Patient-reported outcome measures (PROMs) are tools that assist with patient decision, making, monitoring, and outcome assessment. Modern psychometrics (eg, item response theory) and data science techniques (eg, machine learning) could improve the usefulness of patient-reported outcome (PRO) data initiatives. In this session, Dr. Jen Shin will provide the conceptual basis of item response theory (IRT) and provide examples of how to use IRT to validate PROMs for use in clinical research and practice. Dr. Chris Gibbons will summarize the evidence relating to PROs in clinical practice and introduce data science techniques to facilitate PRO collection and analysis. Dr. Gibbons will introduce and discuss two machine learning algorithms designed to facilitate PRO-based decision-making and monitoring in clinical practice.
Chris Sidey-Gibbons, PhD
Asso. Professor and Deputy Chair, Department of Symptom Research
Chief, Section of Patient- Centered Analytics
Director, MD Anderson Center for INSPiRED Cancer Care (Integrated Systems for Patient-Reported Data)
Jennifer Shin, MD
Vice Chair, Faculty Development
Asso. Professor, Otology and Laryngology, Harvard Medical School