Evaluation of a risk asessment questionnaire used to target tuberculin skin testing in children

Context: Universal tuberculin skin testing of children has been shown to be costly and inefficient. In response, several authorities have recommended targeted screening based on epidemiological risk. In 1996, the New York City Department of Health (NYCDOH) developed questions to identify children who require a tuberculin skin test. Objective: To determine the sensitivity, specificity, and predictive validity of the NYCDOH tuberculosis risk assessment questionnaire. Design: Prospective criterion standard study in which tuberculin skin tests and the NYCDOH questionnaire were administered simultaneously between August 1996 and January 1998. Specific questions asked about contact with a tuberculosis case, birth in or travel to endemic areas, regular contact with high-risk adults, and human immunodeficiency virus infection in the child. Setting: Ambulatory clinic in South Bronx, New York, NY. Participants: Consecutive sample of 2920 children aged 1 to 18 years. Main Outcome Measures: Sensitivity, specificity, positive and negative predictive values of the questionnaire, and odds ratio (OR) of reactive skin test results. Results: The NYCDOH questionnaire identified 413 children (14%) as having at least 1 risk factor. Of these, 23 (5.6%) had a positive skin test result; 4 (0.16%) of the 2507 without risk factors had a positive result. Results for the full NYCDOH questionnaire were sensitivity, 85.2%; specificity, 86.0%; negative predictive value, 99.8%; positive predictive value, 5.4%; and OR, 35.2 (95% confidence interval, 12.1-102.4). Conclusion: The NYCDOH questionnaire is a valid instrument for identifying children for tuberculin skin testing.