Monitoring the quality of life of patients with asthma and COPD in general practice: psychometric properties of a 10-item questionnaire (the RIQ-MON10)

Abstract

Physicians need a good assessment of patients' perceived burden of the disease in order to provide adequate treatment and support. As doctors' judgements often differ from patients'own assessments, a manageable method to incorporate the latter into routine primary care might support patient-centered decision-making. For this purpose we converted a 55-item 7-point Likert scale questionnaire, measuring quality of life of patients with mild to moderate respiratory diseases (QOL-RIQ), into a short form. Prototypes proved applicable in normal care. We analysed the data of three controlled studies performed in primary care (N = 328, 502 and 555 patients). Procedures: inter-item correlations, scale distributions, Cronbach's alpha and factor analysis. A panel of 15 GP-experts judged the clinical relevance of the selection. Dyspnoea, FEV1 % pred, the MRC-ECCS, COOP/WONCA charts and the MOS-SF 36 served as external criteria to test validity and responsiveness. Item reduction resulted in a 10-item short form (alpha's. 87 to. 90), consisting of two 5-item factors: physical and emotional complaints and physical and social limitations. The panel (>70%) considered the selection fit for monitoring in routine care. The correlation of the shortlist with the original instrument (r=.89–.92) was high as well as with dyspnoea (r=.57–.60) and the generic health status instruments (r=.39–.59). As expected there was a low correlation with lung function (r=.10–.15). The short form shows a clinical relevant score difference (>.5) between upper and lower quartiles of the convergent instruments and a high correlation between the repeated scores in a stable group of patients (ICC =.82). The short form reflected the change in a group of 15 improved patients (SRM =.86). The short form questionnaire (RIQ-MON10) maintained the psychometric properties of the original instrument. It has good discriminative properties and seems promising as a practical tool for monitoring quality of life in routine primary care. Large-scaled testing of its reliability and sensitivity to change is planned.