Brief Report: Proportion and Predictors of Adult TB Contacts Accepting HIV Testing During an Active TB Case Finding Intervention in South Africa

BACKGROUND:Many individuals at risk for HIV may be reached through active TB case finding interventions in areas with highly prevalent co-epidemics of TB/HIV. METHODS:We analyzed data from a cluster-randomized trial of 2 TB case finding strategiesfacility-based screening and contact investigation of newly identified TB cases. In both arms, on-site rapid HIV testing was offered to all contacts older than 18 months who did not self-report HIV-positive status. Those who were HIV infected were referred appropriately. All contacts 15 years and older were included in this analysis. RESULTS:Among 2179 contacts identified, 50% (1092) accepted HIV testing and counselling, of whom 6.3% (68) tested HIV-positive. Contacts who were unemployed [adjusted prevalence ratio (aPR) 1.14, 95% confidence interval (CI)1.04 to 1.25], had not been to a clinic (aPR 1.09, 95% CI1.02 to 1.18) or HIV tested (aPR 1.25, 95% CI1.14 to 1.39) 6 months before, and those reporting gastrointestinal symptoms (aPR 1.22, 95% CI0.98 to 1.52) and genitourinary symptoms (aPR 1.30, 95% CI1.17 to 1.45) were significantly associated with accepting HIV testing. Women [adjusted odds ratio (aOR) 2.19, 95% CI1.26 to 3.81], individuals with a past history of tuberculosis (aOR 1.96, 95% CI0.93 to 4.14), and those not HIV tested 6 months before (aOR 2.20, 95% CI1.28 to 3.79) were significantly associated with testing HIV-positive. CONCLUSION:Offering HIV testing in the context of active tuberculosis case finding represents an opportunity to identify a large proportion of previously undiagnosed individuals with HIV in a population that might otherwise not seek testing.