Projecting the Future Pediatric Subspecialty Workforce: Summary and Recommendations

The model, developed by Fraher and colleagues at University of North Carolina at Chapel Hill, in collaboration with Sheps Center for Health Services Research and Strategic Modeling and Analysis Planning Ltd. and staff affiliated with the ABP Foundation, advances our understanding of the pediatric subspecialty workforce by permitting a more nuanced examination of clinical supply for each subspecialty.⁹  Based on historical data, the model forecasts the supply of 14 pediatric subspecialties at the national, census region, and census division levels from 2020 to 2040.^4,9  Workforce projections are expressed in headcount (HC, absolute numbers) and clinical workforce equivalent (CWE, which is the HC adjusted for the reported proportion of time spent in direct clinical or consultative care, including patient billing and charting with or without trainees) for pediatric subspecialists ≤70 years of age. The model also accounts for forecasted changes in the child population at the national and subnational level based on projections from the University of Virginia Weldon Center,¹²  permitting standardization per 100 000 children ≤18 years.¹²  An interactive data visualization of the model is publicly available online.¹³ 

At the national level, the model predicts overall HC for all pediatric subspecialties to increase from 23 289 physicians ≤70 years in 2020 to an estimated 39 253 (+69%) physicians in 2040. When standardized per 100 000 children, this represents an increase of HC from ∼28.6 in 2020 to 42.6 in 2040 (+49%). CWE is projected to increase from 14 344 in 2020 to 24 071 overall (+68%) and from 16.62 to 26.11 per 100 000 children (+48%).¹² 

Combining all subspecialties into a single projection obscures the known heterogeneity of growth trajectories across the subspecialties.¹⁴  For example, the projected increase in HC and CWE per 100 000 children by 2040 for all pediatric subspecialists is 49% and 48%, respectively (Table 1). The same measures are 81% and 78% for pediatric critical care medicine subspecialists and 10% and 8% for child abuse pediatrics subspecialists (Table 1). Although these percent changes may seem large, especially for a smaller subspecialty, these may reflect relatively small numerical changes. Figure 2 displays this visually for CWE per 100 000 children; each subspecialty differs in the intercept in 2020, the projected slope or magnitude of change in CWE from 2020 to 2040, and final estimated CWE in 2040. Conclusions for some subspecialties (eg, pediatric cardiology, pediatric pulmonology) are that future numbers will be sufficient, should historical trends continue.^15,16  Others are struggling to care for current patients (eg, developmental-behavioral pediatrics).¹⁷ 

Examining supply only at the national level also obscures asymmetric geographic growth at the US census region and division level.¹⁸  Recent demographic projections suggest that the South and West regions will increase in child population growth, whereas the Northeast and the Midwest will remain stagnant or demonstrate relatively slower growth.¹²  In contrast, for most subspecialties, the Northeast and Midwest census regions currently have the highest HC and CWE per 100 000 children and the largest projected total subspecialty CWE by 2040 (Table 2). Although this table does not account for potential changes in child health needs, the potential misalignment between areas of child population and subspecialty growth may exacerbate existing access issues and worsen geographic inequities over the next 2 decades.

The model also estimates each subspecialty’s workforce under 10 hypothetical future “scenarios.” The supplement’s introductory and methodology articles describe these scenarios and their drivers.^4,9  Briefly, potential changes include ±12.5% in incoming fellows by 2030, ±7% in clinical time beginning in 2022 and ±1% until 2028, after which the change is permanent, early exit from the workforce by 5 years from 2021 to 2023, +12.5% in the probability of midcareer (10–20 years posttraining) exit, a worst-case scenario (early exit by 5 years, −12.5% in fellow pathway, and −7% in clinical time), and a best-case scenario (+12.5% in fellow pathway and +7% in clinical time).

The fellowship scenarios highlight heterogeneity between the subspecialties and that the effect of a scenario must be interpreted relative to the current size of the workforce. For example, a ±12.5% in fellows for a larger subspecialty (eg, pediatric cardiology) leads to an increase in HC and CWE per 100 000 children of 0.20 and 0.14, respectively. The same scenario applied to a smaller subspecialty (eg, developmental-behavioral pediatrics) yields an increase in HC and CWE per 100 000 of 0.04 and 0.02, respectively.¹⁷  No changes in the number of fellows will address geographic disparities without other solutions.

The scenarios also demonstrate the impact of changes in clinical effort and attrition on HC and CWE. With increases in clinical effort, concerns include unintended consequences, particularly for smaller subspecialties (eg, pediatric endocrinology, pediatric infectious diseases)^19,20  that spend more than 50% of their time in nonclinical activities such as research, teaching, administration, antimicrobial stewardship, and leadership. Short-term attrition hypothesized to reflect pandemic-related burnout and moral injury, defined as “the challenge of simultaneously knowing what care patients need but being unable to provide it because of constraints beyond our control,”²¹  had a significant, albeit temporary, impact.

Results from the model inform our understanding of how potential changes in the numbers of fellows, clinical time, and attrition may influence future supply. The model demonstrates that increasing the size of the pediatric subspecialty workforce in isolation will not address ongoing shortages for some subspecialties. Without changing the location of pediatric subspecialty training, simply increasing fellows’ numbers will only reinforce existing geographic disparities across all subspecialties. Other solutions will be needed.

Below are recommendations from the supplement writers across the domains of education and training, practice, policy, and future workforce research that could potentially bolster the entire pediatric subspecialty workforce and ultimately improve US child healthcare. These recommendations were developed independent of the recent report from the National Academies of Sciences, Engineering, and Medicine (NASEM) Committee on the Pediatric Subspecialty Workforce and Its Impact on Child Health and Well-Being²²  and serve to reinforce those recommendations and provide new insights.

All articles reinforce that pediatric subspecialty workforce supply depends on upstream factors in medical school and residency.^23,24  Creative new strategies are needed to entice undergraduate and medical students to embark on pediatric and pediatric subspecialty careers. The science and practice of pediatrics have never been more exciting and held more promise to impact the lives of children. Suggestions for medical students include the creation of early exposure programs designed to showcase the full spectrum of pediatric subspecialist practice and the allure of the science of pediatrics through novel, engaging mentorship opportunities. Residents also need early exposure; incorporating longitudinal outpatient subspecialty experiences could supplement the inpatient rotations more traditionally experienced by residents. In addition, better understanding of what drives fellowship interest and recruitment will be critical. Some subspecialties, often those that generate higher revenue, demonstrate large growth trajectories in trainees (eg, pediatric cardiology) compared with others (eg, adolescent medicine, developmental-behavioral pediatrics).^25,26  In fact, some subspecialties have shown little to no growth over the last several years.²⁵ 

Targeted recruitment of candidates into pediatrics and pediatric subspecialties who represent the multiple dimensions of diversity of the US pediatric population is a second critical area of focus. Although the proportions of women and physicians with a Doctor of Osteopathy have increased substantially for most subspecialties, racial and ethnic diversification has been slow.¹¹  Diversifying the pediatric subspecialty workforce requires upstream interventions going back as far as elementary school and investments beyond the capabilities of an individual subspecialty. Importantly, supporting diversity should increase the diversity of the medical workforce at large and avoid a “0-sum game” of competition across medical specialties or pediatric subspecialties.⁶  Without a commitment and collaboration across medical and pediatric organizations, these changes will not occur.

A third theme is the notable heterogeneity in the size and location of fellowship training programs by subspecialty, resulting in geographic disparities, coupled with data that many fellows choose to stay in areas close to where they trained.²⁷  Some mechanisms to address this mismatch could include enhanced training in geographic areas of need through the establishment of new training programs, the development of partnerships between large academic medical centers and smaller hospitals in underserved areas with satellite and community locations, incentives to train in areas of shortage, and greater training in the application of telemedicine for subspecialty care.

A fourth theme is that educational innovations and nimbleness are needed to prepare pediatric subspecialty fellows for a rapidly changing world where new diseases, treatments, and technologies such as artificial intelligence are emerging that will impact child health and healthcare delivery. Various interventions to change training length and/or content are suggested. For example, some subspecialties (eg, pediatric infectious diseases, pediatric endocrinology, adolescent medicine, developmental-behavioral pediatrics) question the impact of educational debt and low physician compensation on the pool of trainees and recommend shortening fellowship duration or supporting a combined 5-year residency and fellowship training. Other subspecialties (eg, pediatric hematology-oncology, pediatric gastroenterology, pediatric pulmonary, pediatric critical care medicine) describe a trend toward a fourth year of fellowship in preparation for “sub-subspecialization” (also termed “super-specialization”) in which a subspecialist focuses almost exclusively on a specific condition or patient subpopulation within a subspecialty.^{16,17,19,20,28–31,47}  The small size of the research workforce across pediatric subspecialties is also a concern raised by some subspecialties.^16,28 

Each of these recommendations is aimed at addressing critical workforce needs; however, any potential changes must be examined within the context of possible unintended consequences. For example, one risk of shortening training duration is the potential negative impact on readiness to independently practice subspecialty care, given existing concerns about current graduating fellows’ readiness.^32–34  The recent focus on competency-based medical education across many of the medical subspecialties calls out the importance of ensuring readiness as part of medical training³⁵ ; recent research on the pediatric subspecialty Entrustable Professional Activities provides insight on the current state of readiness to achieve these outcomes at the conclusion of training. In addition, the impact of changes in fellowship training on the research workforce, which is already under-subscribed, must be considered. Changes to training duration may also not influence an individual’s decision to pursue fellowship, the type of fellowship pursued, or their ultimate location of practice. Similarly, although sub-subspecialization may be necessary to address the complexity of children’s needs, downsides include the potential for greater geographic disparities given the concentration of sub-subspecialists in large academic medical centers in urban areas.³⁶  A competency-based medical education-focused approach would allow fellows to tailor their fellowship training to meet the needs of children they anticipate caring for in practice.³⁷ 

Other subspecialties (eg, pediatric emergency medicine, developmental-behavioral pediatrics, adolescent medicine, pediatric nephrology) are highly aware that the numbers of trainees entering their subspecialties will not be sufficient to address children’s needs in the foreseeable future. Recommendations include greater attention to training other child healthcare experts (eg, general pediatricians, emergency medicine physicians, internists, family physicians, advanced practice providers [APP], and mental health professionals) to work as part of an integrated team with board-certified pediatric subspecialists to provide high-quality pediatric subspecialty care.

Any changes in training duration and content will require discussions with the Accreditation Council for Graduate Medical Education, ABP, and other relevant medical credentialing boards. Further research is needed around readiness for practice using the Entrustable Professional Activities framework, potential risks (eg, inadequate achievement of clinical competence, worsening child health outcomes, reduction in research workforce, no change in number of fellows), and solutions will also be necessary.

Practice recommendations are closely intertwined with policy recommendations, as policy changes may be necessary to change practice. Almost every subspecialty article stresses the evolving healthcare needs of children in response to medical advances in diagnosis and treatment, quality improvement initiatives, and environmental factors. They also point out known care disparities in their subspecialty by race and ethnicity, socioeconomic status, geographical location, and exposure to social drivers of health. These inequities in health in the US society for individuals from Black or African American, Hispanic, Latino, or Spanish origin, or Native Hawaiian or other Pacific Islander communities were dramatically and publicly on display during the pandemic.^38–42  Unfortunately, disparities in outcomes among children continue to be common; a more in-depth discussion of these disparities occurs in the article on diversity, equity, and inclusion.¹¹  Robust quality improvement efforts that stratify data and test interventions by race and ethnicity could contribute to addressing these disparities.

A second recommended opportunity to extend the existing subspecialty workforce is supporting primary care providers through immediate virtual consultation when children present in the primary care setting. This approach would allow rapid treatment and management and avoid delays when seeking subspecialty expertise,²⁸  but would require a mechanism to reimburse subspecialists for their time and expertise in the care of remote patients and to provide care across state lines where necessary. In addition, not all care can be provided remotely^3,15 ; more research into the use of telemedicine will be necessary, especially the evaluation of clinical outcomes.

Greater inclusion of APPs in care delivery is another common mechanism identified across the subspecialties. A 2011 study found that pediatric subspecialists were interested in expanding their pediatric nurse practitioner workforce; the articles in this supplement reinforce that need.⁴³  However, the inclusion of APPs in health care teams is limited by small numbers choosing pediatrics, competition for APPs between pediatric generalists and subspecialists, the small number and location of pediatric APP training programs, and unfavorable reimbursement in some markets.^43–45  Ensuring a robust pediatric subspecialty workforce to support APPs in delivering pediatric subspecialty care will continue to be significantly important.

Subspecialties as diverse as pediatric emergency medicine,⁴⁶  pediatric gastroenterology,²⁸  adolescent medicine,⁴⁷  and developmental-behavioral pediatrics¹⁷  also called attention to the burgeoning pediatric mental health crisis. This epidemic existed before the coronavirus disease 2019 (COVID-19) pandemic and has subsequently worsened; the current system is not prepared to care for these children.⁴⁸  Three recent American Academy of Pediatrics (AAP) policy statements reiterate the critical role of (1) all pediatric subspecialists in basic mental health competencies for children, (2) neonatologists in the recognition and management of perinatal depression, and (3) emergency physicians in the management of acute mental health issues.^49–51  Solutions must be developed that integrate behavioral mental health care across the pediatric subspecialties.

Last, recent legal decisions and actions have compounded moral injury by making it more difficult for pediatric subspecialists to provide the care their patients deserve. For example, the AAP supports the rights of adolescents to confidentiality and pregnancy counseling options⁵² ; however, the Supreme Court’s Dobbs v Jackson decision⁵³  that overturned Roe v Wade compromises this care. In addition, recent anecdotal data suggest that antiabortion legislation negatively impacts infant deaths and the number of infants born with congenital anomalies. The AAP also supports access to gender-affirming care,⁵⁴  but increasingly aggressive political agendas are affecting the safety of both patients and adolescent medicine and pediatric endocrinology subspecialists providing care to them.^19,47,55  Child abuse pediatrics subspecialists find themselves increasingly facing negative coverage through media outlets.⁵⁶  Moral injury in response to the pediatric mental health crisis is also being experienced among trainees and subspecialists, specifically in the inpatient management of children with mental health conditions.⁵⁷  Providing both support and solutions to burnout and moral injury will be critical moving forward.

The subspecialty articles raise numerous issues and provide policy recommendations to address concerns. Providing care for children is not always prioritized by health systems, primarily because of unfavorable reimbursement practices compared with adults. Before the pandemic, the pediatric healthcare delivery system was already showing strain. Pediatric inpatient units in the United States decreased by 19.1% from 2008 to 2018, contributing to geographic disparities as units in rural communities were more likely to close.⁵⁸  Hospital consolidations also resulted in loss of inpatient pediatric services.⁵⁹  The COVID-19 pandemic further impacted hospitals; a 2020 Children’s Hospital Association report found that children’s hospitals had a 4% and 11% decrease in inpatient and outpatient revenue in 2020, respectively, and 90% of children’s hospitals experienced financial losses.⁶⁰  Although some have emphasized the unique value of pediatric health careers,⁶¹  the pandemic laid bare the degree to which financing comprehensive pediatric care is less profitable than providing services for adult patients.⁶² 

Ensuring children have access to health insurance will be a critical next step, particularly for children with chronic medical conditions who are often on Medicaid. Both the Affordable Care Act and the public health emergency declared during COVID-19 expanded Medicaid insurance coverage for children; the Kaiser Family Foundation estimated that 8 to 24 million individuals may lose Medicine coverage with the ending of the public emergency on March 31, 2023.⁶³  Given that over 50% of children were covered by Medicaid or another form of public insurance as of November 2022,⁶⁴  this is a potential crisis with respect to access for children cared for by subspecialists. Medicaid reform and parity with Medicare would provide pediatricians with the necessary resources to provide optimal care for children.⁶⁵ 

Policy solutions that address the financial barriers to pursuing a pediatric subspecialty career as a pediatric subspecialist are also recommended. Loan forgiveness programs, such as the recently funded program announced by the Health Resources and Services Administration, are one solution to limit the educational debt experienced by physicians.⁶⁶  Program eligibility for these programs cannot focus only on clinical care as academic subspecialists also serve educational, research, quality improvement, and administrative needs at their institutions. However, research suggests that loan forgiveness or shortened fellowships will minimally impact lifetime earning potential compared with the dramatic differences in lifetime earning potential in an adult subspecialty.^67,68  As educational debt increases, this very real compensation barrier to pediatric subspecialty workforce supply must be addressed.⁶⁹ 

Some subspecialties (eg, pediatric endocrinology, pediatric gastroenterology, pediatric nephrology, and pediatric hematology-oncology) also rely heavily on international medical graduates to provide care.^3,19,28,29  Addressing policy barriers to both training and practice in the United States is an important step for assuring access to care for children and diversifying the pediatric subspecialty workforce. The accompanying article by Orr and Leslie et al discusses policy solutions to reduce barriers to international medical graduates participation in the pediatric subspecialty workforce.¹¹ 

A final overarching recommendation across this supplement is the lack of an obvious feedback mechanism or identified party responsible for assuring the alignment of child health needs, available pediatric subspecialty care across the US healthcare system, and the size, distribution, and content of subspecialty fellowship training programs. The experiences of children and families in accessing care and children’s health outcomes vary across different geographic locales.

There is also a resounding request for continued analysis, monitoring of trends, and use of data to information solutions to increase the capacity of the pediatric subspecialty workforce to meet child health needs. Although these data may exist in isolation, integrating them to inform pediatric care nationally is not yet happening.

Given the complexity of understanding child health needs and the multiple institutions involved in pediatric training and care, an intentional oversight mechanism for tracking child health needs, market demand, and supply of all professionals involved in pediatric subspecialty care would be ideal. Ideally, such an entity would be housed within a neutral third-party organization, such as the Agency for Healthcare Research and Quality (AHRQ), as suggested by the recent NASEM consensus study report,²²  and function to track data collected by other groups. The AHRQ or another party could then convene critical stakeholders to develop recommendations for local and national responses to changes in child health needs, the numbers and scope of work of the broader pediatric subspecialty workforce, the impact of market demand, necessary curricular changes, and other factors.

Robust data will continue to be needed to inform the workforce dialogue, both across subspecialties and specific to a subspecialty. One major gap is appropriate subspecialty-specific metrics for the number of subspecialists needed per 100 000 children. Unfortunately, the science behind measuring child health need and adjusting need based on social drivers of health is limited at best. High-quality outcomes data for children stratified by demographics, different care settings and provider types, and geographic locations will be needed to permit comparisons of different care models and ensure changes don’t exacerbate existing disparities.⁷⁰  Given the growing diversity of children in the United States, it will also be important to determine how to best collect demographic information in a nonstigmatizing way and employ that information to detect and address disparities in outcomes. As a field, we must be intentional about understanding how the diversity of our workforce, from residents to leaders, impacts care and what mechanisms can be applied to increase the diversity of the pediatric workforce or train the nonminoritized workforce to provide more culturally sensitive care. In addition, although the model addresses known limitations of prior supply modeling as described in the opening and methodology articles,^4,9  factors needing attention in future workforce research are provided in Table 3. Subspecialty-specific topics are also discussed in the accompanying articles. These research areas could be prioritized across the AHRQ, National Institutes of Health, the Patient-Centered Outcomes Research Institute, Health Resources and Services Administration, and other federal entities as well as interested foundations.

This article, in conjunction with the larger Pediatrics supplement, seeks to contribute to conversations regarding the pediatric subspecialty workforce by consolidating known information about pediatric subspecialists certified by the ABP using a novel pediatric subspecialty workforce model. National organizations, such as the NASEM and many pediatric organizations, are actively focused on better understanding the barriers and challenges facing the pediatric subspecialty workforce.^22,71  As a field, we must both embrace our clear differences as subspecialties and simultaneously recognize that no particular subspecialty can only consider its own supply needs. The supplement’s authors look forward to further dialogue and the necessary actions to support the health of the nation’s children, now and in the future.

We thank Virginia A. Moyer, John A. Barnard, David L. Turner, Andy Knapton, and Patience Leino for their editorial support; and thank the pediatricians who shared their information with the American Board of Pediatrics Foundation and made this supplement possible.

AAP

American Academy of Pediatrics

ABP

American Board of Pediatrics

AHRQ

Agency for Healthcare Research and Quality

APP

Advanced Practice Provider

CWE

clinical workforce equivalent

HC

headcount

NASEM

National Academies of Sciences, Engineering, and Medicine