HIV Status and Age at First Marriage among Women in Cameroon (WP33)

Recent research has highlighted the risk of HIV infection for married teenage women compared with their unmarried counterparts (Bruce and Clark, 2003; Clark, 2004). This study examines for post-adolescent women age 20-29 in Cameroon whether a relationship exists between HIV status and age at first marriage, or the length of time between first sex and first marriage. Multivariate analysis using the nationally representative sample from the 2004 Cameroon Demographic and Health Survey shows that late-marrying women and those with a longer period of premarital sex have the highest risk of contracting HIV. Although, overall, women in urban areas marry later than their rural counterparts, the positive relationship between HIV risk and age at marriage is stronger in rural areas. For late-marrying women, living in households with higher wealth status and having a larger number of lifetime sexual partners contribute to higher HIV risk. Given that age at first marriage and the gap between first marriage and first sex have increased in recent years, focusing preventive efforts on late-marrying women will be important in reducing HIV prevalence among females. Introduction Young women in sub-Saharan countries have consistently been found to have greater risk of contracting HIV than their male counterparts (Glynn et al., 2001; Gregson et al., 2002; Laga et al., 2001). Although findings in various cities have shown that married teenage women are at particularly high risk of HIV infection, more recently studies indicate that women who marry late may have increased risk of infection because of the longer period between first sex and first marriage (Bongaarts, 2006; Clark, 2004; Glynn et al., 2001). This paper uses data from the 2004 Cameroon Demographic and Health Survey (CDHS) to test whether the risk of HIV infection for post-adolescent women age 20-29 is positively related to age at first marriage or the number of years between first sex and first marriage (INS and ORC Macro, 2006). A number of studies have highlighted married adolescent women’s risk of HIV infection in subSaharan Africa. Findings from cities in Cameroon, Kenya, and Zambia show that married teenage women are more likely to be HIV-positive than unmarried women (Clark, 2004; Glynn et al., 2001; Lydie et al., 2001). It is suggested that married adolescent women in sub-Saharan Africa are more vulnerable to HIV infection than unmarried women because sexual initiation occurs earlier and there is greater frequency of sexual intercourse with their husband, who is more likely to be older and HIV-positive than the partners of unmarried women. Also, these young married women may use condoms more rarely because of a lack of bargaining power in their marriage (Bruce and Clark, 2003; Clark, 2004; Clark et al., 2006). Research in rural Malawi has shown that the husbands of HIV-positive women are more likely to be HIV-positive. Research in rural Uganda indicates that husbands ten or more years older than their wives are especially likely to be HIV positive (Bracher et al., 2003; Kelly et al., 2003). Analysis showing a negative correlation between high median age at first marriage and HIV prevalence in four sub-Saharan cities also supports the greater risk of infection in early marriage (Ferry et al., 2001). In contrast, Bongaarts (2006) has used national-level data to show that higher median age at marriage is positively associated with HIV prevalence. His analysis of 33 sub-Saharan countries used national-level HIV prevalence data from UNAIDS—primarily from the testing of pregnant women in antenatal clinics—and age at marriage data from DHS surveys. Individual-level analyses of Kenya and Ghana by Bongaarts (2006) using HIV data from DHS (controlling for age, residence, and education), show that the length of time between first sex and first marriage more strongly predicts HIV infection than the interval after first marriage. This finding suggests that late age at marriage increases HIV risk primarily because of the longer period of premarital