Jennifer Ross, MD, MPH


Nominated From: University of Washington

Research Site: Uganda

Research Area: HIV/AIDS; Malaria

Primary Mentor: Dr. Judd Walson

Research Project

The relationship between increasing use of co-trimoxazole phrophylaxis among people living with HIV in Kenya and Uganda and observed population-level decreases in malaria incidence in the same regions

Co-trimoxazole prophylaxis (CTP) has been recommended by the World Health Organization since 2006 for individuals with HIV infection and CD4 count <350 or WHO clinical stage 2,3, or 4 due to reduction in overall mortality and disease-specific effects including reduced incidence of clinical malaria. Some trials have shown modest protection from diarrheal disease and pneumonia as well, though this effect has not been consistent among studies. National HIV treatment guidelines in Kenya and Uganda call for CTP prophylaxis for all people with HIV regardless of CD4 count, in part to simplify care in settings where CD4 testing has been unavailable. In contrast, in the United States, CTP is commonly discontinued in individuals with CD4 >200. This is because the primary benefit of CTP in the US is prevention of opportunistic infections such as Pneumocystis jiroveci pneumonia and toxoplasmosis that rarely occur at CD4 >200. Benefits to stopping CTP include reduced pill burden and cost, avoidance of adverse drug effects, and reduced pressure on organisms to develop antimicrobial resistance.

In contrast to the situation in the US, it is not clear if HIV-positive people who live in malaria-endemic areas can safely stop CTP even after initiating antiretroviral therapy and reconstituting their immune systems. Recent studies of adults and children with HIV who were randomized to discontinue CTP suffered excess episodes of clinical malaria and malaria requiring hospitalization, despite CD4 >200. These trials were done in areas of Uganda highly endemic for malaria. Similarly, preliminary results from follow-up randomized controlled trials of CTP discontinuation among adults on antiretroviral therapy in areas of Uganda and Kenya that are highly endemic for malaria recently confirmed the increased malaria risk among those who stopped prophylaxis, though complete data have not been published. Based on these studies, the WHO is expected to offer revised guidelines in fall 2014 with further guidance for continuing CTP in settings of varying malaria incidence. We propose the following aims to determine in whom co-trimoxazole prophylaxis can be discontinued without increasing risk for malaria in the individual or surrounding population.

1. Assess the relationship between increasing use of co-trimoxazole prophylaxis among people living with HIV in Kenya and Uganda and observed population-level decreases in malaria incidence in the same regions.

2. Determine a threshold of malaria incidence that warrants continuing co-trimoxazole prophylaxis among HIV-positive individuals with CD4>350.

3. Predict the change in malaria incidence for the general populations of Kenya and Uganda if adults with CD4>350 in areas of low malaria incidence stop taking co-trimoxazole prophylaxis.

4. Assess the attitudes of Kenyan and Ugandan HIV-positive adults, HIV care providers, and policy-makers toward discontinuing co-trimoxazole prophylaxis in areas of low malaria incidence.

Research Significance

Malaria endemicity and clinical burden have been well mapped and modeled across sub-Saharan Africa, but coverage of co-trimoxazole prophylaxis has not been incorporated into these maps or models. Malaria infection declined across East Africa between 2000 and 2010, a time period that corresponded to major increases in malaria prevention efforts as well as increased implementation of CTP. Utilization of CTP varies across nations in sub-Saharan Africa due to limited drug budgets, drug stock-outs, perceived lack of prioritization by funders, and concerns about development of antibiotic resistance. CTP uptake in Kenya and Uganda among those with known HIV has been high. The impact of widespread use of CTP on population-level malaria incidence has not been estimated.




Differences in clinical manifestations among Cryptosporidium species and subtypes in HIV-infected persons.
Investigating Barriers to Tuberculosis Evaluation in Uganda Using Geographic Information Systems.