Contexts and Mechanisms Related to the Efficacy of Digital Medication Adherence Interventions to Support Medication Adherence Among Adults With Chronic Diseases: Realist Review.

Background: Digital medication adherence interventions for individuals diagnosed with chronic conditions have been developed, but their long-term efficacy in improving medication adherence has been limited. Exploring mechanisms and contexts related to outcomes of digital medication adherence interventions is important for these to be effectively tailored for different populations and contexts. A realist review was conducted to assess which intervention components might work for whom, and under what circumstances. Methods: A realist review of literature published between 2002-2024 was conducted. Studies reporting digital medication adherence interventions for people diagnosed with chronic diseases and experiencing unintentional medication nonadherence issues were included. MEDLINE, CINAHL, PsycInfo, Web of Science, Scopus, Embase, and gray literature databases, Overton, and Policy Commons were searched. Data regarding contexts, mechanisms, and outcomes were extracted and synthesized into program theories. Program theories from the realist review were triangulated with findings from two rounds of workshops (three workshops in each round), in which context-mechanism-outcome (CMO) configurations were cocreated with various stakeholders of a project aimed at implementing a digital medication adherence intervention to reduce unintentional nonadherence among people with chronic diseases. There were 12 participants in each round. Results: The analysis of 83 papers and content of the cocreation workshops led to 35 CMOs theories in seven theory areas, which were (1) building and maintaining medication intake habits; (2) solving medication-related barriers to adherence; (3) enabling collection of prescription medications on time; (4) supporting people with unpredictable health conditions and sensory or motor impairments; (5) supporting individuals with high anxiety, low medication self-efficacy, and low social support; (6) supporting individuals who are uncomfortable with technology; and (7) economic, policy, and organizational factors affecting implementation of smart medication devices. Fourteen CMOs came from both the literature and workshops, 7 CMOs from the literature alone, and 14 CMOs from the workshops alone. Conclusions: Analysis of the program theories suggested a range of intervention components considering contextual factors that may improve medication adherence of individuals. These were used to form recommendations for intervention developers and implementers. Further studies are required on policy and economic factors affecting large-scale implementation of digital medication adherence interventions in different settings. Future interventions should report intervention content and delivery in detail and codevelop effective implementation strategies with professionals who can support the implementation of the intervention.