Background: A winter excess of ischaemic stroke has been found in mortality and hospital-based studies. It is often assumed that this is due to seasonal variation in stroke incidence and several pathophysiological explanations have been proposed. We studied the incidence of stroke in relation to season and outside temperature.
Methods: The data came from a community-based study of first ever in a lifetime stroke in a defined population of about 105 000. 675 such strokes were registered over four years and the month of inset was analysed separately for cerebral infarction, primary intracerebral haemorrhage, and subarachnoid haemorrhage.
Findings: There was no significant seasonal variation. The incidence of primary intracerebral haemorrhage was increased at low temperatures, but there was no significant relation between the incidence of ischaemic stroke or subarachnoid haemorrhage and temperature.
Interpretation: The widely reported winter excess of ischaemic strokes may be an artifact due to referral bias in hospital-based studies and increased case fatality during the winter in mortality studies.