Objective: To evaluate clinical characteristics and outcomes in patients hospitalized for tuberculosis, comparing those in whom tuberculosis treatment was started within the first 24 h after admission with those who did not. Methods: This was a retrospective cohort study involving new tuberculosis cases in patients aged ≥ 18 years who were hospitalized after seeking treatment in the emergency room. Results: We included 305 hospitalized patients, of whom 67 (22.0%) received tuberculosis treatment within the first 24 h after admission (≤24h group) and 238 (88.0%) did not (>24h group). Initiation of tuberculosis treatment within the first 24 h after admission was associated with being female (OR = 1.99; 95% CI: 1.06-3.74; p = 0.032) and with an AFB-positive spontaneous sputum smear (OR = 4.19; 95% CI: 1.94-9.00; p < 0.001). In the ≤24h and >24h groups, respectively, the ICU admission rate was 22.4% and 15.5% (p = 0.258); mechanical ventilation was used in 22.4% and 13.9% (p = 0.133); in-hospital mortality was 22.4% and 14.7% (p = 0.189); and a cure was achieved in 44.8% and 52.5% (p = 0.326). Conclusions: Although tuberculosis treatment was initiated promptly in a considerable proportion of the inpatients evaluated, the rates of in-hospital mortality, ICU admission, and mechanical ventilation use remained high. Strategies for the control of tuberculosis in primary care should consider that patients who seek medical attention at hospitals arrive too late and with advanced disease. It is therefore necessary to implement active surveillance measures in the community for earlier diagnosis and treatment.

Keywords: Tuberculosis; Hospitalization; Time-to-treatment; Emergency medicine; Delayed diagnosis.