Objective: To estimate the prevalence of HIV-1 infection and its relation to socio-economic and obstetric history factors.
Design: A cross sectional study.
Setting: 12 randomly selected villages in rural Gutu District, Zimbabwe.
Subjects: All women of fertile age (15 to 44 years) in the selected villages were invited to participate. In total 1,213 women were interviewed and examined. On average 75% and 81% of those eligible in 1992 and 1993 respectively participated (range for villages 57 to 87%).
Main outcome measures: HIV sero-positivity.
Results: HIV prevalence was high, 21.9%. Single women below 20 years had the lowest prevalence (7.6%), while the highest (30.0%) was found in married women aged 20 to 29 years. Using a case-referent approach, women divorced, widowed or separated had twice as high a risk of being HIV positive as single women (Odds Ratio (OR) = 2.03; 95% confidence interval (CI) 1.40 to 2.93). There was a huge difference in the prevalence of HIV between villages, from 8.4 to 32.8%, thus the risk of having HIV in the two villages with the highest prevalence was more than five times higher (OR = 5.21 and 5.31 respectively) than in the village with the lowest. The multivariate logistic regression confirmed the findings of an increased risk for women aged 20 to 39 years as compared with women 15 to 19 years, and for women that were divorced, widowed or separated as compared with single. It also confirmed that the huge difference in the prevalence found between wards was statistically significant.
Conclusion: Women contract HIV early in marriage. Young men's risk behaviour needs to be addressed and women must be aware of their risks before marriage.