Adherence to medications, lifestyle advice, and self-monitoring for type 2 diabetes and hypertension in sub-Saharan Africa: A systematic review, meta-analysis, and interactive network analysis

Abstract

Background Type 2 diabetes and hypertension are major public health challenges in sub-Saharan Africa, yet treatment control remains poor. Patient adherence to key treatment pillars (medication, lifestyle advice, and self-monitoring) is essential for treatment control, yet adherence across these pillars has not been comprehensively synthesized. We conducted a systematic review and meta-analysis to map adherence to medication, lifestyle advice, and self-monitoring for type 2 diabetes and hypertension in sub-Saharan Africa, and applied a network analysis to identify cross-cutting determinants as priority candidates for future interventional studies. Methods and findings We searched six academic databases, and Google Scholar, for observational and interventional studies in adults with type 2 diabetes or hypertension in sub-Saharan Africa published between January 1, 2004, and May 14, 2026. Risk of bias was assessed using Newcastle–Ottawa Scale and Cochrane RoB 2. We used random-effects models to pool adherence proportions and conducted subgroup analyses and meta-regressions. Cross-cutting determinants were identified using interactive network analysis with noteworthiness scores. We included 312 studies with 108,014 participants from 28 countries. Pooled adherence was 67% (95% CI [60, 73]) for antidiabetic medications, 51% (95% CI [44, 58]) for antihypertensive medications, 44% (95% CI [38, 49]) for dietary recommendations, 42% (95% CI [37, 47]) for physical activity, 85% (95% CI [82, 88]) for alcohol abstinence, 95% (95% CI [94, 96]) for smoking cessation, 18% (95% CI [12, 27]) for glucose monitoring, and 28% (95% CI [16, 45]) for blood pressure monitoring. Overall self-care adherence was 37% (95% CI [30, 46]) for type 2 diabetes and 35% (95% CI [29, 42]) for hypertension. Heterogeneity was high ( I 2 > 95%, p < 0.001 throughout). Education, self-efficacy, and social support emerged as cross-cutting determinants most consistently associated with adherence. A key limitation is the high statistical heterogeneity, which persisted despite random-effects modeling and subgroup analyses. Conclusions Adherence across treatment pillars is suboptimal. Education, self-efficacy, and social support represent priority candidates for future interventional studies aimed at improving adherence. The review was registered with PROSPERO (CRD42024626793).