7/ However, effect size estimates showed less agreement between the two designs (r = 0.74). Similar to the phenotypic findings, differences were primarily due to gene-by-cohort effects, where genetic associations vary across birth years, introducing bias into cross-sectional estimates.
