Background
Despite advances in heart failure care reducing mortality in clinical trials, it remains unclear whether real-life cohorts have had similar improvements in life expectancy across the age spectrum. We aimed to investigate how mortality trends changed in patients with heart failure over the past 25 years, stratified by age groups.

Methods
Using Danish nationwide registries, we identified patients with new-onset heart failure aged 18–95 years. The 5-year all-cause mortality risk and the absolute risk difference of mortality between patients with heart failure and age-matched and sex-matched heart failure-free controls were assessed using Kaplan–Meier estimates and multivariable Cox regression models. Mortality trends were analysed across five calendar periods (1996–2000, 2001–05, 2006–10, 2011–15, and 2016–20) and three age groups (<65 years, 65–79 years, and ≥80 years).

Findings
194 997 patients with heart failure were included. Mortality significantly decreased from 1996–2000 (66% [95% CI 65·5–66·4]) to 2016–20 (43% [42·1–43·4]), with similar results shown in all age groups (<65 years: 35% [33·9–36·1] to 15% [14·6–16·3]; 65–79 years: 64% [63·1–64·5] to 39% [37·6–39·6]; and ≥80 years: 84% [83·1–84·3] to 73% [71·7–73·9]). Adjusted mortality rates supported these associations. The absolute risk difference declined notably in younger age groups (<65 years: 29·9% [28·8–31·0] to 12·7% [12·0–13·4] and 65–79 years: 41·1% [40·3–41·9] to 25·1% [24·4–25·8]), remaining relatively stable in those aged 80 years or older (30·6% [29·9–31·3] to 28% [27·2–28·8]).

Interpretation
Over 25 years, there has been a consistent decrease in mortality among patients with heart failure across age groups, albeit less prominently in patients aged 80 years or older. Further insight is needed to identify effective strategies for improving disease burden in older patients with heart failure.