PURPOSE: The aim of this workshop is to illustrate the key challenges faced when planning and executing machine learning (ML) projects and to describe best practices, which help ensure the integrity and validity of Real-World Evidence (RWE) involving ML. DESCRIPTION: ML is well adopted by many industries. It is commonly used in the drug discovery process but its application to randomized controlled trials and real-world data is in its infancy. As with any methodologies, it has the potential for incorrect or inappropriate use. This is exacerbated due to the lack of established guidelines concerning the ML best practices and its role and use in a dialogue with regulatory and payer bodies. The background ML expertise within pharmacoeconomics and outcomes research is still at an early stage of its evolution, and such domain expertise is an essential component for high quality rigorous studies. In this workshop, through didactic presentation and interactive review of ML abstracts, we aim to present five key challenges for ML projects, and our suggestions for overcoming them. The challenges include; selection of use cases, understanding and reporting predictive performance, preventing overfitting, overcoming the ‘black box’ criticism and dealing with hyperparameters. Our aim is to help ML practitioners produce rigorous outputs and inform reviewers how to assess ML research critically and effectively.