BackgroundNonsuicidal self-injury (NSSI), hereafter referred to as self-injury, is common among adolescents and is associated with elevated risk for suicide attempts and other adverse mental health outcomes. Because most adolescents who self-injure do not receive formal treatment, accessible alternatives are needed. Digital mental health interventions (DMHIs) may offer an accessible and scalable means of support. While previous research has examined DMHIs for youth, less is known about how adolescents who engage in self-injury understand their experiences and use digital technologies, including emerging tools such as artificial intelligence (AI), to support their mental health. This study explored adolescents’ lived experiences of self-injury, help-seeking behaviors, and technology use for coping and support.MethodsWe conducted semi-structured interviews between May 2024 and April 2025 with 21 adolescents (14–18 years old) with lived experience of self-injury. Recruitment occurred through a mental health screener hosted on a large national mental health nonprofit's website. Interviews explored adolescent experiences with self-injury, barriers to seeking help, and the role of digital tools in their self-injury self-management. Interviews were audio-recorded, transcribed, and analyzed via thematic analysis.ResultsAdolescents described self-injury as a coping strategy embedded within broader emotional and social stressors. While they were aware of the harms, they commonly expressed ambivalence about stopping the behavior. Stigma and concerns about disclosure often limited help-seeking and contributed to their use of digital technologies to access information and support. Participants reported using a range of tools, including music streaming platforms, journaling and mood-tracking apps, online communities, and AI systems such as ChatGPT to regulate emotions, reflect on their experiences, seek advice, and discuss sensitive topics. Although these technologies were valued for their accessibility, privacy, and immediacy, participants also highlighted limitations, including inaccurate or generic AI responses, limited moderation in online communities, and the complex emotional impacts of music.ConclusionOur findings underscore the complex and evolving role digital technologies, including AI, play in adolescents’ experiences of self-injury, coping, and help-seeking. These insights have important implications for the development of future interventions and for clinicians, caregivers, and other health professionals supporting adolescents’ mental health.
People experiencing mental health crises frequently turn to open-ended generative AI (GenAI) chatbots such as ChatGPT for support. However, rather than providing immediate assistance, most GenAI chatbots are designed to respond to crisis situations in ways that minimize their developers' liability, primarily through avoidance (e.g., refusing to engage beyond templated referrals to crisis hotlines). Withholding crisis support in these cases may harm users who have no viable alternatives and reduce their motivation to seek further help. At scale, this avoidant design could undermine population mental health. We propose empowerment-oriented design principles for AI crisis support, informed by community helper models. We outline how, as an initial touchpoint in help-seeking, AI chatbots can act as a supportive bridge to de-escalate crises and connect users to more reliable care. Coordination between AI developers and regulators can enable a better balance of risk mitigation and user empowerment in AI crisis support.
This study examined age-related differences in the use of artificial intelligence (AI) for emotional support using survey data from 3,597 respondents recruited through Mental Health America. Chi-square analyses showed that adoption and frequency of AI-based emotional support varied significantly by age, with adolescents under 18 reporting the highest use and adults aged 25 and older reporting lower-than-expected use. Common motivations among users included convenience, 24/7 availability, comfort talking to a bot, free access, and the ability to seek help without informing family members. Non-users most often cited a preference for human support, distrust of AI advice, and privacy concerns. Findings suggest that AI’s non-human nature can both reduce stigma and increase concerns about trust and authenticity. Human-centered systems should therefore provide age-appropriate safety guardrails, transparent limitations, privacy protections, and clear pathways to human support.
Digital peer-to-peer mental health tools have shown promise in supporting the well-being of those receiving help and giving it (i.e. helper therapy), but promoting engagement remains a challenge. We examine whether the framing of helper therapy exercises motivates active user participation and how user characteristics shape differential effects of the framings in a publicly deployed interactive text messaging-based mental health program. Among 3,817 users randomized to different helper therapy framings, we find causal evidence that framings which emphasize helpng oneself increase written engagement rates as much as 4.6% over other framings, with even larger effects seen among minoritized identities. These self-focused framings also elicited messages with more positive, trust, and anticipation-related words and fewer fear, anger, disgust, and sadness words. Our findings highlight the importance of centering the user in the framing of digital intervention content, and personalizing digital mental health tools to align with a diversity of user identities.
Abstract BackgroundSexual and gender minority (SGM) populations experience significant mental health disparities, yet continue to face persistent barriers to care. Mental Health America (MHA) provides free, web-based screening and self-guided resources to millions of visitors, including tens of thousands of SGM visitors each year. Digital mental health (DMH) can facilitate access to mental health resources; however, engagement remains a central challenge that limits effectiveness. MHA faces similar engagement challenges, with preliminary analyses indicating that most visitors exit the website without accessing substantive mental health content pages. The QT-Digital Mental Health Engagement study uses an iterative approach to develop and test strategies for improving engagement with DMH resources among SGM users. ObjectiveThis study provides the rationale and protocol for an iterative series of microrandomized trials testing theory-based engagement strategies embedded within the web infrastructure of a real-world DMH platform. The primary aim is to evaluate whether delivering engagement strategies increases SGM visitors’ naturalistic engagement with MHA resources. MethodsEligible participants will be MHA visitors who complete the Patient Health Questionnaire-9 depression screening test in English and indicate on an optional postscreening survey that they live in the United States, are aged ≥14 years, and identify as LGBTQ+ (lesbian, gay, bisexual, transgender, queer, and other SGM populations). Participants will be randomized at 2 decision points embedded within the MHA Online Screening Program. At each decision point, participants will be randomized with equal probability (0.2) to receive no engagement strategy or an engagement strategy targeting 1 of 4 Health Action Process Approach behavioral determinants, including outcome expectancy, risk perception, self-efficacy, or SGM-specific barriers to engagement. The primary outcome will be proximal engagement with MHA content, operationalized at decision point 1 as click-through to a targeted Next Steps Content Page and at decision point 2 as click-through to any additional MHA content page. The primary analysis will estimate the marginal causal excursion effect of delivering any engagement strategy compared with the control. ResultsThis study received funding from the National Institute of Mental Health in June 2023 (K01MH131795). The protocol was approved by the single institutional review board at the University of Washington on December 3, 2025 (STUDY00017726). Data collection began on March 12, 2026, and is projected to end in July 2026. As of April 13, 2026, a total of 3878 participants have been enrolled, with 52% (n=2004) of participants identifying as SGM visitors aged 14 to 17 years. ConclusionsThis protocol describes an iterative series of microrandomized trials designed to test theory-based engagement strategies within a real-world DMH platform where disengagement can occur within seconds. Results will identify which strategies show promise for improving naturalistic engagement among MHA’s SGM visitors and inform future refinement of engagement strategies and adaptive decision rules for optimizing engagement with DMH resources.
