Menu
Menu
Menu
Menu
Background Avoiding unintended pregnancies through family planning is a WHO strategy for preventing mother to child transmission of HIV (PMTCT) and maternal morbidity/mortality. We investigated factors associated with unintended index pregnancy, unmet contraceptive need, future pregnancy intention and current contraceptive use among Malawian women living with HIV in the Option B+ era. Methods Women who tested HIV positive at 4–26 weeks postpartum were enrolled into a cross-sectional study at high-volume Under-5 clinics. Structured baseline interviews included questions on socio-demographics, HIV knowledge, partner’s HIV status/disclosure, ART use, pregnancy intention and contraceptive use. Logistic regression was used to determine factors associated with outcomes. Results We enrolled 578 HIV-positive women between May 2015-May 2016; median maternal age was 28 years (y) (interquartile-range [IQR]: 23–32), median parity was 3 deliveries (IQR: 2–4) and median infant age was 7 weeks (IQR: 6–12). Overall, 41.8% women reported unintended index pregnancy, of whom 35.0% reported unmet contraceptive need and 65.0% contraceptive failure. In multivariable analysis, unintended index pregnancy was higher in 35y vs. 14-24y (adjusted Odds Ratio [aOR]: 2.1, 95% Confidence Interval [95%CI]: 1.0–4.2) and in women with parity 3 vs. primiparous (aOR: 2.9, 95%CI: 1.5–5.6). Unmet contraceptive need at conception was higher in 14-24y vs. 35y (aOR: 4.2, 95%CI: 1.8–9.9), primiparous vs. 3 (aOR: 8.3, 95%CI: 1.8–39.5), and women with a partner of unknown HIV-status (aOR: 2.2, 95%CI: 1.2–4.0). Current contraceptive use was associated with being on ART in previous pregnancy (aOR: 2.5, 95%CI: 1.5–3.9). Conclusions High prevalence of unintended index pregnancy and unmet contraceptive need among HIV-positive women highlight the need for improved access to contraceptives. To help achieve reproductive goals and elimination of MTCT of HIV, integration of family planning into HIV care should be strengthened to ensure women have timely access to a wide range of family planning methods with low failure risk.
This is a cross-sectional study nested within a nationally representative cohort of HIV-positive Malawian women who were enrolled at 4–26 weeks post-partum in the National Evaluation of the Malawi PMTCT Programme (NEMAPP) study. The parent NEMAPP study used a multistage cluster design to randomly select 54 health facilities across Malawi where mother-infant pairs were consecutively consented, interviewed, and screened for HIV; women testing HIV-positive were invited to participate in the cohort study. In this cross sectional sub-study, we enrolled HIV-positive women presenting with their 4-26-week-old exposed infants at the Under-5 clinics of three government health facilities (1 hospital and 2 health centres) that were purposefully selected because of their large patient volumes and representation of urban and rural settings. Women were interviewed at enrolment on socio-demographics, clinical characteristics, index and future pregnancy intentions and contraceptive use by trained health facility staff using structured questionnaires (S1 Appendix). We defined unintended index pregnancy as the pregnancy of a present 4–26 weeks old infant, which was unwanted or mistimed at the time of conception. Unmet contraceptive need was defined as the proportion of women whose index pregnancy was unintended, but did not report contraceptive use at the time of conception. Future pregnancy intention was defined as the desire to have another child within or after 12 months from the time of enrolment. Current contraceptive use was defined as the proportion of women who were fertile, sexually active and were using a method of contraception to stop child bearing or delay pregnancy for the next 12 months. We designed standardized questionnaires to record socio-demographics, clinical characteristics and family planning-related outcomes of study participants. Exposure variables included age, parity, previous child death, education level, religion, HIV test in index pregnancy, HIV result if tested in index pregnancy, disclosure of HIV status to partner, known HIV status of partner, partner’s HIV result if known, ART in previous pregnancy, timing of ART, health status at ART initiation, and health status at enrolment. Data were pooled across the three sites and the site population effects were not modelled. Our sample also included a small group of women who tested positive during enrolment but reported HIV-negative during index pregnancy, suggesting that they seroconverted later during pregnancy or post-partum. Descriptive statistics were used to characterise study participants and estimate the proportion of each outcome, stratified by socio-demographic and clinical characteristics. Exposure variables which had statistically significant 95% confidence intervals (95%CI) Mantel-Haenszel crude odds ratio for at least one comparison group were considered as potential factors for association with study outcome. We included each potential factor in multivariate logistic models using a step-wise forward algorithm at the significance level of a Likelihood Ratio Test, P <0.05 [25]. Final multivariate logistic models controlled for confounders and risk factors which included age, parity, education level, religion, timing of ART, known HIV status of partner, partner’s HIV result if known, HIV result if tested in index pregnancy, ART in previous pregnancy, health status at ART initiation and future pregnancy intention. Any two given variables were also checked for collinearity, and one variable was omitted if its variance inflation factor was greater than five. Analyses were carried out using Stata version 14.0 (StataCorp, College Station, TX, USA). Ethical approval was obtained from the Malawi National Health Sciences Research Committee (NHSRC, #1262), the Centers for Disease Control and Prevention Center for Global Health Associate Director for Science (#2014-054-7) and the University of Toronto Research Ethics Committee (#30448). Participants provided written informed consent to enrol in the study.
