Effect of multicomponent interventions on tuberculosis notification in mining and pastoralist districts of Oromia region in Ethiopia: a longitudinal quasi-experimental study.

Abstract

To demonstrate the impact of interventions on tuberculosis (TB) case detection in mining and pastoralist districts in southeastern Ethiopia over a 10-year period. Longitudinal quasi-experimental study. Health centres and hospitals in six mining districts implemented interventions and seven nearby districts functioned as controls. Data from the national District Health Information System (DHIS-2) were used for this study; therefore, people did not participate in this study. Directed at training, active case finding and improving treatment outcomes. Primarily, trends in TB case notification and percentage of bacteriologically confirmed TB-as collected by DHIS-2-between pre-intervention (2012-2015) and post-intervention (2016-2021) were analysed. Secondarily, post-intervention was split into early post-intervention (2016-2018) and late post-intervention (2019-2021) to also study the long-term effects of the intervention. For all forms of TB, case notification significantly increased between pre-intervention and early post-intervention (incidence rate ratio (IRR): 1.21, 95% CI: 1.13, 1.31; p<0.001) and significantly decreased between pre-intervention/early post-intervention and late post-intervention (IRR: 0.82, 95% CI: 0.76, 0.89; p<0.001 and IRR: 0.67, 95% CI: 0.62, 0.73; p<0.001). For bacteriologically confirmed cases, we found a significant decrease between pre-intervention/early post-intervention and late post-intervention (IRR: 0.88, 95% CI: 0.81, 0.97; p<0.001 and IRR: 0.81, 95% CI: 0.74, 0.89; p<0.001). The percentage of bacteriologically confirmed cases was significantly lower in the intervention districts during pre-intervention (B: -14.24 percentage points, 95% CI: -19.27, -9.21) and early post-intervention (B: -7.78, 95% CI: -15.46, -0.010; p=0.047). From early post-intervention to late post-intervention, we found a significant increase (B: 9.12, 95% CI: 0.92 to 17.33; p=0.032). The decrease in TB notifications in intervention districts during late post-intervention is possibly due to a decline in actual TB burden as a result of the interventions. The unabated increase in case notification in control districts may be due to continued TB transmission in the community.