Objective: Observational studies have associated use of intramuscular injectable depot medroxyprogesterone acetate (DMPA-IM) with increased risk of HIV-1 acquisition, but limited data are available to assess HIV-1 risk for alternate contraceptive methods.
Methods: Within a randomized trial of the dapivirine vaginal ring for HIV-1 prevention, we assessed HIV-1 incidence by contraceptive method. We limited analyses to participants from South African sites and to women who used DMPA-IM, the alternative injectable norethisterone enanthate, implants, or copper intrauterine devices (IUDs). Contraceptive method was assessed as a time-dependent exposure and multivariate models adjusted for trial randomization arm, age, sexual behaviour, and incident sexually transmitted infections.
Results: A total of 95 incident HIV-1 infections were observed: incidence 5.8 (DMPA-IM, n = 52), 6.2 (norethisterone enanthate, n = 28), 1.9 (implant, n = 3), and 4.5 (IUD, n = 12) cases per 100 woman-years. In multivariable models, there were no statistically significant differences between contraceptive methods in the risk of HIV-1 acquisition. However, compared with the IUD, the three hormonal methods each had point estimates near 1 while the implant had risk that was approximately half that of the IUD. When the three hormonal methods were combined, their relative risk compared with IUD was 0.90 (95% confidence interval 0.45-1.76).
Conclusion: Among women at risk of HIV-1 infections in South Africa, we found no statistically significant differences in HIV-1 incidence by contraceptive method. Implants had the lowest point estimate for HIV-1 incidence, and IUDs had risk comparable with injectable methods in multivariate models. Large, prospective studies are needed to define better the relative HIV-1 risks across different contraceptive methods.
Trial registration: ClinicalTrials.gov NCT01617096.