An Example of How Immortal Time Bias Can Reverse the Results of an Observational Study

To the Editor: Bias and confounding can distort findings from observational studies; adjustment or correction for these sources of error rarely dramatically change the results. Immortal time bias, which is included in the follow-up period during which the study outcome by design cannot occur, is a potential source of bias in longitudinal studies.1–3 Researchers often assume that immortal time bias has, if anything, a negligible effect on...