Tuberculosis treatment outcomes in Ethiopia from 2003 to 2016, and impact of HIV co-infection and prior drug exposure: A systematic review and meta-analysis.

backgroundKnowledge of tuberculosis (TB) treatment outcomes is substantially needed to assess the performance of national TB controls programs (NTPs). To date, the overall estimates of treatment outcomes have not been determined in Ethiopia. Therefore, this meta-analysis was undertaken to produce pooled estimates of TB treatment outcomes and to analyze the impact of prior anti-TB drug exposure and HIV co-infection.

methodsPotentially relevant studies were retrieved from PubMed, EMBASE, and MEDLINE online databases. The unpublished studies have been retrieved from the grey literature through Google and Google Scholar. The pooled estimates were calculated using random effect model. The summary estimates were also presented using Forest plots and Tables. The outcome measures were successful and unsuccessful treatment outcomes. Patients who were cured or with completed treatment defined as successful treatment outcome and patients meeting the definition of death, defaulting and failure are considered as unsuccessfully treated cases.

resultsA total of 34 studies are included for meta-analysis. The pooled estimate of successful TB treatment outcomes amounts to 83.7% (95% CI 81.1%-86.3%). Of successfully treated cases, 33.9% were cured and the remaining completed cases. Besides, among patients with unsuccessful treatment outcome, nearly 50% were dead and the rest were treatment failures and defaulters. Sub-group analysis shows that high treatment success rate was estimated in Afar; 88.9% (95% CI 83.8%-94.2%), followed by Oromia; 88.5% (95% CI 82.6%-94.5%) and Gambella; 86.1% (95% CI 84.4%-87.9%), whereas relatively poor treatment outcome was noted in Tigray; 20.0% (95% CI 2.1%-37.9%) and Amhara; 19.0% (95% CI 12.6%-25.5%). The unsuccessful TB treatment outcome was found to be higher among HIV/TB co-infected cases with an odds ratio of 1.98 (95%CI, 1.56-2.52) and re-treated cases with an odds ratio of 2.17 (95%CI, 1.55-3.03). The time trend was assessed from 2003 to 2016, but it shows insignificant variation with treatment outcome (P = 0.108).

conclusionThe rate of successful treatment outcome in Ethiopia appears generally high, only slightly below the threshold suggested by the World Health Organization. History of tuberculosis treatment and HIV/TB co-infection were inversely associated with favorable treatment outcomes.