Eniko Akom1,3, Daniel Choi1,3, Ezra Musingye2, Anne Nakirijja2, Elizabeth Lee3, Francis Kiweewa2, Monica Millard4, Hannah Kibuuka2

1Henry Jackson Foundation, Bethesda, US; 2Makerere University Walter Reed Program, Kampala, Uganda; 3U.S. Military HIV Research Program, Walter Reed Army Institute of Research, Silver Spring, MD. 4U.S. Military HIV Research Program, Walter Reed Army Institute of Research, Silver Spring, MD


Fishing communities, described as residents of a geographic area whose primary economic activities are tied to fishing, have high risk of acquiring HIV infection. The Ugandan Ministry of Health identifies these communities as priority populations for intensified provision of HIV prevention, care and treatment services. As part of a PEPFAR-funded, multi-phase study exploring optimized ways to deliver HIV services on Koome and Buvuma Islands, we estimated the prevalence of HIV and determined factors associated with infection.


The study was a prospective cohort study that enrolled consenting adult residents of 3008 randomly selected households on the Islands of Koome and Buvuma, to monitor new HIV infections and sexual behavior. The first round of data collection took place between April 2017 and January 2018. Participants were interviewed using a structured questionnaire that collected sociodemographic, economic, and behavioral data, and donated blood for HIV, syphilis, and Hepatitis B testing. Testing was conducted following national testing algorithms; for syphilis testing Rapid Plasma Reagin (RPR) was used. Data management and entry utilized REDCAP and STATA 14.2 was used for analysis. Pearson correlation coefficients, chi-squared and t-tests were used to identify potential predictors of syphilis prevalence. Variables included in the multiple logistic regression model were chosen based on a priori knowledge and significant predictors from the bivariate analysis. Stepwise backward selection (threshold at p-value <0.05) starting with the full multivariate model was employed to assess the most important predictors for the outcome. The study protocol was approved by the Makerere University and WRAIR IRBs as well as the Uganda National Council for Science and Technology. 


Among 3863 individuals initially enrolled, the prevalence of syphilis was 10.8%, or 9.6% among males and 11.9% among females. In multivariate analysis, the following variables were significantly associated with syphilis positivity: female gender [aOR:1.32; 95% CI:1.03, 1.69]; location on Koome Island [aOR:1.73; 95% CI:1.36,2.22]; age over 24 years [aOR:1.51; 95% CI:1.04, 2.20]; having been married more than once [aOR:1.52; 95% CI:1.14, 2.02]; having used condoms at first sexual intercourse [aOR:0.66; 95% CI:0.46, 0.99]; alcohol and illegal drug use [aOR:2.05; 95% CI:1.05, 3.99]; and perfect HIV knowledge [aOR:0.72; 95% CI:0.54, 0.94]. HIV positivity [aOR:1.52; 95% CI:1.14, 2.03] and Hepatitis B antigen positivity [aOR:2.10; 95% CI:1.31,3.36] were also independently associated with syphilis infection. Occupation, current marital status, and migrant status were significantly associated with syphilis infection in bivariate analysis, but the association did not persist when controlling for potential confounding factors.


Residents of Koome and Buvuma islands have a high prevalence of syphilis infection, suggesting need for focused scale up of sexually transmitted infection prevention and testing services, with emphasis on women aged 25 and older, people married more than once, and those who use alcohol and drugs. The association with HIV infection suggests that HIV-infected individuals might benefit from STI screening as part of the ART clinic service package. Of particular interest, the finding that education on HIV might also help beneficiaries protect themselves from syphilis and potentially other STIs warrants further study. 

34 IUSTI Congress - European Congres on Sexually transmitted Infections and HIV/AIDS
Bucharest, September 3-5,