Education and Gender Equality Are Inconsistently Associated with HIV Testing in Eastern Africa

First published online:

In a study that assessed education, gender equality and HIV testing in nationally representative samples of married or cohabiting women in Kenya, Zambia and Zimbabwe, 15–24-year-olds had an elevated likelihood of having ever had an HIV test or having had one in the past year if they had at least some secondary education (odds ratios, 1.8–3.4), but their role in making household financial decisions and attitude toward wife beating were not related to testing.1 Among women aged 25–34, education was not consistently related to HIV testing, but the belief that wife beating is never justified was associated with having had an HIV test in the past year for those living in Kenya and Zambia (1.3 for each), and involvement in financial decision making was associated with a history of testing in Zimbabwe (1.7).

Although reducing gender inequality and promoting females’ education have been cited as potentially useful strategies for preventing HIV, studies have not consistently linked levels of inequality and education to receipt of HIV prevention and treatment services, and none has examined both characteristics. The present study simultaneously assessed the relationship between these variables and testing by analyzing Demographic and Health Survey data from Kenya (2008–2009), Zambia (2007) and Zimbabwe (2005–2006), where national HIV prevalence is high (6%, 14% and 18%, respectively), and rates of HIV testing among women aged 15–49 are relatively low (58%, 40% and 26%, respectively). Analyses focused on married or cohabiting females aged 15–34; because gender inequality may influence HIV testing differentially by age, the samples were stratified into two age-groups (15–24 and 25–34). Weighted samples consisted of 3,221 respondents in Kenya, 3,049 in Zambia and 3,706 in Zimbabwe. The main outcome measures were having ever been tested for HIV and having been tested in the past year; the two gender equality variables were financial decision making (i.e., "Who usually decides how your husband’s/partner’s earnings will be used?") and whether respondents believed that a husband is justified in hitting or beating his wife in five different scenarios (e.g., if she goes out without telling him, argues with him or refuses to have sex). Multivariate logistic regression analysis was used to identify associations between HIV testing and measures of respondents’ gender equality and background characteristics.

Women in Kenya and Zambia had similar education levels, as only a fifth to a third of those in the 15–24 and 25–34 age-groups had at least some secondary education; in contrast, two-thirds of all respondents in Zimbabwe had some secondary schooling. Most women in the younger cohorts had given birth (79–86%), and nearly all of those in the older groups had had a child (97–98%). A majority of respondents lived in rural areas (64–75%), and fewer than one in 10 reported having had an STI in the last year. There was little difference in rates of HIV testing between age-groups in any country: In Kenya, 78–79% of females had ever been tested and 40–43% had been tested in the past year; in Zambia, 46–47% and 25%, respectively, had been tested in these time periods; and in Zimbabwe, 34–37% and 9–11%, respectively, reported such testing. In all three countries, a majority of females in each cohort said they made household financial decisions either jointly or alone (57–62% in Kenya and Zambia; 83–84% in Zimbabwe), and large proportions believed that wife beating was never acceptable (42–49%, 35–39% and 48–59%, respectively).

Regression analysis revealed that associations between HIV testing and education or gender inequality differed by age-group. Among 15–24-year-olds, education was consistently associated with HIV testing in all three study nations: Females who had at least some secondary education were more likely than those with less education to have ever been tested (odds ratios, 1.9–3.4) or to have been tested in the past year (1.8–3.3). Respondents from each country who had ever given birth had elevated odds of having ever been tested (2.3–3.1), as did those from Ken-ya and Zambia who were in the top three wealth quintiles (2.0–3.8). Notably, in this younger age-group, no association was found between lifetime or past-year HIV testing and either financial decision making or attitude toward wife beating.

A different pattern emerged among 25–34-year-olds. Education was related to HIV testing in only two cases: Women with at least some secondary education had an increased likelihood of having ever been tested in Zimbabwe (odds ratio, 1.5), and of having been tested in the last year in Kenya (1.5). In addition, several associations between gender inequality and HIV testing emerged. Kenyan respondents in the older cohort who believed wife beating was never justified were more likely than others to report HIV testing in either period (1.3–1.6), and their Zambian counterparts were more likely to report testing in the past year (1.3). Moreover, Zimbabwean 25–34-year-olds who made financial decisions jointly or alone had increased odds of having ever had an HIV test (1.7). Women who had ever given birth had an elevated likelihood of reporting HIV testing in either period in Kenya and Zambia (4.5–5.3). Interestingly, women’s perception of HIV risk was related to testing only in Zimbabwe, where respondents who reported low or medium risk were less likely than those who thought they had no risk to report lifetime or past-year HIV testing (0.4–0.7).

The researchers note that their use of cross-sectional data did not allow them to establish causal relationships between the gender equality measures and HIV testing. However, they point out that women’s knowledge of their HIV status is a critical step in gaining access to treatment, and they argue that efforts to promote gender equality are important in achieving such access and reducing levels of HIV infection. Furthermore, the investigators suggest that "efforts must be taken … to ensure that women have a voice in how family resources are used," and that community-level programs to improve norms regarding intimate partner violence "can play a crucial role in efforts to protect women and girls from HIV."—J. Thomas


1. Singh K, Luseno W and Haney E, Gender equality and education: increasing the uptake of HIV testing among married women in Kenya, Zambia and Zimbabwe, AIDS Care, 2013, doi:10.1080/09540121.2013.774311.