Reduced fecundity in HIV-positive women.

Abstract

Is HIV associated with increased time to pregnancy (TTP)? HIV-positive women who discontinue a contraceptive method to become pregnant have increased TTP, particularly among those who likely do not know their status. HIV-positive women have fewer children on average than their HIV-negative counterparts due to both behavioral and biological factors. There is a need to better describe and quantify fecundity patterns associated with HIV in the general population. This cross-sectional study was based on data from 12 Demographic and Health Surveys (DHSs) conducted between 2003 and 2013 in 11 African countries. All studies collected dried blood spot samples for HIV testing and included a retrospective calendar module that recorded women's monthly reproductive status in the 5 years preceding the survey. TTP was measured among 3181 women discontinuing a contraceptive method within 2 years of the survey in order to become pregnant. We use Cox proportional hazard models for discrete survival data to model TTP and estimate fecundability odds ratios (FOR) and 95% CIs for the 12-month period following contraceptive discontinuation. In addition to employing a binary measure of HIV status, we also develop an additional explanatory measure that combines HIV status with information on whether respondents had ever been tested for HIV and received their results (which proxies for knowledge of HIV status) to reduce the threat of confounding from behavioral changes following an HIV diagnosis. In our sample, 10.3% of women were HIV-positive, and a little more than half (51.8%) of women received test results and likely knew their status. Over a 12-month observation period, HIV-positive women had a 25% average reduction in fecundity compared to HIV-negative women [adjusted FOR (aFOR) = 0.75 (0.62-0.92)] after adjusting for confounders. The 12-month fecundity patterns differed by women's likely knowledge of their status such that results were primarily driven by HIV-positive women who likely did not know their status. Moreover, reductions in fecundity attributable to HIV were not uniform over time. Among women who were still trying for pregnancy after 3 unsuccessful months, HIV-positive women had half the odds of becoming pregnant compared to HIV-negative women [aFOR = 0.50 (0.35-0.71)]. Conversely, there were no significant differences in FORs between HIV-negative and HIV-positive women in the first 3 months. Because dried blood spot samples for HIV testing were collected at the time of the survey but reproductive calendar data were collected retrospectively, it is possible that we introduced misclassification bias, as we have no knowledge if the acquisition of HIV occurred before or after pregnancy attempt. As life expectancy and quality health status improve due to earlier initiation of antiretroviral (ARV) treatment in HIV-positive women, there has been growing awareness that services should also address the fertility desires of HIV-positive women who want children. These findings indicate that if a pregnancy does not occur after 3 months of attempting pregnancy, HIV-positive women and HIV-discordant couples should request access to HIV and reproductive pre-pregnancy counseling and health assessments. A.G. was supported by the National Institutes of Health (contract T32-HD007275) during the study. During the conceptualization, data collection and analysis time frame, S.vdP. was supported by WHO/RHR/HRP Special Program in Reproductive Health and Research, Geneva, Switzerland, and HRP (the UNDP/UNFPA/UNICEF/WHO/World Bank Special Programme of Research, Development and Research Training in Human Reproduction). The authors have no conflicts of interest to declare.