A Modified Debiased Inverse‐Variance Weighted Estimator in Two‐Sample Summary‐Data Mendelian Randomization
Statistics in Medicine · 2024-12-20 · 43 (29) · 5484-5496
DOI: 10.1002/sim.10245 ↗At a glance
Journal article in Statistics in Medicine, published online on 2024-10-25. Written by Youpeng Su and 8 co-authors. Institution: Department of Epidemiology and Biostatistics, School of Public Health, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, …. Major MEDLINE topics: Mendelian Randomization Analysis, Computer Simulation. Cited 27 times so far (Crossref); 36 references. Funded by: Hubei Provincial Natural Science Foundation of China, National Natural Science Foundation of China, Fundamental Research Funds for School of Public Health, Tongji Medical College, Huazhong University of Science and Technology.
Abstract (original language)
Mendelian randomization uses genetic variants as instrumental variables to estimate the causal effect of exposure on outcome from observational data. A common challenge in Mendelian randomization is that many genetic variants are only modestly or even weakly associated with the exposure of interest, a setting known as many weak instruments. Conventional methods, such as the popular inverse‐variance weighted (IVW) estimator, could be heavily biased toward zero when the instrument strength is weak. To address this issue, the debiased IVW (dIVW) estimator and the penalized IVW (pIVW) estimator, which are shown to be robust to many weak instruments, were recently proposed. However, we find that the dIVW estimator tends to produce an exaggerated estimate of the causal effect, especially when the effective sample size is small. Although the pIVW estimator has better statistical properties, it is slightly more complex, and the idea behind this method is also a bit less intuitive. Therefore, we propose a modified debiased IVW (mdIVW) estimator that directly multiplies a shrinkage factor with the original dIVW estimator. After this simple modification, we prove that the mdIVW estimator not only has second‐order bias with respect to the effective sample size, but also has smaller variance and mean squared error than the preceding two estimators. We then extend the proposed method to account for the presence of instrumental variable selection and balanced horizontal pleiotropy. We demonstrate the improvement of the mdIVW estimator over the competing ones through extensive simulation studies and real data analysis.
Subjects
MeSH terms (MEDLINE)
Author keywords
Authors and institutions
- Department of Epidemiology and Biostatistics, School of Public Health, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, China.
- Clinical Research Centre, Guangzhou First People's Hospital, School of Medicine, South China University of Technology, Guangzhou, China.
- Department of Biostatistics, School of Public Health, Southern Medical University, Guangzhou, China.
- Department of Statistics and Actuarial Science, The University of Hong Kong, Hong Kong SAR, China.
ORCID profiles: Youpeng Su, Jiyuan Zhou, Peng Wang
Funding
- Hubei Provincial Natural Science Foundation of China
- National Natural Science Foundation of China
- Fundamental Research Funds for School of Public Health, Tongji Medical College, Huazhong University of Science and Technology
- Natural Science Foundation of Hubei Province
Cite
APA
Su, Y., Xu, S., Ma, Y., Yin, P., Hao, X., Zhou, J., Fung, W. K., Jiang, H., & Wang, P. (2024). A Modified Debiased Inverse‐Variance Weighted Estimator in Two‐Sample Summary‐Data Mendelian Randomization. Statistics in Medicine, 43(29), 5484-5496. https://doi.org/10.1002/sim.10245
Vancouver
Su Y, Xu S, Ma Y, Yin P, Hao X, Zhou J, et al. A Modified Debiased Inverse‐Variance Weighted Estimator in Two‐Sample Summary‐Data Mendelian Randomization. Statistics in Medicine. 2024;43(29):5484-5496. doi:10.1002/sim.10245
BibTeX
@article{Su2024,
author = {Su, Youpeng and Xu, Siqi and Ma, Yilei and Yin, Ping and Hao, Xingjie and Zhou, Jiyuan and Fung, Wing Kam and Jiang, Hongwei and Wang, Peng},
title = {A Modified Debiased Inverse‐Variance Weighted Estimator in Two‐Sample Summary‐Data Mendelian Randomization},
journal = {Statistics in Medicine},
year = {2024},
volume = {43},
number = {29},
pages = {5484--5496},
doi = {10.1002/sim.10245}
}
EUVIMED