Determination of asthma control using administrative data regarding short-acting beta-agonist inhaler purchase

BACKGROUND: Symptom control is a primary goal in asthma. We hypothesized that administrative data regarding rescue inhaler purchases may correlate with asthma symptom control. METHODS: We identified all patients who purchased short-acting beta-agonist (SABA) inhalers during the course of one year in the database of a Health Maintenance Organization (HMO). Primary physicians identified asthma patients and classified their asthma symptom control into three groups according to the Global Initiative for Asthma (GINA) guidelines. Asthma patients were asked to answer symptom questionnaires and grade their asthma control. SABA inhaler purchases were compared between asthma control groups as classified by the guidelines, the physicians and the patients. We also compared the agreement on asthma control between the three methods of classification. RESULTS: Of 241 asthma patients, 83 completed the questionnaires. Using the GINA guidelines criteria, 26 were symptom controlled, 46 were partially controlled and 11 were uncontrolled. SABA inhaler purchases were not significantly lower in the controlled group. Using patients' overall impression of their asthma control, the mean numbers of SABA inhalers purchased were 1.5, 4.4 and 6.4 per year in the controlled, partially controlled and uncontrolled groups, respectively (p = 0.03). Patients' classification of asthma control had better agreement (kappa = 0.34) with GINA guidelines than physician's' agreement (kappa = 0.05). CONCLUSION: When using administrative data for asthma patients, 2 or more SABA inhaler purchases in one year should alert the physician for the need for asthma control evaluation. Purchase of at least 4 SABA inhalers a year may be regarded as a marker for asthma that is not controlled.