Improving long-term outcomes in atrial fibrillation: temporal decline in stroke and dementia risk among Danish patients – a nationwide registry-based study

Introduction

Atrial fibrillation (AF) is the most common cardiac arrhythmia and a major public health challenge, particularly in ageing populations. Today, improved awareness and screening strategies, and advances in diagnostic technology have enhanced our ability to detect AF earlier and initiate appropriate treatment. This development may offer a unique opportunity not only to prevent stroke, but also to reduce the risk of cognitive decline and dementia, which are increasingly recognized as important complications of AF. As the burden of AF continues to rise, understanding whether improved detection and management translate into better prevention of these long-term outcomes is essential.

Purpose

To investigate temporal trends in incidence of stroke and dementia among AF-patients to assess whether today’s methods for detection and management have improved long-term outcomes.

Methods

Using Danish nationwide registries, all adult patients diagnosed with first time AF diagnosis between January 1995, and May 2024 were identified and included in the study. The population was segregated by age into four groups. Comorbidities and prior use of medication were analyzed at baseline. Fifteen-year cumulative incidences of stroke and all-cause late onset dementia (LOD) were estimated across two calendar periods: 1995–2010 and 2010–2025.

Results

A total of 489.705 patients were newly diagnosed with AF in Denmark between 1995 and 2025. The median age was 75 years, and 55% were male. At inclusion, 26% received oral anticoagulation (OAC). Calculations showed that 15-year cumulative OAC use increased from approximately 55–70% in 1995–2010 to around 75–85% in 2010–2025 across age groups, reflecting earlier initiation and substantially higher overall uptake in the later period.

Figure 1 illustrates the 15-year cumulative incidence of first-time stroke, accounting for death as a competing risk, across two calendar periods. In 1995–2010, the 15-year risk of first-time stroke was 11%, 19%, 21%, and 16% among patients aged <60, 60–69, 70–79, and ≥80 years, respectively. In 2010–2025, the corresponding risks were substantially lower at 4%, 7%, 8%, and 8%, respectively.

Figure 2 illustrates the 15-year cumulative incidence of LOD, accounting for death as a competing risk, across two calendar periods. In 1995–2010, the 15-year risk of first-time LOD was 0.05%, 5%, 11%, and 10% among patients aged <60, 60–69, 70–79, and ≥80 years, respectively. In 2010–2025, the corresponding risks were 0.04%, 4%, 9%, and 9%, respectively.

Conclusion

In this nationwide cohort of AF-patients, the 15-year risk of first-time stroke declined markedly across all age groups over time. A modest reduction in the incidence of all-cause dementia was also observed, possibly reflecting the multifactorial etiology of neurodegenerative disease. The findings suggest that improvements in detection and management of AF have translated into better long-term outcomes.Cumulative Incidence of StrokeCumulative Incidence of Dementia