Construct validity of the EUROHIS-QOL for rehabilitation assessment

Abstract

The use of the EUROHIS as a brief measure of quality of life (QOL) in applied settings is increasingly commonplace, including in rehabilitation assessment contexts. However, there are concerns about the validity of data produced by the measure, chief amongst which is the latent structure underlying scores. This article reports on research conducted to investigate the dimensionality of scores derived from the EUROHIS. In addition, the factorial invariance of the retained model across gender as well as latent mean differences in QOL over age, employment status, and psychiatric severity were examined. Based on 251 responses to the EUROHIS by compensable accident victims, support was found for a complex one-factor model, which was found to be partially replicable across gender. Some evidence for differential item functioning across gender, age, and employment status was found. Finally, a U-shaped effect of age on QOL, characterized by a mid-life nadir, as well as effects of psychiatric severity on QOL, and a marginally significant effect of employment status were evident. Collectively, though the results of the present study yield validation data for the EUROHIS, they also raise concerns about the measure. We offer some tentative guidelines for working with the measure for both researchers and practitioners.