Abstract 129: Enrolling Spanish-Speaking Subjects into Prehospital Research: The FAST-MAG Spanish Line Experience

• DOI: 10.1161/str.44.suppl_1.a129
• Publication Date: 2022-03-20T05:12:15Z
• Authors: Nerses Sanossian; Sidney Starkman; Latisha K Ali; Lucas Restrepo; Miguel Valdes-Sue...; David S Liebeskind; Fiona Chatfield; T. Keta Hodgson; Samantha Cherin; Jeffrey L Saver

Background: Clinical trials should enroll diverse subjects who are representative of the general population. Enrollment in other languages requires substantial effort, including language-specific consent forms and physician-investigators fluent in language of consent. We describe the yield of a dedicated 24-hour Spanish prehospital research enrollment line.

Methods: FAST-MAG is a phase 3, NIH-funded, randomized, placebo-control study of prehospital magnesium sulfate within 2 hours of stroke onset. The study is being conducted in Los Angeles and Orange Counties, home to 12.9 million people of whom just under half (44.8%) identify as Hispanic or Latino. Each of the 353 study ambulances carries: 1) 8 consent forms, 4 in English and 4 in Spanish, for the vehicle’s 4 most common destination hospitals; and 2) a cell phone with a quick-dial button for English and Spanish-speaking physician-investigator.

Results: Data was available on 854 consecutive enrolments of whom 141 (17%) were via the Spanish and 713 (83%) via the English line. Patients enrolled via the Spanish line were younger (65 vs 70 years, p<0.001), less often women (33% vs. 43%, p=0.040), more often self-identifying as Latino (98% vs 16%, p<0.001), and more likely to be of White race vs other (98% vs 72%, p,0.001). Spanish-line enrollments trended to more severe stroke deficits (NIHSS 13.5 vs 11.7, p=0.08), were far more likely to have intracerebral hemorrhage on initial imaging (35% vs. 20%, p<0.001) and to have higher pre-hospital blood pressure (166/94 vs. 159/90, p=0.012). Spanish-line enrolments were more likely to be diabetic (36% vs. 23%, p=0.003), and less likely to have atrial fibrillation (16% vs 27%, p=0.012). Times from paramedic arrival to start of study infusion were only slightly longer in Spanish-line patients (24.6 vs 22.9 minutes, p=0.045).

Conclusions: The FAST-MAG Spanish language enrolling line has accounted for 17% of enrollments and has enrolled a group of subjects distinct from the standard, English language enrolling line. The Spanish language enrollment process has doubled the number of Hispanic/Latino patients enrolled in the trial, allowing the study to more closely reflect the population of Southern California.