Understanding and Predicting Choice Behaviour in Health: Preference Elicitation and Analysis

2025 COURSE DATES: 12 - 14 may  

This course will be available hybrid and in-person. We strongly recommend in-person attendance where possible. The venue is Green Templeton College, Oxford and online via zoom. Please select upon booking if you are attending in-person or online.

book course 

WHAT THE COURSE IS ABOUT

Methods for understanding how choices are applied in health care to assess preferences, analyse policies, help to develop randomised controlled trials, conduct non-market valuation, and forecast behaviour. This course teaches introductory methods for devising experiments and modelling choice data in health care.

Increasingly, health researchers need to understand decision-making in a wide range of health settings. For example, what drives smokers to choose cigarettes or e-cigarettes? What is a realistic non-inferiority margin to use in my clinical trial? How does society value different aspects of quality of life? What drives individuals to take, or refuse, vaccines? What kind of people prefer drug A over drug B?

Whether data are generated in experiments, such as discrete choice experiments, or are taken from recording individuals’ real-world behaviour, choice models are needed to understand these behaviours. The course provides the tools for understanding health-based choice behaviour, an introduction to choice experiments, and an introduction to choice modelling. 

Lectures and practical sessions that form part of the course enable participants to learn and apply these techniques.

The course is designed for those who need to perform discrete choice analysis in healthcare and those who need to understand the issues that health researchers face when performing these analyses. This could include researchers and decision makers from public, commercial, and academic organisations concerned with understanding health preferences.  We welcome participants from a wide variety of organisations and from all over the world. If you are unsure as to whether the course is suitable for you, please email john.buckell@ndph.ox.ac.uk who will be happy to advise.

PREREQUISITES

There are no formal prerequisites for attendance, but participants should have some understanding of basic statistics and regression techniques, including logistic regression and multinomial logistic regression. Choice modelling will be conducted using R (using the Apollo package for choice models), and some familiarity with R software is advised. Ngene software will be used for experimental design, though no prior experience is required.

Laptop computers or tablets with R and R studio installed will be required.

•    To provide detailed study of the methods of discrete choice analysis for health care

•    To provide introduction to the theory and application of discrete choice experiments in health care

•    To provide introduction to the theory and application of discrete choice models in health care

•    To give participants experience of computer-based application with exercises

The course will consist of three days' classroom lectures and practical sessions. These will be held from 9am - 5pm each day. Sessions include lectures for course materials, and practical sessions for choice models and experimental designs.

We will cover the following topics:

• Introduction to choice models, data types, and analysing preferences in health
• Model specification, estimation, post-estimation analyses, graphing outputs, and interpretation
• Experimental design and case studies in health
• Heterogeneity: deterministic tastes, random tastes, and attitudes
• Forecasting choice behaviour

As with past courses, the schedule/content may be altered in response to class participants' questions and needs. Questions and discussion of participants' applications are highly encouraged. 

This course is taught in English and a certificate of participation will be issued post course.

Note:  All exercises require R and R studio software. Ngene software is recommended for the design sessions, but alternatives are available.

Note:  This course requires a minimum number of participants one month before it is due to take place. Please book early to avoid disappointment

For booking enquiries, please contact the HERC Administration team at herc@ndph.ox.ac.uk

COURSE INSTRUCTORS

Professor Stephane Hess is the director of Choice Modelling Center and Professor of Choice Modelling at the University of Leeds. He is an expert in developing advanced choice models and analysing choice behaviour, with theoretical and empirical contributions across different fields. He is the author of Apollo software in R (with David Palma) and is also the founding editor and editor-in-chief of the Journal of Choice Modelling.

Dr John Buckell is a senior research fellow in health economics at the Health Economics Research Centre at the University of Oxford. He has worked on choice models in health markets including tobacco, obesity, and genomics. He has made methodological and policy making contributions in health.

 Dr Thomas Hancock is a Research Fellow at CMC, University of Leeds. His main research interests are understanding the decision-making process, the integration of (econometric) choice modelling and mathematical psychology, model specification and interpretation, and moral choice behaviour. His work thus far has focussed on the implementation of decision field theory and ideas from quantum cognition within choice modelling.