M

I
N
E
R
V
A

M
E
D
I
C
A
C
O
P
Y
R
I
G
H
T

Functional assessment in physiotherapy

A literature review
J.-L. THONNARD, M. PENTA
The present literature review on functional assessment
in physiotherapy was carried out for the following rea-
sons: 1) to identify the functional instruments used in
the field of physiotherapy that were supported by pub-
lished evidence of their psychometric qualities; 2) to
investigate how these instruments relate to the
International Classification of Functioning, Disability
and Health (ICF); and 3) to investigate the use of func-
tional instruments in the financing of physiotherapy.
A search of Medline from 1990 to December 2005, in the
domains of functional evaluation, psychometric qual-
ities, functional classification, and health policy in rela-
tion to physiotherapy resulted in a list of 1,567 stud-
ies. Two reviewers examined the resulting references on
the basis of their title and abstract, in order to select the
studies that presented data on the psychometric qual-
ities of functional evaluation tests, leading to a final
selection of 44 such studies. A selection of functional
tests was identified in four major diagnostic groups
treated in community physiotherapy: musculoskeletal
disorders (including lower back pain), stroke, the elder-
ly, and traumatic brain injuries. The functional tests
authors identified essentially cover the body and activ-
ities dimension of the ICF. The selected tests could be
used as a basis for the standardisation of functional
evaluation of the major diagnostic groups treated in
community physiotherapy. This means that standards
are available for reporting and following the evolution
of patients both longitudinally and transversally.
Nevertheless, in the current literature review no attempt
Funding.This study was funded by a grant from Centre Fdral
dExpertise des Soins de Sant in Belgium.
Address reprint requests to: J.-L. Thonnard, Universit Catholique
de Louvain, Unit de Radaptation, Tour Pasteur (5375), Avenue
Mounier, 53, BE-1200 Bruxelles, Belgium.
E-mail: [email protected]
Vol. 43 - No. 4 EUROPA MEDICOPHYSICA 525
Physical Medecine and Rehabilitation Unit
Catholic University of Louvain, Belgium
at using functional outcomes as a rationale for financ-
ing physiotherapy could be found to date.
KEY WORDS: Physical therapy modalities - Rehabilitation -
Physiotherapy.
P
hysiotherapy is based on an evaluation of the
patients functional health. This evaluation sup-
plements the medical diagnosis, which is primarily
concerned with pathology. For example, in addition
to the medical diagnosis of osteoarthritis, the physical
therapist needs information on pain, joint range of
motion, muscle strength, etc. This information is used
to select appropriate interventions, follow the patien-
ts recovery and assess treatment outcome.
The International Classification of Functioning,
Disability and Health (ICF, World Health Organisation
2001) has been developed to offer an excellent con-
ceptual framework for envisioning the consequences
of health condition or pathology on the function of
individuals. It establishes a common language for
describing the consequences of health that facilitates
the comparison of data across countries, health care dis-
ciplines, services and time. This extensive system
allows up to 1 424 ICF codes to be scored to describe
an individuals functional health Three separate but
related dimensions of functioning are defined: body
functions and structures (body dimension) activity
EURA MEDICOPHYS 2007;43:525-41
M
I
N
E
R
V
A

M
E
D
I
C
A
C
O
P
Y
R
I
G
H
T

THONNARD FUNCTIONAL ASSESSMENT IN PHYSIOTHERAPY

(individual dimension), and participation (social dimen-
sion). 1) Body functions are the physiologic or psy-
chologic functions of body systems (e.g., mental, sen-
sory, neuromusculoskeletal and movement related
functions). Body structures are anatomical parts of the
body such as organs, limbs and their components.
Impairments refer to anomaly, defect, loss or other
significant deviation in body functions and structures.
2) An activity is the execution of a task or action by the
patient. This dimension deals with all aspects of dai-
ly life, envisioning human activities as the purposeful,
integrated use of body functions (e.g., activities relat-
ed to moving about and self-care, and communica-
tion, domestic, or interpersonal activities). Activity lim-
itations are difficulties the patient may have in the per-
formance of these daily activities, whatever the extent
and magnitude of the underlying impairments. Hence,
contextual factors such as the use of assisting devices,
alternative strategies or another persons help do not
eliminate the impairment, but might reduce limita-
tions on activities in specific domains. 3) Participation
is defined as the patients involvement in life situa-
tions in relation to health conditions, body functions
and structures, activities and contextual factors. It refers
to the experience of the patient in the actual context
of life (e.g., social relationships, employment and eco-
nomic life) and it also includes societys response to the
patients level of functioning. Participation restrictions
are the problems the patient may experience in their
involvement in life situations.
Rehabilitation can be defined as a reiterative prob-
lem solving and educational process that focuses on
disability (altered activities) and aims to maximise
participation in society while minimising the stress
on and distress of the patient and family.
1
Community, is used to refer to the physical and
social environment of the patient in the place where
she or he lives. It is usually set in contrast to a hospital,
or another institutional setting where the patient may
go or reside for short periods while receiving a service.
The community might include a nursing home or res-
idential care home if it is the patients long-term res-
idence, but would not normally include a hospital or
short-term stays in a nursing home. The economico-
political profile of rehabilitation in the community
has increased in the last decade. Three factors have
contributed to this rise in interest: 1) the increasing
awareness of disability; 2) the increasing evidence
that active rehabilitation has a beneficial effect in
terms of reducing dependence and care costs, and
in terms of improving the patients quality of life; and
3) the increasing pressure to shorten the time patients
spend in hospital beds. Politicians and health-care
funding agencies have seen an opportunity to control
healthcare expenditure while also appearing to
improve rehabilitation services to disabled people.
They have started to champion community rehabili-
tation in the hope that this will contain or reduce
expenditure by shortening the time patients spend in
the hospital while also improving patient outcomes.
While the outcome of physiotherapy can be
assessed with functional tests, the clinical purpose of
measurement is an important issue that should be
considered. For instance, a diagnostic test should eval-
uate patient functioning in relation to the diagnosis in
question, and variation across repeated measurements
should be minimized. In order to monitor the progress
of patients, the test should provide reproducible results
over time, but also a sufficient range of measurement
with enough sensitivity to detect meaningful changes.
Psychometric qualities of measures are encompassed
under the umbrella terms of validity, reliability (includ-
ing sensitivity and specificity) and responsiveness
2
to
determine if they are meaningful for diagnosis or for
the evaluation of progress.
Validity describes how well a functional test mea-
sures what it purports to measure. Ideally, validity is
assessed against a gold standard, even though such a
standard is not always available for tests measuring
variables that were not previously defined. While
there are many ways to address the issue of validity,
it is generally appraised in terms of the content of
the test or by examining the behaviour of the mea-
surement scale in regard to the underlying theoretical
construct. Reliability refers to the reproducibility of
the measure, and is therefore dependent upon the
amount of measurement error associated with the
measure. A reliable test has items that are internally
consistent (i.e., related to one another) and minimis-
es measurement errors in order to provide the same
results (within confidence limits) regardless of the
particular circumstances in which the measure was
obtained (e.g.,rater, time of evaluation). Finally, mea-
sures intended to measure a patients recovery or lack
of regression should be sensitive to change (have the
ability to detect change given the range of measure-
ment and the measurement error), as well as show
responsiveness (the ability to detect meaningful
change).
The purpose of this literature review on function-
526 EUROPA MEDICOPHYSICA December 2007
M
I
N
E
R
V
A

M
E
D
I
C
A
C
O
P
Y
R
I
G
H
T

FUNCTIONAL ASSESSMENT IN PHYSIOTHERAPY THONNARD

al assessment in physiotherapy is: 1) to identify the
functional instruments used in the field of physio-
therapy that are supported by published evidence of
their psychometric qualities; 2) to investigate how
these instruments relate to the ICF; and 3) to investi-
gate the use of functional instruments in the financing
of physiotherapy.
Methods
A literature search of Medline was undertaken
between October and December 2005 in order to
identify functional evaluation methods of potential
utility for physiotherapy practice within the Belgian
health care system. The literature search focused on
four specific areas: 1) the identification of functional
tests; 2) psychometric qualities of the tests; 3) the ICF
dimensions covered by a functional test; and 4) the use
of functional tests in health care policy. The following
keywords were preliminarily identified in order to
cover the scope of the study: physiotherapy, func-
tional, evaluation, assessment, outcome, disability,
impairment, handicap, participation, quality of life,
life satisfaction, validation, reliability, psychometrics,
health care financing, ICF and ICIDH. The keywords
were then entered into the Medical Subject Heading
(MeSH) in order to find their most relevant counter-
part. The following MeSH terms were retained: phys-
iotherapy, activities of daily living, disability evalua-
tion, outcome assessment, quality of life, evaluation
studies, psychometrics, reproducibility of results, val-
idation studies, ICF, delivery of health care, econom-
ics, health planning guidelines, health policy, health
services research, health services, national health pro-
grams and program evaluation. Each MeSH term was
then combined with the keyword physiotherapy and
with a fixed list of qualifiers (valid*, reliab*, repro-
ducib*, repeatab*, responsiv*, sensitiv*, specificity,
psychometr*) representing psychometric qualities, in
order to limit the search to a manageable number of
references. The complete literature search resulted in
1,567 studies matching these criteria. No systematic
review was found: only papers reporting primary
research were obtained in conducting this literature
review.
Two reviewers examined the resulting references
and selected the studies that presented data on the
psychometric qualities of functional evaluation tests
used in physiotherapy. The references were selected
on the basis of their title and the abstract was consulted
when necessary. Studies with a title or abstract includ-
ing discussion of the psychometric aspects of func-
tional tests or other data relating to functional assess-
ment in community-based physiotherapy were includ-
ed. If studies were published earlier than 1990, were
not written in English, were specific to hospital reha-
bilitation, required the use of sophisticated devices
(e.g.,goniometer, dynamometer), related to psychi-
atric or scar care or focused on country-specific adap-
tations of previously validated scales, they were
excluded. This process allowed 60 references to be
selected for the review.
Two independent readers reviewed the selected
list of references in order to collect data from each
study and encoded them in a standard way. The read-
ers performed a preliminary assessment of 10 studies
together in order to standardise the data collection.
Functional tests were reviewed systematically in order
to describe the clinical utility of each test, identify the
ICF dimension covered by the test and extract the
published data on the tests psychometric qualities,
including its validity, reliability and responsiveness.
Results
Four types of studies were identified among the 60
references reviewed. Forty-four studies presented psy-
chometric qualities of functional tests used in phys-
iotherapy. Fifteen studies discussed clinical, method-
ological or organisational issues, not directly pre-
senting evidence of functional test qualities. One study
discussed issues related to patient satisfaction with
physiotherapy services. The latter 16 studies were not
formally used to review the psychometric qualities of
functional tests (e.g.,clinical utility, metric properties);
instead they were used to develop the arguments pre-
sented in the introduction and in the general discus-
sion of the study.
Each of the 44 studies investigating functional tests
had addressed the metric qualities of one or multiple
tests. The tests were evaluated for validity, reliability
and/or for responsiveness in the following diagnostic
groups: patients with musculoskeletal disorders (N=24)
including lower back pain, spine disorders, chronic
pain, lower limb amputees, lower limb dysfunction
and ankilosing spondylitis; stroke patients (N=16);
elderly patients (N=4) and patients with brain injury
(N=3). Evidence of clinical utility, ICF dimension cov-
Vol. 43 - No. 4 EUROPA MEDICOPHYSICA 527
M
I
N
E
R
V
A

M
E
D
I
C
A
C
O
P
Y
R
I
G
H
T

THONNARD FUNCTIONAL ASSESSMENT IN PHYSIOTHERAPY

ered and metric qualities of functional tests are
described as they were found in the literature.
In the papers we selected, content validity was
assessed using either experts or patients advice or
through a literature review. Concurrent and construct
validity was generally assessed by correlation analy-
ses. The internal consistency was assessed with
Chronbachs alpha.
2
A higher diversity of methods
were used to assess test/re-test reliability: paired t-
test, correlation analysis, percentage of agreement,
intraclass correlation coefficient,
3
weighted or
unweighted Cohens kappa coefficient
4, 5
and
Kendalls coefficient of concordance.
6
Inter-rater reli-
ability was generally assessed by comparing blinded
rater evaluations of videotaped patient performance.
Multiple rater assessments were compared with the
same methods used to address test/re-test reliability.
Responsiveness was generally assessed with analysis
of variance (with or without repeated measures), per-
centage change score, effect size,
7
standardised
response mean (Liang et al 1990) or with the respon-
siveness index.
8
A summary of the functional tests used in muscu-
loskeletal disorders is presented in Table I. Among the
24 studies reviewed, 16 investigated patients with low-
er back pain, two with spine disorders non-specific to
the lower back, two with chronic pain non- specific to
the lower back, two with lower extremity dysfunction,
one with lower limb amputation and one with anky-
losing spondylitis. Three main types of functional tests
were identified. Patient classification based on anato-
mo-pathological or treatment criteria were used by the
physical therapist mostly for treatment planning. Clinical
tests carried out by the physical therapist as part of the
clinical examination (N=8) were used essentially to
evaluate the body structures and functions. Patient-
reported questionnaires, either self-administered or
interview-based, were used essentially to evaluate the
activities of the patient, though some of them also
address the body or participation dimension.
The evidence table of the functional tests used in
stroke patients is presented in Table II. A total of 16
functional tests were reviewed. Among them, 10 clin-
ical tests carried out by the physical therapist addressed
the body dimension. Five tests addressed the whole
body functions after stroke, four were specific to bal-
ance function and one to gait function. The evaluation
of activities was carried out either with clinical tests or
with questionnaires (either self-administered or inter-
view-based). Two test address activities of the whole
body, two tests were specific to the activities of the
upper limb, one to locomotor activity and one to
mobility.
The table of evidence of the functional tests used in
elderly patients is presented in Table III. The four
clinical tests identified addressed the dimension of
activities. Two tests address the performance of whole
body activities and two were specific to locomotor
activities.
The table of evidence of the functional tests used in
patient with traumatic brain injury is presented in
Table IV. Three functional tests were identified. One
of them consists of a classification of functional goals
used for treatment planning. Two other clinical tests
address either whole body activities, though mainly
focussed on locomotor and transfer activities.
Discussion
The literature review allowed a selection of func-
tional tests to be identified in four major diagnostic
groups treated in community physiotherapy: muscu-
loskeletal disorders (including lower back pain),
stroke, elderly, and traumatic brain injuries. The func-
tional tests identified cover essentially the body and
activities dimension of the ICF. Along with function-
al tests allowing one aspect of the patient functioning
to be assessed, a few studies described patient clas-
sification systems that were mostly used by a physi-
cal therapist in order to establish treatment goals and
plan patient treatment.
The primary observation is that most of the tests
identified cover the body and activity dimensions of
the ICF, while very few tests address the participa-
tion dimension. Note that tests addressing participa-
tion have been developed more recently than tests for
the other dimensions of functioning. Although reha-
bilitation aims at improving the performance of activ-
ities and participation in society according to Wade,
1
the clinical practice of physiotherapy puts a higher
emphasis on the body dimension of functioning. This
traditional approach is probably motivated by the fact
that the selection of interventions of physical therapists
is related to the treatment goals that are pursued.
50
Especially treatment goals at the level of impairments
(body dimension) and at the level of activity limitations
(individual dimension) are formulated.
51
Thus, up to
now, treatment goals at the level of participation
restrictions (social dimension) seem to be less specific
528 EUROPA MEDICOPHYSICA December 2007
M
I
N
E
R
V
A

M
E
D
I
C
A
C
O
P
Y
R
I
G
H
T

FUNCTIONAL ASSESSMENT IN PHYSIOTHERAPY THONNARD

Vol. 43 - No. 4 EUROPA MEDICOPHYSICA 529
TABLE I.Evidence of functional tests used in patients with musculoskeletal disorders.
Functional test Clinical utility Metric properties General appraisal
Lower back pain
Diagnostic classification
system for non-specific
lower back pain
Type: classification
9
Treatment based classifi-
cation of lower back
pain patients (TBC)
Type: classification
10
Classification of patients
with LOWER BACK
PAIN
Type: clinical test
11
Modified-Modified Scho-
ber Test (MMST)
Type: CLINICAL test
12
Physical Impairment
Index (PII)
Type: clinical test
13
Diagnostic subclassification of
non-specific lower back pain
consisting of pathoanatomical-
ly labelled syndromes assumed
to refer to a specific pathologi-
cal condition
ICF dimension: body structure,
body unction
Treatment-based classification
system, based on physical
examination and patient self-
reports of pain and disability,
for use in the evaluation and
treatment of patients with acute
lower back pain
ICF dimension: body structure,
body function
Impairment-based classification
system addressing 28 symptom
behavioral items and 22 align-
ment and movement items in
an attempt to define mutually
exclusive categories of lower
back pain problems
ICF dimension: body structure,
body function
Test measuring range of motion
of the lumbar spine in the fol-
low-up of lower back pain
patients
ICF dimension: body function
Evaluate the physical impairment
in lower back pain patients,
through 7 tests, each scored
dichotomously based on pub-
lished cut-offs
ICF dimension: body function,
body structure
Reliability: percentages of agreement ranged from
74% to 100% and kappa coefficients ranged from
0.26 to 1.00. Inter-tester reliability of categorisation
of the syndromes was acceptable
Reliability: the interrater reliability was moderate (kap-
pa=0.56) according to criteria of Landis and Koch.
The percentage agreement between therapists was
65%
Validity: content validated by 4 orthopaedic physical
therapists
Reliability: experienced therapist who had trained
together were able to agree on the results of exam-
ination and obtain acceptable level of reliability
(kappa >=0.75 for all items related to symptoms
elicited and >=0.40 in 72% of items related to align-
ment and movement)
Validity: Pearson correlation coefficient between the
MMST measurements and the gold standard (Xray)
is 0.67 (95%CI: 0.44-0.84)
Reliability: the intra-rater reliability was excellent
(ICC=0.95; 95%CI 0.89-0.97) The inter- rater relia-
bility was excellent (ICC= 0.91; 95%CI 0.83-0.96)
Responsiveness: a change over 1 cm on the MMST
must be observed to be 95% confident that a true
change in ROM occurred
Validity: convergent validity was supported by sig-
nificant correlations with disability (r=0.51 with the
Roland-Morris questionnaire), work loss in the past
year (r=0.43), pain (r=0.27), depression (r=0.26 with
the Zung depression inventory), somatisation (r=0.32
with the modified somatic perception questionnaire),
nonorganic signs (r = 0.49) and nonorganic symp-
toms (r = 0.35). The highest correlation were found
with pain ratings (r=047), the Oswestry disability
questionnaire (r=0.42), nonorganic signs (r=0.42)
and nonorganic symptoms (r=0.36). Smaller but sig-
nificant correlations were found with the physical
component score of the SF-6 (r=-0.28) and the phys-
ical activity subscale of the fear-avoidance beliefs
questionnaire (r=0.24)
Reliability: good to excellent reliability for individual
items (ICC or kappa coefficients ranging from 0.48
to 0.96). The overall score demonstrated excellent
interrater reliability (ICC=0.89); the reliability coef-
ficients for individual components rang
Not applicable since this
study proposes a diag-
nostic classification sys-
tem for non specific lo-
wer back pain, rather
than functional evalua-
tion
Not applicable since this
study proposed a clas-
sification process for
patients in the acute
stage, rather than func-
tional evaluation
Not applicable since this
study proposed a clas-
sification of impair-
ments in lower back
pain patients, rather
than functional evalua-
tion
The metric properties of
this test support its use
in community-based
physiotherapy. The test
takes 5 minutes to be
administered in by the
therapist
Systematic clinical test
using published cut-off
values for scoring phys-
ical impairment in
acute lower back pain
(to be continued)
M
I
N
E
R
V
A

M
E
D
I
C
A
C
O
P
Y
R
I
G
H
T

THONNARD FUNCTIONAL ASSESSMENT IN PHYSIOTHERAPY

530 EUROPA MEDICOPHYSICA December 2007
TABLE I.Evidence of functional tests used in patients with musculoskeletal disorders (continued).
Functional test Clinical utility Metric properties General appraisal
Shuttle walking test
Type: clinical test
14
Standing flexion test
Type: clinical test
15
Aberdeen back pain
scale
Type: self-administered
questionnaire
16
EuroQol (EuroQol)
Type: self-administered
questionnaire
16
Jan Van Breemen Insti-
tuut pain and function
questionnaire
Type: self-administered
questionnaire
17
Lower back SF-36 physi-
cal functioning scale
(Lower back SF-36 PF)
Type: self-administered
questionnaire
18
Measure of walking capacity in
patients with chronic airway
obstruction, chronic heart fail-
ure and lower back pain, requir-
ing the patient to shuttle on a
10-meter course at speeds rang-
ing from 1.8 to 8.53 km/h
ICF dimension: activities
Manual palpation of the sacroili-
ac joint in patients with lower
back pain
ICF dimension: body function
Clinical assessment of patients
with lower back pain made of
19 items of either forced choice
or multiple choice, producing
a back pain severity score
ICF dimension: body function,
activities, participation
Generic instrument that incorpo-
rates descriptions and valua-
tions of health states, applied
in lower back pain patients
ICF dimension: participation
Self-administered questionnaire
consisting of 6 questions ad-
dressing pain and 9 questions
addressing functional capacity
in lower back pain patients
ICF dimension: body function,
activities
Back-specific version of the SF-36
physical functioning scale, com-
bining the advantages of both
generic and specific functional
questionnaires into a single, par-
simonious set of items from
Reliability: the test-reliability is excellent (ICC =0.99)
with a mean difference of 2.5 m between assess-
ments, and upper and lower limits of agreement of
52 m and -47 m
Responsiveness: Patients undertaking fitness training
reached an effect size of 1.2 compared to a control
group of 0.23 and 0.94 for a group undergoing var-
ious orthopaedic treatments
Reliability: intraexaminer reliability data demonstrat-
ed a mean percentage agreement of 68% and a k
coefficient of 0.46 indicating moderate reliability.
Inter-examiner reliability data, with a mean per-
centage agreement of 42% and a k coefficient of
0.052, demonstrated statistically insignificant relia-
bility. These results suggest that the reliability of the
standing flexion test as an indicator of sacroiliac
joint dysfunction still remains questionable
Validity: developed through a review of the clinical lit-
erature and selection of items that reflected areas
of importance in the clinical assessment of patients
with lower back pain. Small but significant correla-
tion (P<0.01) with two pain scales (r=0.24 with worst
pain and 0.25 with best pain), the number of days
absent from work over the past 6 months (r=0.26)
and the number of visits to a general practitioner
for back pain in the last 6 months (r=0.23)
Responsiveness: large responsiveness to treatment as
assessed by the standardized response mean (-7.64
to -13.36), either at 6 weeks, 6 months and 1 year fol-
low-up
Validity: content derived from a review of existing
instruments, consisting of five items measuring mobil-
ity, self-care, usual activities, pain/discomfort and
anxiety/depression. Small but significant correlation
(P<0.01) with two pain scales (r=-0.20 with worst
pain and -0.29 with best pain), the number of days
absent from work over the past 6 months (r=-0.14)
and the number of visits to a general practitioner
for back pain in the last 6 months (r=-0.28)
Responsiveness: Moderate responsiveness to treat-
ment as assessed by the standardized response mean
(0.02 to 0.11), either at 6 weeks, 6 months and 1
year follow-up
Responsiveness: sensitive to clinically important
change (with function scale %change =33 and pain
scale % change =37) after treatment of 4 to 6 weeks
Validity: 18 items pooled from the Australian version
of the SF-36 physical functioning scale, the modified
version of the Oswestry questionnaire and the
Quebec back pain disability scale. The lower back
SF-36 PF scale has fewer misfitting items than the
original SF-36 PF scale and eliminates its floor and
The metric properties of
this test show sensitivi-
ty to clinically mean-
ingful change, a 10 m
walking course is requi-
red and the test takes
15 min to be adminis-
tered by the therapist
The metric properties do
not support the use of
this test in community-
based physiotherapy.
Moreover, the sample
size of the study was
very small (9 patients)
The metric properties do
not support the use of
this test in community-
based physiotherapy
since it is less discrimi-
nating than the well
known Roland Disabili-
ty Questionnaire
The metric properties do
not support the use of
this test in community-
based physiotherapy
since it is a generic
instrument, lacking res-
ponsiveness to treat-
ment
The metric properties
support the use of this
test, nevertheless this
scale is multidimen-
sional
The low-back SF-36 PF18
proposes a revision of
the well established SF-
36 PF scale. It shows
better psychometric
properties than the
(to be continued)
M
I
N
E
R
V
A

M
E
D
I
C
A
C
O
P
Y
R
I
G
H
T

FUNCTIONAL ASSESSMENT IN PHYSIOTHERAPY THONNARD

Vol. 43 - No. 4 EUROPA MEDICOPHYSICA 531
TABLE I.Evidence of functional tests used in patients with musculoskeletal disorders (continued).
Functional test Clinical utility Metric properties General appraisal
Oswestry
Type: Self-administered
questionnaire
17
Roland disability ques-
tionnaire
Type: self-administered
questionnaire
16
Roland-Morris Question-
naire (RMQ)
Type: self-administered
questionnaire
19
Roland-Morris
Type: self-administered
questionnaire
17
Sickness Impact Profile
(SIP)
Type: self-administered
questionnaire
20
which the original SF-36 PF10-
item scale can still be extracted
ICF dimension: body function,
activities
Self-administered questionnaire
of disability in lower back pain
patients
ICF dimension: activities
Self-administered disability mea-
sure in lower back pain reflect-
ing 24 activities of daily living
ICF dimension: activities
Self-administered disability mea-
sure in lower back pain reflect-
ing 24 activities of daily living
ICF dimension: activities
Self-administered disability mea-
sure in lower back pain reflect-
ing 24 activities of daily living
ICF dimension: activities
Self-administered disability mea-
sure in lower back pain reflect-
ing 24 activities of daily living.
ICF dimension: activities.
Behaviorally based measure of
perceived health status applic-
able across a spectrum of ill-
nesses and among various
demographic and cultural sub-
groups, applied in lower back
pain patients
ICF dimension: participation
ceiling effects. Unidimensional and linear scale devel-
oped with the Rasch partial credit model
Reliability: intrarater reliability of the PCBS test were
.94 and .96, respectively. The test-retest of the low-
er back SF-36 PF (ICC=0.91) was comparable to that
of the original SF-36 PF and of the Oswestry
Responsiveness: The responsiveness of the lower back
SF-36 PF was slightly, though not significantly,
improved compared to that of the original SF-36 PF.
The minimum detectable change was improved, as
compared to the original SF-36 PF, from 16 to 12
points on a 0-100 scale expressing a percentage of
the maximum possible score
Responsiveness: sensitive to clinically important
change (%change =40) after treatment of 4 to 6
weeks
Validity: 24 items selected form the Sickness Impact
Profile represent the areas of greatest relevance to
lower back pain. Small but significant correlation
(P<0.01) with two pain scales (r=0.26 with worst
pain and 0.33 with best pain), the number of days
absent from work over the past 6 months (r=0.28)
and the number of visits to a general practitioner
for back pain in the last 6 months (r=0.29)
Responsiveness: Large responsiveness to treatment as
assessed by the standardized response mean (-1.04
to -3.62), either at 6 weeks, 6 months and 1 year
follow-up
Responsiveness: the magnitude of the measurement
error is sufficiently small to detect change after 4 to
6 weeks of physiotherapy treatment in patients with
initial scores in the central portion of the scale; how-
ever, the magnitude is too large to detect improve-
ment in patients with scores of less than 4 and dete-
rioration in patients who have scores greater than 20
Responsiveness: sensitive to clinically important
change (%change =40) after treatment of 4 to 6
weeks
Validity: validity coefficients of moderate magnitude
have been reported between the SIP and various
biologic and clinical measures, and between sub-
sections of the SIP and sub categories of the
Minnesota Multiphasic Personality Inventory
Reliability: A high test-retest reliability coefficient
(r=0.85) has been demonstrated in a lower back
pain patient group
Responsiveness: 20 items of the SIP showed change in
at least 20% of the patients after a 6.4 weeks treat-
ment and have an item-total score (minus item) cor-
original scale for the
assessment of function-
ing in patients with
lower back pain. The
scores are expressed on
a linear scale which
enables quantitative
comparisons of func-
tional states. The test is
self-administered and
takes 5 min
This self-administered
disability measure in
lower back pain is a
well established test
completed in 5 min.
Instructions are avail-
able
This self-administered
disability measure in
lower back pain is a
well established test
completed in 5 min.
Instructions are avail-
able
This self-administered
disability measure in
lower back pain is a
well established test
completed in 5 min.
Instructions are avail-
able
This self-administered
disability measure in
lower back pain is a
well established test
completed in 5 min.
Instructions are avail-
able
The metric properties
support the use of this
self-administered test
addressing participation
in lower back pain
patients
(to be continued)
M
I
N
E
R
V
A

M
E
D
I
C
A
C
O
P
Y
R
I
G
H
T

THONNARD FUNCTIONAL ASSESSMENT IN PHYSIOTHERAPY

532 EUROPA MEDICOPHYSICA December 2007
TABLE I.Evidence of functional tests used in patients with musculoskeletal disorders (continued).
Functional test Clinical utility Metric properties General appraisal
Spine disorders
Fingertip-to-Floor Test
Type: clinical test
21
Passive intervertebral
motion of the cervical
spine
Type: clinical test
22
Chronic pain
Global Physiotherapy
Examination 52 (GPE-
52)
Type: clinical test
23
Chronic pain intensity
measures
Type: telephone inter-
view
24
The finger-to-floor distance might
be used to assess spine stiffness
and the effects of exercise on
spine stiffness in persons with
spine disorders
ICF dimension: body function
Passive motion between adjacent
articular segments is assessed
when diagnosing and treating
dysfunctions of the spine
ICF dimension: body function
Physical examination for muscu-
loskeletal pain patients in 5
domains: posture, respiration,
movement, palpation muscle,
palpation skin
ICF dimension: body structure,
body function
Rating through telephone inter-
view of current, worst, least and
average pain intensity on Visual
Analog Scales over a deter-
mined period of time in chron-
ic pain patients
ICF dimension: body function
relation greater than 0.30. Seven of the 20 items
identified in this study appear in the Roland-Morris
disability questionnaire
Validity: Spearmans correlation coefficient of trunk
flexion assessed by the fingertip-to-floor test and
the radiologic measure was excellent (r=-0.96)
Reliability: The intra-observer reliability of the finger-
tip-to-floor test was excellent with an ICC of 0.99. The
interobserver reliability of the radiologic analysis
was excellent with an ICC of 0.99
Responsiveness: The responsiveness of the fingertip-
to-floor test was evaluated by the SRM (0.97) and the
effect size (0.87)
Reliability: the percentage agreement for both thera-
pists for all tests was 77% (7087%). The kappa coef-
ficient varied between 0.28 and 0.43, considered to
be only fair to moderate. Even if two examiners
have equivalent clinical experience and education-
al backgrounds it is difficult to demonstrate accept-
able inter-examiner reliability in the assessment of
passive cervical joint motion. The results of this study
showed lower concordance than expected in spite
of the optimal testing conditions
Validity: discriminates between patients with local-
ized versus widespread pain. Discriminates between
patients with long-lasting musculoskeletal problems
and people who are healthy. Recently examined on
data from people who were healthy and patients
with long-lasting musculoskeletal pain
Reliability: Good to excellent inter rater reliability of
the total score (ICC=0.91) and individual compo-
nents (ICC=0.65 for posture, 0.60 for respiration,
0.89 for movement, 0.83 for muscle palpation and
0.76 for skin palpation)
Responsiveness: responsiveness to important change,
defined in this study as return to work, was found
only for the total GPE-52 score and within the
movement and respiration domains. Responsiveness
to important change was greater in patients with
localized pain than in patients with widespread
pain
Reliability: the relatively low test-retest stability
(0.550.65) of some of the individual ratings (aver-
age, current and worst) between one pair of assess-
ments (1-month to 2-month follow-up) raises some
issues
Responsiveness: Each of the individual ratings was
able to detect expected changes in pain intensity
from pre-treatment to various points after treatment.
Although the composite measures appeared more
sensitive to treatment effects when compared to the
individual ratings, these differences were not statis-
tically significant
The metric properties of
this test support its use
in community-based
physiotherapy. The test
takes 1 min to be
administered in by the
therapist
The metric properties do
not support the use of
this test in community-
based physiotherapy
This test requires a 3-day
training which is a dis-
advantage compared to
other tests, and it takes
30 min to be completed
The metric properties do
not support the use of
this test in community-
based physiotherapy
since it is not repro-
ducible over time
(to be continued)
M
I
N
E
R
V
A

M
E
D
I
C
A
C
O
P
Y
R
I
G
H
T

FUNCTIONAL ASSESSMENT IN PHYSIOTHERAPY THONNARD

to physiotherapy, and fall into the realm of other med-
ical and paramedical specialities. This notwithstand-
ing, participation remains one of the key research
areas in community rehabilitation according to Wade:
1
Much research is needed to develop and evaluate
the expertise that community-based teams might have.
For example there is minimal research into the
processes of giving support to patients and their fam-
ilies, monitoring, prevention of complications, and
facilitation of social participation. This observation
also suggests that outcome evaluation in physiother-
apy should be in keeping with this framework in
order to monitor functional recovery of the patient
and treatment efficacy.
The functional tests reviewed appeared to address
the dimensions most related to the clinical manage-
Vol. 43 - No. 4 EUROPA MEDICOPHYSICA 533
TABLE I.Evidence of functional tests used in patients with musculoskeletal disorders (continued).
Functional test Clinical utility Metric properties General appraisal
Lower Extremity Functio-
nal Scale (LEFS)
Type: self-administered
questionnaire
25
Questionnaire Rising and
Sitting Down (QR&S)
Type: self-administered
questionnaire
26
Lower limb amputation
Functional Measure for
Amputees (FMA)
Type: Self-administered
questionnaire
27
Ankylosing spondylitis
World Health Organisa-
tion Disability Assess-
ment Schedule II
(WHODAS II)
Type: Self-, Interviewer-
or Proxy- administered
questionnaire
28
Self-reported, condition-specific,
functional measure applicable
to a wide range of patients with
lower-extremity orthopedic con-
ditions
ICF dimension: activities
A self-administered questionnaire
that measures perceived and
actual functional limitations in
rising and sitting down of
patients with lower extremity
dysfunction
ICF dimension: activities
A questionnaire collecting long-
term functional and prosthetic
use information following dis-
charge in lower limb amputees
ICF dimension: activities
Generic instrument measuring the
level of disability across various
conditions and interventions,
applied in ankylosing spondyli-
tis
ICF dimension: activities
Validity: correlations between the LEFS and the SF-36
physical function subscale and physical component
score were r=0.80
Reliability: Excellent internal consistency (Chronbach's
alpha=0.96). Test-retest reliability of the LEFS scores
was excellent (r=0.94 [95% lower limit confidence
interval = 0.89])
Responsiveness: The sensitivity to change of the LEFS
was superior to that of the SF-36 in this population
Validity: scale sum scores correlate 0.30 to 0.41
(P<0.001) with doctors or therapists global assess-
ment of functional limitations. Sum scores correlate
0.20 to 0.59 (P<0.001) with self assessment of func-
tional limitations in walking outdoors and walking
stairs
Reliability: The reliability coefficient Rho for object
and combination scales ranges from 0.77 to 0.91.
The intraclass correlation coefficient ranges from
0.72 to 0.90
Reliability: the FMA questionnaire showed moderate
test-retest agreement
Validity: the scores on the WHODAS II were signifi-
cantly correlated with AS oriented questionnaires
on physical functioning, disease activity, and quali-
ty of life. The WHODAS II and all AS oriented ques-
tionnaires also correlated well with all domains of the
generic health status questionnaire SF-36. The vari-
ables BASDAI and BASFI at baseline were selected
as independently associated with WHODAS II scores
after five years (adjusted r=0.48)
Responsiveness: In all variables an improvement was
seen, with moderate to large responsiveness scores.
The WHODAS II showed a comparable, moderate
responsiveness score. In the short term, a change in
WHODAS II was significantly correlated with a
change in physical functioning (BASFI and DFI), but
not with a change in disease activity or quality of life
The metric properties are
supported by the me-
thodology of the study.
The test can be self-
administered in 5 min
The metric properties of
this test support its use
in community-based
physiotherapy. The test
is self-administered in 5
min
The metric properties of
this test do not support
its use in community-
based physiotherapy
The test is useful for mea-
suring disability in
Ankylosing spondylitis
and can also be self-
administered
Lower extremity dysfunction
M
I
N
E
R
V
A

M
E
D
I
C
A
C
O
P
Y
R
I
G
H
T

THONNARD FUNCTIONAL ASSESSMENT IN PHYSIOTHERAPY

534 EUROPA MEDICOPHYSICA December 2007
TABLE II.Evidence of functional tests used in stroke patients.
Functional test Clinical utility Metric properties General appraisal
Body dimension - global
Fugl-Meyer
Type: clinical test
29
Rivermead Motor Assess-
ment (RMA)
Type: clinical test
30
Stroke Impairment As-
sessment Set (SIAS)
Type: clinical test
31
Stroke Physiotherapy In-
tervention Recording
Tool (SPIRIT)
Type: clinical test
32
Stroke Rehabilitation
Assessment of Move-
ment (STREAM)
Type: clinical test
33
Body dimension - balance
Berg Balance Scale (BBS)
Type: clinical test
34
Assess the physical impairment
following stroke, in terms of
motor performance, balance,
sensation, range of movement
and pain in 155 items
ICF dimension: body structure,
body function
Detailed assessment of gross
motor control, upper limb con-
trol and lower limb and trunk
control after stroke
ICF dimension: body function
The SIAS assesses various aspects
of impairment in hemiplegic
patients (e.g.,motor function,
sensory function, pain)
ICF dimension: body function
Define and describe the content
of physiotherapy intervention
for postural control (posture,
balance, walking) post stroke
ICF dimension: body function,
body structure, activities
Objective and quantitative evalu-
ation of the motor functioning
of individuals with stroke
among 3 subscales: upper-limb
movements, lower-limb move-
ments, and basic mobility items
ICF dimension: body function
Determine change in functional
standing balance over time
ICF dimension: body function
Reliability: the standard error of measurement (com-
bining, rater, occasion and error variance) may be
significant when only small changes in the patient's
level of motor performance are expected. However,
the overall inter-rater reliability was high (ICC=0.96)
as well as the subscore (ICC=0.85 to 0.97), with
exception of pain (ICC=0.61)
Reliability: reliability varied between questions (kap-
pa=0.33 to 0.37), but agreements between assessors
and between observed and asked performance were
always better than expected by chance
Validity: standardized measure of stroke impairment
consisting of subcategories of motor function, tone,
sensory function, ROM, pain, trunk function, visu-
ospatial function, speech, and sound side function.
The scale has been validated with the Rasch model.
The item difficulty patterns were similar for the right-
and left-sided lesion groups
Reliability: Scale reliability is usually analyzed as inter-
nal consistency or unidimensionality. The fit statis-
tics were acceptable, except for a few items.
Justifications for these discrepancies are presented in
the study. The item difficulty patterns were identical
at admission and discharge
Validity: first study in the development of a recording
system for physiotherapy interventions in the reha-
bilitation of postural control post-stroke. Content
based on literature review and validated against
expert advice
Validity: the total STREAM score was moderately to
highly associated with the score of the Barthel Index
(rho=0.67) and Fugl-Meyer motor assessment scale
(rho=0.95)
Reliability: Moderate to excellent agreement inter-rater
on scores for individual items (weighted kappa=0.55
to 0.94). Very high inter-rater reliability (ICC=0.96 for
the total score, 0.95 for upper extremity, 0.92 for
lower extremity and 0.92 for mobility subscales)
Validity: in arm flexion condition, the force platform and
accelerometer outcomes were significantly related to
BBS performance, with CP-flexion explaining 43% and
peak arm acceleration explaining 45% of the variance
in BBS scores. It appears that the ability to show acti-
vation of postural muscles in advance of focal move-
ment is associated with higher BBS scores. In quiet
stance condition, performance on the BBS was found
to be significantly related to CP-stance (r
2
=58%)
Reliability: Paired tests of difference failed to reveal any
differences in the variables over the two days
This widely used test is
hardly applicable, as it
is composed of multi-
ple sub-scales that can-
not be summed in a
total score. The test
takes 30 min to be
administered by the
therapist
The metric properties do
not support the use of
this test in community-
based physiotherapy
The metric properties of
this test support its use
in community-based
physiotherapy. The
scores are expressed on
a linear scale which
enables quantitative
comparisons of func-
tional states. The test
administered by the
therapist in approx. 10
min. Published scoring
guidelines are available
Not applicable since this
study proposes a re-
cording system for
physiotherapy inter-
ventions in postural
control of acute stroke
patients, rather than
functional evaluation
This test requires a 2-day
training, and was most-
ly validated on acute
patients. Moreover, the
total score is composed
of three sub-scales,
which is not suitable for
financing purposes
There is evidence in the
literature to support the
use of the BBS.
Established guidelines
allow the test to be
administered by the
therapist in 20 min
(to be continued)
M
I
N
E
R
V
A

M
E
D
I
C
A
C
O
P
Y
R
I
G
H
T

FUNCTIONAL ASSESSMENT IN PHYSIOTHERAPY THONNARD

Vol. 43 - No. 4 EUROPA MEDICOPHYSICA 535
TABLE II.Evidence of functional tests used in stroke patients (continued).
Functional test Clinical utility Metric properties General appraisal
Berg Balance Scale (BBS)
Type: clinical test
35
Functional balance tests
post stroke
Type: clinical test
36
Postural Control and
Balance for Stroke test
(PCBS)
Type: clinical test
37
Body dimension-gait
Rivermead Visual Gait
Assessment (RVGA)
Type: clinical test
38
Activities - global
Mobility milestones
Type: clinical test
39
Stroke Activity Scale
(SAS)
Type: clinical test
40
Determine change in functional
standing balance over time
ICF dimension: body function
Functional test of postural con-
trol post stroke
ICF dimension: body function
Assess balance in terms of pos-
tural changes, sitting balance,
and standing balance in order to
help physiotherapists in plan-
ning balance interventions and
in rehabilitation follow-up
ICF dimension: body function
The current study aims to devel-
op a procedure and assessment
form which allows practising
therapists to record relevant
aspects of the quality of gait
observed in patients with neu-
rological disease in a reliable
and valid way
ICF dimension: body function
Evaluate mobility after stroke in 4
simple functional tasks provid-
ing a quick, simple, and stan-
dardized outcome measure
ICF dimension: activities
Measure motor function at the
level of disability in stroke
patients
ICF dimension: body function,
activities
Validity: r=0.80 with gait speed, r>0.80 with Barthel
Index, r> 0.70 with Fugl-Meyer, r<0.60 with FIM
Reliability: Excellent test-retest reliability (ICC=0.97).
Excellent inter rater reliability (ICC=0.98)
Responsiveness: A difference of five to seven BBS
points is necessary to conclude with 90% certainty
that patients receiving rehabilitation following CVA
have undergone a real change in BBS performance
when assessed in a between-rater situation
Validity: poor to high: rho=0.33-0.54 with Motor
Assessment scale, rho=0.26-0.70 with Berg Balance
Scale, rho=0.32-0.61 with Rivermead Mobility Index
Reliability: Good to excellent test-retest reliability (ICC:
0.91-1). Good to excellent inter rater reliability
(ICC=0.88-1)
Reliability: the Cronbach alpha for all the items com-
bined was 0.96. The ICC values for the interrater
and intrarater reliability of the PCBS test were 0.94
and 0.96, respectively
Responsiveness: The PCBS test showed an accept-
able level of reliability and the responsiveness results
indicated a good level before 120 days, but not
between 120 and 360 days after stroke
Validity: there was a significant correlation between the
global RVGA score and the various criterion mea-
sures (r=0.53-0.79; P<0.001) and between change
in the RVGA score and change in walking time in
patients who received treatment (r=0.68; P<0.01)
Reliability: inter-rater reliability between multiple raters
was reasonable both for the global scores from the
gait assessment form (Kendalls coefficient of con-
cordance; P<0.001), and for individual items (com-
plete agreement occurred on 63.8% of all observa-
tions)
Reliability: three quarters of the physiotherapists
showed high intrarater reliability, with only 3 subjects
showing poor intrarater reliability. 28 subjects?of the
sample showed good or high test-retest reliability,
with only 3 subjects showing poor reliability. The
percentage agreement on the mobility milestones
between the physiotherapists ratings and our ratings
was very high overall. Percentage agreement
between therapists between 78.3 and 97.5 for all
items
Reliability: the internal consistency of the SAS was
estimated using Cronbach's alpha as 0.68 at time 1
and 0.69 at time 2. A high level of agreement was
observed for the total score (GCC=0.96). Inter-rater
reliability for individual items was good to very good.
A very high level of agreement was observed for
the total score (GCC>0.94)
There is evidence in the
literature to support the
use of the BBS.
Established guidelines
allow the test to be
administered by the
therapist in 20 min
The metric properties do
not support the use of
this test in community-
based physiotherapy
The metric properties do
not support the use of
this test in community-
based physiotherapy
since the criteria for
scoring are doubtful
and the resulting score
is multidimensional.
The test takes 30 min to
be completed
The metric properties of
this test support its use
in community-based
physiotherapy. The test
takes 15 min to be
administered by the
therapist
Not applicable since the
test addresses early
recovery stages after
stroke in a hospital set-
ting
Not applicable since the
test addresses early
recovery stages after
stroke in hospital set-
ting
(to be continued)
M
I
N
E
R
V
A

M
E
D
I
C
A
C
O
P
Y
R
I
G
H
T

THONNARD FUNCTIONAL ASSESSMENT IN PHYSIOTHERAPY

ment of patients in physiotherapy. All tests addressed
either the body or the activities dimension of the
patients functioning, which are the most related to the
clinical management of patients in physiotherapy.
Nevertheless, some of the most widely used tests in
rehabilitation (e.g.,the Functional Independence
Measure or the Barthel Index) were not listed, which
contrasts with a recent inquiry on outcome measure-
536 EUROPA MEDICOPHYSICA December 2007
TABLE II.Evidence of functional tests used in stroke patients (continued).
Functional test Clinical utility Metric properties General appraisal
Activities - upper limb
ABILHAND
Type: self-administered
questionnaire
41
Motor Activity Log (for
the Assessment of Arm
Use in Hemiparetic
Patients) (MAL)
Type: semi-structured
interview
42
Activities-gait
Gait speed
Type: clinical test
30
Activities mobility
New mobility scale
Type: open-ended ques-
tionnaire
43
Self-reported questionnaire on
the perceived difficulty of man-
ual activities of daily living
ICF dimension: activities
Assess the use of the paretic arm
and hand during activities of
daily living in terms of the
amount of use (AOU) and of
the quality of movement
(QOM) of the paretic arm
ICF dimension: activities
Timed walk at the patient's pre-
ferred speed, using the patien-
t's own selection of walking aid
or assistance, over a distance of
5 m, including turn and walk
back
ICF dimension: activities
Measure mobility inside and out-
side the home and in the com-
munity
ICF dimension: activities
Validity: content validated against the involvement of
the affected limb in each activity. The activities that
define the more difficult levels of the scale also tend
to require a higher involvement of the affected limb,
while the easier activities can be achieved in a move-
ment sequence that does not require the affected
limb. ABILHAND measures are significantly corre-
lated to grip strength (r=0.56), motricity (rho=0.73),
dexterity (r=0.60), and depression (rho=-0.21). The
ABILHAND questionnaire results in a valid, unidi-
mensional, person-centered measure of manual abil-
ity in everyday activities. The stability of the item-dif-
ficulty hierarchy across different patient classes fur-
ther supports the clinical application of the scale
Reliability: The overall scale precision is summarized
by a good between-patient separation reliability of
0.90 in this sample. It appears sufficient to discrim-
inate across patients and, presumably, to capture
even subtle functional changes with time
Validity: The cross-sectional construct validity of the
MAL is reasonable, but the results raise doubts about
its longitudinal construct validity
Reliability: Internal consistency was high (Chronbach's
alpha=0.88 for the AOU and 0.91 for the QOM). The
reproducibility is sufficient to detect an individual
change of at least 12% to 15% of the range of the
scale
Responsiveness: The improvement on the MAL during
the intervention was only weakly related to the
improvement on the Action Research Arm test
(rho=0.16 to 0.22)
Reliability: intercorrelations between 3 repetitions of
the test, between 5m and 10m walks, and between
test and retest were very high (r=0.95 to 0.99). For
individual patients, the test-retest variability ranged
from 0 to 40%, with 95% of the patients varying the
walking speed between tests by less than 25% of
the slowest time
Validity: content of mobility scale assessed through
open-ended questionnaire sent to 15 experienced
physiotherapists. They identified important mobili-
ty tasks and important places to access inside the
house, outside the house and in the community as
well as additional mobility abilities required for inde-
pendent mobility in the house and community
The metric properties of
this test support its use
in community-based
physiotherapy. The
scores are expressed on
a linear scale which
enables quantitative
comparisons of func-
tional states. The test
can be self-adminis-
tered in 5 min
The use of the MAL as a
primary outcome mea-
sure in clinical trials is
not recommended be-
cause there are reasons
to doubt the longitudi-
nal construct validity of
the instrument
The metric properties of
this test support its use
in community-based
physiotherapy. The test
can be administered in
5 min
Preliminary study of test
under development;
not applicable
M
I
N
E
R
V
A

M
E
D
I
C
A
C
O
P
Y
R
I
G
H
T

FUNCTIONAL ASSESSMENT IN PHYSIOTHERAPY THONNARD

ment in Physical Medicine and Rehabilitation across
Europe.
52
This observation might be explained by the
fact that the current literature review was focussed
on functional evaluation carried in the community,
where patients are able to live at home, and therefore
located at the higher range of measurement of such
tests. While the tests mostly used in hospital rehabil-
itation are pretty well documented, functional assess-
ment in the community are less standardised.
Moreover, this literature review considered only papers
published since 1990. Consequently, some standard-
ised tests were not reported because they were vali-
dated before 1990. This is particularly the case for
some functional tests in the body dimension like the
nine-hole peg test or the Purdue pegboard test.
The body dimension was mostly addressed in terms
of body function, except in lower back pain patients
or in other spinal disorders that primarily relied on test-
ing body structures through skin palpation. The activ-
ities dimension was generally addressed with clini-
cal tests performed by the physical therapist or by a
self-reported questionnaire relating to the patients
perceived disability in daily life activities. Clinical tests
typically address movement quality or a patients per-
formance in standardized activities, and are considered
more reliable than questionnaires. Nevertheless, ques-
tionnaires present several advantages over clinical
test. First, they address the perception of disability in
activities as they are actually realized by the patient in
their own environment. Second, they capture an aver-
age perception of disability over a longer period of
time (one up to several weeks) and are probably
more representative of the patients disability than
the performance of standardized activities in a con-
sultation room. Third, they are very inexpensive and
can be completed by the patient in the waiting room
in order to reduce the burden of work of the physi-
cal therapist. On the other hand, it can be discussed
Vol. 43 - No. 4 EUROPA MEDICOPHYSICA 537
TABLE III.Evidence of functional tests used in elderly patients.
Functional test Clinical utility Metric properties General appraisal
6-minute Walk Test
Type: clinical test
44
8-item Physical Perfor-
mance Test (PPT-8)
Type: clinical test
44
Functional Gait Assess-
ment (FGA)
Type: clinical test
45
General Motor Function
Assessment Scale (GMF)
Type: clinical test
46
Measure of exercise capacity and
endurance also used as an out-
come measure
ICF dimension: activities
Evaluate the ability to use the
upper and lower extremities in
activities of daily living
ICF dimension: activities
Assess postural stability during
gait tasks in the older adult
(greater than 60 years of age)
at risk for falling
ICF dimension: activities
Compound assessment of three
components (dependence, pain
and insecurity) of daily living
activities among older rehabili-
tation patients
ICF dimension: activities
Reliability: intraclass correlation coefficients for test-
retest reliability were 0.93
Responsiveness: There was no change in 6-minute
walk test distance in the intervention group when
compared with the control group
Reliability: good internal consistency (Cronbach's
alpha=0.78). Intraclass correlation coefficients for
test-retest reliability was 0.88. The intraclass corre-
lation coefficient for interrater reliability was 0.96
Responsiveness: The responsiveness index was 0.8,
indicating a significant difference of effect between
intervention and control group
Validity: Poor to moderate correlation with balance
measurements (rho=0.11 to 0.67)
Reliability: Good internal consistency (Cronbach's
alpha=0.79). Good inter rater reliability (ICC=0.74).
Good inter rater reliability (ICC=0.86)
Validity: Principal components factor analysis demon-
strated individual FGA item loading across 3 extract-
ed factors that may represent separate domains of
performance on the total battery
Reliability: Analysis of reliability showed overall high
values of percentage agreement (PA=0.70) and of
the rank-order agreement coefficient (ra=0.82), and
low degrees of systematic disagreement
The test was unable to
measure change in per-
formance expected
with a functional train-
ing intervention. An
indoor 32-meter course
is also required. Its
applicability is limited
The metric properties are
supported by the me-
thodology of the study.
A published test proto-
col allows the test to be
administered by the
therapist
This test has been devel-
oped for a very specif-
ic group of patients.
The results of this study
should be interpreted
cautiously since the
sample size was very
small (6 patients)
The metric properties do
not support the use of
this test in community-
based physiotherapy
M
I
N
E
R
V
A

M
E
D
I
C
A
C
O
P
Y
R
I
G
H
T

TABLE V.Selected tests according to the dimensions of the International Classification of Functioning, Disability and Health (ICF).
Body Activities Participation
Musculoskeletal disorders
Fingertip-to-floor test Low back SF-36 physical functioning
Modified-Modified Schber Test (MMST) Lower Extremity Functional Scale (LEFS)
Oswestry
Questionnaire Rising and Sitting Down (QR&S)
Roland-Morris
Shuttle walking test
World Health Organisation Disability Assessment Sickness Impact Profile (SIP)
Schedule II (WHODAS II)
Stroke
Berg Balance Scale (BBS) ABILHAND
Rivermead Visual Gait Assessment (RVGA) Gait speed
Stroke Impairment Assessment Set (SIAS)
Elderly
8-item Physical Performance Test (PPT-8)
Brain injury
Step length and step width measurement
THONNARD FUNCTIONAL ASSESSMENT IN PHYSIOTHERAPY
538 EUROPA MEDICOPHYSICA December 2007
TABLE IV.Evidence of functional tests used in patients with brain injury.
Functional test Clinical utility Metric properties General appraisal
Clinical Outcomes Varia-
ble Scale (COVS)
Type: clinical test
47
Step length and step
width measurement
Type: clinical test
48
Taxonomy of patient
goals in acquired brain
injury comprising 21
categories within five
domains, utilizing 125
descriptors
Type: classification
49
Evaluation of functional ability
following traumatic brain injury
in a large range of motor tasks
retrained by physiotherapists
including a measure for transfer
ability to and from bed and
floor surfaces as well as wheel-
chair skill
ICF dimension: body function,
activities
Walking performance after trau-
matic brain injury assessed by
step length and step width mea-
sures
ICF dimension: body function
Classification of client goals in
community-based acquired
brain injury rehabilitation allow-
ing treatment focus and changes
in service delivery to be inves-
tigated
ICF dimension: activities, partici-
pation
Reliability: the intra-class correlation coefficients (ICC)
were very high for both inter-tester reliability
(ICC>0.97 for total COVS scores, ICC>0.93 for indi-
vidual COVS items) and intra-tester reliability
(ICC>0.97)
Validity: concurrent validity was excellent, with cor-
relations between the procedures ranging from 0.93
to 1.00
Reliability: the inter-rater reliability of step length and
width measurements was very high, with intraclass
correlation coefficients between 0.94 and 1.00, for
both procedures
Reliability: the taxonomy demonstrated good inter-
rater consistency and was able to discriminate
between similar but related data sets comprising
goal statements. Out of the 140 goal statements 128
(92%) were placed in the same category by at least
three of the four, and there was full agreement
between all four on 91 of the goal statements (65%).
This indicated that the taxonomy and descriptor
statements had been refined to a stage where there
was significant inter-rater consistency
The metric properties do
not support the use of
this test in community-
based physiotherapy
Potentially interesting test
for neurological pa-
tients, but its metric
properties should be
interpreted cautiously
since the sample size
was small (20 patients).
A 14-meter walk course
is required
Not applicable since this
study proposes a clas-
sification of patient
goals, rather than func-
tional evaluatio
M
I
N
E
R
V
A

M
E
D
I
C
A
C
O
P
Y
R
I
G
H
T

whether it is feasible to base reimbursement on self-

reported questionnaires.
The psychometric qualities of the test were report-
ed as presented in each study. These statistics can be
interpreted by the reader in order to make an optimal
choice of functional tests for their specific purpose..
Some precautions, however, must be kept in mind in
the comparison of statistical indices. For instance,
although validity was generally assessed by correla-
tion analyses, it is well known that correlation coef-
ficients are sensitive to the range and distribution of
the values correlated.
53
The intensity of the reported
correlations is therefore influenced by the difference
in sampling strategies used between studies.
Moreover, different inclusion criteria were used
between studies (e.g.,elderly patients with or without
a previous stroke). The selection bias should be con-
sidered from a clinical perspective in comparing sta-
tistical indices since it influences the relation between
the functional test and the gold standard.
Responsiveness was also assessed with different
types of indicators. Responsiveness is generally defined
as the ability of a test to elicit clinically meaningful
change. Nevertheless, all evaluations with functional
tests are accompanied by a certain amount of mea-
surement error, and some of the statistics used to
assess responsiveness do not take the error of mea-
surement into account (e.g.,percentage of change
score). Thus, when assessing change with such sta-
tistics, scores obtained at both occasions and regard-
ed as significantly different might very well be with-
in the error of measurement and therefore the possi-
bility that the change is due to chance cannot be
excluded. Moreover, the reviewed functional tests
used different levels of measurement.
54
Though the
large majority of results were reported at the ordinal
level (i.e.,allowing for ranking comparisons), some
tests reported their results on nominal scales (i.e.,only
allowing for equivalence comparison) or on interval
or ratio scale (i.e.,allowing difference, or change score,
to be computed on a linear scale). Misinferences due
to linear interpretation of data at the nominal or ordi-
nal level of measurement has been largely reported in
the literature.
55
Conclusions and perspectives
Functional evaluation is part of the process of reha-
bilitation.
1
This process involves identifying the prob-
lems and needs of individuals, defining therapy goals,
planning and implementing interventions and assess-
ing the effect of interventions using measurements
of relevant variables.
56
The literature review present-
ed in this report provides a selection of functional
tests in the major diagnostic groups treated in com-
munity physiotherapy. These functional tests essen-
tially cover the body dimension and the activities
dimension of the ICF, which are the dimensions that
are most pertinent to the current practice of physio-
therapy.
A selection of tests in each dimension of the ICF is
presented in Table V according to their metric prop-
erties and quality appraisal for the four diagnostic
groups emerging from this review. The metric prop-
erties include published evidence of reliability, valid-
ity and responsiveness, but also the unidimensional-
ity and linearity of each scale. The latter properties are
requisite for making quantitative comparisons of func-
tional status over time. Quality appraisal includes a
general evaluation of the methodology of each study,
and the practical applicability of each test in com-
munity physiotherapy (i.e., evaluation time, required
equipment). It is important to emphasize that time
and quality of evaluation are not strictly related.
Indeed, very short, well calibrated scales can be much
more efficient than time consuming tests. The body
and activities dimensions of the ICF are the most well-
represented. Musculoskeletal disorders are the diag-
nostic groups for which the largest number of high
quality tests have been identified. The body dimension
in musculoskeletal disorders is mostly evaluated with
active mobility tests. Although these tests are widely
used, one of their major drawbacks is that they mea-
sure a maximal performance, i.e., maximal mobility,
which can be hindered by pain or other factors.
57
The
activities dimension in musculoskeletal disorders is
the most represented among the four diagnostic
groups. Most tests used in this domain are widely
used, validated questionnaires that can be easily and
inexpensively applied in community physiotherapy.
The body dimension in stroke patients is addressed
either with global tests and/or tests of specific func-
tions (e.g., balance). The activities dimension is eval-
uated either with upper limb or with lower limb spe-
cific tests; no global activity instrument was identi-
fied in stroke community rehabilitation. Although few
tests are available for elderly and brain-injured patients,
one test in each diagnostic group presented enough
quality to be retained in this selection. The summary
FUNCTIONAL ASSESSMENT IN PHYSIOTHERAPY THONNARD
Vol. 43 - No. 4 EUROPA MEDICOPHYSICA 539
M
I
N
E
R
V
A

M
E
D
I
C
A
C
O
P
Y
R
I
G
H
T

15. Vincent-Smith B, Gibbons P. Inter-examiner and intra-examiner reli-

ability of the standing flexion test. Man Ther 1999;4:87-93.
16. Garratt AM, Klaber Moffett J, Farrin AJ. Responsiveness of gener-
ic and specific measures of health outcome in low back pain.
Spine 2001;26:71-7.
17. Stratford PW, Binkley J, Solomon P, Gill C, Finch E. Assessing
change over time in patients with low back pain. Phys Ther
1994;74:528-33.
18. Davidson M, Keating JL, Eyres S. A low back-specific version of the
SF-36 Physical Functioning scale. Spine 2004;29:586-94.
19. Stratford PW, Binkley J, Solomon P, Finch E, Gill C, Moreland J.
Defining the minimum level of detectable change for the Roland-
Morris questionnaire. Phys Ther 1996;76:359-65.
20. Stratford P, Solomon P, Binkley J, Finch E, Gill C. Sensitivity of
Sickness Impact Profile items to measure change over time in a low-
back pain patient group. Spine 1993;18:1723-7.
21. Perret C, Poiraudeau S, Fermanian J, Colau MM, Benhamou MA,
Revel M. Validity, reliability, and responsiveness of the fingertip-
to-floor test. Arch Phys Med Rehabil 2001;82:1566-70.
22. Smedmark V, Wallin M, Arvidsson I. Inter-examiner reliability in
assessing passive intervertebral motion of the cervical spine. Man
Ther 2000;5:97-101.
23. Kvale A, Skouen JS, Ljunggren AE. Sensitivity to change and
responsiveness of the global physiotherapy examination (GPE-
52) in patients with long-lasting musculoskeletal pain. Phys Ther
2005;85:712-26.
24. Jensen MP, Turner JA, Romano JM, Fisher LD. Comparative relia-
bility and validity of chronic pain intensity measures. Pain
1999;83:157-62.
25. Binkley JM, Stratford PW, Lott SA, Riddle DL. The Lower Extremity
Functional Scale (LEFS): scale development, measurement prop-
erties, and clinical application. North American Orthopaedic
Rehabilitation Research Network. Phys Ther 1999;79:371-83.
26. Roorda LD, Roebroeck ME, Lankhorst GJ, van Tilburg T, Bouter LM.
Measuring functional limitations in rising and sitting down: devel-
opment of a questionnaire. Arch Phys Med Rehabil 1996;77:663-
9.
27. Callaghan BG, Sockalingam S, Treweek SP, Condie ME. A post-dis-
charge functional outcome measure for lower limb amputees:
test-retest reliability with trans-tibial amputees. Prosthet Orthot
Int 2002;26:113-9.
28. van Tubergen A, Landewe R, Heuft-Dorenbosch L, Spoorenberg
A, van der Heijde D, van der Tempel H et al. Assessment of dis-
ability with the World Health Organisation Disability Assessment
Schedule II in patients with ankylosing spondylitis. Ann Rheum Dis
2003;62:140-5.
29. Sanford J, Moreland J, Swanson LR, Stratford PW, Gowland C.
Reliability of the Fugl-Meyer assessment for testing motor perfor-
mance in patients following stroke. Phys Ther 1993;73:447-54.
30. Collen FM, Wade DT, Bradshaw CM. Mobility after stroke: relia-
bility of measures of impairment and disability. Int Disabil Stud
1990;12:6-9.
31. Tsuji T, Liu M, Sonoda S, Domen K, Chino N. The stroke impair-
ment assessment set: its internal consistency and predictive valid-
ity. Arch Phys Med Rehabil 2000;81:863-8.
32. Tyson SF, Selley A. The development of the Stroke Physiotherapy
Intervention Recording Tool (SPIRIT). Disabil Rehabil 2004;26:1184-
8.
33. Wang CH, Hsieh CL, Dai MH, Chen CH, Lai YF. Inter-rater relia-
bility and validity of the stroke rehabilitation assessment of move-
ment (stream) instrument. J Rehabil Med 2002;34:20-4.
34. Stevenson TJ, Garland SJ. Standing balance during internally pro-
duced perturbations in subjects with hemiplegia: validation of the
balance scale. Arch Phys Med Rehabil 1996;77:656-62.
35. Stevenson TJ. Detecting change in patients with stroke using the
Berg Balance Scale. Aust J Physiother 2001;47:29-38.
36. Tyson SF, DeSouza LH. Reliability and validity of functional bal-
ance tests post stroke. Clin Rehabil 2004;18:916-23.
37. Pyoria O, Talvitie U, Villberg J. The reliability, distribution, and
table presented here indicates that 1) high quality
functional tests were found for the main diagnostic
groups treated in community physiotherapy (e.g.,mus-
culoskeletal disorders, stroke); 2) the functional tests
identified do not cover all dimensions of the ICF for
all diagnostic groups, 3) in some dimensions, vari-
ous valid scales are proposed, hence emphasising the
need for a standard in each diagnostic group.
This literature review has identified a selection of
valid tests that could be used as a basis for the stan-
dardisation of the functional evaluation of the major
diagnostic groups treated in community physiothera-
py. This means that standards are available for report-
ing and following the evolution of patients.
Nevertheless, no attempt in using functional outcomes
as a rationale for financing physiotherapy was found
to date in the current literature review.
References
1. Wade DT. Community rehabilitation, or rehabilitation in the com-
munity? Disabil Rehabil 2003;25:875-81.
2. Streiner DL, Norman GR. Health Measurement Scales: A practical
guide to their development and use. 2
nd
edition. Oxford: Oxford
University Press; 1995.
3. Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater
reliability. Psychol Bulletin 1979;86:420-8.
4. Cohen J. A coefficient of agreement for nominal scales. Psycho
Meas 1960;20:37-46.
5. Soeken KL, Prescott PA. Issues in the use of kapp to estimate reli-
ability. Medical Care 1986;24:733-41.
6. Maclure M, Willet WC. Misinterpretation and misuse of the kappa
statistic. Am J Epidemiol 1987;126:161-9.
7. Kazis LE, Anderson JJ, Meenan RF. Effect sizes for interpreting
changes in health status. Med Care 1989;27S:178-89.
8. Guyatt GH, Walter S, Norman G. Measuring change over time:
assessing the usefulness of evaluative instruments. J Chronic Dis
1987;40:171-8.
9. Petersen T, Olsen S, Laslett M, Thorsen H, Manniche C, Ekdahl C
et al. Inter-tester reliability of a new diagnostic classification sys-
tem for patients with non-specific low back pain. Aust J Physiother
2004;50:85-94.
10. Fritz JM, George S. The use of a classification approach to identi-
fy subgroups of patients with acute low back pain. Interrater reli-
ability and short-term treatment outcomes. Spine 2000;25:106-14.
11. Van Dillen LR, Sahrmann SA, Norton BJ, Caldwell CA, Fleming DA,
McDonnell MK et al. Reliability of physical examination items
used for classification of patients with low back pain. Phys Ther
1998;78:979-88.
12. Tousignant M, Poulin L, Marchand S, Viau A, Place C. The Modified-
Modified Schober Test for range of motion assessment of lumbar
flexion in patients with low back pain: a study of criterion valid-
ity, intra- and inter-rater reliability and minimum metrically
detectable change. Disabil Rehabil 2005;27:553-9.
13. Fritz JM, Piva SR. Physical impairment index: reliability, validity, and
responsiveness in patients with acute low back pain. Spine
2003;28:1189-94.
14. Taylor S, Frost H, Taylor A, Barker K. Reliability and responsive-
ness of the shuttle walking test in patients with chronic low back
pain. Physiother Res Int 2001;6:170-8.
540 EUROPA MEDICOPHYSICA December 2007
FUNCTIONAL ASSESSMENT IN PHYSIOTHERAPY THONNARD
M
I
N
E
R
V
A

M
E
D
I
C
A
C
O
P
Y
R
I
G
H
T

Outcomes Variable Scale: a reliability study. Aust J Physiother

2002;48:35-9.
48. van Loo MA, Moseley AM, Bosman JM, de Bie RA, Hassett L. Inter-
rater reliability and concurrent validity of step length and step
width measurement after traumatic brain injury. Disabil Rehabil
2003;25:1195-200.
49. Kuipers P, Foster M, Carlson G, Moy J. Classifying client goals in
community-based ABI rehabilitation: a taxonomy for profiling ser-
vice delivery and conceptualizing outcomes. Disabil Rehabil
2003;25:154-62.
50. Dekker J. Application of the ICIDH in survey research on reha-
bilitation: the emergence of the functional diagnosis. Disabil
Rehabil 1995;17:195-201.
51. Dekker J, van Baar ME, Curfs EC, Kerssens JJ. Diagnosis and treat-
ment in physical therapy: an investigation of their relationship. Phys
Ther 1993;73:568-77.
52. Haigh R, Tennant A, Biering-Sorensen F, Grimby G, Marincek C,
Phillips S et al. The use of outcome measures in physical medicine
and rehabilitation within Europe. J Rehabil Med 2001;33:273-8.
53. Bland JM, Altman DG. Statistical methods for assessing agreement
between two methods of clinical measurement. Lancet 1986;1:307-
10.
54. Stevens SS. On the theory of scales of measurement. Science
1946;103:677-80.
55. Merbitz C, Morris J, Grip JC. Ordinal scales and foundations of mis-
inference. Arch Phys Med Rehabil 1989;70:308-12.
56. Steiner WA, Ryser L, Huber E, Uebelhart D, Aeschlimann A, Stucki
G. Use of the ICF model as a clinical problem-solving tool in
physical therapy and rehabilitation medicine. Phys Ther
2002;82:1098-107.
57. Watson P. Non-physiological determinants of physical perfor-
mance in musculoskeletal pain. In: Mitchell M, editor. Pain 1999,
an updated review refresher course syllabus; 9th World Congress
of Pain; Seattle: IASP press; 1999. p. 153-7.
Vol. 43 - No. 4 EUROPA MEDICOPHYSICA 541
FUNCTIONAL ASSESSMENT IN PHYSIOTHERAPY THONNARD
responsiveness of the Postural Control and Balance for Stroke
Test. Arch Phys Med Rehabil 2005;86:296-302.
38. Lord SE, Halligan PW, Wade DT. Visual gait analysis: the devel-
opment of a clinical assessment and scale. Clin Rehabil
1998;12:107-19.
39. Baer GD, Smith MT, Rowe PJ, Masterton L. Establishing the relia-
bility of mobility milestones as an outcome measure for stroke. Arch
Phys Med Rehabil 2003;84:977-81.
40. Horgan NF, Finn AM, ORegan M, Cunningham CJ. A new stroke
activity scale-results of a reliability study. Disabil Rehabil
2003;25:277-85.
41. Penta M, Tesio L, Arnould C, Zancan A, Thonnard JL. The ABIL-
HAND questionnaire as a measure of manual ability in chronic
stroke patients: Rasch-based validation and relationship to upper
limb impairment. Stroke 2001;32:1627-34.
42. van der Lee JH, Beckerman H, Knol DL, de Vet HC, Bouter LM.
Clinimetric properties of the motor activity log for the assessment
of arm use in hemiparetic patients. Stroke 2004;35:1410-4.
43. Stanko E, Goldie P, Nayler M. Development of a new mobility scale
for people living in the community after stroke: content validity.
Aust J Physiother 2001;47:201-8.
44. King MB, Judge JO, Whipple R, Wolfson L. Reliability and respon-
siveness of two physical performance measures examined in the
context of a functional training intervention. Phys Ther 2000;80:8-
16.
45. Wrisley DM, Marchetti GF, Kuharsky DK, Whitney SL. Reliability,
internal consistency, and validity of data obtained with the func-
tional gait assessment. Phys Ther 2004;84:906-18.
46. Aberg AC, Lindmark B, Lithell H. Development and reliability of
the General Motor Function Assessment Scale (GMF)a perfor-
mance-based measure of function-related dependence, pain and
insecurity. Disabil Rehabil 2003;25:462-72.
47. Low Choy N, Kuys S, Richards M, Isles R. Measurement of func-
tional ability following traumatic brain injury using the Clinical