The childbearing preferences of women who think they may be infected with HIV tend to diverge into two very different paths, according to findings from Mozambique.1 On the one hand, compared with women who think it is impossible that they have HIV, those who say they may be infected are more likely to want to have children in the next two years than to do so later (odds ratio, 1.9). On the other hand, these women are also more likely to report that they would rather cease childbearing than have a child later (1.5). In other words, the women tend to want a child now—or never.
Because the growing availability of HIV testing and antiretroviral drugs has lengthened the time during which infected individuals are aware of their status—and the time during which they may have children—understanding the fertility intentions of the more than 22 million people living with HIV in Sub-Saharan Africa has become increasingly important. Studies of the relationship between HIV status and fertility preferences have yielded mixed results, although most have found that the desire to have children is lower among women who suspect or know they are infected, as these women fear transmitting the virus to their infant or dying before the child is grown.
The current study examined fertility intentions among women in Mozambique, where the average desired family size is large (4.3 children) and the prevalence of HIV is among the highest in the world (12–16% among adults aged 15–49, according to recent estimates); in the country’s Gaza province, where the present study was conducted, desired fertility (5.4 children) and the prevalence of HIV (25–27%) are even higher. In 2009, researchers surveyed 1,638 randomly selected women from 56 villages in four rural districts of Gaza. The survey collected information on women’s demographic, social and economic characteristics, as well as their fertility intentions and perceived HIV status. Fertility intentions were classified according to whether the woman wanted a child soon (within two years), later or never. To determine perceived HIV status, the investigators asked women to estimate the likelihood that they were already infected with the virus; response options were "very likely," "a little likely" and "almost impossible." Because the semantic difference between the first two options is small in Changana (the language in which most interviews were conducted), the researchers grouped the "very likely" and "a little likely" responses into a single category that represented women who thought they probably had HIV (the small number of women who spontaneously volunteered that they were infected were also included in this category); the investigators created another category for the substantial proportion of women who said they did not know their likely HIV status. In addition to reporting weighted descriptive statistics, the investigators performed multinomial logistic regression analyses to identify associations between perceived HIV status and fertility intentions. The analyses excluded respondents who were unmarried, childless or missing key survey data, yielding an analytic sample of 1,260 women.
Most women reported that they wanted another child soon (37%) or later (20%), though a substantial proportion (44%) did not want to continue childbearing. About a third of women (35%) thought it likely that they had HIV; 22% said it was impossible that they were infected, and 43% said they did not know their probable status. Women were more likely to report that they probably had HIV if they were aged 20 or younger than if they were 31 or older (47% vs. 32%). The only other measure associated with perceived likelihood of infection was whether the woman had been tested for HIV in the past year; 33% of women who had been tested, but only 20% of those who had not, said it was impossible that they had HIV.
In multinomial regression analyses, women who thought it was probable that they were infected were more likely than those who thought it was impossible that they were infected both to prefer to have another child soon rather than later (odds ratio, 1.9) and to prefer to stop childbearing rather than to have another child later (1.5); hence, they generally wanted to have a child either now or never. In contrast, the childbearing intentions of women who did not know their likelihood of having HIV did not differ from those of women who said it was impossible that they had the virus. None of the three groups differed from the others regarding the preference to stop childbearing rather than to have a child soon. Tests for interactions found no evidence that the relationship between perceived HIV status and childbearing intentions differed according to women’s access to antiretroviral drugs or to medications that prevent mother-to-child transmission.
Not surprisingly, age and parity were also notable predictors of fertility intentions. Women preferred to stop childbearing, rather than to have a child later, if they were aged 31 or older rather than aged 26–30 (odds ratio, 2.0) or if they had a greater number of children (2.1 per child). They were more likely to want a child sooner rather than later if they were aged 31 or older rather than 26–30 (1.8), but less likely if they had a greater number of children (0.7 per child), had at least one child younger than 24 months (0.4) or had had an HIV test in the past year (0.5). Finally, women were more likely to want to stop childbearing, rather than to have a child soon, if they had any religious affiliation (1.9–2.4), had a greater number of living children (2.8 per child), had at least one child younger than 24 months (2.1) or had had an HIV test in the past year (2.3).
Although the findings are from a single country, the researchers note that because Mozambique shares many important characteristics with other eastern African countries, the results may be generalizable to other countries in the region with high HIV rates. They add that study participants probably overestimated their HIV risk, and that as testing becomes more routine, women are likely to have a more accurate perception of their HIV status; if so, the proportion who intend to postpone childbearing, rather than take the now-or-never approach, will likely increase. However, carrying out intentions to delay childbearing is likely to be challenging for women in Mozambique, given their "limited access to effective long-acting contraceptives and safe abortion, [their] low autonomy, and the high social importance of childbearing" in the country.—P. Doskoch
1. Hayford SR, Agadjanian V and Luz L, Now or never: perceived HIV status and fertility intentions in rural Mozambique, Studies in Family Planning, 2012, 43(3):191–199.