PURPOSE: The availability of data to support health economics and outcomes research (HEOR) is increasing at an exponential rate. In the age of big data, cutting-edge machine learning algorithms, which have been tremendously successful in transportation and retail industry (e.g., Uber and Amazon) can play a promising role in HEOR. The objective of this workshop is to introduce the basics and key concepts in machine learning (ML), how machine learning can influence the field of HEOR, what are some of the potential pitfalls of machine learning and applications from health economics outcomes research that utilize machine learning.

DESCRIPTION: In this workshop, we will introduce the current landscape in HEOR through the lens of the state-of-the-art modern machine learning techniques. In particular, we will start with reviewing basics and key concepts of modern machine learning techniques for descriptive, predictive, prescriptive analytics. We will proceed with discussing how ML differs from statistical learning and modeling and provide an overview of selected ML algorithms such as deep neural networks, and random forests. We will then illustrate state-of-the-art applications of ML to HEOR– this will include use of ML for augmenting health economic modeling for cost-effectiveness analysis, developing ML-driven Apps or online tools for running complex models in real time, and synthetic data generation using ML. We will conclude by discussing potential pitfalls of ML in health outcomes research and promising future research directions. This workshop will be primarily geared towards the interested audience from the pharmaceutical industry concerned with health economic evaluation.