ASSESSMENT OF THE RELIABILITY, VALIDITY, AND RESPONSIVENESS OF A SPANISH ASTHMA QUALITY-OF-LIFE QUESTIONNAIRE

A Spanish-language questionnaire designed for measuring the impact of asthma on quality of life in adults was developed. It was derived, by the application of a rigorous translation protocol, from a previously validated, English-language Asthma Quality of Life (AQL) questionnaire which had been developed in Australia. The aim of this study was to e valuate the psychometric properties of the Spanish AQL questionnaire u sing a crass-sectional and longitudinal design. Two hundred ninety-fou r clinically stable subjects with asthma (168 women, mean baseline for ced expiratory volume in 1 sec [FEV1] = 85% predicted), aged 17-70, at tended for the initial baseline assessment. All subjects completed the AQL question nai re and a fu II history and physical examination were performed. The clinical assessment of severity was based on the class ification recommended by the Global Initiative on Asthma (GINA). One w eek after the initial assessment subjects completed the AQL questionna ire for a second time. Six months later, subjects were assessed clinic ally and completed all the assessment measures at baseline. Principal components analysis of the AQL questionnaire responses at the baseline visit revealed a structure that was almost identical to that seen in the original English-language questionnaire. The questionnaire was sho wn to be internally consistent (Cronbach's alpha 0.91 for total score and 0.80-0.86 for the four subscales) and repeatable (intraclass corre lation coefficient 0.91 for the total scale and 0.78-0.92 for the subs cales). The finding of expected strong correlations with the subject's global assessment of severity (rho = 0.70) and dyspnea (rho = 0.63), a weak inverse correlation with FEV1 (rho = -0.17), and good discrimin ation among the four GINA severity categories (F-3,F-291 = 37.16, rho < 0.0001) supports the construct validity of the questionnaire. AQL sc ores increased with age (rho = 0.31) and were higher in women (rho < 0 .005). The AQL was responsive to both improvement (mean change 1.02, r ho < 0.0001) and deterioration (mean change -1.13, rho < 0.001) in the severity of asthma over a 6-month period. This disease-specific, Span ish-language AQL questionnaire was shown to have sound psychometric pr operties which make it suitable for use in cross-sectional or longitud inal studies where it is appropriate to assess the impact of asthma on the quality of life of individual patients.