On the relationship between human all-cause mortality and age

Abstract

Many equations can be used to study the relationship between mortality rates and age: Gompertz, Weibull, logistic, polynomial and age–period–cohort equations a.o. All these equations result in highly significant correlations between ln mortality rates and age in the age range 35–84 years. This applies to all developed countries and is independent of the differences in causes of death between populations. The best fit is obtained by a second-degree polynomial equation (R ² > 0.99), closely followed by the Gompertz equation. This equation is preferred in view of its extreme simplicity. A highly significant correlation exists between the intercept and the slope of the Gompertz equations, pointing to a crossing-over age. Beyond that age, around 85 years, populations with high mortality rates have a lower mortality, due to selective survival of the strongest individuals. The polynomial age² term may be positive or negative, an expression of the acceleration or de-acceleration of mortality at higher ages and is significantly more often positive in women. The equations used are very useful for the study of the aging process and for examining the relationship between possible causal factors and mortality rates in populations.