TUBERCULOSIS PROPHYLAXIS IN THE HOMELESS - A TRIAL TO IMPROVE ADHERENCE TO REFERRAL

Background: Adherence to tuberculosis evaluation is poor in a high-ris k population such as the homeless. Objective: To test two intervention s aimed at improving adherence to tuberculosis evaluation and to ident ify predictors of adherence. Methods: We conducted a randomized clinic al trial in shelters and food lines in the inner city of San Francisco , Calif. We randomized 244 eligible subjects infected with tuberculosi s to (1) peer health adviser (assistance by a peer [n=83]), (2) moneta ry incentive ($5 payment [n=82]), or (3) usual care (referral slips an d bus tokens only [n=79]). The primary outcome of the study was adhere nce to a first follow-up appointment at the tuberculosis clinic, where subjects were evaluated for active tuberculosis and the need for ison iazid prophylaxis. Results: Of the subjects assigned to a monetary inc entive, 69 (84%) completed their first follow-up appointment, compared with 62 subjects (75%) assigned to a peer health adviser and 42 subje cts (53%) assigned to usual care. Adherence was higher in the monetary incentive and peer health adviser groups than in the usual care group (P<.001 and P=.004, respectively). Patients not using intravenous dru gs and patients 50 years of age or older were more likely to adhere to a first follow-up appointment (odds ratios [95% confidence intervals] , 2.5 [1.3 to 5.0] and 3.3 [1.2 to 8.8], respectively). Among the 173 tuberculosis-infected subjects who completed their appointment, isonia zid therapy was started for 72 individuals, and three cases of active tuberculosis were identified.Conclusion: A monetary incentive or a pee r health adviser is effective in improving adherence to a first follow -up appointment in homeless individuals infected with tuberculosis. A monetary incentive appears to be superior. Intravenous drug users and young individuals are at high risk for poor adherence to referral.