PRDV009: Writing Grant Proposals

Improving Patient Flow and Reducing Emergency Department Crowding: A Guide for Hospitals

Appendix C: Example Implementation Plan

Improvement Strategy Name: Mid-Track: The Solution to the ESI 3 Conundrum

Hospital: Good Samaritan Hospital Medical Center

Date: May 14, 2009

1. Goals and Strategies

Problem Statement

In 2007, we identified that our left-without-being-seen (LWBS) rate of 3.5% was higher than acceptable. We implemented a plan to address this issue, and the LWBS rate dropped by nearly 45%. Though this represented a dramatic reduction, this rate eventually "plateaued" over the next 2 years at 2%.

When we analyzed these data, they clearly demonstrated that the Emergency Severity Index (ESI) 3 patients represented the most significant subgroup in the LWBS data. In fact, over 75% of all walk-outs were patients triaged to an ESI category 3, and 85% presented with one of six chief complaints. This patient population also had the longest wait time to be seen by a physician.

Goal Statement

To expedite the care of the ESI 3 subpopulation of patients by reducing walk-out rates by 25% and an average time-to-provider to less than 60 minutes. We hope to achieve this within 3 months of initial implementation. We plan on implementing on 8/4/09. Therefore, we should reach this goal by the end of 10/09.

Strategy Description

We will identify a subset of ESI 3 patients that will be affected by this strategy. This subset will include patients (1) whose chief complaint is any of the following: abdominal pain, vaginal bleeding, pregnancy complication, vomiting, flank pain, or headache; (2) those who meet predefined criteria; and (3) those who arrive to the emergency department (ED) Monday through Friday between 4 p.m. and 11 p.m.

We plan a two-step process for expediting care for ESI 3 patients. The first step is to add a physician to triage Monday through Friday between 4 p.m. and midnight (stretch will be extra hours and 7 days if possible). The subset of ESI 3 triaged patients will be referred directly to the physician in triage who will begin the evaluation of the patient and order appropriate tests. The second step is to utilize the ambulatory surgery unit (ASU) (which is one floor above the ED) as the ESI 3 patient district (mid-track). Here a nonphysician provider (NPP) will receive the patients and coordinate their care with the physician in triage.

To implement this strategy, we first had to identify an area of the ED that we could assign as the mid-track. We attempted to do this within the ED by reassigning one of the four geographic districts. However, the other districts were quickly overwhelmed with ESI level 1 and level 2 patients, and a disproportionate amount of acuity was being handled by the remaining three districts. This resulted in a number of complaints from the staff, and we terminated the pilot after the initial 1-month period. However, we needed to identify another space to house mid-track.

The ASU is directly above the ED, proximate to the ED staff and our radiology services. This area has operations between 6 a.m. and 6 p.m., with a significant decrease in census at 4 p.m. We approached administration, and approval was obtained to use this area after 4 p.m., with certain caveats:

1. We would only occupy one area of the ASU from 4 p.m. through midnight. The other areas would continue to operate, and some areas would be prepped for the next operating day and left undisturbed.
2. Housekeeping had to be involved and would be responsible for cleaning the area used by the ED once we left the ASU after midnight.
3. The ED would be responsible for bringing up supplies needed for our patients.
4. The ED purchased 12 reclining hospital chairs for our patients to use. No stretchers would be used for this project, as we felt patients needed to be ambulatory to qualify for care in this location.
5. The ED identified nursing staff and clinical staff to supervise the patients. We identified the nurse practitioners as the ones to supervise the patients and LPNs to assist them. All care would be coordinated with the physician in triage.
6. Security had to be involved. We placed security personnel on scene in the ASU during the 8 hours of operation. This was done only as a precautionary measure.
7. A protocol had to be developed to identify what types of patients would be best suited for care in this environment. It would also dictate the time of day that new patients would no longer be transferred to the ASU, as well as the procedure for transferring existing ED patients in the ASU back to the ED when the ASU-ED project ended for the day (at midnight).
8. The medical staff had to be informed that patients might be in this area, as this was a new protocol. This could be accomplished at general staff meetings and via notices and letters.
9. The ED attending staff had to familiarize themselves with the protocol and the details outlining the expectations for patient selection as well as hand-off of patients that straddled shifts. This process of education for the ED attending physicians as well as the ED staff was expected to take several months.
10. Once the project was started, feedback would be requested constantly and data reviewed. Protocol adjustments could be made based on this feedback process.

We also had to identify a location within triage that the physician could occupy. We have five triage bays, and one is currently used for performing EKGs. This bay will be used for the physician. It contains a computer for documentation and an exam table/stretcher for evaluations. The physician in triage would only see a patient after the triage nurse assessed the patient and determined that the patient qualified for care under this new protocol. The physician would have the right to reassign the patient to the main ED if he or she felt that the severity of illness warranted it.

Conceptually, we realized that adding more space would not necessarily address the core problem: inpatients occupying ED beds and increasing the throughput times for all ED patients. However, given our options, this approach seemed to allow us to address the issue with expediency, while simultaneously developing programs to address the inpatient aspect of the throughput issue.

2. Approach

3. Estimated Time and Expenses

4. Performance Measures

Source: Agency for Healthcare Research and Quality, https://www.ahrq.gov/research/findings/final-reports/ptflow/appendix-c.html
This work is in the Public Domain.