Does baseline physical function modify the response to a physical activity intervention in adults with arthritis? A secondary analysis of a randomized trial.

Responses to physical activity interventions vary among adults with arthritis, and baseline functional limitations may affect increases in walking. To determine whether baseline physical function moderated changes in accelerometer-measured steps/day after a telephone-delivered Walk With Ease (WWE-T) intervention. This secondary analysis included 267 adults from a randomized trial of WWE-T versus wait-list control. Baseline function was assessed with the 30-Second Chair Stand Test and 6-Minute Walk Test and categorized as high or low using sample means. Steps/day were measured at baseline, 6 weeks, 6 months, and 12 months. Repeated-measures mixed-effects models tested time, physical function category, intervention group, and their three-way interaction, adjusting for age, sex, and body mass index. Participants had a mean age of 64.2 years (standard deviation 9.2); 92% were female, and mean body mass index was 34.2 kg/m² (standard deviation 7.7). The three-way interaction for chair stands was not significant (P = 0.27). Higher chair-stand performance was associated with more steps/day across follow-up (P = 0.0003). In the high-repetition WWE-T subgroup, steps/day increased at 6 weeks by 731 (95% confidence interval 281 to 1182) and at 6 months by 457 (72 to 843). The three-way interaction for walking endurance was not significant (P = 0.20), though higher walking endurance was associated with more steps/day (P < 0.0001). The presence of limitations with physical function did not moderate changes to steps/day following a telephone‑delivered physical activity intervention, though those without limitations had more steps/day.