The Effect of Being Found with Stroke Symptoms on Predictors of Hospital Arrival.

Background: Studies examining predictors of delay in hospital arrival after stroke symptom onset have not accounted for patients who are found with their symptoms and cannot seek help independently. Our objective is to show that inclusion of patients who are "found down" in studies of prehospital delay biases the estimated association of sociodemographic and clinical variables with time of hospital arrival.

Methods: We performed a retrospective analysis of sociodemographic and clinical characteristics of patients with acute ischemic stroke presenting to a tertiary care hospital in the Bronx, New York.

Results: Compared with all other patients with acute ischemic stroke (N = 1784), patients who were found down (N = 120) were more likely to be older (75  ± 13 years versus 68 ± 14 years, P < .0001), female (68% versus 53%, P = .003), Caucasian race (P < .001), have higher socioeconomic status (P = .001), more severe stroke deficits (P < .0001), use emergency medical services (P < .001), and arrive to the hospital more than 3 hours after symptom onset (P < .001). Inclusion of patients who were found down in a model predicting delay in hospital arrival decreased the strength of the association between the predictors and the outcome.

Conclusions: Being found with stroke symptoms confounds the association of sociodemographic and clinical variables with time of hospital arrival. Studies of predictors of prehospital delay should therefore exclude patients who are found down.