PURPOSE: The value of digital health technologies and machine learning (ML) methods are cutting-edge issues in health economics and outcomes research (HEOR). Digital health technologies use dynamic communication and information technology to improve patient health. ML explores pattern recognition and computational learning to construct algorithms that learn from data for classification and prediction purposes. Compared to traditional statistical methods, ML techniques are advantageous in high-dimensional “big data” that digital health technologies can amass. While the use of digital health technologies seems promising, issues of their economic valuation, data ownership, end-user values, and their ability to be linked to existing data have not yet been tackled. DESCRIPTION: The workshop will consist of five topics. First, background on traditional statistical methods vs. ML methods will be introduced. Second, ML algorithms for predicting HEOR outcomes will be explained in easy-to-understand terms. Their comparisons to traditional methods in analyzing HEOR data where high-dimensionality, complex interactive effects, and large sample sizes are a concern, will be highlighted. Third, the added value of ML to the end user due to its increased speed, accuracy, and ability to provide incremental new insights will be discussed. The challenges surrounding security, accessibility, and analysis of digital data (e.g., patient-managed device data connected to cloud-based platforms) will be presented. Fourth, market access issues of digital health technologies, such as the net economic benefit, reimbursement, and value proposition to payers, will be explored. Finally, case examples of ML and digital health solutions that enable personalized insights generation and delivery will be presented. Audience members will be invited to participate in case discussions.