Mortality in a large tuberculosis treatment trial: modifiable and non-modifiable risk factors

SETTING: North America. OBJECTIVES: Tuberculosis (TB) patients in North America often have characteristics that may increase overall mortality. Identifying modifiable risk factors would allow for improvements in outcome. DESIGN: We evaluated mortality in a large TB treatment trial conducted in the United States and Canada. Persons with culture-positive pulmonary TB were enrolled after 2 months of treatment, treated for 4 more months under direct observation, and followed for 2 years (total observation: 28 months). Cause of death was determined by death certificate, autopsy, and/or clinical observation. RESULTS: Of 1075 participants, 71 (6.6%) died: 15/71 (21.1%) HIV-infected persons, and 56/1004 (5.6%) non-HIV-infected persons (P < 0.001). Only one death was attributed to TB. Cox multivariate regression analysis identified four independent risk factors for death after controlling for age: malignancy (hazard ratio [HR] 5.28, P < 0.0001), HIV (HR 3.89, P < 0.0001), daily alcohol (HR 2.94, P < 0.0001), and being unemployed (HR 1.99, P = 0.01). The risk of death increased with the number of independent risk factors present (P < 0.0001). Extent of disease and treatment failure/relapse were not associated with an increased risk of death. CONCLUSIONS: Death due to TB was rare. Interventions to treat malignancy, HIV, and alcohol use in TB patients are needed to reduce mortality in this patient population.