Abstract Body

Background

The advent of antiretroviral treatment (ART) has led to an increase in the number of women living with HIV (HIV+) expressing interest in becoming pregnant. However, the availability of sexual and reproductive health services to support these women in their maternity plans is constrained in sub-Saharan Africa. This study examined intended and unintended pregnancies among adult HIV+ women of childbearing age followed in a 5-year cohort study in Abidjan, Cote d’Ivoire.

Methods

We implemented at 2 HIV-care facilities in Abidjan an interventional non-comparative prospective cohort of adult HIV+ women of childbearing age and on ART for less than 24 months. Participants were followed from 2016 to 2021 and were offered sexual and reproductive health services including a various range of contraceptive methods and systematic 3-monthly pregnancy screening. The incidence of pregnancy was estimated, and following states were identified: not pregnant, intended pregnancy, and unintended pregnancy. The trajectory of each woman who became pregnant at least once was then tracked, and the distribution of these women at a given time was analyzed. A non-homogeneous multi-state model identified factors influencing transitions from not pregnant to intended or unintended pregnancy.

Results

The median age of the 524 participants at baseline was 35 years. A total of 208 pregnancies were recorded overall, 36% were intended and 64% were unintended. The incidence of pregnancy was 12.5 per 100 person-years. We observed a decrease in the proportion of pregnancies over time, and a significant decrease in the proportion of unintended versus intended pregnancies. Figure 1 shows that economic well-being (poor: HR = 3.57; CI95% = [1.95 – 6.56], medium: HR = 1.96; CI95% = [1.29 – 2.99] vs wealthy), parity over 3 (HR = 3.16; CI95% = [1.34 – 7.46] vs 0), number of induced abortion (1: HR = 1.90, CI95% = [1.11 – 3.27]; ≥2: HR = 1.94, CI95% = [1.09 – 3.45] vs 0), live with partner (HR = 1.59; CI95% = [1.05 – 2.40]) and ART delay at baseline over 12 month ( HR = 0.60; CI95% = [0.39 – 0.92] vs < 6 months) influenced transition from being not pregnant to an unintended pregnancy.

Conclusions

Reliable pregnancy incidence estimates were obtained thanks to systematic and regular pregnancy tests. However, pregnancy incidence was much higher than anticipated and we are facing challenges with unintended pregnancies despite the provision of reproductive care services.