Differences in the Access to Sterilization between Women Living and Not Living with HIV: Results from the GENIH Study, Brazil

BACKGROUND In many countries, young women of reproductive age have been especially affected by the HIV epidemic, which have fostered research to better understand how HIV infection influences and shapes women´s fertility and reproductive and sexual decisions. In Brazil, few studies have focused on the impact of the HIV epidemic on contraceptive choices among women living with HIV (WLHIV). OBJECTIVE This study evaluates the impact HIV infection may have in the access to female sterilization in Brazil, using a time-to-event analysis. METHODS A cross-sectional quantitative study (GENIH study) was conducted between February 2013 and April 2014 in the city of São Paulo, comparing two probabilistic samples of 975 WLHIV and 1,003 women not living with HIV (WNLHIV) aged 18 to 49. Sexual and reproductive data was collected retrospectively in order to reconstruct women's reproductive trajectories. Given the objectives of this study, the analysis was restricted to women with parity one or more and, in case of WLHIV, to those sterilized after HIV diagnosis and not infected through vertical transmission. The final sample analysis included 683 WNLHIV and 690 WLHIV. A series of multivariable-adjusted Cox models estimated the probability of being sterilized after HIV diagnosis, compared with WNLHIV. Models were adjusted for schooling, race/color, and stratified by parity at last delivery (1-2, 3+). Hazard ratios were calculated for female sterilization, and separately for interval and postpartum procedures (performed in conjunction with caesarean section or immediately after vaginal delivery). Additionally, information regarding unmet demand for female sterilization was also explored. FINDINGS No statistical difference in the overall risk of sterilization between WLHIV and WNLHIV in the two parity-related groups is observed: HR = 0.88 (0.54-1.43) and 0.94 (0.69-1.29), respectively, among women with 1-2 children and those with three and more. However, significant differences regarding the impact of HIV infection at sterilization are observed depending on the timing and the type of sterilization procedure. The probability of obtaining an interval sterilization is significantly lower for WLHIV compared to those not living with HIV. The reverse occurs regarding postpartum sterilization. Although sterilization is mainly performed in conjunction with caesarean section in Brazil, it is evident that caesarean sections are not the sole factor increasing the risk of sterilization among WLHIV. CONCLUSION The results indicate barriers in the access to services offering interval sterilization for WLHIV and certain facilitation in obtaining the procedure in conjunction with caesarean section. Health policy makers at local and national levels should promote institutional changes in order to facilitate access to interval sterilization and to confront the sensitive discussion of WLHIV's eligibility for postpartum sterilization. It is also urgent to increase access to a wider range of contraceptive methods for WLHIV and promote dual method protection strategies. Moreover, since condom use may decrease in the future in the context of the preventive effect of antiretroviral therapy, promoting dual methods will expand the choices regarding the reproductive rights of women living with HIV.