(1480) Exploring Implementation of the A-F Bundle for Patients Who Speak a Language Other Than English

Introduction: Patients who speak a language other than English (LOE) are often excluded from ICU research, resulting in interventions that are not adapted for the LOE population. As a result, LOE patients may receive fewer A-F bundle interventions, resulting in worse critical care outcomes.

Methods: This is a secondary data analysis of the ICU Liberation Dataset, collected between January 2015-March 2017. The primary exposure is whether an interpreter was required. The primary outcomes are adherence to A-F bundle elements and total bundle. Adherence was calculated as a proportion: dividing number of received interventions by eligible opportunities for each element, as well as the total bundle. Statistical analyses were conducted using odds ratios (OR) and a multilevel random effects model adjusting for mechanical ventilation status.

Results: Comparing the proportional adherence to bundle elements provided by care teams for LOE vs English-speaking patients, LOE patients received significantly higher adherence to A - Assessing pain (58% vs 47%; OR 2.16, 95% CI 1.88-2.49), E - early mobility (66% vs 63%; OR 1.46, 95% CI 1.20-1.78), and F - family engagement (70% vs 62%; OR 1.56, 95% CI 1.27-1.93). Comparing the proportional adherence to bundle elements provided by care teams for LOE vs English-speaking patients, LOE patients received significantly lower adherence to C - choice of sedation (38% vs 45%; OR 0.51, 95% CI 0.44-0.60) and D - delirium assessment (33% vs 50%; OR 0.29, 95% CI 0.25-0.34). There were no differences in the care team’s proportional adherence to B - spontaneous awakening trials (48% vs 50%; OR 0.87, 95% CI 0.67-1.14) and spontaneous breathing trials (64% vs 66%; OR 0.92, 95% CI 0.72-1.16), or total bundle adherence (53% vs 54%; OR: 0.99, 95% CI 0.88-1.12).

Conclusions: While adherence to the total A-F bundle is similar comparing LOE patients to English-speaking patients, there are distinct differences in adherence to individual bundle elements. Significantly lower adherence to choice of sedation and delirium assessments highlights inequitable care in ICU delirium prevention and management for LOE patients. Further research into the causes of these inequities and development of interventions to address them is warranted.