A comparison of the yield and relative cost of active tuberculosis case-finding algorithms in Zimbabwe

Abstract

Setting: 10 districts and 3 cities in Zimbabwe Objective: To compare the yield and relative cost of identifying a case of tuberculosis (TB) if the National TB Programme (NTP) used one of three World Health Organisation (WHO)-recommended algorithms (2c,2d,3b) instead of Zimbabwe’s active case finding (ACF) algorithm Design: Cross-sectional study using data from the Zimbabwe ACF project. Results: 38,574 people were screened from April-December 2017 and 488 (1.3%) were diagnosed with TB. WHO-2d had the least number of people needing a chest X-ray (CXR) at 13,710 (35.5%) and bacteriological confirmation at 2,595 (6.7%). If the NTP had used the WHO recommended algorithms, fewer TB cases would have been diagnosed - 18% (88 cases) with algorithm 2b, 25% (122 cases) algorithm 2d, and only 7% (34 cases) with algorithm 3b. The relative cost-per-case of TB diagnosed for the Zimbabwe algorithm at $565 was over three times that of WHO 3b algorithm ($180) which was the cheapest. Conclusion: The Zimbabwe ACF algorithm had the highest yield but at a considerable cost when compared to WHO algorithms. The trade-off between cost and yield needs to be reviewed by the NTP and changing to use algorithm 3d considered.