As per national guidelines in Uzbekistan, all presumptive tuberculosis patients should be tested using the Xpert MTB/RIF assay for diagnosing tuberculosis. There is no published evidence how well this is being implemented. In this paper, we report on the Xpert coverage among presumptive tuberculosis patients in 2018 and 2019, factors associated with non-testing and delays involved. Analysis of national aggregate data indicated that Xpert testing increased from 24% in 2018 to 46% in 2019, with variation among the regions: 21% in Tashkent region to 100% in Karakalpakstan. In a cohort (January-March 2019) constituted of 40 randomly selected health facilities in Tashkent city and Bukhara region, there were 1940 patients of whom 832 (43%, 95% confidence interval (CI): 41-45%) were not Xpert-tested. Non-testing was significantly higher in Bukhara region (73%) compared to Tashkent city (28%). In multivariable analysis, patient's age, distance between primary health centre (PHC) and Xpert laboratory, diagnostic capacity and site of PHC were associated with non-testing. The median (interquartile range) duration from date of initial visit to PHC to receiving results was 1 (1-2) day in Tashkent city compared to 3 (1-6) days in Bukhara region (p-value < 0.001). While there is commendable progress, universal access to Xpert testing is not a reality yet.
Keywords: SORT IT; central Asia; missing cases; nucleic acid amplification tests; operational research; polymerase chain reaction; pre-diagnosis attrition; rapid molecular diagnostics.