271: PREDICTORS OF ICU CLINICIAN DOCUMENTATION OF DISRUPTED SLEEP IN CRITICALLY ILL MEDICAL ADULTS

Introduction: Disrupted sleep is frequent in the ICU but the factors influencing whether ICU clinicians document it remain unclear. We sought to identify patient variables associated with greater clinician disrupted sleep documentation in medical critically ill adults. Methods: Consecutive adults admitted to 1 of 7 medical ICUs ≥4 days between 1/17 and 6/21 at 2 academic centers were included. Subsequent ICU admissions were excluded. A sample of 30 ICU RNs and MDs were surveyed to identify the terms they most frequently use to document disrupted (poor) and non-disrupted (good) sleep. Pseudo-labels for both RN (from 21-8h) and MD (from 21-11h) ICU night-3 EHR notes, a night when ICU disrupted sleep prevalence is likely greater than nights-1 or 2, were created by finding disrupted sleep terms in RN/MD EHR notes. We identified (through literature review and consensus) and electronically collected 6 factors hypothesized to be associated with greater disrupted sleep documentation [>65 years, white (vs. non-white) race, male sex, pre-ICU sleep disorder, pre-ICU sleep medication use, and intubation on ICU day 1, 2 and/or 3]. We evaluated the association between factors and ICU night 3 disrupted sleep documentation using both univariate and a multiple logistic regression analyses. Results: A total of 3193 disrupted sleep notes (79.6%MD; 20.4%RN) were documented on ICU night-3 for 1982/3407(58.2%) patients. Univariate analysis revealed disrupted sleep documentation to be greater in patients with pre-ICU sleep medication use (64.3% vs. 57.4%, P=0.01) and who were white (59.1 vs. 55.4%, P=0.06), less in intubated patients (55.4% vs. 60.2%, P< 0.01), and not different in patients >65 years old (P=0.66), male (P=0.70), and who had a pre-ICU sleep disorder (P=0.90). The regression analysis revealed disrupted sleep documentation to be greater in patients with pre-ICU sleep aid medication use (aOR 1.35, 95%CI 1.08-1.70; P< 0.01) and less in patients who were intubated (aOR 0.83, 95%CI 0.72-0.95; P< 0.01). Conclusions: Pre-ICU sleep aid use and intubation are each independently associated with disrupted sleep documentation occurrence. Future prospective research focused on characterizing and comparing how ICU physicians and nurses each evaluate and document sleep quality is required.