Abstract:
Introduction: The WHO's 2022 Global Tuberculosis Report shows TB causes 1.3 million deaths
annually, including 167,000 TB-HIV co-infections. It affects 10.6 million people, mainly in Asia
and Africa, with 75% in the economically active age group of 15-54. Treatment interruptions
hinder efforts to end TB by 2030. Survival analysis techniques could enhance understanding and
improve treatment and policies.
Objective: This study was aimed to identify the time to TB treatment cure occurs and its predictors
among tuberculosis patients from January 2021 to December 2023 at a public health facility in
Arba Minch town. South Ethiopia.
Method: An institutional-based retrospective cohort study was conducted among selected 628 TB
patients who are admitted to the TB care unit at Arba Minch General Hospital, Dilfana Primary
Hospital, and both health centers from 2021 and 2023. Kaplan Meier survival curve was fitted to
test the survival time. The Cox proportional hazards model was used to identify predictors with
TB treatment cure. Significance was considered at P-value < or equal to 0.05 with an Adjusted
Hazard Ratio 95% CI in the multivariate analysis.
Results: Out of 628 patient records analyzed, the median time to cure was found 162 days. Key
predictors of TB cure time included sex (AHR 0.3, 95% CI 0.35-0.95), history of TB treatment
(AHR 0.56, 95% CI 0.35-0.95), and BMI.
Conclusions: In conclusion, female patents, history of TB treatment, increase of weight, type of health
facility were found to be independent predicators of cure therefor working on this predicators to improve
TB treatment cure is critical.
