Mediation and Moderation Effects in a Blended Physical Activity Intervention for Office Employees

Physical inactivity is a critical global health challenge, especially for sedentary office employees. Blended interventions show promise for improving moderate-to-vigorous physical activity (MVPA), but their mechanisms remain understudied. This study investigated the mediation effects of theoretical determinants and the moderation effects of sociodemographic characteristics on MVPA changes. Data were drawn from a 24-week randomized controlled trial ( N  = 141) comprising a blended intervention, a web-based intervention, and a control group. The intervention, developed using Intervention Mapping and informed by dual-process perspectives on behavior change, incorporated motivational, self-regulatory, and habit-formation strategies to promote MVPA. The blended intervention combined web-based modules with live e-workshops to enhance engagement and support behavior change. Mediation analyses identified self-efficacy as the only significant mediator, specifically for the effect of the blended intervention versus control on MVPA change from baseline to follow-up 12 weeks after the intervention, suggesting its potential role in short-term maintenance of behavior change. Moderation analyses indicated that the relative effects of the blended intervention varied across sociodemographic subgroups, with stronger effects observed among males, participants aged 29 years or older, participants with lower education levels, and non-single participants in specific group comparisons and time points. These findings provide insights for optimizing MVPA interventions by strengthening self-efficacy and tailoring strategies to sociodemographic subgroups. Future research with larger, adequately powered samples, extended follow-up periods, and more diverse populations is needed to confirm these mediation and moderation pathways and further refine theoretical models underlying blended interventions.