Random measurement error and regression dilution bias

Random measurement error is a pervasive problem in medical research, which can introduce bias to an estimate of the association between a risk factor and a disease or make a true association statistically non-significant. Hutcheon and colleagues explain when, why, and how random measurement error introduces bias and provides strategies for researchers to minimise the problem