HIV-positive women are less likely than their uninfected peers to want to have a child in the future, according to a study conducted in Mbarara, Uganda.1 Although almost all of the HIV-positive women in the clinic-based sample were receiving antiretroviral therapy, only 28% wished to have a child, compared with 56% of HIV-negative women. Even after adjustment for marital status and other characteristics, the odds that an HIV-positive woman desired a child were half those of an uninfected woman (odds ratio, 0.5). In addition, fertility desires were positively associated with income, and negatively associated with age, parity, having a son and having a foster child in the household.

Studies conducted in Sub-Saharan Africa in the 1990s and early 2000s generally found that HIV-infected individuals wanted fewer children than did their uninfected peers—and often wanted to cease childbearing altogether. However, more recent research has yielded mixed results. In some studies, the prospect or availability of antiretroviral therapy has led to renewed interest in childbearing (and in one study, an actual increase in pregnancy among infected women); in others, fertility intentions remained low among infected individuals. The current study examined these issues in Uganda, where the proportion of adults infected with HIV has fallen dramatically but remains substantial (6% in 2009), fertility levels are among the highest in the world (6.7 children per woman in 2006) and antiretroviral therapy is widely available.

In 2010, researchers surveyed 1,594 women who were being treated at a regional hospital in Mbarara. Women were recruited from the hospital’s general outpatient clinic, which serves patients with nonacute conditions, and from its immune suppression clinic, which treats patients with HIV, to yield similar proportions of HIV-negative and HIV-positive women. Women were eligible for the study if they were aged 18–49 and had ever had an HIV test; they were asked about their demographic characteristics, reproductive history, fertility desires, contraceptive use and attitudes toward HIV. The authors used multivariate logistic regression to assess the relationship between women’s HIV status and their fertility desires.

Overall, 60% of the women were HIV-positive; 96% of infected women were taking antiretroviral therapy. Sixty-eight percent of HIV-positive women, but only 3% of HIV-negative women, reported that their husband had HIV. Compared with uninfected women, those with HIV were older (33 vs. 28 years, on average) and less likely to have more than a primary education (25% vs. 47%) and to be married or living as if married (54% vs. 82%). Although HIV-positive women had had one more child, on average, than other women had (3.4 vs. 2.5), the difference was not significant after adjustment for women’s age.

Among the 1,039 women who were married or in other stable unions, only 28% of those who were HIV-positive wished to have another child, compared with 56% of those who did not have the virus. The proportion of women whose partner wanted a child in the future was also smaller among infected than among uninfected women (34% vs. 57%). Regardless of their HIV status, substantial proportions of women had fertility desires concordant with those of their partner. About one in four married women were pregnant; those with HIV were more likely than those without to say that the pregnancy was a “big problem” (37% vs. 11%).

Logistic regression revealed that the odds that an HIV-positive woman wanted another child were about half those of an uninfected woman (odds ratio, 0.5). Women’s fertility desires declined with increasing parity (0.5 per child) and were lower among those who were caring for a foster child than among other women (0.6). In addition, the desire for a child was negatively associated with age and positively associated with household income.

To assess whether a desire for sons was related to childbearing intentions, the researchers calculated a second regression model that replaced the parity measure with a variable indicating whether the woman had any sons. Women with a son were far less likely than those without a son to want a child in the future (0.3); the substitution strengthened the association of childbearing intentions with age but weakened the association with income.

The researchers note that these and other recent findings, together with historical data, suggest that although fertility desires in Sub-Saharan Africa fell during the early years of the HIV epidemic and then rebounded as antiretroviral therapy became widespread, they remain lower among HIV-infected women than among uninfected women, particularly in countries that have been hit hard by AIDS (such as Uganda and South Africa). One of the new study’s limitations, the researchers add, is its lack of information on the reasons for infected women’s reluctance to have children. However, the potential challenge of caring for a child when the woman herself has HIV, as well as the fact that HIV-positive women are more likely than other women to have a low income, to be unmarried, to have an infected spouse and to be caring for a foster child, not only may help explain the disparity in fertility intentions but also may underscore the “numerous and continuing social needs” of women with HIV.—P. Doskoch


1. Snow RC et al., The social legacy of AIDS: fertility aspirations among HIV-affected women in Uganda, American Journal of Public Health, 2013, 103(2): 278–285.