Age trajectories of genetic variance in physical functioning: a longitudinal study of Danish twins aged 70 years and older.

Genetic-evolutionary theories of aging predict that the genetic variance for fitness traits increases with age, while epidemiological-gerontological theories predict an increase in the environmental variance for most traits. In this study we examine the age trajectories of the genetic and environmental variance in physical functioning in a sample of 4731 Danish twins aged 70+ who are being followed longitudinally every second year with up to four assessments completed. A biometric growth model (Neale and McArdle, 2000) was applied to a validated physical ability score. The model included an overall level effect, a rate of linear change effect, and residual effects. The best-fitting model was a sex-specific model including additive genetic and nonshared environmental factors affecting level and rate of change and only nonshared environmental factors affecting the wave-specific levels. For both sexes there is an approximate doubling of both the total variance and the genetic variance in the physical ability score over the four waves and, hence, a rather stable heritability. However, the heritability is approximately.10 for males and.30 for females in all four waves. The heritability of level and slope showed a similar pattern:.11-14 in males and.35-.39 in females. The increase in both additive genetic variance and environmental variance is in agreement with genetic-evolutionary and epidemiological-gerontological theories of aging, respectively. The present study suggests that overall level of strength may be a better phenotype for future molecular genetic studies on physical functioning in the elderly than rate of change, because rate of change is vulnerable to sample attrition due to mortality and dropout and because four waves were needed to be able to detect a heritability for rate of change of the same magnitude as the heritability for level of physical functioning.