Psychometric properties of the PROMIS Preference score (PROPr) in patients with rheumatological and psychosomatic conditions

Abstract

Abstract Background The PROMIS Preference score (PROPr) is a new generic preference-based health-related quality of life (HRQoL) score that can be used as a health state utility (HSU) score for quality-adjusted life years (QALYs) in cost-utility analyses (CUAs). It is the first HSU score based on item response theory (IRT) and has demonstrated favorable psychometric properties in first analyses. The PROPr combines the seven PROMIS domains: cognition, depression, fatigue, pain, physical function, sleep disturbance, and ability to participate in social roles and activities. It was developed based on preferences of the US general population. The aim of this study was to validate the PROPr in a German inpatient sample and to compare it to the EQ-5D. Methods We collected PROPr and EQ-5D-5L data from 141 patients undergoing inpatient treatment in the rheumatology and psychosomatic departments. We evaluated the criterion and convergent validity, and ceiling and floor effects of the PROPr and compared those characteristics to those of the EQ-5D. Results The mean PROPr (0.26, 95% CI: 0.23; 0.29) and the mean EQ-5D (0.44, 95% CI: 0.38; 0.51) scores differed significantly (d = 0.18, p < 0.001). Compared to the EQ-5D, the PROPr scores were less scattered across the measurement range which has resulted in smaller confidence intervals of the mean scores. The Pearson correlation coefficient between the two scores was r = 0.72 (p < 0.001). Both scores showed fair agreement with an Intraclass Correlation Coefficient (ICC) of 0.48 (p < 0.05). The PROPr and EQ-5D demonstrated similar discrimination power across sex, age, and conditions. While the PROPr showed a floor effect, the EQ-5D showed a ceiling effect. Conclusion The PROPr measures HSU considerably lower than the EQ-5D as a result of different construction, anchors and measurement ranges. Because QALYs derived with the EQ-5D are widely considered state-of-the-art, application of the PROPr for QALY measurements would be problematic.