Mendelian Randomization
Mendelian Randomization Analysis · Also known as: MR
Mendelian randomization is a method for estimating causal effects of exposures on outcomes using genetic variants as instrumental variables. Introduced by George Davey Smith in the 1990s, it exploits Mendel's law of segregation to remove confounding bias. It has become a cornerstone technique in epidemiological causal inference.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Use Mendelian randomization when studying causal effects in complex diseases where randomized trials are infeasible or unethical, and when GWAS summary statistics are available for both the exposure and outcome. It is particularly valuable for examining modifiable exposures (diet, lifestyle, biomarkers) in relation to chronic diseases. Avoid using when only weak genetic instruments are available, when there is substantial population stratification, or when the exposure has complex genetic architecture.
Strengths & limitations
- Avoids confounding and reverse causation bias inherent in observational studies by leveraging random genetic assignment
- Uses publicly available GWAS summary statistics, making it cost-effective and enabling large-scale analyses across populations
- Can identify causal effects when randomized trials are ethically or practically infeasible
- Multiple robust sensitivity analysis methods available to assess pleiotropy and violation of assumptions
- Relies on strong instrumental variable assumptions that cannot be fully tested, especially exclusion restriction
- Horizontal pleiotropy (genetic variants affecting outcome through multiple pathways) can bias results
- Weak genetic instruments reduce statistical power and increase weak instrument bias
- Requires large-scale GWAS data that may not be available for all populations or outcomes
Frequently asked
What is horizontal pleiotropy and why does it matter in Mendelian randomization?
Horizontal pleiotropy occurs when a genetic variant affects the outcome through pathways independent of the exposure. This violates the exclusion restriction assumption and can bias causal estimates. Testing for and adjusting for pleiotropy using MR-Egger regression, weighted median, or mode-based estimators is essential for valid inference.
How do I know if my genetic instruments are strong enough?
Weak instruments occur when the F-statistic (from regressing exposure on SNPs) is less than 10. Weak instruments lead to bias and inflated standard errors. Always report the F-statistic and consider using methods robust to weak instruments. For summary statistics, examine the R-squared from SNP-exposure associations.
Can I use Mendelian randomization to study acute outcomes?
Mendelian randomization primarily captures long-term causal effects because genetic effects accumulate over the lifespan. It is less suited for acute outcomes or short-term exposures. Ensure the causal pathway is biologically plausible and temporally reasonable for your research question.
What is the difference between simple Wald ratio and inverse-variance weighting?
Wald ratio uses a single SNP as an instrument, simple and interpretable but low power. Inverse-variance weighting combines multiple SNPs, increasing power and precision. Use inverse-variance weighting as the primary method when multiple SNPs are available, with Wald ratio for sensitivity checks.
Sources
- Davey Smith, G., & Hemani, G. (2014). Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Human Molecular Genetics, 23(R1), R89-R98. DOI: 10.1093/hmg/ddu328 ↗
- Hemani, G., Bowden, J., & Davey Smith, G. (2018). Evaluating the potential role of pleiotropy in Mendelian randomization studies. European Journal of Epidemiology, 33(9), 867-876. DOI: 10.1093/hmg/ddy163 ↗
- Morrison, J., Knoblauch, N., Marcus, J. H., Stephens, M., & He, X. (2020). Mendelian randomization accounting for sample overlap. Nature Communications, 11(1), 574. link ↗
How to cite this page
ScholarGate. (2026, June 3). Mendelian Randomization Analysis. ScholarGate. https://scholargate.app/en/causal-inference/mendelian-randomization
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- 2SLS RegressionEconometrics↔ compare
- Regression DiscontinuityCausal inference↔ compare