(Research Hub Theater) Intrasession Test-Retest Reliability and Minimal Detectable Change of an Ergometer Based Repeated Measures Protocol

Background and/or Objectives: To examine intrasession test-retest reliability, minimal detectable change (MDC), and sample sizes required to detect a 10% change in wheelchair propulsion mechanics during various workloads.

Design: Test-Retest Reliability of a Repeated Measures Protocol

Setting: Research laboratory affiliated with a large academic institution

Participants: N=11 (male=7, female=4) non-disabled adults (mean (SD)); age 24 years (2), BMI 23.8 kg/m² (4.3).

Interventions: N=6 one-minute manual wheelchair propulsion trials (computer-controlled wheelchair ergometer). Two trials with self-selected speed at 25%, 50%, and 75% of participants' estimated maximal capacity. The N=6 trials were part of a N=19 trial repeated measures experiment. A force and moment sensing wheelchair wheel fixed to the right side collected data and was used to generate the main outcome measures.

Main Outcome Measures: Averages of all pushes for each 1-minute trial were computed for average rolling resistance (RR (N)), speed (m/s), average force (Favg(N)), force rate of rise (Fror (N/s)), push frequency (PF (pushes/min)), and push length (PL (deg)). Intrasession reliability (intraclass correlation coefficient (ICC)), MDC estimates, and sample size to detect a 10% change (α=0.05, power=80%, two tailed, paired t-test) were calculated for each variable at each workload.

Results: Reliability increased and MDC/sample size decreased as workload increased. ICC point estimates ranged from 0.45 (Fror, 25% workload) to 0.99 (RR, 75% workload) and were ≥0.80 for all workloads for PF, Favg, and RR. Sample sizes to detect a 10% change ranged from N=5 (RR, 75% workload) to N=226 (Fror, 25% workload). MDCs (sample size) at 25% and 75% workloads respectively were: RR 6 (N=25), 6 (N=5) newtons; Speed 0.27 (N=11), 0.14, (N=8) m/s; Favg 18 (N=33), 5 (N=10) newtons; Fror 533 (N=226), 250 (N=114) newtons; PF 21(N=24), 11(N=8) pushes/min; and PL 22 (N=17), 24 (N=15) degrees.

Conclusions: Using RR ≥75% of participants' maximum capacity would improve power to detect changes in propulsion mechanics. Detecting Fror change may require multi-site studies.