Process evaluation of an app-based intervention with behavioural support to promote brisk walking in people diagnosed with breast, prostate, and colorectal cancer (APPROACH) (Preprint)

BACKGROUND The APPROACH pilot study explored the feasibility and acceptability of an app (National Health Service NHS Active 10) with brief habit-based behavioural support calls and print materials intended to increase brisk walking in people diagnosed with cancer. OBJECTIVE Following UK Medical Research Council guidelines, this process evaluation assessed implementation of the intervention, examined the mechanisms of impact and identified contextual factors influencing engagement. METHODS Adults (>18 years) with breast, prostate or colorectal cancer who reported not meeting the United Kingdom (UK) guidelines for moderate to vigorous physical activity (MVPA; >150 mins/week) were recruited from a single hospital site in Yorkshire, UK and randomised to intervention or control (usual care) and assessed via quantitative surveys at baseline (T0) and 3 month follow-up (T1) and qualitative exit interviews (n=36:82%) at T1. The process evaluation included intervention participants only (n=44). Implementation was assessed using data from the T1 questionnaire exploring use of the intervention components. Perceived usefulness of the app, leaflet, and behavioural support call was rated 0-5. Behavioural support calls were recorded and the fidelity of delivery of 25 planned behaviour change techniques (BCTs) were rated 0-5 using an adapted Dreyfus scale. Mechanisms of impact were identified by examining T0 and T1 scores on the Self-Reported Behavioural Automaticity Index and feedback on the leaflet, app, call, and planner in the T1 questionnaire and qualitative interviews. Contextual factors influencing engagement were identified through qualitative interviews. RESULTS Implementation of the intervention was successful: 98% received a behavioural support call, 78% reported reading the leaflet, 96% reported downloading the app, and 83% reported using the planners. The mean perceived usefulness of the app was 4.3 (SD=0.8) in participants still using the app at T1 (n=33). Participants rated the leaflet as useful (M=3.9, SD =0.6) as well as the behavioural support call (M=4.1, SD =1.0). The intended BCTs in the behavioural support calls were proficiently delivered (Overall M=4.2, SD=1.2). Mechanisms of impact included habit formation, behavioural monitoring, and support and reassurance from the intervention facilitator. Contextual factors impacting engagement included barriers such as the impact of cancer and its treatment and facilitators like social support. CONCLUSIONS The APPROACH intervention was successfully implemented and shows promise for increasing brisk walking potentially through promoting habit formation and enabling self-monitoring. Contextual factors will be important to consider when interpreting outcomes in the larger APPROACH randomised controlled trial. CLINICALTRIAL 18063498 INTERNATIONAL REGISTERED REPORT RR2-10.1186/s40814-022-01028-w