Child Health Needs and the Pediatric Pulmonology Workforce: 2020–2040

A broad spectrum of chronic respiratory disorders of childhood is currently managed primarily by pediatric pulmonologists, though there is overlap with generalists or other subspecialists for some multisystem conditions. Table 1 highlights how division directors, in a 2020 (prepandemic) survey,⁴  ranked the anticipated importance of various disorders over the next decade. Respondents also commented on the future of the subspecialty’s clinical activities. Frequent responses included the changing nature of cystic fibrosis (CF) care with the advent of new therapeutics, transitions of some inpatient domains to hospitalist services, opportunities for a shift toward other facets of respiratory care enabled by new technologies, and factors both increasing and decreasing the roles of pulmonologists in asthma care.

Table 2 speculates the directional change and relative importance of trends for specific factors that may impact clinical need and market demand for our subspecialty. A significant addition to this list, in context of the coronavirus disease 2019 pandemic and subsequent spikes in other respiratory pathogens, is the possibility that pulmonologists will have increasing roles in the management of future infectious epidemics. An overarching issue is also the impact of socioeconomic, racial, and ethnic disparities on the PULM field. These disparities are known to be associated with more severe disease and worse outcomes, which will likely apply across all the conditions mentioned and may increase demand for subspecialty services in at-risk communities. Finally, administrative and regulatory requirements surrounding patient care, as well as the nonclinical aspects of pediatric pulmonologists’ work, are steadily increasing and thus reducing availability of clinicians for direct patient care.

The balance of the PULM clinical portfolio is shifting. Quantifying these changes and their net impact on demand for pediatric pulmonologists is difficult; the field’s scope of practice overlaps to some degree with several subspecialties (eg, allergy and immunology, pediatric critical care medicine, hospital medicine). It is possible that some specific clinical care needs, such as chronic ventilator care, could be partially met by advanced practice providers. However, clinical need for PULM services is anticipated to increase over the next decade with the survival of more children with complex chronic illnesses. The current workforce discussion focuses on the supply of clinicians, but it is important to emphasize that the research and scholarship activities of the subspecialty often inform clinical needs. Thus, the basic, clinical, translational, and health services research workforces in children’s lung disease need to advance in parallel with the increased clinical workforce, as noted by other commentators.^6,12 

During the 20th century, advances in management in neonatology, critical care, infectious diseases, CF, and other areas led to a need for subspecialists focused on the care of children with chronic respiratory disorders. By the mid 1970s, many centers established clinical programs in pediatric pulmonary medicine and subsequently developed training programs to meet the growing need for physicians to care for these children. Because of the rapidly growing clinical need and a desire to standardize training for the discipline, leaders in the field applied to the American Board of Pediatrics (ABP) for certification status. The PULM subboard was approved and implemented in 1984. Board eligibility was contingent on completion of a 3-year accredited fellowship program that included an emphasis on research training.^1,2  Since 1984, the discipline’s core clinical and research portfolio has included CF, chronic lung disease of prematurity, and other disorders (eg, respiratory infections, severe asthma, diffuse and developmental lung diseases, diseases of the thorax), and respiratory support for a variety of neuromuscular disorders. In addition, PULM has become a central discipline for research and clinical care of aerodigestive and sleep disorders, as well as rare, genetic lung diseases.

Based on ABP data through June 2023,¹³  a total of 1582 pediatricians have ever been board-certified in PULM, 79.3% (1254) of whom were enrolled in Maintenance of Certification (MOC). Additionally, 4.4% of ABP-certified pediatric pulmonologists held additional certifications in pediatric critical care medicine, 1.7% in neonatal-perinatal medicine, and 16% in sleep medicine.¹⁴  Initial certifications, issued every 2 years in the PULM discipline, reached a nadir of 47 per exam in 2002, but have since increased slowly, and over the decade from 2010 to 2020 ranged from 84 to 117 per exam offered. Extrapolating from these data, currently 42 to 58 new pediatric pulmonologists complete fellowship training programs and become certified annually in the United States.

The ABP data on currently certified subspecialists include individuals who may not actually be in the workforce because of recent retirement, death, or other factors.¹³  To adjust for this, descriptions of the current workforce below limit the sample to currently certified pediatric pulmonologists ≤70 years. Geographic data are limited to the United States.

Of the 1172 currently certified pediatric pulmonologists ≤70 years in 2023, 49.2% identified as female and 50.8% as male (the ABP has only offered other options since 2021). Demographics of the PULM workforce have shifted over time. In terms of gender representation, most certified physicians in 1986 were male (82.9%), but a rapid increase in the proportion of females certified occurred between 2006 and 2010, and in the latest certification year (2022), 62.4% were female. This parallels the shift toward increased numbers of women in all pediatric subspecialties. Among currently certified pediatric pulmonologists in 2023, the median age was 49 years. Twenty-one percent were 61 to 70 years, providing an estimate that one-fifth may retire in the next 1 to 2 decades. Regarding medical training, 62.6% were American medical graduates (AMG) with a Doctor of Medicine (MD) degree, 4.7% were AMGs with a Doctor of Osteopathy (DO) degree, 20.4% were international medical graduates (IMG) with an MD degree, and 12.3% were IMGs with an international degree.¹³  The proportion of pediatric pulmonologists with DO degrees has increased over the past decade, paralleling the increase in DO clinicians in the pediatric workforce in general.

Detailed, self-reported race and ethnicity data have only been tracked since 2018 via the ABP MOC survey.¹⁵  Race and ethnicity estimates from 2018 to 2022 responses suggest that 60.0% of pediatric pulmonologists self-identified as white, 20.4% as Asian, 7.8% as Hispanic, Latino, or Spanish origin, 3.7% as Middle Eastern or North African, 2.3% as Black or African American, and 0% as Native American or Pacific Islander; approximately 13.1% self-identified as underrepresented in medicine (URiM).¹⁶  These data are consistent with a 2014 survey among American Academy of Pediatrics and American College of Chest Physicians members who identified as pediatric pulmonologists, which reported a similar racial and ethnic breakdown, with only 3% as Black or African American.³  Thus, like other pediatric subspecialties, there is currently a lack of diversity among pediatric pulmonologists compared with the US population, particularly Black or African American, with minimal improvement over the past decade; however, the proportion of trainees from URiM backgrounds appears to be increasing for our subspecialty as discussed below.

Data on the work characteristics of current ABP-certified PULM subspecialists have recently been collected through the ABP’s MOC enrollment surveys.¹⁵  Surveys from 2018 to 2022 had a 61.8% response rate for PULM, reflecting responses from 523 eligible subspecialists ≤70 years. Due to skip patterns in the survey, the percentages reported below are for individual questions and not the entire sample. The majority reported being employed full-time (89.5%); 59.4% reported working ≥50 hours per week on average over the last 6 months, exclusive of time on-call but not working. Women (13.6%) were more likely to indicate part-time employment status compared with men (5.4%). Most (77%) spent ≥50% of their time in clinical care; in contrast, 11.1% reported spending ≥50% of their time in research.¹⁶  These results show a slight increase in part-time work for pediatric pulmonologists, compared with MOC data compiled from 2009 to 2015, although some who self-identified as part-time were working ≥40 hours per week, as noted by Freed et al.^17,18  This may be linked to an increase in women in the field or career decisions made by dual-career households. The largest proportion of respondents (39.7%) endorsed that their primary work setting was within a medical school or parent university, and most (85.7%) had a faculty appointment. The majority (79.2%) reported their primary work setting was within an urban environment. Fifty-nine percent reported that more than half of their patients received public insurance. Although these data are helpful in describing the clinical supply of pediatric pulmonologists, the actual number of clinical full-time equivalents available to see children in need is not known.

When the workforce is limited to the United States, the geographic distribution of currently ABP-certified pediatric pulmonologists is highly variable, with subspecialists concentrated in urban areas within academic medical centers, as highlighted by previous commentators.^1,19,20  In 2023, there were an average of 22.1 currently certified pediatric pulmonologists per state (range 0–111), which translates to 1.5 pediatric pulmonologists per 100 000 children 0 to 17 years (range 0.0–8.0) across the United States (Fig 1). There was wide variability in the distribution of PULM subspecialists within states, with most concentrated in urban settings and few in rural areas. This ranged from 0 (Alaska, Wyoming, North Dakota) to nearly 8 per 100 000 children (District of Columbia). Turner et al²⁰  analyzed geographic variation in pediatric subspecialty access from 2003 to 2019. In 2019, the average driving distance to a certified pediatric pulmonologist in the United States was 25.6 miles; distances ranged from a low of 6.5 miles in New Jersey to a high of 281.8 miles in North Dakota (excluding Alaska, Hawaii, Puerto Rico, and the District of Columbia). There was a decrease in the percentage of US children with driving distances >80 miles to see a pediatric pulmonologist from 13% (2003) to 7% (2019). Over the same interval, mean pediatric pulmonologist availability by hospital referral region increased from 0.62 to 0.98 per 100 000 children (+58%). In comparison, the same metric in 2019 was 0.35 for pediatric rheumatology and 5.57 for neonatal-perinatal medicine.

Much has been written recently concerning the need for improvement in resident recruitment to the subspecialty, the low fill rate of fellowships, and the balance between research and clinical training in our discipline.^6,22–29  Early snapshot data from the Accreditation Council for Graduate Medical Education (ACGME)³⁰  reveal that there was a 14% increase (50 to 57) in the number of US accredited programs between academic years 2012 to 2013 and 2021 to 2022. National Resident Matching Program data for first-year fellows indicates that the number of first-year fellows who match into accredited programs increased from 41 in 2012 to 61 in 2022. The ABP total count of first-year fellows, incorporating individuals who take positions within or outside the Match, demonstrates that the number of first-year pediatric pulmonology fellows over the last decade increased 31%, from 58 in 2012 to 76 in 2022. Over the same interval, the percent match rate for these programs increased from 66% to 73.5%, and the calculated “final fill rate” (which includes fellow positions filled outside the National Resident Matching Program) for 2022 was 85.5%. Thus, although there continue to be unfilled fellowship positions, the absolute number of fellows completing PULM training each year, the most important metric for workforce assessment,³¹  has been increasing. We speculate that programs developed by the American Thoracic Society and Cystic Fibrosis Foundation to engage pediatric residents early in their training and expose them to the subspecialty have contributed to this success. Additionally, academic organizations including the Pediatric Pulmonology Division Directors Association, Pediatric Pulmonary Training Directors Association, and Council on Pediatric Subspecialties have focused efforts on raising awareness of pediatric subspecialty workforce concerns and the need to improve recruitment.

Data from the latest ABP subspecialty trainee dashboard³²  show that among the 197 pediatric pulmonology fellows in training during the current academic year 2022 to 2023 in standard, noncombined US fellowship programs, 65.5% identified as female, 34% as male, and 0.5% declined to disclose their information. Regarding their prior medical training, 54.3% were AMGs with an MD degree, 12.2% were AMGs with a DO degree, 16.2% were IMGs with an MD degree, and 9.6% were IMGs with an international degree. There were 14.7% who self-identified as from a URiM background. Comparison of these data with the previously noted demographic and training data for all certificate holders indicates that the proportion of the discipline’s workforce who are women, AMGs, or DO degree holders continues to rise. In contrast, ACGME census report data suggest that the percentage of PULM fellows who self-identified as URiM improved from 12% (2007) to 18% (2019).³³  Reasons for this difference between recent fellowship demographic trends and current ABP certificate holders are unclear, but factors may include variation in eventual pursuit of board certification, MOC versus initial certification data, or reluctance to respond to various workforce surveys. The increase in the percentage of trainees describing themselves as URiM provides some optimism regarding the diversification of the PULM workforce.

Trainees commonly work in close proximity to their training institution after graduation, and rates of incoming fellows to training programs per regional population are lowest in many of the same areas that are relatively underserved by the discipline (Fig 1).

Recent publications have highlighted differences in financial remuneration across pediatric subspecialties and in comparison, to adult subspecialties.^34,35  Although older reports suggest compensation may be less important for residents in choosing a pediatric subspecialty career, the increasingly high rates of debt could conceivably impact decision-making. For PULM, 37.7% of current fellows owe $200 000 or more (similar to 39.5% for all pediatric subspecialty fellows), based on results from the ABP’s Subspecialty In-Training Exam Survey in 2022 (personal communication, ABP, February 20, 2023).

The workforce forecasting model, an ABP Foundation-led effort to characterize the future supply of pediatric subspecialists ≤70 years in the United States, incorporates data from the ABP and other national resources.¹¹  Baseline workforce estimates are provided for 14 subspecialties at the national and subnational level from 2020 to 2040; alternative forecasts are included that incorporate variables represented by different scenarios. Variables impacting the scenario forecasts include changes in the number of trainees, proportion of time spent in clinical care, and attrition. Workforce projections are expressed in either headcount (HC) or clinical workforce equivalent (CWE), which is HC adjusted for the self-reported time spent in direct clinical or consultative care. The model also accounts for changes in the child population at the national and subnational level based on the US Census Bureau³⁶ ; differences by subspecialty for census regions are discussed in the summary article in this supplement.³⁷  An interactive data visualization tool of the model with projections is publicly available online.³⁸ 

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. Table 3 provides data examining the impact of all scenarios on HC per 100 000 children 0 to 18 years, and Table 4 focuses on CWE per 100 000 children 0 to 18 years. Estimates of 95% confidence intervals are available in Tables 3 and 4 and the visualization tool. The following comments pertain to the baseline model scenarios of critical importance to the PULM workforce and are expressed as CWE per 100 000 children 0 to 18 years.

The baseline model predicts that the overall US supply of clinical pediatric pulmonologists (CWE per 100 000 children) will increase steadily until 2040, when there will be 36% more supply, representing approximately a 1.8% increase per year or 1.1 CWE per 100 000 children. Of the subspecialties modeled, PULM growth is roughly midway between the lowest-growth specialty (child abuse pediatrics at 8%) and highest-growth subspecialty (pediatric critical care medicine at 78%).

In the scenario in which fellows increase by 12.5% from current numbers and the attrition rate remains stable, CWE per 100 000 children increases by 36% (baseline) to 41% by 2040. Similarly, if pediatric pulmonologists increased the proportion of time spent in clinical or consultative care by 7%, the projected CWE growth would rise from 36% to 45%. If both changes occurred concurrently, CWE would increase by 51% by 2040.

However, the experience of the past few years suggests there could be small reductions in the workforce growth, because of burnout, early retirement, and an increase in part-time work. Taking a worst-case scenario with a 12.5% decrease in the number of fellows, a 7% decrease in percent clinical time, and early career exits, the model still predicts a 22% increase in CWE nationally in 2040 compared with the baseline model. Census data, incorporated into the model, project that the US population will increase by 14% over the same period.³⁶  Thus, the model, even in a relatively conservative version, predicts that PULM will be able to keep up with US population growth overall and even absorb a small increase in demand above current levels. Recent data suggest that PULM fellows are quite optimistic about job prospects compared with fellows in other pediatric subspecialties,³⁹  which also seems to bode well for the field’s future.

Despite the positive outlook nationally, there is marked geographic variation in the PULM workforce, and our models suggest this will persist and even worsen for some areas, as examined in Fig 2 for census regions. Tables 3 and 4 delve even deeper and report on HC and CWE per 100 000 children 0 to 18 years by census division, respectively. The CWE per 100 000 children in the highest prevalence division (New England) is anticipated to remain more than double that in the lowest prevalence region (West South Central). Other areas with relatively low workforce HC and CWE include the East South Central, Mountain, and Pacific regions, presenting an access problem that is compounded by driving distances. This is most striking in the Mountain census division, which already has a relatively low estimate of CWE per 100 000 children.

Key trends for PULM include the increase in the proportion of women, the lagging proportion of pediatric pulmonologists from URiM backgrounds, and the slow increase in fellows over the last decade. The model predicts that the number of pediatric pulmonologists normalized to the US population, or HC per 100 000 children, is likely to increase by 37% over the next 20 years. This increase equates to a little less than 1% increase a year. For CWE per 100 000 children, the percent growth (36%) is slightly less. However, increases in burnout, early retirement, and part-time work could limit this growth. There will continue to be marked geographic maldistribution and disparities in access to services, with substantially higher numbers of pediatric pulmonologists per 100 000 children in the Northeast than in other areas of the US and higher concentrations in more urban areas.

It is anticipated that changes in the specialty’s clinical portfolio will significantly impact training needs; careful attention and adaptation of training programs to these issues will be essential. Research and education remain core missions in the field’s discipline, and clinical care for some conditions has been significantly improved as a direct result of research-driven therapeutic breakthroughs and other advances. The need to train outstanding physician-scientists, clinician-educators, and health services experts focused on respiratory disease in children is greater than ever, and in our opinion should be viewed as integral to, rather than in competition with, clinical workforce issues.

Regarding geographic disparities, one approach might be to bolster training opportunities or training partnerships with institutions in lower-access regions, though it may be impractical to develop programs in some rural underserved areas.

Our subspecialty’s clinical workforce needs in the coming years are likely to be impacted by changes in the scope of clinical care (Table 2). For example, the care of children with CF has changed since the initiation of highly effective CF transmembrane conductance regulator modulator therapy, with fewer inpatient admissions. Hopefully the same will become true for other conditions in our clinical portfolio, such as severe asthma, sickle cell lung disease, or childhood interstitial or rare lung diseases. Conversely, the need to provide care for children with complex, technology-dependent disorders has increased in the past decade. Issues such as future pandemics, global warming, environmental degradation, and others will likely pose additional challenges and significantly impact the scope of PULM service needs.

Regarding the geographic variation in access to pulmonologists, a practical approach may be the development of a more robust telemedicine system, in which physicians from urban medical centers facilitate clinical services in rural or underserved regions that cross state lines.

Concentrating specialized respiratory care resources in regional centers may improve some outcomes, but many families do not have the resources to relocate, and adapting program locations to population patterns is likely to be limited, as noted above. Hence, other ways to provide subspecialty care in currently underserved areas must be considered. Loan repayment programs or other incentives for moving graduating fellows from their training location to under-resourced settings should be considered, even though personal finances may not be the highest priority in specialty fellowship decisions. Remote care models enabled by improved telemedicine or other technology-based approaches should be explored, as should novel partnership models with primary care pediatricians.

Assessment of workforce adequacy requires an estimation of future clinical need and market demand, as discussed by Freed³¹  and Leslie et al in the introductory article to this supplement.¹⁰  This PULM workforce model does not include predictions of changing child health needs or market demand for condition-specific subspecialty services, other than referencing overall population trends. Future research must track the workforce, changes in clinical need, and the impact of market forces on demand for pediatric pulmonologists.

Available data and the model developed by Fraher et al¹¹  allow a more detailed analysis of the PULM workforce supply than previously possible. Even after adjusting for factors like early retirements, part-time work, and a small decline in trainees, the PULM discipline appears equipped to meet a modest increase in overall demand for clinical services. These findings are consistent with a prior published assessment of ABP data.⁴  However, there continue to be marked geographic disparities in access to services, with higher concentrations of pediatric pulmonologists per 100 000 children in some regions and urban areas. Future workforce assessments should focus on optimizing care access in the context of need, accounting for the changing clinical portfolio.

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

ABP

American Board of Pediatrics

ACGME

Accreditation Council for Graduate Medical Education

AMG

American medical graduate

CF

cystic fibrosis

CWE

clinical workforce equivalent

DO

Doctor of Osteopathy

HC

headcount

IMG

international medical graduate

MD

Doctor of Medicine

MOC

Maintenance of Certification

PULM

pediatric pulmonology

URiM

underrepresented in medicine