Female adolescents in Sub-Saharan Africa who have lost one or both parents are generally no more likely than their non- orphaned counterparts to marry before age 18, according to a study using population-based data from 10 countries.1 However, in four of the countries, female orphans have increased odds of early sexual debut (odds ratios, 1.4–2.1), and in two countries they have increased odds of early pregnancy (1.7 in each). The risks associated with different types of orphanhood—having lost a mother, a father or both parents—vary across countries. In addition, adolescents who are more socio-economically disadvantaged or have no formal education have elevated odds of some or all of the outcomes studied.
Researchers analyzed data on females aged 15–17 from Demographic and Health Surveys and related household surveys conducted in 2003–2006 in Benin, Chad, Congo (Brazza-ville), Côte d'Ivoire, Lesotho, Malawi, Mozambique, Tanzania, Uganda and Zimbabwe. Young women were considered orphans if either or both of their parents had died or if the status of either or both parents was unknown. The researchers examined associations among a variety of variables—including orphan status, education, religious affiliation, wealth quintile, and urban or rural residence—and three sexual and reproductive outcomes: early marriage (defined as having ever been married), early sexual debut (defined as having ever had sex) and early pregnancy (defined as having ever been pregnant).
The number of female adolescents in the sample ranged from 711 in Côte d'Ivoire to 1,801 in Benin. Overall, 16–40% of the adolescents in the 10 countries were orphans; most were paternal orphans (i.e., they had lost their father). In addition, 7–29% had ever married, 17–61% had ever had sex and 8–28% had ever been pregnant. Levels of all three outcomes were particularly high in Chad, Côte d'Ivoire and Mozambique.
In multivariate analyses, orphan status was not associated with early marriage in any of the countries studied. However, several other social and demographic factors were associated with this outcome. Compared with adolescents with an incomplete primary education, those with no education had elevated odds of early marriage in Benin, Mozambique and Tanzania (odds ratios, 2.4–4.1), and adolescents in Chad, Malawi, Mozambique and Zimbabwe had reduced odds if they had at least some secondary education (0.2–0.4). The odds of early marriage were increased among young women in two or more of the four lowest wealth quintiles (relative to those in the most well-off quintile) in seven countries: Benin, Côte d'Ivoire, Lesotho, Malawi, Mozambique, Uganda and Zimbabwe (1.8– 11.8). Compared with their Protestant peers, Muslim women had elevated odds of early marriage in Chad and Uganda (2.7–3.7), but reduced odds in Benin (0.3). Urban residence was associated with a decreased likelihood of early marriage in Benin (0.5) and with increased odds in Malawi (1.9).
In four countries—Côte d'Ivoire, Lesotho, Mozambique and Tanzania—orphaned female adolescents had greater odds of early sexual debut than did their nonorphaned peers (odds ratios, 1.4–2.1). In Mozambique and Tanzania, young women with no education had elevated odds of having ever had sex relative to those with an incomplete primary education (1.7–2.2); in Chad, Mozambique, Tanzania and Zimbabwe, young women who had at least some secondary education had reduced odds (0.3–0.6). In Côte d'Ivoire, Lesotho, Malawi and Zimbabwe, the likelihood of early sexual debut was elevated among young women in at least two of the four lowest wealth quintiles (1.8–3.8). The odds of being sexually experienced were also elevated among Muslim women in Chad relative to their Protestant peers (3.1) and among urban residents in Tanzania relative to rural women (1.6).
In two countries—Chad and Côte d'Ivoire—orphans had elevated odds of early pregnancy (odds ratio, 1.7 in each country). Early pregnancy was positively associated with a lack of formal education in Benin, Chad, Mozambique and Tanzania (1.9–2.6) and negatively associated with having at least some secondary education in Mozambique, Tanzania and Zimbabwe (0.04–0.5). The odds of early pregnancy were elevated among Muslim women in Chad (3.8), and reduced among Catholic women in Côte d'Ivoire (0.5), relative to their Protestant peers. In addition, the likelihood of early pregnancy was elevated among young women in at least three of the four lowest wealth quintiles in Benin, Congo, Côte d'Ivoire, Lesotho, Mozambique and Zimbabwe. Urban residents in Benin were less likely than their rural peers to have ever been pregnant (0.6).
A final set of analyses examined the importance of the type of orphanhood (maternal, paternal or double) to sexual health outcomes. Early marriage was associated with one or more orphan types in four countries, notably Benin, where young women who had lost both parents had sharply elevated odds of early marriage relative to their nonorphaned counterparts (odds ratio, 7.3). Early pregnancy was associated with one or more orphan types in only three countries: double orphans in Benin (2.6), maternal orphans in Côte d'Ivoire (2.6) and paternal orphans in Chad (1.8).
However, the odds of early sexual debut were elevated among at least one type of orphan in seven of the 10 countries. Those at risk included maternal orphans in Tanzania and Uganda (odds ratios, 1.9–2.2), paternal orphans in Côte d'Ivoire, Lesotho and Mozambique (1.5–2.1) and double orphans in Benin, Lesotho and Malawi (1.8–3.8).
In general, the study's findings provide little evidence that losing one or both parents influences the risks of early marriage and early pregnancy, but they do suggest that it is associated with a higher likelihood of early sexual debut, the investigators contend. The investigators recommend research to ascertain the reasons for the observed differences by type of orphanhood and by country, including any influence that ongoing programs have on sexual and reproductive outcomes, with the aim of informing programmatic efforts and social policy. "Orphanhood status alone may not be a sufficient targeting mechanism for addressing these [adverse sexual and reproductive] outcomes in many countries; a broader, multidimensional targeting scheme including orphan type, schooling, and poverty measures would be more robust in identifying and aiding young women at risk," they conclude.—S. London
1. Palermo T and Peterman A, Are female orphans at risk for early marriage, early sexual debut, and teen pregnancy? evidence from Sub-Saharan Africa, Studies in Family Planning, 2009, 40(2):101–112.