Résumé
Background: Physical activity (PA) benefits young adults living with and beyond cancer, yet participation remains low. The ACCESS intervention aims to address low PA participation in this cohort. Purpose: We analyzed data from a pilot randomized controlled trial on PA collected before (baseline; week 0) and after the 12-week ACCESS intervention (post-intervention [week 13], follow-up [week 24]) to inform the design of a full-scale efficacy trial. Methods: Forty-two young adults (Mage=33 years; 81% female) were randomized post-cancer treatment to a virtual person-centered behaviour change intervention or usual care. The intervention, guided by self-determination theory, comprised 6x60-minute biweekly sessions incorporating behaviour change techniques to affect theory-driven mediators. The primary outcome (total PA hours/week) was assessed using accelerometers and analyzed descriptively using linear regression modelling, accounting for the correlation in repeated measures and prespecified covariates. No significance testing was performed; the correlation for repeated measures and variance estimates were used to inform sample size calculations for a full-scale efficacy trial. Results: The mean ± standard deviation for PA in the intervention and control arms at baseline was 28.8±10.2 and 25.9±9.4 (range=5.6-46.7), respectively. The correlation for repeated measures of PA was r=0.67, and the combined variance was 103. Graphical representation of the model-based means and standard errors by arm with lines depicting individual profiles across assessments revealed important heterogeneity in PA for the whole sample. Conclusion: While efficacy conclusions cannot be drawn, results offer essential parameters for powering and implementing a future full-scale efficacy trial of the ACCESS intervention.
