A comparison of the original chronic respiratory questionnaire with a standardized version

BACKGROUND AND OBJECTIVES: The chronic respiratory questionnaire (CRQ), a widely used measure of health-related quality of life (HRQL) in patients with chronic airflow limitation, includes an individualized dyspnea domain (patients identify five important activities, and report the degree of dyspnea on a 7-point scale). Because the individualized domain is unwieldy in multicenter clinical trials, we developed a standardized version and tested its discriminative and evaluative properties. METHODS: We enrolled 51 patients who completed the standardized and individualized CRQ before starting a respiratory rehabilitation program, and again 3 months later. We calculated both cross-sectional and longitudinal correlations between the two versions and a number of other HRQL instruments, and tested the relative ability of the individualized and standardized versions of the CRQ to detect improvement with rehabilitation. RESULTS: The results of the individualized questions suggested greater dysfunction (lower scores) than did the standardized questions both at baseline (3.18 vs 3.92, p < 0.001) and follow-up (4.62 vs 4.84, p = 0.051). The standardized dyspnea domain showed superior discriminative validity. While both techniques detected important, statistically significant improvement with rehabilitation (individualized domain mean change, 1.44; 95% confidence interval [CI], 1.11 to 1.77 [p < 0.001]; standardized domain mean change, 0.92; 95% CI, 0.61 to 1.24 [p < 0.01]), the difference in effect was substantial and statistically significant (mean difference, 0.52; 95% CI, 0.22 to 0.82; p = 0.001). The two versions showed comparable longitudinal validity. CONCLUSIONS: A standardized version of the CRQ dyspnea domain improves the cross-sectional validity, maintains longitudinal validity, but reduces the responsiveness. By increasing sample size, investigators can use the more efficient standardized version of the CRQ without compromising validity.