Abstract TP369: Home Address is a Reasonable Proxy for Location of Stroke Symptom Onset in Medicare Aged Individuals

Abstract

Background: With increasing focus on optimization of the prehospital transport of acute stroke patients, location of emergency medical services (EMS) pickup is a critical variable. However administrative data do not routinely include this. It is not clear whether we can impute home address as the location of pickup. Our objective was to determine the proportion of EMS transports originating at patients’ homes, and to identify characteristics associated with home pickup. Methods: We used our institutional Get with the Guidelines-Stroke data to identify acute stroke patients presenting 12/01/16-02/20/18. Consecutive charts and EMS run sheets were reviewed for date, time and address of EMS pick-up and patient home address. We categorized pickup as workhours (Monday-Friday, 09:00-17:00), versus not. We used logistic regression to examine the relationship between age (stratified at 65 years), and workhours on home pickup, including an interaction term for age and workhours. Results: Of the 740 charts reviewed, 457 patients arrived by EMS (61.8%). Among EMS-transported patients, 313 (68.5%) were picked up at home, 298 (65.3%) were > 65 years, and 142 were during workhours (31.1%). In bivariate analyses, home pickup was more common among patients > 65 (74.5% vs 57.2%, p<0.001), and was not different during workhours (66.9% vs 69.2%, p=0.62). Relative to other EMS-transported patients, patients with home pickup had similar tPA treatment rates (17% vs 18%, p=0.08) and discharge to home (28% vs 31%, p=0.54). In logistic regression analysis, home pickup was more likely among patients > 65 years (OR 2.23, 95% CI 1.48-3.37), and not impacted by workhours (OR 0.82, 95% CI 0.53-1.26). The relationship between age and home pickup was not mediated by workhours, so for simplicity we report the model results without the interaction term. Conclusion: Among stroke patients transported by EMS, home pickup is more common among patients 65 years and older. Imputing home address would be accurate for 74.5% of Medicare-aged patients. If these findings are validated in larger datasets, these findings suggest that home address may be a reasonable proxy for onset location among older stroke patients.