The tuberculin skin test is considered to be the best method of detecting tuberculosis in patients who have no clinical signs or symptoms of disease. The recommended test is the intradermal Mantoux test, which should be read by a health care professional 48 to 72 hours after application. The Centers for Disease Control and Prevention (CDC), the American Thoracic Society (ATS) consider a positive test to be one with an induration of 15 mm or more in patients with no identifiable risk factors for tuberculosis. The AAP concurs, except in children four years of age or younger. In these children, the American Academy of Pediatrics (AAP) requires an induration of 10 mm or greater. Kendig and colleagues performed an observational study to evaluate the ability of health care professionals to accurately interpret tuberculin skin tests.

An intermediate-strength Mantoux tuberculin skin test was applied to the forearm of an adult man with a history of pulmonary tuberculosis who was known to have a positive skin test. Approximately 64 hours after the Mantoux test was applied, 107 health care professionals were asked to interpret the patient's skin test. The readers were told nothing about the patient's history and assumed they were reading the skin test of a healthy person who “converted” after exposure to a patient with active tuberculosis. They were provided with a tape measure, a ruler and a ball-point pen and were asked to record the results, their initials and their professional designations.

The “gold standard” used for interpretation of the patient's skin test was an induration of 15 mm or more. Of the 52 pediatricians who read the test, 17 (33 percent) interpreted the induration as less than 10 mm. The readings ranged from 5 to 15 mm (median: 10 mm). Of the 33 pediatric residents, 10 (30 percent) read the test as less than 10 mm (range: 5 to 15 mm; median: 10 mm). Of the 12 registered nurses participating in the study, four (33 percent) read the test as less than 10 mm, and three read it as 15 mm or greater. Of the 10 participating academicians, four (40 percent) read the tuberculin skin test as less than 10 mm. Of all the participants, only eight (7 percent) correctly charted an induration of 15 mm or more.

The authors conclude that there is a marked tendency on the part of health care professionals to underread tuberculin skin tests. Even if the previous cutoff level of 10-mm induration had been used, 33 percent of the study participants would have failed to identify this patient's positive tuberculosis skin test. The authors suggest this tendency to under-read raises some questions about the decision of the ATS, the CDC and the AAP to increase the positive induration measurement to 15 mm in low-risk adults and children.

In an accompanying editorial, Reichman regards the study as uncovering a “scandalous incompetence” and demands emergent intervention on the part of professional associations and regulatory bodies.

editor's note: The results of this study should be quite disturbing, particularly to physicians who take care of high-risk patients or practice in areas where the prevalence of tuberculosis is high. Many offices and clinics depend on non-physician personnel to interpret the Mantoux tests of patients who return 48 to 72 hours after the test is applied. If health care workers are indeed missing up to 30 percent of patients with a positive test, the public health ramifications are significant. Many of these patients are theoretically at risk for progressing to active tuberculosis and may be candidates for isoniazid prophylaxis.—j.k.