Estimating the number of incorrect tuberculosis diagnoses in low- and middle-income countries

Tuberculosis (TB) is the greatest cause of infectious disease deaths worldwide. In highly affected countries, effective TB control requires prompt identification and treatment of individuals with active disease. We examined the performance of TB case-finding in low- and middle-income countries based on a comprehensive analysis of TB diagnosis data reported to the World Health Organization. Using these data we estimated the total number of individuals correctly and incorrectly diagnosed with TB, for 111 countries with a collective 6.8 million TB notifications in 2023. Here we estimate that in 2023, 2.05 (1.83-2.27) million individuals were incorrectly diagnosed with TB (false-positives), and 1.00 (0.71-1.36) million received a false-negative diagnosis, at an assumed 25% disease prevalence among individuals evaluated for TB. As many as three of every ten TB notifications may not have TB, and many individuals with TB receive false-negative diagnoses. Compared to current diagnostic performance, scaling-up new polymerase chain reaction-based diagnostics would substantially reduce under-diagnosis but only produce a small reduction in false-positive diagnoses. Major improvements in TB diagnosis will likely require higher-sensitivity bacteriological tests combined with reduced reliance on clinical diagnosis.