High HIV incidence and sexual behavior change among pregnant women in Lilongwe, Malawi: implications for the risk of HIV acquisition

Abstract

Background: HIV incidence is higher among pregnant women than their non-pregnant counterparts in some sub-Saharan African settings. Our aims were (1) to estimate HIV incidence during pregnancy and (2) to compare sexual activity between pregnant, postpartum, and non-pregnant women.

Methods: We examined a retrospective cohort of 1087 women to identify seroconverters using antenatal and labor ward HIV test results. We also conducted a cross-sectional survey, including a quantitative questionnaire (n = 200) and in-depth interviews (n = 20) among women in early pregnancy, late pregnancy, postpartum, and non-pregnancy. Outcomes included measures of sexual activity, reported spouse's risky behavior, and beliefs about abstinence.

Results: 11 of 1087 women seroconverted during pregnancy yielding a 1% seroconversion risk and an incidence rate of 4.0/100 person years (95% CI 2.2-7.2). The reported sexual activity of the early pregnancy and non-pregnancy groups was similar, but significantly higher than the late pregnancy and postpartum groups (p<0.001). During pregnancy, sex acts decreased as gestation increased (p = 0.001). There was no reported difference in the spouse's risky behavior. Most women believed that sex should cease between the 6(th) and 8(th) month of pregnancy and should not resume until 6 months postpartum. Some talked about conflict between their cultural obligation to abstain and fear of HIV infection if their spouses find other partners.

Conclusions: HIV incidence is high among pregnant women in Malawi, and sexual activity decreases during pregnancy and postpartum. Pregnant women need to be informed of their increased risk for HIV and the importance of using condoms throughout pregnancy and the postpartum period.

Figure 1. Matching schematic for HIV incidence calculation in the antenatal population at Bwaila Hospital.

Antenatal and labor ward records were matched manually and systematically resulting in 1302 final matches whose antenatal and labor ward HIV test results were compared. Of 1087 HIV negative women at the initial antenatal test, 11 tested HIV positive at the labor ward indicating a seroconversion risk of 1% during pregnancy.

Figure 2. Average Monthly and Weekly Sex Acts among early pregnancy, late pregnancy, postpartum, and non-pregnant study groups.

Early pregnancy and non-pregnancy sexual activity was similar and significantly exceeded that of the late pregnancy and postpartum groups when comparing both weekly and monthly sex acts.

Figure 3. Monthly Sex Acts by Weeks Gestation.

Comparing the monthly sex acts of all pregnant participants at various weeks gestation, sexual activity significantly declined as pregnancy advanced with Pearson’s correlation coefficient of −0.322 and p value of 0.001.