Effectiveness and theory-based evaluation of a personalised digital intervention (EviBody®) for healthy and sustained lifestyle behaviours and well-being among adults: Study protocol for a real-world quasi-experimental study.

BACKGROUND: Digital interventions offering behaviour change support are warranted to prevent and treat non-communicable diseases, and have been evaluated rigorously in controlled settings. Effectivenss, factors influencing the uptake of scaled-up interventions-such as reach, received dose, usability and acceptability- and predictors and mediators of efficiency are rarely explored in research. The study described herein aims to evaluate the effectiveness of a personally tailored digital intervention (the app EviBody®), intended to support healthy and sustained lifestyle behaviours among the adult population, on well-being and behaviour change. Further aims are to explore context and uptake factors, predictors and mediators for behaviour change over 24 months. METHODS: This is a real-world study, employing a quasi-experimental design and a process evaluation. EviBody® will be marketed and managed by its owner. A four-armed design will allow for comparison between three levels of intervention (basic, standard and premium) and a control group. Adults who sign up for the app will be invited to the research study including sharing app data and answering questionnaires at 0, 1, 3, 6, 12, 18, and 24 months. Study start is Autumn 2025. Controls (n = 200 to evaluate the primary endpoint well-being at 6 months) will be recruited through advertisements on social media and asked to answer the same questionnaires at 0 and 6 months provided by email. For predicting and mediating analyses the intention is to recruit 1500 app users. Well-being (measured with the WHO-5 Well-Being Index), goal achievement, physical activity, eating habits, mental health, mediators (motivation, self-efficacy, and perceived barriers), and demographics will be self-reported. Uptake will be collected using analytics and ratings of usability and acceptability, and described by demographics. Mixed models for repeated measures and structural equation modelling will be employed for data analysis. DISCUSSION: Besides evaluating the effectivenss of a digital intervention, this study also applies a theory-based evaluation to understand which mediators are effective, for whom they are effective, and the specific conditions under which they are most beneficial. TRIAL REGISTRATION: ClinicalTrials.gov, ID: NCT05973383 on 8 July 2023.