Life Expectancy Gains Are Slowing: Will We Ever Reach an Average of 100 Years?
In this article
Looking back through human history, the 20th century witnessed a lifespan revolution. Safer births, childhood vaccines, antibiotics, clean water and sewage systems, better nutrition... Together, these drove an almost linear increase in average life expectancy over a century. Newspapers, magazines and scientific articles often predicted: "At this rate, living to 100 will become ordinary."
But a new study published on 25 August 2025 in Proceedings of the National Academy of Sciences (PNAS) challenges this optimistic picture. Led by Dr. Héctor Pifarré i Arolas of the University of Wisconsin–Madison and colleagues, this extensive study shows that the pace of life expectancy gains has slowed noticeably in recent decades . Yes, we are still living longer, but the gains are no longer coming as quickly.
This article presents the study's findings in accessible, popular-science language . We focus on the numbers, discuss why progress is slowing and explore possible consequences. Above all, we consider: Can average human life expectancy really reach 100 years, or is that a dream?
📌 Why Was This Study Needed?
- Media coverage and some projections assumed life expectancy would always increase at the same rate .
- Period measures, based on calendar years, can show misleading fluctuations during shocks such as pandemics, whereas cohort life expectancy captures the actual longevity experience of birth generations.
- Planning pensions, healthcare spending and the workforce requires a realistic pace of improvement.
Data and Methods: Abbreviations and Plain-Language Explanations
The study uses 23 high-income countries' data from the Human Mortality Database (HMD) . HMDis a reliable database compiled from official death and population records, organized by age and year. People born in 1850–1938 were analyzed as completed cohorts , whose lifetimes have almost entirely been observed. For those born in 1939–2000, age-specific mortality rates (the probability of death at each age in a given year) were projected using six statistical methods .
Abbreviations and What the Methods Mean
- HMD — Human Mortality Database: A standardized open-access archive of national mortality data by age/year. All calculations in this study use HMD.
- Cohort Life Expectancy: The expected lifespan of real people born in the same year, based on the mortality risks encountered throughout their lives. Period life expectancy is based on calendar years; cohort measures avoid exaggerated effects from temporary crises such as pandemics.
- Age-Specific Mortality Rate: A measure of the likelihood of death at a particular age in a particular year; the raw input for life tables.
- Best-Practice Line: The upper curve formed by the country with the highest life expectancy in each cohort. Median shows the middle value across countries.
The 6 Forecasting Methods, Simply Explained
- Lee–Carter (LC): The most widely used period-based model separates common mortality trends (a time factor) from age patterns and extrapolates them into the future. Simply: A statistical answer to: "Will past declines continue at a similar pace?"
- LLC — Linear Lee–Carter: A version of LC with a cohort focus and stronger linear assumptions. Simply: Adapts LC to the cohort axis with more explicit assumptions about linear trends.
- CPS — Smooth Constrained Mortality Forecasting (CP‑Splines): Fits a smooth surface across age and time, constraining unexpected jumps. Simply: Reduces noise and smooths the curve in a reasonable way.
- CoDa — Compositional Data Analysis: Treats death probabilities across ages as parts of a whole , using log-ratio transformations. Simply: Uses compositional modeling to ask how the distribution of deaths across ages changes shape.
- C‑STAD — Cohort Segmented Transformation Age‑at‑death Distributions: Directly projects cohorts' distributions of age at death by transforming and segmenting them. Simply: A cohort model focused on shifts in distribution shape, such as rightward movement or compression.
- WPP‑2024 — UN World Population Prospects 2024: The United Nations' official demographic projection set. Simply: The UN's standard national lifespan projections as used in this analysis.
| Method | Gain per Cohort (Years) | Reduction from Historical Trend (%) |
|---|---|---|
| Historical trend (1900–1938) | 0.46 | — |
| C‑STAD | 0.28 | ≈39% |
| CPS | 0.29 | ≈37% |
| CoDa | 0.24 | ≈48% |
| Lee–Carter | 0.27 | ≈41% |
| LLC | 0.22 | ≈52% |
| WPP‑2024 | 0.23 | ≈50% |
Notes: (1) Figures are based on best-practice line slope estimates. (2) Different methods may vary slightly with data structure and assumptions. (3) LC, CoDa and CPS are period-based and use the Lexis diagonal to estimate cohorts; LLC and C-STAD are directly cohort-based approaches. (4) WPP-2024 is the UN's standard projection set and belongs to the period-based method family.
📊 Numerical Findings: Summary
- Born in 1850–1938: A gain per cohort of +0.46 years (≈5.5 months).
- Born in 1939–2000: The pace of improvement is 37–52% lower, with gains per cohort of ≈2.5–3.5 months.
- The 100-Year Threshold: For no cohort born in 1939–2000 is an average of 100 years projected.
- Contribution by Age: Of the slowdown, 50%+ comes from ages 0–5; 70% comes from ages 0–20 , as gains in these age bands are exhausted.
🧩 Why Has Progress Slowed?
Childhood mortality fell dramatically over history; these early-life gains are now approaching their limit. Continued progress at middle and older ages cannot, on its own, fully compensate for the earlier gains at younger ages.
🚀 What Can Be Done? A Forward-Looking Roadmap
- Midlife mortalityshould be reduced by targeting cardiometabolic disease, cancer and COPD.
- Behavioral interventions: smoking cessation, tackling obesity, physical activity and healthy eating.
- The biology of agingcan be targeted by treatments, alongside early diagnosis (screening and multi-omic risk prediction).
- Inequality-reducing social policies and universal healthcare access.
🔮 Conclusion
Life expectancy appears likely to keep rising , but its pace has slowed considerably. Reaching an average of 100 years will not happen automatically ; scientific breakthroughs and strong public health policies will be needed. We will shape the future.
🔗 Source
Pifarré i Arolas H, Camarda CG, Andrade J. Cohort mortality forecasts indicate signs of deceleration in life expectancy gains. PNAS, 25 August 2025. DOI: 10.1073/pnas.2519179122
SEO (150 characters): According to PNAS 2025, life expectancy gains are slowing: from +5.5 months per cohort to ≈2.5–3.5 months. Causes and possible solutions.
Was this article helpful?
Your feedback helps us improve our content.
Reader comments
Share your comment
Your name and comment are published only after administrator approval. Your email address is never shown to visitors.
Be the first to comment.