Age-dependent seasonality in the incidence of stroke: A 21-year population-based study

Abstract

Introduction: Seasonality in the incidence of stroke has been examined in numerous studies, yet data on whether it differs with age are limited. To fill this gap, we utilized a largescale dataset from Israel. Patients and methods: We retrieved data of all hospitalizations for ischemic stroke (IS), transient ischemic attack (TIA) and intra cerebral hemorrhage (ICH) from 2000 to 2020. We maintained separate datasets for IS/TIA and ICH, divided into five age groups: 18–49, 50–59, 60–69, 70–79, and 80+. We modeled the monthly incidence using a generalized additive model. The seasonal effect was defined by the rate ratio (RR) of each month compared to the annual mean. Results: The analysis included 317,586 and 23,789 events of IS/TIA and ICH respectively. We found an interaction between age and seasonality, accounting for a phase shift with age in the seasonal pattern of IS/TIA incidence. For cases under 70 years, the peak was during summertime and the RRs increased with decreasing age, reaching 1.11 (95% CI 1.09–1.13) at the youngest age group. In contrast, among the elderly, a winter peak was observed and the RRs increased with age to 1.07 (95% CI 1.06–1.08) at the oldest age group. For ICH, a winter/autumn peak was identified and the RRs increased with age to 1.20 (95% CI 1.16–1.24). Conclusions: Our finding of age-dependent seasonal patterns in the occurrence of stroke, suggests closer monitoring of cardiovascular risk factors during wintertime among elderly individuals. The mechanism governing the seasonal phase shift with age in IS/TIA incidence, requires further investigation.