Improving second-generation surveillance: the biological measure of unprotected intercourse using prostate-specific antigen in vaginal secretions of West African women.

Background: Second-generation surveillance for HIV includes measures of high-risk behaviors among the general adult population and sex workers (SW). Questionnaires are prone to social desirability biases because individuals minimize the frequency of behaviors not expected from them.

Objective: Determine whether the prostate-specific antigen (PSA) could be used as a biological marker of unprotected intercourse.

Methods: We measured the presence of PSA in vaginal secretions of women who were (n = 508) or were not (n = 658) SW presenting with vaginal discharge in health facilities of Ghana, Togo, Guinea, and Mali. The cutoff for a positive assay was determined as > or =0.4 microg/L based on a subsample of 95 non-SW claiming abstinence for 3 months.

Results: A positive PSA assay was correlated with infections with Neisseria gonorrhoeae, Chlamydia trachomatis, and Mycoplasma genitalium. Among non-SW, a positive PSA was more common among those with HIV, but less frequent in those better educated. Among SW and non-SW, women from Ghana were less likely to have a positive PSA and had a lower prevalence of sexually transmitted infections than those from elsewhere.

Conclusions: PSA can be used as a biological marker of unprotected intercourse, allowing interventions to target efforts on those at highest risk.