It is certainly possible that a small number of people have mutations or genetic variants that confer notable longevity. The small lineage exhibiting a PAI-1 loss of function mutation springs to mind as an example of this sort of thing. But for the overwhelming majority of long-lived lineages, the evidence on genetic contributions to longevity tends to support the hypothesis that familial longevity arises much more from lifestyle and environment than from inherited genetics. The data from very large genetic databases points to genetic variants contributing little to variation in human life span. The data on exercise, diet, and environmental factors such as particulate air pollution and persistent viral infection regularly results in larger effect sizes for late life mortality.

The familial resemblance in length of adult life is very modest. Studies of parent-offspring and twins suggest that exceptional health and survival have a stronger genetic component than lifespan generally. To shed light on the underlying mechanisms, we collected information on Danish long-lived siblings (born 1886-1938) from 659 families, their 5379 offspring (born 1917-1982), and 10,398 grandchildren (born 1950-2010) and matched background population controls through the Danish 1916 Census, the Civil Registration System, the National Patient Register, and the Register of Causes of Death.

Comparison with the background, population revealed consistently lower occurrence of almost all disease groups and causes of death in the offspring and the grandchildren. The expected incidence of hospitalization for mental and behavioral disorders was reduced by half in the offspring (hazard ratio 0.53) and by one-third in the grandchildren (0.69), while the numbers for tobacco-related cancer were 0.60 and 0.71, respectively. Within-family analyses showed a general, as opposed to specific, lowering of disease risk. Early parenthood and divorce were markedly less frequent in the longevity-enriched families, while economic and educational differences were small to moderate.

The longevity-enriched families in this study have a general health advantage spanning three generations. Having a long-lived parent or grandparent who had at least one long-lived sibling is associated with a substantial health and survival advantage in our study. Most notable is the strength of the associations, and that these are found for a wide range of diseases and causes of death, suggesting a fundamentally slower aging in these families and not just avoidance of specific diseases.

The combination of a particularly low incidence of mental and behavioral disorders and tobacco-related cancers combined with demographic characteristics such as low occurrence of teenage parenthood and early marriage and divorce implicate behavior as a key mechanism underlying the three-generation health and survival advantage observed. We found no evidence that the associations were driven by socioeconomic advantages in the longevity-enriched families either in the 1916 census or in the civil registration system over the last half-century.