Causes of Death among People Affected by Tuberculosis: A Cross-Sectional Study from Trivandrum, Kerala, India

Abstract Background: The case fatality ratio among patients with tuberculosis (TB) in Trivandrum district, India, has been higher than the state and national average since 2020. We did a study to describe the underlying causes of death among patients with TB in Trivandrum, India, in 2022 and to identify the challenges in the health system for systematically conducting audits among people affected by TB. Methods: We did a descriptive cross-sectional study to find out the probable underlying cause of death. We interviewed the relatives of the deceased individuals affected by TB, notified in Nikshay (the web-based case management information system of the National TB Program) during April 1, 2022 to September 30, 2022. We used a modified World Health Organization verbal autopsy tool to determine the cause of death. Two clinicians reviewed the verbal autopsy forms independently to assign the underlying and contributing cause of death. We conducted qualitative interviews with the healthcare workers and program managers involved in the national TB program to identify the challenges in systematically conducting death audits. Results: Among the enlisted 75 deaths, TB was the underlying cause of death for 43 deaths (57.3%), contributing to the cause of death for 19 (25.3%) and neither for 13 (17.3%) deaths. The median (IQR) duration from diagnosis to death was 32 (10, 69) days. Most patients who died due to TB belonged to the vulnerable population (contact with TB patients, migrants, slum dwellers, prison inmates, smoking and alcoholic) and had multiple comorbidities including diabetes, cardiovascular diseases, COPD, chronic kidney disease, chronic liver disease, and cancer). Program staff felt conducting death audits as an extra burden. Death audits conducted by untrained staff, using a limited format questionnaire, difficulties in tracking patients, and unavailability of medical records were other challenges identified for doing systematic death audits under the TB program. Conclusion: The case fatality ratio among notified TB cases may be overestimated because TB was the underlying cause of death in only 57.3% of TB deaths. Timely death audits conducted by trained health staff will help in knowing the underlying cause of death and identifying preventable deaths. This process will contribute to reducing the case fatality rates in the future.