Implementation of universal directly observed therapy at a New York City hospital and evaluation of an out-patient directly observed therapy program

SETTING: Directly observed therapy (DOT) program for tuberculosis (TB) at a New York City hospital. OBJECTIVE: To describe a specific TB DOT program model utilizing active prospective identification of inpatients, and identify factors associated with patient acceptance of voluntary DOT and with their retention in therapy. METHODS: Recruitment for DOT by daily surveillance of in-patients. DOT was offered as the patient's choice together with concrete services and incentives. On-site DOT was offered in an out-patient clinic. Outreach efforts were initiated when a patient missed one or more DOT visit. RESULTS: During the study period, 95% of 176 inpatients with TB were evaluated for DOT. Of the 137 who were eligible for DOT, 85% (95% confidence interval [CI], 77.5% to 90%) elected to receive DOT. Illicit drug use was independently associated with a higher like lihood of acceptance of DOT (odds ratio [OR], 4.88; 95% CI, 1.5-15.7). Among the 101 patients who received onsite DOT, illicit drug use (OR, 0.21; 95% CI, 0.08-0.6) and previous TB therapy (OR, 0.27; 95% CI, 0.27-0.7) were both independently associated with lower retention in therapy. However, with intensive case management, only 1% of this cohort was lost to follow-up and the overall treatment completion index was 98%. CONCLUSION: In-patient surveillance is a highly effective DOT recruitment strategy. A DOT model which elicits patient participation in discharge plans and offers incentives can yield a high rate of voluntary acceptance. Outpatient case management is a highly effective means of ensuring treatment completion, especially in those at risk for poor retention.