Abstract
BACKGROUND: Tuberculosis (TB) remains a global health concern with high mortality despite treatment options. Understanding the underlying risk factors for TB mortality is essential for guiding effective control strategies. This study examined sociodemographic and clinical factors related to TB mortality in Hamadan province, Iran, to inform control strategies. Study Design: A cross-sectional study. METHODS: This study evaluated data (March 2011-March 2022) obtained from a provincial TB surveillance database, encompassing smear-positive pulmonary TB (SPT), smear-negative pulmonary TB (SNT), and extrapulmonary TB (EPT) patients. Demographic and clinical characteristics were investigated, and the death rate for each group was calculated by dividing the number of TB-related deaths by the total number of diagnosed TB cases for that group during the study period. Logistic regression was applied to computed unadjusted and adjusted odds ratios (ORs) with a 95% confidence interval for the death rate using Stata 17 (P<0.05). RESULTS: Among the 942 patients included in the study, 49%, 21%, and 30% were diagnosed with SPT, SNT, and EPT, respectively. The risk of mortality was the highest among SPT patients, with EPT cases showing significantly lower odds of death (OR: 0.38, P<0.001) compared to SPT. Among SPT patients, mortality was associated with older age (OR: 1.04, P<0.001) and positive sputum smear at month 2 (OR: 19.72, P<0.001). Human immunodeficiency virus (HIV) positivity significantly increased the death rate in SNT patients (P=0.037). In EPT patients, mortality was linked to male gender (P=0.042), referral unit (P=0.023), TB hospitalization (P=0.018), and advanced age (P<0.001). CONCLUSION: Targeted interventions focusing on early diagnosis, HIV management, and care for high-risk groups (e.g., elderly) are essential to reduce TB mortality in Hamadan province. However, the findings should be interpreted with caution due to limitations, such as reliance on retrospective registry data, potential information bias, and missing data, particularly regarding HIV status.