Application of decision analytic modelling to cardiovascular disease prevention in Sub-Saharan Africa: a systematic review

Abstract

Background: This systematic review sought to examine the application of decision analytic models (DAMs) to evaluate cardiovascular disease (CVD) prevention interventions in sub-Saharan Africa (SSA), a region that has experienced an increasing CVD burden in the last two decades.

Methods: We searched seven databases and identified model-based economic evaluations of interventions targeting CVD prevention among adult populations in SSA. All articles were screened by two reviewers, data was extracted, and narrative synthesis was performed. Quality assessment was performed using the Philips checklist.

Results: The review included 27 articles from eight SSA countries. The majority of the studies evaluated interventions for primary CVD prevention, with primordial prevention interventions being the least evaluated. Markov models were the most commonly used modelling method. Seven studies incorporated equity dimensions in the modelling, which were assessed mainly through subgroup analysis. The mean quality score of the papers was 68.9% and most studies reported data challenges while only three studies conducted model validation.

Conclusions: The review finds few studies modelling the impact of interventions targeting primordial prevention and those evaluating equitable strategies for improving access to CVD prevention. There is a need for increased transparency in model building, validation and documentation.

Fig. 1. PRISMA Flow Diagram Depicting the Study Selection Process.

The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flow diagram outlines the study selection process. The numbers show the studies selected or excluded at each step of the study selection.

Fig. 2. Characteristics of the included studies.

A Distribution of studies by country. Each country has a unique colour, which corresponds with the colour of the studies. Brown colour represents multi-country studies. B Distribution of studies by type of prevention and year of publication. Blue colour represents primordial prevention; red for primary prevention; yellow for secondary prevention and green are studies that modelled interventions targeting both primary and secondary prevention. C Distribution of studies by type of intervention and level of prevention. The colour codes represent the different types of interventions-black represents diet interventions, yellow for implementation science interventions, blue for studies modelling multiple interventions, green for pharmacological interventions and yellow for interventions targeting tobacco control.

Fig. 3. Types of interventions modelled.

A A graph characterizing the level of prevention by country. Each colour uniquely represents a level of prevention. B A graph presenting the type of intervention by country. Each colour represents an intervention.

Fig. 4. Characteristics of the decision analytic models.

A A graph presenting the type of evaluation performed. B A graph showing the type of model used. C A graph presenting the study perspective adopted. D A graph presenting the time horizon adopted. E A graph showing the cardiovascular disease (CVD) risk equation used. WHO stands for World Health Organization. F A graph presenting the CVD outcomes modelled.

Fig. 5. Model type by country, prevention and type of intervention.

A A graph presenting the distribution of model types by country. Each colour is unique to a model type. B A graph showing the model type by the level of prevention modelled. Each colour is unique to a model type. C A graph presenting the model type by intervention modelled. Each colour is unique to a model type.