Optimizing a Novel Smartphone App to Prevent Postpartum Depression Adapted From an Evidence-Based Cognitive Behavioral Therapy Program: Qualitative Study.

Background: Low-income pregnant patients are at high risk of postpartum depression (PPD). Mothers and Babies (MB) is a cognitive behavioral therapy-based program that prevents up to 50% of de novo PPD when provided in person to low-income Spanish- and English-speaking people who are pregnant without depression. MB is limited by the need for trained personnel to support it. Transforming MB into a smartphone app may mitigate this key barrier. Objective: We aimed to use qualitative data from target end users to create and optimize MBapp, a novel app centered on the MB program. Methods: Draft wireframes of MBapp were created in English and Spanish with cognitive behavioral therapy-based modules adapted from MB. These wireframes included several features shown previously to sustain app engagement: (1) push notifications delivered at participant-preferred times; (2) text-, graphic-, and video-based content; and (3) gamification with digital rewards for app engagement. English- or Spanish-speaking individuals with public health insurance who were between 32 weeks gestation and 6 months post partum and owned smartphones were eligible to consent for individual in-depth interviews. Individuals with prior or current depression were excluded. Interviews were recorded, transcribed, and analyzed using deductive and inductive codes to characterize opinions about MBapp and perceptions of challenges and facilitators of use of MBapp or other perinatal or mental health apps. End user feedback led to major modifications to the wireframes. Each of these changes was categorized according to the FRAME (Framework for Modification and Adaptation), an established method of systematically reporting adaptations and modifications to evidence-based interventions via end user feedback. Recruitment ceased with content saturation, defined as 3 successive participants providing only positive feedback on MBapp's wireframe, without further suggestions for improvement. Results: A total of 25 interviews were completed. Participants were racially and ethnically diverse, generally representing our target end user population, and 48% (n=12) of interviews were conducted in Spanish. Participants' suggestions to improve MBapp were categorized within the FRAME as adaptations that improved either content or context to optimize reach, retention, engagement, and fit for end users. Specifically, the following features were added to MBapp secondary to end user feedback: (1) audio narration; (2) "ask a clinician" nonurgent questions; (3) on-demand module summaries accessible upon module completion; and (4) choice to defer assessments and start the next module. Participants also provided insights into features of perinatal or mental health apps they found appealing or unappealing to understand preferences, challenges, and negotiables or nonnegotiables for MBapp. Conclusions: Adapting MBapp to incorporate end users' perspectives optimized our digital PPD prevention intervention, ideally increasing its appeal to future users. Our team's next steps will confirm that MBapp is a feasible, acceptable intervention among English- and Spanish-speaking perinatal people at risk of PPD.