New York ACEP Updates
Jeffrey S. Rabrich, DO MBA FACEP FAEMS
Senior Vice President
Envision Physician Services
Advancing Leadership, Navigating the Funding Cliff & Addressing Operational Challenges in Emergency Departments
As emergency physicians practicing in New York State, we are at the nexus of three inter-related imperatives: cultivating the next generation of leaders, confronting significant federal health-care funding uncertainties for 2026, and managing the relentless challenge of our departments such as hospital overcrowding, workplace violence and scope creep. Each of these requires vigilant engagement, strategic planning, and advocacy.
Leadership Fellowship: Investing in the Future of Emergency Medicine
As you read this, we are wrapping up our inaugural year of our New York ACEP Leadership and Development Fellowship. We will soon graduate our first two fellows who have done an excellent job, and we look forward to their future involvement as New York ACEP and Emergency Medicine leaders. They have participated in and helped shape our advocacy efforts, written proposals for things such as the rural health transformation grant, as well as contributing at board meetings and other events.
Why this matters: Leadership skills enable us to influence policy, shape systems of care, improve departmental culture, and respond proactively to evolving external pressures. New York ACEP encourages members, especially those early in their career, to explore leadership fellowships, mentorship opportunities, and collaborative roles within the College. By doing so, we not only strengthen our profession—but position ourselves to better protect our patients and our specialty.
Federal Health-Care Funding Challenges for 2026
The horizon for fiscal 2026 brings severe risk to federal health-care funding streams that underpin much of our safety-net infrastructure. For example:
- The One Big Beautiful Bill Act (OBBB) proposes steep reductions in health-coverage subsidies and Medicaid funding, which has been tied directly to ER access and crowding.
- The U.S. Department of Health and Human Services (HHS) has publicly posted its contingency staffing plan for FY 2026 which anticipates potential furloughs and program slow-downs in absence of fully enacted appropriations.
- Children’s hospitals and pediatric emergency services are already bracing for the consequences: anticipated funding shortfalls will contribute to service reductions, staffing challenges and increased strain on EDs.
What that means for emergency departments in New York: As federal and state subsidies shrink, community health centers as well as other outpatient and social services may cut services or close, likely driving more patients toward hospital EDs. At the same time, many hospitals operate on razor-thin margins, leaving little buffer for increased volume, rising acuity, or delayed discharges. As we all know, New York’s healthcare system was already strained and with the potential loss of federal funding for Medicaid and other programs, there is concern that hospitals or emergency departments could be forced to close.
What are we doing about this? The New York ACEP leadership team has met with the Department of Health as well as the legislature and executive to express our concerns as well as offer input on proposals for the rural health transformation grants, protecting Medicaid payments for emergency physicians who as a specialty are disproportionately affected by cuts to Medicaid. We additionally have been meeting with the national ACEP government affairs staff to understand additional impacts and advocacy efforts as policy is enacted.
Hospital Overcrowding, Workplace Violence and Scope Creep
While most of us intuitively know the pressures of ED crowding, recent data underscore its severity and evolving nature. A recent summary of a RAND Corporation-sponsored study warned that the “viability of emergency care as we know it is at risk.” Contributing factors include a rise in high-acuity cases, declining reimbursements, increasing uncompensated care, staffing shortages and rising boarding times. Additionally, we face increasing incidents of violence against healthcare workers with the ED being one of the most high-risk sites for violence to occur. Finally, we expect to see proposals for independent practice and expanded scope for non-physician providers in the state.
These have been our top advocacy priorities at New York ACEP, and these will continue during the 2026 legislative session. We successfully advocated for a bill to help protect healthcare workers from violence which passed earlier this year, and we look forward to the Governor signing it before the end of the year. We have already had several meetings prior to the start of the 2026 session to explain to lawmakers our concerns as well as our patients concerns regarding independent practice of non-physician providers in New York including providing them with data related to costs and outcomes. We are also pushing for more data transparency and public reporting of hospital capacity data and hope to advocate for a bill similar to the one our colleagues in neighboring Connecticut based related to hospital overcrowding and boarding. We will continue these efforts throughout the upcoming session including on our Advocacy Day, which will be February 4th in Albany. Please join us!
In Summary
For New York ACEP members, the weeks and months ahead demand a multi-pronged focus: leadership development, policy engagement, and operational innovation. Our specialty’s ability to deliver high-quality emergency care hinges not only on our individual clinical performance—but on our collective capacity to lead, influence policy and manage systems. I encourage each of you to consider becoming engaged in local/state advocacy efforts around the looming 2026 funding cliff, and to participate actively in departmental efforts to mitigate operational challenges in our EDs.
If you’re interested in learning more about leadership fellowships, policy initiatives, or local crowding-mitigation programs, please reach out to the New York ACEP leadership team. Together, we can serve our patients, support our colleagues, and protect the sustainability of emergency care in our region.
Thank you for all you do on the front lines.
Sound Rounds
Thomas M. Kennedy, MD
Assistant Professor of Pediatrics in Emergency Medicine
Columbia University Vagelos College of Physicians and Surgeons
Department of Emergency Medicine, Division of Emergency Ultrasound
NewYork-Presbyterian Morgan Stanley Children’s Hospital
Allison Havens, DO
PGY-3, Emergency Medicine Resident
University of Rochester Medical Center – Strong Memorial Hospital
Michael Lu, MD FACEP FPD-AEMUS
Associate Professor, Emergency Medicine
Director of Emergency Ultrasound
University of Rochester Medical Center – Strong Memorial Hospital
Beyond the Cuff: Using Doppler to Detect Abnormal Mitral Inflow Velocity Variability in Early Cardiac Tamponade
Case
A 73-year-old female with a history of chronic obstructive pulmonary disease, hypertension, chronic back pain, coronary artery disease and a previously provoked deep vein thrombosis after surgery (not currently anticoagulated) presented for shortness of breath and hypoxia at a free-standing emergency department (ED). For the past three weeks, she has had increasing shortness of breath with a productive cough and lower extremity edema. Pulse oximetry revealed an oxygen saturation of 79% on room air at her primary care office. She was placed on supplemental oxygen via nasal cannula and sent to the ED for further evaluation.
On arrival, she was afebrile with a blood pressure (BP) of 80/45 mmHg, heart rate of 71 beats/minute, respiratory rate of 26 breaths/minute and a pulse oximetry reading of 91% on 4 L/min via nasal cannula. She denied chest pain, back pain, fever or chills. Due to her abnormal vital signs, she was brought back to an exam room immediately and a repeat BP was 107/52 mmHg without any interventions. Her physical exam was remarkable for distant heart sounds, diminished breath sounds and bilateral lower extremity edema. Neck veins were not distended. During the first hour of her ED presentation, her automated BP cuff readings continued to alternate between 83/49 mmHg and 104/69 mmHg.
Given the variable BP cuff readings and the patient’s presenting symptoms, BP was measured manually during inspiration and expiration to assess for pulsus paradoxus, showing a difference of 26 mmHg. Simultaneously, a bedside focused cardiac ultrasound (FOCUS) was completed demonstrating a new pericardial effusion that was not seen on a chest CT scan done one month prior. The inferior vena cava (IVC) was dilated and non-responsive to inspiratory sniff, there was prolonged right atrial (RA) collapse and mitral valve inflow velocity variation >25%, indicative of cardiac tamponade. In addition, there was equalization of the E-wave and A-wave pattern and even E/A reversal in some cardiac cycles, further demonstrating impaired left ventricular (LV) diastolic filling. She was emergently transferred to a tertiary care center.
Discussion
Cardiac tamponade is a life-threatening condition that affects around 1 out of 5,000 people caused by an increase in pericardial fluid that leads to increased external pressure around the heart, resulting in impaired diastolic filling and subsequent obstructive shock. Cardiac tamponade can be caused by many different pathologies including trauma, infection (e.g., tuberculosis and HIV), uremia, myocardial infarction, malignancy, rheumatologic conditions, radiation to the area, or iatrogenic causes. It is associated with Beck’s triad, which includes muffled heart sounds, increased jugular venous distention and hypotension; however, the sensitivity of this triad is poor with less than 50% of patients in cardiac tamponade demonstrating one finding and less than 20% demonstrating all three findings.1 Some other physical exam findings, such as pulsus paradoxus, can also be utilized to assist in the diagnosis of cardiac tamponade, but this may not always be easily detected in many loud, busy ED environments.
Pulsus paradoxus is an abnormally large drop in systolic BP with inhalation due to impaired cardiac filling. Pulsus paradoxus greater than 12 mmHg has a sensitivity of 98% and specificity of 83% for cardiac tamponade.2 To measure pulsus paradoxus using a manual BP cuff, the cuff is inflated 20 mmHg above the patient’s systolic BP and then deflated very slowly. The first Korotkoff sounds will be heard with expiration only and disappear with inspiration; this pressure indicates the expiratory systolic pressure. The cuff will continue to be deflated and respirations monitored and when Korotkoff sounds are consistently heard with both inspiration and exhalation, this pressure indicates the inspiratory systolic pressure. The difference between expiratory and inspiratory pressures yields the pulsus paradoxus. Any value over 10 mmHg is concerning for cardiac tamponade. As mentioned earlier, it may be challenging to perform this in the ED due to environmental reasons or while establishing peripheral IV access during initial resuscitation.
Always perform FOCUS in the evaluation of a hypotensive patient or in the setting of undifferentiated shock, since absence of a pericardial effusion can quickly rule out cardiac tamponade. However, the presence and size of a pericardial effusion alone does not necessarily indicate tamponade. Slower developing pericardial effusions allow the pericardium time to stretch and accommodate to the changes in pressure, while faster developing effusions, such as pericardial effusion caused by trauma, are more likely to cause tamponade with small volumes. Sonographically, the earliest sign of tamponade is systolic RA collapse as the intrapericardial pressure becomes higher than the RA filling pressure. Next, right ventricular (RV) diastolic collapse is seen as the intrapericardial pressure becomes higher than the RV filling pressure. As opposed to RA collapse, RV collapse is highly specific (75-90%), but not very sensitive (48-60%) for tamponade.3 The absence of both RV and RA collapse has a 90% negative predictive value for tamponade.4 However, the visualization of RA and RV collapse on B-mode imaging alone may be difficult to obtain and therefore incompletely evaluated. Incorporating Doppler measurements can be easily done to help detect cardiac tamponade.
Mitral inflow velocity evaluates the speed of blood flow across the mitral valve and can be used to evaluate the same pathophysiology that produces pulsus paradoxus. Using pulsed wave Doppler in the A4CH view, with the gate placed at the tips of the mitral valve leaflets, velocity waveforms are obtained throughout the patient’s respiratory cycle (Figure 4). The first peak during each cardiac cycle is the E-wave, representing early passive diastolic filling and the second peak is the A-wave, representing the atrial contraction during late diastole. As the patient inhales and exhales, the E-wave heights will vary. The tallest E-wave peak during expiration is measured and compared to the shortest E-wave peak during inspiration. A variation of 5-10% may be physiologic, but a difference of >25% is concerning for cardiac tamponade. As demonstrated in Figure 4, the mitral inflow velocity variation for our patient was >30%, supporting the diagnosis of early tamponade.
Another interesting finding in this case is the change in the E-wave and A-wave pattern, which is used in the evaluation of LV diastology. Normally, the E-wave is taller than the A-wave. However, when there is impaired diastolic filling, the normal pressure gradient from the left atrium across the mitral valve into the LV is reduced and more reliant on the atrial contraction. This results in a blunted E-wave and taller A-wave. As seen in this case (Figure 4), the E/A ratio was approximately 1 during expiration and there was E/A reversal during inspiration, indicating mild to moderate diastolic dysfunction due cardiac tamponade.
The definitive treatment for cardiac tamponade is pericardiocentesis. Ultrasound guidance improves the safety and success rate of this procedure. While getting setup to perform the pericardiocentesis, administer intravenous fluid as a temporizing treatment to increase the patient’s preload and cardiac output. Avoid positive pressure ventilation as it decreases venous return and worsens symptoms.
Case Conclusion
At the tertiary care center ED, the patient’s oxygen requirement worsened and she was placed on medium flow nasal cannula to maintain oxygen saturations above 90%. The Cardiology service was urgently consulted and the decision was made to obtain a chest CT scan to further delineate potential causes of her pericardial effusion. The scan was remarkable for a moderate size pericardial effusion, a moderate right pleural effusion, a small left pleural effusion and a suspicious right lung mass with surrounding lymphadenopathy concerning for malignancy. The patient was admitted to the medical intensive care unit and pericardiocentesis was performed with 460 mL of serosanguinous initial output and a drain was placed with an additional 75 mL of output overnight. A thoracentesis was also performed on the right side with 1,050 mL of pleural fluid drained. She was discharged home after a four-day admission with improvement in her symptoms. She was seen by the Oncology service as an outpatient and diagnosed with small cell carcinoma of the lung, for which she has been undergoing treatment and doing well.
Indications
- Cardiac arrest
- Chest pain or palpitations
- Electrocardiogram abnormalities
- Heart murmurs
- Hypotension
- Shortness of breath
Technique
- Lay the patient supine or in the left lateral decubitus position, which will bring the heart closer to the chest wall.
- Use the phased array probe to obtain an A4CH view of the heart.
- Use pulsed wave Doppler with the gate placed at the tips of the mitral valve leaflets (see Figure 4).
- Measure E- and A-wave tracings over multiple cardiac cycles during inspiration and expiration.
- Compare the tallest E-wave peak (expiration) to the shortest E-wave peak (inspiration).
Pitfalls and Limitations
- Views may be limited due to body habitus or irregular respirations.
- Atrial fibrillation makes flow velocity variation measurements unreliable.
- Pulsus paradoxus may also be seen in other pathologies including chronic obstructive pulmonary disease or pulmonary embolism.
References
- Stolz L, Valenzuela J, Situ-LaCasse E, et al. Clinical and historical features of emergency department patients with pericardial effusions. World J Emerg Med. 2017;8(1):29-33.
- Bandinelli G, Lagi A, Modesti PA. Pulsus paradoxus: an underused tool. Intern Emerg Med. 2007 Mar;2(1):33-5.
- Alerhand S, Carter JM. What echocardiographic findings suggest a pericardial effusion is causing tamponade? Am J Emerg Med. 2019;37(2):321-326.
- Merce J, Sagrista-Sauleda J, Permanyer-Miralda G, et al. Correlation between clinical and Doppler echocardiographic findings in patients with moderate and large pericardial effusion: implications for the diagnosis of cardiac tamponade. Am Heart J, 1999. 138(4 Pt 1):759-64.
- Gelmann D, Slagle D, Seaback J, & Schoeneck J. Advanced critical care ultrasound: Use of mitral inflow velocity variation in evaluation of cardiac tamponade. EM Resident, EMRA. Published online 6/12/2024. https://www.emra.org/emresident/article/CCUS-mv-variation.
- Jensen JK, Poulsen SH, Mølgaard H. Cardiac tamponade: A clinical challenge. European Society of Cardiology. Published online 9/27/2017. https://www.escardio.org/Journals/E-Journal-of-Cardiology-Practice/Volume-15/Cardiac-tamponade-a-clinical-challenge.
Abbas Husain, MD FACEP
Associate Professor of Emergency Medicine
Associate Residency Director
Staten Island University Hospital
ACEP 2025 Council Resolutions: Key Takeaways for Emergency Physicians
Introduction
Each fall, leaders from around the country gather in a hotel convention center at ACEP national scientific assembly to discuss and debate topics that are important to EM docs. I want to present a summary of our discussions last month so you the reader can have a window into ACEP policy making and be aware of the current issues.
Workforce & Practice Environment
- Non-Competes (43): ACEP will push to eliminate physician non-compete clauses, calling them unfair and harmful to both physicians and patients.
- Board-Certified Staffing (51): Reaffirmed that every ED should have a board-certified emergency physician on-site.
- APP Scope & Training (77, 78, 65): ACEP will study nurse practitioner/PA practice patterns in independent states and is advocating for standardized, accredited EM-specific postgraduate training for APPs—while reinforcing physician-led teams.
- Employment Negotiations (71): ACEP will provide members with curated resources and webinars on job contract best practices.
Patient Safety & Care Access
- EMTALA Protections (48, 49): ACEP affirmed EMTALA obligations should supersede conflicting state laws and is pushing reforms to reduce transfer delays.
- Staffing Transparency (50, 75): ACEP wants facilities to be transparent if a physician isn’t present 24/7 and to reserve the term “Emergency Department” for sites with continuous physician coverage.
- Medicaid Support (45, 47): Strong stance against Medicaid cuts and DSH payment reductions, which threaten rural and safety-net hospitals.
- Prior Authorization Reform (53): ACEP will lobby to simplify and standardize prior authorizations to prevent care delays.
Physician Protection & Advocacy
- Workplace Violence (44): ACEP will work with the AMA to make workplace violence a top advocacy priority, including mandatory reporting.
- Medical Liability (41): Supporting a shift toward a no-fault liability system that emphasizes patient compensation and system improvement over punishing physicians.
- ICE in the ED (58): ACEP will develop clearer guidance to help EDs handle interactions with immigration enforcement while protecting patient rights.
Education & Training
- Residency Length (36): ACEP reaffirmed support for both 3- and 4-year EM residency accreditation.
- Bias & Equity Training (27, 40): Ongoing commitment to implicit bias, cultural competence and education at all levels of training.
- Firearm Safety (68): Firearm injury prevention and safe storage counseling should be integrated into EM education.
- Resident Rotations (37): ACEP will advocate for CMS funding of residents training at outside, rural and international sites.
Innovation, Systems & Sustainability
- AI in Insurance (56): ACEP will consider human oversight for all AI-driven insurance denials.
- Telemedicine (59): Discussion of cross-state licensure recognition to expand emergency telehealth.
- Environmental Sustainability (61, 62): ACEP supports greener practices, from inhaler alternatives to waste reduction in EDs.
- ED Crowding (80): ACEP will work with surgeons/hospitals on an elective surgery scheduling toolkit to help reduce ED boarding.
Public Health & Community
- Naloxone in Schools (72): Endorsing widespread naloxone access and overdose education in public schools.
- Substance Use Disorders (73): ACEP is pushing for ED-based programs for SUD treatment, harm reduction and linkage to care.
- Transgender Care (76): ACEP supports national guidelines and protections for transgender patients in EM.
- Public Media & Science Integrity (34, 89, 90): ACEP recognizes public media as a public health tool, defends scientific independence and reaffirms support for current vaccine guidelines.
Bottom Line
ACEP 2025 resolutions focused on protecting our workforce (non-competes, workplace violence, APP scope), improving patient access (Medicaid, EMTALA, prior auth), advancing education (residency, bias training, firearm counseling) and defending science and equity. Expect more advocacy on insurance, sustainability and ED operations.
To access all documents and activity from the 2025 council and past councils please visit the Council Website.
Practice Management
Brandon J. Godbout, MD FACEP
Vice Chair, Department of Emergency Medicine
Attending Physician, Department of Emergency Medicine
Lenox Hill Hospital, Northwell Health
Chair, New York ACEP Practice Management Committee
Rajas Karajgikar, MD
PGY-1, Mount Sinai Morningside West
Elyse Lavine, MD FACEP
Associate Professor
Medical Director, Mount Sinai Morningside
Eric Legome, MD FACEP
Professor and Chair
Mount Sinai West & Mount Sinai Morningside Hospitals
Vice Chair, Academic Affairs, Dept. of Emergency Medicine, Icahn Mount Sinai School of Medicine
Pathway-Driven Emergency Medicine: Embedding Clinical Standards into the EHR for Safer, Smarter Care”
Chest Pain, sepsis, stroke? Not only do I need to make the diagnosis, but how do I remember all the treatments? Is there an order set? How do I remember to order everything right? As a new Intern, I am not alone in having these questions. New recommendations seem to come all the time and I am sure even my seasoned seniors and attendings have trouble keeping up. We all know, however, our orders have a real impact on patient care and outcomes.
Given all the clinical, financial and medico-legal issues around getting our orders correct, solutions that can deliver the right care at the right time with consistency and safety is critical. In the inherently chaotic and variable nature of care in the ED, where patient presentations are unpredictable and providers may have differing levels of training, experience, and practice styles, clinical decision support can help promote consistency, safety, and adherence to evidence-based care despite this variability.
While the implementation of integrated pathways in the electronic medical record (EMR) is more current, decision making tools to guide patient care have been a cornerstone of Emergency Medicine for decades. Clinical decision tools arose to decrease variability in common presentations and to provide a way to avoid misdiagnosis while preserving specificity in patient care. Once the clinician has arrived at a reasonable certainty that the area of concern is part of their differential diagnosis, the decision tool can help avoid an expensive or invasive workup if the risk is low, or guide the clinician to the next needed test when concern is heightened. Two common examples of this are the “PERC” and “HEAR(t)” Scores. The “PERC” study demonstrated that when excluding patients classified as high risk for pulmonary embolism and when combining positive D-dimer and “PERC” criteria, the sensitivity in the limited cohort was 100% (95% CI: 0.86–1.00) with no false negative rate.1 Furthermore, a clinical trial funded by the American Heart Association compared outcomes in patients who were risk-stratified with the “HEAR(t)” score vs usual care. In the group with introduction of this tool, the results were significant for a 12.1% decrease in objective cardiac testing at thirty days, a 12 hour decreased length of stay (LOS), and a 21.3% increase in early discharges with none having major adverse cardiac events within thirty days.2 While these decision making tools are incredibly helpful for common ED presentations, pathways will allow us to further build upon patient care that starts in the ED and extends across multiple disciplines. Pathways are a broader tool adapted to translate clinical practice guidelines into specific step-by-step processes to help with workflow, as opposed to common clinical decision making tools which are specific. This is where pathways can provide an answer to adherence to evidence based practice.
One such multidisciplinary example is seen with the implementation of a palliative care pathway. A group spanning social work, palliative care, case management, and Emergency Medicine came together and created guidelines for ED staff to use to help guide clinical workup. Unlike decision tools and scores, these guidelines were developed into a chart-like algorithm that was then uploaded into the electronic health record. Results from this study show that after pathway implementation, ED LOS decreased by 2.9 hours, social work consults for hospice increased, and more patients were identified as hospice candidates.3 Thereby, the pathway proved success in improving patient outcomes across a wide range of measuring criteria. These studies are not limited to the United States. Even in Australia, introduction of an ED vertigo pathway was found to be associated with a reduced length of stay in the Emergency Department.4
The benefit of pathways encompasses all ED staff since a standardized approach allows for improved communication and for members of the care team to have situational awareness and refer to the same chart-like algorithm, a pictorial representation which other decision tools do not have. Additionally, in the ED where patient volume is high, pathways can decrease the stress that comes from decision making fatigue. With constant updates, phone calls, and consults, ED staff may experience less stress with the usage of focused, algorithmic pathways.
Although pathways show promising results, it is also important to acknowledge the limitations. These include over-standardization in the case of atypical cases as well as fatigue secondary to extra alerts. Pathways have a very specific mechanism in that they provide a step-like approach for a singular problem. Patients may have several comorbidities and complexities where pathways may fall inadequate in guiding appropriate treatment. What is indicated as a next step in a patient with a singular concern may be of little value in patients with several comorbidities and complex clinical presentation. It is for this reason that pathways, while quite beneficial, should be used as assistive and supplemental measures rather than a replacement for clinical judgment.
The implementation and usage of pathways will require support from all levels. From trainees to physicians to non-physician staff and administrators, the pathway approach can only be a success with the help of everyone. ACEP has published and released both clinical guidelines and point of care tools online, which serve as examples of evidence-based clinical content created through a collaborative process of top experts and thought-leaders for the Emergency Physician.5,6 Integrating these tools into the electronic health record locally via a pathway encourages utilization of these recommendations more consistently.
It would be prudent for organizations and hospital leaders to advocate for resources to create standardized pathways that can be developed, shared, and disseminated across hospitals. Such models could be further studied using predictive analysis to make modifications as needed. With such measures and the advent of artificial intelligence, once isolated guidelines at the discretion of individual hospitals can grow into well-connected and streamlined networks that promote equitable Emergency Medicine care for all patients. Pathway implementation has the ability to optimize appropriate, efficient, and standardized patient care, regardless of hospital, geography, or training level of the provider.
References
- Karlsson J, Islam MR, Landucci L, Siddiqui AJ. Safety and Diagnostic Utility Pulmonary Embolism Rule-Out Criteria (PERC) and D-Dimer in Emergency Department. J Acute Med. 2024 Dec 1;14(4):145-151. doi: 10.6705/j.jacme.202412_14(4).0002. PMID: 39624146; PMCID: PMC11608861.
- Mahler SA, Riley RF, Hiestand BC, Russell GB, Hoekstra JW, Lefebvre CW, Nicks BA, Cline DM, Askew KL, Elliott SB, Herrington DM, Burke GL, Miller CD. The HEART Pathway randomized trial: identifying emergency department patients with acute chest pain for early discharge. Circ Cardiovasc Qual Outcomes. 2015 Mar;8(2):195-203. doi: 10.1161/CIRCOUTCOMES.114.001384. Epub 2015 Mar 3. PMID: 25737484; PMCID: PMC4413911.
- Sarah K. Wendel, Mackenzie Whitcomb, Ariel Solomon, Angela Swafford, Jeanie Youngwerth, Jennifer L. Wiler, Kelly Bookman, Emergency department hospice care pathway associated with decreased ED and hospital length of stay, The American Journal of Emergency Medicine, Volume 76, 2024, Pages 99-104, ISSN 0735-6757, https://doi.org/10.1016/j.ajem.2023.11.017.
- Stewart V, Rosbergen I, Tsang B, Hoffman A, Kwan S, Grimley R. Do Vestibular Physiotherapy and a Clinical Pathway in the Emergency Department Improve Management of Vertigo? OTO Open. 2022 Aug 12;6(3):2473974X221119163. doi: 10.1177/2473974X221119163. PMID: 35990816; PMCID: PMC9382073.
- https://www.acep.org/resource/categories/search-facets/topics/clinical-guidelines
- https://poctools.acep.org/
Research
Laura Melville, MD MS
Associate Research Director
SAFE Medical Director
NewYork-Presbyterian Brooklyn Methodist Hospital
Chair, New York Research Committee
Alexander Suyunov, MD
St. John’s Riverside Hospital
Adrian Cotarelo, MD MHS
St. John’s Riverside Hospital
Artificial Intelligence in the ED: Triage and Risk Prediction
Artificial intelligence is a broad umbrella, with multiple applications within Emergency Medicine. This discussion focuses on two categories of AI techniques: machine learning (ML) and large language models (LLMs). ML refers to methods that learn patterns from data (numbers, categories, and time-stamped signals) to make predictions or classifications. LLMs are systems “trained” on large bodies of reference text to work with language: they can read free-text, extract key details, and generate summaries. These approaches are used across many industries, but in emergency care the same tools are being applied to support triage and risk stratification. 1-5
In practice, the techniques are complementary. LLMs can help transform triage and early note text into structured inputs, while ML models combine those inputs with vitals and other routinely collected data to estimate the likelihood of outcomes, like ICU need, mortality, or admission. The current evidence is most promising for high-acuity outcomes, and real-world use should pair these tools with clinician oversight and local validation. 3-5
Applications in Emergency Medicine
In the ED, ML and LLM models are being studied for how they can refine or enhance existing triage systems. ML tools often use structured EHR data to identify high-risk patients earlier, while LLMs are being tested to analyze triage narratives, chief complaints, and provider notes to extract information that can improve prediction models. Together, they represent a growing area of research aimed at making triage both faster and more accurate.
Emergency Severity Index Limitations
The Emergency Severity Index (ESI) remains the most widely used triage system in U.S. EDs. However, roughly one-half to two-thirds of encounters are triaged as ESI-3, blending patients who are ultimately stable with those who may require time-sensitive care. Retrospective analysis across multiple hospitals has also highlighted mistriage affecting a substantial fraction of visits, with disproportionate rates among younger, male, and Black patients and those from poorer neighborhoods – an important equity signal for any “improvement” we propose to the status quo. 2 These limitations motivate interest in ML systems that use data available at or near arrival to increase discrimination within ESI-3 and better surface the truly high-risk minority. In a large study of more than 172,000 visits, a machine-learning e-triage system outperformed ESI for predicting ICU admission, need for emergency procedures, and hospitalization, with reported AUROCs in the moderate-to-strong range for the more critical outcomes. Importantly, the model identified a meaningful subset of higher-risk patients otherwise labeled ESI-3. 3 While single-system and retrospective, this line of work consistently shows that data already captured in routine care can sharpen early risk signals beyond ESI alone.
Large Language Model Use in Triage Note Parsing
Where ML often consumes structured inputs, LLMs can parse triage text and early provider notes to assist with complaint recognition or case identification. In a retrospective cohort focused on symptomatic kidney stone visits, OpenAI’s GPT-4 model achieved the best performance among tested LLMs without evidence of race or gender bias, whereas GPT-3.5 showed disparities, underscoring that model choice matters. Authors also documented prompt sensitivity: even small wording changes can influence outputs, a reminder that LLM behavior depends on both inputs and local documentation style. This variability reinforces the need for external validation and careful instruction design before embedding LLMs into triage workflows. As of now, LLMs are best framed as adjuncts that extract or structure text signals which can then feed human judgment or ML models, rather than as stand-alone adjudicators of acuity. 4
Risk Stratification: What the Pooled Evidence Shows
Stepping back, a meta-analysis of ED AI models reported strong aggregate discrimination for mortality and critical-care needs, with somewhat lower performance for predicting hospital admission. Sensitivity and specificity estimates were generally favorable, but heterogeneity in design and endpoints was substantial. The signal is clearest for high-acuity outcomes (mortality/ICU), which likely reflects more distinct physiologic patterns at presentation, whereas “any admission” blends heterogeneous reasons and thresholds that vary by site and crowding. The authors emphasize that most included studies were retrospective and single-center, limiting generalizability. As a result, admission prediction remains a relative weak point and a prime area for prospective work. 5
How to Read and Use These Findings in Practice
For busy ED leaders and QA teams, a few principles can help interpret new AI triage papers:
Match outcome to use-case. If the goal is earlier recognition of the sickest patients, prioritize models that target ICU-level outcomes or mortality; expect stronger performance there than for “any admission.” 3,5
Look beyond AUROC. Calibration (how well predicted risks match observed risks) and subgroup performance are crucial, particularly around equity. A model with an impressive AUROC can still be poorly calibrated in your population or unequal across patient groups. 5
Beware data leakage and portability. Ensure features are available at or near triage and that performance holds in a temporally and geographically distinct cohort. Documentation style and patient mix vary widely across sites; LLMs, in particular, can be sensitive to local phrasing. 4,5
Clarify decision thresholds and actions. The value of a model depends on what changes when the alert fires: expedited clinician review, earlier labs, faster bed request, or imaging prioritization. Without a pre-specified action plan, even accurate predictions may not improve flow or outcomes. 3,5
Equity, Safety, and Implementation
Because AI models learn from historical data, they can either reproduce or reduce existing disparities depending on how they’re trained and monitored. 2 Future work should emphasize prospective, multi-center validation with explicit equity reporting and evaluation across demographic subgroups. 2-5
Overall, ML and LLM applications show early promise for triage and early risk prediction, particularly for identifying high-acuity outcomes. 3-5 Continued research will determine whether these tools can translate retrospective accuracy into meaningful clinical improvement without sacrificing safety or fairness.
Summary
AI can complement current triage tools: ML models have outperformed ESI for ICU-level outcomes and procedures in retrospective cohorts. 3
LLMs provide text-level signal: GPT-4 identified target ED cases from triage/provider notes; prompt sensitivity and external validation remain key. 4
Outcome choice matters: Mortality/ICU predictions are stronger than admission predictions in pooled analyses. 5
Further validation is needed: Prospective, multi-center studies will be important to confirm these findings, assess real-world performance, and guide safe integration into ED practice. 5
Disclosures: The authors report no relevant financial relationships.
References
- IBM. What are large language models? Available at: https://www.ibm.com/think/topics/large-language-models. Accessed 2025.
- Sax DR, Warton EM, Mark DG, Vinson DR, Kene MV, Ballard DW, Vitale TJ, McGaughey KR, Beardsley A, Pines JM, Reed ME. Evaluation of the Emergency Severity Index in US Emergency Departments for the Rate of Mistriage. JAMA Network Open. 2023;6(3):e233404. https://doi.org/10.1001/jamanetworkopen.2023.3404
- Levin S, Toerper M, Hamrock E, Hinson JS, Barnes S, Gardner H, Dugas A, Linton B, Kirsch T, Kelen G. Machine-Learning-Based Electronic Triage More Accurately Differentiates Patients With Respect to Clinical Outcomes Compared With the Emergency Severity Index. Ann Emerg Med. 2018;71(5):565-574.e2. doi:10.1016/j.annemergmed.2017.08.005. Epub 2017 Sep 6. PMID: 28888332.
- Bejan CA, Reed AM, Mikula M, Zhang S, Xu Y, Fabbri D, Embí PJ, Hsi RS. Large language models improve the identification of emergency department visits for symptomatic kidney stones. Scientific Reports. 2025;15(1). https://doi.org/10.1038/s41598-025-86632-5
- Kuo KM, Chang CS. A meta-analysis of the diagnostic test accuracy of artificial intelligence predicting emergency department dispositions. BMC Med Inform Decis Mak. 2025;25:30. doi:10.1186/s12911-025-03010-x. Available at: https://bmcmedinformdecismak.biomedcentral.com/articles/10.1186/s12911-025-03010-x
Elyse Lavine, MD FACEP
Associate Professor
Medical Director, Mount Sinai Morningside
Arvind Balasundaram, MD, MS
PGY-2 Emergency Medicine Resident
Mount Sinai Morningside-West
Eric Legome, MD FACEP
Professor and Chair
Mount Sinai West & Mount Sinai Morningside Hospitals
Vice Chair, Academic Affairs, Dept. of Emergency Medicine, Icahn Mount Sinai School of Medicine
New York Mental Hygiene Law: Emergency Assessments of Mental Health Crises
Last year, the New York City Police Department responded to close to 200,000 mental health crisis calls, close to a doubling of calls over the last decade. As many of these individuals will present to the Emergency Department (ED), the magnitude of the issue is significant and a major increase in complex patients. The New York State Mental Hygiene Law (MHL) addresses the evaluation and treatment of persons who pose a danger to themselves or others, stating that a person can be admitted involuntarily if they were considered a likelihood to result in serious harm to themselves or others. Under Article 9, the law was revised in May 2025 to clarify its scope for all participants, including patients, police, EMS, and physicians. Specifically, it expanded the definition of “likelihood to result in serious harm” to a substantial risk of physical harm to the person due to inability or refusal, as a result of their mental illness, to provide their own essential needs such as food, clothing, necessary medical care, personal safety, or shelter. Given that many of us apply aspects of the law daily, it is important for us as Emergency Physicians (EP) to understand the law, its limitations, and the duty of the EP.
The first laws in New York governing the admission of persons with mental health concerns were passed in the 1960s and have been continually revised over the following decades. Until the most recent change in May 2025, the law allowed for removal from their residency and hospital admission only for individuals who pose a substantial risk of physical, bodily harm to themselves or others. In practice, persons who engaged in broader self-harmful behavior were sometimes admitted, occasionally leading to lawsuits by both affected individuals and advocacy organizations.
New York Appellate courts found that patients who were unable to meet their basic living needs met the criteria of posing a danger to themselves. In May 2025, the NY State statue was updated to reflect the law’s usage and these rulings. Now, an “inability or refusal” to provide for basic needs is explicitly included. Furthermore, the person does not need to have engaged in a recent dangerous act. Evaluating psychiatrists can use a patient’s history including past behavior and actions to determine the need for admission, known as the “Hogue Standard.”
These changes can directly affect our work in the ED. Patients in mental health crises are frequently brought to the ED, and evaluation should include consideration of whether the patient poses broader risks than imminent physical harm. Examples provided in the August 2025 guidance document from the Office of Mental Health (OMH) includes a patient with psychosis who refuses to plan how to obtain food and shelter, and a patient who is acting aggressively and could be a victim or perpetrator of an assault. The examples include evaluating for a history of psychiatric hospitalizations and abnormal behavior. They illustrate how patients that are without direct intent to harm themselves or others may still be appropriate candidates for involuntary psychiatric admission.
Finally, it is important that documentation reflects the consideration of these principles. Ideally, the medical record would include a direct explanation of the reasoning by which the patient’s condition includes “neglect or refusal to care for themselves which presents a real threat of substantial harm to their well-being.” Emergency departments should socialize the new changes and the OMH guideline publication with interpretation of this change. The examples provided in the OMH guidelines provide clear scenarios to involve psychiatry for potential admissions with expanded definitions.
This recent update to the MHL clarifies our responsibilities and the circumstances in which patients with mental health crises may warrant admission. With this awareness, we can more effectively ensure that our patients get the care they need within the appropriate legal framework.
References:
- Sullivan A. Guidance for the Involuntary and Custodial Transportation of Individuals for Emergency Assessments and for Emergency and Involuntary Inpatient Psychiatric Admissions. Published online August 7, 2025. https://omh.ny.gov/omhweb/guidance/interpretative-guidance-involuntary-emergency-admissions.pdf
- Mental Hygiene Law – Admissions Process. Accessed October 27, 2025. https://omh.ny.gov/omhweb/forensic/manual/html/mhl_admissions.htm
- Tucker CT Emma. New York City directive to potentially involuntarily commit someone suffering a mental health crisis can proceed, court rules. CNN. February 4, 2023. Accessed October 27, 2025. https://www.cnn.com/2023/02/04/us/new-york-city-involuntary-commitment-directive
- NYC Public Advocate Calls for Expanding Non-police Response to Mental Health Calls at Hearing on B-HEARD. Accessed October 27, 2025. https://advocate.nyc.gov/press/nyc-public-advocate-calls-for-expanding-nonpolice-response-to-mental-health-calls-at-hearing-on-bheard
- Smith GB. As 911 calls reporting emotionally disturbed people doubled in a decade, training lags and a key response team has been kept out of the loop. THE CITY – NYC News. March 21, 2019. Accessed October 27, 2025. https://www.thecity.nyc/2019/03/21/the-nypd-s-mental-illness-response-breakdown/
- New York City Releases Its First Ever State of Mental Health Report – NYC Health. Accessed October 27, 2025. https://www.nyc.gov/site/doh/about/press/pr2024/nyc-releases-first-ever-state-of-mental-health-report.page
- New York (State). Laws of the State of New York Passed at the Sessions of the Legislature. publisher not identified; 1777. Accessed October 27, 2025. https://catalog.hathitrust.org/Record/003020495
- Legislation: NYS Open Legislation | NYSenate.gov. Available at: https://www.nysenate.gov/legislation/laws/MHY/9.39 Accessed: 29 October 2025.
- Interpretative guidance involuntary emergency admissions. Available at: https://omh.ny.gov/omhweb/guidance/interpretative-guidance-involuntary-emergency-admissions.pdf Accessed: 29 October 2025.
Emergency Medicine Resident Committee
Blake A. Peterson, MD MS
Chair, Emergency Medicine Resident Committee
PGY-3
University at Buffalo
Reflections on Residency Interviews
It’s residency interview season! Medical student applicants have diligently prepared their applications to reflect the culmination of their life’s work and experiences to date – and it’s time to get to know the people behind the applications. As I get ready to assist my own program with their interview process, I am reminded of what it was like on the other side as a fourth-year medical student trying to land a spot in a residency program that would prepare me for a fulfilling career in emergency medicine.
In the current post-pandemic format, most programs are still electing to interview applicants over video conferencing software. This has been shown to “level the playing field” for more economically disadvantaged students that otherwise would not have the financial means necessary to take on the burden of making numerous trips – some across the country – for a shot at attending a program. Although this has clear benefits, the “old” interview process also had benefits. Going to each prospective hospital site in person gives a clear sense of what each program has to offer in terms of training, physical plant and camaraderie among residents – something that cannot be fully replicated on a computer screen over a “Zoom Happy Hour”, program website or a virtual interview.
“Away” or “Audition” rotations remain the best way for you to get to know a program and for a program to get to know you. Being able to interact with a program’s faculty, observing their relationships with their residents and observing the opportunities for clinical training in practice is critical to see if a program would be a good fit. Some of the best advice I received during interview season was “you are interviewing programs just as much as programs are interviewing you”! While visiting, immerse yourself in the program and city; attend journal clubs, attend in-person meet ups with residents or other rotators and take the opportunity to look at as many of the facilities that the program has to offer as possible.
When the actual interview day came, it was essential to know the program inside and out. Program details are publicly listed online, residents were more than willing to share their own experiences during virtual meet-ups and reaching out to current residents to get a feel for the program ahead of time was an essential part of the preparation for interviews. If programs share who will be interviewing applicants, it is easy to find faculty pages that list professional interests – something that will be fun to discuss for both parties. Just as you are preparing for the interview by learning about a program, programs are preparing by learning about you as well. They will know your application well and will be eager to ask specific questions about application points that stand out to them.
Regardless if you are an interviewer or interviewee this season, a lot of work goes into the process of preparing to select the next cohort of residents. This process is paramount for not only the incoming class, but also the program to train the next generation of emergency medicine residents – the ones who will treat any patient, for any complaint, at any time.
Education
Sophia Lin, MD FPD-AEMUS
Assistant Professor of Clinical Emergency Medicine and Clinical Pediatrics
Director of Emergency Ultrasound
Department of Emergency Medicine
Weill Cornell Medicine
Timothy W. Swingle, MD
PGY-1 Emergency Medicine Resident
Penn State College of Medicine
Miles Gordon, MD FPD-AEMUS
Assistant Professor of Emergency Medicine
Columbia University Vagelos College of Physicians and Surgeons
Thomas M. Kennedy, MD FAAP FACEP
Assistant Professor of Pediatrics in Emergency Medicine
Columbia University Vagelos College of Physicians and Surgeons
The use of artificial intelligence (AI) has become increasingly more prominent in multiple domains, including medical education. This issue’s column describes the use of AI in medical student POCUS education.
Artificial Intelligence for Point-of-Care Ultrasound Training in Undergraduate Medical Education
Background
Point-of-care ultrasound (POCUS) is a valuable tool for expediting patient care in the emergency department (ED) and widely used in emergency medicine (EM). As a result, POCUS education in both graduate and undergraduate medical education is important in preparing trainees for EM practice.
At the Columbia University Vagelos College of Physicians and Surgeons (VP&S), the Department of EM offers a POCUS elective to fourth-year medical students. This course provides opportunities for the repetition and practice needed to develop ultrasound skills through both supervised scan shifts and independent scan shifts. With independent scan shifts, though, students often struggle to recognize target structures and optimize their ultrasound views when an attending physician is not available to provide immediate feedback. Ultrasound is highly operator-dependent with a learning curve that is dependent on repetition and practice. Jang et al.1 showed that for novice sonographers performing focused assessment with sonography for trauma (FAST) exams, up to 25% of their first 10 studies are uninterpretable, with that percentage falling to less than 5% by their 50th exam.
Several ultrasound system manufacturers have created applications utilizing artificial intelligence (AI) to aid novices with image acquisition by labeling anatomy and guiding learners with image adjustment. We utilized one of these applications, ScanLab, during our fourth-year medical student POCUS elective and assessed its impact on student confidence and structure identification. ScanLab identifies and labels structures, such as cardiac chambers, valves and major organs, on-screen and in real-time.
Quality Improvement Project
At the beginning of the POCUS elective, ultrasound faculty provided lectures on several POCUS applications, including the FAST and focused cardiac ultrasound (FOCUS) exams. After the lecture, students were assigned to use ScanLab for either the right upper quadrant (RUQ) view of the FAST exam or for the four views of the FOCUS exam for their first scans on a patient.
As some of our elective students had little to no POCUS experience, they were allowed guidance in initial probe placement and orientation. Once a target organ (the liver or kidney for the FAST RUQ view or the heart for the FOCUS exam) was visualized, faculty provided no other guidance.
After scanning, students reported their confidence and ability to recognize organs and structures in the POCUS views they acquired. Students were asked about their confidence across three domains using a 1-5 point Likert-like scale: 1) adequate view was acquired, 2) structures were correctly identified and 3) adequate view would be obtained with the next attempt. Students were also asked to identify 14 structures in the views that they had obtained, including the liver, right kidney and cardiac chambers.
Impact
Eleven students participated in this project and acquired 54 ultrasound images. Students’ mean confidence improved when they used ScanLab for the FOCUS exam (3.6 to 4, p=0.019) but did not improve when they used ScanLab for the RUQ FAST view (3.92 to 3.5, p=0.55). When students used ScanLab they correctly identified a mean of 12.5 structures and when they did not use ScanLab they correctly identified a mean of 13 structures (p=0.71), out of a total of 14 possible structures. These results are not statistically significant. With only 11 students participating in this project, our sample size may have been underpowered to detect a true effect. Also, students did not use ScanLab during the entire elective and this may have limited the impact that ScanLab had on their training.
Discussion
While the results of our project were not statistically significant, we believe AI-generated POCUS education applications may have utility in helping novice learners develop their POCUS skills. During this project, students reported they appreciated the real-time labeling of the structures when using ScanLab, despite this perspective not being reflected in the confidence scores they provided.
Despite these results, we believe AI ultrasound learning tools such as ScanLab merit further investigation. A known barrier to ultrasound education at multiple levels, including the medical student level, is the availability of trained faculty.2 AI may play an important role in mitigating this barrier. AI POCUS education applications can provide learners with real-time guidance when faculty are unavailable, making POCUS education more efficient and effective. For novices, there are AI applications that provide users with instructions in real-time on how to position the ultrasound probe to acquire and optimize views. For more advanced learners, there are AI applications that aid users with image interpretation. The ways in which AI can improve ultrasound education are growing and the potential for AI to improve and transform ultrasound education is tremendous.
References
- Jang T, Kryder G, Sineff S, Naunheim R, Aubin C, Kaji AH. The technical errors of physicians learning to perform focused assessment with sonography in trauma. Acad Emerg Med. 2012 Jan;19(1):98-101.
- Russell FM, Zakeri B, Herbert A, Ferre RM, Leiser A, Wallach PM. The state of point-of-care ultrasound training in undergraduate medical education: findings from a national survey. Acad Med. 2022 May;97(5):723-727.
Membership Engagement and Development
Moshe Weizberg, MD FACEP
Medical Director, Emergency Department
Maimonides Midwood Community Hospital
Chair, New York ACEP Professional Development Committee
Interviewer
Lukasz Cygan, DO FACEP
Assistant Residency Director, Emergency Medicine
NYP Brooklyn Methodist Hospital
Interviewee
Gerardo Chiricolo, MD
Chair, Department of Emergency Medicine
Robert Wood Johnson Barnabas Community Health Center
Chief Medical Officer, New York Red Bull
Navigating Career Opportunities in Emergency Medicine and the Future of Emergency Medicine
I had the honor of interviewing Dr. Gerardo Chiricolo. Dr. Chiricolo is the current chair of the Department of EM and interim Chief Medical Officer at RWJ Barnabas Community Health Center. He is also the Chief Medical Officer of the New York Red Bulls, a franchise in Major League Soccer. We spoke about fellowship training, career trajectories in Emergency Medicine and the future of the field.
LC:
Thank you for joining us today, Dr. Chiricolo. Did you always envision a career in administration, education or leadership roles in emergency medicine?
GC:
No, I didn’t initially have a set path towards academia or administration. It wasn’t a pre-determined goal. However, as I progressed through medical school and residency, I began to recognize my natural strengths in leadership and organization. During my internship year, I found myself gravitating toward leadership roles among my co-residents. This culminated in receiving the “Intern of the Year” award, which naturally led to my role as Chief Resident. Managing schedules, addressing professionalism issues and facilitating communication between residents and faculty were all pivotal experiences. Those responsibilities and the administrative meetings I participated in, shaped the trajectory of my career.
LC:
It sounds like your journey into leadership within emergency medicine was organic and driven by opportunity. Reflecting on your time as an attending physician or faculty, were there any key turning points or experiences that further solidified your administrative path?
GC:
Absolutely. One of the key moments for me was completing my ultrasound fellowship. Upon finishing the program, my fellowship director moved on to a new opportunity. This created an immediate opening for me and I was fortunate to step into the role of Associate Director for the Ultrasound Division and Fellowship Program. It was an incredibly fortunate timing and this leadership position really accelerated my growth in both clinical and administrative spheres. That experience gave me the foundation to pursue further administrative opportunities within the field.
LC:
It’s a great example of how timing and luck play a significant role in career development. Now, on the topic of fellowship, it’s clear that your ultrasound fellowship played a pivotal role in advancing your career. Ultrasound fellowship is one of the largest in Emergency Medicine. Recently, there has been discussion about extending emergency medicine training to 48 months, which could potentially impact fellowship opportunities. How do you think it will affect the future of ultrasound training?
GC:
Expanding residency training to a fourth year would impose a considerable burden on residents, both financially and personally. The prospect of delaying a physician’s career by an additional year at a resident’s salary could deter many from pursuing fellowship training, especially for those who are eager to start their careers as attending physicians. The decision to take on a fellowship would require a deep commitment to the subspecialty and a clear vision for one’s future.
While the quality of fellowship applicants may increase, I expect that the number of applicants will decrease overall. With the increasing number of ultrasound fellowship programs available, I foresee some programs struggling to match residents and in some cases, potentially even closing down.
LC:
That’s an insightful perspective. Given these changes and knowing your passion for ultrasonography, what would you say to residents considering a fellowship, especially in ultrasound, in light of the proposed extended training?
GC:
I think this applies not just to ultrasound, but to all fellowships. In today’s competitive market, it’s essential to bring something unique to the table. Fellowship training offers a specialized skill set that provides significant value to any department or institution. Whether in an academic or clinical setting, individuals who complete a fellowship bring a higher level of expertise that enhances patient care, pushes the boundaries of innovation and drives research.
Moreover, a fellowship enables professionals to become leaders in their niche, training others and advancing knowledge in their field. While others may gain a superficial understanding of certain skills, fellowship-trained specialists are able to master them and contribute at a higher level. We will always need these experts to help guide and mentor future generations.
Beyond the clinical skills I developed, fellowship provided an invaluable professional network. The relationships I built with colleagues across the country—individuals I would never have had the chance to meet otherwise—have been essential. This network continues to be a resource for collaboration, idea exchange and support. Fellowship truly enhances not only your clinical expertise but also your professional network, which is critical for future growth and innovation.
LC:
Networking and building connections is often one of the unspoken benefits of fellowship. Moving on to another hat that you wear, as Chief Medical Officer for the New York Red Bulls: You’re now a chair after 15 years of advancing through the administrative ranks. How did you get involved in sports medicine, especially since it’s not your primary expertise and you weren’t fellowship trained in it?
GC:
Yeah, it’s kind of a strange and random story. I went to medical school with the goal of becoming a sports medicine physician as I was an athlete in multiple sports. But then, I discovered emergency medicine and completely fell in love with it. During medical school, I had a lot of access to urgent and emergent care, which really drew me in.
I was approached by a sports medicine physician who was working with a professional organization and they needed a contact at a nearby emergency department in case players or staff got injured. When this opportunity arose, I made sure I was available, open and collaborative. It led to an invitation to attend a game and then to more invitations. Eventually, I was asked to fill in when the physician was on vacation. From there, I built a strong relationship in the sports world, which led to me becoming the Chief Medical Officer and Head Team Physician for the New York Red Bulls, which was pretty surreal—especially since they’re my favorite team and it’s in my home state. And it’s a great organization. Another example of taking advantage of luck and timing.
LC:
You mentioned saying “yes” to a lot of things along the way. Do you think young faculty should always say yes to every opportunity or are there times when they should say no? And if so, when is it okay to say no?
GC:
I’m a strong believer in saying yes whenever you can. Opportunities lead to more opportunities and saying yes shows you’re a team player and helps you grow. It makes you well-rounded and teaches you a lot along the way.
However, there are times when it’s important to say no. If I feel like I can’t make a positive impact or I don’t have the time or expertise to succeed in a certain role, it’s better to decline. Saying no in those situations is actually a win-win because it gives someone else a chance to step up who may be better suited for the role. It also lets you focus on your existing commitments and prevents you from spreading yourself too thin or setting yourself up to fail.
You really want to avoid the situation where you’re taking on too much and not doing anything well. That’s never a good look.
LC:
In terms of emergency medicine (EM) administration and hospital administration, what are some advances or trends that young faculty should be aware of in the next 5 years?
GC:
A big trend is the consolidation of healthcare systems. There’s a focus on doing more with fewer resources—reimbursements are going down and hospitals and emergency departments are being held to more stringent metrics like throughput times and quality indicators.
The shift towards outpatient and value-based care is growing, which means things like hospital-at-home, telemedicine and observation medicine are going to play a bigger role. EM needs to be at the table during these changes because if we’re not involved in shaping how these new models work, we might get left behind.
One of the biggest areas of focus will be how we adapt to decreased inpatient care while still ensuring we can provide high-quality care for patients in the outpatient setting. For example, using telemedicine not just for initial visits but also to help prevent readmissions, which is a huge priority for most institutions.
We also need to be prepared for an increase in uncompensated care. Emergency departments are the safety net for many people and I take pride in serving our community as that safety net. But and with rising rates of uncompensated care, it’s important that we’re ready for this challenge and have the resources to address it.
LC:
You mentioned doing more with fewer resources. What do you think about the future job market for emergency medicine physicians, especially given the forecasted supply-and-demand issues by 2030?
GC:
That’s a loaded question, but I think it’s important to recognize that moral injury (or burnout, as some call it) is very real in emergency medicine. If you look around, you’ll see many emergency physicians branching out into fields like sports medicine, wellness, concierge medicine, urgent care and even industry roles.
The good news is that emergency medicine physicians have a lot of transferable skills. Our broad experience in handling diverse, high-pressure situations makes us very marketable. I think many physicians will start exploring these alternative paths earlier in their careers and as a result, newer graduates will have a better idea of the marketplace.
As for the job market, I think the major metropolitan areas will continue to have a strong demand for EM physicians. Rural areas have always faced staffing challenges and I think this trend could actually help them, as more physicians may be willing to work in those settings.
While we should definitely monitor these trends, I’m not sure it’ll turn into the catastrophic problem some people are predicting.
LC:
Any final thoughts or advice for young faculty?
GC:
Well, if you’re asking for advice for young faculty, I’d say that you’re going to be in great shape if you stay open to opportunities, collaborate and continue to learn. There’s so much potential in emergency medicine and you have the power to shape its future.
Pediatrics
Maria Tama, MD
Assistant Professor of Pediatrics in Emergency Medicine
Zucker School of Medicine at Hofstra/Northwell
Director, Division of Emergency Ultrasound
Staten Island University Hospital- Northwell Health
Shawn A. Haupt, MD MS FAAP
PGY-6, Pediatric Emergency Medicine Fellow
Good Samaritan University Hospital
Cornelia Muntean, MD FAAP
Director, Pediatric Emergency
Program Director, Pediatric Emergency Medicine Fellowship
Good Samaritan University Hospital
Introduction
In the emergency department (ED), meningitis, encephalitis and meningoencephalitis are high-stakes, time-dependent syndromes where the provider must quickly identify the red flags and adopt a “treat now, clarify later” mindset. As in adults, blood cultures should be obtained, meningeal-dose antibiotics ± acyclovir should be started promptly, and a lumbar puncture (LP) should be performed as soon as it is safe to do so.
Definitions
Meningitis is inflammation of the meninges- the protective membranes covering the brain and spinal cord and is typically caused by bacterial or viral infections.1-2 It typically presents with fever, neck stiffness, photophobia and headache, but this classic triad is unreliable in infants and younger children where the presentations can be non-specific.
Encephalitis refers to inflammation of the brain parenchyma with neurological dysfunction that is typically caused by infectious vs autoimmune causes. It can present with fever and headache with progression to altered mental status (AMS), seizures, behavioral changes and/or psychosis.3 According to the International Encephalitis Consortium, the AMS must persist ≥24 hours without an alternative explanation.4
Meningoencephalitis refers to inflammation of both the meninges and brain parenchyma, with overlapping symptoms of both meningitis and encephalitis. Patients may present with fevers, neck stiffness, AMS, seizures or focal neurological deficits. The distinction between the two may not be clinically apparent at presentation so it is paramount to treat broadly.3
Approach to History & Physical
The assessment of a pediatric patient should begin immediately to avoid delays in care. It is important to remember that the clinical symptoms and signs of meningitis, encephalitis and meningoencephalitis vary depending on the age of the child and duration of disease. Non-specific symptoms may include fever, tachycardia, irritability, lethargy and vomiting.
The red flags for pediatric patients depend on age:
Infants
Fever, hypothermia, poor feeding, irritability/lethargy, bulging fontanelles, hypotonia, weak or high-pitched cry, apnea and/or seizures
Older Children and Teenagers
Fever, headache, neck stiffness, photophobia, vomiting, petechial and purpuric rashes), focal deficits and/or seizure.
When examining a patient, consider their overall appearance, including their mental status. Dig deeper to identify the presence of any focal neurologic signs and perform an extensive skin exam to identify any rashes such as petechiae and purpura. Classic physical exam findings of meningitis may include nuchal rigidity, photophobia, bulging fontanelles and a positive Kernig or Brudzinski’s sign (which is more common in children older than 12 months of age). However, these should be interpreted with caution as patients with meningitis may not have these classic findings.1,3,5
Initial ED Workup
In the ED, workup should start with obtaining labs such as a complete blood count (CBC), basic or complete metabolic panel and a blood culture. Blood cultures must be obtained before antibiotics if feasible; however, if a chil is unstable, do not delay antibiotics to get blood cultures. Consider a serum or urine drug screen, CRP and/or procalcitonin (as supportive markers).5-6
Chest X-rays should be obtained as pneumonia has been described to concurrently occur in up to 22% of patients with pneumococcal meningitis.7
If there are no signs of increased ICP or focal neurological deficits, a lumbar puncture (LP) can be performed without routine head computed tomography (CT). Obtain a CT first if there are any signs of focal neurological deficits, altered mental status, Cushing’s triad, seizures, papilledema, or known mass lesions.
After obtaining cerebrospinal fluid (CSF) from an LP, order your cell count with differential, protein, glucose, gram stain, CSF culture and a HSV PCR.5 When available, consider ordering a multiplex meningitis/encephalitis (ME) PCR panel to rapidly identify common pathogens (especially when encephalitis is strongly suspected or there is CSF pleocytosis), which may reduce the time to targeted therapy and length of stay for the patient.5,8
Empiric Treatment
Start antimicrobials as soon as meningitis, encephalitis or meningoencephalitis is suspected and give all doses of medications at meningeal doses.9-11 Consider adjunctive dexamethasone in children ≥6 weeks of age before or with the first antibiotic dose if Hib or pneumococcal meningitis is suspected, to reduce the risk of hearing loss or mortality.12
Cases
Case 1
A 10-year-old boy with no prior past medical history presents to the ED with 24 hours of fever (as high as 104F), headache, emesis, leg pain and a rapidly spreading petechial rash. He is tachycardic with a heart rate of 140 bpm and hypotensive to 75/36 mmHg with exam findings consistent with poor perfusion. He is placed in droplet precautions. IV access is immediately established and the patient is given IV fluid resuscitation. Blood cultures and other labs (including lactate, point of care glucose, etc) are obtained and Ceftriaxone plus Vancomycin are immediately given. The patient required an epinephrine infusion to maintain perfusion. An LP is deferred until his clinical status improves.
Case 2
A 3-month-old girl, born full term with no past medical history, presents to the ED with a fever as high as 103F and what parents describe as a focal seizure. She has been post-ictal for 2 hours and still has not returned to baseline. Examination reveals no spontaneous movement of her right lower leg and a bulging fontanelle. The patients’ heart rate is 165 bpm and a normal blood pressure of 85/65 mmHg. Blood cultures and other labs are obtained, IV access is established and cefotaxime, vancomycin and acyclovir are given. The patient then gets an LP after imaging.
Case Discussions
Case 1:
The patient’s most likely diagnosis is meningococcemia, which constitutes 20% of meningococcal meningitis cases. Meningococcemia often causes disseminated intravascular coagulation (DIC) which can lead to the characteristic petechial rash on the extremities of the body and purpura. If not intervened on early, meningococcemia can progress to Waterhouse-Friedrichsen syndrome, a rare but life-threatening complication.13
This case underscores the importance of immediate empiric antimicrobial therapy in the setting of shock physiology and rapid disease progress. We stress the importance of early initiation of antibiotics in the unstable child.
The patient is admitted to the pediatric intensive care unit (PICU). The public health department is contacted to facilitate close contacts and any close contacts with the patient are given chemoprophylaxis with rifampin, ceftriaxone or ciprofloxacin.9
Case 2:
The patient’s most likely diagnosis is HSV meningoencephalitis. Her case underscores the importance of NOT anchoring on simple febrile seizure when red flags such as a focal seizure or prolonged post-ictal state are present. In her case, acyclovir is prudent in addition to an empiric anti-microbial while awaiting PCR results from her CSF studies.
She is admitted to the PICU for further testing and management.
Conclusion
When facing a possible case of meningitis, encephalitis, meningoencephalitis, early recognition and intervention is key to preventing significant morbidity and mortality in pediatric patients. Obtain blood cultures before antibiotics (when safe), initiate immediate meningeal-dose antibiotics and acyclovir when indicated and perform an LP as soon as it is safe for the patient to do so. Provide chemoprophylaxis and notify the local public health department as soon as possible if meningococcal disease is suspected.
References
- Zainel A, Mitchell H, Sadarangani M. Bacterial Meningitis in Children: Neurological Complications, Associated Risk Factors, and Prevention. Microorganisms. 2021;9(3):535. Published 2021 Mar 5. doi:10.3390/microorganisms9030535
- Mijovic H, Sadarangani M. To LP or not to LP? Identifying the Etiology of Pediatric Meningitis. Pediatr Infect Dis J. 2019 Jun;38(6S Suppl 1):S39-S42. doi: 10.1097/INF.0000000000002313. PMID: 31205243.
- Messacar K, Fischer M, Dominguez SR, Tyler KL, Abzug MJ. Encephalitis in US Children. Infect Dis Clin North Am. 2018;32(1):145-162. doi:10.1016/j.idc.2017.10.007
- Venkatesan A, Tunkel AR, Bloch KC, Lauring AS, Sejvar J, Bitnun A, Stahl JP, Mailles A, Drebot M, Rupprecht CE, Yoder J, Cope JR, Wilson MR, Whitley RJ, Sullivan J, Granerod J, Jones C, Eastwood K, Ward KN, Durrheim DN, Solbrig MV, Guo-Dong L, Glaser CA; International Encephalitis Consortium. Case definitions, diagnostic algorithms, and priorities in encephalitis: consensus statement of the international encephalitis consortium. Clin Infect Dis. 2013 Oct;57(8):1114-28. doi: 10.1093/cid/cit458. Epub 2013 Jul 15. PMID: 23861361; PMCID: PMC3783060.
- Skar G, Flannigan L, Latch R, Snowden J. Meningitis in Children: Still a Can’t-Miss Diagnosis. Pediatr Rev. 2024 Jun 1;45(6):305-315. doi: 10.1542/pir.2023-006013. PMID: 38821894.
- American Academy of Pediatrics. Meningitis. Pediatric Care Online. 2024. doi:10.1542/aap.ppcqr.396196
- Figueiredo AHA, Brouwer MC, Bijlsma MW, van der Ende A, van de Beek D. Community-acquired pneumonia in patients with bacterial meningitis: a prospective nationwide cohort study. Clin Microbiol Infect. 2020 Apr;26(4):513.e7-513.e11. doi: 10.1016/j.cmi.2019.09.001. Epub 2019 Sep 13. PMID: 31525519
- McDonald D, Gagliardo C, Chiu S, Di Pentima MC. Impact of a Rapid Diagnostic Meningitis/Encephalitis Panel on Antimicrobial Use and Clinical Outcomes in Children. Antibiotics. 2020; 9(11):822. https://doi.org/10.3390/antibiotics9110822
- American Academy of Pediatrics. Red Book: 2024–2027 Report of the Committee on Infectious Diseases. 33rd ed. Itasca, IL: American Academy of Pediatrics; 2024.
- Taketomo CK, Hodding JH, Kraus DM. Pediatric & Neonatal Dosage Handbook. 29th ed. Hudson, OH: Lexicomp; 2022.
- Tunkel AR, Hasbun R, Bhimraj A, et al. 2017 Infectious Diseases Society of America’s Clinical Practice Guidelines for Healthcare-Associated Ventriculitis and Meningitis. Clin Infect Dis. 2017;64(6):e34-e65. doi:10.1093/cid/ciw861
- Brouwer MC, McIntyre P, Prasad K, van de Beek D. Corticosteroids for acute bacterial meningitis. Cochrane Database Syst Rev. 2015 Sep 12;2015(9):CD004405. doi: 10.1002/14651858.CD004405.pub5. PMID: 26362566; PMCID: PMC6491272.
- Stephens DS, Greenwood B, Brandtzaeg P. Epidemic meningitis, meningococcaemia, and Neisseria meningitidis. Lancet. 2007;369(9580):2196-2210. doi:10.1016/S0140-6736(07)61016-2
- Centers for Disease Control and Prevention. Manual for the Surveillance of Vaccine-Preventable Diseases: Chapter 8 – Meningococcal Disease. Accessed September 26, 2025. Available at: https://www.cdc.gov/surv-manual/php/table-of-contents/chapter-8-meningococcal-disease.html
