Archives of Physical Medicine and Rehabilitation

journal homepage: www.archives-pmr.org

Archives of Physical Medicine and Rehabilitation 2016;97:1777-84

ORIGINAL RESEARCH

Effectiveness of Group Wheelchair Skills Training for

People With Spinal Cord Injury: A Randomized
Controlled Trial
Lynn A. Worobey, PhD, DPT, ATP,a R. Lee Kirby, MD,b Allen W. Heinemann, PhD,c,d
Emily A. Krobot, MSW,a Trevor A. Dyson-Hudson, MD,e,f Rachel E. Cowan, PhD,g
Jessica Presperin Pedersen, OTR/L,d Mary Shea, OTR/L,h Michael L. Boninger, MDa
From the aDepartment of Physical Medicine and Rehabilitation, University of Pittsburgh, Pittsburgh, PA; bDivision of Physical Medicine and
Rehabilitation, Dalhousie University, Halifax, NS, Canada; cFeinberg School of Medicine, Northwestern University, Chicago, IL; dRehabilitation
Institute of Chicago, Chicago, IL; eKessler Foundation, West Orange, NJ; fDepartment of Physical Medicine and Rehabilitation, Rutgers New
Jersey Medical School, Newark, NJ; gDepartment of Neurological Surgery, University of Miami, Miami, FL; and hKessler Institute for
Rehabilitation, West Orange, NJ.

Abstract
Objective: To assess the effectiveness of group wheelchair skills training to elicit improvements in wheelchair skills.
Design: Randomized double-blinded controlled trial.
Setting: Four Spinal Cord Injury Model Systems Centers.
Participants: Manual wheelchair users with spinal cord injury (NZ114).
Intervention: Six 90-minute group Wheelchair Skills Training Program (WSTP) classes or two 1-hour active control sessions with 6 to 10 people
per group.
Main Outcome Measures: Baseline (t1) and 1-month follow-up (t2) Wheelchair Skills Test Questionnaire (WST-Q) (Version 4.2) for capacity and
performance and Goal Attainment Scale (GAS) score.
Results: Follow-up was completed by 79 participants (WSTP: nZ36, active control: nZ43). No differences were found between missing and
complete cases. Many users were highly skilled at baseline with a WST-Q capacity interquartile range of 77% to 97%. There were no differences
between groups at baseline in WST-Q measures or demographics. Compared with the active control group, the WSTP group improved in WST-Q
capacity advanced score (PZ.02) but not in WST-Q capacity or WST-Q performance total scores (PZ.068 and PZ.873, respectively). The
average GAS score (0% at t1) for the WSTP group at t2 was 65.6%34.8%. Higher GAS scores and WST-Q capacity scores were found for those
who attended more classes and had lower baseline skills.
Conclusions: Group training can improve advanced wheelchair skills capacity and facilitate achievement of individually set goals. Lower skill
levels at baseline and increased attendance were correlated with greater improvement.
Archives of Physical Medicine and Rehabilitation 2016;97:1777-84
ª 2016 by the American Congress of Rehabilitation Medicine

Presented in part to the American Congress of Rehabilitation Medicine, October 28e31, 2015,
Dallas, TX.
Supported by the National Institute on Disability, Independent Living, and Rehabilitation
Research (NIDILRR) (grant no. 90DP0025).
The NIDILRR is a center within the Administration for Community Living (ACL), Depart-
ment of Health and Human Services (HHS). The contents of this article do not necessarily represent
the policy of the NIDILRR, ACL, HHS, and you should not assume endorsement by the Federal
Government.
Clinical Trials Registration No.: NCT01807728.
Disclosures: none.
The number of people using wheelchairs for mobility in the United
States is estimated to exceed 3.3 million.1 Unfortunately, many
users lack skills crucial for independence, safety, and upper limb
preservation.2,3 Individualized manual wheelchair skills training
can improve skills4-10; however, such training is increasingly
difficult to provide with decreasing lengths of rehabilitation.11
Among those with spinal cord dysfunction, median length of stay
in the United States decreased from 21 to 14 days from 1994 to

0003-9993/16/$36 - see front matter ª 2016 by the American Congress of Rehabilitation Medicine
http://dx.doi.org/10.1016/j.apmr.2016.04.006
1778
2001.12 This trend has serious potential consequences for newly
injured individuals who do not receive training necessary to master
skills for independence and mobility and perform them safely. A
2012 study found that over a patient’s rehabilitation stay
(55.736.6d) an average of 4.25.9 hours was spent on manual
wheelchair mobility skills.13 A recent Canadian study reported an
average of 1 to 4 hours of wheelchair training in rehabilitation
centers, with advanced skills training reported by only 11.8% of
therapists.14 More efficient methods of teaching these skills are
needed in both inpatient and outpatient settings.
Time and resources are cited as barriers by clinicians to
providing wheelchair skills training.14 These barriers increase the
burden for clinicians providing wheelchair skills training in the
community. Group training provides a format that is less
demanding on clinicians’ time because they can train multiple
individuals at once rather than holding sequential 1 on 1 classes on
the same topics. A 2013 study found manual wheelchair mobility
was the third most common type of group therapy provided during
inpatient rehabilitation.15 Wheelchair skills training has been
provided to small (nZ1e3)16 and large cohorts (nZ14e19)17 of
occupational therapy students and medical students (nZ12).6
Through massed practice with a single session of 2 to 4 hours,
participants increased skill capacity and knowledge. Training pairs
of wheelchair users over multiple sessions can increase self-
efficacy and skills.18 However, to our knowledge, no studies
have implemented this training in groups of wheelchair users.
The primary goal of this study was to evaluate the effectiveness
of group wheelchair skills training among people with spinal cord
injury (SCI) to improve wheelchair skills and achieve individual
goals. We hypothesized that people randomized to a group that re-
ceives the Wheelchair Skills Training Program (WSTP) would have
greater improvements in capacity and performance of wheelchair
skills compared with an active control group and be more likely to
achieve individually set goals. Because of ceiling effects,
improvement in capacity for advanced wheelchair skills was also
investigated. Additionally, among those receiving the WSTP, we
sought to determine if subject characteristics, baseline skill level,
and attendance were related to improvements in skill and goal
achievement.
Methods
Design
This study was a double-blind randomized controlled trial; part of
the baseline data of which is being published separately.19
Setting
The study was conducted at 4 Spinal Cord Injury Model System
Centers: Midwest Regional SCI Care System, Northern New
Jersey SCI System, South Florida SCI System, and University of
Pittsburgh Model Center on SCI.
List of abbreviations:
GAS Goal Attainment Scale
SCI spinal cord injury
WSP Wheelchair Skills Program
WSTP Wheelchair Skills Training Program
WST-Q Wheelchair Skills Test Questionnaire
L.A. Worobey et al
Group allocation
Two rounds of training were held at each site with 12 to 20 par-
ticipants enrolled in each round. Participants were randomized to
either a WSTP or active control group, with 6 to 10 participants per
group. Randomization codes were generated prior to the start of the
study for a 1:1 allocation ratio and stored in an Excel spreadsheet.a
Randomization was stratified by site and completed using permu-
tated blocks of 2 or 4 based on level of injury (paraplegia or
tetraplegia) and years since injury (<1y or 1y). Allocation was
concealed with study members at individual sites contacting the
study coordinator at the lead site after completing informed consent
and prior to baseline to receive the randomization assignment.
Blinding
Participants were concealed to randomization and unaware of the
2 training groups; they were only aware that they were enrolling in
a study of training programs for people with SCI. Data collectors
were not aware of group allocation.
Recruitment and screening
A convenience sample of wheelchair users was enrolled from
October 2013 through September 2014 through advertisements,
research registries, and word of mouth. Individuals eligible to
participate in this study were aged 18 to 75 years, had a
nonprogressive SCI (traumatic or nontraumatic), were living in the
community, used a manual wheelchair as a primary means of
mobility (50% mobility), independently propelled a wheelchair,
scored 23 on the Folstein Mini-Mental State Examination, and
completed the baseline evaluation.
Participants
Based on a study by Routhier et al,7 with an effect size of .48,
power of 0.8, and aZ.05, we determined we needed a sample of
110. Self-reported age, sex, height, weight, race, marital status,
place of residence, highest level of formal education, SCI level
(tetraplegia [C8 and higher] or paraplegia [below C8]), length of
time using a wheelchair, and age of wheelchair used most often at
baseline were entered electronically at the National Institutes of
Health Assessment Center.b
Interventions
Each class was taught by 2 trainers who attended a Wheelchair
Skills Program (WSP) course, described in supplemental table S1
(available online only at http://www.archives-pmr.org/). A total of
8 weekly 90-minute WSTP classes were held (6 regular and 2
make-up). WSTP participants were asked to attend a target of 6
classes. Classes involved hands-on demonstrations and practice of
wheelchair skills using the principles and procedures outlined in the
WSP manual and made accessible online through the WSP web-
site.20 Prior to the first session, trainers received the results of the
baseline WST-Q. Sites were provided a sample class schedule
(supplemental table S2, available online only at http://www.
archives-pmr.org/); however, local circumstances, group mix, and
participant goals necessitated that the sessions proceed at different
paces and that some tailoring take place. No adverse consequences
were experienced during the WSTP.
www.archives-pmr.org
Group wheelchair skills training 1779

The active control group was asked to attend two 1-hour general
Table 1 Participant goals grouped into categories
education classes that were scheduled 1 to 3 weeks apart. A reha-
bilitation therapist, counselor, or peer counselor led the classes Participants Mean No.
using an informational PowerPoint presentation,a and participants Participants Achieved of Sessions
had the opportunity to interact as a group. Class topics were aging With Goal Goal Worked on
with an SCI and weight management and nutrition. Classes were Goal Category (n) (%) Goal  SD
held between weeks 4 and 6 of the WSTP group sessions so that Categories relating to ascending curbs, inclines, and stairs
blinding could be maintained during the follow-up data collection. Ascend 20 65 1.40.7
stairs
Outcome measures Ascend 20 60 2.21.5
high curb
Wheelchair Skills Test Questionnaire Ascend 10 90 31.9
We used the Wheelchair Skills Test Questionnaire (WST-Q) 10 incline
(Version 4.2) for self-propelled manual wheelchairs at baseline (t1) Ascend 6 83 3.51.2
and 1-month posttraining follow-up (t2).19,21 The WST-Q consists low curb
of 32 individual skills that are grouped into indoor, community, and Ascend 1 100 30
advanced skill levels. We collected data on capacity and perfor- 5 incline
mance. Total and advanced-level subtotal WST-Q capacity and Categories relating to descending curbs, inclines, and stairs
performance percentage scores were calculated as defined by the Descend 22 64 1.91.3
WSP manual (100%total raw score/possible total raw score). stairs
Down 11 91 2.81.5
Goal Attainment Scale score high curb
Participants in the WSTP only worked with trainers to complete a Descends 4 100 3.51.7
Goal Attainment Scale (GAS) outlining individual skills they steep incline
wanted to improve. These were reassessed in each of the first 3 Categories relating to wheelie skills
classes and entered as free text into a database. Trainers recorded Stationary 17 59 3.72.1
the sessions attended by participants in which each goal was wheelie
addressed and whether participants were successful with goals Wheelie 9 78 3.62.6
(yes/no). A GAS score was calculated for each participant after incline descent
training sessions were completed (number of goals met/number of Wheelie 6 100 3.52.5
goals set100%). The baseline GAS score was 0% by definition. curb descent
Turn in 3 100 4.73.1
Statistical analysis wheelie position
Moving 2 100 32.8
Using intention to treat principles, we solicited follow-up data on all wheelie
participants regardless of the number of classes attended, unless Categories relating to moving wheelchair skills
they withdrew from the study. Complete case analysis was used Efficient 9 100 2.571.2
including only subjects with t2 data. An ordinal variable was wheelchair
created for missing/complete cases, and missing data patterns were propulsion
studied. Descriptive statistics were calculated for all measures, with Get through 3 67 1.37
normality assessed with the Kolmogorov-Smirnov test. To evaluate hinged door
between-group differences, we used independent t tests for nor- Roll across 1 100 10
mally distributed continuous variables (WST-Q performance), side slope
Mann-Whitney U test for non-normally distributed continuous Roll on soft 2 100 10
variables (WST-Q capacity and GAS score), and chi-square test (or surface
Fisher exact test when applicable) for categorical variables. Based Categories relating to stationary wheelchair skills
on skew found in WST-Q capacity, we also investigated between- Folds and 2 0 00
group differences in the advanced-level subscore because this unfolds
measure may be more sensitive to improvement among individuals wheelchair
with a higher baseline skill level.7 We also completed multivariate Reach high 1 100 28
analysis using a 2-way mixed analysis of variance based on group object
and time after a normative quantile transformation for WST-Q ca- Pick up object 1 100 09
pacity total score and advanced subscore.21 Between-group com- from floor
parisons were made based on demographic variables (level of Categories relating to transfer skills
injury, sex, and years since injury), group (WSTP or active control), Transfer floor 17 53 1.61.3
and attrition (complete or missing). Level of injury and years since to chair
injury were treated as dichotomous variables (paraplegia/tetraplegia Transfer between 3 0 00
and <1/1y). Exploratory correlational analysis was completed to surfaces
determine characteristics correlated with improved GAS scores and Transfer chair 2 100 2.50.7
WST-Q capacity scores in the WSTP group. Similar participant to floor
goals were also grouped and analyzed based on achievement in

www.archives-pmr.org
1780
functional categories; these categories are listed in table 1 with the
number of sessions and rate of achievement. Several goals were
removed from analysis because they were not feasible based on the
study environment (pool transfers or escalator use), not related to
wheelchair skills (shoulder care), or not appropriate goals (wheelie
incline ascent). IBM SPSS Statistics 21c was used for statistical
analyses using aZ.05.
Results
Participants
We enrolled 114 manual wheelchair users (WSTP: nZ55, active
control: nZ59); of these, 93 attended at least 1 class (WSTP: nZ44,
Fig 1 Consolidated Standards of Reporting Trials diagram showing the flow of patient participation through each stage of the study.
L.A. Worobey et al
active control: nZ49), and 79 completed follow-up (WSTP: nZ36,
active control: nZ43) (fig 1). There were no differences in de-
mographic, clinical or wheelchair-usage characteristics based on
group assignment or follow-up completion (table 2; supplemental
table S3, available online only at http://www.archives-pmr.org/),
except that fewer women in the WSTP group had missing data
(c2Z6.92, PZ.008). The median number of classes was lower for the
missing data group (1; interquartile range, 0e2) versus the complete
data group (5; interquartile range, 3e6; P<.001). Including make-up
classes, class size ranged from 1 to 9 participants.
Wheelchair Skills Test Questionnaire
No significant differences were found between participants
with missing and complete data for baseline WST-Q capacity or
www.archives-pmr.org
Group wheelchair skills training 1781

of classes had lower baseline WST-Q capacity and performance

Table 2 Participant demographic, clinical, and wheelchair data
scores (RZ .304, PZ.024 and RZ.458, P<.001, respectively).
among those completing 1-month follow-up
Wheelchair Skills Difference
Subject Training Program Active Control Between GAS score
Characteristics Group (nZ36) Group (nZ43) Groups, P
There were 42 WSTP participants who completed the GAS. The
Age (y) 40.111.4 4112.4 .76 average GAS score was 65.6%34.8%, with 4.02.2 goals per
Height (m) 1.70.35 1.70.13 .21 person. Seven people had a GAS score of 0% at t2, whereas 13
Weight (kg) 187.830.4 187.551.6 .78 scored 100%. Reasons documented by trainers for lack of
Years of wheelchair 13.511.8 12.912.5 .83 achievement included participants attending only 1 session,
use missing the session where the goal was worked on, declining to
Age of wheelchair 32.3 3.22.2 .65 attempt the skill, or being able to direct someone to assist with the
Paraplegia 29 (80.6) 31 (72.1) .38* skill but not complete the skill independently. A higher GAS score
<1y since injury 0 (0) 1 (2.3) .77* was found for men (PZ.019). The 5 most frequently listed goals
Male 32 (88.9) 37 (85.0) .71* and achievement rates were descending stairs (64%), ascending
Race .31* stairs (65%), ascending high curbs (60%), stationary wheelies
White 21 (58.3) 22 (51.2) (59%), and floor to wheelchair transfers (53%). The average
Black 11 (30.6) 18 (41.9) number of classes spent on a skill ranged from 0 to 4.7 and was
Asian Indian 0 (0) 1 (2.3) highest for wheelie skills. A higher GAS score was associated
Not provided 4 (11.1) 1 (2.3) with greater class attendance (RZ.531, PZ.001). Those who set
Residence .22* more goals attended more classes (RZ0.362, PZ.018).
Private 32 (88.9) 42 (97.7)
Group living 1 (2.8) 0 (0)
Nursing home 0 (0) 1 (2) Discussion
Hotel or motel 1 (2.8) 0 (0)
Homeless 2 (5.6) 0 (8.6) Compared with the active control group, the WSTP group
Level of education .86* significantly improved in WST-Q capacity advanced score but not
Less than HS 4 (11.1) 4 (9) total score. Change in WST-Q capacity advanced score was
diploma greater for the WSTP group; however, higher scores at follow-up
HS diploma or 15 (41.7) 17 (39.5) were found for both groups. This is likely tied to improvements in
GED higher-level skills, which were set as goals by participants and
Associate’s or 11 (30.5) 18 (41.9) represented during a larger number of classes. Previous studies of
Bachelor’s group training with clinicians and students have shown an increase
degree in wheelchair capacity after intervention.6,16,17 However, these
Graduate degree 5 (13.9) 3 (7.0) studies had lower t1 WST-Q scores secondary to inclusion of able-
Other 1 (2.8) 1 (2.3) bodied participants. The baseline skill of participants in our study
was also higher than other studies of individualized training with
NOTE. Values are mean  SD, n (%), or as otherwise indicated.
Abbreviations: GED, General Educational Development; HS, high wheelchair users, which showed improvement with training from
school. 63.0% to 78.5%9 and 64.9% to 80.9%.8 Perhaps the closest study
* Chi-square, P value. in design to ours was completed by Best et al,18 in which
community-based wheelchair users attended six 30-minute
training sessions. The effect size of our between-group differences
indicates our study may have been underpowered to detect dif-
WST-Q performance. Multivariate analyses revealed no signifi- ferences. A 3% change in WST-Q score corresponds approxi-
cant differences for WST-Q capacity total score. Differences were mately to the mastery of 1 skill. The upper quartile of our
found for WST-Q capacity advanced subscore with main effects of intervention group was 97%, indicating that our outcome measure
time (PZ.003) and group (PZ.047) but no interaction effect could at most measure improvement in 1 item. We did not find
(PZ.128). Baseline, change scores, and between-group compari- increased performance of skills in the WSTP group compared with
sons for WST-Q total and advanced-level subscores can be found the active control group. The large range in WST-Q performance
in table 3. Figure 2 shows box plots of t1 and t2 WST-Q scores. scores may indicate differences in environment or demands of
Change in WST-Q capacity advanced-level subscore was higher activities of daily living dictate how often skills are performed.
for the WSTP group than the active control group (PZ.02). The limited ability to detect a change in WST-Q capacity with
Change in WST-Q capacity total score was higher for the WSTP ceiling effects increases the importance of the GAS results. At t1,
group, but not significantly (PZ.068). There were no significant the GAS score was 0%; therefore, a t2 average score of 65.6% is a
differences in WST-Q performance scores between groups. The significant accomplishment. Percent achievement for stationary
maximum WST-Q capacity score was achieved by 25% of WSTP wheelie (59%) and curb ascent (60%) exceeded that by MacPhee
participants at t2. et al,8 who provided individualized training to wheelchair users
Greater improvement in capacity at follow-up was associated (30% and 0%, respectively). Many wheelchairs users cannot
with lower baseline WST-Q capacity scores (RZ .396, PZ.017). complete a wheelie2; a previous study found this skill took
Baseline WST-Q performance was significantly lower for those comparatively greater time to learn than other wheelchair skills.16
that attended at least 1 class (73.6%12.3%) versus no classes Wheelies accounted for the highest number of sessions in our
(84.2%14.1%; PZ.020). Those who attended a greater number intervention. Future monitoring of the time spent on each skill

www.archives-pmr.org
1782 L.A. Worobey et al

Table 3 Differences between groups in WST-Q capacity and performance scores among individuals completing follow-up
WSTP Group Active Control Group Between-Group Difference
Outcome Measure n Median IQR (Q1 to Q3) n Median IQR (Q1 to Q3) P Effect Size
WST-Q capacity total score
Baseline score 36 88 (77 to 97) 43 84 (76 to 94) .287 .12
Change in score 36 2 ( 0.8 to 6) 43 0 ( 5 to 4) .068 .21
WST-Q capacity advanced subscore
Baseline score 36 60 (30 to 90) 43 50 (30 to 80) .454 .08
Change in score 36 5.5 (0 to 20) 43 0 ( 10 to 7) .020* .26
n Mean  SD n Mean  SD P Effect Size
WST-Q performance total score
Baseline 36 73.412.1 43 72.412.8 .705 .09
Change in score 36 0.4714.4 43 0.057.2 .873 .04
Abbreviations: IQR, interquartile range; Q, quartile.
* Indicates significant between-group difference.

during sessions may help inform our understanding of training
dosage. Group training can provide effective goal attainment with
potentially greater efficiency because less one-on-one time is
required. A structured group training program may be a useful
resource for therapists. Regardless of the setting, group training
provides an opportunity for participants to learn from their
peers, and participants with the same skill levels may have
similar goals.
To maximize goal achievement of participants, clinicians
should be cautious of the effect of class attendance and the
importance of careful goal revision with participants. For goal
categories with <75% participant achievement, most were worked
on for an average of <2 classes. Unlike one-on-one training, the
days and times and content of group classes may not match
each person’s schedule,6,16 and scheduling limitations may have
contributed to decreased GAS scores. Inability to attend group
training secondary to scheduling issues has also been found
in group WSP training with clinicians and students.6,16 Rather
than following a syllabus that includes all WSP skills, training
may be more efficient when focused on skills that group members
want to improve. Clinicians should also be mindful of goal setting
at the start of the intervention. Of the goals listed by participants,
descending stairs and ascending high curbs were the most
prevalent. These advanced skills are not necessarily achievable
(at least for people with tetraplegia) without assistance. Addi-
tionally, many people were not able to formulate 5 goals, leading
to achievement of each goal carrying more weight.
The success of the intervention was related to subject char-
acteristics, with improved outcomes among those who attended
more classes and increased participation among those who had
lower baseline skills and more goals. There was no relation
between baseline WST-Q score and GAS score, indicating base-
line ability does not determine the ability to attain individualized
goals. This lack of correlation may also imply that users with
lower skills were more likely to set achievable goals and therefore
had increased self-awareness of their level of ability.
Although all longitudinal studies are subject to dropout, rates
were particularly high in this study. Dropout may be attributed in
part to the double-blinded study design, which led to enrollment of
people with high baseline skills and people who did not have an
interest in learning wheelchair skills. This possibility is supported
by the fact that those with higher baseline WST-Q scores attended
fewer classes and were less likely to complete the follow-up. A
trial that does not conceal the intervention (eg, using either a
waitlist control group or comparison with one-on-one training)
may help to ensure participants attend classes.

Fig 2 Boxplots of WST-Q[C] score, WST-Q[C] advanced-level subscore, and WST-Q[P] score for WSTP group and ACG at t1 and t2. Abbreviations:
ACG, active control group; WST-Q[C], WST-Q capacity; WST-Q[P], WST-Q performance.

www.archives-pmr.org
Group wheelchair skills training
Study limitations
A convenience sample was used for study of individuals at 4
Spinal Cord Injury Model Systems Centers. This sample may not
be representative of the entire SCI population. A large portion
(30.7%) of the sample did not complete follow-up. Although there
were no statistically significant differences between missing data
cases and complete cases, the exclusion of missing data cases may
have introduced bias. The number of classes attended by partici-
pants varied, and caution should be used when interpreting the
effects of group training and dosage. The follow-up period was
short, and future studies with longer follow-up periods are needed
to evaluate skill retention. The psychometric properties of WST-Q
subscores have not been established. Although all sites followed
the same procedure manual, group mix and individual goals of
participants across sites necessitated tailoring. The t1 testing was
completed after enrollment. The time from t1 to intervention
varied, and this may have affected retention.
Future studies should focus on recruiting people who have
skills to learn, potentially through a cutoff for maximum skill
level. A WSTP trainer at each site was a wheelchair user, and the
influence of peer trainers working in collaboration with therapists
may be an important issue to consider in future studies. Despite
the loss of blinding, improved retention may be found with a wait-
list control design in which participants know the content of the
intervention. Future studies may also benefit from the direct
comparison of group training with individualized training; such
studies may help draw attention to the benefits and shortcomings
of group training. Although the WST-Q has a high correlation with
the WST,19 an objective measure of skills at baseline and follow-
up may prevent potential inaccuracies in self-reporting in the
future. Because the WST is a useful way to identify how a skill is
performed, it is complementary to the WST-Q results; ideally, they
should be used together. Options need to be considered for
eliminating the ceiling effect of the WST-Q and WST. Individuals
<1 year since injury may not have reached complete neurologic
recovery; however, there were no differences between groups
based on time since injury, and this cohort was small. Despite
the limitations of this study and the need for future work, this is
the first randomized controlled interventional study to evaluate the
effectiveness of group wheelchair skills training in a large group
of manual wheelchair users with SCI.
Conclusions
Group training can improve capacity to complete advanced-level
manual wheelchair skills and facilitate achievement of individu-
ally set goals. Lower skill levels at baseline and increased atten-
dance were correlated with greater improvement.
Suppliers
a. Microsoft.
b. Assessment Center platform, 457; National Institutes of Health.
Available at: www.assessmentcenter.net.
c. IBM SPSS Statistics 21; IBM.
www.archives-pmr.org
1783
Keywords
Motor skills; Rehabilitation; Spinal cord injuries; Wheelchairs
Corresponding author
Lynn A. Worobey, PhD, DPT, ATP, Human Engineering Research
Laboratories, VA Pittsburgh Healthcare System, 6425 Penn Ave,
Ste 400, Pittsburgh, PA 15206. E-mail address: law93@pitt.edu.
Acknowledgments
We thank Sara Jerousek, BS, Michelle Oyster, MS, Ian Smith, MS,
Luisa Betancourt, MD, William Weber, MSES, Brian Weiland,
BS, and Arielle Goldsmith, MS, for project planning, imple-
mentation, management, and analysis.
References
1. LaPlante MP, Kaye HS. Demographics and trends in wheeled mobility
equipment use and accessibility in the community. Assist Technol
2010;22:3-17.
2. Hosseini SM, Oyster ML, Kirby RL, Harrington AL, Boninger ML.
Manual wheelchair skills capacity predicts quality of life and com-
munity integration in persons with spinal cord injury. Arch Phys Med
Rehabil 2012;93:2237-43.
3. Lemay V, Routhier F, Noreau L, Phang S, Ginis KM. Relationships
between wheelchair skills, wheelchair mobility and level of injury in
individuals with spinal cord injury. Spinal Cord 2012;50:37-41.
4. Bullard S, Miller S. Comparison of teaching methods to learn a tilt and
balance wheelchair skill. Percept Mot Skills 2001;93:131-8.
5. Öztürk A, Ucsular FD. Effectiveness of a wheelchair skills training
programme for community-living users of manual wheelchairs
in Turkey: a randomized controlled trial. Clin Rehabil 2011;25:
416-24.
6. Kirby RL, Crawford KA, Smith C, Thompson KJ, Sargeant JM. A
wheelchair workshop for medical students improves knowledge and skills:
a randomized controlled trial. Am J Phys Med Rehabil 2011;90:197-206.
7. Routhier F, Kirby RL, Demers L, Depa M, Thompson K. Efficacy and
retention of the French-Canadian version of the Wheelchair Skills
Training Program for manual wheelchair users: a randomized
controlled trial. Arch Phys Med Rehabil 2012;93:940-8.
8. MacPhee AH, Kirby RL, Coolen AL, Smith C, MacLeod DA,
Dupuis DJ. Wheelchair skills training program: a randomized clinical
trial of wheelchair users undergoing initial rehabilitation. Arch Phys
Med Rehabil 2004;85:41-50.
9. Best KL, Kirby RL, Smith C, MacLeod DA. Wheelchair skills training
for community-based manual wheelchair users: a randomized
controlled trial. Arch Phys Med Rehabil 2005;86:2316-23.
10. Kirby RL, Mifflen NJ, Thibault DL, et al. The manual wheelchair-
handling skills of caregivers and the effect of training. Arch Phys
Med Rehabil 2004;85:2011-9.
11. Eastwood EA, Hagglund KJ, Ragnarsson KT, Gordon WA, Marino RJ.
Medical rehabilitation length of stay and outcomes for persons with
traumatic spinal cord injuryd1990e1997. Arch Phys Med Rehabil
1999;80:1457-63.
1784
12. Ottenbacher KJ, Smith PM, Illig SB, Linn RT, Ostir GV, Granger CV.
Trends in length of stay, living setting, functional outcome, and mortality
following medical rehabilitation. JAMA 2004;292:1687-95.
13. Teeter L, Gassaway J, Taylor S, et al. Relationship of physical therapy
inpatient rehabilitation interventions and patient characteristics to
outcomes following spinal cord injury: the SCIRehab project. J Spinal
Cord Med 2012;35:503-26.
14. Best KL, Routhier F, Miller WC. A description of manual wheelchair
skills training: current practices in Canadian rehabilitation centers.
Disabil Rehabil Assist Technol 2015;10:393-400.
15. Zanca JM, Dijkers MP, Hsieh CH, et al. Group therapy utilization in
inpatient spinal cord injury rehabilitation. Arch Phys Med Rehabil
2013;94(4 Suppl):S145-53.
16. Coolen AL, Kirby RL, Landry J, et al. Wheelchair skills training
program for clinicians: a randomized controlled trial with occupa-
tional therapy students. Arch Phys Med Rehabil 2004;85:1160-7.
L.A. Worobey et al
17. Giesbrecht EM, Wilson N, Schneider A, Bains D, Hall J, Miller WC.
Preliminary evidence to support a “boot camp” approach to wheel-
chair skills training for clinicians. Arch Phys Med Rehabil 2015;96:
1158-61.
18. Best KL, Miller WC, Huston G, Routhier F, Eng JJ. Pilot study of a
peer-led wheelchair training program to improve self-efficacy using a
manual wheelchair: a randomized controlled trial. Arch Phys Med
Rehabil 2016;97:37-44.
19. Kirby RL, Worobey LA, Cowan R, et al. Wheelchair skills capacity
and performance of manual wheelchair users with spinal cord injury.
Arch Phys Med Rehabil 2016;97:1761-9.
20. Dalhousie University, Faculty of Medicine, Wheelchair Skills Program.
Wheelchair Skills Training Program (WSTP). Available at: http://www.
wheelchairskillsprogram.ca/eng/trainers.php. Accessed October 8, 2015.
21. Conover WJ. Practical nonparametric statistics. 3rd ed. New York:
Wiley; 1999.
www.archives-pmr.org
Group wheelchair skills training 1784.e1

Supplemental Table S1 Clinician WSP training schedule

Before the Workshop
Check facilities and
wheelchairs, finalize set Faculty and
TBA up for obstacle course logistician
Day of Workshop
Time* Activity Participants
8:30 AM Registration Testers and trainers
8:45 AM Introductions, overview for
the day, spotter training
9:00 AM Practical training of
wheelchair skills
10:15 AM Break
10:30 AM Practical training of
wheelchair skills
11:45 AM Lunch
12:00 PM 1-h didactic presentation General audience,
testers, and
trainers
1:00 PM Break Testers and trainers
1:15 PM Practical training of
wheelchair skills
2:30 PM Break
2:45 PM Community activity
3:45 PM Viewing of videotapes,
practice scoring WSTs,
deciding on training
needed
4:15 PM Discussion
4:25 PM Evaluation of the workshop
4:30 PM Presentation of certificates
Adjournment
Abbreviations: TBA, to be announced; WSTs, Wheelchair Skills Tests.
* Times were adjusted for workshops beginning at noon one day and
finishing at noon the following day.

www.archives-pmr.org
1784.e2 L.A. Worobey et al

Supplemental Table S2 Sample WSTP session schedule

Session No.
Individual Skills 1 2 3 4 5 6 7 8
1. Rolls forward (10m) X X
2. Rolls backward (2m) X X X
3. Moving turns (forward, backward) X X X
4. Turns in place X X X
5. Maneuvers sideways X X X
6. Reaches high object X X X
7. Picks object from floor X X X
8. Relieves weight from buttocks X X
9. Transfers to bench and back X X X
10. Folds and unfolds wheelchair X X
11. Doors X X X
12. Rolls 100m X X X
13. Avoids moving obstacles X X X
14. Inclines (up, down, 5 , 10 ) X X X
15. Side-slope X X X
16. Soft surface X X X
17. Gap X X X
18. Threshold X X X
19. Curbs (up, down, 5cm/15cm) X X X X X
20. Stationary wheelie X X X
21. Wheelie skills (turn, forward, back) X X X X
22. Wheelie incline descent X X X
23. Wheelie curb descent X X X
24. Transfer from ground into wheelchair X X X
25. Descends stairs X X X

www.archives-pmr.org
Group wheelchair skills training 1784.e3

Supplemental Table S3 Participant demographic, clinical, and

wheelchair data among those missing 1-month follow-up
Active Control
WSTP Group Group
Subject Characteristics (nZ19) (nZ16)
Age (y) 36.713.1 3510
Height (m) 1.80.09 1.80.13
Weight (kg) 19043.3 168.140.2
Years of wheelchair use 10.310.5 10.810.2
Age of wheelchair 3.21.8 4.74.7
Paraplegia 16 (84.2) 13 (81.3)
>1y since injury 2 (1.5) 1 (6.3)
Male 19 (100) 11 (68.8)
Race
White 8 (42.1) 5 (31.3)
Black 9 (47.4) 7 (43.8)
Asian Indian 0 (0) 1 (0)
Not provided 2 (10.6) 3 (18.8)
Residence
Private 18 (94.7) 14 (87.5)
Group living 0 (0) 0 (0)
Nursing home 0 (0) 0 (0)
Hotel or motel 0 (0) 0 (0)
Homeless 1 (5.3) 2 (12.5)
Level of education
Less than high school 0 (0) 0 (0)
diploma
High school diploma 14 (73.7) 9 (56.3)
or GED
Associate’s or 3 (15.8) 6 (37.5)
Bachelor’s degree
Graduate degree 2 (10.5) 1 (6.3)
Other 0 (0) 0 (0)
NOTE. Values are mean  SD or n (%). For patterns in missing data:
between-group differences in those completing 1-month follow-up and
those who did not, we found no significant differences based on years
of use (WSTP: PZ.341; AC: PZ.563), level of injury (Fisher exact test,
WSTP: PZ.701; AC: PZ.738), sex (Fisher exact test, WSTP: PZ.286;
AC: PZ.149), or baseline WST-Q capacity score (WSTP: PZ.136).
Abbreviations: AC, active control; GED, General Educational Develop-
ment.

www.archives-pmr.org