Intended for healthcare professionals

CCBYNC Open access
Research

Temporal trends in lifetime risks of atrial fibrillation and its complications between 2000 and 2022: Danish, nationwide, population based cohort study

BMJ 2024; 385 doi: https://doi.org/10.1136/bmj-2023-077209 (Published 17 April 2024) Cite this as: BMJ 2024;385:e077209

Linked Editorial

The growing burden of atrial fibrillation and its consequences

  1. Nicklas Vinter, postdoctoral researcher1 2 3,
  2. Pia Cordsen, statistiscian1,
  3. Søren Paaske Johnsen, professor1,
  4. Laila Staerk, medical doctor4,
  5. Emelia J Benjamin, professor5 6,
  6. Lars Frost, associate professor2 3,
  7. Ludovic Trinquart, associate professor1 7 8 9
  1. 1Danish Center for Health Services Research, Department of Clinical Medicine, Aalborg University, Aalborg, Denmark
  2. 2Diagnostic Centre, University Clinic for Development of Innovative Patient Pathways, Silkeborg Regional Hospital, Silkeborg, Denmark
  3. 3Department of Clinical Medicine, Aarhus University, Aarhus, Denmark
  4. 4Department of Clinical Medicine, Copenhagen University Hospital—Amager and Hvidovre, Copenhagen, Denmark
  5. 5Department of Medicine, Boston Medical Center, Boston University Chobanian and Avedisian School of Medicine, Boston, MA, USA
  6. 6Department of Epidemiology, Boston University School of Public Health, Boston, MA, USA
  7. 7Tufts Clinical and Translational Science Institute, Tufts University, Boston, MA, USA
  8. 8Institute for Clinical Research and Health Policy Studies, Tufts Medical Center, Boston, MA, USA
  9. 9Department of Biostatistics, Boston University School of Public Health, Boston, MA, USA
  1. Correspondence to: N Vinter nicvin{at}rm.dk (@nicklas_vinter on X)
  • Accepted 27 February 2024

Abstract

Objectives To examine how the lifetime risks of atrial fibrillation and of complications after atrial fibrillation changed over time.

Design Danish, nationwide, population based cohort study.

Setting Population of Denmark from 1 January 2000 to 31 December 2022.

Participants 3.5 million individuals (51.7% women and 48.3% men) who did not have atrial fibrillation at 45 years of age or older were followed up until incident atrial fibrillation, migration, death, or end of follow-up, whichever came first. All 362 721 individuals with incident atrial fibrillation (46.4% women and 53.6% men), but with no prevalent complication, were further followed up until incident heart failure, stroke, or myocardial infarction.

Main outcome measures Lifetime risk of atrial fibrillation and lifetime risks of complications after atrial fibrillation over two prespecified periods (2000-10 v 2011-22).

Results The lifetime risk of atrial fibrillation increased from 24.2% in 2000-10 to 30.9% in 2011-22 (difference 6.7% (95% confidence interval 6.5% to 6.8%)). After atrial fibrillation, the most frequent complication was heart failure with a lifetime risk of 42.9% in 2000-10 and 42.1% in 2011-22 (−0.8% (−3.8% to 2.2%)). Individuals with atrial fibrillation lost 14.4 years with no heart failure. The lifetime risks of stroke and of myocardial infarction after atrial fibrillation decreased slightly between the two periods, from 22.4% to 19.9% for stroke (−2.5% (−4.2% to −0.7%)) and from 13.7% to 9.8% for myocardial infarction (−3.9% (−5.3% to −2.4%). No evidence was reported of a differential decrease between men and women.

Conclusion Lifetime risk of atrial fibrillation increased over two decades of follow-up. In individuals with atrial fibrillation, about two in five developed heart failure and one in five had a stroke over their remaining lifetime after atrial fibrillation diagnosis, with no or only small improvement over time. Stroke risks and heart failure prevention strategies are needed for people with atrial fibrillation.

Introduction

Atrial fibrillation is estimated to affect 17.9 million people in Europe by 2060 and 15.9 million people in the US by 2050.12 All cause and cardiovascular mortality among individuals with atrial fibrillation has improved over time, although the excess mortality associated with atrial fibrillation is still high.345 Atrial fibrillation is also associated with increased risks of stroke,6 heart failure,7 and myocardial infarction.8 Therefore, prevention of atrial fibrillation is a major public health priority.

Effective risk assessment for atrial fibrillation and its complications is key to primary and secondary prevention by increasing awareness of the condition.91011 In this context, the residual lifetime risk measures the cumulative risk for developing a disease over the remaining lifespan among individuals who attained a given index age disease-free.1213 Lifetime risk is a tractable quantity that can help to promote lifestyle change using recommendations and is an important communication tool to the public.1415 Population and community based cohort studies have reported on the lifetime risk of atrial fibrillation, which is estimated as affecting one in three in individuals of European ancestry in Europe and the US.161718

Patients are commonly told that the main danger after being diagnosed with atrial fibrillation is the increased risk of stroke, but atrial fibrillation is associated with increased risk of other complications. Quantification of the long term downstream consequences of atrial fibrillation will further improve our understanding of the burden of atrial fibrillation and how to communicate risks to the public, but previous investigations do not appear to have addressed this question. Furthermore, data for temporal trends in the lifetime risks of atrial fibrillation and of subsequent complications are absent. Monitoring temporal changes in atrial fibrillation burden is critical to indirectly measure the effectiveness of improvements in atrial fibrillation management and of primary and secondary prevention strategies, particularly in the new phase of stroke prevention therapy for atrial fibrillation. In this study of the Danish population, we aimed to estimate the lifetime risk of atrial fibrillation and of complications after atrial fibrillation, and to examine their temporal trends over 2000-22.

Methods

Setting, data sources, and study population

We conducted a nationwide cohort study from 1 January 2000 to 31 December 2022 in Denmark. We estimated the lifetime risk of atrial fibrillation among individuals who did not have atrial fibrillation and were aged 45 to 95 years. Among individuals with newly diagnosed atrial fibrillation, we estimated the lifetime risks of subsequent complications, including heart failure, any stroke, ischaemic stroke, myocardial infarction, and systemic embolism. To assess temporal trends, we compared two periods: 2000-10 and 2011-22.

Study populations and data were obtained from administrative nationwide registries. Firstly, we used the Danish National Patient Registry to identify inpatient hospital stays and outpatient contacts. We retrieved individual level information on dates of hospital admission and discharge, procedures performed, and primary and several secondary diagnoses per discharge. Diagnoses were coded according to the International Classification of Diseases 8th Revision (ICD-8) before 1994 and to the 10th revision (ICD-10) from 1994. Supplemental table 1 gives the ICD codes we used. We used the admission dates for outpatient contacts and discharge dates for inpatients contacts. In the third version of the Danish National Patient Registry, which was implemented in early 2019, information to classify hospital contacts as inpatients or outpatients was no longer available, and we used admission dates for all diagnoses. Secondly, we extracted individual level information on sex, date of birth, vital status, and migration from the Civil Registration System.19 Thirdly, we retrieved information on pharmacological treatments from the Danish National Prescription Registry.20 Medications were coded in accordance with the anatomical therapeutic chemical classification system. The codes we used are given in supplemental table 1.

Participants and outcomes for lifetime risk of atrial fibrillation

In each period (2000-22, 2000-10, and 2011-22), we included all Danish individuals aged 45 years or older and who did not have atrial fibrillation. We excluded participants aged 95 or older. To enable medical history assessment, we further excluded individuals who had been Danish residents for fewer than five years. Participants were followed up until the earliest of incident atrial fibrillation, death, age of 95 years, emigration, or the end of the period. In the primary analysis, 45 years was the index age. In secondary analyses, we repeated the analyses for the following index ages: 55 years or older, 65 years or older, and 75 years or older.

To assess newly diagnosed atrial fibrillation, we identified individuals with an incident hospital primary or secondary diagnosis of atrial fibrillation or atrial flutter (supplemental table 1). In the Danish National Patient Registry, the positive predictive value of atrial fibrillation or atrial flutter ICD code is 94-95% for inpatients and outpatients.2122

Participants and outcomes for lifetime risks of complications after atrial fibrillation

In each period, all patients with incident atrial fibrillation were further followed up for complications after their incident hospital diagnosis of atrial fibrillation up until death, age 95 years, emigration, or the end of the period. The population was restricted to patients with newly diagnosed atrial fibrillation in the study period and did not include prevalent patients to prevent survival bias. We identified the first occurrence after atrial fibrillation, if any, of each of the following outcomes: heart failure, any stroke, ischaemic stroke, myocardial infarction, and systemic embolism. For each outcome, we excluded patients with a prevalent diagnosis of the complication of interest at the time of incident atrial fibrillation. For example, when examining heart failure after atrial fibrillation, patients with a history of heart failure at or before the diagnosis of atrial fibrillation were excluded. To limit the possibility of misclassification, we did not use information about previous coronary artery bypass grafting or primary coronary intervention to identify prior myocardial infarction. Both procedures may be used for severe chronic coronary syndrome with no evidence of acute myocardial infarction. We also excluded patients who experienced a complication of interest fewer than seven days after newly diagnosed atrial fibrillation to ensure temporal separation between the diagnoses. All complications were identified as primary or secondary diagnoses (definitions in supplemental table 1). The positive predictive values of ICD codes were more than 80% for heart failure,23 more than 90% for ischaemic stroke,24 more than 70% for intracerebral haemorrhage,24 and more than 97% for myocardial infarction,22 while the value is unknown for systemic embolism. We conducted analyses for each of the index ages 45, 55, 65, and 75 years.

Covariates

To characterise the study populations, we assessed history of hypertension, diabetes, dyslipidaemia, heart failure, myocardial infarction, any stroke, cardiomyopathy, valvular heart disease, chronic obstructive pulmonary disease, chronic kidney disease, ischaemic stroke, and systemic embolism. Furthermore, we assessed family income and educational attainment from Statistics Denmark. Details are in the appendix. Information was assessed among individuals who did not have atrial fibrillation at cohort entry and among people who had atrial fibrillation at the time of incident atrial fibrillation (definitions in supplemental table 1).

Statistical methods

Age was the time scale in all analyses. We used the Aalen-Johansen estimator to calculate the cumulative incidence of atrial fibrillation and of complications after atrial fibrillation, accounting for left truncation and for the competing risk of death. The time at risk started at the exact entry age: index age or older for the lifetime risk of atrial fibrillation, and age at the diagnosis of incident atrial fibrillation for the lifetime risks of complications. The time at risk ended at the individual’s age at the earliest of the following dates: first incident atrial fibrillation or complication after atrial fibrillation, death, age 95 years, emigration, or end of study period. The lifetime risk was the cumulative incidence at the age of 95 years. We used the pseudo-value regression approach on the lifetime risk to obtain 95% confidence intervals (CI) and P values.13 We tested for differences in lifetime risks between men and women and between period 1 and period 2.

To translate the lifetime risk into the time domain, we estimated the restricted mean time lost as the area under the cumulative incidence function from the index age of 45 years up to 95 years.25 For the lifetime risk of atrial fibrillation, the restricted mean time lost gives the mean atrial fibrillation-free time lost between 45 and 95 years. For the lifetime risk of complications after atrial fibrillation, for example heart failure, the restricted mean time lost gives the mean time lost with no heart failure.

We repeated these analyses with adjustment for the covariates listed above. We estimated a propensity score to be in period 2 versus 1 by using all covariates in a logistic regression model. We then applied stabilised inverse propensity weights to the pseudo-value regression.

We also performed subgroup analyses for the lifetime risk of atrial fibrillation and of complications after atrial fibrillation (appendix). Subgroups were defined by sex, cardiometabolic risk factors, clinical comorbidities, and social factors, as aforementioned. We tested for differences in lifetime risks between subgroups and we tested for interaction between temporal trends across periods and subgroups.

Analyses were performed in Stata (StataCorp 2019: release 17.0, College Station, TX, USA).

Patient and public involvement statement

No funding was available to support patients or members of the public in the study design, interpretation of results, or development of the dissemination strategy. We appraised the registry based data and analysed them without public or patient involvement.

Results

Characteristics of participants

We included 3 574 903 individuals who did not have atrial fibrillation at index age 45 years or older, of whom 1 727 703 were men (48.3%) and 1 847 200 were women (51.7%). Flow diagrams showing the selection of participants are in supplemental figure 1. Participant characteristics are summarised in table 1 as well as supplemental tables 2 and 3 by sex. Among participants free of atrial fibrillation at entry, the prevalence of hypertension, diabetes, dyslipidaemia, and stroke increased from 2000-10 to 2011-22. Moreover, we followed up 362 721 individuals with newly diagnosed atrial fibrillation, of whom 194 505 were men and 168 216 were women. At diagnosis of atrial fibrillation, the distribution of age was similar between periods, while the prevalence of hypertension, dyslipidaemia, and diabetes increased over time, and the prevalence of heart failure and myocardial infarction decreased over time.

Table 1

Characteristics of the study populations. Data are number (percentage), unless otherwise specified

View this table:

Lifetime risk of atrial fibrillation

At the index age of 45 years, the lifetime risk of atrial fibrillation over 2000-22 was 27.7% ((95% CI 27.6% to 27.8%); table 2 and supplemental table 4). The lifetime risk of atrial fibrillation was higher among men, individuals with history of hypertension, heart failure, myocardial infarction, cardiomyopathy, dyslipidaemia, valvular heart disease and individuals with higher educational attainment and higher family income. History of stroke, chronic obstructive pulmonary disease, and chronic kidney disease were associated with lower lifetime risk of atrial fibrillation because they were also associated with considerably higher mortality (supplemental table 5 and supplemental figures 2-13).

Table 2

Lifetime risks of atrial fibrillation and of complications after atrial fibrillation at index age 45 years, 2000-22

View this table:

From 2000-10 to 2011-22, the overall lifetime risk of atrial fibrillation increased from 24.2% (95% CI 24.1% to 24.3%) to 30.9% (30.8% to 31.0%), an absolute increase of 6.7% ((6.5% to 6.8%); table 3 and supplemental tables 6). Findings were similar after adjustment for covariates at entry (supplemental table 7). Over time, the lifetime risk of atrial fibrillation increased across all subgroups (supplemental table 8 and supplemental figures 2-13). The increase in lifetime risk was slightly higher among men, individuals with a history of heart failure or stroke, and among individuals with no dyslipidaemia.

Table 3

Temporal trends in lifetime risks of atrial fibrillation and complications after atrial fibrillation, at index age 45 years

View this table:

All findings at index ages 55, 65, and 75 years were consistent. Across 2000-22, the lifetime risk of atrial fibrillation was 27.6% at 55 years, 26.9% at 65 years, and 24.3% at 75 years (supplemental table 9), and the lifetime risk of atrial fibrillation increased between the two periods by an absolute value of 6.5%, 6.3%, and 5.6% at index ages 55, 65, and 75 years (supplemental table 10).

Lifetime risk of complications after atrial fibrillation

Among patients with an incident diagnosis of atrial fibrillation, heart failure was the most frequent complication with a lifetime risk of 41.2% (95% CI 39.8% to 42.7%) from index age 45 years (table 2, fig 1, supplemental table 11-12). The lifetime risks of any stroke and ischaemic stroke after atrial fibrillation were 21.4% (20.6% to 22.3%) and 13.1% (12.4% to 13.8%). The lifetime risk of myocardial infarction after atrial fibrillation was 11.5% (10.9% to 12.2%) and diagnosed systemic embolism after atrial fibrillation was uncommon, with a lifetime risk of 1.8% (1.6% to 2.0%).

Fig 1
Fig 1

Cumulative incidence of complications after atrial fibrillation

Associations between covariates and lifetime risk of complications following atrial fibrillation are summarised in supplemental tables 13-17 and supplemental figures 14-74. Men had a higher lifetime risk of complications after atrial fibrillation compared with women for heart failure (44.2% v 34.6%) and myocardial infarction (12.2% v 10.2%), while the lifetime risk of stroke after atrial fibrillation was lower in men than in women (20.6% v 22.6%). Individuals who had a history of hypertension had a larger lifetime risk of heart failure, ischaemic stroke, myocardial infarction, and systemic embolism compared with those who did not. Additionally, people with history of myocardial infarction (58.4% v 39.6%), cardiomyopathy (84.5% v 39.9%), or valvular heart disease (62.5% v 39.6%) had notably higher lifetime risks of heart failure after atrial fibrillation. Conversely, individuals with prior chronic obstructive pulmonary disease or chronic kidney disease and lower family income generally had lower lifetime risks of complications following atrial fibrillation because of the association that these risk factors have with higher overall mortality rates (supplemental figures 14-74).

Over time, lifetime risk of heart failure after atrial fibrillation did not change (42.9% in 2000-10 v 42.1% in 2011-22) and slight decreases were noted in the lifetime risk for any stroke (−2.5% (95% CI −4.2% to −0.7%)), ischaemic stroke (−5.2% (−6.7% to −3.8%)), and myocardial infarction after atrial fibrillation (−3.9% (−5.3% to −2.4%) (table 3, supplemental tables 18-19). Analyses adjusted for covariates at the time of atrial fibrillation diagnoses showed consistent findings (supplemental table 20).

No evidence suggested differential temporal trends for all complications after atrial fibrillation according to sex and most covariates (fig 2, table 3, supplemental tables 21-25, supplemental figures 14-74). However, the lifetime risk of heart failure decreased from 2000-10 to 2011-22 among patients with atrial fibrillation and a history of hypertension (52.6% v 41.8%) or dyslipidaemia (52.1% v 44.5%). Additionally, the decrease from 2000-10 to 2011-22 in the lifetime risk of myocardial infarction after atrial fibrillation was larger among patients with dyslipidaemia versus people with normal lipids (−11.1% v −3.7%).

Fig 2
Fig 2

Summary plot of temporal trends in lifetime risk of complications by subgroups. CKD=chronic kidney disease; COPD=chronic obstructive pulmonary disease; VHD=valvular heart disease

The analyses of secondary index ages showed consistent findings. The lifetime risk of each complication was smaller with increasing index age (supplemental tables 26-28). Moreover, we found evidence of decrease over time in the lifetime risks for heart failure and systemic embolism after atrial fibrillation at index ages 55 and 75 years.

Discussion

Principal findings

In this nationwide study, the lifetime risk of newly diagnosed atrial fibrillation increased from approximately one in four in 2000-10 to one in three in 2011-22. The lifetime risk of atrial fibrillation increased across all subgroups over time, but the increase was larger among men and individuals with heart failure, myocardial infarction, stroke, diabetes, and chronic kidney disease. Among patients with newly diagnosed atrial fibrillation, heart failure was the most frequent complication after atrial fibrillation, with a lifetime risk of about two in five, twice as large as the lifetime risk of stroke after atrial fibrillation and four times greater than the lifetime risk of myocardial infarction after atrial fibrillation. Overall, lifetime risk of heart failure after atrial fibrillation showed almost no change, and only a slight decrease in the lifetime risks of any stroke, ischaemic stroke, and myocardial infarction after newly diagnosed atrial fibrillation. These slight improvements were similar among men and women. However, the lifetime risk of heart failure decreased from 2000-10 to 2011-22 for patients with history of hypertension and dyslipidaemia.

Comparison with other studies

In Europe, a study reported cumulative incidences of atrial fibrillation of about 27% at the age of 90 years in both men and women based on data from the BiomarCaRE Consortium.16 Another used data from the Rotterdam Study and noted a lifetime risk of atrial fibrillation at the index age of 55 years of 23.8% in men and 22.2% in women.26 A study in China reported a lifetime risk of 21% in women and 17% in men using a Chinese medical insurance database.27 In Taiwan, a study used the National Taiwanese Health Insurance Research Database and reported a lifetime risk of 15% among men and 17% among women.28 Additionally, in the US, data from the Framingham Heart Study were used and a lifetime risk of 37% for the index age of 55 years was reported.17 The Atherosclerosis Risk in Communities study was used to report a lifetime risk of 36% in white men, 30% in white women, 21% in African American men, and 22% in African American women.18 Data from prospective cohort studies are strengthened by systematic follow-up and may be limited by self-selection bias and under-representation of specific groups.29 However, our findings originate from national data and are consistent with prior estimates. No study has thus far examined the lifetime risks of complications following atrial fibrillation.

Interpretation of temporal trends

To our knowledge, the temporal trends in lifetime risks of atrial fibrillation and complications after atrial fibrillation have not been reported previously. The underlying reasons for the observed trends are speculative but likely multifactorial. For lifetime risk of atrial fibrillation, the detection of atrial fibrillation significantly improved over the past two decades, primarily due to advancements in technology, changes in clinical practice, and increased awareness. Approximately a third of patients with atrial fibrillation are asymptomatic,30 and enhanced detection over time may lead to increased lifetime risk of diagnosed atrial fibrillation. Another possible reason is the global increase in life expectancy. Additionally, survival after myocardial infarction and heart failure has improved, which increases the likelihood of developing atrial fibrillation and having atrial fibrillation detected due to increased surveillance. The lifetime risk increased slightly more in some subgroups, particularly those with history of heart failure or stroke, and improved attention on developing comorbidities may partially explain the increase. Finally, the observed increasing prevalence of key risk factors for atrial fibrillation over time, such as hypertension, dyslipidaemia, and diabetes, could also explain the increasing lifetime risk of atrial fibrillation. Although we did not have data for body mass index, the prevalence of obesity has increased over the past two decades in Denmark.31 However, the temporal trend of atrial fibrillation remained after adjustment for all available risk factors at entry.

Regarding complications after atrial fibrillation, lifetime risks of stroke and ischaemic stroke remained high, with only about 4-5% decrease between 2000-10 and 2011-22. Clinical guidelines for the treatment of atrial fibrillation have been published regularly since 2001,32 and new evidence supporting better treatment modalities, including target specific anticoagulants, has been incorporated since then.1133 These improvements in atrial fibrillation management may translate into a larger decline in lifetime risks of complications after atrial fibrillation over longer periods. Another reason for the decreasing risk over time is the potential early identification of patients with atrial fibrillation at low risk because of enhanced awareness. For instance, patients with persistent atrial fibrillation have a higher risk of stroke than patients with paroxysmal atrial fibrillation.34 The lifetime risk of heart failure after atrial fibrillation did not change overall, but over time heart failure substantially decreased among patients with history of hypertension and dyslipidaemia at atrial fibrillation diagnosis. The lifetime risk of myocardial infarction also had larger decrease among patients with dyslipidaemia. These trends may reflect the effect of anti-hypertensive and lipid lowering drugs. Key risk factors have increased over time, particularly hypertension, dyslipidaemia, and diabetes. The increasing prevalence may be explained by improvement of screening protocols for risk factors, leading to more comprehensive detection rates. The criteria for diagnosing hypertension and diabetes have changed over the timeline investigated in this study, with the thresholds for blood pressure and glycaemic concentrations being updated. Finally, lifestyle changes and an increase in risk factors in the general population, such as obesity and sedentary behaviour, may also play a role in the uptick of these conditions. Despite increasing prevalence of risk factors, our analyses adjusting for risk factors at the diagnosis of atrial fibrillation were consistent with the primary analysis.

Implications

Lifetime risk estimates provide epidemiological insights into the public health impact presented by atrial fibrillation and its complications. Communication of lifetime risk estimates may motivate preventive strategies, such as beneficial changes in lifestyle and strengthening of the focus on quality in medical care settings. Preventive strategies require knowledge of different components of long term risk after atrial fibrillation. However, the main body of data on the absolute risk of complications after atrial fibrillation originates from short term studies, and 11 years seems to be the maximum follow-up period.7353637 By adopting a lifetime perspective, we provided evidence on the actual disease burden associated with atrial fibrillation.

Estimates of the lifetime risk of complications after atrial fibrillation indicate the burden of the different complications, and such information may be essential to facilitate prioritisation and development of preventive efforts against complications, prediction of complications, and management of health policy. In Denmark, more than 85% of patients with atrial fibrillation initiate oral anticoagulation, and the one year and two year persistence are both above 85%.38 Despite a high rate of anticoagulation treatment, we observed that stroke risk remains a very important problem after atrial fibrillation. These findings underscore the need for treatments to further decrease stroke risk. Factor XIa inhibitors and left atrial appendage closure are currently evaluated in randomised controlled trials as first line stroke prevention measures. Moreover, although atrial fibrillation guidelines principally focus on stroke prevention,1133 our findings indicate that heart failure was the major complication after incident atrial fibrillation, with a lifetime risk of two in five patients with atrial fibrillation, twice greater than that of stroke. Our findings encourage greater attention to secondary prevention of heart failure after atrial fibrillation, and in line with a recent report from the National Heart, Lung, and Blood Institute, relevant future initiatives include development of non-anticoagulant pharmacotherapies, effective implementation of weight loss, and risk factor modification, and designing successful programmes for cardiac rehabilitation.10 As atrial fibrillation is a common arrhythmia, a lower incidence of complications may reduce the future economic costs in healthcare.

Limitations of the study

The data used in this study originate from a tax supported universal healthcare system with no substantial loss to follow-up of patients. The Danish National Patient Registry provided information on admissions to hospital for atrial fibrillation, but we were unable to examine individuals clinically to confirm the diagnosis. Furthermore, we could not use electrocardiogram monitoring devices such as Holter monitoring, and therefore, we may have missed patients with undiagnosed atrial fibrillation. We had no specific data for the proportion of patients with incident atrial fibrillation who were not captured by the Danish National Patient Registry. However, as the Danish guidelines recommend referral of patients with newly diagnosed atrial fibrillation to a public hospital for further examination, most patients with atrial fibrillation will be registered, and only a few cardiologists work outside of the public healthcare system in Denmark.38 In Sweden, where the healthcare system is very similar to the Danish system, a study using data from 2004 to 2010 from the Swedish National Patient Registry and a central electrocardiogram database found that 93.2% of patients with electrocardiogram confirmed atrial fibrillation had a recording in the Swedish National Patient Registry.39 Additionally, validation studies reported a high positive predictive value of atrial fibrillation and atrial flutter in the Danish National Patient Registry.2122 We could not differentiate between atrial fibrillation and atrial flutter; however, atrial accounts for approximately 5% of the ICD-10 I48 diagnoses.21 The positive predictive values for stroke, heart failure, and myocardial infarction also seem adequate.222324 The positive predictive value of ICD-10 I45 codes for systemic embolism is unknown. The diagnostic validity is likely to be even higher in a cohort of patients with a history of atrial fibrillation because these individuals are likely to be under closer medical surveillance. We had no information on the cause of death, and accordingly, we may have underestimated the true number of incident events and the lifetime risk. We also did not have information on lifestyle factors, such as obesity, smoking status, and physical activity.

Assessing temporal trends in lifetime risk of atrial fibrillation and its complications requires extended follow-up durations of population based samples with minimal loss to follow-up. The Danish registries offer these unique features. However, we were able to assess the lifetime risks over only two periods. Moreover, lifetime risks of atrial fibrillation and subsequent complications may vary according to race and ethnic group, geographical location, health system, and country income, and our results may not be transportable to other countries or settings. Lastly, we did not separate calendar temporal trends from birth cohort effects, although analyses with age as the time scale partially address this limitation.40

Conclusions

Our nationwide study shows that the lifetime risk of atrial fibrillation increased over the past two decades from one in four to one in three. After atrial fibrillation, heart failure was the most frequent complication, with a lifetime risk of two in five, twice greater than the lifetime risk of stroke after atrial fibrillation. The lifetime risks of stroke, ischaemic stroke, and myocardial infarction following atrial fibrillation improved only modestly over time and remained high, while virtually no improvement was noted in the lifetime risk of heart failure after atrial fibrillation. Our novel quantification of the long term downstream consequences of atrial fibrillation highlights the critical need for treatments to further decrease stroke risk as well as for heart failure prevention strategies among patients with atrial fibrillation.

What is already known on this topic

  • Once atrial fibrillation develops, patient care has focused on the risk of stroke

  • Long term downstream consequences of atrial fibrillation, including stroke, heart failure, and myocardial infarction, must be better studied

  • Changes over time in the lifetime risks of atrial fibrillation and of complications after atrial fibrillation remain to be fully explored with new stroke prevention treatment for atrial fibrillation

What this study adds

  • In registry analyses of the entire population in Denmark over 2000-22, the lifetime risks of complications after atrial fibrillation were high

  • The most common complication was heart failure (lifetime risk 41%), followed by stroke (21% for any stroke and 13% for ischaemic stroke) and myocardial infarction (12%)

  • From 2000-10 to 2011-22, virtually no improvement was reported in the lifetime risk of heart failure after atrial fibrillation (−0.8%) and only slight decreases in the lifetime risks of stroke (−2.5% for any stroke, −5.2% for ischaemic stroke) and myocardial infarction (−3.9%) after atrial fibrillation

Ethics statements

Ethical approval

The Danish Health Data Authority, Statistics Denmark, and the Danish Data Protection Agency approved this study. Registry‐based studies do not require approval from an ethics committee according to Danish law.

Data availability statement

Permission to access the data used in this study can be obtained following approval from the Danish Health Authority.

Footnotes

  • Contributors: NV, LF, EJB, and LT developed the hypothesis. NV, LF, and LT developed the study design. NV, PC, and LT did the statistical analysis. NV and LT wrote the first and successive drafts of the manuscript. All authors contributed to analysis and interpretation of data and drafting or critical revision of the manuscript for important intellectual content or additionally to data acquisition. NV and PC had full access to the data and take responsibility for the integrity of the data and the accuracy of the data analysis. NV and LT are the guarantors. The corresponding author attests that all listed authors meet authorship criteria and that no others meeting the criteria have been omitted.

  • Funding: This work was supported by a research grant from the Danish Cardiovascular Academy (PD2Y-2022002-DCA), which is funded by the Novo Nordisk Foundation, grant number NNF20SA0067242 and The Danish Heart Foundation. The sponsor had no role in the study design, data collection, statistical analysis, interpretation of data; in the writing of the manuscript, or in the decision to submit the article for publication.

  • Competing interests: All authors have completed the ICMJE uniform disclosure form at www.icmje.org/disclosure-of-interest and declare NV has served as an advisory board member and consultant for AstraZeneca, no fees were received personally. SPJ has an institutional research grant from BMS/Pfizer (not related to the current study) and personal consulting fees received from BMS and Pfizer. EJB has a grant R01HL092577; American Heart Association AF AHA_18SFRN34110082. LF is supported by the Health Research Foundation of Central Denmark Region and has served as a consultant for BMS/Pfizer and AstraZeneca. LT was supported by a research grant from the American Heart Association (18SFRN34150007). PC and LS declare no competing interests.

  • Transparency: NV and LT affirm that the manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as originally planned have been explained.

  • Dissemination to participants and related patient and public communities: The study's findings will be disseminated through press releases, media engagement, and social media outreach. We also will present findings at conferences oriented towards clinicians that manage patients with atrial fibrillation and those at risk for developing atrial fibrillation. Our target audiences include the general public, including patients with atrial fibrillation and those at risk for developing atrial fibrillation, as well as healthcare professionals (eg, cardiologists, general practitioners, and nurses) and public health and cardiovascular health researchers.

  • Provenance and peer review: Not commissioned; externally peer reviewed.

http://creativecommons.org/licenses/by-nc/4.0/

This is an Open Access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.

References