Bias amplification of unobserved confounding in pharmacoepidemiological studies using indication‐based sampling
Pharmacoepidemiology and Drug Safety · 2023-08 · 32 (8) · 886-897
DOI: 10.1002/pds.5614 ↗Kurz gefasst
Fachartikel in der Zeitschrift Pharmacoepidemiology and Drug Safety, online erschienen am 05.04.2023. Verfasst von Viktor H Ahlqvist und 9 weiteren Autor:innen. Beteiligte Einrichtung: Department of Global Public Health, Karolinska Institutet, Stockholm, Sweden. Hauptthemen laut MEDLINE: Pharmacoepidemiology. Die Arbeit wurde bisher 11-mal zitiert (Crossref) und nennt 34 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: NINDS NIH HHS, NIH HHS, Erik och Edith Fernströms Stiftelse för Medicinsk Forskning.
Abstract (Originalsprache)
Purpose
Estimating causal effects in observational pharmacoepidemiology is a challenging task, as it is often plagued by confounding by indication. Restricting the sample to those with an indication for drug use is a commonly performed procedure; indication‐based sampling ensures that the exposed and unexposed are exchangeable on the indication—limiting the potential for confounding by indication. However, indication‐based sampling has received little scrutiny, despite the hazards of exposure‐related covariate control.
Methods
Using simulations of varying levels of confounding and applied examples we describe bias amplification under indication‐based sampling.
Results
We demonstrate that indication‐based sampling in the presence of unobserved confounding can give rise to bias amplification, a self‐inflicted phenomenon where one inflates pre‐existing bias through inappropriate covariate control. Additionally, we show that indication‐based sampling generally leads to a greater net bias than alternative approaches, such as regression adjustment. Finally, we expand on how bias amplification should be reasoned about when distinct clinically relevant effects on the outcome among those with an indication exist (effect‐heterogeneity).
Conclusion
We conclude that studies using indication‐based sampling should have robust justification ‐ and that it should by no means be considered unbiased to adopt such approaches. As such, we suggest that future observational studies stay wary of bias amplification when considering drug indications.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Department of Global Public Health, Karolinska Institutet, Stockholm, Sweden.
- Centre for Academic Mental Health, Population Health Sciences, Bristol Medical School, University of Bristol, Bristol, UK.
- MRC Integrative Epidemiology Unit, University of Bristol, Bristol, UK.
- A. J. Drexel Autism Institute, Drexel University, Philadelphia, Pennsylvania, USA.
- Centre for Epidemiology and Community Medicine, Region Stockholm, Stockholm, Sweden.
- NIHR Biomedical Research Centre, University of Bristol, Bristol, UK.
ORCID-Profile: Viktor H Ahlqvist
Förderung
- NINDS NIH HHS
- NIH HHS
- Erik och Edith Fernströms Stiftelse för Medicinsk Forskning
- Karolinska Institutet
- National Institutes of Health
- Svenska Sällskapet för Medicinsk Forskning
Zitieren
APA
Ahlqvist, V. H., Madley-Dowd, P., Ly, A., Rast, J., Lundberg, M., Jónsson-Bachmann, E., Berglind, D., Rai, D., Magnusson, C., & Lee, B. K. (2023). Bias amplification of unobserved confounding in pharmacoepidemiological studies using indication‐based sampling. Pharmacoepidemiology and Drug Safety, 32(8), 886-897. https://doi.org/10.1002/pds.5614
Vancouver
Ahlqvist VH, Madley-Dowd P, Ly A, Rast J, Lundberg M, Jónsson-Bachmann E, et al. Bias amplification of unobserved confounding in pharmacoepidemiological studies using indication‐based sampling. Pharmacoepidemiology and Drug Safety. 2023;32(8):886-897. doi:10.1002/pds.5614
BibTeX
@article{Ahlqvist2023,
author = {Ahlqvist, Viktor H and Madley-Dowd, Paul and Ly, Amanda and Rast, Jessica and Lundberg, Michael and Jónsson-Bachmann, Egill and Berglind, Daniel and Rai, Dheeraj and Magnusson, Cecilia and Lee, Brian K},
title = {Bias amplification of unobserved confounding in pharmacoepidemiological studies using indication‐based sampling},
journal = {Pharmacoepidemiology and Drug Safety},
year = {2023},
volume = {32},
number = {8},
pages = {886--897},
doi = {10.1002/pds.5614}
}
EUVIMED