Reliability and predictive validity of the Asthma Control Test administered by telephone calls using speech recognition technology

Abstract

The Asthma Control Test (ACT) has been validated in a paper and pencil version but has not been validated for use by telephone. The purpose of this study was to provide validation data for the ACT administered by interactive telephone calls using speech recognition technology. The ACT was administered to patients who confirmed a diagnosis of physician-diagnosed asthma, and information regarding race/ethnicity, smoking, and asthma course was also obtained during the call. Asthma emergency department visits, hospitalizations, and oral corticosteroid and beta-agonist canister dispensings were assessed for the 12 months after the date of each patient's call. Internal consistency reliability and predictive validity were assessed. Asthma Control Test scores (higher indicates better control) were completed by 2244 patients and were inversely related to black or Hispanic race/ethnicity and smoking. Reliability was 0.83. ACT scores were significantly related to emergency hospital care and oral corticosteroid and beta-agonist dispensings over the period of the subsequent 6 and 12 months. After adjusting for demographic characteristics, a score < or = 15 was associated significantly with an increased 12-month risk of emergency hospital care (odds ratio [OR], 2.5), oral corticosteroid dispensings (OR, 2.6) and dispensing of more than 6 beta-agonist canisters (OR, 6.8) compared with a score > or = 20. These data support the reliability and predictive validity of the ACT administered by interactive telephone calls using speech recognition technology. The ACT can be used for outreach or follow-up by means of interactive telephone calls using speech recognition technology.