Condom use with main and other partners among sterilized and non-sterilized women in the Houston county jail and local residential drug and alcohol treatment centers

Background: Tubal ligation is the most common contraceptive method for poor women, removing one reason for condom use that may leave these women at risk for sexually transmitted infections (STI). Previous studies found sterilized women less likely to report condom-protected sex. As with most STI risk studies, however, they did not control for the nested data structure--individuals can have multiple partner types--a weakness that can increase Type I errors. Objective: We explored the influence of sterility status, partner type, and individual and partner correlates on condom use. Method: We used logistic multi-level modeling to predict any condom use in the 30 days before entering the facility using data from an interview survey of sterilized and non-sterilized women aged 18-44 years, 10-14 days after entering an urban county jail (n=263) or residential drug/alcohol treatment center (n=229). Results: Of 492 pre-menopausal respondents without hysterectomy or oopherectomy, 191 reported tubal ligation. Sterilized women were less likely to report any condom use (OR=0.62, 95% CI: 0.40, 0.97). Women who were black non-Hispanic, in jail, HIV positive, and reported more partners were more likely to use condoms; condom use was more likely with “other” than with “main” partners. Conclusion: Condom use messages have had some success in reaching some high risk subgroups in Houston (HIV positive women, women with more partners). Because jail and treatment centers draw from populations with disproportionate STI risk, interventions promoting condom use should target sterilized women in these settings.