Global, regional, and national burden of atrial fibrillation and its association with alcohol consumption: age, sex, regional, and SDI differences—an analysis of the Global Burden of Disease Study 2021

Figures

• 

  Figure 1. Global trends in age-standardized atrial fibrillation (AF)-related mortality by sex, 1990–2021. Data source: Global Burden of Disease (GBD) Study 2021. Case definition: AF-related deaths coded under ICD-10 I48.0–I48.9, encompassing primary and secondary AF as well as complications including stroke and heart failure. Metric: Age-standardized mortality rate per 100,000 population, calculated using the GBD standard reference population. Model: Mortality estimates were derived from the Cause of Death Ensemble model (CODEm), which integrates national vital registration systems, verbal autopsy, and mortality surveillance data. Adjustment: Estimates are standardized for age and sex; comorbidities (e.g., hypertension, diabetes) and healthcare access were incorporated as covariates within the GBD modeling framework. Uncertainty: Shaded areas represent 95% uncertainty intervals (UI).

• 

  Figure 2. Age- and sex-specific trends in global AF-related mortality, 1990–2021. Data source: GBD Study 2021. Case definition: ICD-10 I48.0–I48.9 (primary and secondary AF, including stroke and heart failure complications). Metric: Age-specific mortality rate per 100,000 population. Age groups: 30–44, 45–64, 65–84, 85–94, and 95+ years. Model: CODEm, integrating multiple mortality data sources with covariate-based ensemble estimation. Adjustment: Age- and sex-specific estimates; comorbidities and healthcare access adjusted within the GBD framework. Uncertainty: Shaded ribbons denote 95% UI. Key observations: Female mortality increased notably in the 90–94 and 95+ age groups, whereas male mortality surged in the 30–44 and 85+ age groups.

• 

  Figure 3. National trends in age-standardized AF-related mortality for selected countries and territories, 1990–2021. Data source: GBD Study 2021. Case definition: ICD-10 I48.0–I48.9. Metric: Age-standardized mortality rate per 100,000 population. Model: CODEm ensemble estimates. Adjustment: Age- and sex-standardized; comorbidities and healthcare access covariates included in GBD models. Uncertainty: Error bands represent 95% UI. Illustrative examples: Greenland (male mortality declined from 15.646 to 13.238 per 100,000), Sweden (male mortality increased from 5.054 to 10.436 per 100,000), and China (male mortality decreased from 7.251 to 5.220 per 100,000; female mortality decreased from 6.874 to 5.325 per 100,000).

• 

  Figure 4. Alcohol-attributable AF-related mortality and disability-adjusted life years (DALYs) by wHO region, 1990-2021. Data source: GBD Study 2021. Case definition: ICD-10 I48.0-I48.9. Metric: Age-standardized alcohol-attributable mortality rate and age-standardized alcohol-attributable DALYs per 100,000 population. Alcohol-attributable fraction methodology: Based on population-level alcohol exposure, relative risks derived from systematic reviews, and the theoretical minimum risk exposure level. Model: CODEm for underlying mortality; GBD comparative risk assessment framework for alcohol-attributable burden estimation. Regions: Europe, the Americas, the Western Pacific, Southeast Asia, Africa, and the Eastern Mediterranean (WHO classification). Adjustment: Age- and sex-standardized; comorbidities (hypertension, diabetes, obesity) and healthcare access incorporated as covariates. Uncertainty: Shaded areas represent 95% UI. Design note: Figure has been split into two panels to improve readability.

• 

  Figure 5. Age-specific alcohol-attributable AF mortality in selected WHO regions by sex, 1990-2021. Data source: GBD Study 2021. Case definition: ICD-10 I48.0-I48.9. Metric: Age-specific alcohol-attributable mortality rate per 100,000 population. Regions shown: Europe, the Americas, and the Western Pacific (selected to highlight major effects). Alcohol-attributable fraction: Estimated using GBD comparative risk assessment methods based on regional drinking patterns, alcohol use disorder prevalence, and AF-specific relative risks. Model: CODEm ensemble with integrated risk factor modeling. Adjustment: Age- and sex-specific; comorbidities and healthcare access covariates included. Uncertainty: Shaded ribbons indicate 95% UI. Design note: Figure simplified by sex-stratified panels and focused on regions with the highest alcohol-related AF burden.

• 

  Figure 6. Socio-Demographic Index (SDI)-specific trends in alcohol-attributable AF mortality, 1990–2021. Data source: GBD Study 2021. Case definition: ICD-10 I48.0–I48.9. Metric: Age-standardized alcohol-attributable mortality rate per 100,000 population. SDI classification: Countries stratified into five quintiles (low, low-middle, middle, high-middle, and high SDI) based on income per capita, educational attainment, and fertility rate. Alcohol-attributable fraction: Population-level alcohol exposure and AF-specific relative risks applied within the GBD comparative risk assessment framework. Model: CODEm ensemble estimates. Adjustment: Age- and sex-standardized; comorbidities and healthcare access adjusted within GBD models. Uncertainty: Shaded areas represent 95% UI.

• 

  Figure 7. Age- and sex-specific alcohol-attributable AF mortality in high and high-middle SDI regions, 1990–2021. Data source: GBD Study 2021. Case definition: ICD-10 I48.0–I48.9. Metric: Age-specific alcohol-attributable mortality rate per 100,000 population. SDI groups: High SDI and high-middle SDI quintiles. Sex/age stratification: Data presented by sex (male/female) and age group (80–89, 85–94, 90–94, 95+ years), focusing on older adult populations where alcohol-attributable mortality is most pronounced. Alcohol-attributable fraction: GBD comparative risk assessment using regional alcohol exposure and AF-specific relative risks. Model: CODEm ensemble modeling. Adjustment: Age- and sex-specific estimates with comorbidity and healthcare access covariates. Uncertainty: Shaded ribbons denote 95% UI. Design note: Presented as a focused panel to highlight the major effect of alcohol in high-SDI settings; detailed multi-SDI age-stratified data are provided in Supplementary Figure X.