A High Prevalence of Manual Wheelchair Rear-wheel Misalignment Could Be Leading to Increased Risk of Repetitive Strain Injuries

Abstract

Purpose:

To determine the prevalence and severity of manual wheelchair rear wheel misalignment in community-dwelling manual wheelchair users and estimate the associated increases in rolling resistance and risk of repetitive strain injuries.

Materials and Methods:

Data was collected in an outpatient rehabilitation clinic, a university research laboratory, and at adaptive sporting events in the United States. Two hundred active, self-propelling manual wheelchair users were recruited. Angular misalignment (referred to as toe angle) while the wheelchair was loaded with the user, and the difference between the maximum and minimum toe angle (referred to as slop) with the wheelchair unloaded.

Results:

Average results for toe angle and slop (movement in the rear wheels) were 0.92 and 0.61 degrees, respectively. Using a lab-based testing method, we quantified the impact of increased rolling resistance forces due to misalignment in increased rolling resistance forces. Our results indicate that the average toe angle while under load and slop, without loading, measured in the community increase required propulsion force by 3.0 N. Combined toe angle and slop (i.e. the worst-case scenario) added increased propulsion force by 3.9 N.

Conclusions:

We found that rear-wheel misalignment was prevalent and severe enough that it may increase the risk for repetitive strain injuries and decrease participation. To mitigate this issue, future work should focus on reducing misalignment through improved maintenance interventions and increased manufacturing quality through more stringent standards.

Introduction

Worldwide there are approximately 75 million people with a disability that requires the use of a wheelchair with an estimated 3.2 million in the United States. There are 250,000 individuals with spinal cord injuries using manual wheelchairs (MWC) in the United States [1]. Manual wheelchair use is associated with an increased risk for repetitive strain injuries (RSI) in upper extremities (UE) including shoulders, wrist, and hands. Increased propulsion forces are associated with increased pain or RSI for manual wheelchair users (MWU) and reducing the forces are one of the goals of common interventions including adjusting the wheelchair setup and training users to propel with certain stroke-styles [1–3]. For MWUs, UE injuries, such as rotator cuff injuries, can pose significant hardship from reduced activity and challenges associated with activities of daily living, especially if surgical repair is required. Higher rolling resistance (RR) forces and high frequency use are risk factors for pain as well as potential RSIs as a result of increased propulsion forces or continued use of MWCs.

Factors that contribute to increased propulsion forces impact MWU’s propulsion and have been an area of research and testing to improve understanding and clinical recommendations. RR is the primary resistive force to propulsion [4]. An illustration of the forces of propulsion and RR is shown in Figure 1. F_t is the tangential force from propelling and F_RR is the RR force that opposes propulsion. RR forces are primarily from inelastic hysteresis and energy loss from the tire, which induces the moment M_ZFRR and therefore incites the RR Force [4]. Many factors affect the energy dissipation of the wheel in the form of RR force, in terms of angle of the wheel (toe angles), load, tire type, tire pressure, and surfaces being propelled upon.

Figure 1.

RR Free body diagram where F_t is the tangential force, V is the angular velocity, and W is the load on the axle, F_RR is the RR force, M_ZFT is the moment due to the tangential force, M_ZFRR is the moment inducing the RR force

Testing shows that RR forces increase non-linearly with increased toe angle (measured in degrees while the manual wheelchair is under load) as shown in Figure 3, whereas weight and RR increase linearly [5,6]. This non-linear relationship underscores the importance of evaluating the prevalence and severity of toe angle and slop that contributes to increased toe angle in manual wheelchairs.

Figure 3.

RR Force versus Toe Angle for Rear Wheels

Misalignment of rear wheels includes toe angle, a lack of parallel alignment of the wheels, and slop (maximum toe angle minus minimum toe angle), which is tolerances in the axle, wheel, or hub connection measured when the device is unloaded. Toe angle manifests as either in or out, as illustrated in Figure 2. Slop and toe angle can combine together, which can significantly increase RR or cancel one another.

Figure 2.

Toe Angle Free Body Diagram, V is velocity, F_t is the tangential force, F_tx is the tangential component in the x direction, F_ty is the tangential component in the y direction

Based on our literature review the prevalence and severity of rear wheel misalignment (toe-in/toe-out) has not studied extensively. We identified only one conference abstract on the topic which measured misalignment of 50 wheelchairs in Kenya. The authors reported an average of 0.2 degrees of misalignment, with over 53% of the wheelchairs having more than 0.5 degrees of misalignment and 24% with greater than 1-degree of misalignment [5]. Other published studies reported RR force increases due to misalignment in lab-based tests [5,7,8], but did not confirm whether it was prevalent in the community.

This research study determines the prevalence and severity of toe-in/out misalignment and slop of rear-wheelchair wheels in the community. In order to accomplish this, we aimed to (1) develop a rigorous testing method to measure rear wheel misalignment, (2) measure misalignment of the rear wheels of wheelchairs used by community-dwelling and self-propelling individuals, and (3) determine which factors are predictors of toe angle or slop.

Methods

The research study was completed in two stages which include design and fabrication of a testing system (Stage 1) and collecting data on the rear-wheel misalignment of community-dwelling MWUs (Stage 2).

Stage 1 Device Development

We iteratively designed and developed a measurement system to measure the misalignment of the rear wheels of MWCs in the community. The conceptual design was developed through team discussions, the specifications outlined in Table 1 were developed. Specifications were determined in order to accommodate as many MWUs as possible, as well as to minimize unnecessary risk during testing.

Table 1.

Design Specifications

Need	Specification	Rationale
Portability	Overall size not bigger than 101 cm wide and 66 cm deep	It will be easily transported in most vehicles.
Toe Angle Measurement	Reach the front and rear of a 61 cm (24”) wheel	61 cm (24”) wheel is a very common size.
Slop Measurement	Apply pressure with the rear-wheels unloaded	Force needs to be applied to take up tolerances in order to measure the amount of tolerance, which cannot be performed with the rear-wheels under load.
Accuracy	1.5 mm or less	Equals less than a 0.25 degree of toe angle on a 61 cm (24-inch) wheel.
Adjustability	Height of measurement from 25 to 35 cm	Will be able to adjust to axle height of 20 to 71 cm (28”) wheels.
Device Accommodation	Fit devices from 35 to 81 cm in width	Very few MWUs would have a device under 35 cm and 81 cm is the width of a standard doorway.
User Accommodation	Lifting capacity of 136 Kg	Most standard MWC can accommodate 113 Kg with a 20% factor of safety.
Rigidity	Flex less than 0.63 cm across the system	Rigidity is important to mitigate flex which ensures accurate measurements.
Speed of Collection	Less than 10 minutes per person	Voluntary study where we did not want to detract from the events going on.

Stage 2 Community Data Collection

Recruitment

The study was submitted for institutional review board approval (IRB #PRO12080311) and found exempt since no personally identifiable information was being collected and there were minimal risks associated with the measurement procedure. Participants were recruited by the primary author via flyers and web-based postings and approached at adaptive sporting events such as wheelchair basketball practice, wheelchair sports competitions, fundraising events, and wheelchair maintenance events. The targeted participants for this research were MWU’s who self-propelled their wheelchairs and used them as their primary means of mobility for at least five days per week.

Data Collection

Descriptive information about the wheelchair and their usage habits were collected including device manufacturer, device model, age of the device, days used per week, hours used per day, tire manufacturer, tire model, tire maximum inflation pressure, tire diameter, and wheel type. Tire pressure was measured for the final 50 participants using a digital tire pressure gauge. Misalignment was collected through the following sequence of steps: (list the methods to collect and record the misalignment, including measuring under load, lifting participants off the ground, applying nominal forces inward and outward and measuring at under each load condition. Further details on the collection procedure can be found in Appendix A. Participants were also compensated with a $5 gift card for their time. Study inclusion criteria and data collection methods are summarized in Appendix A.

Data Analysis

Because there is no previous literature to reference for an effect size, we used a guideline of 15 participants per predictor. With nine predictors (Table A.1), a minimum of 135 participants is required. The study participant goal was set to 200 based on the original ten predictors and an overestimation to account for missing data. Additionally, the power analysis for regression with an alpha level of 0.05 and 80% power was calculated and showed that there is a 0.085 effect size with ten predictors and 200 samples. G*Power was used for this calculation [9]. The effect size is considered small to medium for a regression.

In order to conduct the analysis, the was grouped (by the primary author) in a logical approach that was guided by a clinician seating therapist, a rehab engineer and an industry expert based on the results of the study. Groupings were largely based on the diversity of the responses in order to ensure large enough subgroups for statistical analysis. Device make was grouped by manufacturer, and any device make with less than 9 responses per manufacturer were grouped as other. The Medicare K code was used to group devices by style for statistical analysis. A large number of custom rigid frame devices or former K0009 and a small number of ultralight devices were seen with K0005 codes. Tire type was broken into four categories including high-pressure pneumatic (689 kPa and over), low-pressure pneumatic (under 689 kPa), solid or airless inserts, and a group of unknowns. Tires were also broken down by manufacturer in a similar fashion to device make. Wheel type classified some models as performance due to stronger materials used for construction and enhanced engineering as compared to a standard lite-spoke wheel with additional categories of solid and other.

For statistical analysis, each of the nine predictors from Table A.1 was plotted against toe angle and slop, respectively. A non-parametric data analysis approach was employed due to a positively skewed, non-normal distribution in both toe angle and slop. Nine Kruskal Wallace tests were run using SPSS with each predictor and both toe angle and slop as dependent variables [10].

Results

Measurement Device

The research approach to develop a device was followed in accordance with the design specifications outlined in Table 1. The details of the development of the measurement device are included in Appendix B

Recruitment

200 participants were recruited over the course of approximately two months, mainly at adaptive sporting events, where sizable gatherings of MWUs enabled measurement of many devices before and after matches. Table 2 shows the breakdown of how the participants were recruited. Given the nature of community recruitment, some information was not obtainable based on the MWU’s knowledge or a lack of labeling on the MWC. The oversampling of participants was to ensure that this would not affect the data analysis.

Table 2.

Manual Wheelchair Community-based Study Recruitment

Recruitment Event	N	(%)
Adaptive Sports*	164	(82)
Assisted Living	14	(7.0)
Education & Training	15	(7.5)
Research Lab	5	(2.5)
Seating Clinic	2	(1.0)

Note. Total, N=200

^*Lost three participants

Data

Six wheelchairs manufacturers are represented in the data set (Table 3) as well as 45 unique models. Manufacturer A comprised almost half of all the devices measured.

Table 3.

Manual Wheelchair Community-based Study Manufacturers

Wheelchair Make	N	(%)
Manufacturer A	91	(45.5)
Manufacturer B	40	(20.0)
Manufacturer C	25	(12.5)
Manufacturer D	21	(10.5)
Manufacturer E	9	(4.5)
Other	14	(7.0)

Note. Total, N=200

The wheelchairs fall into two unique Medicare K codes, and an ‘other category’ which was used for statistical analysis (Table 4).

Table 4.

Manual Wheelchair Community-based Study K Codes

Frame Description	N	(%)
Wheelchair Frame
Custom Rigid (Former K0009)	171	(85.5)
K0005	13	(6.5)
Other/Unknown	16	(8.0)

Note. Total, N=200

The age of the wheelchairs spanned from <1 year old, to >4 years old with most of them being in the older category (Table 5).

Table 5.

Manual Wheelchair Community-based Study Age Data

Wheelchair Information	N	(%)
Approximately how many years have you had this wheelchair?
<1 year	27	(13.5)
1–2 years	42	(21.0)
3–4 years	50	(25.0)
>4 years	79	(39.5)
No response	2	(1.0)

Note. Total, N=200

Table 6 shows that 75 percent of the participants were active users, in their device over 10 hours a day and over 5 days a week. Additionally, over 96 percent of participants identified as self-propelling.

Table 6.

Manual Wheelchair Community-based Study Use Data

Wheelchair Use Frequency	N	(%)
Approximately how many hours per day do you use your wheelchair?
<5 hours	13	(6.5)
5–10 hours	34	(17.0)
10–15 hours	73	(36.5)
>15 hours	77	(38.5)
No response	3	(1.5)
Approximately how many days per week do you use your wheelchair?
1 day	3	(1.5)
2–3 days	3	(1.5)
4–5 days	11	(5.5)
> 5 days	181	(90.5)
No response	2	(1.0)

Note. Total, N=200

Roughly half of the participants were using a high-pressure tire as shown in Table 7. Additionally, two tire manufacturers account for 75 percent of the tires observed (Table 8).

Table 7.

Manual Wheelchair Community-based Study Tire Types

Tires by Type	N	(%)
High-pressure (689 kPa and over)	108	(54.0)
Low-pressure (under 689 kPa)	36	(18.0)
Solid or Airless Inserts	37	(18.5)
Unknown	19	(9.5)

Note. Total, N=200

Table 8.

Manual Wheelchair Community-based Study Tire Manufacturers

Tires by Manufacturer	N	(%)
Manufacturer A	79	(39.5)
Manufacturer B	70	(35.0)
Manufacturer C	15	(7.5)
Manufacturer D	8	(4.0)
Manufacturer E	6	(3.0)
Other	22	(11.0)

Note. Total, N=200

Overall, 61 cm (24-inch) wheels were the most common followed by 63.5 cm (25-inch), as seen in Table 9. Table 10 shows that a large portion of MWUs have performance wheels. Lite-spoke wheels were the second most common wheel type.

Table 9.

Manual Wheelchair Community-based Study Wheel Diameter

Wheel Measurements	N	(%)
Overall Wheel and Tire Diameter cm (in.)
56 (22)	7	(3.5)
61 (24)	117	(58.5)
63.5 (25)	53	(26.5)
66 (26)	23	(11.5)

Note. Total, N=200

Table 10.

Manual Wheelchair Community-based Study Wheel Type

Wheels by Type	N	(%)
Performance	123	(61.5)
Lite-Spoke	56	(28.0)
Solid	14	(7.0)
Other	7	(3.5)

Note. Total, N=200

The average amount of toe angle was 9.06 mm or 0.92 degrees with the more common direction being toe-out for 62% of participants. Additionally, slop was an average of 5.98 mm or 0.61 degrees. A break down in the results are shown Figure 4. Out of the last fifty participants, twenty-nine had pneumatic tires and twenty-one had solid or airless inserts. The average tire inflation pressure was forty percent calculated by the result of the digital tire pressure gauge measurement averaged across both rear-wheels compared to sidewall labeled maximum inflation pressure for the last 50 participants.

Figure 4.

Toe Angle and Slop Prevalence in the Community

Statistical Analysis

The graphical analysis did not provide detailed insight into the data but did identify outliers. No predictor was shown to be statistically significant (p < 0.05) in relation to toe angle, but two were significant predictors of slop, which included tire manufacturer (Figure 5 and Table 11) and wheel diameter (Chi square = 12.80, p = 0.03, df = 5 and Chi square = 8.94, p = 0.30, df = 3). Pairwise comparisons were conducted post-hoc with a Bonferroni adjustment and found the tire manufacturer D was significantly lower in slop than manufacturer E (p = 0.02). These results should be interpreted cautiously because of the relatively small numbers of wheelchairs from Manufacturers D (8) and E (6).

Figure 5.

Tire Manufacturer versus Slop

Table 11.

Slop across each tire manufacturer

	Mfr. A	Mfr. B	Mfr. C	Mfr. D	Mfr. E	Other
Average (deg)	0.49	0.54	0.59	0.26	1.04	0.64
Std Dev (deg)	0.37	0.49	0.36	0.21	0.71	0.46
N (qty)	79	70	15	8	6	22

The pairwise comparisons for wheel diameter show that the 56 cm (22-inch) wheels are significantly higher in slop compared to 61 cm (24-inch) wheels (p = 0.02), with the 61 cm (24-inch) grouping dominating the sample. Figure 6 and Table 12 show the graphical representation and descriptive statistics from wheel diameter groupings.

Figure 6.

Wheel Diameter versus Slop

Table 12.

Wheel Diameter versus Slop

	56 cm (22 Inch)	61 cm (24 Inch)	63.5 cm (25 Inch)	66 cm (26 Inch)
Average (deg)	1.14	0.49	0.54	0.64
Std Dev (deg)	0.75	0.37	0.40	0.59
N (qty)	7	117	53	23

Discussion

This community-based study identified the prevalence and severity of MWC rear wheel misalignment and its relationship to wheelchair factors. We found that misalignment is prevalent, and that 61% (122) of the wheelchairs measured had misalignment while loaded of greater than 0.5 degrees. Misalignment of greater than 0.5 degrees is associated with a significant increase of RR that could increase MWUs risk of UE RSIs. Toe angle provides no benefit to the user when propelling and only adds to RR, therefore increasing the required propulsion force for the MWU to move their wheelchair. It is likely that slop leads to increased toe angle RR in a cumulative manner with toe angle. This is an ever-present issue that would not resolve itself and can only get worse over time due to increased slop or reduced tire pressure. Tire pressure can be easily mitigated with routine checks[11]. Furthermore, this sample represented active users, with the majority using their devices more than five days per week for over ten hours per day. Importantly, because these participants are highly active, the increase in RR forces associated with misalignment may disproportionately increase their risk of UE RSI [6].

To convey the impact of toe angle on RR, the loading equation was used to back-calculate the perceived weight equivalent added to the rear axle of a device and is shown in Table 13. Reduced tire pressure at 40% of maximum is equivalent to putting a nine-kilogram weight on the MWU’s lap and having them propel around in that state all the time. While toe angle and slop may not independently lead to a significant increase in RR, when they are combined into a worst-case alignment scenario, the equivalent weight is twenty-one kilograms. Additionally, drum-based testing has shown that the factors act in a cumulative manner if more than one factor is present. Therefore, the absolute worst-case scenario is toe angle + slop + tire pressure + camber for an approximation of combined perceived weight equivalent of 33.6 Kg (F_RR increase of 2.3 N from a standard condition of 2.1 N). That would be a significant addition to most MWUs, and that is the current reality for MWCs in the community.

Table 13.

Average Results from the Community-based Study

Measurement	Average	Standard Deviation	Perceived Weight Equivalent
Toe Angle (mm)	9	12	11 Kg
Toe Angle (Deg)	0.9	1.4
Slop (mm)	6	8	2 Kg
Slop (Deg)	0.6	0.8
Toe Angle + Slop	1.5	N/A	21 Kg
Camber (deg)	3.0	1.5	3 Kg
Tire Pressure	40%	24%	9 Kg

An important result of this work is to provide information for rehabilitation engineers and clinicians and to improve education, guidelines, and standards for wheelchairs. The novel test rig is the only one that exists and therefore could lead to becoming the gold standard measurement of misalignment in manual wheelchairs. Currently, this is the only study to measure the misalignment of MWC wheels in a United States community that we are aware of. End-users, manufacturers, clinicians, and service providers need to be aware of these issues, the prevalence of toe angle, slop and tire underinflation in MWCs. Manufacturers need to ensure they are supplying quality MWCs, where axle tubes are properly aligned, and frames are square. Wheels should also be trued from the factory. During the setup and final fitting of device issuance, care should be taken not to induce toe angle into the rear-wheels. This can be incited due to a camber tube being rotated forward or backward, a change in seat dump, or a change in caster diameter. End-users and providers need to be vigilant of maintenance issues related to age and use of the device such as worn bearings or axles, wheels out of true, and other maintenance issues including underinflated tires. In most cases, it is not recommended to switch to a solid or airless insert because they have a higher RR than severely under inflated pneumatic tire [12].

Limitations

There are a few limitations to note from this study, which include the questionnaire, the novel test rig, the adjustments over the course of data collection, static conditions measurements, results groupings, and that it is a convenience sample. For the results groupings, a team approach was employed to ensure they were grouped in a logical manner, but others may have differing opinions on this approach. This was also a convenience sample of those at adapting sporting events that wanted to participate. While the sample size was relatively large, it may not be representative of a cross-section of the MWU population.

Future Work

Accompanying future testing, future work should include development of a standard or threshold of compliance of misalignment. Toe angle measurements could be implemented into durability testing by strengthening MWC standards to include rear-wheel alignment testing. Additionally, toe needs to be measured under dynamic scenarios to have a more comprehensive understanding of how it effects propulsion. Other options to mitigate this issue is to check toe angle and slop over time, which would require the development of a low-cost test jig that can be easily used in clinics or by providers and a common adjustment mechanism for toe angle or slop on MWCs. One area that could be improved would be expanding the tire pressure data collection because it was only captured on 50 participants (25% of the total population). Another comparison that would strengthen the knowledge base is the measurement of propulsion forces when toe angle or slop are introduced to compare to the measured resistive vales. Continuation of this research could help increase the ability to group predictors and be able to find statistically significant predictors. Furthermore, attention to maintenance issues and methods of adjustment need to be developed and implemented. Further study of the prevalence of tire underinflation levels would be valuable to expand the data available. Improvements to the ease of measuring tire pressure and inflating tires for MWU’s would also be valuable based on our results. There is also potential future work to understand toe angle and slop prevalence and severity differences in developed versus developing countries, and for other demographic and geographic considerations.

List of Abbreviations

MWC

Manual wheelchairs

RSI

repetitive strain injuries

UE

Upper extremities