Using Best-Worst Scaling to Investigate Preferences in Health Care

Kei Long Cheung*, Ben F. M. Wijnen, Ilene L. Hollin, Ellen M. Janssen, John F. Bridges, Silvia M. A. A. Evers, Mickael Hiligsmann

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review


Best-worst scaling (BWS) is becoming increasingly popular to elicit preferences in health care. However, little is known about current practice and trends in the use of BWS in health care. This study aimed to identify, review and critically appraise BWS in health care, and to identify trends over time in key aspects of BWS. A systematic review was conducted, using Medline (via Pubmed) and EMBASE to identify all English-language BWS studies published up until April 2016. Using a predefined extraction form, two reviewers independently selected articles and critically appraised the study quality, using the Purpose, Respondents, Explanation, Findings, Significance (PREFS) checklist. Trends over time periods (aecurrency sign2010, 2011, 2012, 2013, 2014 and 2015) were assessed further. A total of 62 BWS studies were identified, of which 26 were BWS object case studies, 29 were BWS profile case studies and seven were BWS multi-profile case studies. About two thirds of the studies were performed in the last 2 years. Decreasing sample sizes and decreasing numbers of factors in BWS object case studies, as well as use of less complicated analytical methods, were observed in recent studies. The quality of the BWS studies was generally acceptable according to the PREFS checklist, except that most studies did not indicate whether the responders were similar to the non-responders. Use of BWS object case and BWS profile case has drastically increased in health care, especially in the last 2 years. In contrast with previous discrete-choice experiment reviews, there is increasing use of less sophisticated analytical methods.
Original languageEnglish
Pages (from-to)1195-1209
Issue number12
Publication statusPublished - Dec 2016

Cite this