Quantifying the impact of unmeasured confounding in observational studies with the E value

Tobias Gaster; Christine Marie Eggertsen; Henrik Støvring; Vera Ehrenstein; Irene Petersen

doi:10.1136/bmjmed-2022-000366

Quantifying the impact of unmeasured confounding in observational studies with the E value

BMJ Med. 2023 May 4;2(1):e000366. doi: 10.1136/bmjmed-2022-000366. eCollection 2023.

Authors

Tobias Gaster^#¹, Christine Marie Eggertsen^#¹, Henrik Støvring^{2

3}, Vera Ehrenstein⁴, Irene Petersen^{4

5}

Affiliations

¹ Aarhus University, Aarhus, Denmark.
² Steno Diabetes Centre Aarhus, Aarhus, Denmark.
³ Clinical Pharmacology, Pharmacy and Environmental Medicine, University of Southern Denmark, Odense, Denmark.
⁴ Department of Clinical Epidemiology, Aarhus University and Aarhus University Hospital, Aarhus, Denmark.
⁵ Department of Primary Care and Population Health, University College London, London, UK.

^# Contributed equally.

Abstract

The E value method deals with unmeasured confounding, a key source of bias in observational studies. The E value method is described and its use is shown in a worked example of a meta-analysis examining the association between the use of antidepressants in pregnancy and the risk of miscarriage.

Keywords: Epidemiology; Obstetrics; Pregnancy complications.