Developing a multi-level human-centered design framework for digital health implementation: a cross-project methodological synthesis

Background Digital health interventions have considerable potential to improve healthcare access, quality, and efficiency. Despite increasing adoption, many interventions fail to achieve sustainable implementation or progress beyond pilot phases because they are insufficiently integrated with user needs, organizational workflows, and health system contexts. Although human-centered design (HCD) has improved the usability and acceptability of digital health technologies, its application frequently remains confined to early design activities and is rarely integrated with implementation science throughout the implementation lifecycle. Objective This study aimed to develop a multi-level, human-centered design framework informed by a cross-project methodological synthesis integrating participatory design and implementation science to support the design, implementation, evaluation, and scale-up of digital health interventions. Methods A cross-project methodological synthesis was conducted using complementary digital health implementation studies undertaken in Kenya. The empirical corpus included studies spanning adolescent mental health, neonatal clinical decision support, digital health infrastructure, generative artificial intelligence, and frugal digital health innovation. Framework development followed an iterative analytical process combining thematic synthesis, constant comparative analysis, and abductive reasoning. The synthesis organized recurring methodological themes into an integrated framework spanning user, workflow, and system perspectives across the digital health implementation lifecycle. Results The synthesis identified recurring methodological processes that consistently contributed to successful digital health implementation, including contextual understanding, participatory stakeholder engagement, iterative refinement, workflow integration, organizational readiness, governance, and sustainability. These findings informed the development of a multi-level framework structured around six iterative implementation phases: Understand, Co-design, Prototype, Evaluate, Implement, and Scale, and three interconnected analytical levels: user (micro), workflow (meso), and system (macro). The framework conceptualizes digital health implementation as a continuous sociotechnical process in which sustained adoption and scale-up depend upon coordinated alignment among these three levels throughout the implementation lifecycle. Conclusions This study advances human-centered design by integrating it with implementation science through an empirically derived, lifecycle-oriented methodological framework for digital health implementation. The proposed framework may provide researchers, developers, practitioners, and policymakers with a structured approach for designing, implementing, evaluating, and scaling digital health interventions while supporting sustainable, context-responsive, and equitable digital health transformation.