Weak-instrument robust tests in two-sample summary-data Mendelian randomization

Area of Science:

• Genetic Epidemiology
• Statistical Genetics
• Econometrics

Background:

• Mendelian randomization (MR) is widely used to infer causal relationships between exposures and outcomes using genetic variants as instrumental variables (IV).
• Two-sample summary-data MR is particularly popular but susceptible to bias and inflated Type I errors when instruments are weakly associated with the exposure.
• Weak instrument bias is a significant challenge in genetic epidemiology, potentially leading to unreliable causal effect estimates.

Purpose of the Study:

• To develop novel statistical test statistics for two-sample summary-data Mendelian randomization that are robust under weak-instrument asymptotics.
• To extend established econometric tests, including the Anderson-Rubin, Kleibergen, and conditional likelihood ratio tests, to the MR context.
• To provide a robust point estimator and a method for detecting invalid instruments based on the proposed Anderson-Rubin test.

Main Methods:

• Extension of Anderson-Rubin, Kleibergen, and conditional likelihood ratio tests from econometrics to two-sample summary-data MR.
• Development of a point estimator using the proposed Anderson-Rubin test.
• Creation of a method for invalid instrument detection.
• Validation through simulation studies and an empirical analysis.

Main Results:

• The proposed test statistics demonstrate robustness under weak-instrument asymptotics in Mendelian randomization.
• The new methods effectively control Type I errors and mitigate bias associated with weak instruments.
• The Anderson-Rubin test provides a reliable point estimator and aids in identifying invalid instruments.
• Simulations and empirical results show improved power compared to existing methods when dealing with weak instruments.

Conclusions:

• The developed statistical tests offer a more reliable approach to Mendelian randomization analysis, particularly in the presence of weak instruments.
• These robust methods enhance the accuracy and validity of causal inference in genetic epidemiology.
• The findings suggest that the proposed extensions provide superior performance over existing techniques for Mendelian randomization studies with weak instruments.