Research Article: Test–retest reliability and construct validity of the Aspects of Wheelchair Mobility Test as a measure of the mobility of wheelchair users

Date Published: September 08, 2017

Publisher: AOSIS

Author(s): Karen L. Rispin, Kara Huff, Joy Wee.


The Aspects of Wheelchair Mobility Test (AWMT) was developed for use in a repeated measures format to provide comparative effectiveness data on mobility facilitated by different wheelchair types. It has been used in preliminary studies to compare the mobility of wheelchairs designed for low-resource areas and is intended to be simple and flexible enough so as to be used in low-technology settings. However, to reliably compare the impact of different types of wheelchairs on the mobility of users, a measure must first be a reliable and valid measure of mobility.

This study investigated the test–retest reliability and concurrent validity for the AWMT 2.0 as a measure of mobility. For reliability testing, participants in a low-resource setting completed the tests twice in their own wheelchairs at least one week apart. For concurrent validity, participants also completed the Wheelchair Skills Test Questionnaire (WST-Q), a related but not identical validated assessment tool.

Concurrent validity was indicated by a significant positive correlation with an r value of 0.7 between the WST-Q capacity score and the AWMT 2.0 score. Test–retest reliability was confirmed by an intraclass correlation coefficient greater than 0.7 between the two trials.

Results support the preliminary reliability and validity of the AWMT 2.0, supporting its effectiveness in comparing the mobility provided by different wheelchair types. This information can be used to enable effective use of limited funds for wheelchair selection at individual and organisational scales.

Partial Text

Comparative effectiveness studies on the mobility facilitated by wheelchairs designed for use in low-resource areas are lacking (Harniss, Samant Raja & Matter 2015; Matter et al. 2017; Pearlman et al. 2008). The Aspects of Wheelchair Mobility Test (AWMT) was developed as a physical performance measure to provide comparative effectiveness data on wheelchairs designed for low-resource settings (Rispin & Wee 2013, 2015). The AWMT, which is described in a companion paper in this journal, is intended to be used in a repeated measures format to assess the impact of wheelchair type on mobility in commonly encountered rolling environments (Rispin & Wee 2013, 2015).

The purpose of this study was to investigate the reliability and validity of the AWMT 2.0 as a physical performance measure of mobility. Test–retest ICC results well above 0.7 confirm reliability for distance travelled and visual analogue score responses. Significant correlations with the WST-Q capacity scores confirm validity. The WST-Q is a validated physical performance measure. If the AWMT 2.0 is measuring physical performance aspects of mobility, one would expect significant positive correlation between the two measures. This was the case, and concurrent validity of the AWMT 2.0 was confirmed by positive and significant correlation between the total WST-Q capacity score and AWMT 2.0 distance travelled on all tracks. This was also the case for the correlation of AWMT 2.0 mean visual analogue score and WST-Q capacity score. In fact, these positive correlations emphasise the validity of both measures. One might ask why the AWMT 2.0 is needed if the WST-Q is a validated measure. The WST-Q is designed, as the name describes, to test the skill level of an individual wheelchair user. As such, the WST-Q is not primarily designed for comparative effectiveness studies on the impact on mobility of different wheelchair types. If a very strong wheelchair user can roll on rough ground, it is likely that he or she could do that in most wheelchair types, and the 0–2 categorical rating scale might not pick up a slower velocity or a greater difficulty in one wheelchair type as compared to another. Because of the categorical nature of WST-Q questionnaire data for each question, analysis of variance (ANOVA) could not be used to compare the impact of different wheelchair designs on capacity for each skill. If all 32 questions of the WST-Q were to be used, there would be greater discriminative power. However, this would be difficult to do in a repeated measures study because each participant would need to complete or attempt to complete many skills in each wheelchair. This would be very time-consuming and physically wearing. In contrast, because the data are continuous, ANOVA can be applied to AWMT data with rolling environments and wheelchair types acting as factors (see the accompanying paper).

These findings indicate that the AWMT 2.0 is a reliable and valid measure of a wheelchair user’s mobility. This confirms the soundness of comparative effectiveness studies conducted using the AWMT 2.0 for similar populations. As a clinical tool, the AWMT 2.0 could enable wheelchair fitting by allowing direct objective comparisons of the mobility provided by wheelchairs and configurations. In larger studies comparing cadres of wheelchairs of two or more types, the AWMT 2.0 can provide comparative effectiveness data to manufacturers, clinics and stakeholders. Because this study was conducted with a population of adolescent participants in one low-resource area, further studies are needed for other populations and in other locations.




Leave a Reply

Your email address will not be published.