RESUMO
BACKGROUND: Completion of treatment for tuberculosis infection (TBI) with 9 months of self-administered daily isoniazid (9H) has historically been low (<50%) among New York City (NYC) Health Department tuberculosis clinic patients. Treatment of TBI with 3 months of once-weekly isoniazid and rifapentine (3HP) administered under directly observed therapy (DOT) might increase treatment acceptance and completion. METHODS: The study population included patients diagnosed with TBI at 2 NYC Health Department tuberculosis clinics from January 2013 through November 2013. Treatment acceptance and completion with 3HP were compared with historical estimates. Treatment outcomes, side effects, and reasons for refusing 3HP were described. RESULTS: Among 631 patients eligible for TBI treatment, 503 (80%) were offered 3HP; 302 (60%) accepted, 92 (18%) chose other treatment, and 109 (22%) refused treatment. The most common reason for refusing 3HP was the clinic-based DOT requirement. Forty (13%) patients treated with 3HP experienced side effects--9 were restarted on 3HP, 18 switched treatment regimens, and 13 discontinued. Although treatment acceptance did not differ from historical estimates (78% vs 79%, P = .75), treatment completion increased significantly (65% vs 34%, P < .01). CONCLUSIONS: Implementation of 3HP in 2 NYC Health Department tuberculosis clinics increased TBI treatment completion by 31 percentage points compared with historical estimates. More flexible DOT options may improve acceptance of 3HP. Wider use of 3HP may substantially improve TBI treatment completion in NYC and advance progress toward tuberculosis elimination.