- Email: [email protected]
Wheelchair Breakdowns Are Associated With Pain, Pressure Injuries, Rehospitalization, and Self-Perceived Health in Full-Time Wheelchair Users With Spinal Cord Injury
Accepted Manuscript Wheelchair breakdowns are associated with pain, pressure injuries, rehospitalization, and self-perceived health in full-time wheelchair users with spinal cord injury Nathan S. Hogaboom, PhD, Lynn A. Worobey, PhD, DPT, ATP, Bethlyn V. Houlihan, MSW, MPH, Allen W. Heinemann, PhD, Michael L. Boninger, MD PII:
S0003-9993(18)30233-8
DOI:
10.1016/j.apmr.2018.04.002
Reference:
YAPMR 57205
To appear in:
ARCHIVES OF PHYSICAL MEDICINE AND REHABILITATION
Received Date: 19 January 2018 Revised Date:
28 February 2018
Accepted Date: 7 April 2018
Please cite this article as: Hogaboom NS, Worobey LA, Houlihan BV, Heinemann AW, Boninger ML, Wheelchair breakdowns are associated with pain, pressure injuries, rehospitalization, and self-perceived health in full-time wheelchair users with spinal cord injury, ARCHIVES OF PHYSICAL MEDICINE AND REHABILITATION (2018), doi: 10.1016/j.apmr.2018.04.002. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Title: Wheelchair breakdowns are associated with pain, pressure injuries, rehospitalization, and self-
Running Head: Wheelchair breakdowns and health outcomes
SC
Authors:
Lynn A Worobey, PhD, DPT, ATP Bethlyn V. Houlihan, MSW, MPH
TE D
Allen W. Heinemann, PhD
M AN U
Nathan S Hogaboom, PhD
Michael L Boninger, MD
RI PT
perceived health in full-time wheelchair users with spinal cord injury
EP
From the Human Engineering Research Laboratories, VA Pittsburgh Healthcare System, Pittsburgh, PA (Hogaboom, Worobey, Boninger); Department of Rehabilitation Science and
AC C
Technology, School of Health and Rehabilitation Sciences, University of Pittsburgh, Pittsburgh, PA (Hogaboom, Boninger); Department of Physical Medicine and Rehabilitation, School of Medicine, University of Pittsburgh, Pittsburgh, PA (Worobey, Boninger); Spaulding New England Regional SCI Center, Boston, MA (Houlihan); School of Public Health, Boston University, Boston, MA (Houlihan); Center for Rehabilitation Outcomes Research,
ACCEPTED MANUSCRIPT
Rehabilitation Institute of Chicago, Chicago, Illinois (Heinemann); Department of Physical Medicine and Rehabilitation, Northwestern University, Chicago, Illinois (Heinemann).
RI PT
This material is the result of work supported with resources and the use of facilities at the Human Engineering Research Laboratories, VA Pittsburgh Healthcare System. The contents of this paper do not represent the views of the Department of Veterans Affairs or the United States
SC
Government. This project was supported by the National Institute on Disability, Independent Living, and Rehabilitation Research (NIDILRR grant #90SI5008). NIDILRR is a Center within
M AN U
the Administration for Community Living (ACL), Department of Health and Human Services (HHS). The contents of this (insert type of publication; e.g., book, report, film) do not necessarily represent the policy of NIDILRR, ACL, HHS, and you should not assume endorsement by the Federal Government. This material is based upon work supported by the
TE D
National Science Foundation Graduate Research Fellowship (#1247842) and IGERT Fellowship (DGE 1144584). The contents of this publication do not necessarily reflect the view of the
EP
National Science Foundation.
Data were presented at the 2016 American Congress of Rehabilitation Meeting (November 2-4,
AC C
Chicago, Illinois, USA) and the 2017 International Seating Symposium (March 2-4, Nashville, TN, USA).
Address all correspondence to: Lynn Worobey 6425 Penn Avenue, Suite 400
ACCEPTED MANUSCRIPT
Pittsburgh, PA 15206 Phone: 412-822-3700 Fax: 412-822-3699
AC C
EP
TE D
M AN U
SC
RI PT
Email: [email protected]
ACCEPTED MANUSCRIPT
ABSTRACT
2
Objectives: To evaluate the relationship between wheelchair breakdowns, their immediate
3
consequences, and secondary health complications after spinal cord injury (SCI). “Immediate
4
consequences” occur when part of a wheelchair breaks and leaves an individual stranded or
5
injured, or causes them to miss medical appointments, work, or school.
6
Design: Survey, cross-sectional.
7
Setting: Spinal Cord Injury Model Systems Centers (SCIMS).
8
Participants: 771 full-time wheelchair users with SCI from 9 SCIMS centers, with data collected
9
between 2011 and 2016.
M AN U
SC
RI PT
1
Interventions: N/A
11
Main Outcomes: Incidence of self-reported wheelchair breakdowns within the past 6 months that
12
did or did not result in immediate consequences (i.e. injury, being stranded, missing a medical
13
appointment, or an inability to attend school/work); Self-Perceived Health Status (SPHS) scale;
14
pain severity numerical rating scale; rehospitalizations and self-reported pressure injury
15
development within the past 12 months.
16
Results: 610 participants with complete datasets were included in the analyses. When compared
17
to those who reported no breakdowns, participants who reported one or more immediate
18
consequences had worse secondary complications: higher SPHS and pain scores (partial-
19
η2=.009-.012, p<.05), and higher odds of rehospitalization (OR: 1.86, p<.05) and pressure injury
20
development (OR: 1.73, p<.05). Secondary complications were not different in those who
21
reported no immediate consequences compared to those who reported no breakdown.
22
Conclusions: Wheelchair breakdowns that resulted in injury, being stranded, missing medical
23
appointments, and/or an inability to attend work/school appear to have far-reaching impacts on
AC C
EP
TE D
10
ACCEPTED MANUSCRIPT
health and secondary injury. Preventing wheelchair breakdowns, either through better
25
maintenance or manufacturing, may be a means of decreasing secondary disability.
AC C
EP
TE D
M AN U
SC
RI PT
24
ACCEPTED MANUSCRIPT
26
INTRODUCTION
27
Sustaining a spinal cord injury (SCI) has a profound impact on an individual’s life. Independent
29
mobility and participation may be limited by environmental barriers. For those who do not regain
30
the ability to ambulate, many use a wheelchair to achieve full participation. In addition to
31
enhancing mobility, many wheelchairs are equipped with technology that helps to manage pain
32
and pressure; for example, tilt and recline functions in power wheelchairs or adjustable seating in
33
manual chairs.1-3 Despite the wheelchair’s importance, our research group reported that greater
34
than half of wheelchair users with SCI experience a wheelchair breakdown. Between 20% and
35
30% of those individuals experienced an immediate consequence that left them stranded or
36
injured, or prevented them from attending medical appointments, school, or work.4-7 The repair
37
process can be difficult to navigate for wheelchair users8; as a result, nearly 7% of individuals do
38
not receive the necessary repairs for their failed wheelchair components.7
SC
M AN U
TE D
39
RI PT
28
A serious breakdown could also relegate an individual to an inappropriate seating surface, such
41
as a couch or bed, which does not provide adequate postural support or pressure management.
42
By removing a primary means of self-care management, wheelchair breakdowns put the user at
43
risk of exacerbating secondary conditions. These potential secondary effects include pressure
44
injuries, pain, and recurrent hospitalizations, all of which are correlated with lower self-
45
perceived health and quality of life.9-17 Previous investigations using SCI Model Systems data
46
have reported several subject characteristics associated with worse health outcomes: minority
47
race, tetraplegia, public insurance payer, unemployment, less education, and being single. 18-20
48
However, no studies have incorporated wheelchair-related factors in the analyses. By omitting
AC C
EP
40
ACCEPTED MANUSCRIPT
49
the wheelchair in their analyses, investigators may be missing a potentially critical factor that can
50
reduce secondary complication risks.
51
This study is an extension of our previous work, which demonstrated that the incidence of
53
wheelchair breakdowns continues to rise.4, 6, 7 These breakdowns can play an important role in
54
user safety, positioning, and mobility. This study uses the same dataset to investigate whether
55
wheelchair breakdowns are related to secondary health outcomes. The rationale behind this study
56
is to explore and provide evidence of the potential secondary effects of wheelchair breakdowns
57
in the form of worsening secondary health complications. If the following hypotheses are met,
58
this analysis would show the importance of breakdown prevention and the further investigation
59
of preventative methods. Participants who reported one or more wheelchair breakdowns, with
60
and without immediate consequences, were hypothesized to report worse secondary health
61
effects when controlling for other risk factors (older age, minority race, female sex, non-private
62
insurance coverage, single, unemployment, less education, and cervical injury level). Health-
63
related outcome indicators were pain, self-perceived health, rehospitalizations, and self-reported
64
incidence of pressure injuries.
65
EP
TE D
M AN U
SC
RI PT
52
METHODS
67
Subjects
68
Participants were eligible if they were older than 16 years, had a neurologic impairment resulting
69
from a SCI that occurred at least 1 year prior to the study, and used a manual- or power-
70
wheelchair (including power-assisted manual chairs) for at least 40 hours per week. Participants
71
were enrolled at one nine SCI Model Systems Centers March, 2013 and January, 2016.
AC C
66
ACCEPTED MANUSCRIPT
Recruitment methods included approaching individuals who participated in the National SCI
73
Database, local registries, flyers, and identification by clinical staff. Participants completed
74
surveys during interviews, over the phone, or via mail. All centers obtained ethical approval
75
from their local institutional review boards and all participants provided informed consent.
76
Further details of data collection and the specific Model Systems Centers can be found in Toro,
77
et al.7
RI PT
72
79
SC
78
Data Collection
M AN U
80
This is a secondary analysis of the 2016 study conducted by Toro, et al.; additional information
82
about data collection procedures can be found in their study.7 Participants completed a
83
questionnaire that inquired about demographic characteristics and secondary health conditions.
84
Demographic characteristics included age, sex, race, employment, marital status, injury level,
85
and insurance coverage. Participants were asked how many repairs were needed to their
86
wheelchairs within the past 6 months, and whether any immediate consequences occurred as a
87
result of those breakdowns; responses included no consequence, missing a medical appointment,
88
missing work or school, being stranded, or being injured.4, 7 The breakdown questionnaire can be
89
found in the appendix of the 2009 study by McClure, et al.4
EP
AC C
90
TE D
81
91
Regarding secondary health effects, participants were asked about recent pain, self-perceived
92
health status (SPHS), pressure injury development, and rehospitalizations. Pain over the past
93
month was assessed using a 0 (no pain) to 10 (worst possible pain) numerical rating scale – an
94
assessment recommended for use in the SCI population.21 SPHS was determined from item 1 of
ACCEPTED MANUSCRIPT
the Medical Outcomes Study SF-36 scale, which asks individuals to rate how they view their
96
health using a scale of 1 (excellent) to 5 (poor).22 Although validity and reliability have not been
97
established for this item, both full and partial scales have been used extensively in medical
98
research including the SCI population.23 Participants provided information about the number of
99
days hospitalized over the past 12 months, and indicated one of 18 reasons for those
100
rehospitalizations based on a previous study.18 Of interest were those who reported a
101
rehospitalization from diseases of the skin/subcutaneous tissue, including pressure injuries. They
102
were asked whether they experienced the “presence of a pressure sore with open or broken skin
103
in the past 12 months.” Response options were “yes,” “no,” and “I do not know.”
104
M AN U
SC
RI PT
95
Data Reduction and Analysis
106
All statistical analyses were performed using SPSS 23.0 (IBM, Inc., Armonk, New York, USA).
107
Significance was set a priori to α=.05. Descriptive statistics were calculated first, including
108
means and standard deviations for continuous variables and frequencies for nominal variables.
109
Demographic characteristics were categorized to optimize cell size, in accordance with previous
110
studies investigating wheelchair consequences or rehospitalizations4, 6, 18, 24: injury level into
111
tetraplegia vs. paraplegia; race into minority vs. white; education into high school diploma or
112
less vs. post-high school education; marital status into married/living with partner, never
113
married, or divorced/separated/widowed/other; employment status into employed, unemployed,
114
or homemaker/on-the-job-training/retired/student/other; and primary insurance payer into
115
private, Medicare, Medicaid, or other. Participants’ responses for breakdowns and immediate
116
consequences were categorized accordingly into one of three groups: 1) no wheelchair
117
breakdowns, 2) at least one breakdown without immediate consequences, or 3) at least one
AC C
EP
TE D
105
ACCEPTED MANUSCRIPT
breakdown with at least one immediate consequence (resulting in being stranded, injury, missed
119
medical appointment, or missed work/school). Rehospitalizations were recoded into no
120
rehospitalizations, rehospitalized at least once for reasons unrelated to skin disease, or
121
rehospitalized at least once specifically for skin disease.
122
RI PT
118
Participants with missing demographic characteristics, wheelchair breakdowns,
124
rehospitalizations, pressure injury development, self-perceived health, or pain data were
125
excluded. In this way, the same cohort of individuals would be included in each statistical model
126
and thus is sample is more consistent. Potential bias from this additional exclusion was
127
investigated using Pearson’s chi-square tests for nominal variables and independent t-tests for
128
continuous variables (excluded versus included). A follow-up analysis using multiple imputation
129
determined that removing the subjects with missing data did not influence the results. Thus, only
130
the cohort with complete datasets for all independent and dependent variables were included.
131
Furthermore, an insignificant Little’s test suggests data were missing completely at random
132
(Χ2=13.92, df=7, p>.05) and that analysis of the complete cases was valid.25
M AN U
TE D
EP
133
SC
123
Three regression models were built to test the association between wheelchair breakdowns with
135
and without immediate consequences and health-related quality of life variables. Using the
136
general linear model procedure, a linear-regression model with multiple outcomes was built to
137
test whether breakdowns with and without immediate consequences influenced SPHS and pain
138
scores. Two logistic regression models tested whether breakdowns with and without immediate
139
consequences increased odds of self-reported pressure injury development or rehospitalizations
140
related to, and unrelated to skin breakdown. Breakdown groups and demographic characteristics
AC C
134
ACCEPTED MANUSCRIPT
141
were used as predictors: age, sex, race, injury level, education, marital status, employment status,
142
and primary insurance payer. Group assignments and reference categories are located in Tables 1
143
and 2.
145
RI PT
144
RESULTS
146
Subjects
148
The data set consisted of 771 full-time wheelchair users with SCI; 161 (20.9%) had missing data
149
and were excluded. Demographic and wheelchair breakdown data are presented in Table 1.
150
Participants with missing data were more often unemployed and less often in the “other”
151
employment group; more often reported rehospitalizations unrelated to pressure injuries; and less
152
often reported no rehospitalizations (p<.01, Tables 1 and 2). No other differences were found
153
with respect to demographic characteristics, wheelchair breakdowns, or secondary condition
154
outcomes. Missing data sample sizes varied because many subjects with missing data for one
155
variable were not missing data from other variables.
TE D
M AN U
SC
147
EP
156
Of those included in the analysis, 233 (38.2%) used a power chair, 361 (59.2%) were manual
158
wheelchair users, and 16 (2.6%) were power-assisted manual chair users; the latter two groups
159
were combined. The most common immediate consequence from a breakdown was being
160
stranded (12.1%), followed by missing a medical appointment (6.7%), missing work/school
161
(3.6%), and being injured (2.9%). 363 (42.3%) reported having at least one working backup
162
chair.
163
AC C
157
ACCEPTED MANUSCRIPT
Pain and Perceived Health
165
The multivariate linear regression model found that breakdowns without immediate
166
consequences were not associated with pain or perceived-health (Tables 3 and 4). Immediate
167
consequences were found to have small, negative effects on SPHS and pain scores (p<.05,
168
R2=.009-.012; Tables 3 and 4). That is, participants who were stranded, were injured, missed
169
medical appointments, or missed work/school because of a wheelchair breakdown reported
170
worse pain and self-perceived health (Figure 1). Less education and insurance coverage other
171
than private were also associated with more pain (p<.05; Table 3). Unemployment, older age,
172
and Medicaid insurance coverage were associated with worse self-perceived health (p<.05; Table
173
4).
174
M AN U
SC
RI PT
164
Pressure Injuries and Hospitalizations
176
Results of the logistic regression models found no relationship between breakdowns without
177
immediate consequences and rehospitalization or pressure injury development (Tables 5 and 6).
178
Those who experienced immediate consequences after a breakdown had higher odds of
179
rehospitalization both related to (OR: 2.17, p<.10) and unrelated to skin disease (OR: 1.81,
180
p<.05). These individuals also had higher odds of developing a pressure injury within the past 12
181
months (OR: 1.72, p<.05). Individuals with tetraplegia (OR: 1.66, p<.05), in the “other”
182
employment group (OR: 2.37, p<.01), and with Medicaid coverage (OR: 1.66, p<.05) had higher
183
odds of rehospitalization unrelated to skin disease. No other demographic characteristic was
184
related to self-reported rehospitalizations or pressure injury development.
185
AC C
EP
TE D
175
ACCEPTED MANUSCRIPT
186
DISCUSSION
187
The present study advances the available knowledge about wheelchair breakdowns and their
189
secondary effects on health outcomes. This was accomplished by establishing a connection
190
between notable health-related outcomes and breakdowns that resulted in immediate
191
consequences of being stranded, being injured, missing medical appointments, or missing
192
work/school; specifically, pain, pressure injuries, rehospitalizations, and perceived health.
193
Although the sample consisted of only people with SCI, wheelchair failures are not unique to
194
SCI and thus these findings can be applied to active users with different diagnoses who rely upon
195
their wheelchairs for mobility.
196
M AN U
SC
RI PT
188
Users that reported needing a repair on their wheelchair, but not an immediate consequence, did
198
not appear to have worse secondary health outcomes than those who reported not needing any
199
repairs. However, breakdowns that were serious enough to result in being stranded or injured,
200
missing medical appointments, or an inability to attend school or work did have secondary
201
effects (Figure 2). This result was found despite the lack of differences observed in available
202
backup chairs. The findings could mean the immediate consequences of the breakdown, as
203
opposed to the breakdown itself, increase the risk of developing or worsening secondary
204
conditions; for example, missing medical appointments or being injured directly led to poorer
205
health outcomes. It is also possible that the breakdown was serious enough to prevent the user
206
from using their device by disabling seating functions and postural supports, which allow the
207
user to manage pressure and trunk stability. However, it is important to note that, with respect to
208
the latter, the “immediate consequence” classification was considered a surrogate for device
AC C
EP
TE D
197
ACCEPTED MANUSCRIPT
usability. The assumption was made that if an immediate consequence was reported, the
210
breakdown was serious enough to render the device unusable which ultimately led to worse
211
health outcomes. Because of this assumption, the authors cannot definitively conclude that
212
breakdowns led to secondary health consequences. This question does warrant further
213
investigation, however, because immediate consequences from breakdowns can be avoided
214
through early intervention. A more focused and detailed questionnaire about device usability
215
because of the breakdown would provide a more thorough analysis.
SC
RI PT
209
216
These results support previous reports, which indicated between 44% and 64% of wheelchair
218
users experience at least one breakdown in the past 6 months, with between 19.7% and 30.5% of
219
them reporting an immediate consequence.4, 6, 7 In contrast to previous works utilizing SCI
220
Model Systems datasets18, 20, 26, no relationships were observed between demographic
221
characteristics and pressure injury development or rehospitalizations. This finding may reflect
222
our use of self-report data in place of physician documentation20, 26 and inclusion only of
223
wheelchair users.18, 20, 26 It is also possible that wheelchair-related factors mitigate pressure injury
224
development and rehospitalizations more than personal characteristics. Considering these
225
findings, inclusion of wheelchair-related factors in future longitudinal investigations may
226
provide a more complete depiction of the risks of developing secondary conditions.
TE D
EP
AC C
227
M AN U
217
228
Implications for Clinical Practice
229
In many cases, failures result from factors outside of the user’s control, making it difficult to
230
identify supportive interventions. Unfortunately, data concerning the relationship between
231
wheelchair usage and failures is limited. Toro, et al. found that the number of failures was
ACCEPTED MANUSCRIPT
unrelated to the number of hours spent outside of the house, hours out of bed, or nights spent out
233
of the house.7 Toro, et al.’s results, and those of other investigators, suggest that wheelchair
234
type4, 6, 27 (i.e., manual versus power) and manufacturer28, 29 have more influence over failure
235
rates than wheelchair usage. In turn, it would not make sense for clinicians to discourage
236
individuals with SCI from using their wheelchairs as a means to prevent a breakdown and
237
subsequent secondary health issues. Rather, training wheelchair users to maintain their chairs and
238
to recognize a potential breakdown may be a cost-effective way to avoid the breakdowns that
239
lead to negative health outcomes.5, 30
M AN U
240
SC
RI PT
232
There is some evidence that suggests wheelchair seating clinicians are often unfamiliar with how
242
to identify and rectify broken wheelchair components.31 Providing clinicians with the knowledge
243
to help their patients navigate the repair process may lead to more timely repairs, which could
244
reduce the likelihood of an immediate consequence from occurring.32 A wheelchair maintenance
245
training program for clinicians and wheelchair users was recently developed using World Health
246
Organization basic wheelchair service curriculum. Initial testing showed that it improved
247
knowledge on basic maintenance.30 Further development of these training materials, and their
248
dissemination to wheelchair users and clinicians, may be a potential solution to preventing
249
secondary complications caused by wheelchair breakdowns and should be considered for future
250
investigation.
EP
AC C
251
TE D
241
252
Limitations and Subsequent Areas for Future Research
253
This study was exploratory and cross-sectional. As a result, we are unable to describe causal
254
relationships. Future studies could examine these relationships over time. The effect sizes
ACCEPTED MANUSCRIPT
between breakdowns and health outcomes were small, yet similar in magnitude to other subject
256
characteristics in the models, such as insurance type or age; thus, their importance should not be
257
minimized. Future research could ascertain the impact of multiple insurance types on
258
intermediate and distal outcomes.
RI PT
255
259
Participants were asked for a binary report of a limited number of breakdown-related immediate
261
consequences (e.g., being stranded or injured, missing medical appointments or work/school),
262
which may result in under-reporting the number of catastrophic failures. This line of questioning
263
also did not provide any information about whether the wheelchair could be used after the
264
breakdown; the length of time the wheelchair could not be used; or the number of instances the
265
user experienced an immediate consequence after a breakdown. The lack of information made it
266
difficult to determine how and why the breakdown led to the poorer health outcomes that were
267
observed. Asking open-ended questions may provide more detailed information for further study.
268
As well, many individuals reported experiencing more than one of the aforementioned
269
consequences. These individuals were grouped together to maximize cell size and strengthen
270
statistical power; however, this prevented further probing into the effects of specific breakdown-
271
related consequences on health outcomes. A larger sample size would lead to a more thorough
272
analysis.
M AN U
TE D
EP
AC C
273
SC
260
274
The timeframes of the different outcome variables (e.g., breakdowns within the past 6 months,
275
pain within the past week, rehospitalizations within the past 12 months) were not identical. These
276
differences may have affected the results. Health-promoting or self-destructive behaviors were
ACCEPTED MANUSCRIPT
not analyzed; these factors are associated with health status, pain, and pressure injury
278
development33 and may affect how proactive individuals are in maintaining their chairs.
279
Participants were treated at SCI Model Centers and thus results may not be generalizable to the
280
SCI population. Considering the level of care provided at these centers, the outcomes
281
participants achieve may be better than in less-resourced facilities. Finally, participants excluded
282
from the analysis due to missing data were more likely to be unemployed and not rehospitalized
283
than those included in the analyses. It is unclear if the same results would have been reported if
284
fewer cases had missing data.
286
M AN U
285
SC
RI PT
277
CONCLUSIONS
287
In a sample of wheelchair users with SCI, wheelchair breakdowns that led to injury, being
289
stranded, or an inability to attend medical appointments, school, or work were associated with
290
worse pain and perceived health, and greater odds of rehospitalization and pressure injury
291
development. Limiting these immediate consequences by intervening early may be a strategy to
292
prevent secondary health complications and preserve quality of life. Wheelchair factors ,
293
including breakdowns, should be considered in future investigations related to secondary
294
consequences and quality of life. Additional investigation is warranted to better understand the
295
cause-and-effect relationship between breakdowns, their immediate consequences, and
296
secondary health outcomes.
AC C
EP
TE D
288
ACCEPTED MANUSCRIPT
EP
TE D
M AN U
SC
RI PT
1. Hastings JD, Fanucchi ER, Burns SP. Wheelchair configuration and postural alignment in persons with spinal cord injury. Archives of physical medicine and rehabilitation 2003;84(4):528-34. 2. Dicianno BE, Arva J, Lieberman JM, Schmeler MR, Souza A, Phillips K et al. RESNA position on the application of tilt, recline, and elevating legrests for wheelchairs. Assistive technology : the official journal of RESNA 2009;21(1):13-22; quiz 4. 3. Samuelsson K, Larsson H, Thyberg M, Gerdle B. Wheelchair seating intervention. Results from a client-centred approach. Disability and rehabilitation 2001;23(15):677-82. 4. McClure LA, Boninger ML, Oyster ML, Williams S, Houlihan B, Lieberman JA et al. Wheelchair repairs, breakdown, and adverse consequences for people with traumatic spinal cord injury. Archives of physical medicine and rehabilitation 2009;90(12):2034-8. 5. Hansen R, Tresse S, Gunnarsson RK. Fewer accidents and better maintenance with active wheelchair check-ups: a randomized controlled clinical trial. Clinical rehabilitation 2004;18(6):631-9. 6. Worobey L, Oyster M, Nemunaitis G, Cooper R, Boninger ML. Increases in wheelchair breakdowns, repairs, and adverse consequences for people with traumatic spinal cord injury. American journal of physical medicine & rehabilitation / Association of Academic Physiatrists 2012;91(6):463-9. 7. Toro ML, Worobey L, Boninger ML, Cooper RA, Pearlman J. Type and Frequency of Reported Wheelchair Repairs and Related Adverse Consequences Among People With Spinal Cord Injury. Archives of physical medicine and rehabilitation 2016;97(10):1753-60. 8. Fitzgerald SG, Collins DM, Cooper RA, Tolerico M, Kelleher A, Hunt P et al. Issues in maintenance and repairs of wheelchairs: A pilot study. Journal of rehabilitation research and development 2005;42(6):853-62. 9. Consortium for Spinal Cord Medicine Clinical Practice G. Pressure ulcer prevention and treatment following spinal cord injury: a clinical practice guideline for health-care professionals. J Spinal Cord Med 2001;24 Suppl 1:S40-101. 10. Krause JS. Skin sores after spinal cord injury: relationship to life adjustment. Spinal cord 1998;36(1):51-6. 11. Gorecki C, Brown JM, Nelson EA, Briggs M, Schoonhoven L, Dealey C et al. Impact of pressure ulcers on quality of life in older patients: a systematic review. Journal of the American Geriatrics Society 2009;57(7):1175-83. 12. Fuhrer MJ, Garber SL, Rintala DH, Clearman R, Hart KA. Pressure ulcers in communityresident persons with spinal cord injury: prevalence and risk factors. Archives of physical medicine and rehabilitation 1993;74(11):1172-7. 13. Richards S, Waites K, Chen Y, Kogos S, Schmitt MM. The Epidemiology of Secondary Conditions Following Spinal Cord Injury. Topics in Spinal Cord Injury Rehabilitation 2004;10(1):15-29. 14. Leduc BE, Lepage Y. Health-related quality of life after spinal cord injury. Disability and rehabilitation 2002;24(4):196-202. 15. Putzke JD, Richards JS, Hicken BL, DeVivo MJ. Interference due to pain following spinal cord injury: important predictors and impact on quality of life. Pain 2002;100(3):231-42. 16. Budh CN, Osteraker AL. Life satisfaction in individuals with a spinal cord injury and pain. Clinical rehabilitation 2007;21(1):89-96.
AC C
297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340
ACCEPTED MANUSCRIPT
EP
TE D
M AN U
SC
RI PT
17. Charlifue S, Lammertse DP, Adkins RH. Aging with spinal cord injury: changes in selected health indices and life satisfaction. Archives of physical medicine and rehabilitation 2004;85(11):1848-53. 18. Cardenas DD, Hoffman JM, Kirshblum S, McKinley W. Etiology and incidence of rehospitalization after traumatic spinal cord injury: a multicenter analysis. Archives of physical medicine and rehabilitation 2004;85(11):1757-63. 19. Cardenas DD, Bryce TN, Shem K, Richards JS, Elhefni H. Gender and minority differences in the pain experience of people with spinal cord injury. Archives of physical medicine and rehabilitation 2004;85(11):1774-81. 20. Chen Y, Devivo MJ, Jackson AB. Pressure ulcer prevalence in people with spinal cord injury: age-period-duration effects. Archives of physical medicine and rehabilitation 2005;86(6):1208-13. 21. Bryce TN, Budh CN, Cardenas DD, Dijkers M, Felix ER, Finnerup NB et al. Pain after spinal cord injury: an evidence-based review for clinical practice and research. Report of the National Institute on Disability and Rehabilitation Research Spinal Cord Injury Measures meeting. J Spinal Cord Med 2007;30(5):421-40. 22. Ware JEJ, Sherbourne CD. The MOS 36-ltem Short-Form Health Survey (SF-36): I. Conceptual Framework and Item Selection. Medical Care 1992;30(6):473-83. 23. Whitehurst DG, Engel L, Bryan S. Short Form health surveys and related variants in spinal cord injury research: a systematic review. J Spinal Cord Med 2014;37(2):128-38. 24. Toro ML, Worobey L, Boninger ML, Cooper R, Pearlman J. Type and frequency of reported wheelchair repairs and related adverse consequences among people with spinal cord injury. Archives of physical medicine and rehabilitation. 25. Little RJA. A Test of Missing Completely at Random for Multivariate Data with Missing Values. Journal of the American Statistical Association 1988;83(404):1198-202. 26. McKinley WO, Jackson AB, Cardenas DD, DeVivo MJ. Long-term medical complications after traumatic spinal cord injury: a regional model systems analysis. Archives of physical medicine and rehabilitation 1999;80(11):1402-10. 27. Fitzgerald SG, Cooper RA, Boninger ML, Rentschler AJ. Comparison of fatigue life for 3 types of manual wheelchairs. Archives of physical medicine and rehabilitation 2001;82(10):1484-8. 28. Worobey L, Oyster M, Pearlman J, Gebrosky B, Boninger ML. Differences between manufacturers in reported power wheelchair repairs and adverse consequences among people with spinal cord injury. Archives of physical medicine and rehabilitation 2014;95(4):597-603. 29. Rentschler AJ, Cooper RA, Fitzgerald SG, Boninger ML, Guo S, Ammer WA et al. Evaluation of selected electric-powered wheelchairs using the ANSI/RESNA standards. Archives of physical medicine and rehabilitation 2004;85(4):611-9. 30. Toro ML, Bird E, Oyster M, Worobey L, Lain M, Bucior S et al. Development of a wheelchair maintenance training programme and questionnaire for clinicians and wheelchair users. Disability and rehabilitation Assistive technology 2017;12(8):843-51. 31. Worobey L, Pearlman J, Dyson-Hudson T, Boninger M. Clinician Competency with Wheelchair Maintenance and the Efficacy of a Wheelchair Maintenance Training Program. Archives of physical medicine and rehabilitation 2016;97(10):e55. 32. Di Marco A, Russell M, Masters M. Standards for wheelchair prescription. Australian Occupational Therapy Journal 2003;50(1):30-9.
AC C
341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385
ACCEPTED MANUSCRIPT
386 387 388
33. Krause JS, Vines CL, Farley TL, Sniezek J, Coker J. An exploratory study of pressure ulcers after spinal cord injury: relationship to protective behaviors and risk factors. Archives of physical medicine and rehabilitation 2001;82(1):107-13.
AC C
EP
TE D
M AN U
SC
RI PT
389
ACCEPTED MANUSCRIPT
Figure 1. Average pain severity and self-perceived health status scores between the three
391
wheelchair breakdown groups.
392
Pain (bottom) and perceived health status (top) scores are higher in the group who experienced
393
immediate consequences from their breakdowns, when compare to those who reported no
394
breakdowns. Solid white bar = subjects who reported no failures. Solid grey bar = subjects who
395
reported one or more non-catastrophic failures. Striped bar = subjects who reported one or more
396
catastrophic failures. An asterisk denotes a significant difference with respect to the reference
397
group (no wheelchair breakdowns).
M AN U
SC
RI PT
390
398
Figure 2. Flow chart describing the influence of wheelchair failures on health outcomes.
400
Breakdowns that led to being stranded or injured, missing a medical appointment, or being
401
unable to attend work/school were associated with worse health outcomes. In contrast,
402
participants who reported no immediate consequences and no breakdowns at all maintained their
403
health status.
AC C
EP
TE D
399
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 1. Descriptive statistics of demographics and failures for complete and missing datasets, and between-groups analyses. Nominal variables are presented as frequencies (percentages) and continuous variables as means (standard deviations). Complete (N=610) Missing Missing N 151 Age 44.1 (13.8) 44.4 (14.9) 464 (76.1) 111 (76.6) White (r) Race 145 146 (23.9) 34 (23.4) Minority 490 (80.3) 117 (76.5) Male (r) Sex 153 120 (19.7) 36 (23.5) Female 216 (35.4) 53 (34.4) Married/LWP (r) 273 (44.8) 73 (47.4) Marital Status 154 Single 121 (19.8) 28 (18.2) Other 277 (45.4) 65 (42.5) >HS (r) Education 153 332 (54.6) 88 (57.5) HS or less 331 (54.3) 67 (48.9) Paraplegia (r) Injury Level 137 279 (45.7) 70 (51.1) Tetraplegia 161 (26.4) 40 (26.5) Employed (r) 257 (42.1) 41 (27.2) Employment* 151 Unemployed† 192 (31.5) 70 (46.4) Other† 216 (35.4) 40 (28.0) Private (r) 218 (35.7) 51 (35.7) Medicare Payer 143 130 (21.3) 34 (23.8) Medicaid 46 (7.5) 18 (12.6) Other 256 (42.0) 38 (35.5) None (r) 244 (40.0) 47 (43.9) Breakdowns 107 ≥1 BD, No IC 110 (18.0) 22 (20.6) ≥1 BD, ≥1 IC Notes. The Missing N column denotes how many participants were in the “missing” group for each dependent variable. These were subjects who had data for the respective variable but were missing data for one or more other variables. A (r) denotes the reference category for each variable. LWP = living with partner. BD = wheelchair breakdown. IC = immediate consequence (i.e. stranded; injured; missed medical appointments, school, or work). * Significant chi-square test, Χ2(2, N = 761) = 14.722, p<.01, V=.14 † Significant post hoc analysis with Bonferonni correction (p<.05)
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 2. Outcome variable descriptive statistics for complete and missing datasets, and between-groups analyses. Nominal variables are presented as frequencies (percentages) and continuous variables as means (standard deviations). Complete* Missing Missing N SPHS 2.85 (1.1) 2.93 (1.2) 148 Pain 4.26 (2.8) 4.62 (3.0) 147 Pressure Injury None (r) 418 (68.2) 52 (35.6) 146 At least one 195 (31.8) 94 (64.4) Rehospitalization† None (r)‡ 386 (63.3) 37 (47.4) Unrelated to PU‡ 174 (28.5) 37 (47.4) 78 Related to PU 50 (8.2) 4 (5.1) Notes. The Missing N column denotes how many participants were in the “missing” group for each dependent variable. These were subjects who had data for the respective variable but were missing data for one or more other variables. The (r) indicates the reference category for each variable. SPHS = self-perceived health status. * N=610 † Significant chi-square test, Χ2(2, N = 688) =11.718, p<.01, V=.13 ‡ Significant post hoc analysis with Bonferonni correction (p<.05)
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 3. General-linear model statistics of the association between pain and wheelchair breakdowns, and demographic factors. Full statistics are presented only for significant predictors. 95% CI B B B SE Lower Upper p R2 Intercept 2.01 0.22 1.58 2.44 <.001 .124 Breakdowns ≥1 BD, No IC .443 ≥1 BD, ≥1 IC† 0.83 0.31 0.22 1.44 .008 .012 Race Minority .082 Sex Female .073 Marital Status Single .816 Other .965 Education ≤HS Diploma* 0.61 0.24 0.14 1.08 .012 .011 Injury Level Tetraplegia .111 Employment Unemployed .162 Other .635 Payer Medicare† 0.75 0.28 0.19 1.30 .008 .012 Medicaid* 0.78 0.36 0.10 1.46 .025 .008 Other* 0.92 0.45 0.05 1.80 .039 .007 Age .125 Notes. Full model: F(14, 595)=4.247, p<.001, R2=.091, Adj R2 =.069. BD = wheelchair breakdown. IC = immediate consequence (i.e. stranded; injured; missed medical appointments, school, or work). * p<.05 † p<.01
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 4. General-linear model statistics of the association between self-perceived health and wheelchair breakdowns. Full statistics are presented only for significant predictors. 95% CI B B B SE Lower Upper p R2 Intercept 2.01 0.22 1.58 2.44 <.001 .124 Breakdowns ≥1 BD, No IC .762 ≥1 BD, ≥1 IC* 0.26 0.12 0.02 0.50 .032 .009 Race Minority .511 Sex Female .059 Marital Status Single .809 Other .901 Education ≤HS Diploma .057 Injury Level Tetraplegia .426 Employment Unemployed* 0.29 0.12 0.06 0.52 .014 .010 Other .148 Payer Medicare .097 Medicaid* 0.35 0.13 0.08 0.61 .010 .011 Other .908 Age* 0.01 0.01 0.00 0.02 .005 .012 2 2 Notes. Full model: F(14, 595)=3.868, p<.001, R =.083, Adjusted R = .062. BD = wheelchair breakdown. IC = immediate consequence (i.e. stranded; injured; missed medical appointments, school, or work). * p<.05
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 5. Logistic regression model statistics depicting the odds of self-reported pressure injuries within the past twelve months associated with wheelchair breakdowns, and demographic factors. 95% OR B B SE Wald Χ2 df p OR Lower Upper Breakdowns* 6.37 2 0.04 ≥1 BD, No IC -0.05 0.202 0.06 1 0.81 0.97 0.64 1.41 ≥1 BD, ≥1 IC* 0.55 0.25 4.87 1 0.03 1.73 1.06 2.83 Race 0.365 1 0.55 Sex 1.57 1 0.21 Marital Status 1.95 2 0.38 Education 1.81 1 0.18 Employment 0.378 2 0.83 Injury Level 0.621 1 0.62 Insurance 2.68 3 0.44 Age 1.47 1 .225 Notes. Model statistics for insignificant independent variables were omitted. BD = wheelchair breakdown. IC = immediate consequence (i.e. stranded; injured; missed medical appointments, school, or work). *p<.05
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 6. Logistic regression model statistics depicting the odds of rehospitalization unrelated to (top block) and related to skin breakdown (bottom block) within the past twelve months, associated with wheelchair breakdowns and demographic factors. 95% OR B B SE Wald Χ2* p OR Lower Upper Breakdowns ≥1 BD, No IC .123 ≥1 BD, ≥1 IC* .593 .265 5.015 .025 1.810 1.077 3.042 Race Minority .225 Sex Female .307 Marital Status Single .245 Other .366 Education ≤High School .337 Employment Unemployed .284 Other† .862 .282 9.352 .002 2.367 1.363 4.113 Injury Level Tetraplegia* .506 .197 6.610 .010 1.658 1.128 2.437 Insurance Other .347 Medicaid‡ .504 .293 2.965 .085 1.656 .933 2.939 Medicare .499 Age .425 Breakdowns ≥1 BD, No IC .119 ≥1 BD, ≥1 IC‡ .775 .434 3.195 .074 2.171 .928 5.078 Race Minority .584 Sex Female .792 Marital Status Single .776 Other .705 Education ≤High School .925 Employment Unemployed .259 Other .518 Injury Level Tetraplegia .254 Insurance Other .911 Medicaid .874 Medicare .465 Age .327 Notes. Model statistics for insignificant independent variables were omitted. BD = wheelchair breakdown. IC = immediate consequence (i.e. stranded; injured; missed medical appointments, school, or work). *p<.05 † p<.01 ‡ p<.10
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
S0003-9993(18)30233-8
DOI:
10.1016/j.apmr.2018.04.002
Reference:
YAPMR 57205
To appear in:
ARCHIVES OF PHYSICAL MEDICINE AND REHABILITATION
Received Date: 19 January 2018 Revised Date:
28 February 2018
Accepted Date: 7 April 2018
Please cite this article as: Hogaboom NS, Worobey LA, Houlihan BV, Heinemann AW, Boninger ML, Wheelchair breakdowns are associated with pain, pressure injuries, rehospitalization, and self-perceived health in full-time wheelchair users with spinal cord injury, ARCHIVES OF PHYSICAL MEDICINE AND REHABILITATION (2018), doi: 10.1016/j.apmr.2018.04.002. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
Title: Wheelchair breakdowns are associated with pain, pressure injuries, rehospitalization, and self-
Running Head: Wheelchair breakdowns and health outcomes
SC
Authors:
Lynn A Worobey, PhD, DPT, ATP Bethlyn V. Houlihan, MSW, MPH
TE D
Allen W. Heinemann, PhD
M AN U
Nathan S Hogaboom, PhD
Michael L Boninger, MD
RI PT
perceived health in full-time wheelchair users with spinal cord injury
EP
From the Human Engineering Research Laboratories, VA Pittsburgh Healthcare System, Pittsburgh, PA (Hogaboom, Worobey, Boninger); Department of Rehabilitation Science and
AC C
Technology, School of Health and Rehabilitation Sciences, University of Pittsburgh, Pittsburgh, PA (Hogaboom, Boninger); Department of Physical Medicine and Rehabilitation, School of Medicine, University of Pittsburgh, Pittsburgh, PA (Worobey, Boninger); Spaulding New England Regional SCI Center, Boston, MA (Houlihan); School of Public Health, Boston University, Boston, MA (Houlihan); Center for Rehabilitation Outcomes Research,
ACCEPTED MANUSCRIPT
Rehabilitation Institute of Chicago, Chicago, Illinois (Heinemann); Department of Physical Medicine and Rehabilitation, Northwestern University, Chicago, Illinois (Heinemann).
RI PT
This material is the result of work supported with resources and the use of facilities at the Human Engineering Research Laboratories, VA Pittsburgh Healthcare System. The contents of this paper do not represent the views of the Department of Veterans Affairs or the United States
SC
Government. This project was supported by the National Institute on Disability, Independent Living, and Rehabilitation Research (NIDILRR grant #90SI5008). NIDILRR is a Center within
M AN U
the Administration for Community Living (ACL), Department of Health and Human Services (HHS). The contents of this (insert type of publication; e.g., book, report, film) do not necessarily represent the policy of NIDILRR, ACL, HHS, and you should not assume endorsement by the Federal Government. This material is based upon work supported by the
TE D
National Science Foundation Graduate Research Fellowship (#1247842) and IGERT Fellowship (DGE 1144584). The contents of this publication do not necessarily reflect the view of the
EP
National Science Foundation.
Data were presented at the 2016 American Congress of Rehabilitation Meeting (November 2-4,
AC C
Chicago, Illinois, USA) and the 2017 International Seating Symposium (March 2-4, Nashville, TN, USA).
Address all correspondence to: Lynn Worobey 6425 Penn Avenue, Suite 400
ACCEPTED MANUSCRIPT
Pittsburgh, PA 15206 Phone: 412-822-3700 Fax: 412-822-3699
AC C
EP
TE D
M AN U
SC
RI PT
Email: [email protected]
ACCEPTED MANUSCRIPT
ABSTRACT
2
Objectives: To evaluate the relationship between wheelchair breakdowns, their immediate
3
consequences, and secondary health complications after spinal cord injury (SCI). “Immediate
4
consequences” occur when part of a wheelchair breaks and leaves an individual stranded or
5
injured, or causes them to miss medical appointments, work, or school.
6
Design: Survey, cross-sectional.
7
Setting: Spinal Cord Injury Model Systems Centers (SCIMS).
8
Participants: 771 full-time wheelchair users with SCI from 9 SCIMS centers, with data collected
9
between 2011 and 2016.
M AN U
SC
RI PT
1
Interventions: N/A
11
Main Outcomes: Incidence of self-reported wheelchair breakdowns within the past 6 months that
12
did or did not result in immediate consequences (i.e. injury, being stranded, missing a medical
13
appointment, or an inability to attend school/work); Self-Perceived Health Status (SPHS) scale;
14
pain severity numerical rating scale; rehospitalizations and self-reported pressure injury
15
development within the past 12 months.
16
Results: 610 participants with complete datasets were included in the analyses. When compared
17
to those who reported no breakdowns, participants who reported one or more immediate
18
consequences had worse secondary complications: higher SPHS and pain scores (partial-
19
η2=.009-.012, p<.05), and higher odds of rehospitalization (OR: 1.86, p<.05) and pressure injury
20
development (OR: 1.73, p<.05). Secondary complications were not different in those who
21
reported no immediate consequences compared to those who reported no breakdown.
22
Conclusions: Wheelchair breakdowns that resulted in injury, being stranded, missing medical
23
appointments, and/or an inability to attend work/school appear to have far-reaching impacts on
AC C
EP
TE D
10
ACCEPTED MANUSCRIPT
health and secondary injury. Preventing wheelchair breakdowns, either through better
25
maintenance or manufacturing, may be a means of decreasing secondary disability.
AC C
EP
TE D
M AN U
SC
RI PT
24
ACCEPTED MANUSCRIPT
26
INTRODUCTION
27
Sustaining a spinal cord injury (SCI) has a profound impact on an individual’s life. Independent
29
mobility and participation may be limited by environmental barriers. For those who do not regain
30
the ability to ambulate, many use a wheelchair to achieve full participation. In addition to
31
enhancing mobility, many wheelchairs are equipped with technology that helps to manage pain
32
and pressure; for example, tilt and recline functions in power wheelchairs or adjustable seating in
33
manual chairs.1-3 Despite the wheelchair’s importance, our research group reported that greater
34
than half of wheelchair users with SCI experience a wheelchair breakdown. Between 20% and
35
30% of those individuals experienced an immediate consequence that left them stranded or
36
injured, or prevented them from attending medical appointments, school, or work.4-7 The repair
37
process can be difficult to navigate for wheelchair users8; as a result, nearly 7% of individuals do
38
not receive the necessary repairs for their failed wheelchair components.7
SC
M AN U
TE D
39
RI PT
28
A serious breakdown could also relegate an individual to an inappropriate seating surface, such
41
as a couch or bed, which does not provide adequate postural support or pressure management.
42
By removing a primary means of self-care management, wheelchair breakdowns put the user at
43
risk of exacerbating secondary conditions. These potential secondary effects include pressure
44
injuries, pain, and recurrent hospitalizations, all of which are correlated with lower self-
45
perceived health and quality of life.9-17 Previous investigations using SCI Model Systems data
46
have reported several subject characteristics associated with worse health outcomes: minority
47
race, tetraplegia, public insurance payer, unemployment, less education, and being single. 18-20
48
However, no studies have incorporated wheelchair-related factors in the analyses. By omitting
AC C
EP
40
ACCEPTED MANUSCRIPT
49
the wheelchair in their analyses, investigators may be missing a potentially critical factor that can
50
reduce secondary complication risks.
51
This study is an extension of our previous work, which demonstrated that the incidence of
53
wheelchair breakdowns continues to rise.4, 6, 7 These breakdowns can play an important role in
54
user safety, positioning, and mobility. This study uses the same dataset to investigate whether
55
wheelchair breakdowns are related to secondary health outcomes. The rationale behind this study
56
is to explore and provide evidence of the potential secondary effects of wheelchair breakdowns
57
in the form of worsening secondary health complications. If the following hypotheses are met,
58
this analysis would show the importance of breakdown prevention and the further investigation
59
of preventative methods. Participants who reported one or more wheelchair breakdowns, with
60
and without immediate consequences, were hypothesized to report worse secondary health
61
effects when controlling for other risk factors (older age, minority race, female sex, non-private
62
insurance coverage, single, unemployment, less education, and cervical injury level). Health-
63
related outcome indicators were pain, self-perceived health, rehospitalizations, and self-reported
64
incidence of pressure injuries.
65
EP
TE D
M AN U
SC
RI PT
52
METHODS
67
Subjects
68
Participants were eligible if they were older than 16 years, had a neurologic impairment resulting
69
from a SCI that occurred at least 1 year prior to the study, and used a manual- or power-
70
wheelchair (including power-assisted manual chairs) for at least 40 hours per week. Participants
71
were enrolled at one nine SCI Model Systems Centers March, 2013 and January, 2016.
AC C
66
ACCEPTED MANUSCRIPT
Recruitment methods included approaching individuals who participated in the National SCI
73
Database, local registries, flyers, and identification by clinical staff. Participants completed
74
surveys during interviews, over the phone, or via mail. All centers obtained ethical approval
75
from their local institutional review boards and all participants provided informed consent.
76
Further details of data collection and the specific Model Systems Centers can be found in Toro,
77
et al.7
RI PT
72
79
SC
78
Data Collection
M AN U
80
This is a secondary analysis of the 2016 study conducted by Toro, et al.; additional information
82
about data collection procedures can be found in their study.7 Participants completed a
83
questionnaire that inquired about demographic characteristics and secondary health conditions.
84
Demographic characteristics included age, sex, race, employment, marital status, injury level,
85
and insurance coverage. Participants were asked how many repairs were needed to their
86
wheelchairs within the past 6 months, and whether any immediate consequences occurred as a
87
result of those breakdowns; responses included no consequence, missing a medical appointment,
88
missing work or school, being stranded, or being injured.4, 7 The breakdown questionnaire can be
89
found in the appendix of the 2009 study by McClure, et al.4
EP
AC C
90
TE D
81
91
Regarding secondary health effects, participants were asked about recent pain, self-perceived
92
health status (SPHS), pressure injury development, and rehospitalizations. Pain over the past
93
month was assessed using a 0 (no pain) to 10 (worst possible pain) numerical rating scale – an
94
assessment recommended for use in the SCI population.21 SPHS was determined from item 1 of
ACCEPTED MANUSCRIPT
the Medical Outcomes Study SF-36 scale, which asks individuals to rate how they view their
96
health using a scale of 1 (excellent) to 5 (poor).22 Although validity and reliability have not been
97
established for this item, both full and partial scales have been used extensively in medical
98
research including the SCI population.23 Participants provided information about the number of
99
days hospitalized over the past 12 months, and indicated one of 18 reasons for those
100
rehospitalizations based on a previous study.18 Of interest were those who reported a
101
rehospitalization from diseases of the skin/subcutaneous tissue, including pressure injuries. They
102
were asked whether they experienced the “presence of a pressure sore with open or broken skin
103
in the past 12 months.” Response options were “yes,” “no,” and “I do not know.”
104
M AN U
SC
RI PT
95
Data Reduction and Analysis
106
All statistical analyses were performed using SPSS 23.0 (IBM, Inc., Armonk, New York, USA).
107
Significance was set a priori to α=.05. Descriptive statistics were calculated first, including
108
means and standard deviations for continuous variables and frequencies for nominal variables.
109
Demographic characteristics were categorized to optimize cell size, in accordance with previous
110
studies investigating wheelchair consequences or rehospitalizations4, 6, 18, 24: injury level into
111
tetraplegia vs. paraplegia; race into minority vs. white; education into high school diploma or
112
less vs. post-high school education; marital status into married/living with partner, never
113
married, or divorced/separated/widowed/other; employment status into employed, unemployed,
114
or homemaker/on-the-job-training/retired/student/other; and primary insurance payer into
115
private, Medicare, Medicaid, or other. Participants’ responses for breakdowns and immediate
116
consequences were categorized accordingly into one of three groups: 1) no wheelchair
117
breakdowns, 2) at least one breakdown without immediate consequences, or 3) at least one
AC C
EP
TE D
105
ACCEPTED MANUSCRIPT
breakdown with at least one immediate consequence (resulting in being stranded, injury, missed
119
medical appointment, or missed work/school). Rehospitalizations were recoded into no
120
rehospitalizations, rehospitalized at least once for reasons unrelated to skin disease, or
121
rehospitalized at least once specifically for skin disease.
122
RI PT
118
Participants with missing demographic characteristics, wheelchair breakdowns,
124
rehospitalizations, pressure injury development, self-perceived health, or pain data were
125
excluded. In this way, the same cohort of individuals would be included in each statistical model
126
and thus is sample is more consistent. Potential bias from this additional exclusion was
127
investigated using Pearson’s chi-square tests for nominal variables and independent t-tests for
128
continuous variables (excluded versus included). A follow-up analysis using multiple imputation
129
determined that removing the subjects with missing data did not influence the results. Thus, only
130
the cohort with complete datasets for all independent and dependent variables were included.
131
Furthermore, an insignificant Little’s test suggests data were missing completely at random
132
(Χ2=13.92, df=7, p>.05) and that analysis of the complete cases was valid.25
M AN U
TE D
EP
133
SC
123
Three regression models were built to test the association between wheelchair breakdowns with
135
and without immediate consequences and health-related quality of life variables. Using the
136
general linear model procedure, a linear-regression model with multiple outcomes was built to
137
test whether breakdowns with and without immediate consequences influenced SPHS and pain
138
scores. Two logistic regression models tested whether breakdowns with and without immediate
139
consequences increased odds of self-reported pressure injury development or rehospitalizations
140
related to, and unrelated to skin breakdown. Breakdown groups and demographic characteristics
AC C
134
ACCEPTED MANUSCRIPT
141
were used as predictors: age, sex, race, injury level, education, marital status, employment status,
142
and primary insurance payer. Group assignments and reference categories are located in Tables 1
143
and 2.
145
RI PT
144
RESULTS
146
Subjects
148
The data set consisted of 771 full-time wheelchair users with SCI; 161 (20.9%) had missing data
149
and were excluded. Demographic and wheelchair breakdown data are presented in Table 1.
150
Participants with missing data were more often unemployed and less often in the “other”
151
employment group; more often reported rehospitalizations unrelated to pressure injuries; and less
152
often reported no rehospitalizations (p<.01, Tables 1 and 2). No other differences were found
153
with respect to demographic characteristics, wheelchair breakdowns, or secondary condition
154
outcomes. Missing data sample sizes varied because many subjects with missing data for one
155
variable were not missing data from other variables.
TE D
M AN U
SC
147
EP
156
Of those included in the analysis, 233 (38.2%) used a power chair, 361 (59.2%) were manual
158
wheelchair users, and 16 (2.6%) were power-assisted manual chair users; the latter two groups
159
were combined. The most common immediate consequence from a breakdown was being
160
stranded (12.1%), followed by missing a medical appointment (6.7%), missing work/school
161
(3.6%), and being injured (2.9%). 363 (42.3%) reported having at least one working backup
162
chair.
163
AC C
157
ACCEPTED MANUSCRIPT
Pain and Perceived Health
165
The multivariate linear regression model found that breakdowns without immediate
166
consequences were not associated with pain or perceived-health (Tables 3 and 4). Immediate
167
consequences were found to have small, negative effects on SPHS and pain scores (p<.05,
168
R2=.009-.012; Tables 3 and 4). That is, participants who were stranded, were injured, missed
169
medical appointments, or missed work/school because of a wheelchair breakdown reported
170
worse pain and self-perceived health (Figure 1). Less education and insurance coverage other
171
than private were also associated with more pain (p<.05; Table 3). Unemployment, older age,
172
and Medicaid insurance coverage were associated with worse self-perceived health (p<.05; Table
173
4).
174
M AN U
SC
RI PT
164
Pressure Injuries and Hospitalizations
176
Results of the logistic regression models found no relationship between breakdowns without
177
immediate consequences and rehospitalization or pressure injury development (Tables 5 and 6).
178
Those who experienced immediate consequences after a breakdown had higher odds of
179
rehospitalization both related to (OR: 2.17, p<.10) and unrelated to skin disease (OR: 1.81,
180
p<.05). These individuals also had higher odds of developing a pressure injury within the past 12
181
months (OR: 1.72, p<.05). Individuals with tetraplegia (OR: 1.66, p<.05), in the “other”
182
employment group (OR: 2.37, p<.01), and with Medicaid coverage (OR: 1.66, p<.05) had higher
183
odds of rehospitalization unrelated to skin disease. No other demographic characteristic was
184
related to self-reported rehospitalizations or pressure injury development.
185
AC C
EP
TE D
175
ACCEPTED MANUSCRIPT
186
DISCUSSION
187
The present study advances the available knowledge about wheelchair breakdowns and their
189
secondary effects on health outcomes. This was accomplished by establishing a connection
190
between notable health-related outcomes and breakdowns that resulted in immediate
191
consequences of being stranded, being injured, missing medical appointments, or missing
192
work/school; specifically, pain, pressure injuries, rehospitalizations, and perceived health.
193
Although the sample consisted of only people with SCI, wheelchair failures are not unique to
194
SCI and thus these findings can be applied to active users with different diagnoses who rely upon
195
their wheelchairs for mobility.
196
M AN U
SC
RI PT
188
Users that reported needing a repair on their wheelchair, but not an immediate consequence, did
198
not appear to have worse secondary health outcomes than those who reported not needing any
199
repairs. However, breakdowns that were serious enough to result in being stranded or injured,
200
missing medical appointments, or an inability to attend school or work did have secondary
201
effects (Figure 2). This result was found despite the lack of differences observed in available
202
backup chairs. The findings could mean the immediate consequences of the breakdown, as
203
opposed to the breakdown itself, increase the risk of developing or worsening secondary
204
conditions; for example, missing medical appointments or being injured directly led to poorer
205
health outcomes. It is also possible that the breakdown was serious enough to prevent the user
206
from using their device by disabling seating functions and postural supports, which allow the
207
user to manage pressure and trunk stability. However, it is important to note that, with respect to
208
the latter, the “immediate consequence” classification was considered a surrogate for device
AC C
EP
TE D
197
ACCEPTED MANUSCRIPT
usability. The assumption was made that if an immediate consequence was reported, the
210
breakdown was serious enough to render the device unusable which ultimately led to worse
211
health outcomes. Because of this assumption, the authors cannot definitively conclude that
212
breakdowns led to secondary health consequences. This question does warrant further
213
investigation, however, because immediate consequences from breakdowns can be avoided
214
through early intervention. A more focused and detailed questionnaire about device usability
215
because of the breakdown would provide a more thorough analysis.
SC
RI PT
209
216
These results support previous reports, which indicated between 44% and 64% of wheelchair
218
users experience at least one breakdown in the past 6 months, with between 19.7% and 30.5% of
219
them reporting an immediate consequence.4, 6, 7 In contrast to previous works utilizing SCI
220
Model Systems datasets18, 20, 26, no relationships were observed between demographic
221
characteristics and pressure injury development or rehospitalizations. This finding may reflect
222
our use of self-report data in place of physician documentation20, 26 and inclusion only of
223
wheelchair users.18, 20, 26 It is also possible that wheelchair-related factors mitigate pressure injury
224
development and rehospitalizations more than personal characteristics. Considering these
225
findings, inclusion of wheelchair-related factors in future longitudinal investigations may
226
provide a more complete depiction of the risks of developing secondary conditions.
TE D
EP
AC C
227
M AN U
217
228
Implications for Clinical Practice
229
In many cases, failures result from factors outside of the user’s control, making it difficult to
230
identify supportive interventions. Unfortunately, data concerning the relationship between
231
wheelchair usage and failures is limited. Toro, et al. found that the number of failures was
ACCEPTED MANUSCRIPT
unrelated to the number of hours spent outside of the house, hours out of bed, or nights spent out
233
of the house.7 Toro, et al.’s results, and those of other investigators, suggest that wheelchair
234
type4, 6, 27 (i.e., manual versus power) and manufacturer28, 29 have more influence over failure
235
rates than wheelchair usage. In turn, it would not make sense for clinicians to discourage
236
individuals with SCI from using their wheelchairs as a means to prevent a breakdown and
237
subsequent secondary health issues. Rather, training wheelchair users to maintain their chairs and
238
to recognize a potential breakdown may be a cost-effective way to avoid the breakdowns that
239
lead to negative health outcomes.5, 30
M AN U
240
SC
RI PT
232
There is some evidence that suggests wheelchair seating clinicians are often unfamiliar with how
242
to identify and rectify broken wheelchair components.31 Providing clinicians with the knowledge
243
to help their patients navigate the repair process may lead to more timely repairs, which could
244
reduce the likelihood of an immediate consequence from occurring.32 A wheelchair maintenance
245
training program for clinicians and wheelchair users was recently developed using World Health
246
Organization basic wheelchair service curriculum. Initial testing showed that it improved
247
knowledge on basic maintenance.30 Further development of these training materials, and their
248
dissemination to wheelchair users and clinicians, may be a potential solution to preventing
249
secondary complications caused by wheelchair breakdowns and should be considered for future
250
investigation.
EP
AC C
251
TE D
241
252
Limitations and Subsequent Areas for Future Research
253
This study was exploratory and cross-sectional. As a result, we are unable to describe causal
254
relationships. Future studies could examine these relationships over time. The effect sizes
ACCEPTED MANUSCRIPT
between breakdowns and health outcomes were small, yet similar in magnitude to other subject
256
characteristics in the models, such as insurance type or age; thus, their importance should not be
257
minimized. Future research could ascertain the impact of multiple insurance types on
258
intermediate and distal outcomes.
RI PT
255
259
Participants were asked for a binary report of a limited number of breakdown-related immediate
261
consequences (e.g., being stranded or injured, missing medical appointments or work/school),
262
which may result in under-reporting the number of catastrophic failures. This line of questioning
263
also did not provide any information about whether the wheelchair could be used after the
264
breakdown; the length of time the wheelchair could not be used; or the number of instances the
265
user experienced an immediate consequence after a breakdown. The lack of information made it
266
difficult to determine how and why the breakdown led to the poorer health outcomes that were
267
observed. Asking open-ended questions may provide more detailed information for further study.
268
As well, many individuals reported experiencing more than one of the aforementioned
269
consequences. These individuals were grouped together to maximize cell size and strengthen
270
statistical power; however, this prevented further probing into the effects of specific breakdown-
271
related consequences on health outcomes. A larger sample size would lead to a more thorough
272
analysis.
M AN U
TE D
EP
AC C
273
SC
260
274
The timeframes of the different outcome variables (e.g., breakdowns within the past 6 months,
275
pain within the past week, rehospitalizations within the past 12 months) were not identical. These
276
differences may have affected the results. Health-promoting or self-destructive behaviors were
ACCEPTED MANUSCRIPT
not analyzed; these factors are associated with health status, pain, and pressure injury
278
development33 and may affect how proactive individuals are in maintaining their chairs.
279
Participants were treated at SCI Model Centers and thus results may not be generalizable to the
280
SCI population. Considering the level of care provided at these centers, the outcomes
281
participants achieve may be better than in less-resourced facilities. Finally, participants excluded
282
from the analysis due to missing data were more likely to be unemployed and not rehospitalized
283
than those included in the analyses. It is unclear if the same results would have been reported if
284
fewer cases had missing data.
286
M AN U
285
SC
RI PT
277
CONCLUSIONS
287
In a sample of wheelchair users with SCI, wheelchair breakdowns that led to injury, being
289
stranded, or an inability to attend medical appointments, school, or work were associated with
290
worse pain and perceived health, and greater odds of rehospitalization and pressure injury
291
development. Limiting these immediate consequences by intervening early may be a strategy to
292
prevent secondary health complications and preserve quality of life. Wheelchair factors ,
293
including breakdowns, should be considered in future investigations related to secondary
294
consequences and quality of life. Additional investigation is warranted to better understand the
295
cause-and-effect relationship between breakdowns, their immediate consequences, and
296
secondary health outcomes.
AC C
EP
TE D
288
ACCEPTED MANUSCRIPT
EP
TE D
M AN U
SC
RI PT
1. Hastings JD, Fanucchi ER, Burns SP. Wheelchair configuration and postural alignment in persons with spinal cord injury. Archives of physical medicine and rehabilitation 2003;84(4):528-34. 2. Dicianno BE, Arva J, Lieberman JM, Schmeler MR, Souza A, Phillips K et al. RESNA position on the application of tilt, recline, and elevating legrests for wheelchairs. Assistive technology : the official journal of RESNA 2009;21(1):13-22; quiz 4. 3. Samuelsson K, Larsson H, Thyberg M, Gerdle B. Wheelchair seating intervention. Results from a client-centred approach. Disability and rehabilitation 2001;23(15):677-82. 4. McClure LA, Boninger ML, Oyster ML, Williams S, Houlihan B, Lieberman JA et al. Wheelchair repairs, breakdown, and adverse consequences for people with traumatic spinal cord injury. Archives of physical medicine and rehabilitation 2009;90(12):2034-8. 5. Hansen R, Tresse S, Gunnarsson RK. Fewer accidents and better maintenance with active wheelchair check-ups: a randomized controlled clinical trial. Clinical rehabilitation 2004;18(6):631-9. 6. Worobey L, Oyster M, Nemunaitis G, Cooper R, Boninger ML. Increases in wheelchair breakdowns, repairs, and adverse consequences for people with traumatic spinal cord injury. American journal of physical medicine & rehabilitation / Association of Academic Physiatrists 2012;91(6):463-9. 7. Toro ML, Worobey L, Boninger ML, Cooper RA, Pearlman J. Type and Frequency of Reported Wheelchair Repairs and Related Adverse Consequences Among People With Spinal Cord Injury. Archives of physical medicine and rehabilitation 2016;97(10):1753-60. 8. Fitzgerald SG, Collins DM, Cooper RA, Tolerico M, Kelleher A, Hunt P et al. Issues in maintenance and repairs of wheelchairs: A pilot study. Journal of rehabilitation research and development 2005;42(6):853-62. 9. Consortium for Spinal Cord Medicine Clinical Practice G. Pressure ulcer prevention and treatment following spinal cord injury: a clinical practice guideline for health-care professionals. J Spinal Cord Med 2001;24 Suppl 1:S40-101. 10. Krause JS. Skin sores after spinal cord injury: relationship to life adjustment. Spinal cord 1998;36(1):51-6. 11. Gorecki C, Brown JM, Nelson EA, Briggs M, Schoonhoven L, Dealey C et al. Impact of pressure ulcers on quality of life in older patients: a systematic review. Journal of the American Geriatrics Society 2009;57(7):1175-83. 12. Fuhrer MJ, Garber SL, Rintala DH, Clearman R, Hart KA. Pressure ulcers in communityresident persons with spinal cord injury: prevalence and risk factors. Archives of physical medicine and rehabilitation 1993;74(11):1172-7. 13. Richards S, Waites K, Chen Y, Kogos S, Schmitt MM. The Epidemiology of Secondary Conditions Following Spinal Cord Injury. Topics in Spinal Cord Injury Rehabilitation 2004;10(1):15-29. 14. Leduc BE, Lepage Y. Health-related quality of life after spinal cord injury. Disability and rehabilitation 2002;24(4):196-202. 15. Putzke JD, Richards JS, Hicken BL, DeVivo MJ. Interference due to pain following spinal cord injury: important predictors and impact on quality of life. Pain 2002;100(3):231-42. 16. Budh CN, Osteraker AL. Life satisfaction in individuals with a spinal cord injury and pain. Clinical rehabilitation 2007;21(1):89-96.
AC C
297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340
ACCEPTED MANUSCRIPT
EP
TE D
M AN U
SC
RI PT
17. Charlifue S, Lammertse DP, Adkins RH. Aging with spinal cord injury: changes in selected health indices and life satisfaction. Archives of physical medicine and rehabilitation 2004;85(11):1848-53. 18. Cardenas DD, Hoffman JM, Kirshblum S, McKinley W. Etiology and incidence of rehospitalization after traumatic spinal cord injury: a multicenter analysis. Archives of physical medicine and rehabilitation 2004;85(11):1757-63. 19. Cardenas DD, Bryce TN, Shem K, Richards JS, Elhefni H. Gender and minority differences in the pain experience of people with spinal cord injury. Archives of physical medicine and rehabilitation 2004;85(11):1774-81. 20. Chen Y, Devivo MJ, Jackson AB. Pressure ulcer prevalence in people with spinal cord injury: age-period-duration effects. Archives of physical medicine and rehabilitation 2005;86(6):1208-13. 21. Bryce TN, Budh CN, Cardenas DD, Dijkers M, Felix ER, Finnerup NB et al. Pain after spinal cord injury: an evidence-based review for clinical practice and research. Report of the National Institute on Disability and Rehabilitation Research Spinal Cord Injury Measures meeting. J Spinal Cord Med 2007;30(5):421-40. 22. Ware JEJ, Sherbourne CD. The MOS 36-ltem Short-Form Health Survey (SF-36): I. Conceptual Framework and Item Selection. Medical Care 1992;30(6):473-83. 23. Whitehurst DG, Engel L, Bryan S. Short Form health surveys and related variants in spinal cord injury research: a systematic review. J Spinal Cord Med 2014;37(2):128-38. 24. Toro ML, Worobey L, Boninger ML, Cooper R, Pearlman J. Type and frequency of reported wheelchair repairs and related adverse consequences among people with spinal cord injury. Archives of physical medicine and rehabilitation. 25. Little RJA. A Test of Missing Completely at Random for Multivariate Data with Missing Values. Journal of the American Statistical Association 1988;83(404):1198-202. 26. McKinley WO, Jackson AB, Cardenas DD, DeVivo MJ. Long-term medical complications after traumatic spinal cord injury: a regional model systems analysis. Archives of physical medicine and rehabilitation 1999;80(11):1402-10. 27. Fitzgerald SG, Cooper RA, Boninger ML, Rentschler AJ. Comparison of fatigue life for 3 types of manual wheelchairs. Archives of physical medicine and rehabilitation 2001;82(10):1484-8. 28. Worobey L, Oyster M, Pearlman J, Gebrosky B, Boninger ML. Differences between manufacturers in reported power wheelchair repairs and adverse consequences among people with spinal cord injury. Archives of physical medicine and rehabilitation 2014;95(4):597-603. 29. Rentschler AJ, Cooper RA, Fitzgerald SG, Boninger ML, Guo S, Ammer WA et al. Evaluation of selected electric-powered wheelchairs using the ANSI/RESNA standards. Archives of physical medicine and rehabilitation 2004;85(4):611-9. 30. Toro ML, Bird E, Oyster M, Worobey L, Lain M, Bucior S et al. Development of a wheelchair maintenance training programme and questionnaire for clinicians and wheelchair users. Disability and rehabilitation Assistive technology 2017;12(8):843-51. 31. Worobey L, Pearlman J, Dyson-Hudson T, Boninger M. Clinician Competency with Wheelchair Maintenance and the Efficacy of a Wheelchair Maintenance Training Program. Archives of physical medicine and rehabilitation 2016;97(10):e55. 32. Di Marco A, Russell M, Masters M. Standards for wheelchair prescription. Australian Occupational Therapy Journal 2003;50(1):30-9.
AC C
341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385
ACCEPTED MANUSCRIPT
386 387 388
33. Krause JS, Vines CL, Farley TL, Sniezek J, Coker J. An exploratory study of pressure ulcers after spinal cord injury: relationship to protective behaviors and risk factors. Archives of physical medicine and rehabilitation 2001;82(1):107-13.
AC C
EP
TE D
M AN U
SC
RI PT
389
ACCEPTED MANUSCRIPT
Figure 1. Average pain severity and self-perceived health status scores between the three
391
wheelchair breakdown groups.
392
Pain (bottom) and perceived health status (top) scores are higher in the group who experienced
393
immediate consequences from their breakdowns, when compare to those who reported no
394
breakdowns. Solid white bar = subjects who reported no failures. Solid grey bar = subjects who
395
reported one or more non-catastrophic failures. Striped bar = subjects who reported one or more
396
catastrophic failures. An asterisk denotes a significant difference with respect to the reference
397
group (no wheelchair breakdowns).
M AN U
SC
RI PT
390
398
Figure 2. Flow chart describing the influence of wheelchair failures on health outcomes.
400
Breakdowns that led to being stranded or injured, missing a medical appointment, or being
401
unable to attend work/school were associated with worse health outcomes. In contrast,
402
participants who reported no immediate consequences and no breakdowns at all maintained their
403
health status.
AC C
EP
TE D
399
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 1. Descriptive statistics of demographics and failures for complete and missing datasets, and between-groups analyses. Nominal variables are presented as frequencies (percentages) and continuous variables as means (standard deviations). Complete (N=610) Missing Missing N 151 Age 44.1 (13.8) 44.4 (14.9) 464 (76.1) 111 (76.6) White (r) Race 145 146 (23.9) 34 (23.4) Minority 490 (80.3) 117 (76.5) Male (r) Sex 153 120 (19.7) 36 (23.5) Female 216 (35.4) 53 (34.4) Married/LWP (r) 273 (44.8) 73 (47.4) Marital Status 154 Single 121 (19.8) 28 (18.2) Other 277 (45.4) 65 (42.5) >HS (r) Education 153 332 (54.6) 88 (57.5) HS or less 331 (54.3) 67 (48.9) Paraplegia (r) Injury Level 137 279 (45.7) 70 (51.1) Tetraplegia 161 (26.4) 40 (26.5) Employed (r) 257 (42.1) 41 (27.2) Employment* 151 Unemployed† 192 (31.5) 70 (46.4) Other† 216 (35.4) 40 (28.0) Private (r) 218 (35.7) 51 (35.7) Medicare Payer 143 130 (21.3) 34 (23.8) Medicaid 46 (7.5) 18 (12.6) Other 256 (42.0) 38 (35.5) None (r) 244 (40.0) 47 (43.9) Breakdowns 107 ≥1 BD, No IC 110 (18.0) 22 (20.6) ≥1 BD, ≥1 IC Notes. The Missing N column denotes how many participants were in the “missing” group for each dependent variable. These were subjects who had data for the respective variable but were missing data for one or more other variables. A (r) denotes the reference category for each variable. LWP = living with partner. BD = wheelchair breakdown. IC = immediate consequence (i.e. stranded; injured; missed medical appointments, school, or work). * Significant chi-square test, Χ2(2, N = 761) = 14.722, p<.01, V=.14 † Significant post hoc analysis with Bonferonni correction (p<.05)
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 2. Outcome variable descriptive statistics for complete and missing datasets, and between-groups analyses. Nominal variables are presented as frequencies (percentages) and continuous variables as means (standard deviations). Complete* Missing Missing N SPHS 2.85 (1.1) 2.93 (1.2) 148 Pain 4.26 (2.8) 4.62 (3.0) 147 Pressure Injury None (r) 418 (68.2) 52 (35.6) 146 At least one 195 (31.8) 94 (64.4) Rehospitalization† None (r)‡ 386 (63.3) 37 (47.4) Unrelated to PU‡ 174 (28.5) 37 (47.4) 78 Related to PU 50 (8.2) 4 (5.1) Notes. The Missing N column denotes how many participants were in the “missing” group for each dependent variable. These were subjects who had data for the respective variable but were missing data for one or more other variables. The (r) indicates the reference category for each variable. SPHS = self-perceived health status. * N=610 † Significant chi-square test, Χ2(2, N = 688) =11.718, p<.01, V=.13 ‡ Significant post hoc analysis with Bonferonni correction (p<.05)
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 3. General-linear model statistics of the association between pain and wheelchair breakdowns, and demographic factors. Full statistics are presented only for significant predictors. 95% CI B B B SE Lower Upper p R2 Intercept 2.01 0.22 1.58 2.44 <.001 .124 Breakdowns ≥1 BD, No IC .443 ≥1 BD, ≥1 IC† 0.83 0.31 0.22 1.44 .008 .012 Race Minority .082 Sex Female .073 Marital Status Single .816 Other .965 Education ≤HS Diploma* 0.61 0.24 0.14 1.08 .012 .011 Injury Level Tetraplegia .111 Employment Unemployed .162 Other .635 Payer Medicare† 0.75 0.28 0.19 1.30 .008 .012 Medicaid* 0.78 0.36 0.10 1.46 .025 .008 Other* 0.92 0.45 0.05 1.80 .039 .007 Age .125 Notes. Full model: F(14, 595)=4.247, p<.001, R2=.091, Adj R2 =.069. BD = wheelchair breakdown. IC = immediate consequence (i.e. stranded; injured; missed medical appointments, school, or work). * p<.05 † p<.01
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 4. General-linear model statistics of the association between self-perceived health and wheelchair breakdowns. Full statistics are presented only for significant predictors. 95% CI B B B SE Lower Upper p R2 Intercept 2.01 0.22 1.58 2.44 <.001 .124 Breakdowns ≥1 BD, No IC .762 ≥1 BD, ≥1 IC* 0.26 0.12 0.02 0.50 .032 .009 Race Minority .511 Sex Female .059 Marital Status Single .809 Other .901 Education ≤HS Diploma .057 Injury Level Tetraplegia .426 Employment Unemployed* 0.29 0.12 0.06 0.52 .014 .010 Other .148 Payer Medicare .097 Medicaid* 0.35 0.13 0.08 0.61 .010 .011 Other .908 Age* 0.01 0.01 0.00 0.02 .005 .012 2 2 Notes. Full model: F(14, 595)=3.868, p<.001, R =.083, Adjusted R = .062. BD = wheelchair breakdown. IC = immediate consequence (i.e. stranded; injured; missed medical appointments, school, or work). * p<.05
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 5. Logistic regression model statistics depicting the odds of self-reported pressure injuries within the past twelve months associated with wheelchair breakdowns, and demographic factors. 95% OR B B SE Wald Χ2 df p OR Lower Upper Breakdowns* 6.37 2 0.04 ≥1 BD, No IC -0.05 0.202 0.06 1 0.81 0.97 0.64 1.41 ≥1 BD, ≥1 IC* 0.55 0.25 4.87 1 0.03 1.73 1.06 2.83 Race 0.365 1 0.55 Sex 1.57 1 0.21 Marital Status 1.95 2 0.38 Education 1.81 1 0.18 Employment 0.378 2 0.83 Injury Level 0.621 1 0.62 Insurance 2.68 3 0.44 Age 1.47 1 .225 Notes. Model statistics for insignificant independent variables were omitted. BD = wheelchair breakdown. IC = immediate consequence (i.e. stranded; injured; missed medical appointments, school, or work). *p<.05
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 6. Logistic regression model statistics depicting the odds of rehospitalization unrelated to (top block) and related to skin breakdown (bottom block) within the past twelve months, associated with wheelchair breakdowns and demographic factors. 95% OR B B SE Wald Χ2* p OR Lower Upper Breakdowns ≥1 BD, No IC .123 ≥1 BD, ≥1 IC* .593 .265 5.015 .025 1.810 1.077 3.042 Race Minority .225 Sex Female .307 Marital Status Single .245 Other .366 Education ≤High School .337 Employment Unemployed .284 Other† .862 .282 9.352 .002 2.367 1.363 4.113 Injury Level Tetraplegia* .506 .197 6.610 .010 1.658 1.128 2.437 Insurance Other .347 Medicaid‡ .504 .293 2.965 .085 1.656 .933 2.939 Medicare .499 Age .425 Breakdowns ≥1 BD, No IC .119 ≥1 BD, ≥1 IC‡ .775 .434 3.195 .074 2.171 .928 5.078 Race Minority .584 Sex Female .792 Marital Status Single .776 Other .705 Education ≤High School .925 Employment Unemployed .259 Other .518 Injury Level Tetraplegia .254 Insurance Other .911 Medicaid .874 Medicare .465 Age .327 Notes. Model statistics for insignificant independent variables were omitted. BD = wheelchair breakdown. IC = immediate consequence (i.e. stranded; injured; missed medical appointments, school, or work). *p<.05 † p<.01 ‡ p<.10
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT