The conventional wisdom is that common genetic variations have only a very small impact on mortality and health across the majority of the present human life span, but as the molecular damage of aging accumulates in later life, a time of frailty, disability, and high risk of age-related disease, genes make an increasingly significant contribution to determining remaining life expectancy. Here is a very readable open access paper on this topic, covering at a high level a range of the mainstream work on genetics, lifestyle, and aging from the past few decades:
Before the 1990s it was largely considered that aging is ineluctable and that genetics does not control it. It was important, in this view, the idea that aging occurs after reproduction, and then there is no need, but also no opportunity, for selection to act on genes that are expressed during this late period of life. Thereafter studies clearly demonstrated that genetic variability could indeed affect lifespan. This triggered many studies in model organisms in order to disentangle the different biochemical pathways which could affect lifespan, and to highlight the genes coding for the proteins involved in such pathways.
It is of note that some authors suggested the molecular mechanisms modulating lifespan could be due to a pleiotropic effect of genes which have evolved for different purposes (such as the genes in the IGF-1 pathway which have evolved to face presence/absence of nutrients) but can, ultimately affect lifespan; others proposed that some genes may have evolved to program aging and avoid "immortality", as this would hamper the continuous substitution of old subjects with new, younger, ones. It was obviously inevitable that the research of the genetic basis of longevity turned to human beings and investigated whether the common genetic variability of human populations could affect inter individual differences in lifespan but also whether the genes found to prolong lifespan in model organisms, on turn, were correlated to human lifespan.
As to the first question (does common genetic variability affect lifespan, and in particular does it affect longevity?), this has been studied by two approaches. The first one was the reconstruction of the sibships of long-lived subjects and the comparison of their survival curves with those of the birth cohorts born in the same geographical area. This approach demonstrated that brothers and sisters of the long-lived subjects had a clear survival advantage (at any age) with respect to the general population. The second approach, with intrafamily controls, was started in order to distinguish the genetic from the "familiar" effect. Researchers compared the survival function of brothers of centenarians with those estimated for their brothers in law, that is with the men who married their sisters; these men were supposed to share with the brothers of the long lived subjects the familiar environment. By using this second approach, it has been found that the survival advantage of siblings of long-lived subjects was not completely shared by their brothers in law, despite they shared the same environment for most of their life. This suggested that beyond the family environment, there are genetic factors influencing survival and, consequently, lifespan. The genetic component of lifespan in humans has also been analyzed by comparing the age of death of monozygotic and dizygotic twins. This has allowed the estimate that about 25% of the variation in human longevity can be due to genetic factors and indicated that this component is higher at older ages and is more important in males than in females.