Background: A questionnaire alone may not be an adequate screening tool for asthma.
Objective: To determine whether an asthma questionnaire used in combination with an exercise step test is better than a questionnaire alone in screening for asthma in children and to evaluate the validity of a rhinitis questionnaire in determining atopy.
Methods: The International Study of Asthma and Allergies in Childhood (ISAAC) asthma core questionnaire was used to screen for asthma in 307 inner-city first through third graders. All children who had scores consistent with a diagnosis of asthma underwent step testing, as did a subset of children who had negative overall scores. All children who had inconsistent asthma scores and step test results underwent methacholine challenge testing. The same 307 children underwent rhinitis screening and children who had one or more positive responses on the ISAAC rhinitis questionnaire underwent skin testing as did a subset of children who had all negative responses.
Results: Three hundred of 307 asthma and rhinitis questionnaires were returned. Twenty-eight children (9%) had global asthma scores that were considered to be positive (5 or above). Twenty-four of these children underwent step testing as did 34 randomly selected children who had negative global asthma scores. Thirty-one (91%) of the 34 children who had negative global asthma scores had negative step tests. Similarly, 20 of 24 children (83%) of the children who had positive global asthma scores had negative step tests. Only 4 children who had positive global asthma scores were step test-positive or had reversible airway obstruction at baseline. Using a positive methacholine challenge as the gold standard for establishing bronchial hyperresponsiveness, the global asthma score derived from the eight-item ISAAC asthma questionnaire yielded a sensitivity of 64%, a specificity of 11%, a positive predictive value of 47%, and a negative predictive value of 20%. Comparing the six-item ISAAC rhinitis questionnaire results to the gold standard, skin test reactivity, the questionnaire yielded a sensitivity of 76%, a specificity of 21%, a positive predictive value of 56%, and a negative predictive value of 40%.
Conclusions: Step testing was not useful as a screening tool for asthma. In addition, the ISAAC asthma questionnaire may not be a good asthma screening tool for inner-city pediatric populations, especially if the form is self-administered. Investigators should first validate both the ISAAC asthma and rhinitis screening questionnaires in the particular population to be studied before widespread asthma and allergy screening efforts are initiated using these tools.