COVID-19 Real-World Evidence Primer

Chapter 3: Methods in Real-World Evidence Generation - Sources of Error

Authors: Hu Li, MBBS, PhD1; Kueiyu Joshua Lin, MD, ScD2; Christel Renoux, MD, PhD, MSc3; Almut G. Winterstein, RPh, PhD, FISPE4

Real-World Evidence, Gilead Sciences
2Division of Pharmacoepidemiology and Pharmacoeconomics, Brigham and Women’s Hospital and Harvard Medical School
3Department of Neurology & Neurosurgery and Department of Epidemiology, Biostatistics and Occupational Health; McGill University; Centre for Clinical Epidemiology, Lady Davis Institute for Medical Research, Jewish General Hospital
4Department of Pharmaceutical Outcomes and Policy and Center for Drug Evaluation and Safety, University of Florida

Disclosures: none

When using real-world data, key potential sources of systematic error or bias include confounding, selection bias, immortal time bias, and measurement bias resulting from misclassification and missing data. In this chapter we describe the sources of error and provide solutions for detecting and handling these errors.