Directly observed therapy reduces tuberculosis-specific mortality: a population-based follow-up study in Taipei, Taiwan.

OBJECTIVES: To determine the effect of directly observed therapy (DOT) on tuberculosis-specific mortality and non-TB-specific mortality and identify prognostic factors associated with mortality among adults with culture-positive pulmonary TB (PTB). METHODS: All adult Taiwanese with PTB in Taipei, Taiwan were included in a retrospective cohort study in 2006-2010. Backward stepwise multinomial logistic regression was used to identify risk factors associated with each mortality outcome. RESULTS: Mean age of the 3,487 patients was 64.2 years and 70.4% were male. Among 2471 patients on DOT, 4.2% (105) died of TB-specific causes and 15.4% (381) died of non-TB-specific causes. Among 1016 patients on SAT, 4.4% (45) died of TB-specific causes and 11.8% (120) died of non-TB-specific causes. , After adjustment for potential confounders, the odds ratio for TB-specific mortality was 0.45 (95% CI: 0.30-0.69) among patients treated with DOT as compared with those on self-administered treatment. Independent predictors of TB-specific and non-TB-specific mortality included older age (ie, 65-79 and ≥80 years vs. 18-49 years), being unemployed, a positive sputum smear for acid-fast bacilli, and TB notification from a general ward or intensive care unit (reference: outpatient services). Male sex, end-stage renal disease requiring dialysis, malignancy, and pleural effusion on chest radiography were associated with increased risk of non-TB-specific mortality, while presence of lung cavities on chest radiography was associated with lower risk. CONCLUSIONS: DOT reduced TB-specific mortality by 55% among patients with PTB, after controlling for confounders. DOT should be given to all TB patients to further reduce TB-specific mortality.