Background: Daily prophylaxis with trimethoprim-sulfamethoxazole (cotrimoxazole) by persons with HIV reduces morbidity and mortality and is recommended by Joint United Nations Program on HIV/AIDS and World Health Organization (WHO), but there are limited published cost-effectiveness data for this intervention. We assessed the cost-effectiveness of cotrimoxazole prophylaxis for persons living with HIV in rural Uganda.
Methods: We modeled the cost-effectiveness of daily cotrimoxazole prophylaxis based on clinical results and operational data from a prospective cohort study of home-based care delivery to adults and children with HIV in rural Uganda who were older than the age of 5 years. Main outcome measures were net program cost and disability-adjusted life-years (DALYs) gained. We examined the provision of cotrimoxazole prophylaxis for (A) all HIV-infected individuals regardless of immunologic or clinical criteria; (B) those with WHO stage 2 or more advanced disease; (C) those with CD4 cell counts <500 cells/microL; and (D) those meeting criteria B or C, the current WHO recommendation. We calculated the costs and effectiveness of these 4 screening algorithms compared with no cotrimoxazole prophylaxis and calculated incremental cost-effectiveness ratios. We performed univariate and multivariate sensitivity analyses.
Results: Cotrimoxazole prophylaxis for all HIV-infected individuals (algorithm A) produced 7.3 life-years and 7.55 DALYs per 100 persons over 1 year compared with no prophylaxis. Using this screening algorithm, the intervention saved $2.50 per person-year. The program costs and the DALYs gained by algorithms A, B, and D were more favorable than those for algorithm C. Among algorithms A, B, and D, strategies using screening algorithms for WHO stage or CD4 cell counts were more costly and marginally less effective than providing cotrimoxazole prophylaxis to all HIV-infected individuals.
Conclusions: Daily cotrimoxazole prophylaxis for HIV-infected individuals is highly cost-effective in rural Uganda. The use of screening algorithms to identify individuals with advanced HIV disease may result in higher program costs and less favorable cost-effectiveness.