C5e Case-Crossover Analysis of Condom Use and HSV-2 Acquisition

Wednesday, March 10, 2010: 11:30 AM
Cottonwood (M1) (Omni Hotel)
Jeffrey Stanaway, MPH1, Amalia Magaret, PhD1, Emily Martin, PhD, MPH2 and Anna Wald, MD, MPH3, 1Virology Research Clinic, University of Washington, Seattle, WA, 2Children's Hospital Research Institute, Seattle, WA, 3Department of Medicine, Division of Allergy and Infectious Diseases, University of Washington, Seattle, WA

Background: Studies suggest condoms are moderately effective in preventing HSV-2 acquisition, but findings are limited by few HSV-2 events and potential confounding by sexual behavior.

Objectives: We performed a case-crossover analysis to evaluate the effect of condom use on HSV-2 transmission using pooled data from 6 earlier HSV-2 acquisition studies, and compared our findings to a cohort analysis of the same data.

Methods: We summarized HSV-2 acquisition and sexual behavior from prospective studies that had laboratory documentation of HSV-2 status at enrollment and follow-up, and self-reported sexual behavior. Only those who acquired HSV-2 during follow-up were included in the case-crossover analysis. Case periods included the time of likely acquisition (14 to 90 days for serologic diagnosis; 0 to 60 days for culture detection); earlier periods were classified as control periods. Conditional logistic regression was used to assess differences in sexual behaviors between case and control periods. Generalized estimating equations were used to evaluate the full cohort.

Results: 191 participants were eligible for inclusion in the case-crossover and 5324 in the full cohort. Case periods contained more unprotected sexual acts than control periods (12.6 and 7.2, respectively; p<0.001), but contained a similar number of protected acts (3.5 and 4.0, respectively; p=0.677). The odds of HSV-2 acquisition increased by 3.6% with each unprotected act, (95% CI: 1.021, 1.052; p<0.001), but did not increase for protected acts (OR=1.008; 95% CI: 0.987, 1.030; p=0.45). In the full cohort analysis, the odds of HSV-2 acquisition increased by 1.3% for each unprotected act (95%CI: 1.008, 1.018; p<.001), but did not increase for protected acts (OR=0.998; 95% CI: 0.988, 1.008, p=0.675).

Conclusions: Condoms protect against HSV-2 transmission. The case-crossover analysis found a stronger protective effect of condoms than the cohort analysis.

Implications for Programs, Policy, and/or Research: Condoms should continue to be recommended to reduce HSV-2 transmission.