Factors Influencing Incidence of Wheelchair Repairs and Consequences Among Individuals with Spinal Cord Injury

Abstract

Objective:

Investigate frequency and consequences of wheelchair repairs, looking at the relationship to usage, components, out-of-pocket costs, number of days impacting the user, and factors associated with the need for repairs or consequences.

Design:

Survey, cross-sectional.

Setting:

Nine spinal cord injury (SCI) Model Systems Centers.

Participants:

533 wheelchair users with SCI.

Interventions:

N/A.

Main Outcome Measures:

Cost and incidence of wheelchair repairs and consequences and wheelchair usage within the past 6 months.

Results:

310 participants (56%) reported repairs of which 127 (42%) experienced at least one adverse consequence lasting a median [IQR] of 5 [2,17.3] days. Repair rates were highest for the seating system, electronics, and the tires. Participants were most often stranded at home or forced to use a backup chair. Median out-of-pocket costs were $150 [50, 620]. Active users, based on type of mobility and terrain, experienced more repairs and consequences than less active users. Repairs were more common among those who were Black (odds ratio: 2.42) or power wheelchair (PWC) users (odds ratio:1.84) while consequences were more common among those who were Black (OR:2.27), PWC (OR:2.08) or power assist (OR:2.76) users, or those who had public insurance (OR:1.70).

Conclusions:

Wheelchair repairs continue to impact over 50% of wheelchair users with significant financial and personal cost. High repair rates limited participation inside and outside the home. Consequences lasted >2 weeks for many and may be minimized by a working backup chair. Disparities exist based on participant and wheelchair factors; repairs and adverse consequences appear to hit those most vulnerable with the least financial resources. Costs may be a barrier to repair completion for some individuals. This ongoing problem of high repair rates and their associated impact require action such as higher standards, access to quicker service and better training of user on wheelchair maintenance and repair.

The wheelchair is one of the most enabling assistive technologies, making not only mobility possible, but also providing the freedom to change position, manage pressure, complete activities of daily living and participate in the community. Unfortunately, these benefits are not realized to their full potential when the wheelchair breaks down and studies show 45–88% of wheelchair users experience a failure.^1–5 While preventative maintenance has been demonstrated to reduce adverse consequences^6,7, there are no clinical or industry standards for expected wheelchair maintenance. The need for wheelchair repairs has widespread effects that can have a devastating impact on health, quality of life and participation for wheelchair users. Wheelchair breakdowns can result in injury, missing medical appointments, school or work, or being stranded.^2–5,7,8 People experiencing breakdowns that resulted in any of these consequences were 1.9 times more likely to be re-hospitalized, 1.7 times more likely to experience a pressure injury, and more likely to report worse self-perceived health status and pain.⁹

A recent study in Ireland surveyed wheelchair users and stakeholders about their impressions of the wheelchair seating provision process.¹⁰ Worry about repair services was one of the four primary themes identified by respondents. Participants reported that “underlying fear” of breakdown was a barrier to independence and something that affected their plans outside of their homes. Less than 50% of participants received any type of training to maintain their wheelchair. Attending wheelchair follow-up appointments to address repair issues was not common and of the 21% who attended these appointments, 60% were not satisfied with the service secondary to long wait times for repairs or replacement components and a lack of mobility while they were waiting for the repair to be completed.

While previous studies demonstrate high rates of repairs and adverse consequences, questions remain related to temporal aspects of repairs (e.g. how long did they take to complete), the role of wheelchair usage and age, costs of repairs and the relationship between specific types of repairs and resulting consequences. The primary objectives of this study were to determine current incidence of repair and consequences among individuals with spinal cord injury (SCI) in the United States across different wheelchair components, the number of days consequences were experienced, and participant and wheelchair characteristics associated with the need for repairs or consequences.

Methods

Sample

Staff at 9 SCIMS sites (Midwest Regional Spinal Cord Injury Care System, Mount Sinai Hospital Spinal Cord Injury Model System, Northeast Ohio Regional Spinal Cord Injury System, Northern New Jersey Spinal Cord Injury System, Ohio Regional Spinal Cord Injury Model System, Spaulding New England Regional Spinal Cord Injury Center, South Florida Spinal Cord Injury Model System, University of Alabama at Birmingham Spinal Cord Injury Model System, University of Pittsburgh Model Center on Spinal Cord Injury, and Texas Model Spinal Cord Injury System) recruited a convenience sample between May 2017 and June 2020. Institutional Review Boards approved the study at each site.

Recruitment strategies included the National Spinal Cord Injury Database (NSCID), local registries, flyers, and identification by clinical staff. Eligibility criteria included age ≥16 years old and a neurologic impairment caused by a non-progressive SCI that occurred ≥1 year before the study. Those excluded due to progressive SCI would include a SCI secondary to progressive disease with no traumatic event, herniated disc or transverse myelitis. Data were collected by in-person or by phone interview as part of a larger study evaluating equity and quality and assistive technology (n=439 NSCID, n=183 other recruitment strategies), with this analysis focused on full time wheelchair users (≥40 hours/week).

Survey

Participants reported their age, sex, years since injury, level of injury, race, ethnicity, marital status, education level, occupation, veteran status and if they received services at the VA, funding source, and income. They also reported the age of their wheelchair, type (manual, power, power assist), seating functions (tilt, recline, standing, seat elevator, elevating legs, none), usage (household distances, community distances, outdoor terrain), the frequency (daily, monthly, yearly, never) they encountered different terrains (soft surfaces, curbs ≥2 inches, steep inclines), and if they had a backup wheelchair.

Participants reported information about repairs in response to the following question, “In the past 6 months, how many times did you have (insert component) that needed to be repaired or replaced?” This question was asked for 9 manual wheelchair (MWC) components and 12 power wheelchair (PWC) components (Table 1). If respondents reported repairs or replacements, they were then asked about resulting consequences and the associated out of pocket costs. If participants reported any of the following consequences they were asked for how many days they experienced the consequence: stranded outside of the home, stranded in the home, stuck in bed, injured, missed work or school, missed medical appointment, used a backup wheelchair, or missed other social events. Repair questions can be found in the Appendix.

Table 1:

Categories of Repairs

Name of Component	Description
All Wheelchair Types
Tire	Worn out or punctured tires or tubes (wheels or casters)
Brakes	Brakes that wouldn’t lock/unlock or were loose
Caster	Loose or broken caster assembly (includes caster fork or stem)
Bearing	Worn out or broken bearing (wheels or casters)
Frame	Broken or cracked wheelchair frame
Positioning Support	Loose or worn out positioning support (includes foot supports, leg supports, arm supports, head supports, lateral supports, trunk supports, and/or clothing guards)
Seating Component	Worn out or broken seating components (includes sagging seat and back upholstery and cracks in the seat base or backrest or hardware)
Cushion	Worn out or broken cushion (includes torn upholstery, compromised foam, leaking air or gel)
Manual Wheelchair Only
Wheel	Loose or broken wheel component (including spokes, handrims or axle)
Power Wheelchair Only
Seat Function	Broken power seating functions (includes seat elevation, elevating leg rests, tilt, recline, and standing)
Suspension	Broken suspension elements (includes dampers and springs)
Controller	Loose or broken controller box (includes joystick)
Battery	Battery that wouldn’t hold a charge

Data Analysis

The following variables were recoded: level of injury (paraplegia, tetraplegia), marital status (married, not married), occupation (working/student/on-the-job training/sheltered workshop, homemaker/retired/unemployed/disability or medical leave), race and ethnicity (white/non-Hispanic, white/Hispanic, Black), marital status (married, not married), education level (high school diploma or less, 2–4 years of college, graduate degree), and wheelchair age (0–1 years, 2–5 years, >5 years), funding source (private/workers compensation, Medicare/Medicaid, and other). Other funding source comprised less than 10% of the sample and included the Veteran’s Administration, other governed, no pay, and self-pay. Missing demographic and wheelchair characteristic data are reported in Table 2. Repairs and consequences were dichotomized (needed/not needed) for each participant and component.

Table 2:

Participant and Wheelchair Characteristics (n=533)

Participant Characteristic	n (%)	Missing	Difference Based on Repairs (Y/N)	Difference Based on Consequences (Y/N)
		n (%)	X2, p-value	X2, p-value
Male	405 (76%)	2 (1%)	2.619, p=0.106	0.677, p=0.401
Paraplegia	271 (51%)	14 (3%)	0.359, p=0.549	0.056, p=0.813
Race/Ethnicity		18 (3%)	7.422, p=0.024	5.747, p=0.057
Non-Hispanic white	211 (40%)
Hispanic white	250 (47%)
Non-Hispanic Black	54 (10%)
Married	149 (28%)	4 (1%)	0.025, p=0.874	1.166, p=0.280
Education Level		4 (1%)	4.909, p=0.086	0.123, p=0.940
High school or less	302 (57%)
4 years of college or less	177 (33%)
Graduate degree	50 (9%)
Working/student	145 (27%)	58 (11%)	4.477, p=0.034	0.019, p=0.891
Veteran			1.693, p=0.429	2.520, p=0.284
Yes, received services at VA	23 (4%)
Yes, did not receive services at VA	21 (4%)
No	485 (91%)
Funding Source		1 (1%)	6.690, p=0.035	7.092, p=0.029
Private	152 (29%)
Medicare	199 (37%)
Medicaid	128 (24%)
Other	49 (9%)
Income		74 (14%)	2.410, p=0.492	4.152, p=0.245
<$25,000	193 (36%)
$25,000–$49,999	102 (19%)
$50,000–$74,999	68 (13%)
>$75,000	96 (18%)
Income meets household needs		94 (18%)	2.941, p=0.230	0.362, p=0.834
Not well	59 (11%)
Poorly, but I get by	178 (34%)
Pretty well	202 (38%)
Wheelchair Characteristics	n (%)	n (%)	X2 (p-value)	X2 (p-value)
Type of Wheelchair		0	3.785, p=0.050	5.478, p=0.019
Manual	288 (54%)
Power	221 (41%)
Power Assist	24 (5%)
Seating Functions	285 (53%)	0	1.374, p=0.241	2.598, p=0.107
Backup wheelchair	343 (64%)	8 (2%)	0.234, p=0.629	0.043, p=0.835

Note: Bolded values indicate a significant difference with p≤0.05 and values included in the regression models

Descriptive statistics were used to describe the incidence of those experiencing at least one consequence and what type of consequences were experienced. The median and interquartile range (IQR) of days consequences were experienced were determined overall as well as by repair type. Chi-square was used to evaluate differences in repair rates and consequences based on wheelchair usage. When differences were indicated based on surfaces traversed (soft, curbs, inclines), Mann-Whitney U test was used to make comparisons between those who traversed the surfaces (daily, weekly, monthly, and yearly) against those who never traversed the surfaces. Chi-square tests were used to determine if there were differences in those who experienced repairs and consequences based on participant and wheelchair characteristics. Significant associations were used as predictors in logistic regression models to identify factors associated with the need for repair or consequences. The level of significance was set a priori at α=0.05. All statistical analysis was performed using SPSS 27.0.^a

Results

Participants

A total of 533 participants enrolled. Participants were a median [IQR] of 46 [34,57.5] years of age with 13 [5,24] years since injury and the age of their wheelchairs was 3 [1,5] years. 14% of users reported their wheelchair was 6–10 years old and 5% reported their wheelchair was over 10 years old. Additional participant and wheelchair characteristics can be found in Table 2. There were no differences in incidence of repairs on consequences based on participant age. Compared to those with wheelchairs ≤1 year old, individuals with wheelchairs 2–5 years old or >5 years old were more likely to experience repairs (43% vs 61% and 62%, X²(2)=14.504, p=0.001); no differences were found for consequences.

Repairs

Three hundred and one (56%) participants reported the need for repairs. Among those requiring repairs, 212 (34%) participants reported no out-of-pocket costs while 85 (28%) participants reported median [IQR] out-of-pocket costs of $150 [50, 620] which ranged from $6 to $4,000. In order of incidence, the six most common categories of wheelchair repair were the tires, controller (PWC only), positioning supports, battery (PWC only), cushion, and seat function (PWC only) (Figure 1).

Figure 1:

Incidence of repairs with and without consequences based on wheelchair component

Consequences

Of those requiring repairs, 127 (42%) experienced at least one adverse consequence with the median [IQR] number of consequences per person of 2 [1,4] and median [IQR] total days of consequences of 5 [2,17.3] (range: 0–180 days) in the last six months. For those requiring repairs, 19–43% of participants across categories of wheelchair components experienced adverse consequences (Table 3). The percentage of participants experiencing adverse consequences exceeded 25% for all wheelchair components specific to PWCs (seat functions, suspension, controller, battery) as well as tires, casters, bearings, frame, and seating components (Table 3). Repairs to wheelchair components specific to PWCs more often resulted in individuals stranded at home or missing a medical appointment compared to other categories. Across consequences, the highest incidence across participants was forced to use backup chair (36%) and stranded at home (27%). The least frequently reported consequences were being injured (5%) and missing work or school (7%).

Table 3:

Consequences by wheelchair component among those reporting repairs

Wheelchair Component		≥1 Consequence	Days Experienced Consequence(s)	Type of Consequence
				Stranded Outside Home	Stranded Inside Home	Stuck in Bed	Injured	Missed Work/School	Missed Medical Appt	Forced to Use Backup chair	Missed Important Social Event
		n (%)	Median [IQR]	n (%)	n (%)	n (%)	n (%)	n (%)	n (%)	n (%)	n (%)
MWC and PWC (n= 533)	Tire (n=151)	55 (35%)	4 [1,9]	8 (5%)	20 (13%)	9 (6%)	1 (1%)	4 (3%)	7 (5%)	29 (19%)	17 (11%)
	Brakes (n=64)	12 (19%)	2.5 [2,5.5]	3 (5%)	1 (2%)	1 (2%)	4 6%)	0	2 (3%)	6 (9%)	3 (5%)
	Caster (n=65)	28 (43%)	2 [2,7]	3 (5%)	9 (14%)	2 (3%)	1 (2%)	5 (8%)	4 (6%)	13 (20%)	6 (9%)
	Bearing (n=57)	17 (30%)	4 [2,10]	3 (5%)	6 (11%)	2 (4%)	0	3 (5%)	2 (4%)	10 (18%)	1 (2%)
	Frame (n=18)	5 (28%)	7 [2, 67.5]	1 (6%)	3 (17%)	0	0	1 (6%)	1 (6%)	2 (11%)	2 (11%)
	Positioning Support (n=91)	21 (23%)	6 [2.5,16.5]	0	11 (12%)	5 (5%)	2 (2%)	3 (3%)	7 (8%)	7 (8%)	4 (4%)
	Seating Component (n=67)	17 (25%)	4 [1,13]	1 (1%)	5 (7%)	2 (3%)	3 (4%)	1 (1%)	2 (3%)	6 (9%)	2 (3%)
	Cushion (n=75)	16 (21%)	5 [2,7.75]	0	5 (7%)	4 (5%)	2 (3%)	3 (4%)	3 (4%)	7 (9%)	3 (4%)
MWC (n= 288)	Wheel (n=29)	6 (21%)	4.5 [2.75,14]	1 (3%)	2 (7%)	1 (3%)	0	0	1 (3%)	4 (14%)	0
PWC (n= 245)	Seat Function (n=31)	11 (35%)	5 [1,20]	1 (3%)	5 (16%)	3 (10%)	0	0	1 (3%)	6 (19%)	3 (10%)
	Suspension (n=6)	2 (33%)	7 [2, N/A]	0	2 (33%)	2 (33%)	0	0	1 (17 %)	0	1 (17%)
	Controller (n=42)	16 (38%)	7 [1.5, 20]	4 (10%)	9 (21%)	4 (10%)	2 (5%)	0	5 (12 %)	8 (19%)	3 (7%)
	Battery (n=39)	15 (38%)	4 [1,45]	3 (8%)	4 (10%)	1 (3%)	0	1 (3%)	0	10 (26%)	2 (5%)
All participants (n=301)		127 (42%)	5 [2,17.3]	28 (9%)	82 (27%)	36 (12%)	15 (5%)	21 (7%)	36 (12 %)	108 (36%)	47 (16%)

Note: all percentages are reported as the % is of participants requiring a repair for each component, except for All participants (bottom row) that is % of total participants reporting repairs (n=301).

Wheelchair Usage

Those who traversed community and outdoor surfaces were more likely to experience both repairs (X²(1)=7.051, p=0.007 and X²(1)=6.465, p=0.007) and consequences (X²(1)=3.222, p=0.048 and X²(1)=3.553, p=0.037), while no significant differences were found based on traveling household distances. There was a difference in incidence of repairs based on frequency of traversing soft terrain (X²(4)=9.721, p=0.045), the incidence of consequences based on frequency of traversing curbs (X²(4)=4.755, p=0.060 and X²(4)=5.643, p=0.046) and the incidence of repairs and consequences based on frequency of traversing steep inclines (X²(4)=8.266, p=0.012 and X²(4)=7.145, p=0.018). Compared to those who never traversed the terrain, those who traversed soft terrains at least monthly experienced more repairs (p=0.005 to 0.014), those who traversed curbs daily experienced more consequences (p=0.028), and those who traversed steep inclines daily or monthly experienced more repairs and consequences (p=0.007 to 0.021). (Figure 2).

Figure 2:

Incidence of repairs and consequences based on wheelchair usage

*indicates significant difference in repairs (p<0.05).

† indicates significant difference in consequences (p<0.05).

Factors Predicting Incidence of Repairs and Consequences

Logistic regression was used to predict repairs or consequences with 423 and 427 cases included, respectively (Tables 4 and 5). The Hosmer-Lemshow test showed both models were correctly specified (p=0.388 and 0.951, respectively). Repairs were more common among individuals who had a chair that was 2–5 years old (OR: 1.86) or >5 years old (OR: 2.10) compared to those who had a chair that was 1 year or less old, who were Black compared to those who were white (OR: 2.42), and who were PWC users (OR: 1.84). Consequences were more common among those who used a PWC (OR: 2.08) or power assist (2.76) compared to those who used a MWC, those who were Black (OR: 2.27) compared to non-Hispanic white, and those with public insurance (OR:1.70) compared to those with commercial insurance.

Table 4:

Logistic regression model statistics depicting the odds of requiring a repair associated with subject and wheelchair factors

	B	B SE	Wald X2	df	Sig	OR	95% OR
							Lower	Upper
Wheelchair Age (0= 1 year or less)			7.592	2	0.022 *
2–5 years	0.62	0.247	6.301	1	0.012 *	1.86	1.15	3.02
>5 years	0.74	0.32	5.336	1	0.021 *	2.10	1.12	3.92
Race/Ethnicity (0= Non-Hispanic White)			5.97	2	0.050 *
Hispanic White	0.031	0.224	0.019	1	0.890	1.03	0.67	1.60
Non-Hispanic Black	0.885	0.38	5.422	1	0.020 *	2.42	1.15	5.10
Wheelchair Type (0=Manual Wheelchair)			8.21	2	0.016 *
Power Wheelchair	0.608	0.222	7.527	1	0.006 *	1.84	1.19	2.84
Power Assist Wheelchair	0.657	0.508	1.673	1	0.196	1.93	0.71	5.22

Note: Model Statistics for insignificant independent variables were omitted. Model X²(13)=31.089, p=0.003.

Abbreviations: B, unstandardized coefficient; B SE, unstandardized coefficient standard error; Sig, significance

^*indicates p<0.05

Table 5:

Logistic regression model statistics depicting the odds of experiencing a consequence associated with subject and wheelchair factors

	B	B SE	Wald X2	df	Sig	OR	95% OR
							Lower	Upper
Wheelchair Type (0=Manual Wheelchair)			10.553	2	0.005*
Power Wheelchair	0.736	0.249	8.738	1	0.003*	2.09	1.28	3.40
Power Assist Wheelchair	1.016	0.507	4.008	1	0.045*	2.76	1.02	7.47
Race/Ethnicity (0= Non-Hispanic White)			4.661	2	0.097
Hispanic White	0.287	0.267	1.154	1	0.283	1.33	0.79	2.25
Non-Hispanic Black	0.818	0.379	4.646	1	0.031*	2.27	1.08	4.76
Funding (0=Private, 1=Public)	0.532	0.273	3.803	1	0.050 *	1.70	1.00	2.90

Note: Model Statistics for insignificant independent variables were omitted. Model X²=35.471, p<0.001

Abbreviations: B, unstandardized coefficient; B SE, unstandardized coefficient standard error; Sig, significance

^*indicates p<0.05

Discussion

Repairs

Consistent with reports from 2012–2016, the incidence of wheelchair repairs remains high among individuals with SCI with 56% of participants requiring at least one repair in a 6-month period.^3–5 Repair rates were highest among components with frequent use including the seating system, electronics, and the tires. These categories align with recommended preventive maintenance schedules that prioritize daily inspection of the electrical system and weekly inspection of the tires and cushion, compared to other components for which a monthly inspection is recommended.¹¹ Out-of-pocket costs were reported by approximately 25% of participants. It is possible that these costs may be prohibitive for some individuals and resulted in repairs that were needed but not completed.³

Durability testing according to the American National Standards Institute/Rehabilitation Engineering and Assistive Technology Society of North America (ANSI/RESNA) standards is designed to mimic 3–5 years of typical use for a wheelchair.¹² There are three classes of failures within these standards, all of which are likely represented in this sample.^12,13 Class I failures are repairs that can be completed without technical training such as tightening nuts or bolts or adding air to tires. Of the repair types included in this study, these could include loose caster assemblies or positioning components. Consumer education on these types of repairs could address both time to repair and consequences. Class II failures would be completed by a vendor/supplier or other skilled technician and may require ordering parts such as fixing a flat tire or replacing a broken postural support. Class III failures are structural failures that would immobilize the wheelchair such as a malfunctioning controller or a broken caster fork. Independent testing shows that not all wheelchairs meet these safety standards and manufacturers are not required to complete external testing to demonstrate meeting these standards.^14–20

Consequences

The incidence of adverse consequences for this cohort (42%) was higher than previous reports (21–30%).^3–5 Some of these discrepancies may be due to differences in what was classified as a consequence as rates are comparable to a recent report in a Veteran population¹ that included the same response options, which were not part of previous studies (stuck in bed, forced to use backup chair, missed important social event). Previous reports asked participants about being stranded without clarity on whether that was inside or outside of their home. Reported consequences align with user concerns related to wheelchair repairs and maintenance in other studies including fear of “being trapped” where a breakdown occurs, long wait times, and a lack of mobility resulting from a non-operational wheelchair.¹⁰

The majority of individuals experienced between 2 and 17 days of consequences. The number of days was highest for repairs to the frame, battery, seating functions, and controller. These are all class III failures. While only 2 participants reported injuries among these categories, this is likely because users were unable to use their chairs due to safety concerns. Instead, they were likely stranded or forced to use a backup chair as the incidence of consequences in those categories is high for these components. Despite the high percentage of users who had backup chairs (64%), there were still a large number of participants who were stranded in their home. This finding reflects a backup wheelchair that does not meet the user’s seating system needs or it may be in need of repair. An individual’s backup wheelchair is generally the one they used prior to obtaining their current wheelchair; it is usually several years old, in poor working condition, broken, or no longer optimized to the user’s the needs. Most insurers require the cost of repairs to exceed 50% of the cost of a new wheelchair before they will provide a new wheelchair and will not cover repairs or maintenance to a chair that is not the user’s primary wheelchair; as such, while someone may have a backup wheelchair, it may not be operational. The exception to this is individuals who receive their wheelchairs through the Veterans Affairs Healthcare System, which seeks to provide and service quality backup chairs.²¹ Prolonged periods of being stranded at home, potentially in bed, can lead to increased risk of pressure injury development and deconditioning.^22,23 Given the incidence of repairs and absent an improvement in wheelchair quality, back up wheelchairs and their maintenance should be covered by funding sources.

Wheelchair Usage

Previous studies have measured wheelchair usage in terms of hours per day in a chair. Results from this study are novel in reporting the type of usage (indoor, community, outdoor terrain) and the frequency of encountering different types of terrain. It is not surprising that users who were active outside of their home, traveling community distances or encountering outdoor terrain, experienced not only more repairs, but also experienced a greater impact of repairs with more consequences. The frequency of encountering different types of terrain influenced repairs and consequences differently. These differences may be related to different demands on the wheelchair (e.g. soft surfaces and steep inclines that may place more demands on the wheelchair) or different levels of wheelchair skills (more active users encountering obstacles more often). Greater participation is linked to higher quality of life among individuals with SCI.²⁴ Unfortunately, more active users saw not only more repairs, but also experienced more consequences. These types of activity limitations and restrictions to participation can have not only deleterious effects on quality on life, but also life satisfaction, psychological well-being and employment. ^25–32

Participant and Wheelchair Factors

When controlling for other factors, race and ethnicity and type of wheelchair were predictors of both repairs and consequences. Racial and ethnic disparities are consistent with previous reports of wheelchair failures, and differences in wheelchair provision, health quality of life, experiences of discrimination in health care and lower health literacy for minorities with SCI.^4,33–35 These differences were not found in a recent study of Veteran wheelchair users with SCI which suggests the access to VA support and funding may mitigate disparities.¹ PWCs were twice as likely to require a repair and users were 2–2.75 times more likely to experience a consequence. This finding is consistent with previous reports as PWCs are more complex in terms of seating systems and electronics, and a breakdown in either system will likely limit mobility.^1,3,4 While previous studies have not found a relationship between incidence of repairs and wheelchair age³⁶, this study found those with newer chairs (≤1 year old) were less likely to require a repair. Repair rates were similarly high for those with wheelchairs 2–5 years old and >5 years old; this finding suggests that preventive maintenance and repairs are important early in the lifespan of a chair. Lastly, funding source was the final factor associated with consequences. which may be tied to Public insurance as well as minority status and lower income have been tied closely to socioeconomic status and related to disparities in assistive technology provision.³⁴ Socioeconomic status may impact not only one’s ability to pay out-of-pocket costs for a repair, but also challenges with accommodating a breakdown such as missing work or affording to maintain a backup wheelchair.

This publication marks over 15 years’ worth of data, dating back to 2004, demonstrating high and increasing rates of wheelchair breakdown^1–5,36 and needs to serve as a call to action. These breakdowns necessitate repairs, not all of which are completed. The repairs take time to complete, often requiring a wheelchair user to wait for a supplier to come out and assess the wheelchair, obtain pre-authorization to complete the repairs, and complete the repair. Additional time may be needed for the supplier to acquire the needed replacement part(s). Each of these steps draws out the time during the repair process that the user must either continue to use their wheelchair with broken components, remain stranded without a means for mobility, or use a backup or loaner wheelchair that may not meet their seating and positioning needs. There are often out-of-pockets for users to ensure these repairs are completed. This broken process coalesces in over 40% of users experiencing adverse consequences resulting from wheelchair breakdown. Wheelchair users with SCI need a dependable means of mobility, mandates for external strength and durability testing, coverage policies that include maintenance of the wheelchair through its life span and support for functional backup wheelchairs, and maintenance training of not only wheelchair users but also clinicians and suppliers.

Study Limitations

This study relied on recall of events during the previous 6 months. There is a risk of recall bias, though the recall window was relatively short, and participants were asked about the most recent 6 months. A prospective study could address these limitations and provide insight into cause-effect relationships. While we collected the days users experienced consequences, we did not solicit the number of days to repairs. This temporal data could highlight further areas of intervention for improving the service delivery process. Wheelchair usage was self-reported; objective measures of daily mobility would provide greater understanding of the relationship between usage and wheelchair repairs. While 9 sites were involved in data collection representing eastern, southern, central, and midwestern locations there may be additional variability secondary to geography that was not captured with this sample. Future studies may also benefit from investigating the relationship between preventative maintenance and repairs. Finally, data were collected from SCI Model Systems facilities which are centers of excellence in research and clinical care. Thus, the incidence of repairs and consequences may underestimate that of the general population.

Conclusions

Wheelchair repair rates are high with significant impacts on participation inside and outside the home. Consequences were not necessarily short lived with the median upper quartile of total days of consequences exceeding two weeks. A working backup chair may minimize adverse consequences. Repair cost may be a barrier to repair completion for individuals with limited resources. Groups at risk for higher rates of repair and adverse consequence include active users, PWC users, those with a wheelchair >1-year-old, and individuals who are Black. Preventing wheelchair breakdowns or optimizing service to minimize wait time for repairs may reduce secondary disability among wheelchair users.

Suppliers

1. SPSS, 233 S Wacker Dr, 11^th Fl, Chicago, IL 60606

Wheelchair Repair/Replacement Survey Questions

- Did your wheelchair require repair or replacement of parts in the past 6 months?

• No (Skip repair questions)

• Yes

• I do not remember (Skip repair questions)

If yes, then proceed with the following questions:

- What were your out-of-pocket costs (i.e. money you paid) for repairs in the past 6 months?

• Valid currency amount:_____________________________________________

These are screening questions to determine which repair categories should be probed. Check all that apply:

- Did you require repair or replacement to:

All wheelchairs

• Worn out or punctured tires or tubes (wheels or casters)

• Brakes that wouldn’t lock/unlock or were loose

• Loose or broken caster assembly (includes caster fork or stem)

• Worn out or broken bearing (wheels or casters)

• Broken or cracked wheelchair frame

• Loose or worn out positioning support (includes foot supports, leg supports, arm supports,

• head supports, lateral supports, trunk supports and/or clothing guards)

• Worn out or broken seating components (includes sagging seat and back upholstery, a worn

• out seat, cracks in the seat base or backrest or hardware)

• Worn out or broken cushion (includes torn upholstery, compromised foam, leaking air or gel)

Manual wheelchair & Power Assist only:

• Loose or broken wheel component (including spokes, handrims or axle)

• Not applicable, I do not use a manual wheelchair

Power wheelchair & Power Assist only:

• Broken power seat functions (includes seat elevation, elevating leg rests, tilt, recline and

• standing)

• Broken suspension elements (includes dampers and springs)

• Loose or broken controller box (includes joystick)

• Battery that wouldn’t hold a charge

• Not applicable, I do not use a power wheelchair

The following completed for each screening repair category with ‘yes’ response:

- In the past 6 months, how many times did you have [insert repair category] that needed to be repaired or replaced?

• ______________________ (enter any valid number)

- How many times did you need the same repair completed?

• ______________________ (enter any valid number)

- Did you experience any of the following consequences as a result? If yes, for how many days did you experience the consequence?

• Stranded outside of the home

  ▪ Days:_____________________

• Stranded in the home

  ▪ Days:_____________________

• Stuck in bed

  ▪ Days:_____________________

• Injured

  ▪ Days:_____________________

• Missed work or school

  ▪ Days:_____________________

• Missed medical appointment

  ▪ Days:_____________________

• Forced to use backup wheelchair

  ▪ Days:_____________________

• Missed other important social events

  ▪ Days:_____________________

• None of the listed consequences occurred

• Unknown

• Refused

- Was a repair(s) completed?

• No repair(s) completed

• Some, but not all of the repairs were completed

• All of the repairs were completed

• I do not remember

• Not applicable

If yes or Some:

- Who completed the repair?

• Myself

• A family member

• My vendor

• Other

If My Vendor:

- How many days after you reported the repair to your vendor was the repair completed?

• ____________________________________________

• Unknown/ I do not remember

If no repairs or some repairs:

- Why were repairs not completed? (check all that apply)

• Repair was not attempted

• The vendor was contacted but did not complete the repair

• The vendor serviced my chair, but the repairs were not done correctly, additional visits necessary

• Myself or a family member attempted but could not complete it

• Insurance would not provide coverage for the repair

• I could not afford/did not want to pay out of pocket for the repair(s)