Countries with a Reputation for Strong Leadership on AIDS Policies Usually Deserve Their Standing

P. Doskoch

First published online:

Countries that have a good reputation for leadership regarding AIDS policies generally deserve those reputations, according to an analysis of data from 82 nations.1 Such countries, the study reveals, tend to do a better job of providing highly active antiretroviral therapy (HAART) to their citizens than might be expected given their resources, political structure, healthcare system and other factors. For example, in Brazil, Mexico and Thailand—all of which have a reputation for strong leadership on AIDS prevention and treatment—the proportion of people who receive the antiretroviral therapy they need exceeds by 44–56 percentage points the level one might expect on the basis of economic and institutional constraints. Conversely, HAART coverage in South Africa, whose government has repeatedly espoused controversial AIDS policies, falls far short of predicted levels, given the country's resources.

Assessing whether countries with good reputations on AIDS leadership (e.g., Brazil, Mexico, Thailand, Botswana and Cambodia) and those with less stellar reputations (e.g., South Africa, Russia, Ukraine, China and Zimbabwe) deserve such status poses several challenges, not least of which is choosing appropriate criteria. The author of this analysis argues that leadership should be considered in the context of what is possible and reasonable to expect. Therefore, a country was judged to have good leadership if the proportion of people needing HAART in December 2006 who were receiving it was at least 15 percentage points higher than the proportion one might expect on the basis of the country's characteristics. Several of these characteristics concerned a country's resources, including per capita income, whether the country is a focus country in the U.S. President's Emergency Plan for AIDS Relief (PEPFAR) program and whether the country had received a first-round Global Fund grant. Other variables were related to the severity and distribution of the country's HIV epidemic (e.g., prevalence of HIV, proportion of HIV-positive people living in urban areas), the state of the country's health care system (e.g., proportion of births attended by skilled health professionals) and the country's political structure (e.g., whether it is an established democracy). All data were obtained from the World Health Organization, the World Bank and the Joint United Nations Programme on HIV/AIDS.

The analysis was conducted for 82 developing countries affected by AIDS. Associations between country characteristics (most of which were measured in 2005) and HAART coverage were assessed using regression analysis. The researcher supplemented the main regression with two additional models that served as sensitivity analyses. One added language fractionalization to the main model, on the assumption that countries where many languages are spoken may have greater difficulty providing HAART coverage; the other model added this plus region (Latin America/Caribbean, Southern Africa or West Africa) to the analysis. Finally, an additional regression, mirroring the main HAART analysis, examined countries' coverage of treatment to prevent mother-to-child transmission of HIV.

Overall, four variables were related to provision of antiretroviral therapy. On average, HAART coverage increased by 67% if a country was a PEPFAR focus country, and by 55% if it was an established democracy. In addition, coverage increased by 0.2% for every percentage point increase in HIV prevalence and by 2.5% for every percentage point increase in the proportion of the HIV-positive population that lived in urban areas.

In 11 countries, HAART coverage surpassed the expected level by at least 15 percentage points in all three regression models. Of these, the countries that exceeded expectations by the greatest degree in the main analysis were Cambodia (by 59 points), Mexico (56), Thailand (49), Brazil (44) and Paraguay (42); other countries that consistently scored well (17–27 points above expectations) were Burkina Faso, Costa Rica, Mali, Namibia, Suriname and Uganda. Because of missing data, four countries were not included in the two sensitivity models; however, in the main analysis, HAART coverage substantially exceeded expected levels in two of these countries: Cuba (74) and Rwanda (33).

Another 46 countries had coverage scores that fell within 15 points above or below their expected values in all three HAART models, and were thus considered performing as expected. Among these countries was Botswana, which has universal HAART coverage. Results across models were inconsistent for 18 countries. Latvia was particular striking in this regard: Its score in the main model was the lowest of any country (55 points below expectations), but the deficit was negligible (7 points) in the sensitivity model that included both language fractionalization and region. Finally, three countries fell substantially short of expectations in all three models: South Africa (by 36 points in the main model), Trinidad and Tobago (28) and Uruguay (20).

HAART coverage did not necessarily correlate with coverage of treatment to prevent mother-to-child transmission of HIV. For example, Russia and Ukraine both scored below expectations in at least one of the HAART models, but scored 68–75 points above expected levels in preventing mother-to-child transmission.

The researcher notes that HAART coverage is not a direct measure of political leadership, and it is only one of several possible indicators of a country's policy responses to the AIDS crisis. Nonetheless, the results suggest that countries such as Brazil, Cambodia, Mexico, Namibia, Thailand and Uganda—and probably Cuba and Rwanda—have "performed better than expected given their institutional characteristics, demographic challenges and level of development," and as such "their reputation as poster children for good AIDS leadership is probably well deserved." Several other countries with high scores do not have reputations as leaders, possibly indicating that policies and practices such as facilitating the importation of generic HAART drugs (Burkina Faso) and ensuring an efficient, well-organized health system (Suriname) have been underappreciated.

Although Russia and Ukraine, which have negative reputations, did not meet the criteria for poor leadership, their high scores on preventing mother-to-child transmission suggests that those countries are focusing more on helping pregnant women than on providing antiretroviral medications to the intravenous drug users who account for most of their AIDS cases. South Africa's HAART scores, on the other hand, are consistent with the country's "reputation for poor AIDS leadership" and suggests that ideological reasons, rather than economic or institutional constraints, have limited HAART coverage.

—P. Doskoch


1. Nattrass N, Are country reputations for good and bad leadership on AIDS deserved? An exploratory quantitative analysis, Journal of Public Health, 2008, 30(4):398–406.