Background:  HIV voluntary counseling and testing (VCT) is recommended for persons treated for tuberculosis (TB) because of high rates of HIV co-infection. TB clinics serve as a referral point for persons with both TB and a broad range of other pulmonary infections. Significant opportunities to diagnose HIV may be missed by limiting HIV testing to only persons diagnosed with TB. In addition, the ability to test household members of persons presenting to TB clinics may have a high yield in identifying HIV infected persons. Among TB suspects evaluated at an urban TB clinic in Uganda, we determined the HIV prevalence, HIV risk behaviors, and willingness to refer household members for VCT.

Methods:  Adult patients presenting for TB evaluation at the ambulatory National TB and Leprosy Programme clinic in Kampala were systematically invited to receive same-day HIV VCT. Consenting patients received pre and post-test counseling, and were administered a standardized questionnaire to assess HIV risk behaviors and attitudes toward VCT. TB diagnosis data were abstracted from clinical records.

Results:  Among 395 eligible patients approached between October 2004 and June 2005, 325 (82%) consented to VCT. Among these, the median age was 30 years (range 18 to 81 years) and 190 (58%) were male; 137 (42%) were HIV⁺. Of 298 patients with available data, 213 (71%) were diagnosed with TB. HIV seroprevalence was elevated in patients with a non-TB diagnosis (54%) compared to those diagnosed with TB (43%), though this difference was not statistically significant (p = 0.09). Among HIV-infected patients, 44% reported 2 to 5 sexual partners and 12% reported 6 or more sexual partners in the previous 6 months. Fewer than 7% reported routine condom use with regular or casual sexual partners. The vast majority of patients (89%) reported being “very willing” to refer family members for home or clinic-based HIV VCT. Willingness to refer household members for VCT did not differ by HIV serostatus (p = 0.7).

Conclusions:   The very high seroprevalence of HIV in both TB and non-TB cases merits the implementation of routine VCT for all patients evaluated at TB clinics. VCT should be offered to household members of TB clinic patients to identify additional HIV infections and increase access to HIV treatment and prevention. HIV-infected patients with active TB reporting ongoing risky sexual behaviors may be at high risk for HIV transmission, and contribute significantly to HIV and TB spread.