Psychometric Properties of the Asthma Symptom Diary (ASD), a Diary for Use in Clinical Trials of Persistent Asthma

Abstract

No currently available asthma symptom diary has sufficient validation to be recommended for use as a core asthma outcome measure. The objective of this study was to provide validation data for the 10-item asthma symptom diary (ASD). Data were collected in a 4-week prospective, observational study. Subjects completed 3 study visits, completing the ASD twice daily at home for 28 days. Psychometric properties in terms of dimensionality, reliability, validity, and responsiveness were assessed. Data from 276 subjects were analyzed; mean age was 42.9 (standard deviation [SD]= 16.4) years, mean asthma duration was 23.3 (SD= 16.8) years, and 69.6% were female. Confirmatory factor and Rasch analysis supported the ASD as unidimensional and adequately measuring the spectrum of asthma symptom severity. High Cronbach's α (0.94) and intraclass correlation coefficients (0.89-0.95) supported reliability. A high correlation between the 7-day average ASD score and the Asthma Control Questionnaire (ACQ) total score (r= 0.75) and Asthma Quality of Life Questionnaire total scores (r=-0.76), and a moderate correlation with FEV1% predicted (r=-0.30) supported convergent validity. Significant differences (P < .001) between groups classified by ACQ scores supported known-group validity. The 7-day average ASD scores were responsive to change, with significantly higher score changes (P < .001) in responders versus nonresponders. Minimally important differences were calculated and found to be in the range of 0.1-0.3. Results of this study indicated that the ASD is a reliable and valid asthma symptom measure for use in adult and adolescent asthma patients to evaluate the effect of treatment on asthma in clinical trials.