Quantification of dyspnoea using descriptors:development and initial testing of the Dyspnoea-12

Rationale: Dyspnoea is a debilitating and distressingsymptom that is reflected in different verbal descriptors.Evidence suggests that dyspnoea, like pain perception,consists of sensory quality and affective components. Theobjective of this study was to develop an instrument thatmeasures overall dyspnoea severity using descriptors thatreflect its different aspects.Methods: 81 dyspnoea descriptors were administered to123 patients with chronic obstructive pulmonary disease(COPD), 129 with interstitial lung disease and 106 withchronic heart failure. These were reduced to 34 itemsusing hierarchical methods. Rasch analysis informeddecisions regarding further item removal and fit to theunidimensional model. Principal component analysis (PCA)explored the underlying structure of the final item set.Validity and reliability of the new instrument were furtherassessed in a separate group of 53 patients with COPD.Results: After removal of items with hierarchicalmethods (n=47) and items that failed to fit the Raschmodel (n=22), 12 were retained. The ‘‘Dyspnoea-12’’had good internal reliability (Cronbach’s alpha=0.9) andfit to the Rasch model (x2 p=0.08). Items patterned intotwo groups called ‘‘physical’’(n=7) and ‘‘affective’’(n=5). In the separate validation study, Dyspnoea-12correlated with the Hospital Anxiety and Depression Scale(anxiety r=0.51; depression r=0.44, p,0.001,respectively), 6-minute walk distance (r=20.38,p,0.01) and MRC (Medical Research Council) grade(r=0.48, p,0.01), and had good stability over time(intraclass correlation coefficient=0.9, p,0.001).Conclusion: Dyspnoea-12 fulfills modern psychometricrequirements for measurement. It provides a global scoreof breathlessness severity that incorporates both ‘‘physical’’and ‘‘affective’’ aspects, and can measure dyspnoeain a variety of diseases