> ## Content Index
> Fetch the complete content index at: https://www.coios.me/llms.txt
> Use this file to discover other available public pages before exploring further.

# Life expectancy in 28 countries modelled by machine learning on six drivers
- URL: https://www.coios.me/life-expectancy-in-28-countries-modelled-by-machine/
- Published: 2026-10-16T12:50:49.000Z
- Updated: 2026-10-17T08:12:19.000Z
- Description: Machine learning on a 28-country panel found a concave income-longevity link, and projected mean life expectancy of 78.21 years in 2032 under baseline against 78.98 under an integrated prevention scenario.
- Author: Daniel Ryan
- Tags: Data, models and systems, Health systems and access, Forecasts and their record, World, #peer-reviewed, #new, #item, #source-2026-10, #score-50

The study modelled life expectancy at birth against six drivers in 28 countries over 2003-2022\. The abstract names income, health-system capacity, public health interventions, environmental quality and demographic change among them. A range of machine learning techniques allowed nonlinear associations; income showed a concave relationship with longevity. The preferred model projected life expectancy to 2032 under baseline, accelerated health-expenditure, and integrated health-system and prevention scenarios. Mean life expectancy was 78.10 years in 2022 and 78.21, 78.34 and 78.98 years in 2032 respectively. Gains concentrated where coverage gaps remained; the authors call these predictive scenarios, not causal counterfactuals, across 28 countries only.

*Why it is interesting: It separates health-system capacity and prevention from income as predictors of national longevity. The spread across scenarios is under one year by 2032.*

Source

[Population Health Metrics, 15 October 2026](https://link.springer.com/article/10.1186/s12963-026-00517-x?ref=coios.me)

Type

Peer-reviewed article

Design

Cross-country panel of 28 countries, 2003–2022, machine-learning models with scenario projections to 2032

Verdict

New finding

Driver

[Health systems and access](https://www.coios.me/d-health-systems/)

Driver

[Forecasts and their record](https://www.coios.me/d-forecasting/)