TL;DR
This paper uses computer simulation to investigate how nurse-to-patient ratios impact patient flow and delays in emergency departments (EDs). The study reveals that increasing the patient load per nurse significantly lengthens patient "bedtime" (time spent in an ED bed before discharge or admission), highlighting the complex interplay between staffing levels, patient acuity, and operational factors in ED efficiency.
What Problem Does This Paper Address?
Emergency departments (EDs) in the United States frequently experience crowding, a situation where patient volumes overwhelm available resources. This crowding can lead to diminished quality of care and increased patient length of stay (LOS). A primary driver of this issue is a mismatch between patient volume and acuity, and the available staffing. While EDs often rely on trial-and-error or historical data to adjust staffing, there's a need for more systematic approaches to understand the impact of nurse-to-patient ratios. This paper aims to address this gap by using computer simulation to quantify the effects of varying nurse-to-patient ratios on key ED performance metrics, such as patient wait times and "bedtime."
Methods
The researchers developed a 24-hour computer simulation model of ED care. This model incorporates multiple factors influencing patient flow, including:
- Patient Acuity: Modeled using six levels of facility Evaluation and Management (E&M) billing codes, which serve as a proxy for resource utilization.
- Patient Flow: The model simulates a standard ED process: arrival, triage, registration, bed placement, RN assessment, MD assessment, MD discharge, RN discharge, and bed turnover.
- Staffing: The core of the study involves varying nurse-to-patient ratios (1:2, 1:3, 1:4, and 1:5) and observing their impact. Physician (MD) staffing was also varied (1:3 and 1:6 ratios).
- Resource Constraints: The model includes variables like bed availability, physician utilization, and the presence of technicians.
- Crowding Simulation: The study also simulated crowding by increasing patient arrival rates and patient boarding times (admitted patients waiting for an inpatient bed).
The simulation was built using a language called Little-JIL, and task times were derived from clinical studies at the authors' institution. The model was run for 72-hour periods, with the first and last 24 hours discarded to minimize startup and shutdown effects. A significant number of simulations (200 per scenario) were conducted to ensure 95% confidence intervals for key variables were within 5% of the mean.
Datasets
The paper does not explicitly list distinct datasets for training. Instead, it relies on:
- Institutional Data: Task times for various ED processes were derived from clinical studies at the authors' institution.
- National Data: Patient arrival rates, ED size, Emergency Severity Index (ESI) distribution, test frequencies, and nurse hourly costs were based on national data sources (e.g., National Hospital Ambulatory Medical Care Surveys).
- Existing Benchmarks: The model's parameters were calibrated to reflect national averages and the specific characteristics of the study institution's high-acuity ED.
Key Insights and Ideas
- "Bedtime" is a Critical Bottleneck: The study highlights that increasing the number of patients per nurse directly leads to longer "bedtime" (time from bed placement to discharge), indicating delays in care. This is a more sensitive metric than overall LOS in this simulation.
- Nurse Utilization is Key: As nurse-to-patient ratios increase (fewer nurses per patient), nurse utilization rises significantly. High utilization means nurses have less availability for subsequent tasks, exacerbating delays.
- Acuity Matters: Higher-acuity patients, requiring more resources, naturally lead to higher nurse utilization and longer bedtimes, even with optimal staffing ratios.
- Crowding Amplifies Delays: Simulated crowding (increased arrivals or boarding) disproportionately impacts patient delays, especially under less favorable nurse-to-patient ratios.
- Staffing is Multifaceted: The study demonstrates that nurse-to-patient ratios are not the sole determinant of ED flow. Physician utilization, the presence of technicians, and bed availability all play crucial roles.
Key Results
- National ED Model:
- Increasing the RN:patient ratio from 1:2 to 1:5 increased mean bedtime from 92 minutes to 168 minutes.
- Nurse utilization increased from 60% (1:2) to 90% (1:5).
- Nursing costs increased as the ratio became more favorable to nurses (e.g., $4970 for 1:2 vs. $3360 for 1:5).
- High-Acuity ED Model:
- Similar trends were observed, with higher baseline bedtimes and nurse utilization due to increased patient acuity.
- Bedtime increased from 116 minutes (1:2) to 181 minutes (1:5).
- Nurse utilization increased from 64% (1:2) to 88% (1:5).
- Impact of Crowding:
- A 10% increase in arrival rate led to increased bedtime and RN utilization, particularly at a 1:4 ratio.
- Decreasing the number of beds by 50% significantly increased waiting times, especially at a 1:4 ratio.
- Technicians and MDs:
- Adding technicians reduced bedtime for both 1:3 and 1:4 ratios.
- Decreasing MD utilization (increasing MD:patient ratio) led to increased bedtime for all RN ratios.
Limitations and Future Work
- Model Simplification: The model focuses on clinical tasks and does not fully account for non-clinical activities or interruptions that nurses face.
- Input Estimation: While based on data, some input times (e.g., nurse time by billing level) were estimated rather than directly measured.
- Predictive vs. Qualitative: The authors acknowledge that the model is designed to show the qualitative impact of ratios rather than predict exact patient wait times.
- Future Work: The authors suggest that nurse executives should use such innovative modeling tools to develop adaptable staffing models that consider patient acuity, census, and volume while prioritizing safety. Further research could incorporate non-clinical tasks and interruptions to provide a more comprehensive picture.
Summary updated