mHealth intervention including text messaging and behaviour change techniques to support maintenance of physical activity after cardiac rehabilitation: A single-arm feasibility study

Objective: To evaluate the feasibility of a mobile health-supported intervention in patients with cardiovascular diseases after completion of a cardiac rehabilitation programme.

Methods: The feasibility study was performed in two hospitals and one municipality in Region Zealand, Denmark. Eligible participants were ≥18 years old, participated in a supervised cardiac rehabilitation programme, had access to a mobile phone, and could walk 3 m independently. Participants received a 12-week intervention utilizing behaviour change techniques, consisting of action planning, text messages, and phone support. Feasibility was assessed using pre-defined progression criteria, which included recruitment (≥75%), retention (≥80%), accelerometer data completeness (≥80%), coordinator (phone support) time (≤30 min), the response rate on patient-reported outcomes (≥90%), adherence (≥75% respond to ≥75% of messages), and acceptability (≥75%). The secondary outcome of objective physical activity was assessed with accelerometers.

Results: Ten women and 30 men with cardiovascular diseases aged 63.5 (±9.8 SD) participated. The progression criteria for retention (90%), accelerometer data completeness (83%), coordinator time (9.9 min), adherence (83%), and acceptability (82%) were at acceptable levels, exceptions were progression criteria for recruitment (35%) being below acceptable levels for recruitment, and response rate on patient-reported outcomes (75%). High satisfaction (92.6%) with the intervention was found. All objectively measured physical activity levels remained unchanged from baseline to follow-up. No serious adverse events related to the intervention were reported.

Conclusion: Mobile health-supported maintenance of physical activity after cardiac rehabilitation completion was feasible, safe, and acceptable. Yet, changes to improve recruitment and response rate are needed before conducting a large-scale effect evaluation.