Selection bias in Mendelian randomization studies with adjustment for medication use
- 1Harvard T.H. Chan School of Public Health
Abstract
O_ST_ABSBackgroundC_ST_ABSMendelian randomization (MR) studies often evaluate exposures such as LDL cholesterol (LDL-C). The widespread use of lipid lowering medications complicates the interpretation and the validity of MR estimates.
MethodsWe describe two causal estimands in populations with medication use: a lifetime effect, and a lifetime effect under no medication use. In simulations, we compared the common approach of excluding medication users with estimation based on inverse probability (IP) weighting to adjust for medication use. We applied both approaches to a MR analysis of LDL-C and coronary artery disease in the Million Veteran Program (MVP), a large prospective cohort of U.S. veterans with linked electronic health record and genetic data.
ResultsIn simulations, MR analyses that did not adjust for medication use estimated a lifetime effect that reflected a valid estimate of a total effect that included both the harms of higher LDL-C and the benefits of statins. To estimate the effect under no medication use, excluding statin users introduced selection bias. Alternatively, IP weighting could address bias from incident statin users, but could not address bias related to prevalent medication use. Estimates from MVP data varied considerably, reflecting the importance of these analytic choices.
ConclusionsDifferent medication adjustment strategies in MR studies implicitly target different causal estimands and are subject to distinct biases. Transparent analytical choices and careful interpretation are essential for informative MR results in the context of widespread medication use.
Related
- Reporting Quality of Mendelian Randomization Studies Based on STROBE-MR: An Investigation Powered by Large Language Models
- Abdominal Adiposity and Lacunar Stroke: A Mendelian Randomization Mediation Study
- Genetically Proxied SGLT2 Inhibition and Risk of Epilepsy: A Mediation Mendelian Randomization Study of Circulating Metabolites
- Stroke and brain white/gray matter alterations : A Mendelian randomization investigation
- Causal effects of educational attainment on respiratory diseases : A Mendelian randomization study
BibTeX
@misc{shi2026selection,
title = {{Selection bias in Mendelian randomization studies with adjustment for medication use}},
author = {J. Shi and S. A. Swanson and E. W. Diemer and H. Gerlovin and D. C. Posner and P. W. F. Wilson and J. M. Gaziano and K. Cho and M. A. Hernan and on behalf of the VA Million Veteran Program},
howpublished = {medRxiv},
year = {2026},
month = sep,
doi = {10.64898/2026.09.16.26363225},
url = {https://www.medrxiv.org/content/10.64898/2026.09.16.26363225v1},
note = {CC0},
}
RIS
TY - UNPB
TI - Selection bias in Mendelian randomization studies with adjustment for medication use
AU - J. Shi
AU - S. A. Swanson
AU - E. W. Diemer
AU - H. Gerlovin
AU - D. C. Posner
AU - P. W. F. Wilson
AU - J. M. Gaziano
AU - K. Cho
AU - M. A. Hernan
AU - on behalf of the VA Million Veteran Program
PY - 2026
DA - 2026/09/16
PB - medRxiv
DO - 10.64898/2026.09.16.26363225
UR - https://www.medrxiv.org/content/10.64898/2026.09.16.26363225v1
AB - O_ST_ABSBackgroundC_ST_ABSMendelian randomization (MR) studies often evaluate exposures such as LDL cholesterol (LDL-C). The widespread use of lipid lowering medications complicates the interpretation and the validity of MR estimates. MethodsWe describe two causal estimands in populations with medication use: a lifetime effect, and a lifetime effect under no medication use. In simulations, we compared the common approach of excluding medication users with estimation based on inverse probability (IP) weighting to adjust for medication use. We applied both approaches to a MR analysis of LDL-C and coronary artery disease in the Million Veteran Program (MVP), a large prospective cohort of U.S. veterans with linked electronic health record and genetic data. ResultsIn simulations, MR analyses that did not adjust for medication use estimated a lifetime effect that reflected a valid estimate of a total effect that included both the harms of higher LDL-C and the benefits of statins. To estimate the effect under no medication use, excluding statin users introduced selection bias. Alternatively, IP weighting could address bias from incident statin users, but could not address bias related to prevalent medication use. Estimates from MVP data varied considerably, reflecting the importance of these analytic choices. ConclusionsDifferent medication adjustment strategies in MR studies implicitly target different causal estimands and are subject to distinct biases. Transparent analytical choices and careful interpretation are essential for informative MR results in the context of widespread medication use.
ER -