Genetic sensitivity analysis: Adjusting for genetic confounding in epidemiological associations

Area of Science:

• Epidemiology
• Behavioral Genetics
• Biostatistics

Background:

• Epidemiological studies often report associations without adjusting for genetic confounding.
• Genetic factors can influence both exposures and outcomes, potentially distorting observed relationships.

Purpose of the Study:

• To develop and apply a two-stage approach for estimating the extent of genetic confounding in epidemiological associations.
• To assess the impact of genetic factors on the association between maternal education and child developmental outcomes.

Main Methods:

• A two-stage approach involving regression analyses with polygenic scores (PGS) and structural equation modeling (SEM).
• Utilized SNP-based and twin-based heritability estimates to model genetic confounding.
• Examined associations between maternal education and child educational achievement, Body Mass Index (BMI), and Attention Deficit Hyperactivity Disorder (ADHD).

Main Results:

• Polygenic scores explained 14.3% to 23.0% of the observed associations.
• Analyses indicated that observed associations could be almost entirely explained by genetic confounding under SNP- and twin-based heritability scenarios.
• Demonstrated the potential for substantial genetic influence on the studied exposure-outcome relationships.

Conclusions:

• Caution is advised when interpreting associations from epidemiological studies not accounting for genetic confounding.
• The proposed genetically informed sensitivity analysis approach is widely applicable for assessing genetic influences.
• Highlights the importance of considering genetic factors in epidemiological research to avoid misinterpretation.