Incidence of stroke

Method

Incident stroke events ascertained by each cohort's or trial's stated adjudication criteria, pooled as a relative risk or hazard ratio. Ischaemic and haemorrhagic subtypes are recorded on the edge wherever the source separates them.

The method is part of the endpoint identity. Studies measuring the same quantity by a different method are held as separate endpoints.

Notes

Transient ischaemic attack is pooled with stroke in PMID 22493407 and separable elsewhere. TIA is non-infarcting by definition and its ascertainment is far softer, so a composite of the two is dominated by ascertainment behaviour rather than by disease. Also distinct from symptomatic intracranial haemorrhage, which in these trials is a treatment HARM occurring inside stroke care rather than an incident disease event — pooling the two lets a real thrombolytic benefit cancel against a real bleeding harm and report nothing. Distinct again from incidence of cardiovascular disease (composite), which contains stroke among other events, and from cardiovascular mortality, which is fatal outcome rather than incidence.