Implications of scaling up cardiovascular disease treatment in South Africa : a microsimulation and cost-effectiveness analysis

Background: Cardiovascular diseases and their risk factors-particularly hypertension, dyslipidaemia, and diabetes-have become an increasing concern for middle-income countries. Using newly available, nationally representative data, we assessed how cardiovascular risk factors are distributed across subpopulations within South Africa and identified which cardiovascular treatments should be prioritised. Methods: We created a demographically representative simulated population for South Africa and used data from 17 743 respondents aged 15 years or older of the 2012 South African National Health and Nutrition Examination Survey (SANHANES) to assign information on cardiovascular risk factors to each member of the simulated population. We created a microsimulation model to estimate the health and economic implications of two globally recognised treatment recommendations: WHO's package of essential non-communicable disease interventions (PEN) and South Africa's Primary Care 101 (SA PC 101) guidelines. The primary outcome was total disability-adjusted life-years (DALYs) averted through treatment of all cardiovascular disease or microvascular type 2 diabetes complications per 1000 population. We compared outcomes at the aspirational level of achieving access to treatment among 70% of the population. Findings: Based on the SANHANES data, South Africans had a high prevalence of hypertension (24.8%), dyslipidaemia (17.5%), and diabetes (15.3%). Prevalence was disproportionately high and treatment low among male, black, and poor populations. Our simulated population experienced a burden of 40.0 DALYs (95% CI 29.5-52.0) per 1000 population per year from cardiovascular disease or type 2 diabetes complications at current treatment levels, which lowered to 32.9 DALYs (24.4-44.7) under WHO PEN implementation and to 32.5 (24.4-44.8) under SA PC 101 implementation. Under both guidelines, there were increases in blood pressure treatment (4.2 percentage points under WHO PEN vs 12.6 percentage points under SA PC 101), lipid treatment (16.0 vs 14.9), and glucose control medications (1.2 vs 0.6). The incremental cost-effectiveness of implementing SA PC 101 over current treatment would be a saving of US$ 24 902 (95% CI 14 666-62 579) per DALY averted compared with a saving of $ 17 587 (1840-42 589) under WHO PEN guidelines. Interpretation: Cardiovascular risk factors are common and disproportionate among disadvantaged populations in South Africa. Treatment with blood pressure agents and statins might need greater prioritisation than blood glucose therapies, which contrasts with observed treatment levels despite a lower monthly cost of blood pressure or statin treatment than of sulfonylurea or insulin treatment. Funding: Stanford University.