Feasibility of Online and On-Site Physical Activity Delivery Among Older Adults: Mixed Methods Pilot Randomized Controlled Trial

BACKGROUND: Physical activity (PA) is central to healthy aging, yet interventions targeting older adults often report substantial attrition and variable adherence. Although digital and internet-based exercise programs broaden accessibility and flexibility, their ability to sustain engagement compared with traditional on-site formats remains unclear. Understanding how delivery mode influences attrition and adherence-and why participants persist or withdraw-is essential for designing inclusive and sustainable PA interventions. OBJECTIVE: This pilot study evaluates the feasibility of delivering a 26-week PA intervention to older adults through 2 modes: online or on-site. Feasibility was assessed primarily by attrition, adherence, and participants' experiences of factors influencing continued participation or withdrawal. Additionally, the study explores preliminary patterns in PA levels and physical and mental health outcomes to inform the design of a future definitive trial. METHODS: This 26-week pragmatic pilot trial used an embedded convergent mixed methods design. Community-dwelling adults aged 60 years or older (N=35) from 3 residential areas in Sweden were cluster-randomized, without blinding, by site to an online (prerecorded video-based Otago exercises; n=12) or on-site (group-based, in-person Otago exercises; n=23) delivery mode. Feasibility outcomes included attrition and adherence. Exploratory outcomes included PA level (Godin-Shephard Leisure-Time Exercise Questionnaire), lower-body strength (Chair Stand Test), depressive symptoms (Geriatric Depression Scale), and health-related quality of life (EQ-5D-3L), assessed at baseline, midintervention (14 weeks), and postintervention (26 weeks). Exploratory quantitative analyses included t tests, chi-square tests, analysis of covariance, and mixed-design ANOVAs. We analyzed semistructured interviews with completers (n=17) and dropouts (n=7) using qualitative content analysis. Findings were integrated using the COM-B (Capability, Opportunity, Motivation-Behavior) framework. RESULTS: The overall attrition rate was 37% (13/35), with no significant difference between delivery modes (P=.29). Most dropouts occurred within the first 2 weeks. Mean adherence appeared higher in the online delivery mode (mean 17.0, SD 6.7) than in the on-site delivery mode (mean 12.1, SD 10.1) but did not differ significantly (P=.15). Qualitative findings indicated that attrition was primarily driven by physical health issues and a perceived lack of fit, whereas adherence was supported by appeal and reward, routine and commitment, flexibility, feedback and PA overview, and (often) social presence. Mixed methods findings suggest that capability, opportunity, and motivation supported participation across both delivery modes. No intervention-related adverse events were reported. CONCLUSIONS: Both online and on-site delivery modes for PA interventions appeared feasible for many older adults. Although no statistically significant between-group differences were observed, the study was not powered to evaluate effectiveness. Feasibility may depend more on supporting capability, opportunity, and motivation than on delivery mode. Larger, adequately powered trials are needed to test effectiveness, identify optimal formats for different populations, and evaluate strategies to reduce early attrition. TRIAL REGISTRATION: ClinicalTrials.gov NCT05885308; https://clinicaltrials.gov/ct2/show/NCT05885308.