From the Editor's Desk and Ask the Experts

Praveen Mital, MD, MBA, FACEP
Secretary/Editor ACEP Observation Section
Director of ED Clinical Operations, Director of Clinical Decision Unit
North Shore University Hospital, Manhasset, NY

Welcome to the 2025 spring edition of the ACEP Observation section newsletter. The popularity of observation continues to rise, with our membership now exceeding 200 individuals. Our group comprises a diverse array of leaders, ranging from new and prospective observation directors to the nation's foremost observation experts. Our emergency physician colleagues continue to value observation highly, noting its added flexibility, enhanced hospital resources, and additional time for patient care.   

Health insurers remain supportive of observation as an efficient method to avoid unnecessary inpatient admission costs. Additionally, many hospitals' executive teams are prioritizing the expansion of observation services in their business plans, aiming to reduce length of stays, minimize unnecessary admissions, and improve quality ratings while addressing ED holding patients.

In this article, we will delve deeper into the topic of observation, specifically focusing on the management of observation units (OUs) and addressing some of today's pressing issues. Common challenges are frequently encountered across different observation units. Rather than providing answers solely from my own perspective, I sought insights from my mentors, whose responses were remarkably informative. I trust that you will find their answers meaningful and that they will empower you to enhance your own observation units.

Chest pain is usually the top diagnosis placed in Observation Units (OU). When designing a chest pain protocol, what have you tweaked to make it work better?

• Our chest pain protocol encompasses high sensitivity troponin and accepts low and intermediate risk heart score patients.
  - Margarita Pena-Morris, MD, FACEP
• Since the advent of high-sensitivity troponin, our ED has been sending home more low-risk chest pain, and subsequently, we have been putting more high-risk chest pain into our observation unit. We’ve had to tweak our protocol to utilize HEART scores and remind our clinicians to directly admit high-risk patients such as HEART score 7 or those that are too high risk to stress and probably should just be considered for cardiac catheterization.
  - Praveen Mital, MD, MBA, FACEP
• Much is dependent on the imaging modalities available at your institution. Make sure the patient is appropriate for the selected modality. There is no data that chest pain patients need a stress and an echo. Many OU patients don’t want to or are unable to exercise. Patients with high resting heart rates (80s-90s) may not be good CCTA candidates.
  
  It’s best to meet with their directors and iron out processes to ensure OU patients get priority for scans and reads. Come to an understanding of what next steps will be for patients with abnormal imaging. Will they need more testing? Will they go to catheterization? Should they be admitted, or can they stay in the OU? Monitor your chest pain length of stay (LOS) as well as any outliers. Use this data to advocate for streamlining tests and disposition procedures.
  - Matthew Wheatley, MD, FACEP
• It is often confusing for the clinical team to decide what stress test is appropriate for a given patient and what stress test is available. Explaining and delineating the various indications and contraindications for a given test and when to use a given test is very helpful. For example, do not use stress test X if the patient has condition Y.  Absolute and relative contraindications could be listed and checked off as a tool for selecting a given stress test.
  - Sharon Mace, MD, FAAP, FACEP
• To avoid holding all patients placed in OU for stress testing, incorporate expedited outpatient cardiology follow up options at multiple points in the protocol as risk stratification continues. Many patients can be managed outpatient without provocative testing.
  - Christopher Caspers, MD, FACEP

Cellulitis is very commonly placed in OU but sometimes you can experience a high admit rate in this group. What have you done to mitigate this?

• Education is important. Understanding that the erythema is generally not going to resolve in 24 hours can avoid excess LOS. The clinical outcome should be non-progression of symptoms and not solely regression of erythema. Any patient with an infected wound or ulceration should be managed inpatient and not placed in an OU.
  - Christopher Caspers, MD, FACEP
• As with any OU protocol, look at the patients who require admission at your site and modify your inclusion/exclusion accordingly. Previous attempts to derive an admission decision tool for cellulitis have been unsuccessful. So much of the disease treatment and disposition is still based on gestalt. One reason for failure is an unrecognized abscess. Ensure these have been considered and ruled out in the ED. Diabetic foot infections should be admitted due to high failure rates. We allow hand and face infection in our OU because we have specialists who follow these patients but consider limiting these patients if there is no specialist coverage.
  - Matthew Wheatley, MD, FACEP
• Every protocol should have strict inclusion and exclusion criteria. Someone with a diabetic foot does not belong in the OU. If the appropriate patient with cellulitis is placed in the OU, then I don't think there is a high admission to the inpatient service. It is important to list the exclusions, such as patients with high fever, toxic appearing, markedly elevated WBC (over 15,000) or anyone who is immunocompromised. These do not get better in 24 hours. You need strict inclusion and exclusion criteria and these need to be enforced.
  - Sharon Mace, MD, FAAP, FACEP
• For cellulitis, we generally avoid diabetic foot/lower extremity infections because of the high-failure rates and long length of stays. We also started looking into home-infusion of antibiotics – a coordinated effort among case managers, home infusion companies, and infectious disease.
  - Praveen Mital, MD, MBA, FACEP
• We typically consult with the Infectious Disease (ID) specialists who are familiar with our observation model and evaluate our patients early on. They provide their expert opinion on which patients can be: (1) discharged home immediately on oral antibiotics or after one or two doses of intravenous (IV) antibiotics; (2) requiring admission, meaning more than 24 hours of IV antibiotics; and/or (3) considered suitable candidates for home IV antibiotics with peripherally inserted central catheter (PICC) line placement.
  - Margarita Pena-Morris, MD, FACEP

Sometimes, others in the hospital may want to use your OU for other things such as holding inpatients, PACU, or psychiatric patients awaiting transfer. Have you experienced any of these challenges and what should we do in the future if faced with these?

• Yes, I have during the Covid epidemic. Our OU was non-Covid during the peak epidemic time and our census was down, so we split our unit in half. One half were observation patients staffed by us (ED), and the other half were inpatients under the care of a PCP.
  
  I believe it is important to not encourage use of an OU as a holding unit. This can negatively impact efficiency, nursing workload, and short stay culture.
  - Margarita Pena-Morris, MD, FACEP
• This is a numbers game and holding inpatients in the OU should be avoided as it defeats the purpose. For every inpatient held in the OU for 48 hours, we can disposition 4 observation patients with average 12-hour LOS. Moreover, the OU nurse time is extremely important. When placing an inpatient in the OU, the OU nurses need to do a much longer intake evaluation (this may be over an hour longer than for the OU patient) and then when the patient gets admitted to the floor, the floor nurse has to repeat most of the same process.
  - Sharon Mace, MD, FAAP, FACEP
• Avoid missing appropriate OU placements or you risk scattering patients in non-OU beds. If the unit is not filling, then you risk ‘use it or lose it’ scenario with resources. Accommodate others when able and when in the best interest of the patient.
  - Christopher Caspers, MD, FACEP
• We have successfully resisted placing any holding patients in our observation unit, arguing that our unit decreases the number of holding patients. We sometimes have to remind hospital leadership or logistics that our unit will routinely be empty near 5pm (typically when holding starts peaking) but will be full after midnight (when everybody else is done looking at the numbers for the day).
  - Praveen Mital, MD, MBA, FACEP
• The best thing you can do to keep non-observation patients out of the OU is to keep your unit full. Have the providers in the unit look at the ED board and proactively pull patients into the OU. Hospital and ED administrators tend to get uncomfortable with empty beds, especially when the ED and inpatient beds are full.
  
  Get data on observation and short stay inpatient patient... are there missed opportunities or protocols you should consider adding? If you must take non-observation patients, do it as a trial period, limit the hours or number of OU beds that can be used.
  - Matthew Wheatley, MD, FACEP

What advice would you give an EM doc that might be taking over as a director of an established OU?

• Try to think about what the challenges will be in the next 1-5 years: is the unit the right size, do you have the right staffing, are there opportunities for expansion/contraction? A SWOT analysis is a good place to start. You and the OU nurse manager should be a dyad. Develop a good working relationship with them. If there is no OU nurse manager, push for one.
  
  Meet with key stakeholders for your unit. Develop relationships with them. Give them your personal cellphone number. This goes a long way to developing trust across departments. If there are any active problems, create a meeting cadence to assess and address them until they are no longer problems.
  - Matthew Wheatley, MD, FACEP
• Be prepared. Use all the resources out there from the textbooks to the literature. This will help clinically, administratively and financially. Also network with others. Never take on the task of starting an OU without having the needed resources, space, supplies and most importantly, additional staffing (MD, PA, NP, RN, ancillary staff) to cover the unit. You are doomed to fail especially if you fail to secure the extra clinical time needed to take care of the OU patients.
  - Sharon Mace, MD, FAAP, FACEP
• Build strong relationships with your nursing manager, department leaders, and hospital administration. Share your OU successes. Regularly monitor your utilization and quality metrics by condition. Frequently educate others on your protocols and processes.
  - Margarita Pena-Morris, MD, FACEP
• Begin by building relationships with your observation staff, listening to their successes and understanding their challenges. Next, establish connections with key stakeholders in the hospital, such as cardiologists, neurologists, radiologists, hospitalists, and surgical sub-specialists. Many daily issues can be efficiently addressed and resolved through these established relationships and by managing expectations. Maintain a regular schedule of meetings to check in, discuss progress, and express gratitude.
  - Praveen Mital, MD, MBA, FACEP
• Good luck, be patient, use the ACEP observation section for guidance and mentorship. We all learn from each other and ultimately you will need to tailor your unit to fit the needs of your ED and hospital, and not all units look the same.
  - Christopher Caspers, MD, FACEP

Note: some responses were edited for length and clarity.