In South Africa, sexually experienced 15-24-year-old females who report low relationship control are more likely to use condoms inconsistently (odds ratio, 2.1) than are those who report high relationship control.1 Females who have experienced forced sex and those who do not talk to partners about condoms also have increased odds of being inconsistent users (5.8 and 12.9, respectively). In turn, those who report inconsistent condom use are more likely than consistent users to be HIV-positive (1.6).

This study included 4,066 sexually experienced females drawn from a nationally representative 2003 household survey of 15-24-year-old men and women. Participants answered questions on social and demographic factors, HIV risk behavior and two measures of sexual power: degree of relationship control, and experience of forced sex with their most recent partner in the previous year. Relationship control was assessed on the basis of respondents' answers to four questions: whether their partner had more control than the respondent in important relationship decisions; whether he got his way most of the time in arguments; whether he had more control than the respondent over condom use; and whether he had more control over whether the couple had sex. Women who had always used condoms with their most recent partner over the last year were considered to be consistent condom users. All participants were tested for HIV.

Once the sample was weighted to reflect the demographic distribution of 15-24-year-olds in the 2001 census, 88% of participants were black, 64% were 20-24 years old, 73% had not completed high school and 21% were HIV-positive. Thirteen percent had had more than one sexual partner in the past year, and 10% had had sex more than five times in the past month. Seventy-one percent of the participants were inconsistent condom users, and 27% reported low relationship control. Although only 4% reported forced sex in the last year, 10% said they had ever experienced forced sex; the authors believe both proportions are underestimates.

Greater proportions of inconsistent condom users than of consistent users reported low relationship control (33% vs. 14%), forced sex (5% vs. 1%), more than five episodes of sexual intercourse in the last month (13% vs. 4%) and no discussion of condom use with their partner (28% vs. 2%), and they scored lower on condom use self-efficacy (2.2 vs. 2.7 on a scale of 0-3). Inconsistent condom users were also more likely than consistent users to be 20-24 years old (66% vs. 46%) and married (7% vs. 1%), to perceive themselves at high risk for HIV infection (43% vs. 30%) and to have a partner who was 10 or more years older (7% vs. 3%). Moreover, inconsistent users were more likely to have experienced sexual debut before 14 years of age (8% vs. 4%) and to have not completed high school (76% vs. 66%).

The study also found that participants' attributes varied according to their HIV serostatus: Eighty-one percent of HIV-positive females were 20-24 years old (vs. 60% for HIV-negative females), 98% were black (vs. 86%) and 81% had not completed high school (vs. 71%). HIV-positive women were also more likely to have had more than one sexual partner in their life (68% vs. 51%) and to be inconsistent condom users (79% vs. 70%).

Multivariate logistic regression analysis revealed associations among relationship control, condom use and HIV status. Compared with women reporting high relationship control, those with low control had an elevated likelihood of using condoms inconsistently (odds ratio, 2.1). Similarly, women who had experienced forced sex and those who had not talked to their partner about condom use were at increased risk of having used condoms inconsistently (5.8 and 12.9, respectively), as were those who were married (5.4), were 20-24 years old (1.9), reported low condom use self-efficacy (1.9) and had had sex more than five times in the past month (2.9).

Although HIV serostatus was not directly related to the two measures of women's sexual power, it was related to condom use: Inconsistent users had higher odds of being HIV-positive than did consistent users (odds ratio, 1.6). Furthermore, the odds of being HIV-positive were elevated for young women who were black (7.6), were 20-24 years old (2.5), were single (2.1), had not completed high school (2.6) or lived in urban areas (2.4). In addition, participants who reported having had more than one lifetime partner had elevated odds of being HIV-positive (2.5).

Other studies in Africa have examined how women's status or household power may affect contraceptive use, but few studies have looked at sexual power and HIV serostatus among young women. This study considered only two aspects of sexual power - relationship control and experience of forced sex - which may explain why it found no direct association between sexual power and HIV status. However, it did find a significant association between inconsistent condom use and being HIV-positive, and by far the strongest risk factor for inconsistent use was not talking to one's partner about using condoms.

This study thus demonstrated that women with low sexual power were more likely to report using condoms inconsistently, which in turn elevated the likelihood of being HIV-positive. In light of these findings, the authors believe that "efforts to promote consistent condom use, a key element of HIV prevention, would benefit from efforts to enhance women's sexual power," and that improved partner communication would strengthen the sexual decision-making capacity of women.

--J. Thomas


1. Pettifor AE et al., Sexual power and HIV risk, South Africa, Emerging Infectious Diseases, Nov. 2004, <http://>, accessed Jan. 11, 2005.