The Multidisciplinary Critical Care Workforce: An Update from SCCM

Ashish K. Khanna, MD, FCCP, FCCM; Alyssa A. Majesko, MD, MSc; Marcia K. Johansson, DNP, ACNP-BC; Joseph F. Rappold, MD, FACS; Heather H. Meissen MSN, ACNP-BC, FCCM, FAANP; Stephen M. Pastores, MD, MACP, FCCP, FCCM

This appraisal from the Society of Critical Care Medicine (SCCM) is a follow-up to the 2009 Critical Connections article “The Critical Care Physician Workforce: An Update from SCCM.” The 2009 article addressed issues surrounding critical care fellowship programs, the projected shortage of critical care physicians, and measures to increase the number of critical care providers. In this issue of Critical Connections, key opinion leaders from SCCM’s education committees revisit the critical care workforce demographic by:

  • Reviewing efforts to identify the problem
    • Previous efforts
    • Current efforts
  • Analyzing the mix of providers in the intensive care unit (ICU)
    • Residents
    • Fellows
    • Advanced practice providers (APPs)
      • Nurse practitioners (NPs)
      • Physician assistants (PAs)
  • Addressing outside factors - New staffing needs for critical care providers affecting the supply-demand balance - Current measures to increase the number of providers in the ICU
This update also includes granular data on the number of fellows trained, number of available training programs per specialty, and percentage of matched positions for select specialties. Further, we summarize the work done by Halpern et al, who used the American Hospital Association (AHA) Annual Survey Database to clearly delineate the current supply and distribution of intensivists. The landscape of our critical care workforce has undergone a paradigm shift in the past decade. This change is evident by the addition of the word Multidisciplinary to the title of this update.

The demand for both critical care physicians and APPs is projected to grow. However, these projections are limited by concurrent estimates of an aging population and growth in numbers of ICU beds, ratios of ICU beds to hospital beds, and the need for critical care providers in non-ICU environments (such as rapid response teams and tele-ICU services) across the United States, which may be increasing more than the calculations of these projections show and may change with time. In addition, the lack of clear estimates of numbers of APPs working in critical care environments, hospital medicine physicians staffing ICU systems, and non-privileged intensivists staffing ICUs hinders a clear calculation of the need and availability mix. Critical care medicine (CCM) as a team sport is now more evident than ever before as we continue to need more providers for extended critical care work outside the ICU, lag in physician intensivist staffing, and benefit from an appropriate mix of multidisciplinary teams to allow for safe and effective care of the critically ill.

Critical Care Physicians: The Unmet Need
Halpern et al analyzed the AHA Annual Survey Database (fiscal year 2015) to determine the total numbers of privileged and full-time-equivalent intensivists in acute care hospitals with intensivists.1 The study also compared the characteristics of hospitals with and without intensivists. Of 2,814 acute care hospitals, only about half had any intensivists. Specifically, 28,808 privileged and 19,996 fulltime-equivalent intensivists were documented as available in 1,469 hospitals with intensivists. These intensivistinclusive hospitals were significantly more likely to be located in metropolitan areas (91% vs. 50%; p < 0.001) and had nearly thrice the number of aggregate hospital beds (403,522 [75%] vs. 137,146 [25%]), 3.6 times the number of ICU beds (74,222 [78%] vs. 20,615 [22%]), and almost twice as many ICUs (3,383 [65%] vs. 1,846 [35%]) compared with hospitals without intensivists. Hospitals with privileged intensivists had significantly more hospital beds (median, 213 vs. 68; p < 0.0001) and ICU beds (median, 32 vs. 8; p < 0.0001), a higher ratio of ICU beds to hospital beds (15.6% vs. 12.6%; p < 0.0001), and a higher number of ICUs per hospital (2 vs. 1; p < 0.0001) than hospitals without intensivists.1 Overall, nearly half of acute care hospitals with ICU beds do not have intensivists. However, up to 75% of ICU beds are covered by intensivists (largely because many of these ICU beds are in large academic centers in urban locations).

With nearly 105,000 ICU beds in the United States, defining the optimal number of CCM physicians needed remains elusive.2 Based on recent data from Halpern et al and from the Association of American Medical Colleges, there appears to be a significant shortage in the CCM physician workforce.1,3 Some innovative mechanisms, such as tele-ICU, are used to work around this shortage. The ever-increasing and pivotal role of APPs to help stabilize the supply-demand balance is significant and continues to evolve. As the U.S. population ages, disease burden expands, and expectations remain high for successful outcomes, the need for intensivists will continue to grow. Additionally, evidence demonstrates that intensivistdirected care results in more appropriate, efficient, high-quality care and improved outcomes for critically ill patients. Mortality benefits occur particularly among patients with major trauma, sepsis, and neurologic injury.4,5,6

Over the past several years, specialties contributing to the CCM workforce have increased their available fellowship positions, with varying degrees of success in filling those spots. This, combined with intensivist burnout, part-time positions originating from contributing specialties (anesthesiology, emergency medicine, pulmonary medicine, and surgery), and retirement, are leading to a negative growth curve.7 The critical care specialty is at a crossroads in terms of being prepared for future demand.7 Clearly, the need for CCM providers may not be adequately matched by the supply of CCM physicians only. Further, creative ideas are needed to encourage recent primary residency entrants to pursue a career in CCM—to communicate to them that they can have a fulfilling future and that they will be mentored by welltrained CCM specialists who are passionate about their career choice. Finally, a collaborative multidisciplinary model with a well-designed nonphysician ICU workforce will result in better management of critical care needs.

Critical Care Physician Training Evolution Over the Past Decade
Since 2009, some key changes have occurred in the critical care provider mix. The inclusion of critical care APPs as an integral part of the workforce is an essential part of this change. Also, the scope of practice of CCM providers continues to expand beyond the boundaries of the ICU (eg, hospital rapid response teams and tele-ICU systems). These changes, along with a larger aging population and improvements in healthcare delivery, may not yet translate into a replete supply of providers. 

Data from the Accreditation Council for Graduate Medical Education (ACGME) for the past decade show exponential increase in the number of critical care fellows across specialties. However, this seems to be less than appropriately matched by a requisite increase in training programs (Table 1). There were 369 accredited training programs in critical care in 2008-09, which has now increased by 25% to 462 in 2018-19. The number of fellows in these programs (a close surrogate for filled fellowship positions) was 2,023 in 2008-09 and has dramatically risen to 3,074 as of the most recent count—an increase of nearly 50% and nearly 1,000 fellows. More importantly, this is much higher than the 18.5% increase reported from 2003-04 to 2007-08 (306 filled positions or fellows). A depiction of this “fellow burden” per program is evident in the derived program-to-fellow ratio (Table 1). A higher program-to-fellow ratio indicates fewer fellows trained per program, while a lower ratio indicates more fellows trained per program. Overall, the past decade has seen an increase in trainees per program. Pediatric, pulmonary, and two-year CCM programs have always been high-volume programs and have shown a slight decline in program-to-fellow ratios, incurring an even higher fellow burden per program. On the other hand, surgical critical care and anesthesia critical care programs maintain better ratios, but anesthesia critical care has shown a dramatic increase in fellows per program, especially evident in the 2010-11 to 2015-16 time frame (Figure 1). 

In terms of specialties, pulmonary critical care programs continue to have the highest numbers of fellows in training. Over the past decade, new training pathways for critical care physicians have emerged. One that is increasing in popularity is the emergency medicine critical care fellowship. There are five pathways to subspecialty board certification in critical care for diplomates of the American Board of Emergency Medicine (ABEM). These are cosponsored by the American Boards of Internal Medicine (ABIM), Surgery (ABS), and Anesthesiology (ABA), and the United Council for Neurologic Subspecialties. Another relatively recent venture by the American Board of Medical Specialties was to adopt neurocritical care as a new subspecialty of the American Board of Psychiatry and Neurology, ABA, ABEM, and American Board of Neurosurgery. Neurocritical care subspecialty programs are new and allow for uniformity in the training and skill sets of neurointensivists through an ACGME-accredited fellowship. Currently, data for these emerging avenues are limited, if available, but more granular reporting may be forthcoming in the coming years. Clearly these pathways will allow more board-certified intensivists to join the workforce and contribute to the growth and development of the specialty.

The update in 2009 showed that, overall, critical care fellowship programs were able to fill about 85% of all approved fellowship positions from 2005-06 to 2007- 08. Whereas these data are no longer reported by the ACGME, over the past five years (2014-2018), critical care fellowships have stayed increasingly competitive. This is evident from the 70% applicant match rate for pulmonary critical care (the specialty with the highest number of fellowship positions) (Figure 2). Surgical critical care and pediatric critical care do slightly better, with an 80%-90% match rate over the past five years. 

The Changing Workforce Demographic
New and innovative staffing models, including utilization of APPs (both NPs and PAs) have been implemented in many critical care units to alleviate physician shortages. Formal training in critical care for APPs has evolved over the past decade as healthcare institutions recognized a gap in critical care preparation.

A deficit in data exists regarding numbers of APPs employed in critical care. According to the National Commission on Certification of Physician Assistants, approximately 1,370 PAs identify their primary practice as critical care. These data are likely lower due to PAs identifying with their department rather than their practice setting (ie, surgery or anesthesia). The American Academy of Physician Assistants reports that 6.8%-8.5% (8,200-10,460) of PAs identify their primary practice setting as a critical care or coronary care unit. Due to the nature of the questionnaire, wherein a discrepancy exists in reporting between practice setting and departmental affiliation, the actual number of practicing critical care PAs is probably 3,000-8,000.

Data regarding the number of NPs practicing in critical care is equally limited. Kleinpell et al8 identified 12.1% (approximately 2,380) acute care NPs working primarily in critical care practice settings. These data do not account for NPs with a different certification such as Family Nurse Practitioner (FNP) or Adult Nurse Practitioner (ANP). The American Association of Nurse Practitioners National Sample Survey does not specify critical care as a practice setting, which would capture other certification types practicing in critical care. Therefore, the number of NPs practicing in critical care is probably 2,500- 4,000.

Based on the PA and NP surveys, the number of APPs practicing in critical care is approximately 3,650-14,500. The number of APPs seeking employment specifically in critical care is growing. Many APPs entering the field have had no exposure to critical care in their educational programs. A small percentage of academic programs offer elective training courses or intensivist specialty tracks. For most APPs, critical care learning occurs during orientation. Orientation varies greatly; most APPs do not receive the fundamental critical care training they need to perform to the top of their license. Because of the deficit in training during orientation, residency and fellowship programs have expanded exponentially over the past decade (Figure 3). Currently 14% of all APP residency and fellowship programs across the United States are focused on critical care (Figure 4).  These 15 programs receive approximately 20-40 applications per cycle but have funding for only 2-4 trainees per year. This equates to 500-1,000 new graduate APPs seeking additional training in critical care while only 50-100 are afforded the opportunity each year. Significant attention must be paid to the deficit of funded training seats. Healthcare institutions, federal payers, and academic centers must find novel ways to fund programs in which structured critical care training is available to all new graduate APPs seeking employment in critical care.

Critical Care Workforce Training and Utilization
Changes in the Landscape of Graduate Medical Education
Between 2011 and 2017, ACGME modifications have impacted resident work hours and supervision requirements. Until 2017, most interns were limited to 16 consecutive hours and residents could work up to 24 consecutive hours. While ACGME work hour standards were intended to mitigate fatigued decision-making, evidence suggested that they may not have reduced medical errors as expected.8 Improvement in patient outcomes may no longer be the only desired endpoint for duty hour restrictions as clinician burnout and wellness become priorities. In 2018, ACGME changed its policy to allow interns to work 24 consecutive hours. In addition, proficiency in ICU procedures may become less common for some trainees because some primary training pathways, such as ABIM’s, do not require placement of central lines or arterial lines for board certification. Many residency programs are also attempting to ensure 24-hour in-house coverage by fellows and intensivists despite the lack of conclusive data on ICU mortality with nighttime or round-the-clock in-house intensivist coverage. The broad implications and unintended consequences of these changes will impact procedural proficiency and competency of providers without fellowship training. These modifications in training requirements for residents will continue to add to the current critical care workforce shortage.

Directions for Hospitals Without Dedicated In-House Critical Care Providers
Over the years, SCCM has worked to increase the supply of providers able to initiate evidence-based critical care. SCCM continues to encourage participation in its Fundamental Critical Care Support (FCCS) program, with special attention paid to those in rural and underserved areas. FCCS primarily serves non-intensivist physicians, nurses, APPs, and other providers who often lack critical care training. Growth in FCCS participation has been steady. Overall, there has been a 35.63% increase in the number of FCCS courses held between 2012 and 2017, resulting in a 16.2% increase in the number of FCCS certificates requested between 2012 and 2017. These programs are designed to help mitigate the ICU provider shortage but may not be able to fully address complex issues that may arise in the middle of the night when seasoned providers are not in house.

The Case for ICU Telemedicine
ICU telemedicine programs are a way to address the urgent need for intensivist consultation. These programs account for up to 20% of critical care services in the United States, primarily in populations without access to adult critical care services. Implementation of an ICU telemedicine program is a practical way to increase access and reduce mortality and length of stay.10 The Centers for Medicare and Medicaid Services (CMS) has made some concessions for payment for telehealth in its 2019 guidelines. CMS will pay for telehealth in a limited number of circumstances—rural hospitals with live videoconferencing and an approved telemedicine provider. The current options for telehealth reimbursement are very restrictive but additional data obtained from the use of telehealth may lead to broader acceptance. Data from a pilot innovation project in electronic ICUs started 3 years ago have found reduced length of ICU stays, fewer readmissions, and costs reduced by millions of dollars. Preliminary data from this project also suggest savings at university medical centers.11 The growth and expansion of telemedicine programs shows great promise and could resolve many of the issues hospitals face related to the current shortage of intensivists.

Looking Ahead
The next 10 years will be a challenging and interesting time for the practice of critical care medicine in the United States—challenging because the need for more trained intensivists is constantly increasing. On one hand, we may be doing well, largely because the workforce demographic has changed to incorporate APPs as a solid backbone of our CCM machinery. On the other hand, this may be misleading since a lot of acute care hospitals do not have privileged intensivists staffing their ICUs. In fact, only about 50% of a large sample of acute care hospitals with ICU beds that completed the AHA Annual Survey reported having intensivist coverage.1 This will get very interesting in the years to come as the needs for ancillary critical care services, such as telemedicine and rapid response teams, increase outside the ICU and the complexity of the care model includes an aging population along with the needs of long-term acute care facilities and post-ICU services.

References
  1. Halpern NA, Tan KS, DeWitt M, Pastores SM. Intensivists in U.S. acute care hospitals. Crit Care Med. 2019 Apr;47(4):517-525.
  2. Halpern NA, Goldman DA, Tan KS, Pastores SM. Trends in critical care beds and use among population groups and Medicare and Medicaid beneficiaries in the United States: 2000-2010. Crit Care Med. 2016 Aug;44(8):1490-1499.
  3. Association of American Medical Colleges. Workforce Studies.  https://www.aamc.org/data/workforce/. Accessed March 5, 2019.
  4. Pronovost PJ, Jenckes MW, Dorman T, et al. Organizational characteristics of intensive care units related to outcomes of abdominal aortic surgery. JAMA. 1999 Apr 14;281(14):1310-1317.
  5. Van der Sluis FJ, Slagt C, Liebman B, Beute J, Mulder JW, Engel AF. The impact of open versus closed format ICU admission practices on the outcome of high risk surgical patients: a cohort analysis. BMC Surg. 2011 Aug 23;11:18.
  6. Iyegha UP, Asghar JI, Habermann EB, Broccard A, Weinart C, Beilman G. Intensivists improve outcomes and compliance with process measures in critically ill patients. J Am Coll Surg. 2013 Mar;216(3):363-372.
  7. Pastores SM, Halpern NA, Oropello JM, Kvetan V. Intensivist workforce in the United States: the crisis is real, not imagined. Am J Respir Crit Care Med. 2015 Mar 15;191(6):718-719.
  8. Kleinpell R, Cook ML, Padden DL. American Association of Nurse Practitioners National Nurse Practitioner sample survey: update on acute care nurse practitioner practice. J Am Assoc Nurse Pract. 2018 Mar;30(3):140-149.
  9. Accreditation Council for Graduate Medical Education. History of Duty Hours. http://www.acgme.org/What-We-Do/Accreditation/Duty-Hours/History-of-Duty-Hours. Accessed June 8, 2017.
  10. Critical care telemedicine: evolution and state of the art: erratum. Crit Care Med. 2015 Feb;43(2):e64.
  11. Abt Associates. Evaluation of Hospital-Setting HCIA Awards. Third Annual Report. Prepared for Centers for Medicare and Medicaid Services. November 1, 2016. Cambridge, MA: Abt Associates. https://downloads.cms.gov/files/cmmi/hcia-hospitalsetting-thirdannualrpt.pdf. Accessed March 5, 2019.