Use of mathematical modelling to estimate the impact of interventions and risk factors on cardiovascular diseases in Africa: A scoping review

Source abstract

BackgroundAchieving health-related Sustainable Development Goals (SDGs) in Africa is increasingly threatened by the rising burden of cardiovascular diseases (CVD), exacerbated by limited healthcare infrastructure and constrained resources. Mathematical modellingoffers a powerful tool for informing CVD prevention and control strategies; however, its application in Africa remains poorly understood. This scoping review aims to map existing modelling studies, identify methodological and contextual gaps in the literature,and inform future model development for effective decision-making.MethodsThis scoping review followed Levac et al.’s framework and PRISMA-ScR guidelines to explore how mathematical models have been used to study CVD in Africa. We included studies published since 2000 that used structured models to simulate how CVD progresses and responds to risk factors or interventions. Our search, conducted in PubMed and Google Scholar, focused on models relevant to local settings and policy. Data were extracted and analysed thematically. As the review used only publicly available data, no ethical approval was needed. The protocol was registered with the Open Science Foundation.ResultsCVD modelling studies in Africa span nearly two decades, with a geographic concentration in South Africa, Nigeria, and Kenya, while many low-resource and fragile settings remain underrepresented. A range of modelling approaches were used, including decision- analytic models, microsimulation, and WHO-CHOICE frameworks, with recent studies increasingly incorporating dynamic and stochastic elements. Interventions modelled included pharmacological treatments (e.g., statins, antihypertensives), lifestyle modifications (e.g., salt reduction, increased physical activity), and community-level prevention strategies. Many models assessed combinations of interventions and accounted for multiple risk factors. However, limitations in local data availability led to reliance on extrapolated estimates, assumptions about treatment effects, and expert-informed parameterization. Few studies calibrated models to local epidemiological data or disaggregated outcomes by equity dimensions such as gender, income, or geography.ConclusionWhile CVD modelling in Africa has progressed, its utility for informing real-world policy remains limited by insufficient incorporation of health system constraints and equity considerations. To maximize impact, future efforts should prioritize the development of context-specific microsimulation frameworks that integrate system dynamics, reflect local implementation realities, and use countryspecificdata for calibration.