Social Emergency Medicine
Background: The underuse of interpreters for limited English proficiency (LEP) patient encounters is pervasive, especially in the emergency department (ED). The objective of this pilot study was to measure the impact of multifaceted, behavior-theory based implementation strategies to improve the use and documentation of interpreters by ED providers.
Methods: Pre- and post- intervention evaluation of the unmet need for language assistance (LA) in an urban, academic ED where most patients have LEP and remote LA is available 24/7. From 2016 to 2018 we used implementation science strategies to influence ED provider decision-making to improve interpreter use. Rooted in the Behavior Change Wheel framework, strategies included: education on the right to LA, hands-on training with video interpreter machines, technology-based facilitators (interpreter icon on tracking board, more available phones, more dedicated internet bandwidth), local champions (interpreter, EM faculty, and administrators) and environmental cues (screen savers and posters). We sampled and reviewed all charts for a 48-hour period of ED registrations pre- and post- intervention to characterize the rate of documented interpreter use and the remaining unmet need for LA. We recorded preferred language, interpreter request, whether the provider was certified bilingual, documentation of LA use, and language concordance of discharge instructions.
Results: On pre-intervention chart review, of the 110 interpreter requests, 17 (15.5%) had either documentation of an interpreter-mediated encounter or were seen by a language concordant, certified bilingual provider (unmet need=84.5%). Post intervention, of the 159 interpreter requests, 47 (29.6%) had documentation of an interpreter-mediated encounter or were seen by a certified bilingual provider (unmet need=70.4%). Pre-post difference= +0.14 (CI=0.03-0.23). The rate of language concordant written discharge instructions for LEP patients discharged from the ED did not change. (pre= 66/95, 69.5%, post=97/134, 72.4%, difference 0.029, CI -0.08 to 0.14).
Conclusion: In this single-center pilot study of behavior-theory based implementation strategies to improve LA use, we found a statistically significant increase in the documented use of interpreters, although a large performance gap remains.