User journey facilitates digital intervention development and implementation

Background: Digital mental health interventions often face low behavioral engagement and challenges in translation from research to real-life environments. These problems could be alleviated by evaluating digital interventions as services that consist of numerous elements that the user interacts with over time. We advance here the user journey method, which structures the measurement of such user-service interactions and facilitates identifying intervention-specific usage barriers and constraints.Methods: We applied the user journey method in a clinical trial that investigated the efficacy of a novel game-based intervention for adult Major Depressive Disorder. We modelled the user journey, which included four technological (recruitment, website, questionnaires, intervention software) and two interpersonal elements (assessment, support), and their integration into a service. We operationalized the behavioral engagement with usage data, particularly the number of participants remaining in a given study phase. We measured engagement with six sources of usage data: social media analytics, website usage data, signup data, clinical study coordinator (CSC) interview data, symptom questionnaire data, and behavioral intervention usage data. These measurements were complemented with the qualitative analysis of the study discovery sources and support contacts. Results: The modelled user journey structured the usage data measurement. The study recruitment reached approximately 440 000 people, with social media, word-of-mouth, and news and web sources being the most effective recruitment channels. The study website received 16 243 visitors, which led 1 007 sign ups. 895 participants were evaluated with interview assessment and online questionnaires, which led to 735 accepted participants. 498 participants were assigned to the intervention software, of whom 457 used the intervention at least once: on average, for 17.3 hours (SD = 20.4 h) on 19.7 days (SD = 20.7 d) over a period of 38.9 days (SD = 31.2 d). The 28 intervention levels were associated with an average dropout rate of 2.6%, with two sections exhibiting an increase against this baseline. 150 participants met the minimum adherence goal. 58% of the support contacts concerned a specific intervention element, and 42% the service element integration. Conclusion: We found that user dropout occurred throughout the service, its elements, and phases, and that some users exhibited growing engagement with the service. The user journey method allowed identifying specific service-element-related usage barriers, which expedites iterative intervention development and implementation.