American translation, modification, and validation of the St. George's Respiratory Questionnaire

Abstract

Background The St. George's Respiratory Questionnaire (SGRQ) is a 50-item health status survey specific for chronic obstructive pulmonary disease (COPD) and other respiratory diseases that is available in British English but not American English. The SGRQ's symptom-reporting component requires a 1-year reporting period, which may be too long for reliable and accurate patient recall. Objectives The objectives of the present study were to translate the SGRQ from British to American English, modify the reporting period of the symptom-reporting component from 1 year to 1 month, and assess the reliability, validity, and sensitivity to change of this translated modified version in a sample of patients with COPD. Methods Based on input from American patients with COPD and health professionals, the SGRQ was translated into American English (SGRQ-A) and then translated back to British English. For SGRQ-A reliability and validity studies, patients were asked to report symptoms experienced over 1 year (reporting period in the original SGRQ) and 1 month (modification made to SGRQ-A). We evaluated 102 patients with COPD (50% female; mean age, 68 years; mean forced expiratory volume in 1 second [FEV,], 1.01 L) at an administrative session before and after completion of a pulmonary rehabilitation program. The SGRQ-A, Chronic Respiratory Disease Questionnaire (CRQ), 36-Item Short Form Health Survey (SF-36), 6-minute walk (6MW), Medical Research Council (MRC) Dyspnea scale, and pulmonary function tests (FEV 1 and % predicted FEV 1) were used in the assessment battery. Results The SGRQ-A showed good agreement with the original SGRQ when translated back to British English. Internal reliability (Cronbach a) was >0.70 for all SGRQ-A components except the 1-year symptom-reporting component. Test-retest intraclass correlations were 0.795 to 0.900. Construct validity was strengthened when all SGRQ-A components (except 1-year symptoms and most 1 -month symptoms) correlated (P ≤ 0.01) with the MRC Dyspnea scale, 6MW, all SF-36 concept scores, and 80% of CRQ domains (r = 0.30-0.72). Discriminate validity was demonstrated when all components of the SGRQ-A with the modified 1- month symptom-reporting period were shown to discriminate better between disease-severity groups (based on patient selfreports of disease severity) than did pulmonary function tests and the 6MW. Responsiveness of the SGRQ-A to change in health status was demonstrated when scores on the Symptoms-1 month and Total-1 month components detected significant improvements in patients' health status (P = 0.02 and P = 0.04, respectively). Conclusion The SGRQ-A with a modified 1-month symptom-reporting period demonstrated reliability and validity in this sample of patients with COPD.