Abstract Body: Introduction: Among Black Americans, the time between stroke symptom onset and hospital arrival is longer than among White Americans. We aimed to adapt and test a stroke preparedness intervention for emergency department patients in Flint Michigan, a majority Black American community.

Methods: In this randomized controlled trial (2/2022–8/2023), eligible participants were ≥18 years old, English-speaking, and lacked severe pain or any other medical condition that might distract from participation. Participants were administered the video-STAT, which queries response to 4 stroke and 4 non-stroke video vignettes (scored 0-8 for correctly designating intent to call 911 for stroke) and tests recognition of stroke vignettes (scored 0-4). Only responses to stroke video vignettes are scored. Participants were randomized to the intervention condition (a pamphlet and video teaching stroke symptoms and the importance of calling 911) or the control condition (a pamphlet about the American Heart Association’s Life’s Simple 7). The primary outcome was the intent to call 911 score immediately post-exposure (secondary endpoint) and at 1 month (primary endpoint). The secondary outcome was the stroke recognition score. Linear mixed models adjusting for pre-exposure intent to call 911 or stroke recognition score, and demographic and clinical covariates evaluated the association between treatment group and the outcomes.

Results: A total of 330 people participated (n=171 control group; n=159 intervention group; mean age= 45, 61% female, 49% Black participants). Pre-exposure, the control and intervention groups did not differ in intent to call 911 (mean ± standard deviation (SD) score of 5.3 ± 2.2 and 5.4 ± 2.1 in the control and intervention groups) or recognition of stroke vignette scores (mean ± SD score of 2.9 ± 1.0 and 2.8 ± 1.1 in the control and intervention groups). Post-exposure, the intervention group had a higher intent to call 911 score than the control group (adjusted mean between-group difference, 1.14 [95% CI, 0.78 – 1.50] immediately post-exposure; 1.12 [95% CI, 0.64 -1.59] at 1-month post-exposure). The intervention group also correctly recognized more stroke vignettes than the control group (adjusted mean difference, 0.40 [95% CI, 0.23-0.56] immediately post-exposure; 0.51 [95% CI, 0.30-0.72] at 1-month post-exposure).

Conclusions: The ED-based educational intervention resulted in greater intent to call 911 for stroke and greater stroke symptom recognition.