Child Health Needs and the Pediatric Emergency Medicine Workforce: 2020–2040

Children facing health emergencies require treatment by physicians trained in the unique health care needs of such children. In the United States, approximately 30 million children visit an ED annually.¹  Before the coronavirus disease 2019 pandemic, pediatric ED visits accounted for ∼25% of all US ED visits, of which 40% were children aged 0 to 5 years. Approximately 5% of patients presenting to EDs required hospital admission.⁴  During the pandemic, overall ED visits declined by 45.7%.⁵  Simultaneously, the proportion of children aged 15 to 18 years and those with chronic conditions seeking emergency care increased from 10.3% to 12.9% and 23.7% to 27.8%, respectively.⁵ 

As pediatric ED visits return to prepandemic levels, the leading causes of pediatric morbidity and mortality continue to be motor vehicle collisions and firearm-related injuries.⁶  Minor illnesses and trauma are also commonly seen in EDs and other urgent care settings. Furthermore, there has been a nearly 45% increase in ED visits for pediatric mental health, with an almost 300% increase in visits for attempted suicide or self-inflicted injury.^7,8 

Almost 85% of pediatric patients are seen in general EDs where adults are cared for, often by non-PEM board-certified physicians.^9–13  Children treated in general EDs are more likely to be discharged without prescriptions, be transferred, or have caregivers leave against medical advice compared with pediatric EDs.¹⁴  Pediatric patients presenting with mental health concerns are also more likely to have longer lengths of stay, increased costs associated with the visit, and be transferred to another facility.^7,8,15,16  Better outcomes have been demonstrated for children cared for in pediatric EDs compared with general EDs, including increased survival from nontraumatic out-of-hospital cardiac arrests, decreased use of ionizing radiation imaging studies, and improved care of febrile neonates and children with asthma.^17–21  These differences in clinical outcomes show that PEM plays a critical role in treating pediatric patients.

PEM is a relatively new pediatric subspecialty. In the mid 1980s, the first federal funding was allocated to improve the emergency care of children through Emergency Medical Services for Children and the National Pediatric Trauma Registry.^22–24  Data from these organizations identified unique aspects of care for pediatric patients.

Pediatricians working in general EDs were also concerned about the scope of work and education and training related to PEM. PEM-specific professional organizations were formed within the American Academy of Pediatrics (AAP), the American College of Emergency Physicians, the Emergency Nurses Association, and the National Association of Emergency Medical Services Physicians.²⁵  The first pediatric emergency care textbook was published in 1983, and the first dedicated journal was launched in 1985.²⁵  The first PEM fellowship was established in 1980, and in 1991, PEM fellowships became accredited by the Accreditation Council for Graduate Medical Education (ACGME).^25,26  In 1992, the American Board of Pediatrics (ABP) administered the first PEM certifying exam.²⁵  Two pathways were developed to grant board certification in PEM: (1) completion of a pediatrics residency followed by a PEM fellowship, or (2) completion of an emergency medicine (EM) residency followed by a PEM fellowship.²⁷  To date, the ABP develops and administers the PEM Initial Board Certification Exam and PEM Maintenance of Certification. The American Board of EM (ABEM), which certifies individuals training in EM, is responsible for notifying its candidates of test results for PEM-certifying board exams and for certifying its diplomates.

On the basis of ABP data through June 2023, a total of 3493 pediatricians have ever been board-certified in PEM, of which 86.4% were actively maintaining certification.²⁸  Certifications in EM granted by ABEM are only covered in some portions of this report but accounted for another 382 subspecialists as of June 2022. A recent cross-sectional study using the American Medical Association Physician Masterfile database linked to American Board of Medical Specialties board certification information showed that 2403 clinically active physicians self-identified as treating pediatric patients in EDs, of which 68% were PEM board-certified; 74% reported EM residency training and 24% reported pediatric residency training.²⁹  Because pediatric emergency care may be provided by physicians other than those certified by the ABP and ABEM, data on the total number of board-certified physicians providing PEM care are only sometimes available across multiple variables as discussed below.

The ABP data on currently certified PEM subspecialists include individuals who may not be in the workforce because of recent retirement, death, or other factors. Consequently, the current workforce descriptions are limited to 2972 certified pediatricians aged ≤70 years. Of these, 60.8% identified as female and 39.2% as male (the ABP has only offered alternative options since 2021).²⁸  The median age was 46 years, 14.9% of which were aged 61 to 70 years.²⁸  Those with PEM board certification were older than those who treat pediatric patients but only had general EM board certification, neither general nor PEM board certification, or without any board certification.²⁹  Regarding medical training, 75.7% were American medical graduates (AMGs) with a Doctor of Medicine (MD), 7.3% were AMGs with a Doctorate of Osteopathy (DO), 11.2% were international medical graduates (IMGs) with an MD, and 5.8% were IMGs with an international degree.²⁸  Race and ethnicity estimates from 2018 to 2022 suggest that ∼12.5% of PEM subspecialists self-identified as underrepresented in medicine (URiM), with 4.9% Black/African American, 5.3% Hispanic/Latino/Spanish origin, and 0.1% American Indian/Native Alaskan.³⁰ 

Demographic and work characteristic data of PEM board-certified physicians are collected through the ABP’s Maintenance of Certification enrollment surveys. Surveys from 2018 to 2022 had a 58.6% response rate for PEM, reflecting responses from 1419 eligible subspecialists aged ≤70 years (because of skip patterns in the survey, percentages reported below are for individual questions and not the entire sample). The majority were employed full time (84.3%), of which 24.0% reported working ≥50 hours per week on average over the last 6 months.³¹  Women (19.5%) were more likely to indicate part-time employment status than men (6.7%). Most (79.6%) spent ≥50% of their time in clinical care; only 2.2% reported spending ≥50% of their time in research. One-third of respondents (36.1%) reported working in a nongovernmental hospital or clinic; another 24.6% endorsed working within a medical school or parent university. Most (80.5%) had a faculty appointment. The majority (81.4%) reported that a primary work setting was in an urban environment. Approximately 41.9% reported that ≥50% of their patients received public insurance.³¹  Although these data are helpful, they do not clearly define the number of clinical full-time equivalents available to serve children seeking emergency care.

When the subspecialty workforce is limited to the United States, the average number of PEM subspecialists per US state (range 0–325) is ∼4.4 PEM subspecialists per 100 000 children aged 0–17 years (range 0.0–31.8) (Fig 1).³²  This accounts for estimates across both ABP and ABEM certifications. PEM physicians are variably distributed across the United States, with most concentrated in urban settings.³²  Four states (Wyoming, South Dakota, North Dakota, and Alaska) do not have any PEM board-certified physicians, and 2 states (Montana and Idaho) have PEM board-certified physicians in only 1 county.³²  In 2019, the average driving distance to a certified PEM subspecialist was 24.4 miles; distances ranged from 4.1 miles in Rhode Island to 327.1 miles in North Dakota (excluding Alaska, Hawaii, Puerto Rico, and the District of Columbia).³³  Those physicians treating pediatric patients in EDs in nonurban areas are more likely to have completed medical training >20 years ago, less likely to have PEM board certification, and more likely to be board-certified in another specialty.²⁹ 

Yearly snapshot data from the ACGME showed an increase of 19.7% in United States-accredited programs between the academic years 2012 and 2022, 71 PEM fellowship programs in 2012 to 2013, and 56 PEM fellowship programs (pediatrics) and 29 PEM fellowship programs (EM) in 2021 to 2022.^34,35  Review of the National Resident Matching Program data for first-year fellows from 2012 to 2021 shows, at minimum, 94% of programs filled, with several years of 100% fill rate (2016, 2019, 2021).³⁶  The ABP total count of first-year fellows, incorporating individuals who take positions before or after the Match, demonstrates that the number of first-year PEM fellows over the last decade increased by 39.1%.³⁷  These numbers support the growing popularity of PEM.

Among current PEM fellows (600) in training levels 1 to 3 during the academic year 2022 to 2023 and in standard, noncombined US fellowship programs, 67.3% identified as female and 32.7% as male. Race and ethnicity data from 2022 to 2023 suggest that ∼18% of PEM fellows self-identified as URiM, with 7.2% Black or African American; 8.0% Hispanic, Latino, or Spanish origin; and 0.4% American Indian or Native Alaskan.³⁰  Regarding medical training, 62.5% were AMGs with an MD degree, 15.8% were AMGs with a DO degree, 6.0% were AMGs with an unknown degree, 12.0% were IMGs with an MD degree, and 2.8% were IMGs with an international degree.³⁸  These data suggest that more trainees with URiM backgrounds, AMGs with DOs, and IMGs are entering PEM. Among current and graduated PEM fellows, 90% and 93%, respectively, have completed a pediatric residency before starting a fellowship.³⁴  Fellows’ first postgraduate clinical position is usually near their training location and likely at academic medical centers.^39–41 Figure 1 shows the variability in PEM fellowship locations in academic year 2021 to 2022, highlighting how this can lead to a geographic maldistribution of practicing PEM physicians. The attrition rate has not been estimated broadly in PEM, but a single-center study showed a fellowship attrition rate of <10%.⁴² 

Although recent publications have suggested that increasing debt rates during medical training may impact decision-making, this has not necessarily been reflected in the PEM workforce. For PEM, approximately 47.0% of current fellows owe $200 000 or more compared with 39.5% for all pediatric subspecialty fellows, on the basis of results from the ABP’s Subspecialty In-training Exam Survey in 2022.⁴³  Additionally, previous studies consistently show that PEM generates negative financial returns (−$69 637 over a lifetime) compared with general pediatrics.^44,45  Despite this, financial implications only ranked 10th as a reason for selecting a subspecialty on the AAP Graduating Residents Survey in 2019, and there continues to be growth in the number of PEM fellows and programs.⁴⁶ 

This article is the result of a national collaboration involving the ABP Foundation and others to better define the US pediatric subspecialty workforce and how it might change in response to various factors in the next 2 decades. Data reported here come from the ABP workforce data repository and other national resources incorporated into a microsimulation model.^3,47  The model takes into account changes in the child population at the national and subnational level on the basis of the US Census Bureau⁴⁸ ; differences by subspecialty for census regions are discussed in the summary article in this supplement.⁴⁹  In addition to baseline estimates, alternative scenarios are included that examine the impact of potential changes in fellowship trainee numbers, retirement, and attrition. An interactive, Web-based visualization of the model with projections is publicly available online.⁵⁰ 

The 2 most pertinent scenarios for PEM are changes in the fellowship pathway and in percentage of clinical time. The results below reflect predictions based on permutations of the aforementioned relevant scenarios. Details of scenarios not discussed in detail and estimated 95% confidence intervals are provided in Tables 1 and 2 and the interactive tool. The numbers reported below may differ from those in the previous section because of differences in years (2020 vs 2023), sample selection criteria, and inclusion of self-reported clinical time.

The model predicts that, by 2040, there will be 39 253 US subspecialist pediatricians, of which 4809 (12.3%) will be board-certified PEM physicians; this represents an 84% increase in headcount of board-certified PEM physicians over 20 years or 5.2 PEM physicians per 100 000 children. The clinical workforce equivalent (CWE) for PEM physicians in 2040 is predicted to be 2.8 per 100 000 children.

The model calculates that there will be an increase in board-certified PEM physicians throughout the United States. CWE per 100 000 children is expected to increase across all 4 census regions: West (+83%, CWE = 2.44), Midwest (+77%, CWE = 2.72), South (+56%, CWE = 2.41), and Northeast (+52%, CWE = 4.44) (Fig 2). Similar trends in CWE per 100 000 children are seen among the 9 census divisions, the largest divisions being in Mountain (+90%), West South Central (+87%), and East North Central (+82%) (Table 1). The findings are similar when expressed as headcount per 100 000 children (Table 2).

Some of the supply projections of the PEM workforce do not parallel the predicted population growth of children aged 0 to 18 years. For example, the second-highest region of PEM growth will occur in the Midwest; however, the child population in this region is the lowest in the country and is only expected to grow by 2% by 2040. In contrast, the West’s PEM supply projection is the largest (83%); this parallels the children’s population growth projection, which also has the largest projected growth (23%). Although the model accounts for these anticipated changes in population, it does not include measures of population need specific to PEM or market demand for PEM positions.⁴⁷ 

Changes in the number of graduating fellows each year will have different implications for the workforce. A 12.5% decrease in fellows, implemented gradually from 2026, with a 2.5% decrease each year until 2030, would produce little change in the PEM workforce as compared with baseline at census region and census division level (Tables 1 and 2). A 12.5% increase in fellows, starting in 2026, with a 2.5% increase each year until 2030, would result in an overall 69% CWE increase in PEM (2.89 per 100 000 children from a baseline of 1.71), with differing increases across census regions: West (+90%), Midwest (+87%), South (+62%), and Northeast (+59%). These changes reflect an increase of 2 to 3 board-certified PEM physicians per census division from baseline (Table 2).

Changes in the proportion of time spent in direct clinical care would also influence workforce patterns. A 7% reduction in clinical hours worked would result in a slower growth rate of CWE per 100 000 children across all census regions and divisions, with the lowest growth in the New England division (+31%; Table 1). On the other hand, a 7% increase in clinical hours worked starting in 2022 with a 1% increase each year until 2028 would result in an increase in CWE per 100 000 children from baseline across all census regions: West (+96%), Midwest (+90%), South (+67%), Northeast (+62%), and highest in the Mountain division (+104%).

The model predicts there will not be any long-term change if PEM subspecialists exit the workforce 5 years earlier than in the past. However, early exit would create shortages from 2020 to 2030, with the largest decrease in supply occurring in 2024. If rates of midcareer PEM subspecialist exits from the workforce increase, there is essentially no change of supply across time.

Over the next 20 years, PEM is projected to grow for each census region and division. The number of new PEM fellowship programs is increasing, and many existing programs adding additional training spots each year.³⁴  Even if there were a decrease in the number of fellows entering PEM, a decrease in clinical time, and an increase in the exit of physicians at all ranks as indicated in the worst-case scenario, the PEM field would still show a positive (albeit smaller) change from baseline. However, geographic distribution remains uneven. Previous data show that pediatric subspecialty fellows are most likely to practice in the region where they train.^39,40  This is also reflected in the model projections in which the West South Central, West North Central, and Pacific census divisions will have <2.5 board-certified PEM physicians per 100 000 children. In contrast, the already saturated New England and Middle Atlantic census divisions will have nearly double the number of board-certified PEM physicians. At the same time, 4 states (Alaska, North Dakota, Wyoming, and South Dakota) do not have any board-certified PEM physicians, EM residency programs, or PEM fellowship programs. Pediatric patients in these “PEM deserts” will likely be treated by general EM physicians and primary care pediatricians. On the basis of this projected growth, PEM can potentially expand its reach into other communities.

PEM fellowship education may need to include clinical training in community or rural settings to address geographic differences.⁵¹  Training future PEM physicians to work in academic-based community hospitals or nonacademic-affiliated hospitals and establishing PEM fellowship programs in community-based hospitals with general EM departments may be a means to address health care in underserved areas.³⁹  Given the growth of PEM fellowship programs in EM academic departments, PEM/EM board-certified physicians may be more likely to work in an EM department where they can see both adults and children. These individuals are well positioned to serve in rural areas. However, if rural areas continue to be served only by general EM physicians, pediatric-specific training requirements in EM should be enhanced. For example, pediatric mental health training is not currently an ACGME requirement for EM.⁵²  Addressing these gaps in EM residency training programs may better prepare frontline physicians for treating children.

At the same time, acute pediatric mental health care education should also be emphasized during PEM fellowship. The United States is currently facing a pediatric mental health care crisis.^7,8  The nearly 50% increase in ED visits for pediatric mental health is reflected in pediatric hospitals, nonchildren’s hospitals, and rural EDs.^7,8,15  Less than half of children’s hospitals have an inpatient psychiatric unit, and nonchildren’s hospitals have even fewer resources. As access to outpatient psychiatric services is greatly limited by high demand, PEM physicians have been and will likely continue to address the significant need for acute mental health care services.^53,54  As such, PEM fellowship education must also target this need. Recent updates of PEM ACGME requirements state that fellows must “provide care that is sensitive to the developmental stage of the patient with common behavioral and mental health issues…and demonstrate[ing] the ability to refer and/or comanage patients with common behavioral and mental health issues.”⁵⁵  Furthermore, although there is no current requirement for telehealth education in PEM fellowship, this could help address acute mental health access needs in rural areas and many other gaps in caring for the acutely ill child.⁵⁵ 

Flexibility in the length of PEM fellowship training may help address the geographic maldistribution of PEM physicians. Funding and accreditation for 2 tracks for PEM training (2-year program for those preferring to work only in a clinical capacity or 3-year program for those desiring a career with traditional academic pursuits) may allow for a greater increase in the number of fellows entering the workforce. The Infectious Diseases Society of America has already recommended 3 adult infectious diseases fellowship tracks (clinician, clinical investigator, basic investigator) to address the various needs of their workforce.⁵⁶  Curricular content for a 2-year PEM fellowship program could include community-based medicine, telehealth, and critical care medicine in resource-poor settings. Furthermore, a decrease in the duration of fellowship training, along with implementing a federal loan repayment program, may reduce the potential negative financial burdens for physicians entering PEM.^43–45 

As a field, PEM may need to provide a bridge for developing urgent care centers. The Society of Pediatric Urgent Care and the AAP Provisional Section of EM Special Interest Group for Urgent Care are newer professional medical organizations supporting urgent care medicine. To date, seven 1-year pediatric urgent care medicine fellowships in the United States focus on evidence-based approaches to acute illness, management of minor trauma and musculoskeletal injuries, procedural training, radiology interpretation, approach to higher-acuity patients, and developing efficiency in patient flow.⁵⁷  Fostering support for such programs, as well as encouraging general pediatrics residency programs to tailor curricula for trainees entering practice in underserved areas, will help address this geographic maldistribution of board-certified PEM physicians.⁵⁸ 

Finally, discussions around the longevity of board-certified PEM physicians’ careers have arisen. Among 42 medical specialties, PEM physicians report experiencing the most satisfaction with their careers.^59,60  Factors that play prominently in career satisfaction (eg, decreased off-hours of clinical administrative load including following up on test results), shift work allowing for work–life balance, and supportive physician colleagues promoting academic success have enabled this field to flourish.⁶¹  PEM physicians also have relatively high earning potential, given the skills needed in critical care resuscitation and the ability to complete procedures, compared with other pediatric subspecialties.⁶²  However, PEM physicians who are clinically active work ∼30 clinical hours per week, 50% work overnight shifts, 97% work on weekends, and 94% work on holidays, leaving nearly two-thirds of PEM physicians questioning “how long can I last?”⁶³  Factors that play into burnout in PEM include lack of appreciation by patients and supervisors, difficult clinical schedules, dissatisfaction with promotion opportunities, and frustration with electronic medical record systems.⁶⁴  These factors have led nearly 46% of PEM physicians to plan to change clinical activity in the next 5 years, including reducing hours, changing shifts, or retiring.⁶⁵  Although the future workforce may differently define what qualifies as a full-time equivalent in PEM, the current model takes into account this potential for burnout and still shows growth of PEM in the coming decades, likely because of the continued increase in the number of PEM fellows.

Emergency Medical Services for Children federal funding may also need to consider overall pediatric readiness for nonpediatric EDs. Given that some clinical outcomes for pediatric patients treated in general EDs are below those seen in pediatric EDs, perhaps there should be federal requirements that, if a non-pediatric ED treats a certain percentage of children annually, then they would be required to have a PEM physician on staff to provide care, serve as a consultant, or be a point person for pediatric education.^7–21  A recent study suggested that national hospital accreditation organizations could adopt similar state-level accreditation practices to ensure pediatric readiness in facilities caring for children.⁶⁶  Furthermore, ED pediatric readiness could be tied to reimbursement given that >60% of children seeking emergency care have public insurance, thus following already established efforts of the Centers for Medicare & Medicaid Services.⁴ 

Although the model shows potentially underserved areas, the true demand for PEM providers in these regions is unclear and should be investigated further. Research is needed to determine how PEM will address reaching geographically underserved areas through educational interventions, educational loan forgiveness/repayment plans and financial incentives, telehealth, and implementing more urgent care and telehealth options. Finally, there have been few studies conducted by PEM board-certified physicians themselves on understanding the workforce, clinical needs, work in community settings, and the definition of CWE. Addressing the importance of research conducted by those in the specialty is paramount to understanding the future workforce needs.^29,67 

PEM will experience growth over the next 2 to 3 decades. Although this growth will bring highly skilled care to millions of children, there will likely be a geographic maldistribution of board-certified PEM physicians. Diligent effort must be placed on PEM fellowship education and EM residency training and on supporting providers in caring for the changing clinical needs presenting to EDs and pediatric patients in underserved areas.

We thank Emily McCartha, Andrew Knapton, Adam L. Turner, Adriana R. Gaona, Crista Gregg, and Diana Gonzalez for their review of the data presented. In addition, we thank Virginia A. Moyer and Patience Leino for their editorial support. Last, we thank the pediatricians who shared their information with the ABP Foundation and made this supplement possible.

AAP

American Academy of Pediatrics

ABEM

American Board of Emergency Medicine

ABP

American Board of Pediatrics

ACGME

Accreditation Council for Graduate Medical Education

AMG

American medical graduate

CWE

clinical workforce equivalent

DO

Doctor of Osteopathy

ED

emergency department

EM

emergency medicine

IMG

international medical graduate

MD

Doctor of Medicine

PEM

pediatric emergency medicine

URiM

underrepresented in medicine