Premium
Discrete choice experiments for complex health‐care decisions: does hierarchical information integration offer a solution?
Author(s) -
van HelvoortPostulart Debby,
Dellaert Benedict G. C.,
van der Weijden Trudy,
von Meyenfeldt Maarten F.,
Dirksen Carmen D.
Publication year - 2009
Publication title -
health economics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.55
H-Index - 109
eISSN - 1099-1050
pISSN - 1057-9230
DOI - 10.1002/hec.1411
Subject(s) - construct (python library) , mixed logit , logit , consistency (knowledge bases) , health care , construct validity , guideline , scale (ratio) , logistic regression , ordered logit , computer science , operations management , actuarial science , medicine , nursing , economics , machine learning , patient satisfaction , artificial intelligence , physics , pathology , quantum mechanics , programming language , economic growth
Abstract This paper describes an application of hierarchical information integration (HII) discrete choice experiments. We assessed theoretical and construct validity, as well as internal consistency, to investigate whether HII can be used to investigate complex multi‐faceted health‐care decisions (objective 1). In addition, we incorporated recent advances in mixed logit modelling (objective 2). Finally, we determined the response rate and predictive ability to study the feasibility of HII to support health‐care management (objective 3). The clinical subject was the implementation of the guideline for breast cancer surgery in day care, which is a complex process that involves changes at the organizational and management levels, as well as the level of health‐care professionals and that of patients. We found good theoretical and construct validity and satisfactory internal consistency. The proposed mixed logit model, which included repeated measures corrections and subexperiment error scale variations, also performed well. We found a poor response, but the model had satisfactory predictive ability. Therefore, we conclude that HII can be used successfully to study complex multi‐faceted health‐care decisions (objectives 1 and 2), but that the feasibility of HII to support health‐care management, in particular in challenging implementation projects, seems less favourable (objective 3). Copyright © 2008 John Wiley & Sons, Ltd.