A Pilot Comparison of the Retention Rates of FAST and BEFAST Stroke Warning-Sign Mnemonics

Shujing GuoJeffrey L Saver^*

• University of California, Los Angeles, Los Angeles, United States

Background: Several pre-hospital delays prevent stroke patients from arriving within the optimal 4.5-hour therapeutic window, including failure to recognize stroke symptoms and lack of urgency in perceiving them as requiring immediate medical attention. Community stroke education is critical in reducing pre-hospital delays. The Face-Arm-Speech-Time (FAST) mnemonic has been the AHA/ASA's official stroke warning-signs public education message since 2013. Recently, a more inclusive but potentially harder to remember mnemonic, adding Balance-Eyes (BEFAST), has been proposed to increase the sensitivity of warning-sign messaging in detecting strokes. We undertook a pilot randomized trial to assess the feasibility of, refine methods for, and provide information regarding required sample size for a larger, pivotal randomized trial comparing the retention rate of FAST versus BEFAST. Methods: This study randomized adult participants without history of stroke to a comprehensive community stroke education intervention that included the definition of a stroke, its risk factors, outcomes, and either FAST or BEFAST mnemonics and assessed retention after 14-21 days. The primary endpoint was retention of mnemonic letter and warning-sign knowledge. The secondary outcome was improvement at 14-21 days in participant knowledge of stroke definition, risk factors, and outcomes. Results: Among the 50 adult participants, mean age was 37.2 (±14.8) and 30 (60%) participants were women. Study procedures were completed in all participants. At initiation, 23 (46%) participants knew the definition of stroke and could describe a mean 1.6 signs, 0.98 risk factors, and 0.74 outcomes. Full mnemonic recall rates at 14-21 days were 68% for FAST and 56% for the BEFAST cohorts (P=0.39). Secondary paired analysis found significant improvements in participant knowledge of stroke signs (to 2.72), risk factors (to 3.30), and outcomes (to 2.24) with P=0.007 or smaller. Discussion: This pilot trial demonstrated the feasibility of performing a large, pivotal trial and indicates a sample size of 512 participants is needed to provide sufficient power. The preliminary pilot data suggest a generally higher recall performance in the FAST group. Regardless of the mnemonic taught, participants had significant improvements in their knowledge of stroke signs, risk factors, and outcomes.