Background. The efficiency and access to existing perinatal resources has become a focus of debate. Despite inconsistent references to the number of neonatologists and unsubstantiated personnel requirement recommendations, recent commentaries have suggested a current 30% to 50% excess in workforce.
Objective. To describe the current neonatology workforce and its practice patterns.
Design. Using a questionnaire developed by the Committee on Practice of the Section on Perinatal Pediatrics of the American Academy of Pediatrics and distributed to 675 neonatology practices identified in the United States Neonatologists Directory 1996, a survey was conducted from July 1, 1995 to June 30, 1996 requesting specific information relating to personnel, type and size of practice, and clinical services provided at practice hospitals.
Results. Respondents included 420 neonatology practices (62.2% response rate) representing 2006 neonatologists providing clinical care in 695 hospitals, 652 with delivery services that accounted for 1 646 881 live births in 1994. More than 95% of practices and neonatologists identified themselves as based in university, private, or hospital settings. Eighty percent of neonatologists were <50 years old. There was an overall 2:1 male to female gender distribution. Sixty percent of practices consisted of 4 or fewer neonatologists, 25% of practices 5 to 7 neonatologists, and 15% of practices 8 or more neonatologists. Sixty percent of practices provided clinical care in only 1 hospital and 1 neonatal intensive care unit (NICU) as compared with 15% of practices in 3 or more hospitals and <5% of practices in 3 or more NICUs. Of the total 478 NICUs (22 in children's hospitals), 67% had <501 annual admissions and 33% had more then 500 admissions. Of the 456 NICUs in 652 practice hospitals with delivery services, 61% of hospitals had <2501 annual deliveries (57% with NICUs) and 39% of hospitals had more than 2500 annual deliveries (90% with NICUs). The average inborn admission rate for these practice hospitals was 11.7%. University, private, and hospital practices had consistent rates of admissions for inborn and outborn NICU and special care nursery admissions. More than 60% of neonatology practices were involved in normal newborn care on a routine basis, in addition to staffing developmental clinics and providing inpatient and outpatient pediatric care. Additional information was analyzed for utilization of residents and neonatal nurse practitioners. By 1999, 50% of practices anticipated hiring 279 neonatologists and 575 neonatal nurse practitioners.
Conclusion. Significant discrepancies between earlier projected neonatologist requirements and current neonatology workforce and service responsibilities are discussed in relation to demands of reallocation of subspecialty resources within an evolving health care system.
- NICU =
- neonatal intensive care unit •
- HMO =
- health maintenance organization •
- SCN =
- special care nursery •
- NNP =
- neonatal nurse practitioner
If you don't know the kind of person I am and I don't know the kind of person you are, a pattern that others made may prevail in the world and following the wrong god home we may miss our star.
—From “A Ritual to Read to Each Other” by William Stafford
The current quality, efficiency, and access to existing perinatal services has become a focus of debate.1 2The reallocation of resources and personnel within an evolving health care system has reassigned neonatology, as well as all subspecialties, to a position of cost-contained justified utilization.3During the past 20 years, 3069 physicians have been certified by the American Board of Pediatrics: Neonatal-Perinatal Medicine (W. W. Tunnessen, Jr, MD, American Board of Pediatrics, personal communication, 1997). The recently published United States Directory of Neonatologists 1996 4 listed 3517 physicians, including: 2635 board-certified in neonatal-perinatal medicine, 652 board-admissible after having completed fellowship training in an accredited program and 230 others identified as participating in neonatology.
In the 1976 consensus publication Toward Improving the Outcome of Pregnancy, 5 the Committee on Perinatal Health issued recommendations for the regional development of maternal and neonatal health care services. During the next 10 years, there was expansion of university training programs and subsequent increases in sophisticated services and specialty personnel resulting in documented improvements in neonatal and perinatal outcomes.6 7 In addition, there were recurrent recommendations that standards be established and regulations enforced to maintain and assess neonatology personnel requirements.8 9
A 1985 statement by the American Academy of Pediatrics Committee on Fetus and Newborn projected the number of neonatologists needed to provide acute and convalescent care.10 Using two reasonable, but unsubstantiated, assumptions of patient to physician clinical service ratios, the statement projected an estimated maximum requirement of 1497 neonatologists and concluded that there was “presently an adequate supply of neonatologists to fulfill all level II and level III program responsibilities.” An accompanying national survey identified 1509 physicians practicing neonatology, demonstrated no regional deficiencies in geographic distribution of personnel, and speculated about the potential for an excessive number of neonatologists.11
The most recent Committee on Perinatal Health published its 1993 conclusions and recommendations for maximizing the quality, efficiency, and access to risk-appropriate perinatal resources and emphasized the need to develop evidenced-based outcome measurements to evaluate existing clinical services.12 A coincident commentary comparing the numbers of neonatologists in seven developed countries was critical of a purported excess of United States neonatologists competing in an unrestrained marketplace, perpetuated by an irresponsible medical system.13 Yet there were inconsistencies in referenced estimates of physicians participating in newborn care and no outcome measurements provided to support such an opinion.14
As an initial step toward a more meaningful and valid assessment of the current neonatology workforce, a survey of all United States neonatology practices was undertaken to establish a reference baseline. Specific information was obtained on characteristics and distribution of neonatology practices, neonatologists, non-neonatologist personnel, hospitals, neonatal intensive care units (NICUs), clinical service responsibilities, and anticipated employment opportunities.
Under the auspices of the American Academy of Pediatrics, the Executive Committee of the Section on Perinatal Pediatrics and the Committee on Fetus and Newborn coordinated resources to identify and survey neonatology practices in the United States. The Committee on Practice of the Section on Perinatal Pediatrics developed a questionnaire that was distributed on July 1, 1995 to 675 neonatology practices identified and cross-referenced in the United States Neonatologists Directory 1996.4
The questionnaire was directed to only one physician in each practice (senior partner or division chief) and requested specific information on personnel characteristics, type and size of practice, and the scope of clinical services (1994 statistics) provided in participating practice hospitals. All neonatology practices were asked to identify themselves as one of five specific practice types: private, hospital, university, health maintenance organization (HMO), or military. The number of neonatologists represented absolute number of individuals and were not delineated as to full time or partial time equivalency. Hospitals included only those facilities in which practice neonatologists provided direct clinical care, specifically excluding referral hospitals in which clinical services were limited to stabilization before neonatal transport.
The definitions of nurseries were based upon clinical services provided and duration of assisted mandatory ventilation. Special care nurseries (SCNs) were capable of providing up to, but not necessarily all inclusive of: intravenous therapy, supplemental oxygen, continuous positive airway pressure, and short-term (≤7 days) assisted mandatory ventilation. NICUs were capable of providing all SCN treatment, in addition to assisted ventilation greater than 7 days.
Data entry was performed by an independent commercial service allowing for strict confidentiality of individual proprietary practice information. Data analysis was accomplished using a relational database (Access 2.0, Microsoft, Redmond, WA).
There were 420 of 675 (62.2%) questionnaires completed by June 30, 1996. These 420 respondent neonatology practices and their 2006 neonatologists represented an overall 60% (52% to 71% by district) of the 3313 United States Neonatologists Directory 1996 4 neonatologists listed by the nine American Academy of Pediatrics districts. There were 1645 of 2006 (82%) board-certified surveyed neonatologists as compared with 2635 of 3517 (75%) board-certified neonatologists in the United States Neonatologists Directory 1996.4 The 741 of 889 (83%) surveyed university practice neonatologists represented the highest percentage of United States Neonatologists Directory 1996 4 neonatologists identified by practice type.
Age and Gender of Neonatologists
The age and gender distribution of 2006 neonatologists was compared among four age groups and five practice types (Table1). More than 80% of neonatologists were <50 years old. The overall gender distribution of neonatologists was 65.8% male and 34.2% female. The percentage of male neonatologists consistently increased across all practice types with increasing age group and represented 57.6% of 30 to 39 year olds as compared with 88.2% of neonatologists 60 or more years old. Female neonatologists represented higher percentages of university (39.7%) and hospital (38.5%) practices than private (27.5%), military (15.2%), and HMO (12.5%) practices.
Type and Size of Neonatology Practice
The distribution of 420 neonatology practices and their neonatologists was identified by practice type and size (Table2). There were 257 (61.2%) practices with 4 or fewer neonatologists as compared with 163 (38.8%) practices with 5 or more neonatologists. Military, private, hospital, and HMO practices were typically smaller, 235 (73.0%) had 4 or fewer neonatologists (average 4.7, 4.2, 3.6, and 3.6 neonatologists per practice, respectively) compared with 76 (76.2%) university practices with 5 or more neonatologists (average 7.6 neonatologists per practice).
Distribution of Non-neonatologist Personnel
The distribution of resident physicians used in 183 (43.5%) neonatology practices were categorized by practice type and size (Fig1). University (91.2%) and military (85.7%) practices worked with residents more frequently than hospital (37.3%), HMO (33.3%), and private (19.2%) practices. The percentage use of residents increased with increasing size of practice and represented only 10.3% of 1 neonatologist practices as compared with 92.9% of practices with more than 12 neonatologists.
The distribution of neonatal nurse practitioners (NNPs) used in 233 (55.3%) neonatology practices were categorized by practice type and size (Fig 1). University (77.2%), private (52.9%), and hospital (46.3%) practices worked with NNPs more frequently than HMO (33.3%) and military (14.3%) practices. Likewise, the percentage use of NNPs increased with increasing size of practice and represented only 35.0% of practices with 1 neonatologist as compared with 85.7% of practices with more than 12 neonatologists.
Distribution of Practice Hospitals
The 695 practice hospitals, 652 (93.8%) with delivery services that accounted for 1 646 881 live births in 1994, were categorized by type and size of practice (Fig 2). Clinical care was provided in only 1 hospital by 258 (61.4%) practices and in 2 or more hospitals by 162 (38.6%) practices. University (51.6%) and private (46.5%) practices more frequently provided clinical care in more than 1 hospital as compared with military (28.6%), HMO (22.2%), and hospital (20.9%) practices. The number of hospitals within which clinical services were provided increased directly with the size of practice. There were 27 (44.3%) practices with 8 or more neonatologists that treated patients in 3 or more hospitals as compared with only 22 (8.6%) practices with 4 or fewer neonatologists.
Characteristics of NICUs and SCNs
The 456 NICUs and 173 SCNs in 652 practice hospitals with delivery services were characterized by a direct relationship between increasing size of delivery service and greater intensity of care (Fig3). NICUs were in 42 of 117 (36.2%) hospitals with <1000 deliveries as compared with NICUs in 83 of 86 (96.5%) hospitals with 4001 or more deliveries. In contrast, SCNs were in 60 of 117 (53.4%) hospitals with <1000 deliveries as compared with only 3 of 86 (3.5%) hospitals with 4001 or more deliveries. An additional 43 hospitals identified with NICUs included 22 of 478 NICUs (4.6%) in free-standing pediatric facilities and 21 hospitals for which the number of deliveries or types of admissions could not be determined.
The distribution of the total 478 NICUs were categorized by type and size of practice (Fig 4). There were 322 (76.9%) practices that treated patients in only 1 NICU. University (33.7%), military (28.6%), and private (19.2%) practices more frequently provided care in 2 or more NICUs as compared with hospital (3.7%) and HMO (0.0%) practices. Participation in 2 or more NICUs was directly related to practice size. There were 31 (50.8%) practices with 8 or more neonatologists that provided care in 2 or more NICUs as compared with only 24 (9.3%) practices with 4 or fewer neonatologists.
The total 478 NICUs were categorized by the number of admissions. There were 123 NICUs (25.7%) with 0 to 250 admissions, 198 NICUs (41.4%) with 251 to 500 admissions, 96 NICUs (20.2%) with 501 to 750 admissions, 45 NICUs (9.4%) with 751 to 1000 admissions, and 16 NICUs (3.3%) with more than 1000 admissions.
The average admission rates (ratio of inborn NICU and SCN admissions to total live births in practice hospitals) for hospitals with delivery services were delineated by size of delivery service (Fig5) and practice type and size (Fig6). The 190 735 inborn admissions had an overall average admission rate of 11.7%. There were higher admission rates for military practices (16.2%) and lower admission rates for practices with more than 12 neonatologists (9.2%).
The distribution of the total 243 656 (inborn and outborn) NICU and SCN admissions were identified by practice type and size. HMO, private, hospital, and university practices had remarkably similar rates of percent admissions relative to percent neonatologists. The higher the rate of percent admissions relative to percent neonatologists, the smaller the practice size.
Allocation of Time
Table 3 identifies the distribution of time (based upon percentages of a 52-week calendar year) allocated by neonatologists. Clinical service consumed the highest percentage of time in private (64%) and hospital (62%) practices as compared with HMO (47%), military (47%), and university (39%) practices. In contrast, research commitments were much greater for neonatologists in university (22%) and military (16%) practices than hospital (5%), HMO (5%), and private (2%) practices. There were striking similarities across all practice types with regard to time allocated for administration, education, committees, conferences, volunteer commitments, and vacation.
Clinical service responsibilities additional to NICU and SCN care were identified (Table 4). Private (57%) and university (55%) practices provided normal newborn care less frequently than military (100%), HMO (89%), and hospital (70%) practices. University (78%), HMO (78%), and military (71%) practices more often staffed developmental follow-up clinics as compared with private (26%) and hospital (32%) practices. Inpatient pediatric care was provided least often by university practices (9%) as compared with all other practices types (93 of 323, 29%). A majority of HMO (67%) and military (57%) practices participated in general office pediatric care.
There were 200 of 420 (47.4%) neonatology practices that anticipated recruiting 279 neonatologists within the next 2 to 3 years (1998 to 1999), representing a 13.9% (279 of 2006) increase in existing personnel. There were 172 private practices (716 neonatologists) that expected to hire 66 (13.6% increase) neonatologists and 98 university practices (741 neonatologists) expected to hire 100 (13.5% increase) neonatologists. Military and HMO practice anticipated only 3 additional employment opportunities for neonatologists.
There were 221 of 420 (52.6%) neonatology practices that projected recruiting 575 NNPs within the next 2 to 3 years (1998 to 1999). There were 73 of 98 (73%) university practices that anticipated hiring 221 (575, 38.4%) NNPs, 75 of 172 (44%) private practices expecting to hire 155 of 575 (27%) NNPs, and 65 of 134 (49%) hospital practices that expected to hire 156 of 575 (27.1%) NNPs. Military and HMO practices anticipated hiring a total of 8 NNPs. Recruitment of NNPs was anticipated by 154 of 233 (66.1%) neonatology practices already utilizing NNPs and 67 of 187 (35.8%) neonatology practices not currently working with NNPs. By 1999, it was projected that 300 of 420 (71.4%) neonatology practices would be utilizing NNPs.
Information obtained from this national neonatology practice survey constitutes the most comprehensive demographic database to date that describes the neonatology workforce, its hospital practice patterns, and the diversity of its service responsibilities. It provides an accurate reference baseline to which individual practices can be compared.
There were pragmatic limitations inherent in the survey design. Hospital and clinical service information represented 1994 statistics, whereas personnel figures and anticipated employment projections reflected the time period 1995 to 1996. The accuracy of requested information provided was dependent upon a single individual in each responding practice and although most numbers represented objective data, percentage estimates of time allocation might reflect responder bias. There was no differentiation between full time and partial time equivalency for neonatologists. The distribution of nonresponding practices was unknown, so percentages of neonatologists, practices, hospitals, and NICUs may not accurately reflect the entire neonatology workforce. There was no information obtained on patient acuity, morbidity, or mortality.
Distribution differences in age and gender of neonatologists amongst the three predominate types of identified practices were consistent with recently published characteristics of pediatric practice patterns.15 Although the overall male to female ratio of neonatologists was 2:1, there was a noticeable trend toward a higher percentage of women in the younger age groups. University practices consisted of the highest percentage of neonatologists 50 or more years old and the highest percentage of women. This compared with private practices with the highest percentage of neonatologists <50 years old and the lowest percentage of women.
Despite specific individual differences between types of practice, practice characteristics were most consistently related to size of practice. Sixty percent of neonatology practices were identified as consisting of 4 or fewer neonatologists providing services most often without the use of residents or NNPs. These neonatologist groups worked in a single hospital with <2501 annual deliveries and 1 NICU with 500 or less annual admissions. A majority of their time was allocated to patient care that was not limited to the NICU and SCN, but included significant responsibility for normal newborn and inpatient pediatric care. In contrast, almost 15% of neonatology practices were identified as consisting of 8 of more neonatologists almost always utilizing both residents and NNPs. They most often worked in 2 or more hospitals, each with more than 4000 annual deliveries, and 2 or more NICUs, each with 501 or more annual admissions. Patient care accounted for 50% of their time and included NICU, SCN, and normal newborn care, in addition to staffing developmental follow-up clinics. The difference in clinical time allotment was accounted for by a significant time commitment to research.
The surveyed practice hospitals with delivery services accounted for 41.2% (1 646 881 of 3 997 00016) of the total live births in 1994. A previously published needs assessment projected that 7% to 10% of newborns required acute and convalescent care17 as compared with an 11.7% average inborn admission rate for practice hospitals. This slightly higher inborn admission rate was most likely attributable to high-risk maternal and neonatal transfers from nonpractice referral hospitals whose additional number of deliveries was excluded from requested survey information. Such an assumption of maternal transfer reflecting an established degree of perinatal regionalization is reinforced by the 78.9% (190 735 of 243 656) rate of inborn admissions to total (inborn and outborn) admissions.
Previous recommendations18 19 and observations20 have suggested as much as a 30% to 50% excess of neonatologists and that an oversupply of physicians most likely would result in increased cost and decreased quality of care.21 22 Despite these projections, one-half of all surveyed neonatology practices anticipated recruitment of more neonatologists and more than twice as many NNPs. Striking differences of involvement in research activities and responsibility for a diversity of previously unacknowledged clinical service commitments confounds previous workforce projections.
How many of the 3069 board-certified physicians are deceased, retired, or no longer active in neonatology practice? Are previously estimated annual attrition rates of 1% to 2% still accurate?23 If <20% of practicing neonatologists are 50 or more years old, are there recognizable limitations to an individual's professional longevity in the NICU?24 Review of American Academy of Pediatrics District VIII physicians listed in the United States Neonatologists Directory 1996 4 revealed 47 of 263 (17.9%) not currently active in neonatology practice. Such a rate of clinical inactivity projected nationwide would significantly reduce the number of physicians actively participating in neonatal care.
There are currently 102 neonatology fellowship programs accredited by the Accreditation Council for Graduate Medical Education Residency Review Committee and responsible for graduating 150 to 165 neonatology fellows each year.25 Are these too few, too many, or the appropriate number of physicians to maintain an optimum number of neonatologists? What consequences do such workforce decisions have on existing neonatology training programs? What are the implications of a disproportionately higher number of anticipated employment opportunities for NNPs than neonatologists?
The American Academy of Pediatrics, the American Board of Pediatrics, and the Association of Medical School Pediatric Chairmen have recently convened the Future of Pediatric Education II, a multidisciplinary task force dedicated in part to answering questions of education, training, and supply of subspecialty pediatric physicians. The Executive Committee of the Section on Perinatal Pediatrics has assumed a leadership role on behalf of its constituency. With completion of this national neonatology practice survey, the Committee on Practice has provided a critical first step toward assessment of neonatal care and its workforce. The rigorous application of standardized evidenced-based outcome measurements to compare a diverse number of practice models will be required to project credible, meaningful neonatology personnel requirements.
This project was supported in part by grants from MeadJohnson Nutritionals and Ross Laboratories.
The identification of neonatology practices and correspondent neonatologists would not have been possible without the work of Dr Dilip R. Bhatt. The authors acknowledge the assistance in development of the survey questionnaire by our Committee on Practice physician colleagues: Stuart A. Weisberger, Paul B. Yellin, Lawrence M. Skolnick, David G. Wells, Steven M. Donn, John J. Fangman, William J. Daily, and Frank L. Mannino. We recognize the editorial assistance of Dr David G. Wells, Dr David K. Stevenson, Dr Ann Stark, Dr Gerald Merenstein, and Dr Richard A. Molteni. We appreciate the illustration designs of Dr Harry Harrison and the logistical support services of Ms Nicole Blankenship and the American Academy of Pediatrics Division of Sections.
- Received May 7, 1997.
- Accepted October 6, 1997.
Reprint requests to (L.D.P.) Northwest Newborn and Pediatric Services, PO Box 95287, Seattle, WA 98145-2287.
- ↵American Academy of Pediatrics. United States Neonatologists Directory 1996; Section on Perinatal Pediatrics. Elk Grove Village, IL: American Academy of Pediatrics; 1996
- ↵Committee on Perinatal Health. Toward Improving the Outcome of Pregnancy: Recommendations for the Regional Development of Maternal and Perinatal Health Services. White Plains, New York: The National Foundation-March of Dimes; 1976
- ↵Silverman WA. How many neonatologists? Lancet. 1982:772–773
- American Academy of Pediatrics, Committee on Fetus and Newborn
- Merenstein GB,
- Rhodes PG,
- Little GA
- ↵Committee on Perinatal Health. Toward Improving the Outcome of Pregnancy: the 90's and Beyond. White Plains, New York: March of Dimes Birth Defects Foundation; 1993
- Silverman WA
- Ratner IM
- Brotherton SE,
- Tang SS,
- O'Connor KG
- Guyer B,
- Stobino DM,
- Ventura SJ,
- Singh GK
- Jung AL,
- Streeter NS
- ↵Gagnon DE. The impact of manage care on the delivery of perinatal health services. In: Proceedings of the Annual Meeting of the National Perinatal Association, San Diego, California, November 17–20, 1994. San Diego, CA: National Perinatal Association; 1994
- American Academy of Pediatrics, Committee on Fetus and Newborn, Committee of the Section on Perinatal Pediatrics
- Clarke TA,
- Maniscalco WM,
- Taylor-Brown S,
- et al.
- ↵American Academy of Pediatrics. Directory of Training Programs in Neonatal-Perinatal Medicine, Section on Perinatal Pediatrics. Elk Grove Village, IL: American Academy of Pediatrics; 1996
- Copyright © 1998 American Academy of Pediatrics