Survival-based Prognostic Factors for Mortality in Drug-resistant Tuberculosis: A Systematic Review and Meta-analysis

Mortality among patients with drug-resistant tuberculosis (DR-TB) remains substantial despite advances in diagnosis and treatment. Individual studies have reported multiple prognostic factors for death, but findings remain heterogeneous. The objective of this study is to systematically review and quantitatively synthesize evidence on prognostic factors associated with mortality among patients with DR-TB using hazard ratios (HR) derived from survival analyses. A systematic search of PubMed, Scopus, and Web of Science was conducted to retrieve articles published from database inception to the final search date on March 31, 2026. Observational longitudinal studies involving patients with any form of DR-TB were included if they reported adjusted HR for all-cause mortality from survival models. Study selection followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2020 guideline. Risk of bias was assessed using the Quality in Prognosis Studies tool. Fixed-effects and random-effects meta-analyses using restricted maximum likelihood were performed for prognostic factors reported in at least three comparable studies, whereas other factors were synthesized narratively. The study protocol was registered in the International Prospective Register of Systematic Reviews (PROSPERO) ID: CRD420261355163. A total of 34 studies published between 2006 and 2025 were included in the study. Most studies were retrospective cohort studies conducted mainly in Asia and Africa. The pooled analysis demonstrated that several factors were significantly associated with increased mortality including human immunodeficiency virus infection (pooled HR = 2.40; 95% confidence interval [CI] 1.73, 3.33), smoking (pooled HR = 2.07; 95% CI 1.48, 2.90), anemia (pooled HR = 1.79; 95% CI 1.51, 2.12), low body mass index or underweight (pooled HR = 2.08; 95% CI 1.69, 2.56), and diabetes mellitus (pooled HR = 1.77; 95% CI 1.29, 2.44). Gender was not significantly associated with mortality (pooled HR = 0.92; 95% CI 0.77, 1.11). Between-study heterogeneity ranged from low to substantial across the meta-analyses. Age, bacteriological positivity, previous TB treatment history, and drug resistance classification were synthesized narratively because of substantial heterogeneity in exposure definitions and comparator groups across studies. Mortality in DR-TB appears more consistently associated with clinical vulnerability and comorbidity rather than demographic characteristics or resistance profile alone. These findings support a risk-based approach to DR-TB care including early identification of high-risk patients, nutritional and comorbidity assessment and closer treatment monitoring.