Moderating the covariance between family member's substance use behavior

Area of Science:

• Behavioral Genetics
• Quantitative Genetics
• Biostatistics

Background:

• Traditional twin and family studies assume constant genetic and environmental covariation across family members.
• Age-specific gene expression and cohort effects can influence correlations between relatives, potentially biasing estimates.
• Existing models do not account for age-related moderation in familial resemblance.

Purpose of the Study:

• To develop and validate a statistical model for age-moderated covariance in twin and sibling studies.
• To assess the power of this model to detect age-moderation effects.
• To investigate the age-dependency of genetic and environmental influences on smoking behavior.

Main Methods:

• Development of a statistical model incorporating age difference as a moderator of covariance between relatives.
• Simulation studies to evaluate parameter recovery and model fit with varying parameter values.
• Power analyses to determine sample size requirements for detecting moderation effects.
• Application of the model to empirical data on cigarette smoking.

Main Results:

• The developed model accurately recovers simulated parameters, demonstrating its validity.
• Detecting age-moderation effects requires large sample sizes due to relatively low statistical power.
• Genetic covariance for smoking behavior among siblings decays rapidly with increasing age difference.

Conclusions:

• Age-moderation significantly influences familial resemblance, particularly for behaviors like smoking.
• Failure to account for age differences can lead to reduced power in genome-wide association studies (GWAS) and misinterpretation of genetic effects.
• Age-moderation may contribute to the 'missing heritability' paradox by explaining variance previously attributed to other factors.