Public-private collaboration for multidrug-resistant tuberculosis control in New York City.

SETTING: An urban tuberculosis control program where an enhanced multidrug-resistant tuberculosis (MDR-TB) management plan coordinated care with multiple providers. OBJECTIVE: To evaluate treatment outcomes of primary MDR-TB patients treated by multiple providers. DESIGN: Retrospective cohort study of tuberculosis patients from 1992-1997 provided that 1) their Mycobacterium tuberculosis isolates were resistant to at least isoniazid and rifampin, and 2) they had had < or = 30 days of anti-tuberculosis treatment prior to the collection of the first MDR-TB specimen. RESULTS: More than 100 facilities and providers reported 856 MDR-TB patients. Treatment completion reached 70% among non-HIV-infected and 30% among HIV-infected persons; 57.2% of the cohort died prior to treatment completion, 26.5% completed treatment, 16.0% transferred out, refused treatment or were lost to follow-up and 0.2% are still in care. Diagnosis in the later years of the study or cavitation on chest radiograph was independently associated with increased completion among HIV-infected patients. Eight of the 227 (3.5%) patients who completed treatment relapsed (relapse rate 1.01/100 person-years), two with drug-susceptible strains. CONCLUSION: A comprehensive MDR-TB control program improved the outcomes of both HIV-infected and non-infected individuals, despite management by multiple providers. Relapse was infrequent among patients who completed the recommended regimens.