When “digital works” isn't enough: Cultural fit, engagement, and the real-world efficacy of E-mental health in low and middle-income countries

Low-and middle-income countries (LMICs) bear a disproportionate burden of mental disorders yet continue to face substantial treatment gaps due to limited service capacity, maldistribution of the workforce, stigma, and systemic barriers; consequently, e-mental health has often been positioned as a scalable solution. However, accumulating evidence indicates that the transition from efficacy in controlled trials to effectiveness in routine settings is far from automatic, largely because engagement declines and dropout increases during real-world implementation. This article presents a critical narrative review to explain why claims that “digital works” are insufficient as a policy foundation for e-mental health in LMICs, foregrounding cultural fit, user engagement, and structural determinants as central drivers of real-world value. Evidence was identified through targeted searches of PubMed and Scopus, complemented by backward and forward citation tracking, and interpreted through an implementation science lens (e.g., acceptability, adoption, feasibility, fidelity, and sustainability) alongside process evaluations of complex interventions. We propose a causal chain in which the depth of cultural adaptation (surface–structural–deep) shapes acceptability, trust, and cultural credibility, which in turn influence engagement and therapeutic dose, ultimately determining clinical outcomes and the plausibility of effectiveness at scale. The framework also highlights LMIC-relevant moderators—including digital inequality, stigma and privacy risks, family and community contexts, and health-system support—that strengthen or weaken each link in the pathway. The implications are that engagement and human support (e.g., blended care and digital navigators) should be treated as core design and implementation components; cultural fit must be operationalised as a measurable design target; and evaluation should shift toward asking “for whom, under what conditions, through which mechanisms, and with what system supports” digital interventions produce sustained benefits.