Child Health Needs and the Pediatric Infectious Diseases Workforce: 2020–2040

Infectious diseases remain a significant cause of morbidity and mortality among infants, children, adolescents, and young adults (hereafter, “children”) in the United States. Infections account for ∼25% of all pediatric hospitalizations, with respiratory infections, urinary tract infections, and sepsis being among the most prevalent and costly inpatient conditions.^2–5  PID subspecialists identify, treat, and prevent existing, reemerging, and novel infectious diseases that affect children in hospital and community settings. PID subspecialists also play key roles in multidisciplinary teams caring for children with impaired immunity, medical complexity, critical illness, and antibiotic-resistant infections.

More than half of children admitted to US hospitals receive antibiotics.⁶  Antimicrobial stewardship programs are required for hospitals to receive accreditation and federal funding; there is increasing evidence that these programs reduce medical expenditures.^7–9  Diagnostic stewardship, including new diagnostic tests such as molecular pathogen panels, aims to optimize the use of laboratory testing and can have synergistic benefits when paired with antimicrobial stewardship.^10,11  Finally, the complexity of antimicrobial resistance mechanisms and slow development of new antibiotics require expertise in childhood antibiotic-resistant infections.^9,12 

Although the advent of antibiotics and vaccines heralded hope for eradicating infectious diseases, global changes have countered these advances. In particular, the threat of emerging and reemerging pathogens escalates as the global population grows, global connectivity increases, and climate change expands the range of locations suitable to pathogens, vectors, or reservoir species.^13–15  The past decade has seen the emergence of the Middle East Respiratory Syndrome coronavirus in 2012, the Ebola virus epidemic in 2013-2016, the Zika virus epidemic in 2015, and the global COVID-19 pandemic in 2019, all of which resulted in substantial pediatric morbidity and mortality. We anticipate that these demographic and climate shifts will drive the emergence of new infectious diseases and that expertise will be needed to understand, diagnose, prevent, and treat these infections.¹⁶ 

Survival of children with inherited and acquired immunodeficiencies has improved substantially over the past several decades, resulting in rapid growth in the number of immunocompromised children.^17–22  These children are vulnerable to infections caused by routine, opportunistic, or multidrug-resistant pathogens associated with high morbidity and mortality.^22–24  Many health systems have developed consultation services dedicated to preventing and treating infections in these vulnerable children. Several institutions have also developed 1-year, nonaccredited sub-subspecialty fellowship programs following a 3-year PID fellowship that provides additional training in pediatric immunocompromised host infectious diseases. Expansion of these training opportunities will be needed to care for this growing population of immunocompromised children.

Over the centuries, much childhood morbidity and mortality have resulted from infectious diseases, particularly diarrheal and respiratory illnesses. The establishment of major pediatric organizations, including the American Pediatric Society (1888), the American Academy of Pediatrics (1930), the Society for Pediatric Research (1931), and the American Board of Pediatrics (1933), fostered a strong clinical, public health, and research focus on pediatric infectious diseases.²⁵  PID became an American Board of Pediatrics (ABP) certified subspecialty in 1994.

Based on ABP data through June 2023, 1876 physicians have ever been board-certified in PID, of which 1368 (72.9%) were actively enrolled in Maintenance of Certification.²⁶  ABP data on currently certified subspecialists include individuals who may not be in the workforce because of recent retirement, death, or other factors. To account for this, the following workforce descriptions are limited to the 1290 currently certified PID subspecialists aged ≤70 years. Geographic data are for the United States only. Of these, 58.8% identified as female, 41.1% as male, and 0.1% declined to answer.²⁶  This represents a shift in the gender of the PID workforce from 66.3% male at the time of initial certification in 1994, paralleling trends in other pediatric subspecialties over the past several decades.²⁶  The median age of PID subspecialists was 49 years; 19.8% were aged 61 to 70 years.²⁶  Approximately 69.5% were American medical graduates (AMGs) with a Doctor of Medicine (MD) degree, 3.1% were AMGs with a Doctor of Osteopathy (DO) degree, 16.6% were international medical graduates (IMGs) with an MD degree, and 10.9% were IMGs with an international degree.²⁶  Based on survey data from 2018 through 2022, approximately 60.0% identified as White, 17.0% as Asian, 3.3% as Middle Eastern or North African, and 17.8% as underrepresented in medicine, with 9.4% of Hispanic, Latino, or Spanish origin, 4.2% as Black or African American, and the remainder as multiracial.²⁶ 

Data on practice characteristics of the PID workforce were collected through the ABP’s Maintenance of Certification enrollment surveys from 2018 through 2022 (59.7% response rate for PID).²⁷  The vast majority (90.9%) of PID subspecialists reported being employed full-time, with 57.5% working an average of ≥50 hours per week. Women were slightly more likely to report part-time employment than men (10.3 vs 4.5%). Only 49.2% of PID subspecialists report spending ≥50% time on clinical care compared with 77.8% of all pediatric subspecialists. Most (76.7%) PID subspecialists reported an urban primary work setting and 49% reported that more than half of their patients had public insurance.

When the workforce is limited to the United States, there are 1.7 PID subspecialists per 100 000 children aged 0 to 17 years.²⁸  If these physicians were distributed evenly across the country, this would equate to an average of 24 PID subspecialists per state.²⁸  However, the geographic distribution of PID subspecialists ranges from 11.1 physicians per 100 000 children in the District of Columbia to 0 PID subspecialists in Wyoming and Montana.²⁸  Only 15 states have 2 or more PID subspecialists per 100 000 children.²⁸  Additionally, there is wide variability in the distribution of PID subspecialists within states, with most concentrated in urban areas. Using data from 2019, the average driving distance to a PID subspecialist within the United States was 23.9 miles, ranging from 5.8 miles in Rhode Island to 320.1 miles in Montana (excluding Alaska, Hawaii, Puerto Rico, and the District of Columbia).²⁸  After training, fellows commonly accept a first position near their training location, contributing to geographic disparities in access to PID care (Fig 1).

According to the National Resident Matching Program, there were 59 PID fellowship programs and 84 PID first-year fellowship positions in the United States in 2022, a 22.0% increase in the number of programs and a 31.3% increase in the number of positions since 2014.²⁹  During this same period, the proportion of programs that were not full increased from 41.7% to 59.3%, whereas the number of unfilled positions doubled from 20 (31.3%) in 2014 to 40 (47.6%) in 2022.²⁹  Notably, there is a discrepancy between the number of applicants who match through the National Resident Matching Program and the number of fellows who start training each year, which likely reflects fellows accepting positions outside of the match and programs filling unmatched positions. A total of 62 individuals started PID fellowship in 2008 and 57 in 2022, representing an 8.1% decline.³⁰  Although most individuals starting PID fellowship identified as female, this percentage has declined from 73.3% to 63.8% since 2008, despite rising numbers of female fellows across pediatric subspecialties. More PID fellows have DO degrees and fewer are IMGs with an international degree compared with certified PID specialists; of the 183 fellows in training levels 1 through 3 in 2022 through 2023, 97 (53.0%) were AMGs with an MD degree; 17 (9.3%) were AMGs with a DO degree; 2 (1.1%) were AMGs with an unknown degree; 32 (17.5%) were IMGs with an MD degree; 13 (7.1%) were IMGs with an international degree; and 22 (12.0%) were IMGs with an unknown degree.³⁰ 

Lifetime earning potential is associated with gaps in the pediatric subspecialist workforce and fellowship fill rates.³¹  Using data from national physician surveys conducted in 2018 and 2019, higher subspecialist lifetime earning potential was associated with shorter travel distances for care, a higher number of subspecialists, and higher fellowship fill rates.³¹  Moreover, pediatric subspecialties for which lifetime earning potential increased the least between 2007 and 2018, of which PID had the second smallest increase, and experienced less growth than subspecialties with larger increases in earning potential.³¹  In 2007, PID subspecialists were estimated to have a net –$1 million in lifetime earnings compared with community-based general pediatricians, a figure that had widened to a net –$1.6 million based on 2018 data.^32,33  Moreover, PID subspecialists are paid on average 25% less than adult infectious diseases specialists, representing a net –$1.2 million in lifetime earnings.³⁴ 

The ABP Foundation partnered with the Carolina Health Workforce Research Center at the University of North Carolina at Chapel Hill’s Cecil G. Sheps Center for Health Services Research and Strategic Modeling and Analysis Planning Ltd. to develop an interactive, web-based model that forecasts the supply of physicians aged ≤70 years for 14 pediatric subspecialties between 2020 and 2040.³⁵  The model contains projections at the national, Census region, and Census division levels and includes both a baseline projection and alternative projections reflecting 10 scenarios related to changes in the supply of fellows, the proportion of time in clinical care, and exit from the workforce. Workforce projections presented in the model and below are in headcount (HC; absolute numbers) and clinical workforce equivalent (CWE; HC adjusted by the reported proportion of time spent in direct clinical or consultative care, including patient billing and charting with or without trainees), per 100 000 children aged 0 to 18 years. The model also accounts for changes in the child population at the national and subnational level based on the US Census Bureau; differences by subspecialty for census regions are discussed in the summary article in this supplement.^36,37  An interactive data visualization tool with projections is publicly available online.³⁸  Numbers reported later in this article 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. Estimates of 95% confidence intervals for all scenarios are included in Tables 1–2 and the online tool; next, we focus on scenarios critical to PID.

Between 2020 and 2040, HC is projected to increase by 41%, with a slightly smaller projected increase in CWE. By comparison, the overall pediatric subspecialty workforce HC is projected to increase by 69%, with an accompanying increase in CWE of 68%. Of the 14 subspecialties, PID is projected to have the fourth smallest growth in HC and the third smallest increase in CWE. Moreover, adjusting for anticipated growth in the pediatric population between 2020 and 2040, the HC per 100 000 children is projected to increase by only 25%, with a corresponding increase in CWE per 100 000 children (+23%).^38,39 

Disparities in the geographic distribution of PID subspecialists are projected to worsen between 2020 and 2040 (Tables 1–2, Fig 2). In general, census regions with a lower density of PID subspecialists in 2020 will have smaller improvements in access to PID subspecialists over the subsequent 2 decades. In 2020, the model estimates that the Northeast Census region has 63% and 47% more PID subspecialists per 100 000 children than the Midwest and South regions. Moreover, the Northeast region is projected to experience 28% growth in the number of PID subspecialists between 2020 and 2040, whereas growth in the Midwest and South is expected to be slower (Midwest, mean: 19%; South, mean: 15%). This is particularly concerning for the South Census region, which has the highest poverty rate, the lowest insurance coverage rate, and some of the worst child health outcomes of any census region.⁴⁰  In contrast, although the number of PID subspecialists in the West region is lower than in other Census regions (1.29 per 100 000 children in 2020), this number is projected to grow by 46% between 2020 and 2040. Despite this growth, substantial disparities in access to PID care are projected to persist in the West at the Census division level. The Mountain division had the lowest number of PID subspecialists of any Census division in 2020 (0.92 per 100 000 children), including 37% fewer PID subspecialists than the neighboring Pacific division. Between 2020 and 2040, the number of PID subspecialists in the Mountain and Pacific divisions is predicted to grow at comparable rates (+49% and +46%, respectively), with the Mountain division projected to remain the census division with the fewest PID subspecialists in 2040.

Although the baseline model estimates the projected supply of PID subspecialists based on historical data, the alternative projections provide information on the supply of PID subspecialists should future trends deviate from data that informed the baseline model. Factors projected to have the largest impact on the PID workforce between 2020 and 2040 include changes in the number of trainees entering fellowship and the percentage of time devoted to direct clinical care (Tables 1–2). For instance, a 12.5% reduction in the number of PID fellows by 2030 would reduce projected growth in HC between 2020 and 2040 from 25% to 20%, with a corresponding reduction in growth of CWE from 23% to 18%. Similarly, a 7% decrease in the proportion of time devoted to clinical care by PID subspecialists would reduce projected population-adjusted growth in CWE from 23% to 18%. Alternative scenarios modeling an increase in attrition from the workforce are projected to have a comparatively small impact on the PID workforce. For instance, if all PID subspecialists exited the workforce 5 years earlier over a 3-year period (2021–2023) compared with baseline (2019), this would have a minimal effect on the number of PID subspecialists in 2040 and would only decrease CWE by 8%. Notably, the cooccurrence of alternative scenarios results in approximately additive effects, with a worst-case scenario that combines the assumptions of the 3 aforementioned alternative scenarios reducing projected growth of the population-adjusted number of PID subspecialists from 25% to 20% and CWE from 23% to 10%. Moreover, the impact of these alternative scenarios is anticipated to be most significant in census divisions that currently have fewer PID subspecialists and are projected to have relatively little growth in the PID workforce. For instance, in the West South Central Census division, which had the second lowest number of CWE in 2020, the worst-case scenario is projected to result in a 2% reduction in CWE per 100 000 children between 2020 and 2040 compared with a 9% increase projected by the baseline model. Thus, there is the threat that current disparities in access to PID care across census divisions could worsen with deviations from the baseline model.

There are several reasons to anticipate that PID workforce growth may differ from these projections. First, the model held constant the annual number of fellows starting PID fellowship at 60 fellows per year based on 2019 data. In the past decade, there have been several years with fewer than 60 trainees starting fellowships, and there has been a general downward trend in the numbers starting PID fellowships since 2008.³⁰  As such, the alternative scenario that decreases the incoming fellows by 12.5% may approximate the current reality.

Similarly, estimates of attrition from the PID workforce are based on data collected before the COVID-19 pandemic. The pandemic has been associated with heavy work demands for PID subspecialists, typically without additional compensation, and often hostile public portrayals of and dialogue with those working to inform policy focused on containment and mitigation measures.^41–44  Trainees may be less attracted to PID after witnessing this increased workload and considering the challenges that PID physicians face regarding vaccine hesitancy and the implementation of infection control measures. With the virus that causes COVID-19 becoming endemic and the ongoing threat of other emerging infectious diseases, attrition from clinical PID positions or reductions in clinical hours worked by PID subspecialists could exceed what was considered in alternative model scenarios without further efforts to mitigate the increased demands placed on PID subspecialists. As such, the early exit, CWE decrease, and worst-case scenarios may more accurately represent future changes in the PID workforce.

Finally, because this modeling analysis used self-reported survey data to calculate CWE, the extent to which reported clinical effort includes non–patient-facing clinical activities such as antimicrobial stewardship and infection prevention and control is unknown. Clinical time may remain low or decrease if PID subspecialists are concentrated in academic medical centers, where nonclinical activities such as research may be valued more than in other settings. Because direct clinical revenue generated by PID subspecialists is generally lower than other pediatric subspecialties, health systems may be reluctant to provide additional salary support for clinical care, driving PID physicians to pursue other activities such as research and administrative or education roles. If PID fellowship positions and programs continue to go unfilled, consolidation of training into fewer locations could worsen geographical disparities. Next, we provide some potential implications for the PID subspecialty workforce.

School-based programs introducing children and adolescents to PID enable exploration of pediatrics before medical school, when most education is focused on the care of adults and as little as 4 weeks may be devoted to pediatrics. Nearly three-quarters of medical students graduate in debt, with a median medical school debt of $200 000 in 2019.⁴⁵  Educational debt may play a role in subspecialty choice, and disparities in debt by socioeconomic status, race, and ethnicity could limit efforts to increase PID workforce diversity.^45–47  Since becoming a recognized pediatric subspecialty, PID fellowship has required a minimum of 3 years of training. In recent years, there has been increased interest in shortening this training to 2 years, in part because of increasing numbers of PID fellows pursuing positions focused on antimicrobial stewardship, quality improvement, and hospital epidemiology, with an accompanying decrease in fellows pursuing basic science research careers. Notably, internal medicine has successfully maintained 2-year fellowship programs for decades; the initial board pass rate for adult infectious diseases was 92% in 2021.⁴⁸  Shortening the required duration of PID fellowship training could lessen the financial burden assumed by trainees and stimulate interest in PID among the next generation of pediatricians. Furthermore, additional flexibility could be built into training programs to support the varied interests of individuals pursuing a career in PID. Creating pathways for sub-subspecialty training in disciplines such as transplant and immunocompromised host infectious diseases, antimicrobial stewardship, and infection prevention would enable interested fellows to complete an additional year of training in these disciplines during a 3-year training program.

Improving children’s access to PID care may require changes to how PID subspecialists provide patient care and alternative staffing for roles traditionally occupied by PID subspecialists. PID subspecialists could increase the proportion of time they devote to clinical care, although without other changes in care delivery, this would leave less time for other critical activities (eg, research, infection prevention). Increasing the use of advanced practice providers and shifting nonclinical work currently performed by PID subspecialists to other health professionals (eg, antimicrobial stewardship pharmacists) could increase the time that PID subspecialists spend in clinical care, though some of these providers also face workforce challenges.⁴⁹  Additionally, general pediatricians could be trained to provide care for some infections that are currently comanaged with PID subspecialists through an increased emphasis on infectious diseases in pediatric residency or through short courses focused on specific PID competencies, such as has been done to train physicians in HIV care.^50,51  Consideration should be given to increasing the involvement of general pediatricians in public health efforts focused on preventing infections among children in the community. Finally, the expansion of telemedicine is a promising solution to reducing current geographic disparities in access to PID care. Previous studies reported that pediatric subspecialty care delivered via telemedicine was associated with high parental satisfaction and, in some cases, improved health outcomes.^52–55  Recently, a step-by-step guide was proposed for the provision of remote infectious diseases support for patient consultations and antimicrobial stewardship activities in underserved communities in the United States.⁵⁶ 

The physician compensation structure in the United States is complex and related in part to the value that insurers place on the performance of procedures rather than investigation through history and physical examination. With the exception of 3 procedure-intensive pediatric subspecialties (cardiology, critical care medicine, and neonatal-perinatal medicine), pediatric subspecialists are compensated less than their general pediatric colleagues and their internal medicine subspecialty counterparts.^33,34  Leadership in several pediatric departments have recently taken steps to improve equity of subspecialist compensation by adjusting the salaries of individuals in low-paying pediatric subspecialties (including PID) to be commensurate with those of pediatric hospital medicine subspecialists or their internal medicine subspecialist counterparts. Differences in insurance coverage of children and adults also contribute to the disparity in compensation among pediatric and internal medicine subspecialists. A significant and increasing percentage of children (>50% as of November 2022) are insured through Medicaid or the Children’s Health Insurance Program, with Medicaid reimbursement remaining well below that of Medicare and private insurance.^57,58  Working with legislators to reform Medicaid and move beyond fee-for-service models is necessary to remedy these inequalities in physician compensation, stimulate growth of the pediatric subspecialty workforce, and demonstrate the value of children in our society.⁵⁹ 

To anticipate and effectively respond to changes in the future PID workforce, research priorities include evaluating the impact of changes in clinical practice on access to and delivery of PID care, understanding factors that contribute to shifting demographics in the PID workforce, and determining the effects of emerging infections on PID subspecialist supply. Although telemedicine has the potential to substantially expand access to PID subspecialty care, few previous studies compared in-person and telemedicine-based PID care. Notably, a recent study conducted at 3 community hospitals found that a shift from in-person to telemedicine-based infectious diseases consultation was associated with more consultations, although length of hospital stay and mortality were unchanged.⁶⁰  Additionally, future studies could investigate the use of telemedicine to expand pediatric antimicrobial stewardship and infection prevention initiatives to rural US areas. Relatively little is also known about how the COVID-19 pandemic may influence the decisions of pediatric residents pursuing fellowship training or PID subspecialists exiting the workforce. More than one-quarter of the current PID workforce are IMGs, yet specific data are lacking on practice locations of IMGs after PID fellowship and how many of these individuals leave the US PID workforce to pursue positions abroad. Finally, a better understanding of factors contributing to the declining proportion of PID fellows who identify as female is needed and learning how PID can offer desirable career pathways to all pediatric trainees.

Although the PID workforce is anticipated to grow between 2020 and 2040, growth is lower than that of other subspecialties; it is uncertain if this growth is realistic and will be sufficient to meet anticipated or unexpected growth in the need for PID care. In particular, increasing populations of immunocompromised or medically complex children, rising antimicrobial resistance, and emerging pathogens in an increasingly connected world threaten to strain the PID workforce over the coming decades. Additionally, the availability of PID care is projected to remain concentrated in urban areas, with geographic disparities in access worsening between 2020 and 2040. A multifaceted approach will be necessary to ensure that the PID subspecialty thrives over the next several decades and is able to meet the needs of children.

The authors thank Emily McCartha, Andrew Knapton, and Adriana R. Gaona for their review of the modeling data presented. They also thank Virginia A. Moyer and Patience Leino for their editorial support. Last, they thank the pediatricians who shared their information with the American Board of Pediatrics Foundation and made this supplement possible.

ABP

American Board of Pediatrics

AMG

American medical graduate

CWE

Clinical workforce equivalent

DO

Doctor of Osteopathy

HC

Headcount

IMG

International medical graduate

MD

Doctor of Medicine

PID

Pediatric infectious diseases