In Cameroon, women who marry later and those who have longer periods between age at first sex and age at first marriage have elevated rates of HIV infection, most likely because they have more partners than other women, according to a nationally representative survey.¹ Among married women aged 20–29, those who first married at age 20 or older are more likely than those who married by age 16 to be HIV-positive (odds ratio, 2.7), as are women who had longer periods between first sex and marriage (1.1 for each additional year). Women aged 20–24 have elevated odds of infection for each additional year between these two events (1.2), whereas they have reduced odds for each year they postponed having first sex (0.8). In addition, women of higher socioeconomic status are at elevated risk of HIV infection, as are those who report a greater number of lifetime sexual partners.

The study examined whether HIV risk among married Cameroonian women in their 20s varied by their age at first marriage or the length of time between first sex and first marriage. While previous studies in Cameroon have been limited to women living in large cities, this one was based on the 2004 Cameroon Demographic and Health Survey, a nationally representative survey and the first to include HIV testing. More than 90% of eligible women agreed to be tested, yielding a sample of 1,481 women with both interview data and conclusive HIV results. Age at first marriage was defined as the age at which a woman began living with a husband or consensual partner, and was categorized as 16 or younger, 17–19, or 20 or older. Initial regression analyses were conducted separately for place of residence (urban or rural) and age-group (20–24 or 25–29), and further analyses controlled for age, residence, household wealth, number of lifetime sexual partners and use of condoms at last sex.

Between 1991 and 2004, the median age at first marriage among Cameroonian women aged 20–24 rose from 17.3 to 18.3, and the interval between median age at first sex and median age at first marriage increased from 1.1 to 1.6 years. Nonetheless, nearly half (48%) of women in the current analysis had married by age 16. Age at first marriage varied by socioeconomic characteristics; women who were poorer, had less education and lived in rural areas married earlier. Mean age at first sex was 15.8, and it increased with higher age at first marriage. However, the number of lifetime partners was greater among those who married later, partly because of the longer duration of their premarital sex lives.

Eight percent of married respondents aged 20–29 tested positive for HIV. In multivariate analysis controlling only for current age, women who had first married at age 20 or older had higher odds of being HIV-positive than those who married by age 16 (odds ratio, 2.7); this was particularly the case among rural women (3.4). Among women aged 20–24, the odds of having HIV were reduced for each additional year they had postponed having first sex (0.8). In a similar analysis, the risk of being HIV-positive increased with each additional year between first sex and first marriage for the entire sample (1.1), as well as among rural women and those aged 20–24 (1.2 each).

Regression analysis that controlled for residence, wealth and several sex-related behaviors found no significant association between HIV infection and age at first marriage. However, household wealth and number of lifetime sexual partners were associated with infection: Women in the three highest wealth quintiles had higher odds of having HIV than did those in the lowest quintile (odds ratios, 2.7–3.4), and respondents' odds of infection increased with each additional lifetime partner (1.1). In a separate analysis, HIV infection was not associated with the number of years between first sex and first marriage, but it was again associated with being in the three highest wealth quintiles (2.8–3.5) and with having more partners (1.1 for each additional partner).

The researcher believes that the increased HIV risk among later-marrying women, as well as among those with a longer period of premarital sex, is mostly explained by the greater number of partners these women tended to have. Hence, he recommends that policy and program interventions focus on this premarital period, and specifically on reducing the number of partners. Furthermore, he warns that if the trend toward "long-term increase[s] in the age at marriage and the length of time between first sex and marriage in Cameroon" continues, "there will be an even greater need in [the] future to protect women during the premarital sexual period from contracting HIV," and effective interventions will require significant financial support.<

—J. Thomas

REFERENCE

1. Adair T, HIV status and age at first marriage among women in Cameroon, Journal of Biosocial Science, 2008, 40:743–760.