INTRODUCTION: The EQ-5D provides community-based preference weights (utilities) for calculating quality adjusted life years (QALYs) in cost-utility analysis. This study aimed to analyze the responsiveness of EQ-5D-based utilities in patients with alcohol dependence. METHODS: In an observational longitudinal cohort study of alcohol-dependent patients (N=52), three different EQ-5D-based utilities (EQ-Index United Kingdom, United States, and Germany) were calculated and compared with the scores of the WHOQOL-BREF (mental domain), HoNOS (total score), and GAF at baseline and at 18 months. Patients were categorized with more/same/less problems according to their self reported consumption of alcohol (patient-based anchor) and their problem status due to alcohol consumption using a question of the HoNOS scale (clinician-based anchor). Effect sizes (ES) were used to compare longitudinal changes in scores within each group; regression analysis was used to derive difference scores in health status associated with a shift from "same" to "less" problems according to the two anchors. RESULTS: ES were rather trivial to medium for all EQ-5D indices (ranging from |0.10| to |0.59|) related to a shift from "same" to "less" problems in the two anchors. Differences in scores of the EQ-5D indices revealing a shift from "same" to "less" problems according to the two anchors were not significant. CONCLUSIONS: These results suggest that the EQ-5D indices are less responsive and require larger patient samples to detect meaningful clinical differences compared to the other used instruments. Additional research is needed to compare societal and clinical views on relevant change in health status in this patient group.
INTRODUCTION: The EQ-5D provides community-based preference weights (utilities) for calculating quality adjusted life years (QALYs) in cost-utility analysis. This study aimed to analyze the responsiveness of EQ-5D-based utilities in patients with alcohol dependence. METHODS: In an observational longitudinal cohort study of alcohol-dependent patients (N=52), three different EQ-5D-based utilities (EQ-Index United Kingdom, United States, and Germany) were calculated and compared with the scores of the WHOQOL-BREF (mental domain), HoNOS (total score), and GAF at baseline and at 18 months. Patients were categorized with more/same/less problems according to their self reported consumption of alcohol (patient-based anchor) and their problem status due to alcohol consumption using a question of the HoNOS scale (clinician-based anchor). Effect sizes (ES) were used to compare longitudinal changes in scores within each group; regression analysis was used to derive difference scores in health status associated with a shift from "same" to "less" problems according to the two anchors. RESULTS:ES were rather trivial to medium for all EQ-5D indices (ranging from |0.10| to |0.59|) related to a shift from "same" to "less" problems in the two anchors. Differences in scores of the EQ-5D indices revealing a shift from "same" to "less" problems according to the two anchors were not significant. CONCLUSIONS: These results suggest that the EQ-5D indices are less responsive and require larger patient samples to detect meaningful clinical differences compared to the other used instruments. Additional research is needed to compare societal and clinical views on relevant change in health status in this patient group.
Authors: Jeffrey M Pyne; Shanti Tripathi; Michael French; Kathryn McCollister; Richard C Rapp; Brenda M Booth Journal: Addiction Date: 2011-03 Impact factor: 6.526
Authors: Holly N Essex; Ian R White; Zarnie Khadjesari; Stuart Linke; Jim McCambridge; Elizabeth Murray; Steve Parrott; Christine Godfrey Journal: Qual Life Res Date: 2013-09-13 Impact factor: 4.147