| dc.description.abstract | Objectives
To summarise treatment success rate (TSR) among adult bacteriologically confirmed pulmonary tuberculosis (BC-PTB) patients in sub-Saharan Africa (SSA). Design We searched MEDLINE, EMBASE, Google Scholar and Web of Science electronic databases for eligible studies published in the decade between 1 July 2008 and 30 June 2018. Two independent reviewers extracted data and disagreements were resolved by consensus with a third reviewer. We used random-effects model to pool TSR in Stata V.15, and presented results in a forest plot with 95% CIs and predictive intervals. We assessed heterogeneity with Cochrane’s (Q) test and quantified with I-squared values. We checked publication bias with funnel plots and Egger’s test. We performed subgroup,
meta-regression, sensitivity and cumulative metaanalyses. Setting SSA. Participants Adults 15 years and older, new and retreatment BC-PTB patients. Outcomes TSR measured as the proportion of smearpositive TB cases registered under directly observed therapy in a given year that successfully completed treatment, either with bacteriologic evidence of success (cured) or without (treatment completed). Results 31 studies (2 cross-sectional, 1 case–control, 17 retrospective cohort, 6 prospective cohort and 5 randomised controlled trials) involving 18 194 participants
were meta-analysed. 28 of the studies had good quality data. Egger’s test indicated no publication bias, rather small study effect. The pooled TSR was 76.2% (95% CI 72.5% to 79.8%; 95% prediction interval, 50.0% to 90.0%, I2 statistics=96.9%). No single study influenced the meta-analytical results or conclusions. Between 2008 and 2018, a gradual but steady decline in TSR occurred in SSA but without statistically significant time trend variation (p=0.444). The optimum TSR of 90% was not achieved. Conclusion Over the past decade, TSR was heterogeneous and suboptimal in SSA, suggesting context and country-specific strategies are needed to end the TB
epidemic. | en_US |
