The starting hypothesis is that a multilevel educational campaign, specifically developed for the local community, can increase public stroke awareness and reduce pre-hospital delay.

The effectiveness of such intervention will be evaluated according to a cluster randomized, stepped wedged design. The clusters are the four communities of the Area Vasta Emilia Nord, AVEN (Parma, Piacenza, Modena e Reggio Emilia). As analysis Units, we will consider the patients consecutively admitted to the six participating hospitals throughout the study period, for suspected stroke or transitory ischemic attack (TIA).

After a baseline 3 month period, the educational campaign will be sequentially launched in the four communities over four 3 month periods, according to a computer-generated list. The comparison will be the"usual care".

Primary outcome measures: The proportion of patients arriving at the Emergency Department (ED) with suspected stroke or TIA within two hours.

Secondary outcome measures: the proportion of patients with confirmed stroke or TIA diagnosis; the proportion of ischemic stroke patients evaluated for recominanat Tissue Plasminogen Activator (rTPA) therapy; the proportion of patients treated with rTPA; time interval between arrival at the ED and CT scan; for patients treated with rTPA, time interval between arrival at the ED and therapy initiation (door to needle time); death and disability (assessed as modified Rankin Score 3-5) at 1 and 3 month from stroke onset.

Inclusion Criteria:

• Patients consecutively admitted to the six participating hospitals for suspected stroke or transitory ischemic attack (TIA), defined as abrupt onset of focal or generalized neurologic deficit of vascular origin.

Exclusion Criteria:

• No information available on the time of stroke onset.
• No informed consent from patient or caregiver