To the Editor:
When assessing health, physicians traditionally compare perceived and chronologic age. Among adults, “looking old for your age” is often interpreted as an indicator of poor health. The sparse data available on the relation between perceived age and survival indicate an inverse association.1 It is not known whether “looking old for your age” is primarily a result of lifestyle and other environmental factors or whether genetic factors play an important role. Here we use a population-based survey of Danish twins, age 70 years and older, to assess whether perceived age is influenced by genetic factors and whether “looking old for your age” is associated with an increased mortality.
In the 2001 survey of the Longitudinal Study of Aging Danish Twins,2 91% of cognitively intact participants agreed to have their picture taken (using a digital camera at 0.6-meter distance, with a neutral background, whenever possible). For a total of 387 same-sex twin pairs, we had a high-quality picture of both twins: 82 sets of monozygotic males, 93 monozygotic females, 94 dizygotic males, and 118 dizygotic females. The twins were not all photographed with neutral facial expressions, but according to Sheretz et al.,3 this has no effect on age estimation.
We engaged 20 female nurses (age 25 to 46 years) to estimate the twins’ ages based on digital photographs. The nurses were not informed beforehand about the age range of the twin pairs. They estimated the ages of all first-born and second-born twins on different days. The mean of the nurses’ age estimates for a twin was used as the twin's perceived age. The reliability of the mean age rating was estimated at 0.94 from a one-way analysis of variance. The correlation between real age and perceived age was 0.40 (95% confidence interval [CI] = 0.32–0.48), and the table shows that the nurses’ estimates regressed toward a mean of 77.
The intrapair correlation for perceived age within monozygotic twins (r = 0.59; CI = 0.49–0.68) is approximately twice the correlation for dizygotic twins (r = 0.29; CI = 0.16–0.41) (Table 1). These findings indicate an effect of additive genetic factors4 influencing perceived age. The correlations did not vary by age group or gender. Biometric models4 confirmed that the twin similarity is best explained by a model including additive genetic factors and nonfamily environment, and that the heritability (ie, the proportion of the variance explained by genetic factors) of perceived age is approximately 60% with no sex or age differences.
By January 2003, nearly 2 years after having been photographed, at least one of the pictured twins in 49 pairs had died. Among these 49 pairs, the longer-surviving twin had been rated as looking younger, on average, than his or her cotwin (mean of 1.15 years; CI = 0.11–2.19). This difference, however, owed entirely to those twin pairs who were perceived to be discrepant in age. Among the 26 pairs for which perceived age differed by 2 or more years, the oldest-looking twin died first in 19 (73%) cases, verifying that perceived age is associated with mortality. Stratifying these analyses for zygosity revealed zygosity differences: among 16 dizygotic twin pairs, the oldest-looking twin died first in 13 (81%) cases, whereas for the 10 monozygotic pairs, the oldest-looking twin died first in 6 (60%) cases. The numbers are small, but this pattern suggests that there are common genetic factors influencing both perceived age and survival, because controlling for genetic factors (the within monozygotic comparison) attenuates the association between perceived age and survival.
In summary, we show that approximately 60% of the variation in perceived age among twins can be explained by genetic factors and that “looking old for your age” is associated with increased mortality. Genetic factors are important for a broad range of traits and conditions from behavioral phenotypes, anthropometrics, and diseases to physical and cognitive functioning.5 Our design cannot determine how the genetic influence on perceived age is mediated. Nonetheless, our biometric models indicate that the genetic effects are acting additively, which suggests that “looking old” (or “looking young”) is a trait likely to run in families.
1.Borkan GA,Bachman SS,Norris AH. Comparison of visually estimated age with physiologically predicted age as indicators of rates of aging. Soc Sci Med
2.Christensen K,Gaist D,Vaupel JW, et al. Genetic contribution to rate of change in functional abilities among Danish twins aged 75 years or more. Am J Epidemiol
3.Sheretz EF,Hess SP. Stated age. N Engl J Med
4.Neale MC,Cardon LR. Methodology for Genetic Studies of Twins and Families
. Dordrecht: Kluwer Academic Publishers; 1992.
5.Plomin R,Owen MJ,McGuffin P. The genetic basis of complex human behaviors. Science