ABSTRACT
BACKGROUND: To reduce transmission and improve patient outcomes, rapid diagnosis and treatment of rifampicin-resistant tuberculosis (RR-TB) is required. OBJECTIVE: To conduct a systematic review and meta-analysis assessing time to treatment for RR-TB and variability using diagnostic testing methods and treatment delivery approach. DESIGN: Studies from 2000 to 2015 reporting time to second-line treatment initiation were selected from PubMed and published conference abstracts. RESULTS: From 53 studies, 83 cohorts (13 034 patients) were included. Overall weighted mean time to treatment from specimen collection was 81 days (95%CI 70-91), and was shorter with ambulatory (57 days, 95%CI 40-74) than hospital-based treatment (86 days, 95%CI 71-102). Time to treatment was shorter with genotypic susceptibility testing (38 days, 95%CI 27-49) than phenotypic testing (108 days, 95%CI 98-117). The mean percentage of diagnosed patients initiating treatment was 76% (95%CI 70-83, range 25-100). CONCLUSION: Time to second-line anti-tuberculosis treatment initiation is extremely variable across studies, and often unnecessarily long. Reduced delays are associated with genotypic testing and ambulatory treatment settings. Routine monitoring of the proportion of diagnosed patients initiating treatment and time to treatment are necessary to identify areas for intervention.
