Abstract
Background
Oral anticoagulants (OACs) effectively reduce stroke and mortality in patients with atrial fibrillation (AF), but their use increases the risk of gastrointestinal (GI) bleeding. The combined mortality burden from deaths in which AF and GI bleeding are co-listed has not been comprehensively characterized. This study evaluated national trends and demographic disparities in AF- and GI bleeding–related mortality from 1999 to 2020.
Methods
Mortality data were obtained from the CDC WONDER database for U.S. residents from 1999 to 2020. Deaths were included only when both AF and GI bleeding were recorded anywhere on the death certificate as underlying or contributing causes. Crude and age-adjusted mortality rates (AAMR) per 100,000 population were calculated, and Joinpoint regression estimated annual percent changes (APC).
Results
Deaths involving AF and GI bleeding increased from 1,486 in 1999 to 4,855 in 2020. AAMR rose from 0.55 to 1.16 per 100,000, with an accelerated rise between 2010 and 2020 (APC = 5.65%, p < 0.05). Males showed higher AAMR than females (AAPC = 3.61% vs. 2.83%), and White populations had the highest rates, while American Indian/Alaska Natives had the lowest. Nonmetropolitan areas showed faster increases than metropolitan ones (AAPC = 4.20% vs. 2.88%), with the central and western regions experiencing the steepest growth. Mortality was highest among older adults.
Conclusions
AF- and GI bleeding–related mortality more than doubled from 1999 to 2020, disproportionately affecting elderly men, rural populations, and western U.S. regions. Targeted preventive strategies are essential to mitigate these rising disparities.
Keywords: Atrial fibrillation, Gastrointestinal bleeding, Direct oral anticoagulants, Epidemiology, CDC WONDER
Introduction
In the United States (US), atrial fibrillation (AF) affects up to 1 in 3 people in their lifetime and is estimated to affect approximately 15.90 million people by 2050 [1–3]. AF is associated with a significantly increased mortality and morbidity, mainly via stroke [4]. Over one in five AIS hospitalizations in the US have comorbid AF, and this proportion increased over the period 2004–2013 [5]. Direct Oral anticoagulants (DOACs) are the preferred anticoagulation therapy for patients with atrial fibrillation [6, 7]. Oral anticoagulation (OAC) with warfarin or several DOACs is highly effective in mitigating AIS risk from AF [8], and OAC utilization in AF patients in the United States increased significantly over the period 2010–2020 [9, 10]. Anticoagulation to prevent ischemic stroke is one of the cornerstones of AF treatment [11], and OACs reduce stroke and mortality in patients with AF. Previous studies have examined AF or GI bleeding mortality independently. In this study, we examine deaths in which both AF and GI bleeding were co-listed as underlying or contributing causes.
Despite the availability of effective anticoagulation therapies, the management of AF remains challenging due to complications such as bleeding, particularly gastrointestinal (GI) bleeding, which represents a major adverse effect of OAC use. GI bleeding not only limits the continued use of anticoagulation but also contributes significantly to morbidity and mortality in this population [11]. Moreover, the aging demographic, increasing prevalence of comorbidities, and expanded indications for anticoagulation have collectively intensified the clinical and public health burden associated with AF and its complications. Understanding how mortality patterns related to both AF and GI bleeding have evolved over time is therefore crucial to guide preventive strategies and improve therapeutic decision-making in real-world settings.
This study aims to elucidate the temporal trends in AF and gastrointestinal bleeding-related mortality in the US from 1999 to 2020. Using the US Centers for Disease Control and Prevention’s Wide-ranging Online Data for Epidemiologic Research (CDC WONDER) database, we seek to identify demographic disparities in sex, age, race, urbanization, and geographic region to inform targeted interventions, optimize clinical practices, and shape public health policies for this vulnerable population.
Methods
Study setting and population
This descriptive study analyzed mortality trends using death certificate data from the CDC WONDER database. The primary aim was to assess mortality rates associated with atrial fibrillation and gastrointestinal bleeding among individuals between 1999 and 2020. The study utilized records from the Multiple Cause of Death (MCD) Public Use dataset, which provides comprehensive mortality data across all 50 US States and the District of Columbia [12]. AF related deaths were identified using the ICD-10 code I48, while gastrointestinal bleeding-related deaths were classified under the ICD-10 code K25-K27, K92. Individuals with both atrial fibrillation and gastrointestinal bleeding in their death certificates were included in the analysis. The study includes deaths in which both AF and gastrointestinal bleeding were recorded as underlying or contributing causes on the death certificate. Given that the study was based on publicly available, de-identified data from a federal database, institutional review board (IRB) approval was not required. The study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting checklist to ensure methodological rigor.
Data abstraction
Mortality data were stratified by key demographic and geographic variables, including sex, race/ethnicity, urbanization level, census region, and state of residence. Racial classifications are designed as Asian/Pacific Islander, Indian/Alaska Native, White, and Black/African American consistent with classifications commonly used in prior analyses of the WONDER database. Urban and rural designations were assigned using the National Center for Health Statistics(NCHS) Urban-Rural Classification Scheme, which defines urban areas as large metropolitan regions (populations exceeding 1 million) and medium/small metropolitan regions (populations between 50,000 and 999,999), while rural areas include counties with populations below 50,000, as per the 2013 US Census data. Geographic regions were classified in accordance with the US Census Bureau’s regional divisions into Northeast, Midwest, South, and West [13].
Statistical analysis
Age-adjusted mortality rates (AAMR) per 100,000 individuals were calculated to evaluate national mortality trends related to atrial fibrillation and gastrointestinal bleeding, standardized to the 2000 US population [14]. Temporal changes in mortality rates were analyzed using the Joinpoint Regression Program (Version 5.2.0, National Cancer Institute), which employs log-linear regression models to estimate the annual percent change (APC) in AAMR, along with corresponding 95% confidence intervals (CIs) [15]. Trends were classified as increasing or decreasing based on deviations from the null hypothesis of no change, with statistical significance set at P < 0.05, using a two-tailed t-test. A parallelism test was also conducted to assess whether mortality trends differed significantly across demographic subgroups; a significant P-value in this test indicated a meaningful divergence in average annual percentage change (AAPC) trends [16].
Results
In 1999 and 2020, atrial fibrillation and gastrointestinal bleeding accounted for 1,486 and 4,855 deaths, respectively (Table 1). This study initially included 1,486 patients with atrial fibrillation combined with gastrointestinal bleeding in 1999, of whom 41.38% were male, 78.86% lived in Metro, and 91.04% were Caucasian. By 2020, the cohort had increased to 4,855 participants, of whom 48.49% were male, 79.84% lived in Metro, and 88.69% were Caucasian.
Table 1.
Frequency and age-adjusted mortality rates per 100,000 in older adults with atrial fibrillation and Gastrointestinal bleeding concomitantly stratified by sex, race, census region and urbanization
| 1999 | 2020 | ||||
|---|---|---|---|---|---|
| Counts | AAMR | Counts | AAMR | AAPC | |
| All | 1486 | 0.55 (0.52, 0.58) | 4855 | 1.16 (1.13, 1.20) | 3.03 (2.54 to 3.6) |
| Sex | |||||
| Female | 871 | 0.49 (0.46, 0.52) | 2321 | 0.94 (0.90, 0.98) | 2.83 (2.53 to 3.23) |
| Male | 615 | 0.63 (0.58, 0.68) | 2534 | 1.48 (1.42, 1.54) | 3.61 (3.09 to 4.06) |
| Census Region | |||||
| Northeast | 360 | 0.61 (0.55, 0.68) | 923 | 1.16 (1.09, 1.24) | 2.16 (1.51 to 2.92) |
| Midwest | 365 | 0.55 (0.50, 0.61) | 1122 | 1.25 (1.18, 1.32) | 3.81 (3.25 to 4.45) |
| South | 489 | 0.51 (0.47, 0.56) | 1751 | 1.13 (1.07, 1.18) | 3.44 (3.05 to 3.88) |
| West | 272 | 0.51 (0.45, 0.57) | 1059 | 1.14 (1.07, 1.21) | 3.76 (3.23 to 4.73) |
| Urbanization | |||||
| Metro | 1172 | 0.53 (0.50, 0.56) | 3876 | 1.10 (1.07, 1.14) | 2.88 (2.48 to 3.36) |
| Nonmetro | 314 | 0.60 (0.53, 0.67) | 979 | 1.42 (1.33, 1.51) | 4.2 (3.69 to 4.86) |
| Race | |||||
| Asian or Pacific Islander | 21a | 0.98 (0.59, 1.53) | 142 | 0.96 (0.64, 1.38) | 1.77 (0.11 to 3.47) |
| Black or African American | 112 | 0.51 (0.41, 0.60) | 376 | 0.68 (0.57, 0.80) | 3.1 (2.25 to 4.06) |
| White | 1353 | 0.55 (0.52, 0.58) | 4306 | 0.96 (0.86, 1.06) | 3.36 (3 to 3.83) |
| American Indian or Alaska Native | 23b | 0.38 (0.24, 0.57) | 31 | 1.21 (1.18, 1.25) | -2.17 (-14.76 to 19.22) |
a: indicates that the classification is available from 2013 onwards; b: indicates that data is available from 2001
Annual trends for atrial fibrillation and gastrointestinal bleeding-related mortality
The age-adjusted mortality rate (AAMR) for atrial fibrillation-related gastrointestinal bleeding deaths rose sharply from 0.55 to 1.16 per 100,000 between 1999 and 2020. The average annual percentage change (AAPC) was 3.03% (95% CI: 2.54–3.6). Joinpoint regression identified 2010 as the pivotal trend transition year, with the AAMR increasing modestly at 0.71% annually before 2010 but accelerating markedly to 5.65% annually thereafter (Fig. 1).
Fig. 1.
Annual trends for atrial fibrillation and gastrointestinal bleeding -related mortality AAPC Average annual percent change
Atrial fibrillation and gastrointestinal bleeding -related mortality trends stratified by sex
The AAMR was consistently higher among males than females, with a more rapid increase (AAPC = 3.61%, 95% CI: 3.09–4.06 vs. 2.83%, 95% CI: 2.53–3.23). In males, the AAMR exhibited a steady rise from 2009 to 2018, followed by a sharper increase from 2018 to 2020 (APC = 12.59%, P < 0.05). Among females, the AAMR showed a transient decline in 2006 (APC = − 3.32%, P > 0.05) but accelerated significantly after 2009 (APC = 5.00%, P < 0.05) (Fig. 2).
Fig. 2.
Sex-based trends in atrial fibrillation and gastrointestinal bleeding related age adjusted mortality rates from 1999 to 2020. AAPC Average annual percent change
Atrial fibrillation and gastrointestinal bleeding -related mortality trends stratified by race
The white group exhibited a significant increase in AAMR after 2009 (APC = 5.88%, P < 0.05). Similarly, the black group demonstrated a marked upward trend beginning in 2011 (APC = 6.82%, P < 0.05). Among Asian/Pacific Islanders (API), AAMR rose sharply starting in 2015 (APC = 9.56%, P < 0.05), while American Indian/Alaska Native (AI/AN) populations showed a declining trend (AAPC = -2.17%, 95% CI: -14.76 to 19.22), revealing substantial ethnic heterogeneity (Fig. 3).
Fig. 3.
Race -based trends in atrial fibrillation and gastrointestinal bleeding related age adjusted mortality rates from 1999 to 2020. AAPC Average annual percent change
Atrial fibrillation and gastrointestinal bleeding -related mortality trends stratified by census region
The central and western regions exhibited the most pronounced AAMR increase (AAPC = 3.81%, 95% CI:3.25–4.45), particularly during 2018–2020 (APC = 12.72%, P < 0.05). Southern and western regions followed with AAPCs of 3.44% (95% CI: 3.05–3.88) and 3.76% (95% CI: 3.23–4.73), respectively, while the northeast region showed the lowest growth (AAPC = 2.16%, 95% CI: 1.51–2.92). Regional disparities in joinpoint years suggest widening geographic variation in mortality burden over time (Fig. 4).
Fig. 4.
Census Region -based trends in atrial fibrillation and gastrointestinal bleeding related age adjusted mortality rates from 1999 to 2020. AAPC Average annual percent change
Atrial fibrillation and gastrointestinal bleeding-related mortality trends stratified by urbanization
The AAMR rose more sharply in non-metropolitan areas than in metropolitan regions (AAPC: 4.20%, 95% CI: 3.69–4.86 vs. 2.88%, 95% CI: 2.48–3.36). Between 2018 and 2020, non-metropolitan areas experienced an exceptionally high APC of 11.52% (P < 0.05), contrasting with a transient decline from 2005 to 2008 (APC: −3.63%, P > 0.05). Metropolitan areas showed sustained rapid growth after 2010 (APC: 5.52%, P < 0.05) (Fig. 5).
Fig. 5.
Differences in the level of urbanization based trends in atrial fibrillation and gastrointestinal bleeding related age adjusted mortality rates from 1999 to 2020. AAPC Average annual percent change
Atrial fibrillation and gastrointestinal bleeding-related mortality trends stratified by age
The figure above shows the trend of the composition of mortality from this cause by age group (Fig. 6A) and crude mortality (Fig. 6B) by age group from 1999 to 2020. The results showed that the deaths were mainly concentrated in the elderly population, of which the “85 years old and above” people accounted for the highest proportion for a long time, which remained stable at about 50%, followed by the “75–84 years old” people, which accounted for more than 80% in total, suggesting that the death of this cause was highly concentrated in the old age. From the perspective of the change trend of crude mortality (Fig. 6B), the mortality rate of all age groups showed a slow upward trend, especially the mortality rate of people aged 85 and above, which increased from about 17/100,000 in 2000 to nearly 30/100,000 in 2020. The mortality rate in the “75–84 years” group also increased steadily, reaching about 10/100,000 in 2020. The mortality rate in the 45–54 and 55–64 age groups remained low with little significant change. There are significant differences in mortality risk and growth rate gradients between different age groups, reflecting that the elderly population is the main source of burden.
Fig. 6.
Age based trends in atrial fibrillation and gastrointestinal bleeding related age adjusted mortality rates from 1999 to 2020
Atrial fibrillation and gastrointestinal bleeding-related mortality stratified by geographic regions
The figure above shows the geographic distribution of age-standardized mortality rates (ASR, Fig. 7A) and their average annual percentage change from 1999 to 2020 (AAPC, Fig. 7B) by U.S. state in 2020. The results show that ASR is generally higher in the North Central region, especially in South Dakota, North Dakota, Missouri, Wisconsin, and Georgia, with an ASR of more than 2.0/100,000, which is significantly higher than the national average. In contrast, some parts of the West (such as California, Arizona) have lower ASR. AAPC maps reflect geographic differences in the rate of mortality growth over the past two decades. The fastest growing states include Nevada (AAPC > 20%), followed by southern states such as Oklahoma, Tennessee, and Louisiana, all showing higher growth trends (AAPC > 10%). In contrast, some states in the north-central and New England region (such as Montana, North Dakota, Maine, etc.) are missing and are shown in grey.
Fig. 7.
Geographic Regions based trends in atrial fibrillation and gastrointestinal bleeding related age adjusted mortality rates from 1999 to 2020. AAPC Average annual percent change
Overall, the mortality rate from this cause shows obvious spatial heterogeneity in the United States, and many states in the southwest and south are at a high level in terms of existing death burden and growth rate, suggesting that targeted intervention and resource allocation should be strengthened in high-burden areas.
Discussion
This study analyzed mortality trends related to atrial fibrillation and gastrointestinal bleeding in the US from 1999 to 2020 using CDC WONDER data. The findings revealed an overall rise in AAMR until 2010, after which a pronounced surge was observed. Males consistently exhibited higher AAMR than females. Among racial groups, the white group had the highest mortality rates, followed by Black or African American individuals. Finally, differences in mortality rates were revealed on a state and regional basis, in addition to comparing urban and rural areas. Because we used a multiple-cause-of-death approach, our results reflect deaths where AF and GI bleeding were co-listed, not only those in which AF was the underlying cause. This approach captures a wider spectrum of clinically relevant fatal events.
There are several factors that could have played a role in the overall increase in mortality in the US during this time. This trend is driven by the aging US population [17]. The occurrence of atrial fibrillation across all patient populations has been increasing [18]. Patients with AF and risk factors for thromboembolism require anticoagulant treatment to reduce the risk of stroke [19, 20]. The major complication with anticoagulant treatment is the increased risk of bleeding [21], particularly gastrointestinal bleeding [22–24]. Gastrointestinal bleeding is a serious event with high case fatality, and this study supports the results from the UK national audit 2007 [25]. Gastrointestinal bleeding is associated with a high mortality, especially among elderly patients with atrial fibrillation and multiple comorbidities who take antithrombotics [26]. In this Danish cohort study (1996–2012), 39.9% of patients died within two years after gastrointestinal bleeding [27]. These findings reinforce the importance of gastrointestinal bleeding as a powerful indicator of death. While we have observed changes in the prescribing of anticoagulants for patients with atrial fibrillation and a rise in mortality over time, our ecological study methodology does not allow for definitive conclusions regarding the causal nature of these events. Specifically, although a marked increase in anticoagulants prescribed to patients with atrial fibrillation has occurred since 2010, we cannot definitively associate their increased use with an increased incidence of atrial fibrillation–gastrointestinal bleeding or increased cardiovascular-related deaths within the same time period. Accordingly, our findings should be viewed as an initial step in generating hypotheses rather than proving a direct relationship between the adoption of new oral anticoagulants (DOAC) and adverse outcomes related to atrial fibrillation. To fully understand this relationship in further detail, future studies must utilize individual-level data (for example, type of anticoagulant prescribed and adherence to dosing instructions) so that the impact of each factor on the incidence of thromboembolic events and the associated risk of strokes can be assessed. Finally, it is essential to acknowledge that patients with AF aged over 65 years have a higher baseline risk than younger patients for both thrombosis and bleeding regardless of whether they are taking anticoagulation therapy. This underlying risk may be part of the reason for the increase in mortality that occurred between 2010 and 2011, as well as the rapid spike that occurred between 2019 and 2020. The latter spike occurred during the early part of the COVID-19 pandemic, at which point there were widespread disruptions to routine medical care (e.g., reduced access to routine medical care), delayed presentations for medical attention, and poor management of chronic diseases [28]. There are a number of indirect factors that may have contributed to both the mortality trends and potential excess mortality that were observed between 2010 and 2011, and between 2019 and 2020, and these explanations are speculative until such time that they can be confirmed or ruled out through the use of robust clinical research methodologies.
Our findings indicate males consistently exhibited higher AAMR than females. Celina et al. demonstrated lower anticoagulation with DOACs among women, supporting older findings from North America (US Medicare population and a Canadian study) [29–32]. The Registry of US ambulatory encounters also demonstrated lower DOAC use among women [33]. These will reduce the risk of gastrointestinal bleeding in women, and studies have shown that smoking and alcohol consumption can induce atrial fibrillation [34], and alcohol consumption also increases the risk of bleeding in patients with atrial fibrillation [35]. The rate of drinking in men is relatively higher than that of women [36]. Another reason might be that female are under-prescribed with OACs, so more OAC therapy in men more bleeding. The above is the reason why AAMR is consistently higher in men with atrial fibrillation and gastrointestinal bleeding than in women.
Racial differences in AF–GI bleeding mortality likely reflect a combination of biological and structural factors, including AF prevalence, comorbidity burden, anticoagulant prescribing patterns, healthcare access, and health literacy. Importantly, lower AF prevalence in certain groups does not necessarily translate into lower mortality from AF-related bleeding once anticoagulation is prescribed. The higher mortality observed among older White and Black populations in our analysis likely results from the intersection of these factors, highlighting that severe bleeding risk is modulated by both patient-level and system-level determinants.
Mortality rates were consistently higher in non-metropolitan areas and varied across U.S. regions. These patterns likely reflect the combined influence of healthcare access, physician density, socioeconomic conditions, and local public health initiatives. State-level differences may additionally be affected by variations in death certificate coding, population age structure, and reporting practices. While the ecological design precludes causal inference, these observations generate hypotheses for future studies investigating how local healthcare resources and social determinants contribute to disparities in AF–GI bleeding mortality.
The studies have shown that Black individuals have a lower prevalence of AF compared to whites [37]. A meta-analysis incorporating data from the ARIC and CHS studies indicated that a 10% increase in European ancestry was associated with a 13% higher risk of AF [38]. Genetic factors play a key role in these disparities. A protective minor allele of pituitary homeobox 2 related single-nucleotide polymorphism is more common in Black individuals, reducing their AF risk by 11.4%–31.7% [39]. This may partly explain why AF incidence and mortality are higher in Whites. Detection differences also contribute. White men tend to have larger left atria, making AF easier to detect, while Black individuals may have more undiagnosed cases due to studies missing paroxysmal or silent AF episodes [40,41]. Lower awareness of AF, reduced health literacy, limited healthcare access, and provider bias further worsen disparities [42, 43].
Lastly, this study found geographical differences in AAMR throughout the United States. AAMR were consistently higher in rural regions in comparison to urban regions. This is congruent with findings from previous literature that suggest increased rural mortality may be due to treatment delays, socioeconomic inequality in rural communities, physician shortages, and lack of health insurance in rural communities [44–46]. Regional and state-level differences in atrial fibrillation and gastrointestinal bleeding mortality trends were also noted in this study. Regions with more rural communities, such as the Midwest United States, had higher increases in AAMR. The Northeast had the lowest AAMR, and this region notably has many more massive urban locations than the other regions. This further supports the tie between atrial fibrillation and gastrointestinal bleeding mortality, healthcare access, and socioeconomic inequality. However, a specific trend for state–level differences remains unclear. Many states with increasing AAMR, including South Dakota, North Dakota, also have a large rural community. Although some parts of the West (such as California, Arizona) have decreasing AAMR. It is possible these differences are due to state health initiatives or simply state population size.
This study has noted many differences between the mortality rates associated with AF and GI bleeding. The differences exist by sex, race, age, geographic location, and by whether patients live in urban areas. There is a need for all patients to receive a full range of care. The patterns of mortality reflect a combination of newly diagnosed individuals with AF as well as those with chronic AF over a 20-year period. Newly diagnosed patients have different trajectories of risk and vulnerability to bleeding when compared to patients with chronic AF. Each of these groups cannot be differentiated from each other using death certificate data and, thus, the trends observed in this study are likely the result of a combination of increased incidence of AF and improved long-term survival of patients with AF. We recommend linking claims data to mortality data and/or AF registries to determine the timing of excess deaths in patients with AF, whether they occur at the time of diagnosis of AF or later in the disease course.
There is also a need to discuss the importance of clinical complexity and multimorbidity on the outcomes in AF. The higher mortality rates observed among older adults, rural residents and residents of geographic areas that have higher levels of multiple chronic illnesses can be associated with a number of factors, including, but not limited to: frailty, heart failure, chronic kidney disease, increased susceptibility to infections, and polypharmacy, all of which dramatically increase the vulnerability to bleeding and competing causes of death beyond the impact of selecting the right anticoagulant.
Changes in time such as increases in risk factors for AF (anticoagulant use) as well as advanced age (60 + years), in contrast to older adults, are an indication that bleeding risk is dynamic and is determined by age, co-morbidity, and medications being taken. As such, patients’ risk should be assessed on multiple longitudinal assessments and not simply at a baseline.
Temporal relationships between increased use of anticoagulants and patients’ demographic background, socioeconomic status, access to care, and other factors may demonstrate that increased risk of bleeding for patients with AF is due to multiple variables; hence, the association should not be seen as having a cause-and-effect relationship.
These results demonstrate that integrated, evidence-based AF care is critical to reducing death, stroke, and major bleeding in both high-risk and underserved populations and that comprehensive models such as AF-CARE, SOS and ABC Pathway can reduce these risks through randomised clinical trials as well as real-world studies of these models. To support comprehensive AF care, mobile health may greatly improve patients’ ability to adhere to their recommendations and facilitate monitoring and rapid decision-making regarding clinical care. This will assist in counteracting the demographics and geography of the reduced care access demonstrated in the results of this study.
Limitations
Several limitations are to be acknowledged. One limitation of our study lies in the use of the CDC WONDER database itself due to the potential underestimation of mortality rates. The database depends on the accurate reporting and coding of death certificates by healthcare professionals, so different interpretations or errors could lead to inaccurate data. Importantly, the dataset does not include information on oral anticoagulant prescription, drug type (warfarin vs. DOACs), dosing, switching, duration of therapy, or adherence. Therefore, the contribution of anticoagulation patterns to mortality cannot be assessed. Individuals with AF often possess a high baseline risk of both thromboembolism and bleeding, particularly when multiple comorbidities are present, which may independently influence mortality trends. Second, our analysis could not account for various confounding factors, including underlying health conditions, socioeconomic status, health insurance type, education, and occupation. Thirdly, it is important to mention potential double-counting or misclassification bias, especially in older adults with multimorbidity. Lastly, as this study relies on aggregate data, it is also subject to potential limitations such as ecological bias when drawing conclusions at the population level.
Importantly, this study defined outcomes based on death certificates in which atrial fibrillation and gastrointestinal bleeding were co-listed anywhere as underlying or contributing causes of death. As such, the analysis cannot distinguish whether AF or gastrointestinal bleeding directly caused death, contributed causally, or was incidentally recorded in the context of severe multimorbidity. Consequently, the reported mortality rates should not be interpreted as deaths caused by atrial fibrillation or gastrointestinal bleeding alone, but rather as deaths in which these conditions were present at the time of death. This limitation is inherent to multiple-cause-of-death analyses and underscores that our findings are descriptive in nature and intended to be hypothesis-generating rather than causal.
Conclusion
This study identified an overall rise in mortality among deaths in which AF and GI bleeding were co-listed from 1999 to 2020, with a pronounced acceleration after 2010. This increasing trend in mortality was seen across gender, race, age, and regional factors. Gastrointestinal bleeding-related mortality in AF patients was significantly higher in the male, white, and 85 + years old populations, with the Midwest and Nonmetro having the highest gastrointestinal bleeding-related mortality rates as well. Our findings stress the need to increase both patient and provider education on the clinical signs of gastrointestinal bleeding in high-risk AF patient populations to allow for proper, timely treatment. It is also crucial to address the underlying socioeconomic disparities that leave the aforementioned patient populations with an unjust mortality burden.
Acknowledgements
All of our authors are grateful to the CDC WONDER database officials for sharing such valuable data.
Authors’ contributions
SX and LD designed and supervised the study. SJ analyzed the data. YZ, and SX wrote and edited the draft. LW contributed with a critical revision of the manuscript. LM provides financial support. LM and LD contributed language editing. All authors have read and approved the final manuscript.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
Data availability
The datasets generated and/or analyzed during the current study are available on the CDC Wonder Database, https://wonder.cdc.gov/.
Declarations
Ethics approval and consent to participate
No ethical approval required for use of anonymous publicly available data.
Consent for publication
Permission from all authors to publish this article.
Competing interests
The authors declare no competing interests.
Clinical trial number
Not applicable.
Footnotes
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Shuai Jin, Shigang Xu, Yingchuan Zhang, Lixue Wu, Liwei Duan and Linhao Ma contributed equally to this work.
Contributor Information
Shuai Jin, Email: jssss@gmc.edu.cn.
Linhao Ma, Email: macro118118@163.com.
References
- 1.Mou L, Norby FL, Chen LY, et al. Lifetime risk of atrial fibrillation by race and socioeconomic status: ARIC study (Atherosclerosis risk in Communities). Circ Arrhythm Electrophysiol. 2018;11(7):e006350. 10.1161/CIRCEP.118.006350. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.WengLC. PreisSR,HulmeOL,et al.Genetic predisposition, clinical risk factor burden, and lifetime risk of atrial fibrillation. Circulation. 2018;137(10):1027–38. 10.1161/CIRCULATIONAHA.117.031431. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.NoubiapJJ,TangJJ,TeraokaJT. Minimum National prevalence of diagnosed atrial fibrillation inferred from California acute care facilities. J Am Coll Cardiol. 2024;84(16):1501–8. 10.1016/j.jacc.2024.07.014. [DOI] [PubMed] [Google Scholar]
- 4.Askarinejad A, Lane DA, Sadeghipour P, et al. Stroke prevention in atrial fibrillation: A narrative review of current evidence and emerging strategies. Eur J Clin Invest. 2025;55(9):e70082. 10.1111/eci.70082. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Otite FO, Khandelwal P, Chaturvedi S, et al. Increasing atrial fibrillation prevalence in acute ischemic stroke and TIA. Neurology. 2016;87(19):2034–42. 10.1212/WNL.0000000000003321. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.JoglarJA,ChungMK,ArmbrusterAL,et al;Peer, et al. ReviewCommitteeMembers.2023ACC/AHA/ACCP/HRSguidelineforthediagnosis and management of atrial fibrillation: a report of the AmericanCollegeofCardiology/American heart association joint committee on clinical practice guidelines. Circulation. 2024;149(1):e1–156. 10.1161/CIR.0000000000001193. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Van Gelder IC, Rienstra M, Bunting KV, et al. 2024 ESC guidelines for the management of atrial fibrillation developed in collaboration with the European association for Cardio-Thoracic surgery (EACTS). Eur Heart J. 2024;45(36):3314–414. 10.1093/eurheartj/ehae176. [DOI] [PubMed] [Google Scholar]
- 8.Katsanos AH, Kamel H, Healey JS, et al. Stroke prevention in atrial fibrillation: looking forward. Circulation. 2020;142(24):2371–88. 10.1161/CIRCULATIONAHA.120.049768. [DOI] [PubMed] [Google Scholar]
- 9.Ko D, Lin KJ, Bessette LG, et al. Trends in use of oral anti-coagulants in older adults with newly diagnosed atrial fibrillation, 2010–2020. JAMA Netw Open. 2022;5(11):e2242964. 10.1001/jamanetworkopen.2022.42964. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Maddox TM, Song Y, Allen J, et al. Trends in U.S. Ambulatory cardiovascular care 2013 to 2017: JACC review topic of the week. J Am Coll Cardiol. 2020;75(1):93–112. 10.1016/j.jacc.2019.11.011. [DOI] [PubMed] [Google Scholar]
- 11.Hart RG, Pearce LA, Aguilar MI. Meta-analysis: antithrombotic therapy to prevent stroke in patients who have nonvalvular atrial fibrillation. Ann Intern Med. 2007;146(12):857–67. 10.7326/0003-4819-146-12-200706190-00007. [DOI] [PubMed] [Google Scholar]
- 12.Multiple Cause of Death Data on CDC WONDER. Accessed: Mar. 04,2025. Available: https://wonder.cdc.gov/mcd.html
- 13.Ingram DD, Franco SJ. 2013 NCHS Urban-Rural Classification Scheme for Counties. Vital Health Stat 2. 2014;166:1-73. https://pubmed.ncbi.nlm.nih.gov/24776070/. [PubMed]
- 14.Klein RJ, Schoenborn CA. Age adjustment using the 2000 projected U.S. Population. Healthy People 2010 Stat Notes. 2001;201–10. 10.1037/e583772012-001. [PubMed]
- 15.Joinpoint Regression Program. Accessed: Mar. 04, 2025. Available: https://surveillance.cancer.gov/joinpoint/
- 16.Kim HJ, Fay MP, Yu B, et al. Comparability of segmented line regression models. Biometrics. 2004;60(4):1005–14. 10.1111/j.0006-341X.2004.00256.x. [DOI] [PubMed] [Google Scholar]
- 17.Bureau UC. U.S. Older Population Grew From 2010 to 2020 at Fastest Rate Since 1880 to 1890, Census.gov, accessed November 6, 2024, https://www.census.gov/library/stories/2023/05/2020-census-united-states-older-population-grew.html
- 18.Ko D, Chung MK, Evans PT, Benjamin EJ, Helm RH. Atrial fibrillation: A review. JAMA. 2025;333(4):329–42. 10.1001/jama.2024.22451. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Lip GH, Lane DA. Stroke prevention in atrial fibrillation: a systematic review. JAMA. 2015;313(19):1950–62. 10.1001/jama.2015.4369. [DOI] [PubMed] [Google Scholar]
- 20.Camm AJ, Lip GY, De Caterina R, ESC Committee for Practice Guidelines-CPG. 2012 focused update of the ESC guidelines for the management of atrial fibrillation: an update of the 2010 ESC guidelines for the management of atrial fibrillation–developed with the special contribution of the European heart rhythm association. Europace. 2012;14(10):1385–413. 10.1093/europace/eus305. Document Reviewers. [DOI] [PubMed] [Google Scholar]
- 21.Sam C, Massaro JM, D’Agostino RB, et al. Framingham heart Study. Warfarin and aspirin use and the predictors of major bleeding complications in atrial fibrillation (the Framingham heart Study). Am J Cardiol. 2004;94(7):947–51. 10.1016/j.amjcard.2004.06.038. [DOI] [PubMed] [Google Scholar]
- 22.Hansen ML, Sørensen R, Clausen MT, et al. Risk of bleeding with single, dual, or triple therapy with warfarin, aspirin, and clopidogrel in patients with atrial fibrillation. Arch Intern Med. 2010;170(16):1433–41. 10.1001/archinternmed.2010.271. [DOI] [PubMed] [Google Scholar]
- 23.Choudari CP, Palmer KR. Acute Gastrointestinal haemorrhage in patients treated with anticoagulant drugs. Gut. 1995;36(4):483–4844. 10.1136/gut.36.4.483. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Man-Son-Hing M, Laupacis A. Balancing the risks of stroke and upper Gastrointestinal tract bleeding in older patients with atrial fibrillation. Arch Intern Med. 2002;162(5):541–50. 10.1001/archinte.162.5.541. [DOI] [PubMed] [Google Scholar]
- 25.Hearnshaw SA, Logan RF, Lowe D, et al. Acute upper Gastrointestinal bleeding in the UK: patient characteristics, diagnoses and outcomes in the 2007 UK audit. Gut. 2011;60(10):1327–35. 10.1136/gut.2010.228437. [DOI] [PubMed] [Google Scholar]
- 26.Lau JY, Barkun A, Fan DM, et al. Challenges in the management of acute peptic ulcer bleeding. Lancet. 2013;381(9882):2033–43. 10.1016/S0140-6736(13)60596-6. [DOI] [PubMed] [Google Scholar]
- 27.Staerk L, Lip GY, Olesen JB, et al. Stroke and recurrent haemorrhage associated with antithrombotic treatment after Gastrointestinal bleeding in patients with atrial fibrillation: nationwide cohort study. BMJ. 2015;351:h5876. 10.1136/bmj.h5876. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Nwankwo OT, Eun Y, et al. Age and Racial disparities in Gastrointestinal bleeding incidence among atrial fibrillation patients on direct oral anticoagulants: A retrospective cohort study. Circulation. 2023;148(1):A17633doi. 10.1161/circ.148.suppl_1.17633. [Google Scholar]
- 29.Lip GY, Rushton-Smith SK, Goldhaber SZ, et al. Does sex affect anticoagulant use for stroke prevention in nonvalvular atrial fibrillation? The prospective global anticoagulant registry in the FIELD-Atrial fibrillation. Circ Cardiovasc Qual Outcomes. 2015;8(2 Suppl 1):S12–20. 10.1161/CIRCOUTCOMES.114.001556. [DOI] [PubMed] [Google Scholar]
- 30.Bhave PD, Lu X, Girotra S, et al. Race- and sex-related differences in care for patients newly diagnosed with atrial fibrillation. Heart Rhythm. 2015;12(7):1406–12. 10.1016/j.hrthm.2015.03.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Avgil Tsadok M, Jackevicius CA, Rahme E, et al. Sex differences in Dabigatran Use, Safety, and effectiveness in a Population-Based cohort of patients with atrial fibrillation. Circ Cardiovasc Qual Outcomes. 2015;8(6):593–9. 10.1161/CIRCOUTCOMES.114.001398. [DOI] [PubMed] [Google Scholar]
- 32.Yong CM, Tremmel JA, Lansberg MG, et al. Sex differences in oral anticoagulation and outcomes of stroke and intracranial bleeding in newly diagnosed atrial fibrillation. J Am Heart Assoc. 2020;9(10):e015689. 10.1161/JAHA.120.015689. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Thompson LE, Maddox TM, Lei L, et al. Sex differences in the use of oral anticoagulants for atrial fibrillation: A report from the National cardiovascular data registry (NCDR®) PINNACLE registry. J Am Heart Assoc. 2017;6(7):e005801. 10.1161/JAHA.117.005801. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Ye X, Wu Q, Lv Q, et al. Smoking, alcohol Consumption, and atrial fibrillation: Mendelian randomization study. Cardiovasc Toxicol. 2025;25(3):341–53. 10.1007/s12012-025-09964-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Oraii A, Conen D, Johnson LS, et al. Alcohol intake and cardiovascular outcomes in patients with atrial fibrillation: RE-LY AF registry analysis. Ann Noninvasive Electrocardiol. 2025;30(4):e70096. 10.1111/anec.70096. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Fama R, Le Berre AP, Sullivan EV. Alcohol’s unique effects on cognition in women: A 2020 (Re)view to envision future research and treatment. Alcohol Res. 2020;40(2):03. 10.35946/arcr.v40.2.03. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Lipworth L, Okafor H, Mumma MT, et al. Race-specific impact of atrial fibrillation risk factors in Blacks and Whites in the Southern community cohort study. Am J Cardiol. 2012;110(11):1637–42. 10.1016/j.amjcard.2012.07.032. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Marcus GM, Alonso A, Peralta CAT, et al. European ancestry as a risk factor for atrial fibrillation in African Americans. Circulation. 2010;122(20):2009–15. 10.1161/CIRCULATIONAHA.110.958306. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Stamos TD, Darbar D. The double paradox of atrial fibrillation in black individuals. JAMA Cardiol. 2016;1(4):377–9. 10.1001/jamacardio.2016.1259. [DOI] [PubMed] [Google Scholar]
- 40.Psaty BM, Manolio TA, Kuller LH, et al. Incidence of and risk factors for atrial fibrillation in older adults. Circulation. 1997;96(7):2455–61. 10.1161/01.cir.96.7.2455. [DOI] [PubMed] [Google Scholar]
- 41.Soliman EZ, Prineas RJ. The paradox of atrial fibrillation in African Americans. J Electrocardiol. 2014;47(6):804–8. 10.1016/j.jelectrocard.2014.07.010. [DOI] [PubMed] [Google Scholar]
- 42.van Ryn M, Fu SS. Paved with good intentions: do public health and human service providers contribute to racial/ethnic disparities in health? Am J Public Health. 2003;93(2):248–55. 10.2105/ajph.93.2.248. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Rigutini AG, Bucci T, Rossi M et al. Clinical complexity in patients with atrial fibrillation: exploring differential risk profiles from European and Asian cohorts. Europace 2025 Sep 19:euaf229. 10.1093/europace/euaf229 [DOI] [PMC free article] [PubMed]
- 44.Gong G, Phillips SG, Hudson C, et al. Higher US rural mortality rates linked to socioeconomic Status, physician Shortages, and lack of health insurance. Health Aff (Millwood). 2019;38(12):2003–10. 10.1377/hlthaff.2019.00722. [DOI] [PubMed] [Google Scholar]
- 45.Lip GY, Kamath S, Jafri M et al. Ethnic differences in patient perceptions of atrial fibrillation and anticoagulation therapy: the West Birmingham Atrial Fibrillation Project. Stroke. 2002;33(1):238–242. 10.1161/hs0102.101817. PMID: 11779916. [DOI] [PubMed]
- 46.Mann HK, Streiff M, Schultz KC, et al. Rurality and atrial fibrillation: patient perceptions of barriers and facilitators to care. J Am Heart Assoc. 2023;12(21):e031152. 10.1161/JAHA.123.031152. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The datasets generated and/or analyzed during the current study are available on the CDC Wonder Database, https://wonder.cdc.gov/.