Does HIV/AIDS Funding Undermine Health Systems?

Dear Sir: We read with interest the recent article by Shepard and others, 1 which attempted to evaluate the impact of human immunodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS) funding on Rwanda's health system. The headline of the associated press release is assertive: " Six-Year Study in Rwanda Finds Influx of HIV/AIDS Funding Does Not Undermine Health Care Services for Other Diseases. Study Addresses Long-Standing Debate about Funding Imbalances for Global Diseases. " 2 However, the study's results are far from a definitive answer to the policy question of whether AIDS funding has undermined or enhanced efforts on non-AIDS service provision. The main threat to the validity of this study relates to the assignment of treatment of HIV/AIDS funding to health centers. In 2008, Shepard and others 1 collected retrospective data on the performance of 25 health centers that received HIV/ AIDS funding between 2002 and 2007. These health centers are said to be " randomly selected as the intervention group " by Shepard and others 1 and then " perfectly matched " to 25 other health centers serving as control units. 1 Unfortunately, the intervention group was " randomly " selected only in the sense that Shepard and others 1 chose to study them, not that the health centers in the intervention group were randomly assigned to treatment. Without random assignment and appropriate strategies for causal inference, there is likely to be treatment imbalance error, which arises from observable and unobservable differences between intervention and control centers. The non-random assignment of the intervention group compromises the internal validity of the study and its assertions on the causal inference regarding the effect of the intervention (AIDS funding to health centers) on the outcomes (performance on non-HIV/AIDS service delivery indicators). A serious problem that the work by Shepard and others 1 faces is that it is unclear, if not unknown, why treatment was assigned to those 25 centers in the first place. Even if Shepard and others 1 had enough money to get data for all the HIV/ AIDS health centers and did not have to take a random subset of 25 HIV/AIDS centers, the work by Shepard and others 1 needs to address why these health centers were chosen to receive HIV/AIDS funding to start. For example, it would not be surprising if the government assigned centers to receive HIV/ AIDS funding because those centers were more likely to have better outcomes, …