Solving the worldwide emergency department crowding problem - what can we learn from an Israeli ED?

Medical practitioner perspectives on AI in emergency triage

Introduction

A proposed Diagnostic AI System for Robot-Assisted Triage ("DAISY") is under development to support Emergency Department ("ED") triage following increasing reports of overcrowding and shortage of staff in ED care experienced within National Health Service, England ("NHS") but also globally. DAISY aims to reduce ED patient wait times and medical practitioner overload. The objective of this study was to explore NHS health practitioners' perspectives and attitudes towards the future use of AI-supported technologies in ED triage.

Methods

Between July and August 2022 a qualitative-exploratory research study was conducted to collect and capture the perceptions and attitudes of nine NHS healthcare practitioners to better understand the challenges and benefits of a DAISY deployment. The study was based on a thematic analysis of semi-structured interviews. The study involved qualitative data analysis of the interviewees' responses. Audio-recordings were transcribed verbatim, and notes included into data documents. The transcripts were coded line-by-line, and data were organised into themes and sub-themes. Both inductive and deductive approaches to thematic analysis were used to analyse such data.

Results

Based on a qualitative analysis of coded interviews with the practitioners, responses were categorised into broad main thematic-types, namely: trust; current practice; social, legal, ethical, and cultural concerns; and empathetic practice. Sub-themes were identified for each main theme. Further quantitative analyses explored the vocabulary and sentiments of the participants when talking generally about NHS ED practices compared to discussing DAISY. Limitations include a small sample size and the requirement that research participants imagine a prototype AI-supported system still under development. The expectation is that such a system would work alongside the practitioner. Findings can be generalisable to other healthcare AI-supported systems and to other domains.

Discussion

This study highlights the benefits and challenges for an AI-supported triage healthcare solution. The study shows that most NHS ED practitioners interviewed were positive about such adoption. Benefits cited were a reduction in patient wait times in the ED, assistance in the streamlining of the triage process, support in calling for appropriate diagnostics and for further patient examination, and identification of those very unwell and requiring more immediate and urgent attention. Words used to describe the system were that DAISY is a "good idea", "help", helpful, "easier", "value", and "accurate". Our study demonstrates that trust in the system is a significant driver of use and a potential barrier to adoption. Participants emphasised social, legal, ethical, and cultural considerations and barriers to DAISY adoption and the importance of empathy and non-verbal cues in patient interactions. Findings demonstrate how DAISY might support and augment human medical performance in ED care, and provide an understanding of attitudinal barriers and considerations for the development and implementation of future triage AI-supported systems.

Lessons learned from an evaluation of referrals to the emergency department

BACKGROUND

Emergency department (ED) crowding is an international phenomenon dependent on input, throughput, and output factors. This study aims to determine whether patterns of potentially unnecessary referrals from either primary care physicians (PCPs) or urgent care centers (UCCs) can be identified, thereby to reduce ED visits by patients who could be treated elsewhere. Literature from the United States reports up to 35% unnecessary referrals from UCCs.

METHODS

A retrospective cohort study was conducted of patients referred to an ED in Jerusalem by either their PCP or a group of UCCs with a full range of laboratory tests and basic imaging capabilities between January 2017 and December 2017. The data were analyzed to identify referrals involving diagnoses, specialist consultations, and examinations unavailable in the PCP's office or UCC (e.g., ultrasound, CT, echocardiogram, or stress test); these referrals were considered necessary for completion of the patient work-up. If patients were evaluated by an ED physician and sent home after an examination or laboratory test available at least in the UCC, the referrals were considered potentially unnecessary.

RESULTS

Significantly more referrals were made by PCPs than UCCs (1712 vs. 280, p < 0.001). Significant differences were observed for orthopedics, general surgery, and obstetrics/gynecology referrals (p = 0.039, p < 0.001, p = 0.003). A higher percentage of patients referred by PCPs had potentially unnecessary visits compared to patients referred by UCCs (13.9% vs. 7.9%, p = 0.005).

CONCLUSION

A robust UCC system may help further reduce potentially unnecessary visits (including complex patients) to the ED.

Prediction of emergency department patient disposition decision for proactive resource allocation for admission

We investigate the capability of information from electronic health records of an emergency department (ED) to predict patient disposition decisions for reducing "boarding" delays through the proactive initiation of admission processes (e.g., inpatient bed requests, transport, etc.). We model the process of ED disposition decision prediction as a hierarchical multiclass classification while dealing with the progressive accrual of clinical information throughout the ED caregiving process. Multinomial logistic regression as well as machine learning models are built for carrying out the predictions. Utilizing results from just the first set of ED laboratory tests along with other prior information gathered for each patient (2.5 h ahead of the actual disposition decision on average), our model predicts disposition decisions with positive predictive values of 55.4%, 45.1%, 56.9%, and 47.5%, while controlling false positive rates (1.4%, 1.0%, 4.3%, and 1.4%), with AUC values of 0.97, 0.95, 0.89, and 0.84 for the four admission (minor) classes, i.e., intensive care unit (3.6% of the testing samples), telemetry unit (2.2%), general practice unit (11.9%), and observation unit (6.6%) classes, respectively. Moreover, patients destined to intensive care unit present a more drastic increment in prediction quality at triage than others. Disposition decision classification models can provide more actionable information than a binary admission vs. discharge prediction model for the proactive initiation of admission processes for ED patients. Observing the distinct trajectories of information accrual and prediction quality evolvement for ED patients destined to different types of units, proactive coordination strategies should be tailored accordingly for each destination unit.

Emergency department performances during overcrowding: the experience of the health protection agency of Brianza

Background: Hospital emergency departments (ED) can contribute to improve health outcomes and reduce costs of health care system. This study evaluated ED admissions during a twelve months period, analyzing characteristics of patients who underwent to emergency care in order to understand factors involved in ED overcrowding and promote adequate management. Methods: This retrospective study analyzed a twelve months window, with in-depth focus on December/January when almost all EDs reported overcrowding. All ED admissions were recorded in electronic schedules including: demographic characteristics, time/date of the access, incoming triage code, diagnosis, performed procedures, discharge, time/date of discharge. A backward multivariable logistic regression model was used to estimate relationships between investigated variables and ED pattern mortality. Results: A total of 416,299 ED admissions were analyzed. During the overcrowded period there was an increase in patients admissions (+32 patients per day, p = 0.0079) with a statistically significant rise of critical patients (+1.7% yellow codes and +0.7% red codes, p < 0.001) and older subjects (+1.4% patients aged 75 or more years, p < 0.001). Moreover, there were statistically significant increases in waiting times and in length of visits, a higher percentage of patients who were hospitalized (13.3% vs. 12.2%, p < 0.001), left ED (4.46% vs. 4.15%, p < 0.001) and died (0.27% vs. 0.17%, p < 0.0001). This latter result maintained a marginal statistical significance (OR = 1.16, 95% CI = 0.98–1.38, p = 0.075) after adjustment for confounding. Conclusion: Our study highlights that ED crowding can determine measurable worsening in ED services and patient outcomes as mortality, waiting times, lengths of stay, percentage of abandonment without being seen and, probably, costs. Thus, address ED crowding has to be considered an important public health priority requiring policymakers involvement.

Optimal Strategy in Queueing Systems in Emergency Department

The authors' study a noncooperative game problem for queueing control in emergency department (ED). One of the challenges to emergency department (ED) is the control of the urgent patients and the non-urgent patients. The urgent patient which is the primary customer, can be considered as the service interruption in a queueing system. The service interruptions occur frequently and can incur significant delays for the non-urgent patients. Therefore, a non-urgent patient needs to decide whether to join the queue or leave. The scenario is modeled as an M/M/1 queueing game with server interruption where each patient wants to optimize his benefit. It is shown that the individually optimal strategy for joining the queue is characterized by a threshold of queue length. The socially optimal threshold of queue length is also obtained. To bridge the gap between the individually and socially optimal strategies, a pricing mechanism is proposed to toll the service of each non-urgent patient, thus equalizing the two optimal strategies.

Two emergency departments, 6000km apart: Differences in patient flow and staff perceptions about crowding

INTRODUCTION

Emergency department (ED) crowding is a worldwide public health issue. In this study, patient flow and staff perceptions of crowding were assessed in Pakistan (Aga Khan University Hospital (AKUH)) and in the Netherlands (Haaglanden Medical Centre Westeinde (HMCW)). Bottlenecks affecting ED patient flow were identified.

METHODS

First, a one-year review of patient visits was performed. Second, staff perceptions about ED crowding were collected using face-to-face interviews. Non-participant observation and document review were used to interpret the findings.

RESULTS

At AKUH 58,839 (160visits/day) and at HMCW 50,802 visits (140visits/day) were registered. Length of stay (LOS) at AKUH was significantly longer than at HMCW (279min (IQR 357) vs. 100min (IQR 152)). There were major differences in patient acuities, admission and mortality rates, indicating a sicker population at AKUH. Respondents from both departments experienced hampered patient flow on a daily basis, and perceived similar causes for crowding: increased patients' complexity, long treatment times, and poor availability of inpatient beds.

CONCLUSION

Despite differences in environment, demographics, and ED patient flow, respondents perceived similar bottlenecks in patient flow. Interventions should be tailored to specific ED and hospital needs. For both EDs, improving the outflow of boarded patients is essential.