Matt Fox & Lisa Bodnar
Validation of key study variables is a critical part of ensuring validity of study results and yet we find that few students have the opportunity to learn about how to design appropriate validation studies during their training, and few practicing epidemiologists have strong experience with conducting them. Good validation studies can be used to adjust estimates of effect using quantitative bias analysis in order to get a more accurate picture of total study error. However, poorly designed validation studies can lead to incorrect adjustments that lead to even greater bias. Below are readings on the design and use of validation studies.