PTSD-Coach App Use in a Mass Trauma Context: A Qualitative Study of Engagement and AI-Driven Feature Integration (Preprint)

BACKGROUND Digital mental health applications are increasingly deployed during large-scale crises to provide scalable psychosocial support. Following the mass-trauma events of October 7, 2023 in Israel, the Hebrew version of PTSD-Coach became a widely accessible tool for individuals seeking support outside formal clinical settings. While prior research has examined adoption intentions, little is known about how users engage with and interpret such applications during the acute and ongoing phases of collective trauma. OBJECTIVE This study explored how adult users experienced, interpreted and adopted the PTSD-Coach during a period of national emergency, and how user-informed recommendations may guide AI-driven personalization in crisis contexts. METHODS A qualitative study was conducted with 12 adult users in two focus groups who had downloaded the Hebrew version of PTSD-Coach during the emergency period. Focus groups examined lived experiences of app use, emotional regulation practices, perceived benefits and limitations, and expectations for AI-driven feature integration. Data were analyzed using an inductive thematic approach with theoretically informed interpretation grounded in medical anthropology frameworks of trauma, ritual, and symbolic care. RESULTS Analysis identified four distinct modes of engagement with the app: Ritual, Sharing, Expert, and Shelter. Through these modes, participants sought structure and routine, social connection, authoritative guidance, and emotional containment. The app functioned as both an individual self-regulation tool and a shared symbolic resource during crisis. Participants also articulated concrete expectations for adaptive features, highlighting the potential value of AI-driven personalization to accommodate changing emotional states, varying literacy levels, and contextual demands. CONCLUSIONS PTSD-Coach functioned not only as a symptom-management tool but as a culturally embedded form of digital care during mass trauma. Engagement mechanisms extended beyond therapeutic content to include ritualization, symbolic containment, institutional trust, and moral citizenship. Integrating user-informed AI personalization may enhance responsiveness, sustain engagement, and support scalable, context-sensitive digital interventions in future collective crises.