BACKGROUND: The tuberculin skin test (TST) is the most widely used test for detecting tuberculosis (TB) infection. Accurate interpretation of TST requires consideration of three dimensions-the size of the skin reaction, the positive predictive value (PPV) and risk of disease.
METHODS: We developed a web-based algorithm incorporating epidemiological, medical and radiographic risk factors to help in the interpretation of positive TST results in adults (http://www.meakins.mcgill.ca/meakins/NEW TST Calculator/homeE.htm). We used summary estimates from published reviews on the prevalence of latent TB infection, the likelihood of false-positive TST and risk of active TB disease.
RESULTS: The algorithm calculations show that the most important determinants of risk of active disease are the presence of medical and radiographic risk factors, while the size of the reaction is of modest importance. In persons who have received bacille Calmette-Guérin vaccination after infancy, the algorithm calculations show that the PPV will be low. In such persons, the risk of disease is predicted to be very low, unless there are medical or radiographic risk factors that increase the risk of reactivation.
CONCLUSIONS: Our web-based algorithm can generate clinically useful estimates of the annual and cumulative lifetime risk of developing TB in adults with a positive TST.