Improving interventions for physical activity and depression

Abstract

Background: There is increasing evidence supporting the potential use of physical activity as an intervention for depression. However, effective behavioural interventions which focus on encouraging physical activity in people with depression are needed. Aim: This thesis reports on the development and pilot evaluation of a web-based intervention based on behavioural activation (BA) and physical activity promotion for people with depression (eMotion). There is also a specific focus on intervention fidelity. Method: A systematic review was conducted to identify how fidelity is assessed in behavioural interventions promoting physical activity in adults. Findings from the systematic review and the Centre for eHealth and Disease Management (CeHReS) Roadmap informed the development of eMotion. A novel “design fidelity” methodology was devised, which involved using an independent coder to systematically verify the presence or absence of intended and non-intended behaviour change techniques (BCTs). A pilot RCT of eMotion was conducted to provide evidence on the feasibility and acceptability of eMotion, and exploratory data on outcomes. A mixed methods process evaluation of the understanding, feasibility and enactment of the main BCTs used in eMotion, and participant’s perceptions of eMotion was conducted. Results: The systematic review presents a critical appraisal of the wide range of methods and measures used to assess intervention fidelity. The eMotion intervention consisted of weekly modules designed to re-engage people in routine, pleasurable, necessary or physical activities. Design fidelity of the intervention was excellent with discrepancies between intervention materials and the intended design for 12 out of 221 (AC1 =0.91) and 14 out of 221 (AC1 = 0.88) BCTs relating to BA and physical activity respectively. In the pilot trial, 62 people (≥10 on the PHQ-8) were randomised, and the two-month attrition rate was 19% (95% CI: 11 to 30%). Study completers in the eMotion group (n=25) had a significantly larger reduction in depressive symptoms than the waiting list control group (n=25) (Adj Mean Diff -3.6, 95% CI: -6.1 to -1.1). Participants using eMotion also reported significantly higher levels of understanding, ability and enactment of most BCTs compared with controls. Interviews with participants revealed a number of possible mechanisms by which eMotion may have changed behaviour (e.g. increased awareness of the link between mood and activity) and reduced depression (e.g. through distraction). Conclusion: This thesis has reported on the development and pilot evaluation of a behavioural intervention promoting physical activity for depression while focussing on fidelity. Further insight into a number of methodological and clinical implications for the conduct of future RCTs in this area has also been provided.