Fig. 5

Performance of cutoffs and decision curve analysis for tuberculosis diagnostic nomogram among people with HIV/AIDS. A The sensitive and specific performance of the tuberculosis nomogram at different cut-offs. B Decision curve analysis compares the standardized net benefit of different strategies. The clinical utility of four diagnostic strategies is compared by plotting the net benefit (y axis) for the threshold based on the risk of bacteriological confirmed tuberculosis (x axis). Net benefit calculated by summing the benefits (true positives) and subtracting the harms (false positives, weighting by the relative harm of withdraw treatment and an unnecessary treatment). The red line represents the developed nomogram yielded highest net benefit and area under decision curve for potential thresholds (ranging from 10 to 41%). The green line represents the net benefit of XPHACTOR study rule (included ART status, CD4 count, BMI, and the number of WHO symptoms). The blue line represents the net benefit of W4SS algorithm. The black line represents the assumption that all PLHIV diagnosis with confirmed TB. The grey line represents the assumption that no PLHIV diagnosis with confirmed TB. PLHIV people with HIV/AID, TB tuberculosis, W4SS WHO-recommended four symptom screen