetter
'10hcalthyelderly('704!4yr)
.Most dImb IIIh
completedIn<10sce
.Mostamb50+Jd
(meanM, range71Osce)
.Most ro outIIonc

2t).29JCCa'ftdable
... --

-
MeaamnentTool 1m:lof 1beofttlcal
Measure
Construtt
1\'111
PrIIcIlcdlty CommeaII
. IntheUt_hire
I
ISensitiveto
TatPopulation
I EqufpmCftt
Strengths
I
Waknates RellablBty vandlty
ChAnge Needed
Slopwatth 14287111C11, 100women DoesllOlhelp PointBlscerlal lD:r1cX"GrossSCI'CCIlInJ
22lCUteunilateral
Inler-tester No Physical Physiologle:: 1!mdemRomberg'."
Blindfold tot balance(4trI2IsIIIIX detcmIIneCIIISCIof
1sIlffpenon
r=.9913 (5harpcncdSa) Ftrformmce ' Proprioci:pti\'\:
60eec-mu:ICOn:240) llbyrlnlhlno(&lltolfI'ected TImed , Vestlbulu
side:) notpredic:tl\'\:of dJmmIc:bdmcc inteMl
1!mdemlIm,ums
StandIm
fa1Is
'111
MII.c ECScO!I'! probIem:tIIldcm
fl'om'l'1li110
SWIc:talofanklemd crossdlat,cyesdosed
1630 227(SI)34) ''71healthy_en6O-86,r posIdoIlftqIIIrcs
distinguishnonnal
blp$Il2ttg1cswithand
COIItt'Ollimal 31-40 22O(SD43) cIltI'a:eIItpoebII'ld
without'I'Isu2llllput
from-n:stlbular
60RCtdalu:44br
cues(nrItJ'...
41-4' 206(SD6O) losenslttretocbtonlc:unIIat
1IntCJT-
patient1=1.83714
JlreslcTnorms
withdIIWIcc labyrintblllcddlc:\ts(Barany) 190(SD67)
I:JPk'2II1used.
I'IomIWIonary n60 181(SI)72)
maynotbev:l!1d object) CctcbelIupallcnts&ll 0belIelIIIJ'IIOlbe
llIdk:atorofbalance bac:kwml IhIeIOIIJIIIIIIIe
Indderly
JrI&a"l!!:mIk
1!OII!O.tIbors/DIIIlC posI!IoIl
'"
-notcorrelated ti064-56156 25140
wlthMls

31130
24130
7,.'1'9-401-43 14/14
81)86'.4"42 2'lfll
SD(maD)19121 21/24
LacbfImcdonal
PafonDtnee
Pointbl:rcerlnl SOU!CTImeSc:ore ''Tested'Y2Ildltywithmales No MOft:ecnsItMthinS1l PbysIoI Physiological
andperlpberallabyrintbllle lnIerICOo.9923 ra.647 YIlldiIJ ProprJoccpIm:/
patients TImed Ftq!t:J1':fmDIImllt!
OClCdIaJcc,hlp1It:1Itr.I1. P.C,
mtIbulat AnIIIcrossthear.oneleg
SlopW.llth r.rlll'rllln-Nn Intl'MI 16-50BC 130(3%) MInrdclcdror
50RCoruntlImo\'\:etllnee
BlIndfold l!oiIgrecmenl" Concurrent-No hcalthJ' 31-40BC U8(4%) IncreasedIrn:Iof
SOLEQoo8().86% Imtrpmon 2().79rr,sboesoff 4104511C 110(47) dlfllcultyfrom I1CIII'OlosIC
dIapoiIIs
arms,or
SOI.l!(".-S0.67% Wc::Ikorquick , SbIrpenedRomI!eIB 84(47) openCJeS
I
btlgueblpv$ IIOf beabletodo
'trialsIIIIlCICOI'tort50
''71bcaltbywomen6Q.86yr. (Sit) '1-60 BC 77(<<>
tmlance lbrlalOllt
20-40%gaittime:Is
nodIlI'bctM:erIrlshtJlet\,
aardIfcdlO shoesonIotr l'IIIhImI!IslIJ
Notpredlctl'll:of bItIacc
bell:of,trials
lI'Bdgarmutskle,4Ssec. spenton1reg EOJmSD)
IlShca1tby,18w:stIbuIlr 29- ,30 29(2)
patlentstor:IlIlm'1tertaIInJ DtlIIcaIttoduslt't m9- 30 Z8(.S)
10normal,10'RStlbullr .40-4,. 30(1) 24(8) dIlFelatt fII'IJI
rlPtJlelhtulcc,comblned p!ltlCIItStot taI-retal 50.5,. 211(2) 21(10) CnrItJ'hq\I!:nq\
aaxIcnIIInorms /I:IbbUIty 1IIIpIItudc)
J""> .fG)
MustItIlIdunasslsted
JrI&a:"
6o.$s 23(1) 10(9)
Weakhanlstrlnponraised

LB Ol'weakhipof8IllIICCmay
6Q.64.2BJ28 6/8
1lI1ert'cn::
4!6
7O-74a1812O 4/(
713
"""'11110
fIO.86ooJlIJ2
5J3
SDoaJ71J7 314
) /
(,.hamc:lerillllc:, .......kl
/
le
I .)
PrlctlcaUty C'.omIIleIltIJ
In the UICflImre .
M_aremeat Tool Lm:lor 11Ieotet1ca1
I
ISensltlYeto . I Eqalpmem
I
Measure Ccmstruct
ValldJty Test Population
Stra1Jths 'WeIkne!Ises ReUabWty
Change Needed
,
WaIt on floor Eya Closed Pllysical Physlologlc Testietest=I.IIO' DIstinguish between No! Reponed "Tested 287 IIOrmal men and Open 5rbll\le.cutojf1II:IIlI:!2[ fJIIas=o12
(WOfEC', Perfor vcstibular deficient 100 norlllll women; 22 walkway II.CIIIIIIl
N
Not dIscdlIlIIIatIY
mince NatlevcloC and normal 1'=.838
14
Continuous vestibular deficient persons lS!2lfperson
""mea17{j1 yr acore 30 ofleslon IJPC 01' side
count of stcps 'JluxIcm w:dk,EC.ums dlfJIcuIty from SIh"dJe-29
Test for vestibular takc:n CIOfSc:hest,count steps No difference between genders Cannot dI.stIlIguIsh single 11mb stance
_ wtllIlCI2 tlWlS yr
ataxia only If rule out wlo sidestep (max=to) dIfIi:tcna:s IlOmI2b
ICOI.'e 3O,drop to 23111 cerebellar or AcMnccd ambulatory patient
Stop Ifarms down, stop, . proprioceptive With subtle balance dd1c11S StMruJe
problems first by Ifproprlocepd.t iIIdeS!ep, orEO
Atutdchto!llc:1aIIbulat Iestlog with EO cetebdIat Elderly without balance probs
pltlelllSwlllAIlthlste$t l1li}' have dlIllculty"pU5lng" ddldrsCl!st Best 3of' IrlaIs sum of
stepPSCOre (maJ:1t3(
QInIa1Test or 11me seore Ice Concurrent with NOI teported Nlshnc:r's model Ih:l8,2-3" Maypolnt to IOWtC of Stand unwlJted WhIt IslIOIm2I 'fJ impair'
...1)911 Scmory Interacdon dynamic tmbaImce,butmmy ment Eldaty,vcstIbulu pallents,dIzzy, thIdt Mcdlum iImormallllllO!lllt of
Continuous ConditIons re!lect pOSlUrognphy kids wltIt IeamIng dlsabUltles, odler &aors must be 'FI'oami IW'I.'fI
time score patlents' ability 10 Movemc:nl strategy CVA patlenlS tested lantern CODSIdercd as
lntcgtlte and use ICC Kappa=,42.72 domc;VlsuaI f.QIm IDd Notm=InIanced 1-6,(SWJ)' 5,6 0JIlttIbuIInI to instability Less ICIISIdn: tbm
(good agreement)l) 6conditions: visual, vestibular, and ...S7.90
11
target Conditions 13 determine sorUSlD&dyDmIk
l-EO,llnn lIlImalUlICIlllory IO'my; vlsloil dependency,ablc s,stcmatlolly poslIII'IIIIph)'
2-EC./lnn Information under (100% agreement on Spearman suppress intccurale tIsual Stopwatch determl!Ies abilityto
cl)ndllPlIIlIl)( !-Dame,lIrm CAlcfficlcnt wllh tllt2fl'pcnon Pmb 3&: <l=Abnormal visual. select approp 8C1lsory MayIlOt kIaItII'f
Ind YiIIIml/\ I'IIIII-Meycrl' dCPl'lllleru:y I111l1lIlnaCCIIIIII('. Inbmldon Ind
c:hronIe 1\!SIlbuIJf
S-!C.loam somatOtlCn5<ll')'lnput Tcst'n:fCllt feet sc:n.wry rho-.55 Yisll2l YS IO/IWOICII!!IIr or Maycnhance IIIppJallnaa:'Clrate pI1!cut
6-Dome,Ibam togetltcr25 balance rho=.77 analysis of vcstlbulu Informalloll, I1112ble lllput and points to 1lIIIeSI head 1Ikalto
Mec)lanl5tlc LE motor rho=.69 " agreement to select useftlllnpUIJ stDtegles It impairments In balance ..de oppoelte lesion
Eac:b tinted (max 30') for TImescorc (Cecd-back " fecd usc\'ldco aymm PlOb' a: 6=tack'mtlbular
IIIIInblned balance (noroot forward) 8()'1OO%
n:tercncc bced to use.
Does IIOtmeume
hind momnent,or EO); Movement score
vestlbwu becauselIston
fimClloDll bIlmcc:
mcm:ment IIntegy so.tlli:l%
absent or In2ccurate and
usIat ncaled,abIIItJ
o-ankIe
somatosensory intccumc
to "*ftItk: loads
'-hip
!'lob 6a_tosensory
orbody dlt,racIIon
ZaoIher
depend
to Intt:maJICltemII
PlOb 2J,S,6=rriIIon dependent
peIlIllbittIoII (Deed
Prob2-6=depcndent'ris1on A
tocomblllewlth

1'IIIK:tIoadtats)
l'b3aInI StJas1l:5t'NI
Pb)'$ial Mcc:hanlstlc: 89" Igrccmcnt Not tested on rehab 9-polnt ordinal FraU c:IderIy with JlgDU\canI PuIIq ICt1Ip
PossIbJc IIIdk:a1Ion of lh'llld1trtofimc:IIoD ccntftIlntcgtatlon Perfor. bcl:ween two raletS on pallen" yet balanec problems Vi"IIIl belt
dIqeInIons IataIq CcnterofII1UI dlspllccd mance ordinal poslIlml Poor (or or motor " sensory
DbC1Id&hts
teSpOIIIClillell!ethc by droppInl wdShts for postual n:flcxes: stratc&y score Not correlated 10 LB ImrIICIISIr:Mtr healtltynon- Unsafe (or patients with up to 101
poswraI reacdoIl III:lCbcd to belt via pulley EquUJbrlum and (n=88}\'Ideo IlMG" Callers (subtle hcmlpleglal aeatc
IJ'SICID;obsene lbr 9 posturaI righting r7 wIIb IICIII'O assessment
l
' change)OIC
PwIbly,oung tICUIOpatientS Vldeoamera
BS$(mvm
rt
stntegIes to gm: IlaIance strategies III n:sponse popuIaIfoo (Do CrIterion: scores of Corfnil (population not etudled yet) 3staffpeople
s-.,.kon: (IISS)
\OQtcmal pcItUIba+ Cronb:lch alpha: CaUerslnNH elderly fallers daIlIIItrd.ttI,)
lions at Mist 6O-10211ea1thJa21. .99" palienlS<non-hllcrs"
Must iIt2nd Wl2ss1sted
JII&fIItIffeDIt I:IdIID
..t
.... "."1 ....." .. _ ...... ___
I
TestClJlIl'lIcteristles&tabllshc:d
PractlcaUty CommeDtI
IntheUteI'lIture
1beoredca1 Lm:lof
MeaslltcmentTool
Construct StrcrJgthI TestPopulatlon I JIqIIlpment Measure
RdlabWty
Valldlty ISen=to
Needed
I
I
f\JII:adaStepplDgTest"
ArmsoutInfMlt,EC, mm:h
III pIace(becsblgb),
50or100stepsItnormal
,specd.obKm:III09'C!IlCRt
I1I'IJfromJWtpoint
DabMobIIltrSkllla
PnIfoCOl3,..
12itemSfromdyn2mlc
welshtsbIftIngIIIIdmobUlty
tub
BerJIIa1ance'I""
t4c:ommoDIISbof
abIIIt)'toIIIIInIaIn
posItIoaJor I'IIOft:IlICIItof
lIIa'euIrlacII1Bcu1tyby
dcaasIIII_ofsuppon
_ sit, ItIIIIdEOsinglelea
cognItIvt StcpstoOl ordinalrcale
suppon
NOIJIIII=fomrd1m(arely Impalrmestts'l IIt2ffperson
PhysIcal
2conCCllllk bacIr.Tmds)
Pcrfonnance
<4S( rotation ckdl:s(lmeter, dlsta,nceo(dbplaccmem:
2meterdlamclCl) fromlWt,posIIIonand
Abnol.'l1l2l=labyrlnthdellcltat dlYldCInto30-
'.of1'OtI1I0I'I
IIIcn:mcIlIsdmin
,
tosideofdelldt
middlecarorhlgbcr,devllte
01'1 deJtplutIe
IIt2ffperson
Func:tlonal
UsedIIIassessInSmobWty
andtalllist: 111306mile Om:nl broadnngeof
eldCIly(7Q.IMyr)27
Y:udst1ct:
$top'IntdI moblll!rskillsftel!dc:d
0IIIr lot homemobility
Seemsapplicabletomost Shoebox
rchabpopulatlons
Pencil
IIt2ffperson
Testsamplc:"
m PIlIIelderlymean83.Syr SlopwItch NotdIIgnoslsor JP
60CVAmean7t.6yr
, ChaIt
eped1Ic
Punc:tlonal
LabyJlntbine
function
Abilitytoremain
uprightdurlng
progress\'dymote
dlfBcuItsenesof
mobWtylUb
J.!unc:tlonal
coordinationof
Testsm(lbUlty
taSksthat1IIwm:
functionalcoordination
ofmultiplesystems:
sttength,lIcxlblllty.
bIltnce
3Dimensions:
Malnlllinposition
Posturaladjustmentto
YOIImtIrymotion
ReaCllonEO external
penurbatlon
NotteStcd
IIItra'l1Ilcr
r=.76-.9828
'IbIalSCore..
Inter-l1Itc:r=.99
Jtemscores
=.71-.99
Testtest-.98
(YldCOlapc)"
Intcmalconslstcnc:y
Cronbach1=.96"
Nottested
DIstinguishes
bctwtentallC1sand
non-tailsInpredicting
rec:urrenttaIls
17
Com:!atcswithfRat
r=.6S'

onIntcrvlew5with
geriatrIcS&
proksslonals
s:anmm"
60aCUleCVA
correlationwith:
Ilanhcl(.8-.94)
FugJ.Meyer(.62'.94)
Conc:lJl1t!lf'
Correla!c:swith
elderlyIl1l1ngsot
balmce(.3!M1)
withcaregivers
(.470.61)
withCOPE-.55
wllh'nnelll.91
withUp&:Go-.76
..

multiplenlls
Nottested
Notteponcd
3categories
ordinalscale
NO! reported
Spolnt
3 tcommunitydwclIlngelderly
83yr .
RcUabDltytestJns:l'
14patiestts6S+ratedby4Prle
llcl!administrator
'3eldClresidentsofICPfwlth
meanIlIt83yrIIIIdmild
Ruler
GreatersensIttIttJthin
DukeMobUltyIIId
'l'Inettl(5polntns3)
95"Conftdc1'IccIntemI
"
pRdlc:tl
AmbubtloDDmcc
Needed.'29
NoA1dc-47.,.,1.1
CaneOutdoor-47-4'"


,-
IIerBIIItcs!hatc!InIcal
aperlcnce'll'llllld
supportI c:uHI1FICOIe
!
!
of-(SCoutof5S)for
Independentsate
ambulation.
10-1'IIlImI'IaIO
admIDIstctI
l
Notl=ommasdcd
dueto/actof
taI!ng
on1dIIbIIIt)'/nIkIItr
DoesDOtclJstIDpbb
betw=usIsthe
dcr!cell:llc:d
WouldDOtappc:u10
bclSas\tm:to
cbqcISdcslrc:d
NopitII1CiUIIt'CS

mudtftmctIoD2l
1IIIdltyuDate
MobilitylotIdghcr
Im:lIIIObIIII.y
,
f
Y

F
)
J
,

)
"
)
,
TestCh:IracterlslksEslllblbhcd
1'rIctIc:aUty C".ommeatII .
In theUterature
I
Mtu1I1'ell1entTool Levdot Theoredcal
I EquIpment
WeUncsses
.
I
I to
Strengths Measure Construct
Rdlabillty VaIldlty
Change
TestPopuJatlon
Needed
AmC!lolls (thIIl!eedstesting)
FIlbEffiacySale"-l6
rlsk forfalls
Handicap
Context
Psychoemot!onal:
cognldveprocess
Func:donal
Ttstretest=.71"
(pearson)
No(IItar.llerdata

Conscn5lls ofexperts/
Not Itported
Not testedyet
0-10visual
Not reponed
)165+yr(a\'Clll78) cognltlvely
intact,ambulatory(0-74)
78.6yr)
Scaleand
Wot1cshect
hcfhome
Identtrrc:ognItM:IcmotioItII
areaneedlngWorIc to
homeIe!IInI
NottestedJetIn
rebabpopuIatlon
ScIf confidenceat
m1d1ngf.IIls during10
relathdyessential
IlOIIImIrdousmobility
IcttttIIes;meuurcs
impact offatoffI1llng
onbehmar&: function
underlying
emotionoffearor
f.llllng
intarnter-Nomill
elderlyOil 10most
impol'tllntlndependent
livingsitnation

$cOlt Increuedparallel
toselfreportS offcaroC
hUlng
analogicscale tested
1I60 communitydwdUng
ambulatoryelderly(65-95yr)
Istaffperson Increase confidenceIn
mobWty
lacbdemollSU1ted
/ntl:mtCrrelbbWty
and ICIISItIYItyto
cIw!&e (dacn
dtqe Irllilfs
c:f!!c:acyn:Iatc to
I cIw!&e In baImce
Amtt!oa7)

Independentil5soclat(on
with sodal,physical,
an(llhlJumncuon
Fllgl-MC11="DIIlnnce 1I11jlalrmc.1l1 l'ulIlIIl:llhac:kKmlllldfllr 1l'Sln1r51'" l.&l1ll:1I\'" NOI I'Clltll'1cd "HlCVA Blindfold YAIIdItCd onlJ'lot
Sub-ItemJMI volitionalmuscle =.119-.98
UrunnltromSllIgea lIII5CiApatients 1yrpast SmaUltems CIApopubllan
actlv:!t1on ofmRtotrtCm'Cry ;polnl "12left " rightCiApatients (eg,pen.
SitwithoutIIIIPptlfl 1'c.filNlIlIIlC\ Interr.II.".II;(;" nrdimll:I\"ak "1\1hculthy.1I1IcR GVA patterns L'Yllmlcr,hall, Not dcstsncdIS.
ProIectlYereactionnon-
ItTectcd .
.
95%Om8dcncc

Abnormalp651Ur.1I
"12(4l1il(iyr)(:VApattenlJ<Cimll
postwithmoderatNCYCre
etc)
Imffpel'lOll
bI.IInce miIIItlon
IepUlte !TommotQI'
reactionaffected Interval=.82.97 adlustmentin 10left Impairment rec:omypauem
Standwith5IlJll'lnl1
C"NA patk:nt.1cnrrcl:lIClI
StllldwldKlutIUppon
(!) LE motorscore
StandonlIQM/fectedLE
S!2IIdonIfmed LE Conslrucr'
SOTSpearmanrho:.77
'l1ndtlBIIInceScIle'l.tJ.4' Functional IntCSllltlonoC Intemler
Not reported "elderly(orInterrntertesting') Armiw cbaIr RequiresDttlcequipmentor Categorydcsaipllom
PetforDIIIlCl: OrIented physlal.c:ognltive, agreementonItems Bag...91
Pencil InlnIng somewblt "BUt
MobilityAssessment
'"psychological 85",
Darth Mob=.76 3categories 1sI2ll person
componentsof lotalSC:OIt 90%43 normal,
C'ma:sangeotltem bcbIIIIIII!latTscore
ISItansidentifiedIS mohllll)';po5ltlnn

dIfIIcuIty
componentsoCbdmce cb2nga7gait Test-fctCst-Nodala OfbUsInelderly abnormal lacb'nIIdatIonwith
maneuversusedInADL communitydwelling
tdlIb
andICPrelIdents
hncdoall Home Ptrformance Intcm.tcrICOO "24elderlyveterans(mean Worbhect Loob It theImportant Needsnlldltt
Anessment ProflJe'" InHome Totalscun:=.92
componentoCbIIance II.1IdIes 10teeIf
Individual rooms:.72-
Access10 andmoblllty WfthInthe /mpmI'aDcIIIsIn
IdClltlllcsbadsIn .95 patient III hili validcontextotbb dlDlallIICIS1IrI:Sot
homeIndthe frequency
IlIImcepn:dk:I
wItb ...tdcbdiepatientcome3
1ft conIIctwith heard to
ImpRmdSll'ctJ'it
home
determineenvImnrm:nlal
--
....
\,..u.....AI..... w.J'WUW,..,",-"
PrIdIc:IJ1ty
la the UteratIIte
_Tool J.......
'I'heordbI
I
ISensitlYe to
I '1:'
MI!UIIte CDIIIInICt
IdlllbWty Validity
ChIn,:
Tat PopaIatIon .
CoII:ImemI
1
SUeftathI
I

D)'DImIc

lmpaltment
of
VcstIbuJo.
spinal
system
Theoretical
Construct
CNS integIItIon of
YCSIlbubr,ll:
somatosensory
with final motor
output
denIonJtRted by
postulllt sway
lnuz.te5t ICe (1;1)
..J,..9P
lntr.ttest correlation
...92"
Intel'testet correlation
Not reponed Yes
for platform
mclSUrcs
Low sensitivity
todetect/.ng
vestibular
Need to get nOl'l1lS
Any patlerlt able to
unassi.'IICd
SM}' Inc:retses with age
0Ianea:
BaJance
Systent
474
or
Ncutocom
Equlte&t"
or
QuIDIfIIes ICIISOrJ
0JPI1Iza'd0Il tolIlIInt2III
trance IIldmotorCDl'lllOl
topafotmpostuI3I
I'CIctIoI1l'1r.1feJ1cS
(SwIy.bocIywcfght
StatIc mc:uure
?1tdmnccro
dJnrmIC ftmcdon
(GIlt orb2m!en)
0Iattea: not rrrq
I
I Nasbner's model for
-.9" defldts4' patlents Porc:cpbtc
plat!otm
dIstribudoII, limits of
atablllty)

'tImd COIIfIk:t
.
SCIISOr)' orpnlz:lllon
and .mOlar stmcglcs
With platform
Validity
1staJfperson Om posItlOQ toorpIatcs
to mClSlll'C sw:rrIn
IItenW
. p:tIIIdl\It1oD"
mlm:lllent

(sitting OQdlt bomf,sIt

pertudmIon
AssUmeS 'llC5!lbular
I'J"ICII1 actS as Intemal
to_cl, slngleUmb,
. tmdc:m, U7IIIIlIett!c
Is rrrqIIneIt11
rdcrena: against
_cc) rdated tobiIl:ancc1
whlchsenSOl')'
information Is
I Upper aImIIItr
compared
b Intl'lClus
corrcfaIIon
codlIck:Dt
C Nursingbome
d I!yaopcI'IICJCI
dOKd
c 0I!c 11mb_cc
time (SIaCd OQC
Jcg.CJCI dosed)
. IlntemIedlatc care
lidlit)'
J 1nsIrw:rJ0IlII
aatrIdcs ofd2IlJ'
lImB
- ....-
, ...._,
B")t2IfB g.o::l g C 0.:;1'0.0 &0.
0
0!3
l\) l\) l\) tIl 'c.. ,)1<) ::I ." tII U\ ::I
a( .gg n El:a.r; er ::I r;; "'" 0 n n VI "": 0 C O.E.. 11 ." ;:\0"0'.::'\1\1,.--
'0 c:a.....
l.1J
i.Il'" iil"n fIIl:f 1:1 t'l.n n. n Q 0 ,.'; C 1\1 '. ,......
I) c::. ;t fI*" tit -.. ...
III
17
y"(Y
r

'f
t
r
~
:t
I;
IS
u
d
ry
at
i).
lIS
lid
an
in
b.'i
\
Ill-
12;
ml
~
lie
17.
D'-
lid
et,
t
,
I
~
l
I
t
I

I
I
I

t
l
,
I
I
~
I
t
l
I
I
t
l
!
I
I
l

l
l
I
l
l
I
I
t
t
l
Ph'JsioMapy &search International, 2(1). 1997 Whurr Publishers Ltd
Physiotherapists' use of evidence based
practice: a cross ... national study
PAT TURNER La Trobe University, Melbourne, Australia (on secondment from
University of Teesside, Middlesbrough, UK)
T.w.ALLAN WHITFIELD Swinbume University, Melbourne, Australia
ABSTRACT This stud, in4lestigated clinical ph'Jsiotherapists' reaspns for their use of
tTeatment techniques, with a parricular focus on their utilisation of journal retJiew and
research literature; A questionnaire was completed b:Y 180 ph'Jsiotherapists inEngland and
141 ph'Jsiotherapists. in Australia. Despite the greateT premlence ofpre-registration degree
respondents inAus.r.m1ia, rhere weTe no differences between the twonationalgroups in their
reasons given for choice oftreatment t e c ~ The basis ofO4Ier 90% of each groups'
choice of techniques reflected what was taught during their initial training. Experience of
tTeatmenteffects onpriorpatients, andinformation gained in practice-related courses, were
alsoprimaryreasons. Research liteTature ranked least in importanceas a basisfor choosing
tec1miques, andreview articles fared liule bette'l'. The resulas indicated that these ph'Jsiother.
apists relied mainl'J on information gained from formal pre- and post registration courses.
B, implication, the most ob4Iiousmeans of influencing ph'Jsiotherapists' attitudes to research
utilisation lies with those TesponsibIeforph'JsiotheraP'J education.
Key words: research utilisation, treatment techniques, physiotherapy education.
INTRODUCTION
According to Bohannon and LeVeau (1986), the objective of human service profes.
sions such as physiotherapy is to improve thestatus of the client or patient, and all
interventions aimed at achieving this goal should be founded on knowledge that is
research based. The need for such evidence based physiotherapy practice was recog
nised more"than two decades ago (e.g., Campbell, 1970; Hislop, 1975), and has been
emphasised since by many authors (e.g., Bohannon and LeVeau, 1986; Piper, 1991).
Signmcandy, it has emerged recendy as a major issue in the Research & Develop.
"\
Turner and Whitfield Physiol
ment (R&.D) initiative of the Department of Health (DoH) in the United Kingdom.
demon
In fact, the key issues identified by this initiative are the promotion of better utilisa-
Societ'
tion and dissemination of research findings by the therapy professions (Newham.
activit
1994).
above,
Evidence based practice is a method of clinical decision making that requires the
papers,
results of primary research' to be made accessible to those involved in the clinical
import
decision making process (Silagy and Lancaster, 1995; Rosenberg and Donald, 1995). and CL
Guidelines for evidence based practice have been developed in various disciplines,
Des
including public and community health (Oyorkos et al., 1994; Woolf et al., 1996),
dence I
general practice (Sackett and Rosenberg, 1995), critical care medicine (Cook et al.,
researcl
1996), obstetrics (Grimes, 1995), and nursing (Ciliska et al., 1996). MethodS of dis-
deficit
'seminating research findings, and teaching clinicians at all levels, have also been
investil
defined (e.g., Rosenberg and Donald, 1995; Robinson, 1995; Silagy and Lancaster,
don wc
1995).
niques,
Criticisms of the lack of research utilisation in the clinical setting have been lev-
formed
elled at a number of health care professions, including the medical profession educati,
(Bohannon and LeVeau, 1986; Rosenberg and Donald, 1995). Surveys into the din,
The
ical application of research findings have been undertaken in disciplines such as pists w(
social work. nursing, psychology (Bohannon and LeVeau, 1986), and general medi-
compar:
cine (Ellis et al., 1995). .
differen
The DoH initiative to promote evidence based practice amongst the therapy pro-
tion wa:
fessions can be seen in the specific context of criticisms of the physiotherapy profes-
in AUSb
sion. It has been perceived as a profession which bases much of its practice on apy stue .
anecdotal evidence, and which uses treatment techniQUes that have little scientific
pilot wc
foundation (Basmajian. 1975; David. 1985; Riddoch and Lennon, 1991: Which?
ness wa!
magazine, 1995). If this perception is to be changed. then, it has been argued. the tee prior
profession must embrace scientific method (David, 1985; Turner et al., 1996) and
become users of research (Bohannon and LeVeau,1986: Rothstein, 1990).
METH(
Inevitably, the solution to this problem lies in education, in all of its aspects.
Physiotherapy education is multi-faceted. It encompasses formal education at
The que:
both pre- and post-registration levels; and formal and informal continuing profes-
characte
sional development (CPD). Pre-regiscration education has made the transition to physioth
degree status, with the envisaged outcomes of improved professional credentials and
Back,
the acquisition of research skills (Finlay et al., 198J.). Formal posHegistration educa-
tion has developed to consist of practice-related courses. diplomas and. higher
1 Numl
degrees. In the United Kingdom, post,qualification, practice-related courses (PRC)
2 Whet
appear to be of particular importance to practising physiotherapists; for example, sur-
diplol
veys of recently qualified physiotherapists reveal that a major factor in their choice
3 Post...
of employment was the opportunity to attend PRCs (Francis. 1983; Warriner and degre.
Walker, 1996). C.pD, on the other hand, encompasses more than formal courses.
4 Whet
Several stUdies, (e.g., Hightower, 1913; Bohannon, 1990), reveal that physiothera,
degret
pists rely on a variety of sources and activities for gaining information relevant to
physiotherapy practicej these include reading (books, notes, protocols - as well as
The "
journals), discussion with colleagues, prior experience of the effect of a treatment,
participaJ
19 ! r
..
I
!
[
..
Physiotherapists' useofevidencebased practice: a cross-national stud,
demonsttations,andlectures.InarecentbriefingpaperconcerningCPD(Chartered
Society ofPhysiotherapy (CSP), 1994), a wide range ofbothformal and informal
activities were identifiedas being partofCPO. Inaddition to those mentioned
above, the briefingpaperspecifically identifiedthereadingofjournalsandresearch
papers, and the presentation ofresearchpapers. as vital to CPO. This echoed the
importance ofjournalsas partofCPOidentifiedseveral years earlierby Arsenault
andOeather.(1982). .
Despite the clear acknowledgement inthe literature ofthe importanceofevl-
.dencebased practice, itis notable thatresearch has notbeen undertaken intothe
research utilisationpracticesofclinicalphysiotherapists. Itis alsonotable thatthis
deficitwas highlighted adecade agoby Bohannonand Leveau. (1986). In orderto
investigatethistopic.astudywasdevisedtodetermine(a)whichsourcesofinforma,
tionwere used by hospital based physiotherapists when choosingtreatment tech-
niques, Cb) to what extentjournal literature ' particularly research literature'
formed a basis for choosing treatmenttechniques, and (c) towhatextentdegree
educationwasassociatedwiththeuseofjournalliteratureinpractice
Thestudy involved theadministtationofa postalquestionnaireto
pists working inseveral hospitals inEngland. Inorder to provide a cross-national
comparison,thestudywas thenreplicatedinseveralhospitals inAustralia.A major
differencebetweenthetwocountriesisthatwhiledegreelevelphysiotherapyeduca-
tionwas introducedonlyrecentlyinEngland.itwas institutedovertwodecades.ago
inAusttalia.Inbothcountriesthehospitidschosenwereallprovidersofphysiother-
apy studentclinicaleducation.-The questionnaire was designedfollowingextensivD
pilotwork withpractisinghospitalphysiotherapistsinEngland.andits
nesswasverifiedbyformalapprovalbyaphysiotherapyprojectsandethicscommit- , 1
teepriortoitsadministtationinAustralia. &.tAfL-,

ME'lHOD

Thequestionnaire was designed to elicit informationconcerning (a) background
characteristicsofparticipatingphysiotherapists. and (b) theirreasonsfor choice of
physiotherapytreatmenttechniques.
Backgroundcharacteristicswereasfollows:
1 Numberof yearssinceoriginalqualificationasaphysiotherapist.
2 Whethertheoriginalqualification as a physiotherapist was diploma. graduate
piplomaordegree.
3 Post-qualification education undertaken; practice-relatedcourses. diploma,
degreeorhigherdegree.
4 Whethertheparticipantwascurrentlyregisteredfor adiploma,degree orhigher
degree.
The'reasons for use oftechniques' covered treatment'techniquesused by each
participantduringthesixmonthsprecedingcompletionofthequestionnaire.Akey

/
:al
n.
est
5),
d.,
lis,
en
:er,
ev'"
.on
in,
,as
:di,
,ro-

on
:mc
,ch?
the
and

t\at
)fes-
t\to
and
uca-

RC)
sur-
LOice
and
IISe$.
lera
ltto
as
lent,
!

10
I

,
,
I
I

i
I
I
t

i

i
i
t
i


i
I
l

l
;
I
r
I
l
j
L
;
,
I

I

I

i
l
I
I

!

Turner and Whitfield
Plr,sior
list of reasons was provided and, from this list, participants were requested to indi,
have I\
cate why they had used each specified technique. Respondents were requested to list
hospiu
a maximum of three reasons per technique. They could also indicate any reason not
AI
included in the list by completing the 'other, please specify' category. The list of rea'
manag.
sons for use of techniques was compiled from a pilot studYt and was based initially on
the th(
the results of studies by Hightower (1913) and Bohannon (1990). The list of reasons
staff ut
and techniques is given in Table 1.
manage
coverir
pre,pab
TABLE 1: list of teclmiques and reasons used in the questionnaire
Neuro-motor (e.g., &bath) Taught in original training
Passive mobilising/manipulation Suggested by a colleague
(e.g., Maitland. Cyriax eb::.)
General exercises Experience effect on prior patienr/$
Special respiratory techniques Following practice related course
McKemie techniques Following reading - journal or
Massage (including friction) FoJlowin& reading - journal or
PNF (proprioceptive Always use treatment for
neuromuscular faciUtation) specific: condition
SWD -pulsed Pan of research project
SWD continuous Following group presentatioo/ di.scussion
UltraSOUnd Orher please spe<:ify
inrerferential
Acupuncture
Ice therapy
Local heat (e.g., wax, hot packs eb::.)
Infra red radiation
TENS (Tr3llSCUtaneous nerve
stimulation)
Electrical stimulation
Hydrotherapy
Strapping! splints
Suspension therapy
Other techniques (please specify)
review article >
research article
The
apists b
pitals w
RESUI
Questiol
In Engl;
and twe
returnec
to parti<
pletedq
The
poses of
teted fOI
main po
e The.
eAlmc
again
e Then
q u l ~
origir
e Ahig
regist .
.;t;) Convenience samples were used in both countries. Hospitals trusts or groups
known to be centres for physiotherapy clinical education were selected from the
For each
North (North East and North West) and South (South East and South West) of
listed n ~
England, and from the Australian states of Victoria and Tasmania. The selection of
tailed bit
the hospitals in each country was to obtain a sufficiently broad sample, and so to
from the
counter possible bias that could result from using only local. regional hospitals. In
occurred
addition, clinicians at hospitals known to be providers of clinical education would 'taught i
n
1\
r>ups
,the
t) of
nof
10 to
s. In
'OUld
l
Physiotherapists' useofe4lidence based practice: a cross,national stud,
21
l
have reasonable aCcess to university libraries, and therefore journal literature. Ten
hospitalswereapproachedinEnglandandtwenty-oneinAustralia.
l
A letter, togetherwith asample questionnaire, was sent tothe physiotherapy
managersateachoftheselectedhospitals.Theletterrequestedtheco,operationof
thetherapymanagersinthedistributionofthequestionnaires tothephysiotherapy
staffunder theirrespective management. Thefollowing were sent toeach therapy

.
managerwhorespondedintheaffmnative;therequirednumber'ofquestionnaires,a
,covering explanatory lettertoeach prospective participantphysiotherapist and a
I
; pre'paidenvelope.
;
Thequestionnaires, lettersandenvelopesweredistributedtohospitalphysiother,
!
apistsbythetherapymanagersateachofthepartiCipatinghospitals. IndiVidualhos,
pitalswerenotidentified,asanonymityforindividualsandhospitalswasagreed.
RESULTS
Questionnaire responseand background characteristics
In England,eighthospitals (80%) agreedtoparticipateinthestudy.Threehundred
andtwentyquestionnairesweredistributed, and 180completedquestionnaireswere
returned- aresponserateof56.25%.Seventeenhospitals(81%)inAustraliaagreed
toparticipate.Twohundredandfortyquestionnairesweredistributed, and 141 corn.
pletedquestionnaireswerereturned,aresponserateof58.75%.
ThebackgroundcharacteristicsofrespondentsarepresentedinTable 2. Forput'
poses ofthestudy, respondents who hada degree qualification, orwho were regis.
teredforadegreeorhigherdegree, wereclasSed as havingadegreebackground.The
mainpointsofnoteare:
Themajorityofrespondentshadmorethan10yearssinceoriginalqualification.
Almostall'oftheAustralian physiotherapists had degree experience (99%), as
against22%oftheEnglishphysiotherapists.
There was asignificantassociation (X
2
,,,= 137.06, p<O.OOO1) betweenoriginal
qualificationand nationality: 75.2%oftheAustralian physiotherapistsqualified
originallybymeansofadegree,asagainst12.2%oftheEnglishphysiotherapists.
AhigherpercentageofAustralianphysiotherapistseitherhadcompleted,orwere
registeredfor, post-qualificationdiplomas,degreesorhigherdegrees.
Reasons for using- treatment teclmiques
For each respondent, thedatawere tabled according to whetherornotthey had
listedanyofthegivenreasonsforanytechnique,evenif onlyonce (Table
tests were performed (Table 3) toestablish the extentofdeparture
from the test propottionof50%. It is dearthat for both countries thisdeparture
occurred for all buttwoofthe reasons. Further, over 90% ofall respondents listed
'taughtinoriginal training', and 'priorexperience oftechnique' as t:'easons for

ts
TurnerandWhit;Jie1d Ph,sio
TABLE2:BackgroundcharacI:eristiaofrespondents
Characteristics Australian
(n=141)
English
('1'1=180)
Yearssincequalifying
Lessthan2years 22 (15%)
25years 28(20%)
5.10years 32(23%)
10ormoreyears 59(.42%)
Typeofqualification
Diploma 27 (19%)
Graduatediploma 8(6%)
Degree 106(75%)
Postquali6cationeducation(PQE)
None +2(30%)
PRC 81 (57%)
Diploma 22(16%)
Degree 8(6%)
Higher degree 8(6%)
RegisteredforPQE
Diploma 11 (8%)
Degree 2(1.5%)
Highetdegree 12(9%)
34(19%)
37(20%)
30(17%)
79(44%)
7+(+1%)
&4(47%)
22 (12%)
37(21%)
134(74%)
11 (6%)
8(4%)
3 (1.7%)
2(1%)
11 (6%)
..(2%)
lndudedrespondenuwith. orwOOwererqi5reredfar. otherPQE.
TABJ
Reaso
Origir
Colla
PriorI
Taugt.
Jounu
Jounu
Alwal
Res::aJ
Discua
'Othet
byace
tion'c:
In(
nation:
reason
theCOt
4}. Au
physiot
performingtechniques.butjustoveronethirdofrespondents listed'readinga jour
nalreviewarticle'.The'readingof researcharticles'asthereasonforchoiceoftech Ap:
niqueswaslistedbylessthan30%ofrespondents. remark
timese
Educationeffecu
Asignificantassociation(Xl,.=5.70,P<0.02)wasfoundbetweenuseofreview \
desanddegreebackgroundfor theEnglishrespondents;ofthosewithadegree "")
ground. 52% listed usingreview articles, as against32.6% ofthosewithouta degree ,-/ A priOI
background.AmongsttheAustralianphysiotherapistsasimilarsigniftcantassociation theme
(Xl ..,=4.06.pd).OS)wasfoundbetweenuseofresearcharticlesandcurrent theord
tionfor courses.44%of thoseregisteredforpostqualification tation/
tionlistedresearcharticlesasareason,against24.1%ofthosenotregistered. lowing
r
necessa
thekey
Nationaldifference5
respone
Theuse ofPeaISQU' revealed a significantassociation between thetwo ferent (
national groups for two ofthereasonsj7tt:rstralian physiotherapists cited 'suggested several
m
Ph':!siomerapists' useof evidencebased practice: a cross.-national stud,:!
)
i
TABLE3:Percentageuseofeach reason
Numberof respondents
and percentageuseof reasons
Reason Ausaalian English
Physiotherapists Physiotherapists
Originaleducation 134(95.0%) 167(92.8%)-
Colleaguesuggests 96(68.1%)* 101(56.1%)
Priorexperience 134(95.0%) 162(90.0%)-
Taughtin PRe 112(79.4%)* 15S(86.1%)-
Journal -review 50(35.6%) 68(,n9%)*
Joumal research 39(27.7%) 37(20.6%)*
Alwaysuseforcondition 70(49.7%) .67{37.2%)*
. Researchproject 8(5.8%) 15(8.3%)-
Discussion!presentation 41 (29.2%)- 45(25.0%)
'Other'category 15(10.6%)- 11 (6.1%)-
* denotesIK0.01 (departure/ramBinomial ratproponion of 0.50)
byacolleague'(X'l.. = 4.78,p<O.05) and'alwaysuseatechniqueforaspecificcorufi
tion'(XZ", :: 5.00, p<O.05) morefretiuentl'Y thantheir English counft1'PO.11S (Tabk 3).
Inorderto betterdetermine theextentof reportingthevariousreasons by each
nationalgroup, thefollowing were calculated; (a) themeanf!.umber of times each
reasonwaslistedperrespondent.and(b)thepercentageratioof eachreasonagainst
thecombinedtotalnumberof reasonsgivenby eachgroupforalltechniques(Table
4). Australianphysiotherapistslisted morereasonsperrespondentthan theEnglish
physiotherapists(22.98.as against 18.14).A Wile>xon
testrevealedthisdifferencetobe significant(Z- p<O.OOl). --, --
Apartfrom thedifferencesalready acknowledged, the twogroups' responsesare
remarkablysimilar (Table ..).particularlywhencomparing the mean numberof
timeseachreasonwaslistedpersubject.
, am
Priori"orderofreasons
back--
A priorityorderof reasons for usinganytreatmenttechnique was ascertained from
Iegree
themeanslistedin Table4,andisgiveninTable5.Therewas onlyonedifferencein
iation
pstra.
/'"theorderofprioritybetweenthetwonationalgroups;thereason'followingapresen.
tation!discussion' was positionedhigherby the Englishphysiotherapists than 'fol
duca
lowingreadinga journal research anicle'. It is notable that respondentsdidnot
necessarilylistthethreereasonspertechniqueinthenumericalorderofreasonsin
thekey. andmayhavelistedthereasonsinorderof theirownpriority.Fifty English
respondents(21.8%) and46Australianrespondents(32.4%) listedreasonsin adif
ferent orderfrom thekey. Not all respondents listed three reasons per technique;
e two
severallistedonlyone.
lested
23

OO
, '
. .
. 24 Turner and Whit/ield
t
t

TABLE 4: Reasoos given. their percentage of the total and per teSpOndent
Re4son Australian respondents English respondents
(n:14U (n= 180j
nper %0{ mean nper %0{
reason coral per reason total
retISOnS subject rtaSOllS
mean
per
subject
1 Original naining 1120 34.57 7.94 1281 37.99
2 Colleague 261 8.06 1.85 153 7.51
SUgge5tJ
3 Experience 1039 32.07 7.37 993 29.43
4 Special coune 307 9.48 2.18 377 11.17
5 Review article 106 3.27 0.75 131 3.88
6 Research article 95 2.93 0.67 67 1.99
7 Always use 198 6.11 1.4 143 4;24
8 Project 8 0.25 0.06 17 0.5
9 Presentation/ 82 2.53 0.58 89 2.64
discussion
10 Other 24 0.74 0.17 21 0.65
1.11
1.41
5.52
2.09
0.73
0.37
0.79
0.09
0.49
0.12
Torah'e4SON (n) 3209 n/fJ. 22.98 3159 n/a 18.74
TABLE 5: Priority order of reasons listed by respondena
Australian Respondenr:s .
1 Taught in initial naining 1 'Taught in initial training
2 Experience on prior patiena 2 experience on prior patients
3 Practice related COUJSe 3 Practice related coune
4 Suggested by colleague 4 Suggested by colleague
5 Always use technique 5 Always use technique
for specific conditiOll for specific condition
6 Following reading journal 6 FoI1owina reading
review anicle jounW, review article
7 Following reading.joumal 7 FoI1owina presentatiOn/
research article discussion
8 follOWing presentation,l 8 Fo11owina reading
discussion jcumaI research anide
9 'Other' reason 9 Part ofresearch project
10 Part of research project 10 'Other' reason
PIrySiol
TABL
Paden
Doctol
Lutre
philosc
Books
POSt..
Aspect
Prelimi
Mentor
NOt av.
NOt sui
NOt tra
DISCL
Twoma
groups i
by the J
on the i:
The
means (
degree J
cation. I
one Aut
courses
tionalrE
cation iJ
theoppc
The:
literatl.lrt
for a po
recent d
journal I
therapy I
Austt
'OtM-, please spedf" car.e.gory
techniqu
A total of 26 respondents completed this category (Table 6), of whom 15 (10.6%) reasons
were from Australia and 11 (6.1%) were from England. It can be seen that several tion'. Tb
responses were reasons for not performing a specific technique and, apart from emphasis
'books', literature or research did not feature as an alternative reason. ofresearc
f
.6%)
vera1
hom
Ph,siotherapists' use of eWlence basedpractice: a CTOS5.-national stutl,
25

,"

TABLE 6: List of reasons in 'other, pI.eau specify' category
Reason Australian Respondents English Respondents
Patient's request
Doctor's request
Lastreson
philosoph.ic:al objection
Books
POSt-grad trainina
Aspect of work placement
Preliminary to omer treatment
Mentor
Not available
Not suitable for cUent group
Not ttained for technique
1
1
1
1
3
1
1
1
O.
1
..
0
0
1
0
0
2
0
1
1
1
2
1
2
DISCUSSION
Two main points emerge from the srudy: fust, the differences between the two' natiOnal
groups in pre and post-qualification education and in the wider rangeof reasons given
by the Australian physiotherapists; and secondly, the emphasis placed by bothgroups
on the importance.ofinitial training in the selection of treaanent techniques.
The fact that significantly fewer English physiotherapists qualified originally by
means of degree education' is a reflection on the relatively recent introduction of
degree physiotherapy education in the United Kingdom; degree phYSiotherapy
cation, however, was introduced in Victoria andTasmania in1973. Further, all but
one Australian respondent who had qualified prior to the introduction ofthe degree
COUT5eS in Australia later completed a physiotherapy degree qualification. An
tional reason for thisnational difference is that with the introduction ofdegree
cation in Australia, diplomate physiotherapists in the state ofVictoria were given
the oppornmity of converting to a physiotherapy degree.

d
The significant association found for each national groupbetween use of journal
terature and either degree background (English respondents) or curtent registration
for a post.qualification course (Australian respondents), suggests that the more
recent the higher education experience the more likely are physiotherapists to
journal literature as a basis for practice. This may reflect recent changes in
therapy teaching, with greater emphasis being placed on research.
Australian physiotherapists gave awider range of reasons for usingthe treatment
techniques. Thev also made significantly greater useof the anecdotal andrecipe-like
reasons 'suggested bya colleague' and 'always use a technique for a specific condi-
tion', This wassurprising. given their greater experience of degree education, which
emphasises the development of skills such as critical thinking and an understandiDg
of research methodology.
. Ji",
I
t"DYI

'''' Z6
I
Tumet and Whitfield
Intel
I
I Startye
I
Theresults clearly indicate thatbothAustralian and English respondents rely
/

I
I
heavily upon their initial trainingwhen selecting techniquesfor treatment. This
dependence oninitialtrainingwould inevitablyalsoinfluencethephysiotherapists' .I
,,#experience ofthe effect ofa technique onpriorpatients - which was thesecond
I
mostfrequentlygivenreasonfortechniqueselectionlistedbyallrespondents.These
I
(
tworeasons,infact,accountforovertwo-thirdsofallreasonsgivenbybothnational
I
Thereliancealsoontheskills andinfonnationprovidedbypractice-related
coursesisa further indicationofthe importancethatformalcoursesappeartohave : Shipto=
inunderpinningphysiotherapypractice. I Address
Giventhatfew clinicalphysiotherapistswereinvolvedinresearchprojects,andthat
, City---':
thecategoryof'other,pleasespecifYlargelyelicitedreasonswhichwerevariedandnot
I Zi
researchrelated. thentheuseof research literatureranksleast inimportanceforselect- I P-
ingany treatmenttechniqueamongstthe Englishphysiotherapists. Itranks only 1----
ginallyhigherthan'presentation!discussion' amongstthe Australians.Theuse of
I Physi
review literaturefares littlebetter,rankingsixthoutof eightforbothgroups;lessimpor-
tantthaneither theanecdotal'suggestedbya colleague'orthe recipe-like 'alwaysuse
: Inter
thetechniqueforaspecificcondition'.Regardingthelatter,physiotherapistsweretwice
I Startyo'
as likelytochooseitasa reasonfortechniqueselection, thanuseof researchliterature.
I
Researchandreviewliteraturecombinedaccountedforlessthan6%of allreasonsgiven
I
byrespondentsforselectingtreatmenttechniques;aresultwhichsuggeststhatany
I
,.
ificationof practiceisbasedalmostentirelyonformal trainingoranecdotalsources.
Theseresultsaresimilartothose reported by Bohannonand LeVeau (1986)for
I
professionssuchas nursing,psychologyandsocialwork.Themedicalprofessionalso (
I
hasbeenaccused ofptoviding less than25%ofevidencebased treatments (EUis et
I
.,
al. 1995),butinastudyconductedby thelatterauthors,theyfound thatover80%
oftreatments provided in a hospital medical departmentwere evidence based. I Shipto=_
Though thedata inthis latterstudy are comparablewith the current one, the
,
I Address
methodology differed inthat practitioners were notasked why theychose treat-
City---=:
ments; themedicalnoteswere audited, anda databasesearchconductedto deter-
f
POSIaI Cod
minewhethertreatmentschosendid,infact,havea researchbase.
I
Fromaneducationalstandpoint. the resultsofthestudyare disappointing. The 1------
aims ofdegree education indudetheencouragement ofwiderreading, and the
I
developmentofcritical thinkingandresearchskills (FinIay etaL, 1983).However,
I
therewasnodifferencebetweenthetwonationalgroupsintheirdeclaredlackof use
I
ofjournalliterature - researchorotherwise - as a basisfor justifyingphysiotherapy
I
f
IhaveexaJ
treatment.InAustralia,physiotherapyhashadthebenefitofmorethantwentyyears
ofdegreeeducation.againstthemorerecenttransitiontothisfonnofeducationin
: SigDature_
England. It was plausible, therefore, to assume thatAustralianphysiotherapists
wouldhavedemonstratedagreaterimplementationofresearchfindingsasabasisfor
f
clinicalpractice.Infact. thesimilaritiesbetweentheEnglishandAustralianrespon-
I
dents'choiceofreasons aresopronouncedastomarginaliseanyeffectdueeitherto
I
nationalityordegree status. Theextent towhich these similarities are due to the
I
influenceof eitherpre-registrationeducationorpost-registrationnormsofpractice- .
I
NorlhAmel
orboth- dearlywarrantsinvestigation.
I
Physiothen
P.O.Box U
I
Lawrence,J
I
800-627"()6
I
-- -
Physiotherapists' use ofetJidence based praclice: acrosHuuional study
27
IftheiQfonnationprovidedby bothpreregistrationphysiotherapyeducationand
practice.relatedcourses hassucha major influenceonclinicalpractice, thenclearly
muchoftheresponsibility for promoting theutilisationofresearch findings within
physiotherapy practice lies with theeducators. Academicandclinicalstafwho are
responsiblefor delivering theoretical and clinical education tobothundergraduate
andqualifiedphysiotherapistsmustensurethatrecourseismadetotheresearchliter-
ea aturetojustifynotonlyphysiotherapytreatments, butalsomeaSurementand patient
assessment(Turneretal., 1996), InpartiCular, physiotherapistswhoprovide clinical
- educationandsupervisestudentphysiotherapistsmustbeseentoimplementtheprac-
tice ofreading and using joumalliterature.According toRothstein (1990). such .
practiceensuresafuturegenerationofmorescientificphysiotherapists.Ofmajorcon-
cerninthe presentstudy is thefact thatthesurvey inbothEngland and Australia
---I was conducted- amongst physiotherapists working inhospitals thatprovided physio-
MP
POST
IlLL
lVBIl
rlOtTf
lE
therapy clinical training tostudentphysiotherapists; and thatthe qualified physio-
I
therapists whocompleted thequestionnaire indicatedan extremely limited use of
I
researchorotherjoumalliteratureasabasisfortheirchoiceoftreatmenttechniques.
I
AccordingtoSilagy andLancaster(1995).failure toimplementevidencebased
I
treatments can result ina considerable time lapse before effective therapies are
I
introduced.orbeforeineffectiveonesarediscontinUed. Forthisreason itis impera
I
tivethatresearchfindingsareadequatelydisseminatedtomembersof anyprofession.
I'
Onemethodofachievingthis inthemedicalprofessionisthe'CochraneCollabora
I
tion' (SilagyandLancaster. 1995;Robinson. 1995);whichis aninternationaleffort
I
toco-ordinate health care research, and,tosystematically disseminate critical
I
I
reviewsofresearchfindings. Clearly, a similarinitiativewouldbe oneway forward
forphysiotherapists. .
I
I
Atpresent,theassimilationofresearchjournalinformationdoesnotappeartobe
I
amajoractivityamongsthospitalphysiotherapists;inarecentsurveyof physiothera.
I
pistsworkinginteachinghospitalsinEngland(TurnerandWhitfield,1996).journal
I
readershipwasfoundtobeextremelylimited.and mainlyconfinedtotheCSPjour
I
nal.Physiotherapy. Thelatterauthorsfound. however.thatrespondentswithadegree
----I
background had a significandygreaterreadershipofjournalsotherthantheCSP
journal. indicatingthatdegree education may have a positiveinfluence in the
future. though-thisoptimism must be tempered with caution.given the present
resultsfromtheAustralianrespondents. .
CONCLUSION
Onthebasisofthepresentresults. therecentcriticismslevelledatthephysiother-
apyprofession would seem tobe justified. Physiothetapy treatments appear to be
[}
basedonanecdotal.ratherthanresearchbased.evidence.Thebenefitofmanyyears
ofdegreebasededucation,as experiencedby theAustralianphysiotherapists.seems
tohavehadonlylimited impactonpromotingevidencebasedpractice. Infact, the
startlingfeature ofthestudy is thesheersimilarityofbothnationalgroups intheir
reasonsgivenforchoosingtechniques. -
Turner and Whitjielil
Physioth
WhUe it is acknowledged that the results of this survey reflect only the reasons
Robinson
given by respondents. and that the results are also sample,specjfic. and therefore
cannot legitimately be generalised, they are nonetheless informative, and provide an
indication of current physiotherapy practice. It would be of interest to determine to
Assoc
RosenbefJ
Medic
Romstein
what extent these results reflect current practice amongst those involved directly in Sackeu[
physiotherapy education, at both pre' and post-registration levels. Medic
It appealS that there may be considerable room for improvement in aspects of
physiotherapy education. Academic and clinical educators alike must ensure that
future generations of physiotherapists are educated in a climate where reference to,
and implementation of, research fmdings is the norm; and that physiotherapy prac-
SilagyC.
dence
Turner PI
Phytk
TumerP.
tice is based on scientific evidence. even if this means rejecting the techniques that physic
have little scientific basis or evidence of effectiveness.
Warriner;
82: Z9

REFERENCES
her:
WoolfSH
Arsenault AB, Cleamer J. A perspective on Physiotherapy Canada: Results of 1981 readership survey.
Physioc:herapy Canada 1982; 3"': 2836.
lines: :
1996;
BasmajianN. Research or retrench: The rehabilitation professions challenged. Physical Therapy 1975;
55.: 607-610.
Address co
Bohannon RW. Information accessing behaviour of physical merapists. Physiomerapy .Theoty and
sit:!. Melbo
Practice 1990; 6: 215225.
BOOannon RW, LeVeau BF. Clinicians' use of research findings. Physical Therapy 1986; li6: "'550.
Campbell EDR. The purpose of research. Physiomerapy 1970; 6: 480481.
Chartered Society of Physiotherapy. Briefing paper 'Continuing Professional Development'. Physio-
merapy 199+, 80: 623.624.
Ciliska 0, Mitchell A, BaWft;llJU1 A, Sheppard K, Van Berkel C, Adam V, Underwood J, Soumwell D.
Changing nursing practice ttisectoral collaboration in decision making. Canadian Journal of
Nursing Administration 1996; 9: 6073.
Cook DJ, Sibbald W, Vincent JL, Cerra FB. evidence based medicine: What it is and what it can do for
us? Evidence based medicine in a Critical Care group. Critical Care Medicine 1996; Z4: 33+337.
David H. The poor image of Physiotherapy are you contributing to me ptoblem or helping wim me
solution? South African Journal of Physiotherapy 1985; 41: 1s..19.
E1Jis J, Mulligan L, Rowe J, Sackeu DL. Inpatient general mediCine is evidence based. Lancet 1995;
340: 407-410.
Fmlay A. Weaver M, Hearsey S. McCorquedale K. Diploma to degcee: A descriptive survey. Physio-
merapy Canada 1983; 35: 105.
Frands A. A survey into me employment prospects of newly qualified pbysiorherapists. Physiotherapy
1983; 69: 231232. .
Grimes DA. Introducing evidence based medicine into a departrnent of obstetrics and gynaecology.
Obstetrics and Gynaecology 1995; 86: ...51 ....57.
Gyodc.os TW. Tannenbaum TN, Abrahamowia M, Oxman AD. Scou Eo Millson ME. Rasooly I, Frank
J. Riben P, Mathias R. An approach to the development of practice guidelines for community
healm interventions. Canadian Jou.mal of Public Health 199+. 85 SI: 58- 513.
Hightower AB. Continuing education in Physical Therapy. Physical Therapy 1973; 53: 16-24.
Hislop HJ. The dream. Physical Therapy 1975; 55: 10691080.
Newham D. Practical research. Physiomerapy 199"'; 80:
Piper M. Phvsiomerapyand research future visions. Physiomerapy Canada 1991; 43: 710.
Riddoch J, Lennon S. Evaluation of practice: the single case study approach. Physiomerapy TheOty and
Practice 1991; 7: 311.
n
,
l'
Ph,siotherapists' use ofevidence based practice: aCTOSS--ruuional stud,
29
I
Robinson A. Research. practice and me CochraneCollaboration.Journal ofthe Canadian Medical
Association1995:152:883889.
RosenbergW.DonaldA.Evidencebased m.edk:ine: Anapproach toelinlcalprobIem-solving. British
MedicalJoumalI995;310:11221126.
RothsteinJ.Upontheserocks(editorial).PhysicalTherapy1990;70:467-168. I
r
I
SaekettD1.. Rosenberg WM. Theneed forevidenceba&ed medk:ine.Journal01 the ROyal Societyof
Medicine1995:88:620-624.
SilagyC.LancasterT.TheCochraneCollaborationinprimarycare:AnInternational forevi-

dencebasedpracticeoffamjlymedicine.FamilyMedicine1995:27:302305. !
:pt
Turner PAt WhitfieldTWA. A multivariate analysisofphysiotherapyclink:iar.i'journal readership.
PhysiotherapyTheorYandPractice1996;(inpn:ss) .
TurnerPA,WhitfieldTWA.8rewst'erS.HalliganM.KennedyJ. Theassessmentofpain:Anauditof
physiotherapypractice.TheAusttalianJournalofPhvsiothempy1996;42:55-62.
WaninerJM.WalJcerAM.FactorsaffectingphYSiorherapygraduatejobselection. Physiotherapy1996:
82:291-294. .!
Which?Magazine.PhysiotherapyConsumerAssociationreport.WhichlWaytoHeal*1995:Octo-
ber: I
WoolfSH.DiGuiseppeC.Addns0,KamerowDB.Developingevidencebasedclinical guide.
IIneI: lessons learntby theUS Preventive ServicesTaskForce. AnnualReview oflPublic Health
1996;17:511538. L
",75;
Address com.spondence to ABan Whi&fieJd. DivisiOn of Science. Engineeringan4 Design, S 1' Urnm'-
stt" Melboume.Awerafia
.and
rsio-
I
aD.
dof
,for
,
the
:
)95,
t
sic-
apy
lIlY.

mk
tity
t
l