You are on page 1of 10

Archives of Physical Medicine and Rehabilitation

journal homepage: www.archives-pmr.org


Archives of Physical Medicine and Rehabilitation 2014;-:-------

REVIEW ARTICLE

Scales to Assess Gross Motor Function in Stroke


Patients: A Systematic Review
Mara Dolores Gor-Garca-Fogeda, BcPT, MSc, Francisco Molina-Rueda, BcPT, MSc, PhD,
Alicia Cuesta-Gomez, BcPT, MSc, Mara Carratala-Tejada, BcPT, MSc,
Isabel M. Alguacil-Diego, MD, PhD, Juan Carlos Miangolarra-Page, MD, PhD
From the Faculty of Health Sciences, Motion Analysis, Ergonomics, Biomechanics and Motor Control Laboratory (LAMBECOM), Rey Juan Carlos
University, Madrid, Spain.

Abstract
Objective: To assess the clinical and psychometric properties of stroke motor assessment scales.
Data Sources: The databases consulted for the literature research were MEDLINE, PEDro, ISI Web of Knowledge, and Cumulative Index to
Nursing and Allied Health (CINAHL). The search was carried out between March 2011 and January 2014.
Study Selection: Studies that describe and validate a measurement scale designed to assess gross motor function in stroke. The articles were
classified according to the levels of evidence and grades of recommendation for diagnosis studies of the Oxford Center for Evidence-Based Medicine.
Data Extraction: General characteristics of the studies, including number of patients, motor function assessment scales analyzed, and their
psychometric properties, were collected.
Data Synthesis: After the literature search, 19 articles were included in this review; 32 articles were excluded for not meeting the inclusion
criteria. Four of the 19 articles studied the Motor Assessment Scale, 5 the Fugl-Meyer Assessment, 3 investigated the Sodring Motor Evaluation
for Stroke Patients, 4 the Stroke Rehabilitation Assessment of Movement, 2 were about the Motricity Index, and 2 about the Rivermead Motor
Assessment. All of them were classified as level 2b according to the levels of evidence and grades of recommendation.
Conclusions: All the scales compiled in this review have been shown to be useful both in clinical practice and in terms of research. The most
suitable scales to be used in the clinical field would be the short versions of the Fugl-Meyer Assessment and the Stroke Rehabilitation Assessment
of Movement. A real consensus about the measurement of gross motor function in patients with stroke is not available in the recent literature.
Archives of Physical Medicine and Rehabilitation 2014;-:------ 2014 by the American Congress of Rehabilitation Medicine

Stroke is one of the main causes of death worldwide, has the


highest economic cost, and is the leading cause of disability.1-4
Stroke survivors can suffer several neurologic deficits or impairments, such as hemiparesis, communication disorders, cognitive
deficits, or disorders in visual-spatial perception.2,5 After
completing standard rehabilitation, approximately 50% to 60% of
the patients with stroke still experience some degree of gross
motor impairment.5 Gross motor function refers to global body

This study was carried out as part of the research project Hybrid NeuroProsthetic and
NeuroRobotic Devices for Functional Compensation and Rehabilitation of Motor Disorders
(HYPER) within the program CONSOLIDER-Ingenio 2010 and the 6th Spanish National Plan for
Scientific Research, Development and Technological Innovation 2008-2011.
No commercial party having a direct financial interest in the results of the research supporting
this article has conferred or will confer a benefit on the authors or on any organization with which
the authors are associated.

movements executed by the coordinated action of several muscle


groups. Recovering gross motor function after a stroke is indispensable for the reestablishment of functionality, including such
basic tasks as transferring from sit to stand or walking.
Analysis of the available scales in this field is essential to
establish an adequate patient evaluation and to compare results of
the rehabilitation treatment for this growing group of patients.
However, in daily clinical practice, it is not always easy to choose
the most appropriate assessment tools. There are several scales for
patients with stroke that have been used inconsistently in previous
studies,6 and their application criteria are not well defined.
Scales that measure functionality and activities of daily living
provide objective information about the autonomy and independence
of the patient. However, they do not assess motor recovery of the
limbs, because they do not restrict compensations.7

0003-9993/14/$36 - see front matter 2014 by the American Congress of Rehabilitation Medicine
http://dx.doi.org/10.1016/j.apmr.2014.02.013

M.D. Gor-Garca-Fogeda et al

2
For a scale to be considered suitable for clinical use and
research, its psychometric properties need to be established.8,9
When reviewing the existing literature about clinical scales that
assess motor function in patients with stroke, we find that even
though many clinical trials have been conducted to investigate
psychometric properties, very few systematic reviews regarding
this subject have been written so far. The only systematic review
found in the literature was written by Poole and Whitney.10 They
investigated 11 motor function scales. According to their findings,
the Motor Assessment Scale (MAS) could be the most useful
because of its fast administration. However, the Fugl-Meyer
Assessment (FMA) has more sound psychometric evidence than
do the other scales reviewed. This systematic review omitted some
scales that would have been useful to include, such as the Sodring
Motor Evaluation for Stroke Patients (SMES) or the short versions
of FMA and Stroke Rehabilitation Assessment of Movement (STREAM).
Regarding studies that include only upper limb motor function
scales, we find the review written by Croarkin et al,11 which, for
the first time, studies in depth the use of these types of scales
among patients with stroke, rating the tests relative to their psychometric properties (interrater reliability, test-retest reliability,
convergent validity or concurrent validity, and predictive validity),
and considering the Nine Hole Peg Test the only one for which 3
of these properties have been proven, whereas for the rest of the
scales only 2 or 1 of these properties has been proven.11
We also find less specific studies, such as the one by GellezLeman et al12 that aggregates a large number of scales that assess
function after stroke. The article briefly explains each of the tests,
without in-depth discussion of the study of their psychometric
properties or further investigation into the advantages and disadvantages of their use.12
Considering the large number of clinical trials that study the
use of motor function scales among patients with stroke, and
regarding the lack of systematic reviews of this subject, we proposed a systematic review whose primary goals were (1) to
compile all scales available in the scientific literature that assess
gross motor function in stroke, (2) to establish which specific
aspects of gross motor function they assess, and (3) to study their
psychometric properties.

Methods
Search strategy
A complete literature search on MEDLINE, PEDro, ISI Web of
Knowledge, Cumulative Index to Nursing and Allied Health

List of abbreviations:
BI
FAI
FMA
ICC
MAS
MI
MMAS
RMA
S-FM
SMES
SRM
STREAM
S-STREAM

Barthel Index
Franchay Activity Index
Fugl-Meyer Assessment
intraclass correlation coefficient
Motor Assessment Scale
Motricity Index
modified MAS
Rivermead Motor Assessment
short version of the FMA
Sodring Motor Evaluation for Stroke Patients
standardized response mean
Stroke Rehabilitation Assessment of Movement
15-item simplified version of STREAM

(CINAHL), and the metasearch engine Tripdatabase without limits


on publication data was carried out. The searches were accomplished between March 2011 and January 2014. Only full-text
articles published in English, French, or Spanish were selected.
The combinations of keywords used are described in detail in
appendix 1.

Study selection
The included articles were selected according to the following
inclusion criteria: (1) study describes a measurement scale
designed to assess gross motor function; (2) scale assesses gross
motor function; scale may also measure other aspects of functioning, but it must include at least 2 gross motor function items;
(3) study reports the main psychometric properties8,9 (at least 1
reliability, validity, or sensitivity coefficient); and (4) study includes adults diagnosed with cerebrovascular accident, without
distinction between acute, subacute, and chronic phases in the
stage of stroke.
This systematic review excluded articles according to the
following exclusion criteria: (1) study only reports other psychometric properties such as hierarchical order or the minimum
detectable difference and (2) scale only assesses upper limb
function, because they were mostly dedicated to evaluating fine
motor function.

Data collection
General characteristics of the studies, including number of patients, stage of stroke (acute, subacute, chronic), motor function
assessment scales analyzed, and their psychometric properties,
were collected.
Two of the authors (physiotherapists with >5y of clinical
experience in neurologic rehabilitation and with psychometric
experience: M.D.G.-F. and F.M.-R.) carried out 2 independent
screenings of the titles and abstracts obtained from the electronic
research. The reviewers decided which articles could potentially
meet the inclusion criteria, and the articles selected by each
reviewer were compared. Those articles on which there was no
consensus were discussed by the reviewers. For studies that met
the inclusion criteria, the full-text articles were obtained. The
reviewers executed a new screening for all articles to confirm their
relevance, with any disagreement about the selection of the articles settled by discussion between the reviewers. The reviewers
were in complete agreement.
The articles were classified according to the levels of evidence
and grades of recommendation for diagnosis studies established
by the Oxford Center for Evidence-Based Medicine (table 1).13
The levels of evidence include 9 different items representing
different designs of diagnosis studies. These items are classified
from level 1 (maximum level of evidence) to level 5 (minimum
level of evidence). In addition, each item is related to a grade of
recommendation. These grades are established from grade A
(consistent level 1 studies, maximum grade of recommendation) to
grade D (level 5 evidence or troublingly inconsistent or inconclusive studies of any level, minimum grade of recommendation).

Results
A total of 60 articles were found in the literature search. Once the
abstracts were read, 51 of them were selected for further review
and critical reading. Also, the references of these articles were
reviewed to avoid loss of information that was not found in the
www.archives-pmr.org

Gross motor function in stroke


Table 1 Oxford Center for Evidence-based MedicinedLevels of
evidence and grades of recommendation
Level

Diagnosis

1a

SR (with homogeneity) of level 1 diagnostic studies; CDR


with 1b studies from different clinical centers
1b
Validating cohort study with good reference standards; or
CDR tested within 1 clinical center
Independent blind comparison of an appropriate
spectrum of consecutive patients, all of whom have
undergone both the diagnostic test and the reference
standard; or a clinical decision rule not validated on a
second set of patients
1c
Absolute SpPins and SnNouts (An Absolute SpPin is a
diagnostic finding whose specificity is so high that a
positive result rules in the diagnosis. An Absolute
SnNout is a diagnostic finding whose sensitivity is so
high that a negative result rules out the diagnosis)
2a
SR (with homogeneity) of level >2 diagnostic studies
2b
Exploratory cohort study with good reference standards;
CDR after derivation, or validated only on split-sample
or databases
Any of: (1) independent blind or objective comparison;
(2) study performed in a set of nonconsecutive
patients, or confined to a narrow spectrum of study
individuals (or both) all of whom have undergone both
the diagnostic test and the reference standard; (3) a
diagnostic clinical rule not validated in a test set
3a
SR (with homogeneity) of 3b and better studies
3b
Nonconsecutive study; or without consistently applied
reference standards
4
Case-control study, poor or nonindependent reference
standard
5
Expert opinion without explicit critical appraisal, or based
on physiology, bench research, or first principles
Grades of recommendation
A
Consistent level 1 studies
B
Consistent level 2 or 3 studies or extrapolations from level
1 studies
C
Level 4 studies or extrapolations from level 2 or 3 studies
D
Level 5 evidence or troublingly inconsistent or
inconclusive studies of any level
CDR, critical design review; SR, systematic review.

bibliographic search, without obtaining any results. In these articles, we identified 18 motor function assessment scales for patients with stroke, from which 6 met the inclusion criteria and
were included in the review. Finally, 19 articles were included in
this review14-32 and 32 were excluded (fig 1).33-64
The excluded tests are described in table 2, whereas the scales
that were included are as follows: MAS, FMA, SMES, STREAM,
Rivermead Motor Assessment (RMA), and Motricity Index (MI).
Four of the 19 articles studied the MAS,14-17 5 the FMA,18-22
2 investigated the SMES,23,24 5 the STREAM,22,25-28 2 were
about the RMA,29,30 and 2 about the MI.31,32
All the 19 articles were classified as level 2b according to the
levels of evidence and grades of recommendation (see table 1)
because they were all exploratory cohort studies (systematic reviews were not included) with good reference standards consistently applied, but they could not be considered high quality
www.archives-pmr.org

3
because they presented possibilities of bias due to a lack of a
thorough follow-up of the sample, poorly defined inclusion
criteria, or a failure to specify whether they were blind studies or
randomly assigned.
The scales whose psychometric properties are investigated in
the largest number of studies are the MAS and STREAM, with
STREAM being the one with the soundest psychometric evidence.
The psychometric properties investigated in the articles were
predictive validity; content, construct, criterion, and concurrent
validity; interrater and test-retest reliability; and internal consistency and sensitivity. Reliability is the most frequently studied
psychometric property: all 6 scales demonstrated good to excellent
reliability (table 3).

Synthesis of the results


Motor assessment scale
The MAS was designed by Carr et al14 in 1985 to evaluate motor
function in patients with stroke. It includes 8 different items
representing 8 areas of motor function and 1 item related to
muscle tone on the affected side, and each item is scored on a 7point scale from 0 to 6. Although 2 of the items assess hand motor
function, the primary focus of the MAS is gross motor function.14
To evaluate the interrater reliability, Carr14 selected a sample
of 5 patients with stroke, and 1 of the authors assessed them using
the MAS, videotaping the assessment sessions. Later, 20 therapists, who were previously trained in the use of the scale, were
asked to view the videotape and use the MAS to score each patients performance. Then, the scores obtained were compared
with those from the first assessment. To evaluate the test-retest
reliability, a sample of 15 patients was scored by one of the authors14 on 2 occasions separated by a 4-week interval. They found
a high percentage agreement between the raters (87% agreement),
as well as a high Pearson correlation coefficient between the
scores obtained by different raters (.95) and those obtained by
the same rater on different occasions (.98), which suggests that the
MAS is a reliable tool to assess motor function.14
In 1988, Poole and Whitney15 conducted a study to establish
the concurrent validity of the MAS by comparing it with the FMA,
and to establish its interrater reliability. A sample of 30 subjects
with stroke was tested with the MAS and the FMA on consecutive
days. To establish interrater reliability, 2 different evaluators
observed the MAS testing and scored each subject independently.
The correlation between the total score on the MAS and that on
the FMA was .88, resulting in strong and significant correlations in
all tests except for sitting balance (due to differences in the nature
of the item on the 2 tests). The interrater correlation coefficient for
the total score on the MAS was .99, with all coefficients scoring
high results except for muscle tone, which suggests that physical
contact is necessary to assess it. This work concludes that the
MAS can be used instead of the FMA to assess motor recovery
after stroke because the former has several advantages when
compared with the latter: it requires little time to administer, and it
provides further information about the functional capacities of
the patient.15
Also, in 1988, Loewen and Anderson16 modified the MAS to
improve its sensitivity. They created the modified MAS (MMAS)
to assess changes in patient status more effectively. In addition,
they investigated its interrater and test-retest reliability in a sample
of 7 patients with stroke.16 Fourteen therapists, trained in the use
of the MMAS, scored the patients by watching videotaped MMAS
assessments on 2 occasions, occurring 1 month apart. High

M.D. Gor-Garca-Fogeda et al

Fig 1

Summary of the selection process.

interrater reliability was established when comparing the results


obtained by each therapist, showing a high Spearman correlation
coefficient (.97), a high percentage of agreement (72%e100%),
and a Cohen kappa of .84. When comparing the scores obtained
on the 2 separate assessments, good test-retest reliability was
proven, obtaining a high Spearman correlation coefficient (.98),
and for Cohen kappa, 85% were on an excellent level of agreement. However, it is difficult to establish correlations between the
MAS and the MMAS from the results obtained in this study.16
Subsequently (in 1994), Malouin et al17 also investigated
concurrent validity by establishing the correlation between the
MAS and the FMA. They included a sample of patients with acute
stroke. Nevertheless, in this case, they also compared the levels of
motor recovery measured by each scale by normalizing the scores
in percent of maximal value. This study analyzed the differences
between the 2 scales by using the Wilcoxon test, showing that the
FMA measured a higher level of motor recovery. The results

Table 2

obtained were the same as those observed in the study conducted


by Poole and Whitney15: high correlation between the scores of
each scale (Spearman correlation coefficientZ.96), and high
correlation between items (.65e.92) except for sitting balance
( 0.1), with the strongest correlation being the one for the upper
limb items.17
Fugl-Meyer Assessment
Presented by Fugl-Meyer et al18 in 1975, the FMA is a scale
created to measure motor recovery in patients who have suffered a
stroke. The FMA comprises 4 different sections: motor function
(voluntary movements and reflexes of limbs), balance, sensation,
and passive joint motion and joint pain. The items are scored in a
3-grade ordinal scale, with 0 as minimum and 2 as maximum, and
a maximum score of 226.18
In the same article, the authors also described an investigation
carried out with a sample of 28 patients with acute stroke; they

Motor function scales excluded from the study

Scale

Exclusion Criteria

Frenchay Arm Test


Wolf Motor Function Test
Action Research Arm Test
Jebsen Hand Test
Motor Status Scale
Arm Motor Ability Test
Blox and Block Test
Nine Hole Peg Test
Reaching Assessment Scale
Trunk Control Test
Trunk Impairment Scale
Chedoke McMaster Stroke Assessment

They do not specifically assess gross motor function, but they are focused on measuring
hands fine motor function

It
It
It
It
It
It
It

does
does
does
does
does
does
does

not
not
not
not
not
not
not

assess motor function but functional capacity (it allows compensations)


assess gross motor function; it has not been validated in patients poststroke
assess upper limb motor function but finger dexterity
assess motor function but trunk compensations
assess motor function but postural control of trunk
assess motor function but postural control of trunk
specifically assess motor function

www.archives-pmr.org

Main results of the studies

Study

Scale

Psychometric Properties

Statistical Analysis

Results

Carr et al
Poole and
Whitney15
Loewen and
Anderson16
Malouin et al17
Fugl-Meyer
et al18
Duncan et al19
Sanford et al20
Hsieh et al21

MAS
MAS

Interrater and test-retest reliability


Concurrent validity (correlation with FMA) and
interrater reliability
Interrater and test-retest reliability

Pearson correlation coefficient


Spearman rank correlation coefficient

rZ0.95 for interrater reliability and 0.98 for test-retest


rZ0.88 for MAS/FMA and rZ0.99 for interrater reliability

Spearman rank correlation coefficient and


Cohen k coefficient
Spearman rank correlation coefficient
Correlation coefficients

rZ0.97 and kZ0.84 for interrater reliability; and rZ0.98


and kZ0.80 for test-retest
rZ0.96 between MAS and FMA
rZ0.88 between the different items

Pearson correlation coefficient


ICC
Pearson correlation coefficient and SRM

Hsueh et al22

FMA and S-FM

Sdring et al23

SMES

Halsaa et al24
Daley et al25
Daley et al26

SMES
STREAM
STREAM

Hsueh et al27

S-STREAM

rZ0.98 for interrater and test-retest reliability


ICCZ0.96
rZ0.92 between S-FM and FMA; rZ0.48-0.59 with ADL
scales; SRMZ0.62-0.71 for S-FM and 0.60-0.67 for FMA
rZ0.70 for FMA and 0.73 for S-FM (correlation with
Barthel); SRMZ0.99 for FMA and 0.94 for S-FM,
ICCZ0.99 for FMA and 0.95 for S-FM
rZ0.83-0.94 (correlation with BLMA), Cronbach
aZ0.94-0.99
kZ0.77; ICCZ0.95
rZ0.49-0.95; kZ0.32-1.00
rZ0.99 for interrater and test-restest reliability; Cronbach
aZ0.96-0.99
Rasch reliability coefficient >0.91; ICC>0.99

Hsieh et al28

STREAM and
S-STREAM
STREAM and
S-STREAM
RMA

14

Hsieh et al21
Lincoln and
Leadbitter29

M-MAS
MAS
FMA
FMA
FMA
S-FM

Kurtais et al30

RMA

Collin and
Wade31
Cameron and
Bohannon32

MI
MI

Concurrent validity (correlation with FMA)


Reliability (internal consistency) and content
validity
Interrater reliability and test-retest
Interrater reliability
Concurrent validity (correlation with FMA),
predictive validity, and sensitivity
Concurrent validity (correlation with FMA and
STREAM), predictive validity, sensitivity, and
test-retest reliability
Concurrent validity and construct validity
Interrater reliability
Internal consistency and interrater reliability
Interrater and test-retest reliability, internal
consistency
Realiability and concurrent validity (correlation
with STREAM)
Predictive validity and sensitivity
Concurrent and predictive validity, test-retest
reliability and sensitivity
Validity, interrater and test-retest reliability

Construct validity (correlation with FIM),


reliability, and sensitivity
Construct validity (correlation with RMA),
reliability, and sensitivity
Internal consistency and criterion validity
(correlations with dynamometer measurements)

Spearman correlation coefficient, SRM, and


ICC
Spearman correlation coefficient and
Cronbach a
Cohen k and ICC
Pearson correlation coefficient and Cohen k
Pearson correlation coefficient and
Cronbach a
Rasch analysis and ICC
Pearson correlation coefficient and ICC
Spearman correlation coefficient, ICC, and
SRM
Reproductivity and scalability coefficients
(Guttman scale); correlation coefficients;
ANOVA (F test)
Cronbach a, ICC, Spearman correlation
coefficient, and SRM
Spearman correltion coefficient
Cronbach a and Pearson correlation
coefficient

Gross motor function in stroke

www.archives-pmr.org

Table 3

rZ0.39-0.52 for STREAM, 0.51-0.60 for S-STREAM;


ICCZ0.90-1.00 for S-STREAM and 0.56-1.00 for STREAM
rZ0.75 for S-STREAM and 0.72 for STREAM; ICCZ0.96 and
0.98, respectively; SRMZ1.10-1.26 for S-STREAM
Reproductivity and scalability coefficientsZ0.79-0.98;
rZ0.66-0.93
ICC and Cronbach aZ0.88-0.95; rZ0.38-0.86; SRMZ0.511.20
rZ0.75-0.81 between MI and RMA, and 0.88 for interrater
reliability
Cronbach aZ0.77; rZ0.78-0.91

Abbreviations: ADL, activities of daily living; ANOVA, analysis of variance; BLMA, Brigitta Lindmark Motor Assessment.

6
were assessed with the FMA during the first week after hospital
admission.18 The authors established correlations between the
motor function subscales for both upper and lower limbs,
obtaining a mean correlation coefficient of .88, which suggests
good internal consistency and shows that the scale is a reliable
measure. They also observed that the items of the scale were
consistent with the motor recovery pattern, showing the scales
content validity.
In 1983, Duncan et al19 investigated the reliability of the FMA
in 18 patients with chronic stroke (at least 1y after a stroke). To
evaluate test-retest reliability, the 18 patients were scored using
the FMA by 1 rater at 3-week intervals. To evaluate interrater
reliability, 2 different raters scored them again (one on the same
day as the second test and the other on the same day as the third).
There was high test-retest reliability for the total score (rZ.98), as
well as for each section of the test (rZ.89e.99). Interrater reliability was also found to be high (rZ.86e.99).19
Sandford et al20 also established (in 1993) the interrater reliability of the FMA. They used 3 different therapists who were
previously trained in the use of the scale. They assessed 12 patients with acute and subacute stroke on different occasions (each
within 1 working day of the previous assessment). The results
obtained after the statistical analysis showed a high intraclass
correlation coefficient (ICC) for the total score (.96), as well as for
different parts of the test (ranging from .61 for the pain to .97 for
the upper limb), which means that the FMA is a reliable measure
to assess motor function in patients with stroke.20
In 2007, Hsieh et al21 created a short version of the FMA
(S-FM) by selecting 6 items from each motor function subscale
(upper and lower limb) and keeping only 12 items from the
original 50 (on the basis of expert opinions and the results of
Rasch analyses). To evaluate its psychometric properties, they
used a sample of 279 subjects who were scored on different occasions (14, 30, 90, and 180d after a stroke). They used the FMA
to establish concurrent validity, as well as 2 activities of daily
living scales (Barthel Index [BI] and Franchay Activity Index
[FAI]) to establish predictive validity. Sensitivity was calculated
by the standardized response mean. Pearson r was used to
correlate the results, obtaining a high correlation between the
S-FM and the FMA (rZ.92), and a moderate correlation between
the S-FM and the BI and FAI (rZ.48e.59), which would create
some doubts about the ability of the scale to predict functional
capacity in the future. Sensitivity was shown to be moderate for
both scales (standardized response mean [SRM]Z.62e.71 for the
S-FM and .60e.67 for the FMA).21
This short version of the FMA was investigated again in 2008
in a study conducted by Hsueh et al,22 in which they compared its
psychometric properties (concurrent and predictive validity,
sensitivity, and test-retest reliability) with those of the FMA,
STREAM, and a 15-item simplified version of STREAM
(S-STREAM) in a sample of 50 patients with stroke.22 Both the
FMA and the S-FM showed acceptable predictive validity when
correlated with the BI (rZ.66e.72 for the FMA and rZ.70e.74
for the S-FM), excellent concurrent validity (rZ.91e.99 when
comparing all scales), high sensitivity (SRMZ0.3e1.16 for the
FMA and 0.99e1.00 for the S-FM), and high test-retest reliability
(ICCZ.95e.98 for the FMA and .93e.96 for the S-FM).22
Sodring motor evaluation for stroke patients
The SMES was developed in 1995 by Sdring et al.23 It measures
motor function and activities in patients with stroke by evaluating
the motor function of the upper limb and the lower limb and gross

M.D. Gor-Garca-Fogeda et al
motor function. The first 2 subscales assess the execution of
simple voluntary movements, while the third evaluates the
execution of functional tasks, which include trunk movements,
balance, and gait. The SMES comprises 32 different items, which
are scored using a 5-point scale.23
Sdring23 administered the scale to a sample of 93 patients
with subacute stroke (fewer than 14d after a stroke) on 3 different
occasions: 2 to 6 days, 8 to 10 days, and 3 months after admission.
They also scored them using the Brigitta Lindmark Motor
Assessment, eliminating the trunk control and gait items because
they were not validated. High correlations were found for upper
and lower limb subscales between the 2 tests (.83e.94), suggesting good concurrent validity. The authors23 also established
construct validity by using Cronbach alpha, finding high correlations between each item and the rest (.94e.99).
In 1999, Halsaa et al24 investigated the psychometric properties of the SMES; more specifically, its interrater reliability.24
They selected a sample of 30 patients with acute stroke and
subacute stroke, who were then scored by 3 different therapists on
3 consecutive days in groups of 10. They used Cohen kappa to
evaluate the reliability of each item, and the ICC to compare total
scores, finding high reliability for each item (kZ.77) as well as for
the sum scores (ICCZ.95).24
Stroke rehabilitation assessment of movement
The STREAM was designed by Daley et al25 in 1997 to evaluate
the recovery of voluntary movement and basic mobility after a
stroke. The scale includes 30 items divided into 3 different sections: voluntary movements of upper and of lower limbs, and basic
mobility (functional tasks). Each item is scored on an ordinal
scale, with 1 as a minimum, and as a maximum, 3 for the upper
and lower limb subscales and 4 for the basic mobility subscale,
with 70 as the maximum possible score.25
Regarding the development of STREAM, the authors created a
prototype version comprising 43 items and studied its internal
consistency and interrater reliability using a sample of 26 patients
with stroke, who were assessed by 10 therapists. Internal consistency was analyzed by calculating Pearson correlation coefficient
between the items and the sum scores. Following this, the next
step was to eliminate those items that showed low or high correlation with the others, which would suggest poor internal consistency or reiteration, respectively. The mean correlation between
each item and the total score was .49 to .95. The authors used
Cohen kappa to find the interrater reliability (obtaining kZ0.32e
1.00) and subsequently the items that showed low reliability were
eliminated.25
The internal consistency and reliability of STREAM were
evaluated again by the same authors, but on this occasion they
used the modified version with only 30 items.26 They conducted 3
different substudies, which were presented in the same article. In
the first substudy, a sample of 20 patients was scored by pairs of
raters from a group of 6 therapists through direct observation to
evaluate interrater reliability. In the second, they asked 20 therapists to view and score 4 videotaped assessments (4 patients from
the first study who agreed to be videotaped and participate in the
second one) on 2 occasions with a 1-month interval. The internal
consistency was evaluated in the third substudy by using the
scores obtained for the 20 subjects from the original sample used
in the first study and for 6 new patients who presented lower
scores (none of the original 20 patients scored <30 points on the
scale). The authors observed a high correlation between the results
obtained from direct observation (interrater) for each subscale
www.archives-pmr.org

Gross motor function in stroke


(.98e.99) and for the total scores (.99). They also observed
excellent internal consistency demonstrated by Cronbach alpha
(.96e.99).26
Hsueh et al27 modified STREAM in 2006, creating S-STREAM
by selecting 5 items from each subscale, and then analyzed its
psychometric properties (reliability and concurrent validity). A
sample of 351 subjects was scored using STREAM by 1 rater. Each
subscale of STREAM was then simplified by deleting redundant
items on the basis of expert opinion and the results of the Rasch
analysis. They found high reliability (Rasch reliability coefficients
of >.91) for the modified scale (S-STREAM), as well as adequate
concurrent validity when comparing the scores of STREAM and
those of the S-STREAM (ICCZ.99).27
In 2007, Hsieh et al28 carried out a new investigation about the
psychometric properties of the S-STREAM; more specifically,
about its predictive validity and sensitivity. A sample of 388
patients was assessed by 1 therapist from a group of 12 at the
moment of admission, discharge, and finally, 6 months after the
stroke. To study the predictive validity, the authors compared
the results obtained from the administration of the S-STREAM 6
months after the stroke with those obtained from the BI and the
FAI, and also with those obtained from STREAM. When correlating the results from STREAM and the S-STREAM with the
activities of daily living scales (BI and FAI) using Pearson correlation coefficient, acceptable correlation was found for
STREAM (.39e.62) and the S-STREAM (.51e.60), suggesting
moderate predictive validity. Sensitivity results were strong for the
S-STREAM (SRMZ0.90e1.06) and moderate for STREAM
(SRMZ0.56e1.00).28
In a study conducted in 2008 in which the FMA and the S-FM
were compared, Hsueh22 also investigated STREAM and the SSTREAM. The results obtained were the same for both scales,
showing good concurrent and predictive validity (rZ.75 for the
S-STREAM and .72 for STREAM) and excellent test-retest reliability (ICCZ.98 for the S-STREAM and .98 for STREAM), with
the S-STREAM being the most sensitive (SRMZ1.10e1.26).22
Rivermead motor assessment
The RMA was developed in 1979 by Lincoln and Leadbitter.29 It
includes 38 items divided into 3 sections: gross motor function,
lower limb and trunk, and upper limb. A 0 to 1 scoring system was
adopted, so the patient scores 1 if he/she does the activity according to the specific instructions and 0 if he/she does not.
Lincoln and Leadbitter29 created a first version of the scale and
established its validity by assessing a sample of 51 patients and
finding the coefficient of reproducibility and scalability, which
was shown to be high (.79e.98), indicating that the scale was
valid. To evaluate the test-retest reliability, 1 therapist assessed 10
patients on 2 occasions 1 month apart, and correlations were
established between the results from the first and the second assessments, showing moderate to good reliability (rZ.66e.88).
Last, to evaluate interrater reliability, 7 assessments of patients
were videotaped, 1 given by each rater, and all 7 raters scored each
of the assessments. The authors used the 1-way analysis of variance to analyze the results, showing significant differences between the patients, but not between the raters, except in the upper
limb section. Subsequently, the scale was modified by deleting
some items on the basis of being difficult to score, and by arranging all items in order of difficulty for each section, thus
creating the present-day version of the scale. The test-retest reliability of the new scale was investigated in a sample of 10 patients
who were assessed by 1 rater on 2 occasions at a 4-week interval.
www.archives-pmr.org

7
The correlation coefficients (.66e.88) indicated that the test had
adequate test-retest reliability.29
In 2009, Kurtais et al30 investigated the RMAs construct validity, reliability, and sensitivity. They selected a sample of 107
subjects, including patients in the subacute and chronic phases
after stroke who were assessed with the RMA and the FIM at
hospital admission and on discharge. Reliability was established
by evaluating internal consistency using Cronbach alpha and the
ICC; construct validity was calculated by comparing the results
from the RMA and the FIM; and sensitivity was evaluated through
the effect size and SRM. The results obtained showed good internal consistency (ICC and Cronbach a between .88 and .95),
moderate to high construct validity (rZ.39e.86), and good
sensitivity (SRMZ.60e.86; effect sizeZ.38e.51).30
Motricity index
The MI was developed in 1980 by Demeurisse et al,33 and it is
based on the Medical Council Research grades to assess muscular
strength. The large number of movements was reduced to 1
movement at each joint, which represented the general strength of
movement at the joint (3 for the upper limb: thumb-index pinch,
elbow flexion, and shoulder abduction, and 3 for the lower limb:
ankle dorsiflexion, knee extension, and hip flexion), giving a rapid
overall indication of a patients limb impairment.
In 1990, Collin and Wade31 investigated the MIs construct
validity by comparing it with the RMA, and they also investigated
its reliability and sensitivity. Forty subjects were assessed on 3
occasions (6, 12, and 18wk after stroke) by 2 raters who administered the MI and 2 additional new raters who administered the
RMA. All assessments were performed within 5 days of each
other on each patient. When calculating the Spearman correlation
coefficient between the MI and the RMA, the authors found good
construct validity (rZ.75e.81), as well as high interrater reliability for the MI (rZ.88). Results at 6 weeks were compared with
walking ability at 18 weeks to determine the predictive value of
the test, finding high correlations, which suggests that the scale
has good predictive validity.
Ten years later, Cameron and Bohannon32 assessed the lower
extremity MI in a sample of 15 patients with stroke. Cronbach
alpha was used to establish internal consistency, which was good
(Cronbach aZ.77), and criterion validity was established by
comparing the results from the MI with the measurements obtained using a handheld dynamometer, finding high correlations
(rZ.78e.91).

Discussion
This review gives an overview of gross motor function scales for
patients with stroke. A complete literature search of the electronic
literature databases without limits on publication data was carried
out. Our primary objectives in this study were (1) to compile all
scales available in the scientific literature that assess gross motor
function in stroke; (2) to establish which specific aspects of gross
motor function they assess; and (3) to study their psychometric
properties.
We found that a multitude of measurement scales are available to
assess gross motor function in patients with stroke. However, they
are not specific scales because they dedicate only 1 or more subsections to assess gross motor function. All the scales included
in the review have been shown to have several psychometric
properties, showing their usefulness in both clinical practice
and research.

M.D. Gor-Garca-Fogeda et al

8
According to the data extracted, the most suitable scales to be
used in the clinical field would be the short versions of the FMA
and STREAM because they have been shown to have sound
psychometric evidence and are easy and fast to administer. The
MAS has also proven to be useful because of its high reliability,
and when compared with the FMA, we observe that its administration is shorter and less complex, so it could be used as an
alternative to the FMA. However, we cannot conclude whether it
can be administered instead of the FMA or whether both scales
should be used together. The short version of the MAS has been
shown to be reliable as well, although it has not been compared
with the MAS. The FMA has proven to have a large number of
psychometric properties; nonetheless, its internal consistency is
doubtful (due to the sitting balance item), and so it would be
constructive to effectuate some modifications to improve
the scale.
It has been noted that there is no well-defined criterion standard when investigating validity and that there is no consensus on
the use of a specific scale. The FMA is used in 2 of the articles to
investigate the MAS15,17; the RMA is compared with the MI in
another study31; and even the FIM, which is an activity of daily
living scale, is used to investigate validity in the RMA.30 The
dynamometer is also used as a criterion standard in one of the
studies to evaluate the MI.32 Therefore, it would be interesting to
take this lack of consensus into account for further investigation,
and compare motor function scales with objective data obtained
from instrumental analysis systems.
Furthermore, the inclusion criteria and the selection strategies
were not well described in some of these studies. Several investigations included patients with associated problems, which
could have interfered with the results, with an example being the
study conducted by Duncan et al19 in 1983 in which, according to
the authors, a patient with aphasia obtained the minimum score
possible in the test because of the comprehension troubles he
presented. In the 1997 study by Daley et al,25 3 patients with
aphasia were included, as well as another 4 with cognitive problems, although the authors do not mention any incidents during the
study caused by these issues.
Moreover, in most of the studies, the recovery phase of the
patients included in the sample14-16,22,25-27,29,32 is not taken into
account, and only a few distinguish between acute, subacute, and
chronic patients.17-21,23,24,28,30,31 Furthermore, many of them do
not explain the process used to assign patients to therapists,
though some of the studies do specify that the patients were
randomly assigned and that the therapists did not have previous
knowledge of the state of the patients, and did not know the
results obtained by other therapists or from the criterion standard.
In most of the studies, there is a decline in the number of
subjects, either because of the death of subjects or another cause;
however, most of them included those subjects in the final results.
In the remainder of the studies, this reason is unspecified, and so
we cannot be sure whether the final results include the subjects
who were lost during the study or not.

Study limitations
Although this systematic review was conducted with care, there
are some methodological limitations such as not hand-searching
conference proceedings, missing outcome data, the language restriction, or not performing meta-analyses of individual patient
data. In addition, this systematic review has included articles with

several methodological limitations. Therefore, our conclusions


should be interpreted with caution.

Conclusions
In this study, we found the following: (1) Six measurement scales
for patients with stroke that include specific subsections to assess
gross motor function. (2) All the scales compiled have been shown
to be useful both in clinical practice and in terms of research. (3)
The scales for which the most psychometric properties have been
established in clinical trials are the FMA and STREAM. (4) The
most suitable scales to be used in the clinical field would be the
short versions of the FMA and STREAM. (5) A real consensus
about the measurement of gross motor function in patients with
stroke is not available in the recent literature.

Keywords
Motor activity; Motor skills disorders; Paresis; Rehabilitation;
Stroke

Corresponding author
Francisco Molina-Rueda, BcPT, MSc, PhD, Physical Therapy,
Occupational Therapy, Rehabilitation and Physical Medicine
Department, Rey Juan Carlos University, Avda, de Atenas, s/n, CP
Alcorco n, Madrid 28922, Spain. E-mail address: francisco.
molina@urjc.es.

Appendix 1 Search strategy


1. Motor function AND test AND (stroke OR cerebrovascular accidents
OR hemiplegia)
2. Motor function AND measure* AND (stroke OR cerebrovascular
accidents OR hemiplegia)
3. Motor function AND assessment AND (stroke OR cerebrovascular
accidents OR hemiplegia)
4. Motor function AND scale AND (stroke OR cerebrovascular accidents
OR hemiplegia)
5. Motor recovery AND evaluation AND (stroke OR cerebrovascular
accidents OR hemiplegia)
6. Motor function AND test AND (stroke OR cerebrovascular accidents
OR hemiplegia)

References
1. World Health Organization. The atlas of heart disease and stroke.
Available at: http://www.who.int/cardiovascular_diseases/resources/
atlas/en. Accessed March 10, 2013.
2. Daz-Guzman J, Egido-Herrero JA, Gabriel-Sanchez R, et al. Incidence of strokes in Spain: the Iberictus study. Data from the pilot
study [in Spanish]. Rev Neurol 2008;47:617-23.
3. Instituto Nacional de Estadstica. Defunciones segun la causa de
muerte 2008. Available at: www.ine.es. Accessed March 10, 2013.
4. Duran-Heras MA. Informe Sobre el Impacto Social de los Enfermos
Dependientes por ICTUS. INFORME ISEDIC. 2nd ed. Madrid:
Engraf, SA; 2004.
5. Hallett M. Plasticity of the human motor cortex and recovery from
stroke. Brain Res Brain Res Rev 2001;36:169-74.
6. Kasner SE. Clinical interpretation and use of stroke scales. Lancet
Neurol 2006;5:603-12.

www.archives-pmr.org

Gross motor function in stroke


7. Levin MF, Kleim JA, Wolf SL. What do motor recovery and
compensation mean in patients following stroke? Neurorehabil
Neural Repair 2009;23:313-9.
8. Task Force on Standards for Measurement in Physical Therapy.
Standards for tests and measurements in physical therapy practice.
Phys Ther 1991;71:589-622.
9. Fermanian J. Validation des echelles devaluation en medecine
physique et de readaptation: comment apprecier correctement leurs
qualites psychometriques. Ann Readapt Med Phys 2005;48:281-7.
10. Poole JL, Whitney SL. Assessments of motor function post stroke: a
review. Phys Occup Ther Geriatr 2001;19:1-22.
11. Croarkin E, Danoff J, Barnes C. Evidence-based rating of upperextremity motor function tests used for people following a stroke.
Phys Ther 2004;84:62-74.
12. Gellez-Leman MC, Colle F, Bonan I, Bradai N, Yelnik A. Evaluation
des incapacites fonctionnelles chez le patient hemiplegique: mise au
point. Ann Readapt Med Phys 2005;48:361-8.
13. Marzo Castillejo M, Viana Zulaica C. Calidad de la evidencia y
grado de recomendacion. Guas Clnicas 2009;9:6. Available at:
http://www.fisterra.com/guias2/fmc/sintesis.asp. Accessed March 10,
2013.
14. Carr JH, Shepherd RB, Nordholm L, Lynne D. Investigation of a
new motor assessment scale for stroke patients. Phys Ther 1985;65:
175-80.
15. Poole JL, Whitney SL. Motor assessment scale for stroke patients:
concurrent validity and interrater reliability. Arch Phys Med Rehabil
1988;69:195-7.
16. Loewen SC, Anderson BA. Reliability of the Modified Motor
Assessment Scale and the Barthel Index. Phys Ther 1988;68:1077-81.
17. Malouin F, Pichard L, Bonneau C, Durand A, Corriveau D. Evaluating motor recovery early after stroke: comparison of the FuglMeyer Assessment and the Motor Assessment Scale. Arch Phys
Med Rehabil 1994;75:1206-12.
18. Fugl-Meyer AR, Jaasko L, Leyman I, Olsson S, Steglind S. The poststroke hemiplegic patient, 1: a method for evaluation of physical
performance. Scand J Rehabil Med 1975;7:13-31.
19. Duncan PW, Propst M, Nelson SG. Reliability of the Fugl-Meyer
assessment of sensorimotor recovery following cerebrovascular accident. Phys Ther 1983;63:1606-10.
20. Sanford J, Moreland J, Swanson LR, Stratford PW, Gowland C.
Reliability of the Fugl-Meyer assessment for testing motor performance in patients following stroke. Phys Ther 1993;73:447-54.
21. Hsieh YW, Hsueh IP, Chou YT, Sheu CF, Hsieh CL, Kwakkel G.
Development and validation of a short form of the Fugl-Meyer motor
scale in patients with stroke. Stroke 2007;38:3052-4.
22. Hsueh IP, Hsu MJ, Sheu CF, Lee S, Hsieh CL, Lin JH. Psychometric
comparisons of 2 versions of the Fugl-Meyer Motor Scale and 2
versions of the Stroke Rehabilitation Assessment of Movement.
Neurorehabil Neural Repair 2008;22:737-44.
23. Sdring KM, Bautz-Holter E, Ljunggren AE, Wyller TB. Description
and validation of a test of motor function and activities in stroke
patients: the Sdring Motor Evaluation of Stroke Patients. Scand J
Rehabil Med 1995;27:211-7.
24. Halsaa KE, Sdring KM, Bjelland E, Finsrud K, Bautz-Holter E.
Inter-rater reliability of the Sdring Motor Evaluation of Stroke patients (SMES). Scand J Rehabil Med 1999;31:240-3.
25. Daley K, Mayo N, Danys I, Cabot R, Wood-Dauphinee S. The Stroke
Rehabilitation Assessment of Movement (STREAM): refining and
validating the content. Physiother Can 1997;49:269-78.
26. Daley K, Mayo N, Wood-Dauphinee S. Reliability of scores on the
Stroke Rehabilitation Assessment of Movement (STREAM) measure. Phys Ther 1999;79:8-19.
27. Hsueh IP, Wang WC, Wang CH, et al. A simplified stroke rehabilitation assessment of movement instrument. Phys Ther 2006;86:936-43.
28. Hsieh YW, Lin JH, Wang CH, Sheu CF, Hsueh IP, Hsieh CL.
Discriminative, predictive and evaluative properties of the simplified
stroke rehabilitation assessment of movement instrument in patients
with stroke. J Rehabil Med 2007;39:454-60.

www.archives-pmr.org

9
29. Lincoln N, Leadbitter D. Assessment of motor function in stroke
patients. Physiotherapy 1979;65:48-51.
30. Kurtais Y, Kucukdeveci A, Elhan A, et al. Psychometric properties of
the Rivermead Motor Assessment: its utility in stroke. J Rehabil Med
2009;41:1055-61.
31. Collin C, Wade D. Assessing motor impairment after stroke: a pilot
reliability study. J Neurol Neurosurg Psychiatry 1990;53:576-9.
32. Cameron D, Bohannon RW. Criterion validity of lower extremity
Motricity Index scores. Clin Rehabil 2000;14:208-11.
33. Demeurisse G, Demol O, Robaye E. Motor evaluation in vascular
hemiplegia. Eur Neurol 1980;19:382-9.
34. Wyller TB, Sdring KM, Sveen U, Ljunggren AE, Bautz-Holter E.
Predictive validity of the Sdring Motor Evaluation of Stroke Patients (SMES). Scand J Rehabil Med 1996;28:211-6.
35. Hsieh YW, Wang CH, Sheu CF, Hsueh IP, Hsieh CL. Estimating the
minimal clinically important difference of the Stroke Rehabilitation
Assessment of Movement measure. Neurorehabil Neural Repair
2008;22:723-7.
36. Lin JH, Hsu MJ, Sheu CF, et al. Psychometric comparisons of 4
measures for assessing upper-extremity function in people with
stroke. Phys Ther 2009;89:840-50.
37. Hsieh YW, Wu CY, Lin KC, Chang YF, Chen CL, Liu JS. Responsiveness and validity of three outcome measures of motor function
after stroke rehabilitation. Stroke 2009;40:1386-91.
38. Langhammer B, Stanghelle JK. Co-variation of tests commonly used
in stroke rehabilitation. Physiother Res Int 2006;11:228-34.
39. Rabadi MH, Rabadi FM. Comparison of the action research arm test
and the Fugl-Meyer assessment as measures of upper-extremity
motor weakness after stroke. Arch Phys Med Rehabil 2006;87:962-6.
40. Beckerman, Vogelaar TW, Lankhorst GJ, Verbeek AL. A criterion for
stability of the motor function of the lower extremity in stroke patients using the Fugl-Meyer Assessment Scale. Scand J Rehabil Med
1996;28:3-7.
41. Filiatrault J, Arsenault AB, Dutil E, Bourbonnais D. Motor function
and activities of daily living assessments: a study of three tests for
persons with hemiplegia. Am J Occup Ther 1991;45:806-10.
42. Crow JL, Harmeling-van der Wel BC. Hierarchical properties of the
motor function sections of the Fugl-Meyer assessment scale for people
after stroke: a retrospective study. Phys Ther 2008;88:1554-67.
43. Miller KJ, Slade AL, Pallant JF, Galea MP. Evaluation of the psychometric properties of the upper limb subscales of the Motor
Assessment Scale using a Rasch analysis model. J Rehabil Med
2010;42:315-22.
44. Van der Lee JH, Beckerman H, Lankhorst GJ, Bouter LM. The
responsiveness of the Action Research Arm test and the Fugl-Meyer
Assessment scale in chronic stroke patients. J Rehabil Med 2001;33:
110-3.
45. Pickering RL, Hubbard IJ, Baker KG, Parsons MW. Assessment of
the upper limb in acute stroke: the validity of hierarchal scoring for
the Motor Assessment Scale. Aust Occup Ther J 2010;57:174-82.
46. Brauer SG, Bew PG, Kuys SS, Lynch MR, Morrison G. Prediction of
discharge destination after stroke using the motor assessment scale
on admission: a prospective, multisite study. Arch Phys Med Rehabil
2008;89:1061-5.
47. Aamodt G, Kjendahl A, Jahnsen R. Dimensionality and scalability
of the Motor Assessment Scale (MAS). Disabil Rehabil 2006;28:
1007-13.
48. English CK, Hillier SL, Stiller K, Warden-Flood A. The sensitivity of
three commonly used outcome measures to detect change amongst
patients receiving inpatient rehabilitation following stroke. Clin
Rehabil 2006;20:52-5.
49. Sabari JS, Lim AL, Velozo CA, Lehman L, Kieran O, Lai JS.
Assessing arm and hand function after stroke: a validity test of the
hierarchical scoring system used in the motor assessment scale for
stroke. Arch Phys Med Rehabil 2005;86:1609-15.
50. Lannin N. Reliability, validity and factor structure of the upper limb
subscale of the Motor Assessment Scale (UL-MAS) in adults
following stroke. Disabil Rehabil 2004;26:109-16.

10
51. Hsueh IP, Hsieh CL. Responsiveness of two upper extremity function
instruments for stroke inpatients receiving rehabilitation. Clin
Rehabil 2002;16:617-24.
52. Safaz I, Yilmaz B, Yasar E, Alaca R. Brunnstrom recovery stage and
motricity index for the evaluation of upper extremity in stroke:
analysis for correlation and responsiveness. Int J Rehabil Res 2009;
32:228-31.
53. Adams SA, Ashburn A, Pickering RM, Taylor D. The scalability of
the Rivermead Motor Assessment in acute stroke patients. Clin
Rehabil 1997;11:42-51.
54. Adams SA, Pickering RM, Ashburn A, Lincoln NB. The scalability
of the Rivermead Motor Assessment in nonacute stroke patients. Clin
Rehabil 1997;11:52-9.
55. Lu WS, Wang CH, Lin JH, Sheu CF, Hsieh CL. The minimal
detectable change of the simplified stroke rehabilitation assessment
of movement measure. J Rehabil Med 2008;40:615-9.
56. Woodbury ML, Velozo CA, Richards LG, Duncan PW, Studenski S,
Lai SM. Longitudinal stability of the Fugl-Meyer Assessment of the
upper extremity. Arch Phys Med Rehabil 2008;89:1563-9.
57. Woodbury ML, Velozo CA, Richards LG, Duncan PW, Studenski S,
Lai SM. Dimensionality and construct validity of the Fugl-Meyer
Assessment of the upper extremity. Arch Phys Med Rehabil 2007;
88:715-23.

M.D. Gor-Garca-Fogeda et al
58. Ahmed S, Mayo NE, Higgins J, Salbach NM, Finch L, WoodDauphinee SL. The Stroke Rehabilitation Assessment of Movement
(STREAM): a comparison with other measures used to evaluate effects of stroke and rehabilitation. Phys Ther 2003;83:617-30.
59. Collen FM, Wade DT, Bradshaw CM. Mobility after stroke: reliability of measures of impairment and disability. Int Disabil Stud
1990;12:6-9.
60. Van de Winckel A, Feys H, Lincoln N, De Weerdt W. Assessment of
arm function in stroke patients: Rivermead Motor Assessment arm
section revised with Rasch analysis. Clin Rehabil 2007;21:471-9.
61. Johnson L, Selfe J. Measurement of mobility following stroke: a
comparison of the Modified Rivermead Mobility Index and the
Motor Assessment Scale. Physiother 2004;90:132-8.
62. Tucak C, Scott J, Kirkman A, Singer B. Relationships between initial
Motor Assessment Scale scores and length of stay, mobility at
discharge and discharge destination after stroke. NZ J Physiother
2010;38:7-13.
63. Bohannon RW. Motricity Index scores are valid indicators of paretic
upper extremity strength following stroke. J Phys Ther Sci 1999;11:
59-61.
64. Chen HF, Lin KC, Wu CY, Chen CL. Rasch validation and predictive
validity of the action research arm test in patients receiving stroke
rehabilitation. Arch Phys Med Rehabil 2012;93:1039-45.

www.archives-pmr.org