ISPOR 20th Annual European Conference
Organized session: Wednesday, November 8, 2017: 08:45-09:45
COMPARING, CONTRASTING, AND VALIDATING HEALTH ECONOMIC DECISION MODELS: EXPERIENCES FROM THE LATEST MT. HOOD CHALLENGE IN DIABETES AND LESSONS FOR OTHER DISEASE AREAS
Discussion Leaders: Alan Brennan, PhD, Professor of Health Economics and Decision Modelling, School of Health and Related Research (ScHARR), University of Sheffield, Sheffield, United Kingdom; Mark Lamotte, MD, Cardiologist, Senior principal, RWES, QuintilesIMS, Zaventem, Belgium; Talitha Feenstra, PhD, Senior Researcher / Assistant Professor, RIVM /UMCG, Bilthoven, Netherlands; Michael Willis, PhD, Research Director, The Swedish Institute for Health Economics, Lund, Sweden
PURPOSE: In 1999, two groups of diabetes simulation modelers met on the slopes of Mt Hood near Portland, Oregon, simulated their models for a set of identical scenarios, and compared the results. Now, there have been 8 Mt. Hood Challenges, with up to 10 modeling groups meeting biennially to validate their models, and discuss methodology and data in what is arguably the most collegiate and yet competitive environment for debating models in the HEOR community. The purpose of this workshop is to share the concept of disease-specific, cross-modeling working meetings with modelers from all disease areas and to discuss practical considerations and best methods for this kind of forum.
DESCRIPTION: This session will start by reviewing the purpose, scope, structure, and practical implementation of the Mt. Hood Challenge congresses, at a level relevant for modelers without diabetes experience. A brief summary of the methods and results of the transparency challenge from 2016 will then be presented, illustrating the difficulty of reproducing results based on publications (even when the same model was used). The 2016 challenge has led to development of guidelines for the reporting of model-based results in diabetes. With a focus on a minimum set of key parameters and scenario characteristics, they substantially extend existing more general model transparency guidelines. The draft version will be presented and the floor will be opened for discussion. Important issues around academic intellectual property versus model sharing and transparency will be discussed. A similar venture for cross model validation and comparison in COPD will also be briefly discussed, with key learnings shared. The workshop will include an extended interactive Q&A session and live electronic voting via smartphone app, during which the audience will have the opportunity to discuss aspects of validation by cross model comparisons as in the Mt Hood Challenges with the panel.