Articulo Valoracion Radial

1. Splinting and Radial Nerve Palsy:
A Single-subject Experiment
Susan D. Hannah, BScOT (Reg), OTR, CHT
The Toronto Western Hospital Hand Program
The University Health Network '
399 Bathurst Street
Toronto, Ontario, Canada
Pamela 1. Hudak, MSc, BScPT
Doctoral candidate, Institute for Medical Sciences,
The University of Toronto;
The Toronto Western Hospital Hand Program;
The Toronto Western Hospital and Hospital
for Sick Children
Toronto, Ontario
The radial nerve is a commonly injured upper
extremity peripheral nerve. Although its superficial
position in the spiral groove makes the nerve most
vulnerable to injury at the mid-humeral level, injury to
the radial nerve can also occur at the neck of the
humerus and brachial plexus.1
Common causes of
radial nerve palsy are humeral fractures, elbow dis-
locations, and Monteggia fracture-dislocations.1
Although trauma is the most common cause of radial
nerve palsy, other conditions such as diabetes mellitus
and polyarteritis may cause radial nerve dysfunction.1
Radial nerve palsy causes significant impairment
of hand function, presenting clinically as the classic
drop wrist. The inability to extend the wrist results in
loss of tenodesis action1
and diminished use of the
fingers for power grip and grasp-and-release
actions.1
-
3
On a positive note, the loss of sensation
associated with radial nerve palsy is not functionally
significant, because palmar sensation remains intact.3
Various authors1
have reported that the radial
nerve recovers spontaneously, and estimates of this
recovery period range from 5 weeks to 24 months.
One of the challenges for hand therapists during this
period of nerve regeneration is to fabricate a splint
that prevents over-stretching of denervated extensor
musculature while maximizing hand function.3
A
Correspondence and reprint requests to Susan D. Hannah, BScOT
(Reg), OTR, CHT, The University Health Network Hand
Program, The Toronto Western Hospital, 399 Bathurst Street,
Toronto, Canada, M5T 2S8
ABSTRACT: This study examines which of three splint designs
most effectively improved hand function in a patient with radial
nerve palsy, and demonstrates the application of a single-subject
experimental design. The static volar wrist cock-up splint
(splint 1), dynamic tenodesis suspension splint (splint 2), and dor-
sal wrist cock-up with dynamic finger extension splint (splint 3)
were evaluated. Each splint was worn for 3 weeks, and hand func-
tion was assessed by means of standardized measures of function
and disability. Statistical significance was calculated using the
minimal level of detectable change (MDC) at the 95% confidence
level. Only with splints 2 and 3 did a true change in function
occur, compared with baseline scores (no splint). In addition, the
patient completed all tasks while using splints 2 and 3 but did not
complete three tasks while using splint 1.The hand therapists'
goal is to fabricate a splint that improves function and that the
patient will wear. Only splint 3 met these criteria. This experiment
highlights the need to evaluate both the statistical and the clinical
significance of treatment interventions.
JHAND THER. 2001;14:195-201.
variety of splinting options have been described-a
static wrist extension splint,4 a light-weight spring
splint that allows wrist movement,S a modified
spring splint, or "lively splint,,,6 a dorsal slab splint
with loop suspension for digits} and a splint that
"harnesses" the tenodesis pattern of the hand using
static lines to support the digits.1
McKee and
MorganS summarized static and dynamic splinting
options used following radial nerve palsy at the mid-
humeral level but did not report on splint efficacy.
Although splinting options are described in the liter-
ature, we were unable to locate (in the CINAHL and
MEDLINE databases, 1980-99) published work report-
ing evidence of splint efficacy for improving the
hand function of patients with radial nerve palsy.
Hand therapists practice in a continually changing
health care environment in which policy makers,
health care providers, and consumers expect docu-
mentation of clinical change and accountability.9
From a professional as well as a clinical perspective,
hand therapists are required to justify treatment
interventions such as splinting.
We often refer to the published literature for
research evidence to support our treatment choices.
One of the more rigorous research designs is the dou-
ble-blind randomized controlled trial. Although the
radial nerve is a commonly injured peripheral nerve,
the incidence of this condition is relatively low. Thus,
even multicenter randomized controlled trials are
infeasible because of small sample sizes.1O
-
12
In addi-
tion to the issue of sample size, a randomized con-
July-September 200 I 195

2. FIGURE 1. Static volar wrist cock-up splint. This splint stati-
cally holds the wrist in extension.
FIGURE 2. Dynamic tenodesis suspension splint. Through
the use ofa tenodesis action, this splint provides adynamic, pas-
sive assist to substitute for weak or absent wrist and finger
extensors.
FIGURE 3. Dorsal wrist cock-up with dynamic finger exten-
sion splint. This splint provides adynamic, passive assist to sub-
stitute for weak or absent finger extensors. The wrist is held stat-
ically in extension.
196 JOURNAL OF HAND THERAPY
trolled trial is not the best way to determine the treat-
ment of choice for a specific patient, since results are
based on the average improvement scores for all sub-
jects and therefore do not provide information on the
performance of individual subjects.9
Since the focus and accountability of therapists are
at the level of the individual patient, we need an alter-
native research method that allows for the evaluation
of individual treatment interventions. Single-subject
experimental designs have been proposed as one alter-
native to the randomized controlled trial.9
-
14
These
designs are advantageous for a number of reasons:
results can indicate whether treatment benefits a par-
ticular patient, the patient is his or her own control,
and results can be used to fine-tune clinical hypothe-
ses.12
,14 In addition, a single-subject experimental
design is a natural extension of therapists' systematic
evaluation of patients and can therefore be easily
incorporated into clinical practice. If this approach is
used to study enough patients with the same condi-
tion over time, results may be generalizable.14
This study was designed to evaluate which of three
splint designs was most effective in improving hand
function in an 83-year-old woman with radial nerve
palsy, and to demonstrate the application of a single-
subject experimental design to a clinical problem.
METHODS
Subject Profile
The patient was an active, cooperative 83-year-old
woman who had sustained a subglenoid dislocation
of her left shoulder after slipping and falling at home.
She experienced full recovery of the biceps, triceps,
supinator, and pronator teres muscles following a left
brachial plexopathy. Residual impairments included
lack of wrist, finger, and thumb extension and short-
ening of the flexor pollicis longus musculotendinous
unit. Six months after the injury, she was referred to
hand therapy for fabrication of a splint to improve
performance of activities of daily living. At the time
of referral, the treating therapist reported that the
patient's physical status had stabilized.
Evaluators
The three evaluators were hand therapists working
in a university-affiliated program specialized in the
care of upper extremity and hand conditions. This pro-
gram is located in a major tertiary-care hospital in a
large urban center in Canada. Each therapist had at
least five years' experience working in hand therapy.
Research Design
A single-subject experimental design was used to
compare the patient's responses to four treatment
interventions-no splint (baseline), static volar wrist

3. cock-up splint (Figure I), dynamic tenodesis suspen-
sion splint (Figure 2), and dorsal wrist cock-up with
dynamic finger extension splint (Figure 3). These
three splint designs were selected becal,lse of their
common use clinically and frequent description in
the literature. The order of splint design was ran-
domly allocated. Each splint was fabricated by the
same therapist, and standard instructions for wear
and care of each splint were provided. Because of
shortening of the patient's flexor pollicis longus mus-
culotendinous unit, thumb abduction components
were added to the dynamic tenodesis suspension
splint and the dorsal wrist colck-up with dynamic fin-
ger extension splint to improve thumb abduction.
On the basis of the treating therapist's clinical judg-
ment and manual muscle testing results collected at
3-week intervals during the study, the patient's con-
dition was assumed to be stable. Although the
patient was blinded to the weekly assessment results,
it was not possible to blind the patient and therapists
to which splint was being evaluated, since the splints
were worn during evaluation of the patient's physi-
cal function.
Approval of the hospital's ethics review committee
and informed consent of the patient were obtained
prior to initiation of the study.
Outcome Measures
Four established outcome measures were chosen to
assess the following variables-performance of the
upper extremity during activities of daily Living (Test
Evaluant Les Membres Superieurs des Personnes
Agees, TEMPA), self-reported level of disability
(Disabilities of the Arm, Shoulder and Hand ques-
tionnaire, DASH), self-perceived performance in
activities of daily living (Canadian Occupational
Performance Measure, COPM), and strength of spe-
cific muscle groups to monitor stability of physical
status (manual muscle testing). Two nonstandard-
ized questionnaires designed specifically for this
study were also used-a splint-wearing schedule and
a splint-preference questionnaire. .
The TEMPA is an upper extremity function test
designed for geriatric populations. High reliability as
well as content, face, and preliminary construct
validity in this population have been reported.1S
,16
The TEMPA consists of five bilateral and four unilat-
eral tasks related to daily activities (Figure 4). Both
the right and left hands are evaluated in the perform-
ance of unilateral tasks. Scoring is based on speed of
performance, a functional rating score (independence
in performing each task), and task analysis (range of
motion, strength, prehension, and precision of gross
and fine movement). A combined total score is calcu-
lated by adding the functional rating and task analy-
sis scores for both unilateral and bilateral tasks.
Speed of execution is examined independently of the
Pick up and move a jar Unilateral
2 Open a jar and remove a spoonful of coffee Bilateral
3 Pick up a pitcher and pour water into a glass Unilateral
4 Unlock a lock and open a pill container Bilateral
5 Write on an envelope and stick a stamp on it Bilateral
6 Tie a scarf around one's neck Bilateral
7 Shuffle and deal playing cards Bilateral
8 Handle coins Unilateral
9 Pick up and move small objects Unilateral
FIGURE 4. TEMPA Tasks
combined total score. We report the functional rating
score in this study, since it provides the most global
indication of functional independence.
The DASH self-completed questionnaire measures
disability and symptoms related to upper extremity
musculoskeletal disordersY The patient is asked to
rate his or her ability to perform certain activities dur-
ing the previous week. Because this measure is intend-
ed to capture disability, patients are asked to describe
their ability to do different tasks regardless of which
hand or arm is used to complete the activity. The 30-
item questionnaire includes 21 physical function items
(e.g., prepare a meal, garden or do yard work, and par-
ticipate in recreational activities), 6 symptom items
(e.g., pain, tingling, weakness, stiffness) and 3
social!role function items (extent to which arm, shoul-
der, or hand problem interferes with normal social
activities). Two optional modules address the effects
of the arm, shoulder, or hand problem on sports, per-
forming arts, and work. Initial results suggest the
DASH is sensitive to upper limb disability and is able
to discriminate across severity of condition.18
,19
Preliminary results provide evidence of the reliability
and convergent validity of the DASH, although sup-
port for its divergent validity is poor.18
,19 The original
DASH score out of 30 is transformed to a score out of
lOa, where 0 corresponds to "no disability" and 100
corresponds to "considerable disability."
The COPM is an individualized measure designed
to detect change over time in a patient's self-percep-
tion of occupational performance?O The patient is
asked to generate a list of occupational performance
issues in the areas of self-care (hygiene, feeding,
finances), productivity (paid/unpaid work,cleaning,
play, homework), and leisure (hobbies, sports, social-
ization). Each issue is rated by the patient on a ten-
point scale, in terms of importance, performance, and
patient satisfaction with performance. We report on
performance in this study because it is most reflective
of function. Preliminary results provide evidence of
July-September 200 I 197

4. I· I I+t---_ I+t - - -_
3 Weeks 3 Weeks 3 Weeks
Stallc Volar Wrisl
Cock-Up Spllnl
Dynamic Tenodesis Dorsal Wrlsl Cock-up
Suspension Splinl wilh Dynamic Finger
Extension Spllnl
FIGURE 5. Method: single-subject experimental design. Data
were collected at baseline (no splint) and after each 3-week splint
trial.
test-retest reliability of the COPM as well as of con-
tent validity and responsiveness to clinical change;
results with respect to construct validity are mixed.21
The TEMPA, DASH, COPM, and muscle strength
test were assessed at baseline (no splint) and after the
patient wore each splint (Figure 5).
For each 3-week period of splint intervention, the
patient completed a splint-wearing schedule on
which she documented, daily, the total number of
hours that the splint was worn, the. activities per-
formed while the splint was worn, and the length of
time the splint was worn during each activity. At the
end of the study, the patient was asked to complete a
final 3-week splint-wearing schedule. During this
time, she chose the splint design she preferred to use
for each activity and documented the splint chosen,
the length of wearing time, and the activity per-
formed. Finally, at the end of the study, she complet-
ed a splint preference questionnaire outlining her
splint preferences and rationale for selection.
Data Analysis
Results were examined from a statistical as well as
a clinical perspective. For change to be statistically
significant, it is important to determine that the
change is real and not due to chance or measurement
error.22
One method for assessing statistical change is
called the minimal level of detectable change
(MDC).9,22,23 The MDC is based on the standard error
of measurement of each outcome measure and pro-
vides a confidence interval around a point estimate.
Clinicians can be confident that any value greater
than the MDC represents a true change,z2 The MDC
at the 95% confidence level was calculated for the
TEMPA, DASH, and COPM using published values
(see Appendix). The following MDC values were cal-
culated-TEMPA, 3.13; DASH, 17.23; and COPM,
2.815. This means, for example, we can be 95% confi-
dent that a true change in score on the DASH has
occurred if the difference between the baseline and
post-splint score is greater than 17.23 points.
Statistical analysis of single-subject experiments is
still relatively new, and there are controversies
regarding appropriate methods of analyzing such
data,z4 After the literature was reviewed, the MDC
method was chosen, because the patient's physical
status was stable at baseline and throughout the
study as assessed by manual muscle testing. It was
198 JOURNAL OF HAND THERAPY
reasonable to assume that the patient was clinically
stable, because she did not report changes in strength
or function over the course of the study. Methods
based on trend estimation are more appropriate if a
patient's baseline data are not stable or are fluctuat-
ing, since this may reveal a trend that continues dur-
ing the treatment intervention.25
To assess clinical significance, the number of
TEMPA tasks completed during the use of each
splint, trends identified in the splint-wearing sched-
ule, and information gathered in the splint prefer-
ence questionnaire were considered.
RESULTS
Statistical Findings
Scores on the TEMPA, DASH, and COPM at base-
line and while wearing each splint are summarized
in graph form. Figure 6 shows results for the TEMPA
functional rating scores, where a reduction in score
reflects improvement in upper extremity function.
Scores at baseline and with the static volar wrist
cock-up splint (splint I), dynamic tenodesis suspen-
sion splint (splint 2), and dorsal wrist cock-up with
dynamic finger extension splint (splint 3) were 13, 12,
7, and 3, respectively. The 95% confidence interval
around the baseline TEMPA score is shown in black.
Scores obtained during use of the dynamic tenodesis
suspension splint and the dorsal wrist cock-up with
dynamic finger extension splint fall outside the 95%
confidence interval, indicating improved hand func-
tion. Hand function did not improve during use of
the static volar wrist cock-up splint.
Figure 7 illustrates results for the DASH, in which
a reduction in score reflects improvement in the
patient's upper extremity disability and symptoms.
Scores at baseline and with the static volar wrist
cock-up splint (splint 1), dynamic tenodesis suspen-
sion splint (splint 2), and dorsal wrist cock-up with
20
16. 1
15
X = 13
X = 12
10 9.8
X =3
Sa line 3
Slatlc Volar Oyn.mlc Dorsal Wrist
Wrlsl Tenod••is Cock-up with
Cock-Up Su pension OyNmlc Flng r
Solint Splint ExtoMlon Splinl
FIGURE 6. Statistical results for the Test Evaluant Les Membres
Superieurs des Personnes Agees functional rating scores.

5. dynamic finger extension splint (splint 3) were 62.5,
51.6,36.25, and 26 respectively. The 95% confidence
interval around the baseline DASH score is shown in
black. DASH scores obtained during use of the
dynamic tenodesis suspension splint and the dorsal
wrist cock-up with dynamic finger extension splint
fall outside this confidence interval, indicating with
95% confidence that these splints improved the
patient's disability and symptoms, whereas the static
volar wrist cock-up splint did not.
Finally, the COPM performance rating scores are
shown in Figure 8. These numbers reflect the
patient's perception of performance while wearing
each of the three splints; improvement on this meas-
ure is indicated by an increase in score. Scores at
baseline and with the static volar wrist cock-up splint
(splint 1), dynamic tenodesis suspension splint
(splint 2), and dorsal wrist cock-up with dynamic fin-
ger extension splint (splint 3) were 3, 4.8, 7.2, and 7.6,
respectively. The 95% confidence interval is shown in
black. Again, scores with the dynamic tenodesis sus-
pension splint and the dorsal wrist cock-up with
dynamic finger extension splint fall outside of the
95% confidence interval, indicating improvement in
hand function. Hand function did not improve when
the patient used the static volar wrist cock-up splint.
Clinical Findings
Functionally, the patient was unable to complete
three TEMPA tasks both at baseline and with the stat-
ic volar wrist cock-up splint. However, with the
dynamic tenodesis suspension splint and the dorsal
wrist cock-up with dynamic finger extension splint,
she was able to complete all nine TEMPA tasks.
The total number of hours that each splint was
worn while the patient performed daily tasks and the
average number of hours that the splint was used
each day were calculated from the splint-wearing
schedule. The static volar wrist cock-up splint was
100
90
80
70
60
50
40
30
20
10
X= 62.5
--
Baseline
X = 51.6
Static Volar
Wrist
Cock·Up
Splint
1- - - - 1- 79. 7
45.3
x =36.25
X =26
3
Dynamic Dorsal Wrist
Tenodesis Cock·up with
Susponslon Dynamic Finger
Splint Extension Splint
FIGURE 7. Statistical results for the Disabilities of the
Shoulder, Arm and Hand questionnaire scores.
.0
6 ;-_ __
a....".
X-4.8
StaUcVolar
Wrll!
Cock-tJp
Splint
X =7.2
2
DyNlm1c
T.nod....
Suspension
SpUnt
X =7.&
Ooru. Wrlst
Cock-upwlth
DyNlm1c Finger
Ext.nslon Spin!
5.8
.2
FIGURE 8. Statistical results for the Canadian Occupational
Performance Measure performance rating score.
worn 4.5 hours a day, compared with the dynamic
tenodesis suspension splint and the dorsal wrist
cock-up with dynamic finger extension splint, which
were both worn 6.8 hours a day. Although these data
were not analyzed statistically, the cumulative time
that each of the two dynamic splints was worn, when
compared with the time the static volar wrist cock-up
splint was worn, appears to be longer. A list of activ-
ities performed while each splint was worn and the
length of time the splint was worn during each activ-
ity were also collected. No obvious trends in these
data were apparent.
The patient was asked to complete the splint-wear-
ing schedule one more time over a 3-week post-study
period. During this time, the patient was allowed to
choose whichever of the three splints she preferred
for performing activities of daily living. She chose the
static volar wrist cock-up splint (2.4 hours per day)
and the dorsal wrist cock-up with dynamic finger
extension splint (1.8 hours per day). She did not use
the dynamic tenodesis suspension splint at any time
during this 3-week period.
The splint preference questionnaire showed that the
patient believed that her hand function improved with
both dynamic splints, more with the dorsal wrist cock-
up with dynamic finger extension splint than with the
dynamic tenodesis suspension splint. Interestingly,
however, despite the improvement in function, she
preferred the static volar wrist cock-up splint because
it offered support, was easy to put on, and was less
conspicuous to wear than the other two splints.
DISCUSSION
The importance of demonstrating treatment effec-
tiveness is acknowledged and accepted by clinicians.
Splinting is an intervention used frequently by hand
therapists to treat patients with radial nerve palsy.
Although splinting options for radial nerve palsy are
described in the literature, published works compar-
ing designs could not be located.
July-September 200 I 199

6. This study shows that the dynamic tenodesis sus-
pension splint and the dorsal wrist cock-up with
dynamic finger extension splint statistically im-
proved (in comparison with baseline scores on the
TEMPA, DASH, and COPM) the hand function of
this 83-year-old female patient with radial nerve
palsy, whereas the static volar wrist cock-up splint
did not. In addition, the patient was able to complete
all TEMPA tasks while using the two dynamic splints
but was unable to complete three tasks while using
the static volar wrist cock-up splint.
Regardless of evaluation scores, our clinical goal is
to fabricate a splint that improves function and that
the patient will wear. In this single-subject experi-
ment, only the dorsal wrist cock-up with dynamic
finger extension splint met these criteria. According
to our statistical analyses, this splint improved the
patient's hand function, and, as documented in the
post-study diary, the patient used it for daily activi-
ties. In the splint preference questionnaire, she
reported better hand function while wearing the dor-
sal wrist cock-up with dynamic finger extension
splint than with the dynamic tenodesis suspension
and static volar wrist cock-up splints. Although the
static volar wrist cock-up splint did not statistically
improve her hand function, she preferred to wear
this splint because it was supportive, easy to apply,
and less noticeable. This reinforces the importance of
ensuring that dynamic splints are comfortable, light-
weight, and inconspicuous.
The study has three limitations. First, we were
unable to blind testers to the intervention being eval-
uated because the patient was required to wear each
splint when completing the TEMPA assessment.
However, the use of standardized assessment proce-
dures should help minimize any potential bias this
may have introduced. Second, re-administration of
the TEMPA without splint at the end of each 3-week
interval may have further supported our clinical
impression that no change in physical status had
occurred over the 12-week period of this study.
Finally, the two dynamic splints had an added com-
ponent to facilitate thumb abduction for grasp-and-
release activities (e.g., picking up small objects such
as coins or a lipstick, picking up a glass or container,
turning a key, opening a tight jar), whereas the static
volar wrist cock-up splint did not. The addition of
this component may have magnified the functional
differences identified statistically and clinically
between the two dynamic splints and the static volar
wrist cock-up splint.
CONCLUSION
This paper describes a process for the systematic
evaluation of treatment interventions using a single-
subject experimental design. This approach should
encourage experimentation by clinicians, because it is
200 JOURNAL OF HAND THERAPY
easier than a randomized controlled trial to conduct
in clinical settings, particularly with relatively rare
conditions like radial nerve palsy.
In this experiment, the patient's hand function
improved statistically and clinically during use of the
dorsal wrist cock-up with dynamic finger extension
splint and the dynamic tenodesis suspension splint.
However, when given the choice of wearing one of
these two dynamic splints, the patient wore only the
dorsal wrist cock-up with dynamic finger extension
splint during activities of daily living.
Although results of this study cannot be general-
ized to all patients with radial nerve palsy, a series of
Single-subject experiments comparing splint effec-
tiveness in patients in this group could make gener-
alization possible in the future.
REFERENCES
1. Colditz Je. Splinting for radial nerve palsy. J Hand Ther.
1987;1:18-23.
2. Reid RL. Radial nerve palsy. Hand Clin. 1988;4:179--85.
3. Colditz Je. Splinting the hand with a peripheral nerve injury.
In: Hunter JM, Mackin EJ, Callahan MS (eds). Rehabilitation of
the Hand: Surgery and Therapy. St. Louis, Mo.: Mosby,
1995:679-700.
4. Ellis M. Orthoses for the hand. In: Lamb DW, Kuczynski K
(eds). The Practice of Hand Surgery. London, UK: Blackwell
Scientific,1981:529-41.
5. Thomas FB. An improved splint for radial (musculospiral)
nerve paralysis. JBone Joint Surg. 1951;33B:272-3.
6. Capener N. Lively splints. Physiotherapy. 1967;53:371-4.
7. Penner DA. Dorsal splint for radial palsy. Am J Occup Ther.
1972;26:46-7.
8. McKee P, Morgan L. Orthotics in Rehabilitation: Splinting the
Hand and Body. Philadelphia, Pa.: F. A. Davis, 1998:112-4.
9. Ottenbacher KJ, Johnson MB, Hojem M. The significance of
clinical change and clinical change of significance: issues and
methods. Am J Occup Ther. 1988;42:156-63.
10. Elder JH. Single-subject experimentation for psychiatric nurs-
ing. Arch Psychiat Nurs. 1997;11:133--8.
11. Guyatt G, Sackett D, Taylor DW, Chong J, Roberts R, Pugsley S.
Determining optimal therapy: randomized trials in individual
patients. N Engl J Med. 1986;314:889-92.
12. Ottenbacher KJ. Clinically relevant designs for rehabilitation
research: the ideographic model. Am J Phys Med Rehabil.
1991;70:S144-50.
13. Kazdin AE. Single Case Research Designs: Methods for
Clinical and Applied Settings, New York: Oxford University
Press, 1982:103--25.
14. Gonnella e. Designs for clinical research. Phys Ther.
1973;53:1276-83.
15. Desrosiers J, Hebert R, Dutil E, Bravo G. Development and
reliability of an upper ~xtremity function test for the elderly:
the TEMPA. Can JOccup Ther. 1993;60:9-16.
16. Desrosiers J, Hebert R, Dutil E, Bravo G, Mercier L. Validity of
the TEMPA: a measurement instrument for upper extremity
performance. Occup Ther JRes. 1994;14:267-81.
17. Hudak P, Amadio P, Bombardier C, and the Upper Extremity
Collaborative Group. Development of an upper extremity out-
come measure: the DASH (Disabilities of the Arm, Shoulder,
and Hand). Am JInd Med. 1996;29:602--8.
18. Upper Extremity Collaborative Group. Measuring disability
and symptoms of the upper limb: a validation study of the
DASH questionnaire. Arthritis Rheum. 1996;39:S112.
19 Turchin DC, Beaton DE, Richards RR. Validity of observer-
based aggregate scoring systems as descriptors of elbow pain,

7. function and disability. J Bone Joint Surg. 1998;80A:154-62.
20. Law M, Baptiste S, Carswell A, McColl M, Polatajko H, Pollock N.
Canadian Occupational Performance Measure, 2nd ed. Toronto,
Ontario, Canada: CAOT Publications ACE, 1994.
21. Chan CCH, Lee TMC. Validity of the Canadian Occupational
Performance Measure. Occup Ther Int. 1997;4:229-47.
22. Stratford PW, Finch E, Solomon P, Binkley J, Gill e. Moreland
J. Using the Roland Morris Questionnaire to make decisions
about individual patients. Physiother Can. 1996;48:107-10.
23. Jacobson NS, Follette We. Revenstorf D. Psychotherapy out-
come research: methods for reporting variability and evaluat-
ing clinical significance. Behav Ther. 1984;15:336-52.
24. Hojem MA, Ottenbacher, KJ. Empirical investigation of visual-
inspection vs. trend-line analysis of single-subject data. Phys
Ther. 1988;68:983-8.
25. Nourbakhsh MR, Ottenbacher, KJ. The statistical analysis of
single-subject data: a comparative examination. Phys Ther.
1994;74:768-76.
ApPENDIX
Calculation of Minimal Level of Detectable Change22
The minimal level of detectable change (MDC) is calculat-
ed as follows:
MDC = SEM x1.96 x,j2.
where SEM is the standard error of measurement.
The SEM is calculated as:
SEM=utb-R
where u t is the total variance for the test.
Test Evaluant Les Membres Superieurs des
Personnes Agees (TEMPA)16
One hundred and four men and women who had various
upper limb impairments participated in this study. Their
ages ranged from 60 to 94 years. Concurrent validity was
tested by examining the agreement of TEMPA scores with
scores for the Action Research Arm Test and the Box and
Block Test. Construct validity was explored by testing two
hypotheses relating upper limb performance with func-
tional independence.
ICC (test-retest for total functional rating score) = 0.98
SD (total functional rating score) = -8.0
MDC = (~-82 xb -0.98) x1.96 x,j2.
MDC = 3.13
Disabilities of the Arm, Shoulder and Hand
(DASH) Questionnaire19
Sixty-nine patients were treated at one of two tertiary refer-
ral clinics for problems related to the elbow-either non-
surgical management or postoperative reassessment.
ICC (test-retest) = 0.92
SD=22
MDC = (,J-22z x~1-0.92) x1.96 x,j2.
MDC = 17.23
Canadian Occupational Performance
Measure (COPM)20
Twenty-seven patients over the age of 65 years were receiv-
ing rehabilitation services for a variety of neurologic and
musculoskeletal disorders.
ICC (test-retest for performance component) = 0.63
SD (performance component) = 1.67
MDC = (~1.672 xb -0.63) x1.96 x,j2.
MDC =2.815
July-September 200 I 20 I