A comparison of the validity of different diagnostic tests in adults with asthma.

STUDY OBJECTIVES: Diagnosing asthma is not always easy, and there are times when objective tests can be helpful. The extent to which these tests alter the probability of asthma depends on how much more commonly the test result is positive in subjects with asthma compared to healthy subjects and particularly subjects with conditions that are commonly confused with asthma. We set out to compare the sensitivity and specificity of different tests in this setting. DESIGN: Single-center, cross-sectional, observational study. SETTING: Teaching hospital. PATIENTS: Twenty-one healthy control subjects, 69 patients with asthma, and 20 subjects referred to the hospital with a diagnosis of asthma who were found to have alternative explanations for their symptoms (i.e., pseudoasthma). INTERVENTIONS: We measured methacholine airway responsiveness, the maximum within-day peak expiratory flow amplitude mean percentage (derived from twice-daily readings for > 2 weeks), the FEV(1)/FVC ratio, the percentage change in FEV(1) 10 min after the administration of 200 microg inhaled albuterol, and the differential eosinophil count in blood and induced sputum. We derived normal ranges (from the 95% upper or lower limit for healthy subjects), sensitivity, and specificity (ie, the percentage of subjects with pseudoasthma who had negative test results). RESULTS: Most tests were less specific when the reference population was composed of subjects with conditions that can be confused with asthma. Methacholine airway responsiveness and the sputum differential eosinophil count were the most sensitive (91% and 72%, respectively) and specific (90% and 80%, respectively) tests. CONCLUSION: We conclude that methacholine airway responsiveness and the sputum differential eosinophil count are the most useful objective tests in patients with mild asthma.