cal analyses to correlate changes in key outcome or impact indicators with explanatory variables in order to compare countries with variable levels of PEPFAR funding over time. Several key factors contributed to this decision.

One such factor was that there are critical differences among PEPFAR countries in demographics, social and economic factors, the epidemiology of the epidemic, and the availability of appropriate data. Similarly, in order to compare PEPFAR countries, which were not chosen randomly, to non-PEPFAR countries, an analysis would need to account for important differences related to economic, political, and health factors; population sizes; the stage of the epidemic; and available infrastructure and capacity prior to the introduction of PEPFAR. Furthermore, many countries receive some level of PEPFAR investment (see Chapter 1), and where PEPFAR has not been implemented or has been implemented with less intensity, programs with support from other external or national funding sources may have implemented similar interventions to achieve similar objectives. Therefore, it is a critical challenge to identify control countries that can be appropriately compared to PEPFAR countries.

Another important factor was the lack of reliable data across all of the countries of interest for key benchmark indicators, including HIV-related deaths and all-cause mortality, despite intense efforts reviewing multiple sources for these data. Potential sources of mortality data that were considered are described in more detail in Appendix C. In addition, for a number of measures of interest for this evaluation, data are not collected across PEPFAR and non-PEPFAR countries.

Finally, as described in more depth in Chapter 4, complete and reliable data on annual PEPFAR expenditures by country were not readily available. Ideally the committee would have designed a model to determine if a bigger annual investment of PEPFAR funding over time, across all PEPFAR-funded partner countries, had led to a greater impact on health. This would cover a larger scope of countries than prior analyses, and the use of a continuous funding variable as the input to the model rather than the dichotomous comparison of focus versus non-focus countries would address some of the limitations of the existing analyses.

Ultimately, the committee determined that the limitations were too great to design and carry out analyses in the time available that would meaningfully add to the existing analyses in the published literature. Although these limitations prevented the committee from quantitatively modeling the impact of PEPFAR, the many data sources reviewed by the committee and presented in this chapter did make it feasible to conclude that PEPFAR’s support for care and treatment services has had a major positive effect in partner countries.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement