In search of childhood asthma: questionnaire, tests of bronchial hyperresponsiveness, and clinical evaluation.

BACKGROUND The definition or diagnosis of asthma is a challenge for both clinicians and epidemiologists. Symptom history is usually supplemented with tests of bronchial hyperresponsiveness (BHR) in spite of their uncertainty in improving diagnostic accuracy. METHODS To assess the interrelationship between respiratory symptoms, BHR, and clinical diagnosis of asthma, the respiratory symptoms of 1633 schoolchildren were screened using a questionnaire (response rate 81.2%) and a clinical study was conducted in a subsample of 247 children. Data from a free running test and a methacholine inhalation challenge test were available in 218 children. The diagnosis of asthma was confirmed by a paediatric allergist. RESULTS Despite their high specificity (>0.97), BHR tests did not significantly improve the diagnostic accuracy after the symptom history: area under the receiver operator characteristic (ROC) curve was 0.90 for a logistic regression model with four symptoms and 0.94 for the symptoms with free running test and methacholine inhalation challenge results. On the other hand, BHR tests had low sensitivity (0.35-0.47), whereas several symptoms had both high specificity (>0.97) and sensitivity (>0.7) in relation to clinical asthma, which makes them a better tool for asthma epidemiology than BHR. CONCLUSIONS Symptom history still forms the basis for defining asthma in both clinical and epidemiological settings. BHR tests only marginally increased the diagnostic accuracy after symptom history had been taken into account. The diagnosis of childhood asthma should not therefore be overlooked in symptomatic cases with no objective evidence of BHR. Moreover, BHR should not be required for defining asthma in epidemiological studies.