Objective: To study the relationship of temperature and stroke.
Methods: When long time trend, season and relative humidity were under control, the relationship between temperature and incidence rate of acute stroke was explored by generalized additive models (GAMs) for the whole population and stratified by age, sex, disease history and stroke type.
Results: Cold appeared to be a risk factor to stroke incidence. The incidence of stroke was higher than the baseline data when the temperature was lower than 8.5 degrees C but the incidence increased along with the decrease of temperature. Cold was more likely to affect females as well as patients older than 65-year-old with first-ever episode. However, heat was more likely to influence females and those over 65-year-old patients having had suffered stroke before. Cold endangered both haemorrhagic and ischaemic stroke while heat only affected ischaemic stroke.
Conclusion: The interaction between average weekly temperature and stroke was related to the types of stroke. Age, gender and history of stroke in the past had different effects on the relationship between weekly average temperature and stroke incidence.