Unintended pregnancy leads to poorer treatment outcomes for women living with HIV in South Africa

Women living with HIV in South Africa who did not plan to get pregnant were three times more likely to have an elevated viral load than women who had an intended pregnancy – and this persisted up to 36-60 months post-partum.

The study, published in AIDS journal, is the first research to look at the intendedness to get pregnant on long-term viral load outcomes.

Pregnant women living with HIV should start antiretroviral treatment (ART) as soon as they learn they are HIV-positive, but there is so far little research into how intention to get pregnant can affect maternal and child health outcomes. This is despite unintended pregnancy being very common – between 2010 and 2014, an estimated 44% of all pregnancies were unintended globally.

In the context of low- and middle-income countries, estimates are similarly high. A 2012 survey from South Africa revealed that 65% of all pregnancies were unplanned. In Uganda, between 2001 and 2013, an estimated 46% of pregnancies were unplanned, and in eSwatini, 69% of women have reported that their most recent pregnancy was unplanned.

Well established research tells us that unintended pregnancy may be associated with mental health concerns, including depression, relationship dissolution, intimate partner violence, or delayed initiation of treatment if the woman is living with HIV. The findings from this current research now indicate that biomedical outcomes of HIV treatment, such as viral suppression, are also a factor.

In a cohort of 459 women starting ART during pregnancy, researchers sought to understand the association between intendedness of the current pregnancy, measured antenatally, and elevated viral load up to five years post-partum.

Women were visited up to a maximum of seven times as part of the study, from six weeks postpartum, up to 36-60 months postpartum. At all study visits, women were given a viral load test, and infants were tested for HIV at the 12-month study visit. Intention to get pregnant was assessed at study enrolment using the widely used and validated London Measure of Unplanned Pregnancy (LMUP) tool, which categorises women in intended, ambivalent, or planned pregnancy groups. Incidence of post-partum depression, alcohol misuse, and intimate partner violence were also assessed at different study visits.

Overall, ambivalent (20%) and unplanned pregnancies (60%) were commonly reported. In the month that they got pregnant, all the women reported using contraceptives, but just 2% of the sample were using contraceptives every time they had sex. The large majority had not intended to get pregnant (64%), while 7% had mixed intentions and 28% had wanted to conceive. Only 8% of the respondents actually wanted to have a baby and 67% of the respondents had never discussed having a child with their partner. But of those women who reported not wanting a baby, and not intending to get pregnant, 49% said the timing was OK, but not quite right.

Overall, 56% of the women reported elevated viral load at one visit post-partum – that increased to 60% of the women reporting unplanned pregnancy, 62% of women who reported ambivalence, and 41% who reported that their pregnancy was planned.

Over time, an elevated viral load was significantly associated with ambivalence and unplanned pregnancy – this persisted even when adjusting for demographic, clinical and psychosocial factors. Despite having adjusted for these factors, the authors were careful not to draw direct causation, saying that unintended pregnancy is one of several risk factors that contribute to women’s vulnerability.

The authors did note, however, that several factors were associated with both a lower level of pregnancy intendedness and elevated viral load – each was independently more common in women who were single or not living with a partner; had entered into antenatal care late into their pregnancy; and in those who had reported risky alcohol use.

“We posit that an unintended pregnancy heightens women’s vulnerability as part of a constellation of risk factors. In addition, we posit that unintended pregnancy may be a marker of some degree of ambivalence both towards pregnancy and towards one’s own health, with these ambivalences reflected in women’s adherence to ART,” comment the authors. Further long-term research is needed into socio-behavioural factors that have implications for these study results.

Indeed, overall, the authors note that we need to improve our understanding of the drivers of suboptimal adherence during the post-partum period and how unintended pregnancy plays a role here.