Archives of Physical Medicine and Rehabilitation

journal homepage: www.archives-pmr.org

Archives of Physical Medicine and Rehabilitation 2015;96:1360-3

BRIEF REPORT

Prevalence of Low Mobility and Self-Management

Self-Efficacy in Manual Wheelchair Users and the
Association With Wheelchair Skills
Brodie M. Sakakibara, PhD,a,b William C. Miller, PhD, FCAOTb,c
From the aFaculty of Health Sciences, Simon Fraser University, Vancouver, BC; bRehabilitation Research Program, GF Strong Rehabilitation
Center, Vancouver Coastal Health Research Institute, Vancouver, BC; and cDepartment of Occupational Science and Occupational Therapy,
Faculty of Medicine, University of British Columbia, Vancouver, BC, Canada.

Abstract
Objective: To estimate the prevalence of low wheelchair-mobility and self-management self-efficacy and to evaluate the association with
wheelchair skills.
Design: Cross-sectional.
Setting: Community.
Participants: Community-dwelling manual wheelchair users (NZ123) who were
50 years of age (mean, 59.77.5y) and from British
Columbia and Quebec, Canada.
Interventions: None.
Main Outcome Measures: The 13-item mobility and 8-item self-management subscales from the Wheelchair Use Confidence ScaleeShort Form
(standardized scores range, 0e100) measured self-efficacy, and the 32-item Wheelchair Skills Test, Questionnaire Version (scores range, 0e100)
measured wheelchair skills. A score of 50 was used to differentiate individuals with high and low self-efficacy, and a score of 72 differentiated
between high and low wheelchair skills.
Results: The prevalence of low wheelchair-mobility and self-management self-efficacy was 28.5% (95% confidence interval [CI], 20.6e36.4)
and 11.4% (95% CI, 5.8e17.0), respectively, and their bivariate association with wheelchair skills was rZ.70 and rZ.39, respectively. Of the
sample, 16% reported conflicting mobility self-efficacy and skill scores; 25% reported low self-efficacy and high skills. Of the participants,
30% reported conflicting scores between self-management self-efficacy and wheelchair skills, with 8.1% reporting lower self-efficacy than
skill.
Conclusions: Low self-efficacy was relatively high in this sample as was its discordance with wheelchair skills. Interventions to address low self-
efficacy and/or offset the discordant self-efficacy/skill profiles are warranted.
Archives of Physical Medicine and Rehabilitation 2015;96:1360-3
ª 2015 by the American Congress of Rehabilitation Medicine

Self-efficacy is the belief individuals have in their ability to overall health.1 Furthermore, evidence shows that various forms of
perform specific behaviors to achieve desired outcomes.1 Ac- the construct have the potential to be modified.1,2 Self-efficacy
cording to social cognitive theory it is a central construct for specific to wheelchair use is a new construct defined as the
behavior change because it has both direct and indirect influences belief individuals have in their ability to use their wheelchair in a
on what people do.1 In general, higher self-efficacy specific to variety of challenging situations.3 Given evidence on the benefits
health-related behaviors has positive effects on what people do, is of high self-efficacy specific to other areas of health, wheelchair-
associated with lower health risks, and is associated with better use self-efficacy is currently receiving research attention and
demonstrating positive results related to social participation and
wheelchair mobility.4,5
Supported by the Canadian Institutes of Health Research with a postdoctoral fellowship and
grant (grant no. CIHR IAP-107848).
Our previous research indicates that 39% of wheelchair users
Disclosures: none. have low wheelchair-use self-efficacy6 measured using the

0003-9993/15/$36 - see front matter ª 2015 by the American Congress of Rehabilitation Medicine
http://dx.doi.org/10.1016/j.apmr.2015.03.002
Prevalence of low wheelchair-use self-efficacy 1361

Wheelchair Use Confidence Scale (WheelCon).3 Moreover, 27% Wheelchair skills were captured using the Wheelchair Skills
report having disproportionate levels of self-efficacy and wheel- Test, Questionnaire Version, a self-report measure comprised of
chair skills, which may lead to sedentary lifestyles if people have 9 advanced skills and 23 basic indoor and community skills.8
less self-efficacy than ability or unsafe performance of activities if Individuals reported their ability (yes/no) to complete each
self-efficacy exceeds ability. Although these estimates contribute skill. Total percentage scores were derived by dividing the
to an appreciation of the potential impact of health care and number of skills individuals can do by the total number of
rehabilitation strategies to prevent and minimize the consequences applicable skills. Higher scores indicate more wheelchair skill.
of low self-efficacy in wheelchair users, the estimates are based on Scores from this measure are highly correlated with measure-
a composite score from a multidimensional measure. There re- ments from the performance-based Wheelchair Skills Test,
mains a lack of evidence on the prevalence of the various and Version 4.1 (Spearman rZ.89).8 A score of 72 (ie, 23 basic
more specific forms of wheelchair-use self-efficacy (ie, mobility skills, 32 total skills) differentiated between high and low
and self-management self-efficacy found within the WheelCon wheelchair skills in this study.
measure) that have recently been established using principal Ability to perform activities of daily living and depression and
components analysis, along with item response theory,7 and their anxiety were measured using the Barthel Index9 and Hospital
association with wheelchair skills. Such specific knowledge will Anxiety and Depression Scale,10 respectively, and were used as
further aid researchers and clinicians to develop and plan appro- sample descriptors. Evidence supports the hypothesized magni-
priate rehabilitation services. tude and direction of the associations between the Barthel Index
The objectives of this study are to estimate the prevalence of and Hospital Anxiety and Depression Scale with relevant variables
low wheelchair-mobility and self-management self-efficacy. We in wheelchair users.3
also estimate the association and amount of discordance between
the self-efficacy constructs and wheelchair skills. Data analyses
Descriptive statistics are presented as frequencies and percentages
Methods and means  SDs.
Prevalence of low wheelchair-mobility and self-management
self-efficacy are estimated using proportions and 95% confidence
Study design and participants
intervals (CIs).
This is a secondary analysis of cross-sectional data from The bivariate association between the self-efficacy constructs
community-dwelling volunteers from British Columbia and and skill was estimated using Pearson correlation coefficient, and
Quebec, Canada, who were aged
50 years, had at least 6 the discordance between the variables was evaluated using cross-
months of experience with manual wheelchair use on a daily tab analyses and reported as proportions and 95% CIs.
basis, and were able to communicate in either English or
French.4 Individuals with a Mini-Mental State Examination Results
score <23 and/or those who were not medically stable were
The mean age of this sample (NZ123) of experienced wheelchair
excluded from study. Rehabilitation therapists from various
users was 59.77.5 years; 74 participants (60.2%) were men. Of
health authorities in British Columbia, seating clinics in Quebec,
the participants, 59 (48%) had a spinal cord injury. Sample
and community groups provided study information to recruit
characteristics are detailed in table 1.
potential participants.
The prevalence of low wheelchair-mobility and self-
management self-efficacy (figs 1A,B, quadrants 3 and 4) was
Study protocol
28.5% (95% CI, 20.6e36.4) and 11.4% (95% CI, 5.8e17.0),
After participants provided consent, they met with a trained respectively.
research assistant who gathered demographic information and The bivariate association between the wheelchair-mobility
explained and administered the self-efficacy and wheelchair skill self-efficacy and wheelchair skills was rZ.70, and the associa-
measures. The ethics boards from all participating sites approved tion between self-management self-efficacy and wheelchair skills
the study protocol. was rZ.39.
Conflicting wheelchair-mobility self-efficacy and wheelchair
Measures skill scores were reported by 16.3% of the sample. Of these in-
dividuals, 25% (ie, 4% of the entire sample) reported low self-
The demographic information questionnaire gathered data on sex, efficacy and high skill (see fig 1A, quadrant 3). In addition, 24.4%
age, marital status and health status, and wheelchair- of the entire sample reported having both low wheelchair-mobility
related variables. self-efficacy and wheelchair skills (see fig 1A, quadrant 4). Thirty
The self-efficacy constructs were estimated using the 21-item percent reported discordant self-management self-efficacy and
Rasch-derived WheelConeShort Form,7 which is comprised of wheelchair skills. Of these individuals, 8.1% (ie, 2.4% of the
mobility (13 items) and self-management (8 items) subscales. entire sample) reported lower levels of self-efficacy than skill (see
Standardized scores from each subscale range from 0 to 100, with fig 1B, quadrant 3). Only 9% reported having both low self-
higher scores indicating higher self-efficacy. A standardized score management self-efficacy and wheelchair skills (see fig 1B,
of 50 was used to differentiate between high and low self-efficacy. quadrant 4).

List of abbreviations: Discussion

CI confidence interval
Approximately 25% of wheelchair users reported both low
WheelCon Wheelchair Use Confidence Scale
wheelchair-mobility self-efficacy and wheelchair skills and

www.archives-pmr.org
1362 B.M. Sakakibara, W.C. Miller

self-efficacy lead people to do things regardless of ability,1 in-

Table 1 Descriptive statistics (NZ123)
dividuals with high mobility self-efficacy who are less able to use
Variable Value their wheelchair may attempt risky maneuvers, potentially leading
Age 59.77.5 to injury.
Sex (male) 74 (60.2) The anticipated increase of wheelchair users aged
50 years
Education (high school graduate) 85 (69.1) because of the population aging combined with the results of this
Married (yes) 58 (47.2) study are reasons to believe that there will be greater numbers of
Employed/volunteer (yes) 46 (37.4) wheelchair users with low mobility and self-management self-
Comorbidities 2.72.4 efficacy. Furthermore, given theory that self-efficacy in older in-
Activities of daily living (0e20)* 14.42.8 dividuals tends to diminish with age,1 it is plausible that the
Depression symptoms (0e21)y 3.83.1 prevalence rates of both mobility and self-management self-effi-
Anxiety symptoms (0e21)y 5.13.9 cacy will also increase. Although wheelchair mobility requires
Diagnosis physical ability and self-management requires greater cognitive
Spinal cord injury 59 (48.0) abilities, our findings indicate greater issues concerning beliefs
Multiple sclerosis 16 (13.0) related to physically using the wheelchair than those related to
Stroke 12 (9.7) problem-solving, decision-making, and action planning. There-
Otherz 36 (29.3) fore, finding that the prevalence of low mobility self-efficacy was
Wheelchair factors relatively high and that discordance exists with wheelchair skill
Formal training (ever) (yes) 21 (17.1) suggest research on efficacy-enhancing interventions focusing on
Wheelchair assistance (as of now) (yes) 38 (30.9) wheelchair mobility is warranted. This may be especially true for
Wheelchair experience (years) 22.516.0 those wheelchair users who have greater ability than mobility self-
Daily use (h) 12.34.3 efficacy (ie, 25% of those reporting discordance) because high
Wheelchair-mobility self-efficacy (0e100) 61.018.7 skill alone is likely insufficient to promote participation
Self-management self-efficacy (0e100) 69.820.0 and mobility.
Wheelchair skills (0e100) 75.515.0
NOTE. Values are mean  SD or frequency (%).
* Barthel Index.
y
Hospital Anxiety and Depression Scale.
z
Parkinson disease, cerebral palsy, brain injury, polio, arthritis, and
amputation.

therefore are at risk for low mobility and low participation fre-
quency.4,5 Although these individuals may benefit from efficacy
and/or skill intervention, it is important for clinicians to be
cautious when prescribing treatment because those individuals
who may appear to be in greatest need of efficacy enhancements
may first require development of their wheelchair skills, and vice
versa. Complete understanding of reasons for low self-efficacy is
necessary for treatment planning and may warrant a multidisci-
plinary approach that includes important physical and cognitive
considerations using the 4 sources of information (performance
accomplishment, vicarious learning, verbal persuasion, and
interpretation of physiological and affective states1) theorized to
modify self-efficacy.
The shared variance between self-management self-efficacy
and wheelchair skills might be low (15.2%) because items in the
self-management self-efficacy measure largely focus on decision-
making and problem-solving, and not on ability to use a wheel-
chair. This does, however, speak to issues relevant to the validity
of the measures with respect to discrimination and convergence.
Although the shared variance between mobility self-efficacy and
wheelchair skills was higher (49%), 16% of the sample reported
discordant mobility self-efficacy and wheelchair skills, with most
reporting a high self-efficacy and low skill profile. This finding has
both positive and potentially negative implications. Recent evi- Fig 1 Prevalence of low wheelchair mobility SE (A) and low self-
dence hypothesizes a causal path, in which skills mediates the management SE (B), and their discordance with wheelchair skills.
self-efficacy and life-space mobility association.5 This suggests Scores of 50 and 72 were used to differentiate between low and high
that the high self-efficacy reported by those individuals in our SE and wheelchair skills, respectively. Abbreviations: SE, self-efficacy;
study might act to improve their wheelchair skills, which may 1, high self-efficacy/low skill; 2, high self-efficacy/high skill; 3, low
lead to better mobility. Alternatively, because high levels of self-efficacy/high skill; 4, low self-efficacy/low skill.

www.archives-pmr.org
Prevalence of low wheelchair-use self-efficacy 1363

Study limitations References

This study has several limitations, including the small sample size
1. Bandura A. Self-efficacy: the exercise of control. New York: Freeman;
used to generate the prevalence estimates. In addition, generaliz-
1997.
ability of the findings is limited to community-living, wheelchair 2. Sakakibara BM, Miller WC, Souza M, Nikolova V, Best KL.
users with similar characteristics to that of the sample in Wheelchair skills training to improve confidence with using a manual
this study. wheelchair among older adults: a pilot study. Arch Phys Med Rehabil
2013;94:1031-7.
3. Rushton PW, Miller WC, Kirby RL, Eng JJ. Measure for the assess-
Conclusions ment of confidence with manual wheelchair use (WheelCon-M)
version 2.1: reliability and validity. J Rehabil Med 2013;45:61-7.
Prevalence of low self-efficacy is high and in 16.3% to 30% of 4. Sakakibara BM, Miller WC, Routhier F, Backman CL, Eng JJ. The
cases does not match well with level of skill. Both self-efficacy association between self-efficacy and participation in community-
and skills have important implications on rehabilitation outcomes; dwelling manual wheelchair-users, aged 50 and over. Phys Ther
therefore, research on interventions to improve low mobility self- 2014;94:664-74.
efficacy and/or offset the discordant mobility self-efficacy skill 5. Sakakibara BM, Miller WC, Eng JJ, Backman CL, Routhier F. In-
profile is warranted. fluences of wheelchair-related efficacy on life-space mobility in adult,
community-dwelling manual wheelchair-users. Phys Ther 2014;94:
1604-13.
6. Miller WC, Sakakibara BM, Rushton PW. The prevalence of low
Keywords confidence with using a wheelchair and its relationship to wheelchair
skills. Gerontologist 2012;52(Suppl 1).
Prevalence; Rehabilitation; Self-efficacy; Wheelchairs 7. Sakakibara BM, Miller WC, Rushton PW. Rasch analyses of the
Wheelchair Use Confidence Scale. Arch Phys Med Rehabil 2015;96:
1036-44.
8. Rushton PW, Kirby RL, Miller WC. Manual wheelchair skills:
Corresponding author objective testing versus subjective questionnaire. Arch Phys Med
Rehabil 2012;93:2313-8.
William C. Miller, PhD, FCAOT, Department of Occupational 9. Gompertz P, Pound P, Ebrahim S. A postal version of the Barthel
Science and Occupational Therapy, University of British Index. Clin Rehabil 1994;8:233.
Columbia, T325-2211 Wesbrook Mall, Vancouver, BC V6T 2B5, 10. Zigmond AS, Snaith RP. The Hospital Anxiety and Depression scale.
Canada. E-mail address: bill.miller@ubc.ca. Acta Psychiatr Scand 1983;67:361-70.

www.archives-pmr.org