Mapping tuberculosis treatment outcomes in Ethiopia

Area of Science:

• Epidemiology
• Geospatial Analysis
• Public Health

Background:

• Tuberculosis (TB) is a leading infectious cause of death in Ethiopia, claiming over 30,000 lives annually.
• Significant geographical disparities in poor TB treatment outcomes exist within Ethiopia at district and zone levels.

Purpose of the Study:

• To determine the geographical variation of poor TB treatment outcomes across Ethiopian districts and zones.
• To investigate the association between these variations and socioeconomic, behavioral, healthcare access, and climatic factors.

Main Methods:

• Geospatial analysis of national TB data (2015-2017) from the Health Management Information System (HMIS).
• Calculation of poor TB treatment outcome prevalence (treatment failure, death, loss to follow-up).
• Binomial logistic regression and spatial analysis using a Bayesian framework with MCMC simulation, Getis-Ord Gi*, and Moran's I statistics.

Main Results:

• Analysis of 223,244 TB patients revealed an overall poor treatment outcome rate of 9.0% (range 1-58%).
• Hot-spots for poor outcomes were identified in border districts of Afar, Gambelia, and Somali regions; cold spots were in Oromia and Amhara.
• Poor outcomes correlated with low wealth index (OR 1.01), poor TB knowledge (OR 1.02), and higher annual mean temperature (OR 1.15).

Conclusions:

• Significant spatial variation in poor TB treatment outcomes exists in Ethiopia.
• Socioeconomic status, TB knowledge, and climate are associated with these spatial variations.
• Targeted interventions in identified hot-spot areas are crucial for reducing poor TB outcomes and achieving national End-TB Strategy goals.