Estimating mean lifetime

Abstract

The life expectancy implied by current age-specific mortality rates is calculated with life table methods that are among the oldest and most fundamental tools of demography. We demonstrate that these conventional estimates of period life expectancy are affected by an undesirable "tempo effect." The tempo effect is positive when the mean age at death is rising and negative when the mean age is declining. Estimates of the effect for females in three countries with high and rising life expectancy range from 1.6 years in the United States and Sweden to 2.4 years in France for the period 1980-95.

Variance in death and its implications for modeling and forecasting mortality

“…The pioneering model of mortality translation due to Bongaarts and Feeney (2003) provides an appealingly simple description of mortality change, although it is controversial. A recent edited volume by Barbi, Bongaarts and Vaupel (2008) provides a thorough overview of the method, which is related to the concept of tempo effects in mortality (Bongaarts and Feeney, 2002).…”

Linking age, survival, and transit time distributions

“…There is considerable interest in estimating age and transit times of elements in a physical system or, equivalently, of individuals in a population, in disciplines as diverse as population dynamics and demography [M' Kendrick, 1925;von Foerster, 1959;Preston et al, 2000;Bongaarts and Feeney, 2003], chemical engineering [Nauman, 1969[Nauman, , 2008, and hydrology and geophysics (e.g., among many others, Eriksson [1971], Cvetkovic and Dagan [1994], Goode [1996], Ginn [1999], Delhez et al [1999], McGuire and McDonnell [2006], Duffy [2010], Botter et al [2011], McDonnell and Beven [2014], Harman [2014], Porporato and Calabrese [2015], and Benettin et al [2015]). It has in fact become increasingly clear that the age, survival time, and the total time spent by each element in a system may provide additional key insights into specific aspects of a system's behavior.…”

Tempo effects may distort the interpretation of trends in life expectancy

“…Current techniques to estimate this change in mortality conditions first reconstruct the full history of the currently living cohorts, taking into account mortality rates of the past, and then quantify the change in the average survival of these cohorts from one year to the next. (24,38) However, these techniques have been criticized because they do not provide falsifiable predictions. (10,18,39) This fundamental problem could be tackled by identifying the presence or absence of a large fraction of short-term shifts in death events between two periods for a population, in particular related to the presence of persons with a 10 high mortality risk.…”

“…However, starting in 2002, a series of papers surprised the demographic community, claiming that life tables are distorted whenever mortality is changing, due to so-called mortality tempo effects (7,. 16,24) These effects belong to a larger class of distortions defined as "an undesirable inflation or deflation of a period[...] indicator of a life-cycle event" (25). The general idea is that any period measure is prone to timing shifts of the events it counts, which in the case of mortality refers to postponements of deaths.…”