Background: Tuberculosis (TB) is a major public health problem in low and middle-income countries. Early detection and enrolment of TB cases is a challenge for National TB Programs.
Objective: To understand the performance and feasibility for scale-up of Xpert MTB/RIF assay for the TB diagnosis in Nepal.
Design: Implementation research employed mixed-method sequential explanatory design. The results of Xpert MTB/RIF assay were analysed in 26 TB diagnostic centres where Xpert machines had been installed before 2015. In-depth interviews and focus group discussions were conducted with stakeholders, purposively selected to represent experiences in centres that were functioning well, poorly or not functioning.
Results: During a one-year period in 2015/16, 23,075 Xpert MTB/RIF assays were performed in 21 diagnostic centres with 22,288 people also tested by sputum microscopy. Among these, 77% had concordant (positive or negative) results, demonstrating fair agreement (Kappa score, 0.3) between test results. Test failure and positivity rates in diagnostic centres ranged from 2.6% to 13.4% and 6.5% to 49%, respectively. The number of cartridges per positive result varied from 2.3 to 10.2. Xpert assay was positive in 3314 (15% of all cases) sputum smear microscopy negative cases. Of 4280 bacteriologically confirmed cases by Xpert assay, 355 (8%) were rifampicin resistant. Xpert machines were no longer functioning regularly throughout the year in 5 diagnostic centres. The main barriers for effective implementation of Xpert in Nepal were the lack of: timely supply of cartridges; replacement of damaged modules; maintenance of Xpert machines; and stock verification for timely procurement of cartridges. Inadequate laboratory infrastructure for maintaining functional Xpert equipment further challenges implementation and scale-up.
Conclusion: The implementation of Xpert MTB/RIF assay has increased case-finding of TB and MDR-TB in Nepal. However, there is a need to improve laboratory performance and strengthen laboratory infrastructure for optimal utilisation and scale-up of Xpert.