Modeling the Survival of Tuberculosis Patients in Eastern Zone of Tigray Regional State

Abstract

Background

Tuberculosis (TB) is still a public health problem and amongst the top ten leading causes of death. The aim of this paper was to identify the factors that significantly affect the survival of tuberculosis patients.

Methods

A retrospective cohort study was carried out in Adigrat General and Wukro hospitals, Eastern Zone of Tigray region, Ethiopia. Data for this study were obtained from medical records of all TB cases registered from September 2016 to August 2017 in the two hospitals. Log-rank test and Kaplan–Meier plot were used to evaluate the survival pattern of TB patients. A multivariable Cox proportional regression model was employed to identify the predictors of mortality. Factors with a P-value smaller than 0.05 were taken as statistically significant facilitators of TB death.

Results

Of the 397 patients studied over the specified period, 23 (5.8%) had died. A statistically significant survival difference was observed among gender, residence, HIV status, treatment category, and age category of patients. In multivariable cox regression, lower survival rates were observed among patients aged ≥45 years (HR = 5.315, 95% CI: 1.231–22.959), relapse cases (HR = 4.069, 95% CI: 1.636–10.119), patients with extrapulmonary TB (HR = 3.054, 95% CI: 1.044–8.940), patients from rural areas (HR = 2.834, 95% CI: 1.161–6.916), patients with a bodyweight of ≤50 kg and HIV-positive patients.

Conclusion

Based on the survival experience of TB patients, advancing age, extrapulmonary TB infection, living in rural residence, lower bodyweight at beginning of treatment, HIV co-infection, and being a retreatment patient were predictors of mortality. To achieve the “End TB Strategy” goal of zero death, proper targeting of care to these vulnerable groups should be advised.

Keywords:

• hazard probability
• mortality
• tuberculosis
• survival

Acknowledgments

The author acknowledges Adigrat and Wukro hospital staff for their support throughout the data collection. I am also thankful to Adigrat University for funding this project.

Abbreviations

AIC, Akaike information criterion; ART, antiretroviral therapy; CI, confidence interval; CFR, case fatality ratio; CPT, cotrimoxazole prophylaxis therapy; HR, hazard ratio; HIV, human immunodeficiency virus; MDR, multidrug-resistant; NTB, national tuberculosis program.

Data Sharing Statement

The raw data set used for this study is offered on reasonable request.

Disclosure

The author declares no conflict of interest concerning this paper.