Discontinuation of oral contraceptives and depot medroxyprogesterone acetate among women with and without HIV in Uganda, Zimbabwe and Thailand.

Background: We examined hormonal contraceptive (HC) discontinuation and factors associated with discontinuation among HIV-uninfected women and the effect of HIV diagnosis on subsequent contraceptive use.

Study Design: We analyzed 4461 HIV-uninfected women from a prospective study of HC and HIV acquisition in Uganda, Zimbabwe and Thailand. Participants were ages 18-35 years, not pregnant, and using oral contraceptives (OCs) or injectable depot medroxyprogesterone acetate (DMPA) for at least 3 months before enrollment (median duration of OC and DMPA use before enrollment was 11.7 and 8.9 months, respectively). We compared the probability of OC and DMPA discontinuation using survival analysis and factors related to discontinuation using Cox regression. We also analyzed contraceptive patterns among 194 women who became infected with HIV.

Results: Median duration of use after study enrollment was 15.6 months for OCs and 18.5 months for DMPA. Continuation rates for both methods were highest in Thailand. Factors associated with OC discontinuation included, nausea, breast tenderness, condom use, and no sex. Factors associated with DMPA discontinuation included young age, breast tenderness, nausea, irregular bleeding, high-risk sexual behaviors, partner risk, condom use, and no sex. Following an HIV diagnosis, 135 (98.5%) of 137 hormonal users continued HC and 14 (25%) of 57 nonusers began using HC.

Conclusions: Contraceptive continuation for OCs and DMPA was relatively high over 2 years. Young women, those reporting side effects, and those using condoms are more likely to discontinue and need ongoing contraceptive counseling. Many women receiving HIV-positive diagnoses desire effective contraception.