Comparing Performance of Manual Wheelchair Skills Using New and Conventional Rear Anti-Tip Devices: Randomized Controlled Trial

Article Outline

• Abstract
• Methods
  • Study Design
  • Participants
  • Inclusion and Exclusion Criteria
  • Ethical Issues
  • Group Allocation
  • Wheelchairs
  • Wheelchair Skills
  • Safety
  • Participants’ Perceptions
  • Statistical Analysis
• Results
  • Demographics
  • Wheelchair Skills
  • Individual Skills
  • Participants’ Comments
• Discussion
  • Study Limitations
• Conclusions
• References
• Copyright

Abstract 

Kirby RL, Corkum CG, Smith C, Rushton P, MacLeod DA, Webber A. Comparing performance of manual wheelchair skills using new and conventional rear anti-tip devices: randomized controlled trial.

Objective

To test the hypotheses that, compared with participants using manual wheelchairs equipped with conventional rear anti-tip devices (C-RADs), those using a new RAD design that deploys through an arc (Arc-RAD) perform RAD-relevant wheelchair skills better and as safely.

Design

A randomized controlled study.

Setting

A rehabilitation center.

Participants

Participants (N=30) including 16 able-bodied and 14 wheelchair users.

Intervention

Participants were provided with wheelchair skills training (up to 2.4h).

Main Outcome Measures

Total percentage score on a set of 23 RAD-relevant skills of the Wheelchair Skills Test (WST, version 3.2) administered a minimum of 3 days after training.

Results

For the C-RAD and Arc-RAD groups, the mean ± standard deviation RAD-relevant WST scores were 32.3%±8.5% and 85.1%±18.9% (Kruskal-Wallis, P<.001). Of the 23 RAD-relevant individual skills, the success rates for the Arc-RAD group were at least 20% higher (the criterion we set for clinical significance) in 17 (74%). For the C-RAD group, the success rate was 0% for the 12 wheelie-dependent skills, the 13-cm–high obstacle, and the 15-cm level change ascent. There were no serious adverse effects in either group.

Conclusions

The new RAD design allows much better performance on relevant wheelchair skills than the conventional design without compromising safety.

Key Words: Rehabilitation, Safety, Wheelchairs

ALTHOUGH THE WHEELCHAIR is among the most important therapeutic devices used in rehabilitation, about 5% to 18% of wheelchair users sustain wheelchair-related injuries each year.¹, ² Of these injuries in the United States, each year approximately 100,000 of them are serious enough to result in an emergency department visit³ and 50 or more result in death.⁴ Over two thirds of wheelchair-related injuries are due to tips or falls.⁵, ⁶ Of the tips that occur, those in the rear direction are particularly worrisome because of the difficulty that users have in protecting their heads from striking the ground.⁷ Rear stability is affected by a number of wheelchair and user factors.⁸

Currently available wheelchairs are often fitted with conventional rear anti-tip devices (C-RADs) that extend from the rear frame of the wheelchair (fig 1). Although able to limit the extent of rear tip if they are properly adjusted, most C-RADs require a trade-off between safety and performance.⁹, ¹⁰ For instance, with C-RADs in place, most wheelchairs cannot be tipped back far enough to ascend a curb. For this reason, C-RADs are often removed by the wheelchair user or adjusted into ineffective positions.⁹

• 
  • View Large Image
  • Download to PowerPoint

• Fig 1. 

  C-RADs on a wheelchair.

The limitations of C-RADs have led to alternative designs that permit more rear tip. One such design is the Arc-RAD.¹¹ The Arc-RAD is out of the way during normal wheelchair use (fig 2A) but self-deploys when needed. When the wheelchair tips backward sufficiently, the lower end of the arm (on which a small swivel caster is located) contacts the ground. The force of the tipping chair rotates the arm through an arc until it hits a cam that acts as a stop to prevent any further tipping (fig 2B). Although participants using wheelchairs equipped with Arc-RADs have been compared with those using wheelchairs without anti-tip devices,¹² no study has compared Arc-RADs and C-RADs from the perspective of the ability of wheelchair users to perform wheelchair skills.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 2. 

  The new Arc-RAD design. The Arc-RADs (A) remain out of the way at rest but (B) permit moderate rear tip when they self-deploy because of ground contact.

The primary objective of this study was to test the hypotheses that, compared with participants using manual wheelchairs equipped with C-RADs, those using a new Arc-RAD design perform RAD-relevant wheelchair skills better and as safely. Our secondary objectives were to obtain data that might help us refine our training methods and to obtain qualitative observations from participants that might be of use in refining the Arc-RAD design.

Back to Article Outline

Methods 

Study Design 

This was a randomized controlled study of 2 groups, 1 using wheelchairs equipped with C-RADs and the other using wheelchairs equipped with Arc-RADs.

Participants 

We studied 30 participants, a sample of convenience. The sample-size determination was based on estimates of the population standard deviation (SD) (from the variance data of an efficacy study of a wheelchair-skills training program on new wheelchair users),¹³ in which the overall difference between the means of the groups was 12.5%±7.4%. For the purposes of the power analysis, we treated the Wheelchair Skills Test (WST, version 3.2) value as a continuous, normally distributed variable. We defined a difference of 20 WST percentage points in the RAD-relevant WST scores between the groups as the minimum clinically significant difference. To detect a difference of at least this magnitude between the Arc-RAD and C-RAD groups, a minimum of 5 participants needed to be studied in each group to have 95% power, assuming that a 2-sided, 2-sample t test was used with an α level of .05. We elected to use a larger sample, both to obviate any difficulty should our variance be greater than in the earlier study and to help us meet our secondary objectives. Although wheelchair users were preferred, we also studied able-bodied persons in the interest of completing the study in a reasonable period of time.

Potential participants were recruited by word of mouth (for able-bodied participants) or by their clinicians (for wheelchair users). Wheelchair users were recruited from the Nova Scotia Rehabilitation Site of the Queen Elizabeth II Health Sciences Centre. Demographic and wheelchair-usage data were recorded. Clinical data were obtained by interview and chart review.

Inclusion and Exclusion Criteria 

The inclusion and exclusion criteria were evaluated by self-report for the able-bodied participants and by the wheelchair-using participants’ attending physicians. All participants met the following inclusion criteria: 18 years of age or older; able to provide informed consent; willing to participate; alert, able, and willing to follow instructions; able to answer questions about wheelchair use; and had body sizes that fit the wheelchairs used. To ensure that each potential participant had the necessary skill and strength to perform the RAD-relevant skills, he/she had to be able to perform or be able to learn all of the indoor skills and the 5-cm curb ascent of the WST.¹⁴ Participants were excluded if they had any unstable medical conditions or had any emotional problems that might make testing or training unpleasant.

Ethical Issues 

The study was approved by the Research Ethics Board of the Capital District Health Authority. All participants provided informed consent.

Group Allocation 

Participants were randomly assigned (by a number randomization code) to either the Arc-RAD or C-RAD groups.

Wheelchairs 

Two standard rear-wheel–drive lightweight (15.4-kg) manual wheelchairs^a were used for all testing and training. The wheelchair assigned to the Arc-RAD group was fitted with a pair of prototype Arc-RADs. The Arc-RADs were adjusted for each participant, using criteria reported earlier.¹² The C-RADs were adjusted to their most stable positions (as rearward as possible) consistent with the recommended clearance of 3.8cm in the manufacturer’s user instruction manual. The wheelchair set-up (eg, axle position), which can affect wheelchair handling, was otherwise identical for both wheelchairs.

Wheelchair Skills 

Testing and training were performed by the investigators using the principles and procedures of the standardized Wheelchair Skills Program.¹⁴ Each participant was trained on any of the skills that he/she was unable to perform. Training sessions were between 30 and 60 minutes in length, at target intervals of 1 to 7 days. When all of the RAD-relevant skills had been successfully learned or abandoned (by the agreement of the participant and trainer), training was deemed to be complete.

The WST¹⁴ was administered a minimum of 3 days after training.¹⁵ The only modification in the WST procedures, for the purposes of this study, was that participants were not permitted to get out of their wheelchairs to adjust or remove the RADs. As is routinely permitted in the WST, participants were permitted to lean against the RADs during the performance of wheelie-dependent skills.¹² The main outcome measure was the total percentage score of a set of 23 RAD-relevant skills.

Safety 

Each wheelchair was fitted with a spotter strap that was held by a spotter during training and testing of skills that had the potential for rear tips. Adverse incidents were recorded.

Participants’ Perceptions 

A brief questionnaire was administered at the end of the study with questions about how the participants perceived their experiences. Unstructured comments were recorded.

Statistical Analysis 

Statistical analyses were performed using SAS statistical software.^b Descriptive statistics, including evaluations of normality, were performed to describe the sample and to guide the choice of parametric or nonparametric statistical tests. The 2 groups were compared from the perspectives of age (Kruskal-Wallis test), sex (chi-square test), height (t test), weight (t test), clinical status (able-bodied vs wheelchair user) (chi-square test), and duration of training (Welch modified 2-sample t test). For the between-group comparison of RAD-relevant WST scores, we used a 2-sided Kruskal-Wallis test and an α level of .05. We evaluated the effects of clinical status (ie, able-bodied vs wheelchair user) and demographic variables by using mixed models in a multivariate analysis. For this process, the dependent variable was the RAD-relevant WST percentage score and the independent variables were group (Arc-RAD vs C-RAD), clinical status (wheelchair user vs able-bodied), height, weight, sex, and age. The success rates on the individual skills were assessed descriptively, because there was not sufficient power in the study to do so statistically.

Back to Article Outline

Results 

Demographics 

We studied 30 participants. Sixteen of them were able-bodied persons simulating paraplegia (ie, not permitted to use their legs to perform skills), and 14 were wheelchair users. Their demographic and clinical data are shown in table 1. The major diagnoses accounting for wheelchair use among wheelchair users included unilateral lower-limb amputation (n=5), bilateral lower-limb amputation (n=4), spinal cord injury (n=2), stroke (n=2), cerebral palsy (n=1), and bilateral leg contractures (n=1). The CONSORT diagram¹⁶ for the wheelchair-using participants is shown in figure 3. There were no statistically significant differences between the C-RAD and Arc-RAD groups for age (P=.88), sex (P=.30), height (P=.48), weight (P=.17), or clinical status (P=.26).

Table 1. Study Participants’ Demographic and Clinical Data (N=30)

Abbreviations: AB, able-bodied participant; WCU, wheelchair-using participant.

• 
  • View Large Image
  • Download to PowerPoint

• Fig 3. 

  CONSORT diagram for wheelchair-using participants. During the 289-day recruitment period, there were a total of 401 inpatients admitted to the rehabilitation center.

Wheelchair Skills 

Wheelchair-using participants in the Arc-RAD group required more training time than those in the C-RAD group, with mean values ± SD of 89.8±34.9 and 40.3±10.8 minutes, respectively (P=.0317). The mean RAD-relevant total WST percentage scores ± SD for the Arc-RAD and C-RAD groups were 85.1%±18.9% and 32.3%±8.5%, respectively (P<.001). As illustrated in figure 4, the 2 groups did not overlap (ie, the minimum Arc-RAD score of 43.5% was greater than the maximum C-RAD score of 39.1%).

• 
  • View Large Image
  • Download to PowerPoint

• Fig 4. 

  RAD-relevant percentage scores on the WST for the Arc-RAD and C-RAD groups. The horizontal lines in the middle of the boxes are the median values, the lower and upper ends of the boxes are the 25th and 75th percentiles, the whiskers extend to the furthest nonoutlier values, and outliers are displayed as dots.

When using multivariate analysis to look at the factors that influenced RAD-relevant WST percentage scores, both group (Arc-RAD vs C-RAD) (P<.001) and clinical status (able-bodied vs wheelchair user) (P<.001) were significant (Arc-RAD and able-bodied groups higher). Height, weight, and sex were not significant when included in this model as additional terms or as the only variables in the model. Wheelchair-using participants were generally older than able-bodied participants.

Individual Skills 

The success rates for the individual skills are shown in table 2. Of the 23 RAD-relevant skills, the success rates for the Arc-RAD group were at least 20% higher (the criterion we set for clinical significance) in 17 (74%). For 14 (61%) of the RAD-relevant skills, the success rates for the C-RAD group were 0%, whereas the success rates on those skills for the Arc-RAD group ranged from 21.4% to 92.9%. The 6 RAD-relevant skills for which there were no clinically significant differences between the groups were the 2-cm–high obstacle, the gravel, the 15-cm pothole, the 5° incline ascent, the 5° incline descent, and the descent of the 5-cm level change. There were no clinically significant differences between the groups for any of the 34 non–RAD-relevant skills.

Table 2. Success Rates for Individual Skills of the WST 3.2

Legend: +, skill considered RAD-relevant.

Participants’ Comments 

Most (72% and 86%, respectively) of the wheelchair-using participants did not find the training or testing to be physically or emotionally stressful. Thirteen (93%) of them found the training sessions useful and perceived that they had improved their abilities. Their comments on the training experience included, “The session better equipped me for getting around in my wheelchair at home and outside,” “Useful for me when I have to be on my own,” and “I have more confidence.” Their comments on the RADs included, “I like the [Arc-RADs] because you can get over things better than with the [C-RAD] kind” and “These skills are a lot easier to do in the [Arc-RAD] wheelchair than a normal one.”

Back to Article Outline

Discussion 

This study corroborated the hypothesis that participants using manual wheelchairs equipped with Arc-RADs perform better on RAD-relevant wheelchair skills than participants using wheelchairs equipped with C-RADs. The extent of the difference was substantial, with the mean score of those in the Arc-RAD group being 2.6 times greater that of those in the C-RAD group. There were no adverse effects of Arc-RAD use on safety. The success rates for individual skills were higher to a clinically significant extent (ie, at least 20%) for the Arc-RAD group in 74% of the 23 RAD-relevant skills. The differences between the groups were particularly striking for the wheelie-dependent skills. The direction and extent of the differences between the groups in this study were consistent with those of the earlier study on Arc-RADs.¹²

There were no significant differences between the groups with respect to demographic factors or clinical status. The able-bodied participants were younger than the wheelchair users, and age had a significant effect on WST scores (as has been reported earlier¹⁵), but this affected both the Arc-RAD and C-RAD groups.

Regarding the secondary objective of learning more about wheelchair skills training, the fact that some participants in both groups had difficulty with some of the relatively easy skills (eg, armrest handling) suggests that more training might have been advisable. Most of the wheelchair-using participants tolerated the training well and found it useful. The most likely explanation for the fact that the C-RAD group required less training time was because so many of the wheelie-related skills were impossible with the C-RADs in place. Regarding the secondary objective of making observations that might affect the Arc-RAD design, no obvious design flaws came to light.

Study Limitations 

We recognize that there were a number of study limitations. The sample size was small. Although we had sufficient power to identify the large differences between the groups for the WST scores, there was insufficient power to statistically analyze the different success rates for the individual skills. The sample was one of convenience and therefore not random. Differences between groups may be more or less evident in a larger sample, among persons with different diagnoses, or in older subjects. Although the use of able-bodied participants was a study limitation, we considered this to be an acceptable decision given that this was a preliminary study and that the focus of the study was on comparing 2 RAD designs rather than on wheelchair users themselves. Also, the multivariate analysis showed that the group differences were independent of whether the subjects were wheelchair users or able-bodied.

Although we believe that our study policy of not allowing participants to get out of their wheelchairs to adjust or remove their anti-tip devices was justifiable given the objectives of the study, this could have led to an underestimation of the WST scores for any participants in the C-RAD group who were capable of getting out of their wheelchairs to do so. Our use of a pair of standardized wheelchairs minimized 1 source of variability between the groups, but wheelchair users might have performed differently in their own wheelchairs or wheelchairs of different design. The Arc-RADs used were experimental prototypes, not products that are commercially available.

Future research and development are needed to address the limitations noted. In particular, the study should be replicated with a larger, more representative sample, including wheelchair users at the more limited end of the performance spectrum (eg, those who require the assistance of caregivers). Despite the study limitations and the need for further study, this study is the first to identify a significant difference between conventional RADs and the new design. Our results, although preliminary in nature, are encouraging and suggest that refinements of wheelchair design can enhance the performance of wheelchair users without compromising safety.

Back to Article Outline

Conclusions 

The new rear anti-tip device design allows much better performance on relevant wheelchair skills than the conventional design without compromising safety.

Suppliers

Back to Article Outline

References 

1. Kirby RL, Ackroyd-Stolarz SA, Brown MG, Kirkland SA. Wheelchair-related accidents caused by tips and falls among noninstitutionalized users of manually propelled wheelchairs in Nova Scotia. Am J Phys Med Rehabil. 1994;73:319–330
   • View In Article
   • MEDLINE
   • CrossRef
2. Berg K, Hines M, Allen S. Wheelchair users at home: few home modifications and many injurious falls. Am J Public Health. 2002;92:48
   • View In Article
   • MEDLINE
   • CrossRef
3. Xiang H, Chany AM, Smith GA. Wheelchair related injuries treated in US emergency departments. Inj Prev. 2006;12:8–11
   • View In Article
   • MEDLINE
   • CrossRef
4. Calder CJ, Kirby RL. Fatal wheelchair-related accidents in the United States. Am J Phys Med Rehabil. 1990;69:184–190
   • View In Article
   • MEDLINE
   • CrossRef
5. Ummat S, Kirby RL. Nonfatal wheelchair-related accidents reported to the National Electronic Injury Surveillance System. Am J Phys Med Rehabil. 1994;73:163–167
   • View In Article
   • MEDLINE
   • CrossRef
6. Gaal RP, Rebholtz N, Hotchkiss RD, Pfaelzer PF. Wheelchair rider injuries: consequences for wheelchair design and selection. J Rehabil Res Dev. 1997;34:58–71
   • View In Article
   • MEDLINE
7. Kirby RL, DiPersio M, MacLeod DA. Wheelchair safety: effect of locking or grasping the rear wheels during a rear tip. Arch Phys Med Rehabil. 1996;77:1266–1270
   • View In Article
   • Abstract
   • Full-Text PDF (1506 KB)
   • CrossRef
8. Kirby RL. Wheelchair stability: important, measurable and modifiable. Technol Disabil. 1996;5:75–80
   • View In Article
9. Kirby RL, Thoren F, Ashton B, Ackroyd-Stolarz SA. Effect of the position of rear antitippers on safety and maneuverability. Arch Phys Med Rehabil. 1994;75:525–534
   • View In Article
   • MEDLINE
10. Kirby RL, Smith C, Seaman R, Macleod DA, Parker K. The manual wheelchair wheelie: a review of our current understanding of an important motor skill. Disabil Rehabil. 2006;1:119–127
    • View In Article
11. Kirby RL. Anti-tip devices for wheeled conveyances including wheelchairs and methods related thereto. US patent 6,530,598. 2003 Mar 11.
    • View In Article
12. Kirby RL, Lugar JA, Breckenridge C. New wheelie aid for wheelchairs: controlled trial of safety and efficacy. Arch Phys Med Rehabil. 2001;82:380–390
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (175 KB)
    • CrossRef
13. MacPhee AH, Kirby RL, Coolen AL, Smith C, MacLeod DA, Dupuis DJ. Wheelchair skills training program: a randomized clinical trial on wheelchair users undergoing initial rehabilitation. Arch Phys Med Rehabil. 2004;85:41–50
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (58 KB)
    • CrossRef
14. Dalhousie University. The Wheelchair Skills Program manual (Version 3.2). http://www.wheelchairskillsprogram.ca/eng/manual.htmAccessed October 9, 2007
    • View In Article
15. Kirby RL, Dupuis DJ, MacPhee AH, et al. The Wheelchair Skills Test (version 2.4): measurement properties. Arch Phys Med Rehabil. 2004;85:794–804
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (253 KB)
    • CrossRef
16. Begg C, Cho M, Eastwood S, et al. Improving the quality of reporting of randomized controlled trials: the CONSORT statement. JAMA. 1996;276:637–639
    • View In Article
    • MEDLINE

• a Quickie LXI; Sunrise Medical Canada Inc, 237 Romina Dr, Concord, ON L4K 4V3, Canada.
• b Version 9.1; SAS Institute Inc, 100 SAS Campus Dr, Cary, NC 27513.