Is best-worst scaling suitable for health state valuation?: A comparison with discrete choice experiments

Nicolas Krucien; Verity Watson; Mandy Ryan

doi:10.1002/hec.3459

Is best-worst scaling suitable for health state valuation? A comparison with discrete choice experiments

Nicolas Krucien^* (Corresponding Author), Verity Watson, Mandy Ryan

^*Corresponding author for this work

Research output: Contribution to journal › Article › peer-review

37 Citations (Scopus)

25 Downloads (Pure)

Abstract

Health utility indices (HUIs) are widely used in economic evaluation. The best-worst scaling (BWS) method is being used to value dimensions of HUIs. However, little is known about the properties of this method. This paper investigates the validity of the BWS method to develop HUI, comparing it to another ordinal valuation method, the discrete choice experiment (DCE). Using a parametric approach, we find a low level of concordance between the two methods, with evidence of preference reversals. BWS responses are subject to decision biases, with significant effects on individuals' preferences. Non parametric tests indicate that BWS data has lower stability, monotonicity and continuity compared to DCE data, suggesting that the BWS provides lower quality data. As a consequence, for both theoretical and technical reasons, practitioners should be cautious both about using the BWS method to measure health-related preferences, and using HUI based on BWS data. Given existing evidence, it seems that the DCE method is a better method, at least because its limitations (and measurement properties) have been extensively researched. Copyright © 2016 John Wiley & Sons, Ltd.

Original language	English
Pages (from-to)	e1-e16
Number of pages	16
Journal	Health Economics
Volume	26
Issue number	12
Early online date	4 Dec 2016
DOIs	https://doi.org/10.1002/hec.3459
Publication status	Published - Dec 2017

Bibliographical note

The University of Aberdeen (UoA) and the Chief Scientist Office (CSO) of the Scottish Government Health and Social Care Directorates fund the Health Economics Research Unit (HERU). We thank all participants who took part in the study and the WH Ross foundation that supported the data collection. We also thank authors of the original study (Mary Kilonzo, Jennifer Burr and Luke Vale) for their contribution to questionnaire design and data collection. The views expressed in this paper are those of the authors only and not those of the funding bodies.

Keywords

Best-worst scaling
Discrete choice experiments
Health utility
Stated preferences

Access to Document

10.1002/hec.3459Licence: Unspecified

Accepted Manuscript
"This is the peer reviewed version of the following article: Krucien, N, Watson, V & Ryan, M 2017, 'Is best-worst scaling suitable for health state valuation?: A comparison with discrete choice experiments' Health Economics, vol 26, no. 12, pp. e1 - e16. DOI: 10.1002/hec.3459 . This article may be used for non-commercial purposes in accordance with Wiley Terms and Conditions for Self-Archiving."
Accepted author manuscript, 842 KBLicence: Unspecified

Mandy Ryan
- School of Medicine, Medical Sciences & Nutrition, Health Economics Research Unit - Director of H E R U
- Institute of Applied Health Sciences
Person: Academic

Cite this

@article{9aabeb977b0f44a985d8ee2dde2a18e8,

title = "Is best-worst scaling suitable for health state valuation?: A comparison with discrete choice experiments",

abstract = "Health utility indices (HUIs) are widely used in economic evaluation. The best-worst scaling (BWS) method is being used to value dimensions of HUIs. However, little is known about the properties of this method. This paper investigates the validity of the BWS method to develop HUI, comparing it to another ordinal valuation method, the discrete choice experiment (DCE). Using a parametric approach, we find a low level of concordance between the two methods, with evidence of preference reversals. BWS responses are subject to decision biases, with significant effects on individuals' preferences. Non parametric tests indicate that BWS data has lower stability, monotonicity and continuity compared to DCE data, suggesting that the BWS provides lower quality data. As a consequence, for both theoretical and technical reasons, practitioners should be cautious both about using the BWS method to measure health-related preferences, and using HUI based on BWS data. Given existing evidence, it seems that the DCE method is a better method, at least because its limitations (and measurement properties) have been extensively researched. Copyright {\textcopyright} 2016 John Wiley & Sons, Ltd.",

keywords = "Best-worst scaling, Discrete choice experiments, Health utility, Stated preferences",

author = "Nicolas Krucien and Verity Watson and Mandy Ryan",

note = "The University of Aberdeen (UoA) and the Chief Scientist Office (CSO) of the Scottish Government Health and Social Care Directorates fund the Health Economics Research Unit (HERU). We thank all participants who took part in the study and the WH Ross foundation that supported the data collection. We also thank authors of the original study (Mary Kilonzo, Jennifer Burr and Luke Vale) for their contribution to questionnaire design and data collection. The views expressed in this paper are those of the authors only and not those of the funding bodies.",

year = "2017",

month = dec,

doi = "10.1002/hec.3459",

language = "English",

volume = "26",

pages = "e1--e16",

journal = "Health Economics",

issn = "1057-9230",

publisher = "John Wiley and Sons Ltd",

number = "12",

}

TY - JOUR

T1 - Is best-worst scaling suitable for health state valuation?

T2 - A comparison with discrete choice experiments

AU - Krucien, Nicolas

AU - Watson, Verity

AU - Ryan, Mandy

N1 - The University of Aberdeen (UoA) and the Chief Scientist Office (CSO) of the Scottish Government Health and Social Care Directorates fund the Health Economics Research Unit (HERU). We thank all participants who took part in the study and the WH Ross foundation that supported the data collection. We also thank authors of the original study (Mary Kilonzo, Jennifer Burr and Luke Vale) for their contribution to questionnaire design and data collection. The views expressed in this paper are those of the authors only and not those of the funding bodies.

PY - 2017/12

Y1 - 2017/12

N2 - Health utility indices (HUIs) are widely used in economic evaluation. The best-worst scaling (BWS) method is being used to value dimensions of HUIs. However, little is known about the properties of this method. This paper investigates the validity of the BWS method to develop HUI, comparing it to another ordinal valuation method, the discrete choice experiment (DCE). Using a parametric approach, we find a low level of concordance between the two methods, with evidence of preference reversals. BWS responses are subject to decision biases, with significant effects on individuals' preferences. Non parametric tests indicate that BWS data has lower stability, monotonicity and continuity compared to DCE data, suggesting that the BWS provides lower quality data. As a consequence, for both theoretical and technical reasons, practitioners should be cautious both about using the BWS method to measure health-related preferences, and using HUI based on BWS data. Given existing evidence, it seems that the DCE method is a better method, at least because its limitations (and measurement properties) have been extensively researched. Copyright © 2016 John Wiley & Sons, Ltd.

AB - Health utility indices (HUIs) are widely used in economic evaluation. The best-worst scaling (BWS) method is being used to value dimensions of HUIs. However, little is known about the properties of this method. This paper investigates the validity of the BWS method to develop HUI, comparing it to another ordinal valuation method, the discrete choice experiment (DCE). Using a parametric approach, we find a low level of concordance between the two methods, with evidence of preference reversals. BWS responses are subject to decision biases, with significant effects on individuals' preferences. Non parametric tests indicate that BWS data has lower stability, monotonicity and continuity compared to DCE data, suggesting that the BWS provides lower quality data. As a consequence, for both theoretical and technical reasons, practitioners should be cautious both about using the BWS method to measure health-related preferences, and using HUI based on BWS data. Given existing evidence, it seems that the DCE method is a better method, at least because its limitations (and measurement properties) have been extensively researched. Copyright © 2016 John Wiley & Sons, Ltd.

KW - Best-worst scaling

KW - Discrete choice experiments

KW - Health utility

KW - Stated preferences

U2 - 10.1002/hec.3459

DO - 10.1002/hec.3459

M3 - Article

C2 - 27917560

SN - 1057-9230

VL - 26

SP - e1-e16

JO - Health Economics

JF - Health Economics

IS - 12

ER -

Is best-worst scaling suitable for health state valuation? A comparison with discrete choice experiments

Abstract

Bibliographical note

Keywords

Access to Document

Fingerprint

Profiles

Mandy Ryan

Cite this