HIV incidence and associated risk factors among Peruvian male participants of the NIMH collaborative HIV/STD prevention trial

Published: July 21, 2010

HIV incidence and associated risk factors among Peruvian male participants of the NIMH collaborative HIV/STD prevention trial

E. Segura1, A. Silva-Santisteban1, S. Leon1, T. Coates2, C. Caceres1

1Universidad Peruana Cayetano Heredia, Lima, Peru, 2University of California at Los Angeles, Los Angeles, CA, United States

Background: In recent years, HIV incidence in Peruvian men who have sex with me (MSM) has been estimated from surveillance data, differential sensitivity EIA and other sources, raising questions of external validity. Based on secondary data from a large cohort, we aimed to estimate HIV incidence, and its associated factors, in urban men (including MSM) in three Peruvian coastal cities (Lima, Trujillo and Chiclayo).

Methods: Methods and outcomes of the overall 5-country NIMH Collaborative Trial have been described elsewhere. In this secondary analysis, only men with consistently negative HIV test results at baseline, and a total of two visits minimum(including baseline) were included. Incident HIV infections were defined as reactive EIA/reactive WB at any visit after negative results in a previous visit. A multilevel Poisson regression model was developed to estimate the HIV incidence rate and explore associations.

Results: The final population included 1888 “Esquineros” (high-risk heterosexually identified men who sometimes have sex with other men) and 421 MSM. HIV incidence rate (per 100 person-years) was 0.24 (0.12-0.48; 95%CI) and 2.2 (1.32 3.68; 95%CI) respectively. In multivariate analysis, none of factors assessed (age, employment, number of sex partners and unprotected anal intercourse-both within last 6 months) were found to be associated with HIV acquisition.

Conclusions: These results are important because: first, they convey the first measurement, high for a concentrated epidemic, of HIV incidence in a male, heterosexually identified, yet socially marginalized population; second, they convey a measurement of HIV incidence among MSM that is lower and arguably more useful that previous ones since

  1. it was based on a more diverse range of MSM rather than solely high risk individuals,
  2. it prospectively assessed HIV status in a cohort, and
  3. it originated from a trial testing a community-level intervention with no statistically significant effect detected as explained elsewhere.


Download the e-Poster

Leave a Reply