Understanding how a digital mental health intervention can be optimised to ensure effectiveness in the longer-term: findings from a causal mediation analyses of the CONEMO trials

Background Two CONEMO trials in Lima, Peru and São Paulo, Brazil evaluated a digital mental health intervention (DMHI) based on behavioural activation (BA) that demonstrated improvements in symptoms of depression between trial arms at three-months, but not at six-months. To understand how we can optimize CONEMO in the longer-term, we therefore aim to investigate mediators through which the DMHI improved symptoms of depression at six-months, separately for the two trials and then using a pooled dataset. Methods We used data that included adults with depression (Patient Health Questionnaire – 9 (PHQ-9) score ≥10) and comorbid hypertension and/or diabetes. Interventional effects were used to decompose the total effect of DMHI on symptoms of depression at six months into indirect effects via: understanding the content of the sessions without difficulty; number of activities completed that were self-selected to improve levels of BA; and levels of activation measured using the Behavioural Activation for Depression Short Form (BADS-SF). Findings Using the pooled dataset, understanding the content of the sessions without difficulty mediated a 10% [0.10: 95% CI: 0.03 to 0.15] improvement in PHQ-9 scores at six months; completing self-selected activities mediated a 12% improvement [0.12: 0.01 to 0.23]; and, lastly, BA mediated a 2% [0.02: 0.01, 0.05] improvement. Conclusions Our findings suggest that targeting participants to complete activities they find enjoyable will help to improve levels of activation and maintain the effect of the CONEMO intervention in the longer-term. Improving the content of the sessions to facilitate understanding can also help to maintain improvements.