A Cost-Effectiveness Analysis of a Program to Control Rheumatic Fever and Rheumatic Heart Disease in Pinar del Rio, Cuba

BACKGROUND: Acute rheumatic fever (ARF) and rheumatic heart disease (RHD) persist in many low- and middle-income countries. To date, the cost-effectiveness of population-based, combined primary and secondary prevention strategies has not been assessed. In the Pinar del Rio province of Cuba, a comprehensive ARF/RHD control program was undertaken over 1986 – 1996. The present study analyzes the cost-effectiveness of this Cuban program. METHODS AND FINDINGS: We developed a decision tree model based on the natural history of ARF/RHD, comparing the costs and effectiveness of the 10-year Cuban program to a “do nothing” approach. Our population of interest was the cohort of children aged 5 – 24 years resident in Pinar del Rio in 1986. We assessed costs and health outcomes over a lifetime horizon, and we took the healthcare system perspective on costs but did not apply a discount rate. We used epidemiologic, clinical, and direct medical cost inputs that were previously collected for publications on the Cuban program. We estimated health gains as disability-adjusted life years (DALYs) averted using standard approaches developed for the Global Burden of Disease studies. Cost-effectiveness acceptability thresholds were defined by one and three times per capita gross domestic product per DALY averted. We also conducted an uncertainty analysis using Monte Carlo simulations and several scenario analyses exploring the impact of alternative assumptions about the program’s effects and costs. We found that, compared to doing nothing, the Cuban program averted 5051 DALYs (1844 per 100,000 school-aged children) and saved $7,848,590 (2010 USD) despite a total program cost of $202,890 over 10 years. In the scenario analyses, the program remained cost saving when a lower level of effectiveness and a reduction in averted years of life lost were assumed. In a worst-case scenario including 20-fold higher costs, the program still had a 100% of being cost-effective and an 85% chance of being cost saving. CONCLUSIONS: A 10-year program to control ARF/RHD in Pinar del Rio, Cuba dramatically reduced morbidity and premature mortality in children and young adults and was cost saving. The results of our analysis were robust to higher program costs and more conservative assumptions about the program’s effectiveness. It is possible that the program’s effectiveness resulted from synergies between primary and secondary prevention strategies. The findings of this study have implications for non-communicable disease policymaking in other resource-limited settings.