Estimates of how genetic influence on complex traits changes across the lifespan depend substantially on study design. Schoeler and colleagues report that cross-sectional and longitudinal analyses yield divergent age-varying effect estimates, largely driven by birth cohort differences and selective participation rather than true biological change with age. Interpretation of polygenic risk at different ages therefore requires explicit accounting for these design artifacts.
Key Points
- Cross-sectional and longitudinal designs produce conflicting age-varying genetic estimates
- Birth cohort differences and participation bias drive much of divergence
- Age-specific polygenic risk interpretation requires design-aware modeling
Longevity Analysis
Polygenic scores are increasingly used to stratify risk and personalize intervention timing, yet their apparent strength at a given age may reflect who was recruited and when they were born rather than any shift in underlying biology. Separating measurement artifact from genuine change in how inherited variation shapes metabolic, cardiovascular, and cognitive trajectories determines whether a risk estimate is actionable or misleading. Accurate signal interpretation at the level of study design precedes any meaningful downstream decision about who benefits from early intervention and when.
Original published by Nature Aging, by Tabea Schoeler.