Emergency department observational unit usage by patients with cancer: A retrospective review

Background

Emergency department observation units (EDOUs) are effective and increasingly utilized in the delivery of short-term care. These units decrease length of stay and cost while increasing patient satisfaction and safety. Currently, EDOUs are underutilized for patients with cancer. We aimed to characterize the care of patients with cancer in an academic type 1 EDOU.

Methods

Retrospective review of EDOU quality data at an academic medical center with a Comprehensive Cancer Center (CCC). Eligible encounters included patients ≥ 18 cared for in the EDOU between July 1, 2019, and December 31, 2022. Patients with cancer were identified using an oncology triage screen during ED intake defined as a self-report of an active cancer or receiving care at the associated CCC. Descriptive statistics were employed to describe demographics, chief complaint, and assigned EDOU protocol. Pearson's chi-squared test was performed to determine statistically significant differences among patients with and without cancer. Due to dataset limitations, it was impossible to remove EDOU subjects with cancer treated for noncancer-related reasons.

Results

From 15,089 eligible EDOU encounters, 1,711 (11.3%) involved patients with cancer – 348, 548, 444, and 371 visits in 2019, 2020, 2021, and 2022 respectively. In the cohort with cancer, most patients were White (1,353, 79.1%) with a slight female predominance (905, 52.9%) noted. Furthermore, unspecified (148, 8.6%), chest pain (127, 7.4%), and abdominal pain (93, 5.4%) were the top chief complaints. In patients with cancer, 1,234 (72.1%) encounters resulted in discharge, significantly lower than that observed in patients without cancer (10,408, 77.8%; p<0.001). In both groups, the general observation protocol was the most utilized order set. However, further analysis revealed a significantly increased use of the general observation protocol in patients with cancer compared to noncancer patients (25.4% and 14.0%, respectively; p<0.001).

Conclusions

Compared to patients without cancer, patients with cancer experience a significantly higher use of the general observation protocol suggesting that further optimization and development of novel cancer-specific EDOU protocols is warranted.