Demographic and health surveys indicate limited impact of condoms and HIV testing in four African countries.

Condom promotion and HIV testing for the general population have been major components of HIV prevention efforts in sub-Saharan Africa's high prevalence HIV epidemics, although little evidence documents their public health impact. Recent enhancements to the large, population-based demographic and health surveys (DHS) and AIDS information surveys (AIS) allow use of these data to assess the population-wide impact of these strategies. We analysed the latest DHS and AIS data from four sub-Saharan African countries with high prevalence, heterosexually transmitted HIV epidemics (Côte d'Ivoire, Swaziland, Tanzania and Zambia; N = 48 298) to answer two questions: 1) Are men and women who use condoms less likely to be HIV-infected than those who do not?; and 2) Are men and women who report knowing their HIV status more likely to use condoms than those who do not? Consistent condom use was associated with lower HIV infection rates for Swazi men but with higher HIV infection rates for women in Tanzania and Zambia; it made no significant difference in the other five sex/country subgroups analysed. Inconsistent condom use was not significantly associated with HIV status in any subgroup. Knowing one's HIV status was consistently associated with higher rates of condom use only among married people who were HIV-positive, even though condom use in this group remained relatively low. Effects of knowing one's HIV status among other subgroups varied. These results suggest that condoms have had little population-wide impact for HIV/AIDS prevention in these four countries. HIV testing appears to be associated with increased condom use mainly among people in stable partnerships who test positive. HIV testing and condom promotion may be more effective when targeted to specific groups where there is evidence of benefit rather than to general populations.