The curse of the plateau. Measuring confidence in human mortality estimates at extreme ages

The question of the human mortality plateau: Contrasting insights by longevity pioneers

“…With a rela-tively comparable sample size, a higher level of b grants higher chances of detecting the true trajectory instead of the simpler constant baseline hazard model that supports the existence of a mortality plateau. The low estimate of b in the study using Italian data might be one of the reasons why constant baseline hazard model is selected even if the true underlying mortality pattern could have been more likely Gompertzian (Camarda 2022).…”

Using a Penalized Likelihood to Detect Mortality Deceleration

“…The theory of unobserved heterogeneity and the associated frailty model (Vaupel et al, 1979) predicts a downward deviation at the oldest ages, to which only the most robust individuals in the population survive. Detecting such a deceleration in real data is not always successful (Gavrilova and Gavrilov, 2015;Newman, 2018), even though the vast majority of studies indicate that death rates at older ages increase at lower rates and can even level off (Curtsinger et al, 1992;Fukui et al, 1993Fukui et al, , 1996Carey et al, 1995;Khazaeli et al, 1998;Gampe, 2010Gampe, , 2021Rootzén and Zholud, 2017;Alvarez et al, 2021;Camarda, 2022;Belzile et al, 2022). In a frailty model setting, testing for mortality deceleration is equivalent to testing whether the non-negative frailty parameter is strictly positive.…”

Using a penalized likelihood to detect mortality deceleration

“…The theory of unobserved heterogeneity and the associated frailty model [2] predicts a downward deviation at the oldest ages, to which only the most robust individuals in the population survive. Detecting such a deceleration in real data is not always successful [3,4], even though the vast majority of studies indicate that death rates at older ages increase at lower rates and can even level off [5][6][7][8][9][10][11][12][13][14][15]. In a frailty model setting, testing for mortality deceleration is equivalent to testing whether the non-negative frailty parameter is strictly positive.…”