Jewkes R, Levin J; International Conference on AIDS.

Int Conf AIDS. 2000 Jul 9-14; 13: abstract no. WePpD1416.

R.Jewkes, Cersal Medical Research Council, Private Bag X385, Pretoria 0001, South Africa, Tel.: +27 12 339 8525, Fax: +27 12 339 8582, E-mail: rjewkes@mvc.ac.za

Issues: Gender inequalities are increasingly recognised as critical influences on relationships and barriers to the adoption of HIV preventive practices. In order to describe connections in South Africa we explored available data. Methods: We examined two major data sets, the 1998 South Africa Demographic & Health Survey, which interviewed almost 12 000 women and in-depth study of violence against women, with 1306 participants, which was conducted to provide external validation of the DHS results. Results: Women who had been beaten in the year prior to the DHS were more likely to have used a condom on last intercourse, than those never abused (10.8% v. 8.4%), abused women were more likely to have suggested condom use in the validation study (sparking violence or threats of violence in 1% of cases); in the DHS women raped as children were more likely to have used condoms than other women (10.2% v. 8.5%), as were women who were molested (but not raped) as children (9.9% v. 8.5%). These associations were not markedly changed by adjustment for age, race or level of education. Findings running counter to this pattern were a highly significant protective association between discussing HIV prevention and domestic violence found in the validation study. Conclusions: Experiences of assault were not important determinants of condom use. This finding is surprising and may be artifact as it is possible that the magnitude of the effect of abuse has been diluted by contamination of the `non-abused' group by abused women who deny it in an interview. None the less, this points to a need to explore the role of other aspects of gender relations and HIV prevention, particularly communication in relationships and those operating at an ideational level. Perhaps the ultimate test will be to see whether behavioural interventions which deliver changes in gender inequalities in relationships also result in substantial increases in the level of condom use. This may be shown with the proposed evaluations in South Africa and Gambia of the package Stepping Stones. Accurate measurement of primary and secondary outcomes of HIV preventive practices will be critical.

Publication Types:

• Meeting Abstracts

Keywords:

• Acquired Immunodeficiency Syndrome
• Battered Women
• Child
• Condoms
• Domestic Violence
• Evaluation Studies
• Female
• Gambia
• HIV Infections
• HIV Seropositivity
• Humans
• Interpersonal Relations
• Rape
• Safe Sex
• Socioeconomic Factors
• South Africa
• Violence
• methods

Other ID:

• GWAIDS0003351

UI: 102240847

From Meeting Abstracts