Improving Language Access for Patients with Limited English Proficiency (LEP) During Emergency Room Care

ABSTRACT

Purpose/Objectives: Patients with Limited English proficiency (LEP) are at greater risk for hospital admissions and representations to the emergency department (ED) in comparison to patients with English proficiency (EP). Improved communication with language services has been shown to decrease these health disparities. In our free-standing Children's Hospital yearly ED representation rates for patients with LEP (4.6%) are over 20% higher than for our patients with EP (3.7%). We aimed to decrease the 7-day representation rate for patients with LEP in the emergency department by 20% over 18 months using the Plan Do Study Act (PDSA) method for quality improvement. Design/Methods: Our first cycle was the deployment of video remote interpretation (VRI). Second and third cycles included focused teaching for Faculty, House staff and staff on VRI use. PDSA cycles were interrupted with the unexpected event of Covid-19 and issues obtaining Cerner data. We evaluated outcomes quantitatively based on data from our language service provider and electronic medical record (EMR). We also conducted an electronic survey of Faculty and Staff to evaluate quantitatively over time VRI reception and frequency of obtaining preferred language. Results: LEP representation rates initially increased followed by a decreasing trend with the Covid19 pandemic and opening of an ED observation unit but this trend was not sustained and average representation rates after PDSA cycles increased to 6.1%. EP representation rates overall remained unchanged at 3.7%. Utilization of interpretation carts increased from 0 to >250 total calls made per month over the course of 18 months. The monthly average number of identified patients with LEP decreased after changes were made to the recording system in the EMR. Monthly rates of interpretation carts used for un-identified patients with LEP were found to range from 27% to 78%, signifying significant under identification in the EMR. 75% of ED Faculty and staff reported that they only ask preferred language or offer interpretation services for medical discussions infrequently when patients appear to speak English well but is clearly not their native language. Total ED visit numbers decreased after March of 2020 to 30-50% of expected due to Covid-19 pandemic. Conclusion/Discussion: ED Patients with LEP had an unexpectedly high rate of un-identification in addition to ED faculty and staff reporting not asking preferred language if the patient seems to speak English well. This has likely led to an underestimation of our LEP representation rate and lack of representative change during our PDSA cycles. Without identifying this vulnerable population, we will always be lacking in providing care in the preferred language which is critical for improving outcomes for our patients with LEP. Our next steps are to develop interventions to increase the identification of patients with LEP to provide safe and comprehensive care for this vulnerable population.