Sex‐Based Differences in Atrial Fibrillation: Epidemiology, Prehospital Care, Hospital Management, and Outcomes

Figures

• 

  Figure 1. Age‐standardized AF event rates for women (blue) and men (red), showing higher rates overall among women, especially in major cities and inner regional areas. AF indicates atrial fibrillation; and IRR, incidence rate ratio.

• 

  Figure 2. Age‐specific AF event rates showing higher rates among men in age groups <70 years and higher rates among women in age groups ≥70 years. The right‐hand column presents IRRs and 95% CIs comparing women with men. AF indicates atrial fibrillation; and IRR, incidence rate ratio.

• 

  Figure 3. Fully adjusted associations between female sex and prehospital and hospital care measures. Relative differences (adjusted OR) are shown. Fully adjusted analyses were calculated using a multilevel logistic regression model with age, sex, comorbidities, clinical status, socioeconomic status, and remoteness included as fixed effects. For prehospital care, local government region was included as a random effect to account for clustering, whereas for hospital care, hospital facility was included as a random effect to account for clustering. Additional adjusted estimates for key prehospital outcomes are summarized in Table S3. *Hospital length of stay is continuous, hence described as a ratio of means. ED indicates emergency department; and MICA, mobile intensive care ambulance.

• 

  Figure 4. Fully adjusted associations between female sex and inpatient cardioversion and AF ablation. Relative differences (adjusted OR) are shown on the left and absolute differences (adjusted risk difference) shown on the right. Fully adjusted analyses were calculated using a multilevel logistic regression model with age, sex, comorbidities, clinical status, socioeconomic status, and remoteness included as fixed effects, with hospital facility included as a random effect to account for clustering. AF indicates atrial fibrillation.

• 

  Figure 5. Fully adjusted associations between female sex and AF reattendance and all‐cause mortality. Adjusted analyses were calculated using a multilevel parametric survival regression model based on the Weibull distribution with age, sex, comorbidities, clinical status, private hospital status, hospital capabilities, socioeconomic status, and remoteness included as fixed effects. Local government region was included as a random effect to account for clustering. Median follow‐up was quantified using the reverse Kaplan–Meier method. 22 AF indicates atrial fibrillation.