Description: Despite a considerable accumulation of empirical and theoretical evidence in the last two decades in favor of a slowdown in the rate of mortality increase with age, the veracity of mortality deceleration observed at very old ages among humans was recently called into question on the basis of: (1) inaccurate data, (2) invalid assumptions, and/or (3) inadequate mixture of birth cohorts with different mortality experiences. In this article, we use a highly reliable set of individual-level data on French-Canadians centenarians born between 1870 and 1896 and who died between 1970 and 2009 to compute new estimates of the age-trajectory of mortality at ages 100 and over. Our results suggest that the rate of mortality increase with age tends to slow down at very old ages, even when special care is taken to avoid issues (1) to (3) outlined above.

Description: Abstract: Despite a considerable accumulation of empirical and theoretical evidence in the last two decades in favor of a slowdown in the rate of mortality increase with age, the veracity of mortality deceleration observed at very old ages among humans was recently called into question on the basis of: (1) inaccurate data, (2) invalid assumptions, and/or (3) inadequate mixture of birth cohorts with different mortality experiences. In this article, we use a highly reliable set of individual-level data on French-Canadians centenarians born between 1870 and 1896 and who died between 1970 and 2009 to compute new estimates of the age-trajectory of mortality at ages 100 and over. Our results suggest that the rate of mortality increase with age tends to slow down at very old ages, even when special care is taken to avoid issues (1) to (3) outlined above.