Process evaluation of digital mental health interventions for psychosis: A mixed–methods systematic review with framework synthesis. (Preprint)

BACKGROUND Digital mental health interventions (DMHI) are a promising treatment mode for those with psychosis, though their implementation is a persistent challenge. A process evaluation, or studies conducted alongside trials, is one research method that may address this issue. However, a synthesis of process evaluation data in this literature is missing. OBJECTIVE To review and synthesize process evaluation data from trials evaluating DMHIs for people with psychosis to explore what is known about context, implementation, and mechanisms of impact. METHODS A mixed–method systematic review using a two–phase search strategy underpinned by the Medical Research Council (MRC) process evaluation framework was conducted. Database searches of CENTRAL and PsycInfo first identified an index sample of peer–reviewed trials evaluating a relevant DMHI that was predominantly conducted in the United Kingdom (50≥% United Kingdom sample in multi–country studies). Associated papers linked to this index sample were then systematically searched for and included if process evaluation data as operationalized in the MRC framework were reported. Two authors independently screened references, extracted summary data and assessed the quality of index trials. One author qualitatively analyzed process evaluation data using a framework synthesis approach using the MRC framework and synthesized findings using narrative synthesis. RESULTS Searches identified 10 DMHIs and 36 papers reporting process evaluation data, though only two were labelled as such. Studies indicated most DMHIs supported users to respond to challenging situations in more helpful ways by changing their habitual responses, indicating a possible mechanism of impact. Implementation strategies aimed to enhance delivery in a trial by preparing delivery staff, providing support to users and developing interventions that were flexible and relevant to users. Implementation data also demonstrated DMHI use varied, and contextual factors affecting implementation often differed from stakeholder expectations and included users’ perceptions of the DMHI, life commitments and some psychosis symptoms, and staff availability in services for blended DMHIs. CONCLUSIONS Our findings highlight possible considerations for future DMHI development, delivery and evaluation. Intervention personalization and user–centered design may be important DMHI design features. Further work establishing how DMHIs work is recommended. Implementation efforts should address actual, rather than perceived, determinants of DMHI implementation, including users’ clinical and general context, and staff availability for blended interventions. Application of process evaluation methods in trials could strengthen this work. CLINICALTRIAL PROSPERO: CRD42024439117.