Spotlight Sessions

Spotlight sessions highlight timely topics in HEOR and promote areas of innovation of interest to the ISPOR community.

Monday, May 20, 2019, 2:15PM - 3:15PM

Global Developments in Artificial Intelligence and Machine Learning in Healthcare

Description
In this Spotlight Session, an international panel of experts will focus on global developments in artificial intelligence (AI) and machine learning (ML) in healthcare. The session will clarify the distinctions between AI and ML and the different forms of each. Trends in AI and ML will be discussed from the perspectives of the North America and European regions. The panelists will discuss developments in the healthcare data infrastructure within each region and the challenges and opportunities posed by these developments. For example, complex data structures and large volumes of unstructured data create challenges for traditional statistical methods but opportunities for machine learning methods. On the other hand, ML methods have traditionally been used to characterize the underlying structure of data, as well as for prediction and classification. Can they also be used for causal inference? How are AI and ML methods currently being used in healthcare delivery, drug discovery, health technology assessment, regulatory approval and safety surveillance? How might this change over the next decade?
Moderator and Speakers

Moderator:

William Crown, PhD, Chief Scientific Officer, Optum Labs, Cambridge, USA

Speakers:

  • Pall Jonsson, PhD, Associate Director, Research and Development, National Institute for Health and Care Excellence, London, United Kingdom

Monday, May 20, 2019, 2:15PM - 3:15PM

Back To the Future in Value in Health

Description

Over the last 20 years 3 major themes have dominated the content of Value in Health. First, many papers have discussed the nature of ‘value’ in health and healthcare, including the role of patient-reported outcomes, the construction and use of the quality-adjusted life-year and the methods and application of the various preference-based measures. Secondly, decision-analytic modeling has been the predominant methodology in the empirical studies published in the journal. There has been debate about alternative types of models, whether particular models are fit for purpose in decision-making and the need for more validation and transparency in modeling. Thirdly, several papers have explored the use of health economics and outcomes research (HEOR) in the broader context of health technology assessment (HTA) and comparative effectiveness research. Papers have compared and contrasted the applications of HEOR in different decision-making contexts, its implications for the pricing and reimbursement of health technologies, and the implications for patient access.


The speakers in this session, who have all been Editors-in-Chief of Value in Health, will discuss how these themes have evolved over the years, which issues have been resolved and which not. More importantly, they will give their views on which issues are likely to have more prominence in the future and whether new themes will emerge. C. Daniel Mullins will address the theme of the nature of value in health, Joel Hay will address the theme of decision-analytic modeling in health and healthcare and Josephine Mauskopf will address the theme of HEOR in the broader context of HTA and comparative effectiveness research. In the general discussion there will be opportunities for members of the audience to share their views on the likely directions for HEOR in the future.

Moderator and Speakers

Moderator:

Michael Drummond, PhD, Professor of Health Economics, Centre for Health Economics, University of York, York, UK

Panelists:

  • C. Daniel Mullins, PhD, Professor and Chair, PHSR Department, University of Maryland, Baltimore, MD, USA
  • Joel Hay, PhD, Professor of Pharmaceutical and Health Economics, USC School of Pharmacy, Los Angeles, CA, USA
  • Josephine Mauskopf, PhD, Vice President of Health Economics, RTI Health Solutions, Durham, NC, USA

 

 

Monday, May 20, 2019, 2:15PM - 3:15PM

Beyond One Size Fits All: The Importance of Heterogeneity in Preference Research

Description
This workshop will help participants to make informed decisions about how to analyse and interpret preference heterogeneity in stated preference studies, based on methodological and pragmatic considerations.

 

Policy makers have shown an increased interest in considering preference data in their decision making. For example, the US Food and Drug Administration (FDA) has issued guidance on the use of patient preference methods to inform quantitative benefit-risk assessments. A central aspect of such policy making is the definition of and distinction between target populations. Preferences may be heterogeneous between and within populations. Not considering such preference heterogeneity can lead to bias, and ultimately to suboptimal policy advice. However, issues with survey development and the study sampling framework can hinder the detection of preference heterogeneity. Additionally, a wide range of methods for capturing preference heterogeneity exists, such that best-practice and pragmatic considerations need thoughtful balancing..

Dr Ho will present on the relevance of preference heterogeneity for decision makers, including: how heterogeneity has influenced FDA’s decisions, and guidance on capturing preference heterogeneity. Dr Gonzalez will use case studies to illustrate implications of preference heterogeneity for product development out to outline important methodological considerations. Dr Heidenreich will present best practice for measuring preference heterogeneity, instructions on interpreting findings, and recommendations for reporting results. Dr Marsh will chair and facilitate discussions between workshop leaders and with the audience.



Discussion Leaders

Speakers:

  • Kevin Marsh, PhD, Executive Director, Patient-Centered Research, Evidera Ltd, Newport Pagell, UK
  • Martin Ho, MS, Associate Director for Quantitative Innovation, Center for Devices and Radiological Health, US Food and Drug Administration, Silver Springs, MD, USA
  • Juan Marcos Gonzalez Sepulveda, PhD, Assistant Professor, Population Health Sciences, Preference Evaluation Research Group, Duke University, Durham, NC, USA
  • Sebastian Heidenreich, PhD, Senior Research Associate, Patient Preferences, Evidera, Aberdeen, UK