Is there a clinically significant seasonal component to hospital admissions for atrial fibrillation?

Abstract

Abstract Background Atrial fibrillation is a common cardiac dysrhythmia, particularly in the elderly. Recent studies have indicated a statistically significant seasonal component to atrial fibrillation hospitalizations. Methods We conducted a retrospective population cohort study using time series analysis to evaluate seasonal patterns of atrial fibrillation hospitalizations for the province of Ontario for the years 1988 to 2001. Five different series methods were used to analyze the data, including spectral analysis, X11, R-Squared, autocorrelation function and monthly aggregation. Results This study found evidence of weak seasonality, most apparent at aggregate levels including both ages and sexes. There was dramatic increase in hospitalizations for atrial fibrillation over the years studied and an age dependent increase in rates per 100,000. Overall, the magnitude of seasonal difference between peak and trough months is in the order of 1.4 admissions per 100,000 population. The peaks for hospitalizations were predominantly in April, and the troughs in August. Conclusions Our study confirms statistical evidence of seasonality for atrial fibrillation hospitalizations. This effect is small in absolute terms and likely not significant for policy or etiological research purposes.

Keywords

• Atrial fibrillation
• seasons
• stochastic processes
• statistics
• time factors