Probabilistic forecasting of maximum human lifespan by 2100 using Bayesian population projections

Area of Science:

• Gerontology and Demography
• Statistical Modeling
• Survival Analysis

Background:

• Quantifying human lifespan and forecasting maximum reported age at death (MRAD) is a complex statistical challenge.
• Understanding the probability of reaching extreme ages is crucial for demographic and societal planning.
• Previous models require updates with recent data and extended forecasting windows.

Purpose of the Study:

• To probabilistically forecast the maximum reported age at death (MRAD) through the year 2100.
• To quantify the likelihood of individuals reaching extreme ages, such as 120 and above.
• To provide updated, unconditional estimates of MRAD distribution using recent data and advanced statistical methods.

Main Methods:

• Utilized an updated exponential survival model for supercentenarians (over 110 years old).
• Incorporated Bayesian population projections to account for population uncertainty.
• Extended the forecasting window and performed a fully Bayesian analysis using the latest International Database on Longevity (IDL) data.

Main Results:

• The updated supercentenarian survival model aligns with recent IDL data.
• Population projections for individuals aged 110-114 through 2080 are deemed sensible.
• An unconditional distribution of MRAD by 2100 was estimated by integrating model and projection uncertainties.

Conclusions:

• There is a greater than 99% probability that the current MRAD of 122 will be broken by 2100.
• Estimated probabilities for reaching at least 126, 128, or 130 years by 2100 are 89%, 44%, and 13%, respectively.
• This study presents the first fully Bayesian, unconditional probabilistic projection of MRAD by 2100.