The late-life mortality plateau

node · earth · Italy — validated cohort

The observation that, at extreme ages, the death rate stops rising as the Gompertz law predicts and appears to level off. Barbi, Lagona, Marsili, Vaupel and Wachter reported in Science in 2018, using individually validated Italian records for people aged 105 and over, that the annual hazard of death flattens at roughly one half — that is, a person of 108 has about the same chance of dying within the year as a person of 112. If that is real, there is no age at which death becomes certain, and a strict biological ceiling on human lifespan is not implied by the mortality data. The finding is disputed on grounds of age misreporting: at these ages the number of people is tiny, a single misrecorded birth year distorts the estimate, and a systematic tendency for ages to be overstated would produce an artefactual plateau. The Italian data were assembled specifically to answer that objection, which is why the paper matters.