COIOS
Power of data Item new peer-reviewed

A Bayesian model extends excess-death estimation to six years, applied to New Zealand

A Bayesian hierarchical model fitted to 22 years of monthly deaths by age and sex extends expected-death estimation beyond the one-to-two-year horizon of standard methods, and in New Zealand finds few excess deaths during the pandemic and a return to pre-pandemic trends by 2025.

The paper develops a Bayesian hierarchical model for expected monthly deaths by age and sex, built on long-term mortality forecasting methods and fitted to 22 years of New Zealand data, including sparse confidentialised counts. Applied from February 2020 to December 2025, it reproduces pre-pandemic features of the series and yields excess-death estimates with wide uncertainty. The estimates show relatively few excess deaths during the pandemic and mortality back on its pre-pandemic trend by 2025.

Why it is interesting: Existing excess-mortality baselines break down after one to two years, which is precisely the window needed to judge whether post-pandemic mortality has returned to trend.

Source
Demographic Research, 2026-09-17
Type
Peer-reviewed article
Design
Bayesian hierarchical forecasting model fitted to 22 years of monthly deaths by age and sex, applied to New Zealand, February 2020 to December 2025
Verdict
New finding
Driver
Measuring and projecting life expectancy
Driver
Infectious disease and pandemics