BackgroundDeaths among tuberculosis patients due to varied causes are frequently reported. Usually, the death rate is high during the first six to eight months, i.e., the intensive phase of treatment. The healthcare personnel working under National Tuberculosis Elimination Program are reluctant to ascribe death to tuberculosis and may prefer to attribute the death to other causes. On the other hand, in hospitals, most death are attributed to tuberculosis without careful death audits and reasonable investigations. Comorbidities like diabetes, HIV, drug resistance, and drug-induced toxicity consequent to administering secondary lines of drugs need due consideration while ascertaining the cause of death. The study aimed to measure mortality within one year and a maximum of two years after registration. We also determined the cause of death by conducting a verbal autopsy by medico-social workers using the WHO modified autopsy tool. Two pairs of physicians ascertained cause of death by reviewing the collected information. We obtained the consensus opinion of a pair of physicians. In case of disputed opinion, opinion of senior physician from the other group was considered final. Material and methodsWe obtained the list of registered tuberculosis patients in 2019–2020 in Pune Municipal Corporation. The authors conducted the study in 2021–22. The authors prepared a verbal autopsy tool based on the WHO tool. We trained the experienced medical social workers. They called registered phone numbers to know the patient's present status. The social workers paid home visits and confirmed the present status if the phone was not connected even after repeated calls. They collected all the required details, reviewed information and decided the cause of death. The collected detailed information was shared with pairs of senior physicians who ascertained the cause of death. We calculated standardized mortality ratio and Kaplan-Meier survival analysis. ResultsThe eligible patients were 7461, and there were 506 deaths. The death rate among males was higher (8.29%) compared to females (5.02%) [χ2 (with Yates correction) = 30.73; p < 0.00001]. The mean age of deceased and living patients was 42.77 (S.D. = 18.07) and 38.46 (S.D. = 17.54) years, respectively (t = 5.33; p < 0.0001). The death rate was 3.88 per 1000 person-months. More than 60% of patients died within six months of initiation of treatment. The overall standardized mortality ratio was 9.61; it decreased as the age advanced. Kaplan-Meier survival analysis showed that the overall survival for two years was 92.7%. The common causes were pneumonia, pulmonary tuberculosis, acute coronary syndrome, and some liver injury/disease. ConclusionsThe standardized mortality ratio due to tuberculosis is high. The common causes of death are pneumonia and acute coronary syndrome.