Predictors of poor treatment outcomes for drug resistant tuberculosis in eastern Uganda :

Abstract

The outcome of treatment in drug resistant tuberculosis (DR-TB) patients is often poor. but predictors of poor treatment outcome have remained obscure. Uganda is one of the 30 high TB/HIV burden countries in the world. The Uganda multidrug resistant tuberculosis (MDR-TB) treatment success rate dropped from 82% among patients enrolled into care in 2010 to 74% among the 201 3 cohort and the mortality rate increased from 11% in the 2012 cohort to 18% in the 2013 cohort, I assessed treatment outcomes and predictors of poor treatment outcome among DR-TB patients in Eastern Uganda in order to identify enabling and disabling factors that would inform MDR TB intervention programs to ensure better treatment outcomes. Methods A retrospective cohort study was conducted using data from four selected DR TB treatment initiation hospitals in Eastern Uganda: Iganga, Mbale, Soroti and Lira. All DR-TB patients who initiated TB treatment between June 2013 and December 2016 were included in the study. Data were abstracted from DR TB register and patients' files using a data abstraction form. Data from the records of 269 DR-TB patients were analyzed using STATA version 14. Univariate, bivariate and multivariable analyses were done. Outcomes were evaluated at 6, 12 and 24 months. Results The median age was 37 years, and 62.4% were female DR-TB/HIV co-infection was at 54.7% (147). Overall, 69% (203) of patients had favorable treatment outcomes. Death, loss to follow up, not evaluated and failure rates among the study participants were 32 (11.9%) of which 50% of the deaths occurred in the first 6 months of treatment, 9 (3.3 0/0), 24 (8.9%) and 1 (0.3%) respectively. Predictors of poor treatment outcomes were negative HIV status (OR 0.39, 95% CI 0.20-0.76), Directly Observed Treatment (DOT) model of care (OR O. I l, 95 0/0 Cl 0.05-0.23), DOT site (OR 3.8, 95% CI 1.65-8.76) and drug unavailability (OR 2,7, 95% CI 1.20-5.95), The majority of the MDR TB patients (n=193, 71.7%) had prior exposure to TB treatment, and 66 (25%) of MDR TB patients’ treatment had poor outcomes. The community based DOT model of care yielded more treatment success (195/207 (94.2%)) compared to the facility based model (1 0/207 (4.8%). Conclusion The treatment success of DR TB patients in Eastern Uganda is lower than the global and national targets. This can be majorly attributed to the high death rates. My study shows that being on DOT does not translate to favorable treatment outcomes centrally to what is globally known. The high proportion of unevaluated patients also plays a significant role in the low treatment success rate just as the irregular supply of medicines. It is recommended that the. TB program reviews the DOT model of care and ensure constant drug supply.