[Skip to Content]
[Skip to Content Landing]
Views 721
Citations 0
Comment & Response
February 26, 2019

Case-Control Studies

Author Affiliations
  • 1Melbourne School of Population and Global Health, University of Melbourne, Melbourne, Australia
  • 2Department of Medical Statistics, London School of Hygiene and Tropical Medicine, London, England
  • 3School of Public Health, University of Adelaide, Adelaide, Australia
JAMA. 2019;321(8):806-807. doi:10.1001/jama.2018.20253

To the Editor The JAMA Guide to Statistics and Methods article on case-control studies1 contains several misconceptions.

First, Dr Irony stated that “the RR [relative risk] cannot be determined from a case-control study. A case-control study can only estimate the OR [odds ratio], which is the ratio of odds and not the ratio of probabilities.” This is true for studies using epidemic sampling (also known as cumulative incidence sampling), in which controls are selected from those who did not develop the outcome by the end of the risk period. However, the vast majority of case-control studies sample controls from the source population over the entire risk period under study (density sampling). With such sampling, the OR exactly estimates the rate ratio from the corresponding cohort study.2-4