Should linking replace regression when mapping from profile-based measures to preference-based measures?

Peter M Fayers; Ron D Hays

doi:10.1016/j.jval.2013.12.002

Should linking replace regression when mapping from profile-based measures to preference-based measures?

Value Health. 2014 Mar;17(2):261-5. doi: 10.1016/j.jval.2013.12.002.

Authors

Peter M Fayers¹, Ron D Hays²

Affiliations

¹ Institute of Applied Health Sciences, University of Aberdeen, Aberdeen, UK; Department of Cancer Research and Molecular Medicine, Faculty of Medicine, Norwegian University of Science and Technology (NTNU), Trondheim, Norway. Electronic address: [email protected].
² UCLA Department of Medicine, Los Angeles, CA, USA.

Abstract

Background: Profile instruments are frequently used to assess health-related quality of life and other patient-reported outcomes. However, preference-based measures are required for health-economic cost-utility evaluations.

Results: Although regression-based approaches are commonly used to map from profile measures to preference measures, we show that this results in biased estimates because of regression to the mean.

Conclusions: Linking (scale-aligning) is proposed as an alternative.

Keywords: linking values; mapping functions; patient-reported outcomes; preference-based measures; profile instruments; scale-aligning; test equating.

Publication types

Research Support, N.I.H., Extramural
Research Support, U.S. Gov't, P.H.S.

MeSH terms

Clinical Trials as Topic / methods
Cost-Benefit Analysis
Humans
Observational Studies as Topic / methods
Outcome Assessment, Health Care / methods*
Quality of Life*
Regression Analysis
Research Design*
Self Report

Abstract

Publication types

MeSH terms

Grants and funding