In Malawi, Polygamy Is Associated with Reduced Contraceptive Use
In Malawi, women in polygamous marriages are less likely than their counterparts in monogamous marriages to use modern contraceptives, surveillance data from more than 2,500 couples indicate.1 The proportion of women who wanted to stop childbearing was higher among those in polygamous marriages than among those in monogamous unions (54% vs. 40%), although the difference was not significant after adjustment for age and parity. Nonetheless, the odds of contraceptive use were substantially lower among women in polygamous marriages than among those in monogamous ones (odds ratio, 0.7). Moreover, when partners had conflicting fertility desires, the associated decrease in the odds of contraceptive use was larger among men and women in polygamous marriages (0.4–0.5) than among those in monogamous unions (0.6–0.7).
More than 20% of married women in West and East Africa are in polygamous relationships. Although studies have examined fertility desires and contraceptive use in such marriages, they have not taken into account differences that may exist among the co-wives in these unions. For example, one co-wife may desire a future birth, but another may not; similarly, a husband may want to have another child, but only with a specific wife. To explore whether wife-specific fertility preferences and contraceptive use in polygamous marriages differ from those in monogamous marriages, researchers analysed data collected from a demographic surveillance site in Karonga district, Malawi. Although Karonga is the most rural region in Malawi, it has the country’s highest literacy rates; nearly all residents are Christian, and about a quarter of women live in polygamous relationships.
In 2008–2009, a survey module on fertility intentions was added to ongoing surveillance efforts in Karonga. Husbands and wives were asked about their demographic characteristics, fertility history, fertility desires, perceptions of spousal fertility desires and contraceptive use; men in polygamous marriages answered these questions separately for each wife. A total of 2,636 confirmed couples completed the module; the current analyses were restricted to the 2,243 couples in which the wife was aged 18–49 and was not pregnant. The researchers compiled descriptive statistics and performed bivariate and logistic regression analyses to compare outcomes by type of marriage.
Overall, 78% of the wives were in mono-gamous marriages and 22% were in poly-gamous ones; nearly a quarter of those in polygamous relationships had more than one co-wife. Compared with wives in monogamous marriages, those in polygamous unions were older (32 vs. 29 years), had been married longer (12 vs. 10 years), were more likely to have been previously married (32% vs. 16%), had a greater number of living children (3.5 vs. 3.3), were more likely to have had a child who died (36% vs. 27%) and were less likely to have at least a secondary education (13% vs. 21%). Similar differences in age, parity and education were observed by marriage type among husbands.
The proportion of respondents who wanted to stop childbearing was higher in polygamous marriages than in monogamous unions, among both wives (54% vs. 40%) and husbands (54% vs. 38%). Similarly, both partners reported wanting to stop childbearing in 37% of polygamous husband-wife pairs, but in only 27% of monogamous pairs. None of these differences were significant, however, after adjustment for the older age and higher parity of polygamous respondents. Men and women in polygamous marriages were more likely than those in monogamous unions to think that their fertility preference matched that of their spouse; in reality, agreement in fertility desires was lower among polygamous couples than among monogamous ones.
The prevalence of contraceptive use was lower among respondents in polygamous marriages than among those in monogamous marriages, according to both wives (35% vs. 44%) and husbands (38% vs. 47%); the findings did not change after adjustment for whether couples had had sex in the past three months. Clandestine contraceptive use appeared to be greater in polygamous than in monogamous marriages; among husband-wife pairs in which the wife reported contraceptive use, 71% of monogamous husbands, but only 59% of polygamous husbands, also reported use.
Logistic regression analyses confirmed that polygamous respondents were less likely than monogamous ones to be using a modern method of contraception, according to wives’ reports (odds ratio, 0.7). Although the odds of contraceptive use were lower among couples in which only one spouse wanted to stop childbearing than among those in which both partners wanted to stop, the results did not differ substantially according to the sex of the partner who wanted to stop. However, the odds of use were reduced to a greater extent when polygamous women and men disagreed about continued childbearing (0.4–0.5) than when monogamous partners disagreed (0.6–0.7). Among polygamous couples, monogamous couples or both, contraceptive use was negatively associated with age and positively associated with level of education and number of living children. If the husband had HIV, monogamous couples were more likely to practice contraception (1.5), whereas polygamous couples were less likely to do so (0.4).
The findings, according to the researchers, suggest that although polygamous couples are at least as likely as monogamous couples to want to stop childbearing, “the translation of preferences into behaviour is less strong in [these] couples, leading to a lower use of contraception.” While the study was not designed to elucidate the reasons for this, the investigators speculate that the ability of women in polygamous marriages to share responsibilities (including child care) with their co-wives softens the impact of having an unplanned birth, and thus may reduce women’s motivation to practice contraception.—P. Doskoch
1. Baschieri A et al., Reproductive preferences and contraceptive use: a comparison of monogamous and polygamous couples in northern Malawi, Journal of Biosocial Science, 2013, 45(2):145–165.