0
Editorials: Point and Counterpoint |

COUNTERPOINT: Does the United States Need More Intensivist Physicians? No FREE TO VIEW

Gordon D. Rubenfeld, MD
Author and Funding Information

CORRESPONDENCE TO: Gordon D. Rubenfeld, MD, Program in Trauma, Emergency and Critical Care, Interdepartmental Division of Critical Care Medicine, Sunnybrook Health Sciences Center, University of Toronto, 2075 Bayview Ave, Room D108c, Toronto, ON M4N 3M5, Canada


Copyright 2016, . All Rights Reserved.


Chest. 2016;149(3):625-628. doi:10.1016/j.chest.2015.11.031
Text Size: A A A
Published online

The current “debate” about the need to train more intensivists is, of course, not really a debate. If the models that predict the growing demand for critical care services are valid and every patient in an ICU bed needs an intensivist to improve outcome, and if health care is organized in the future as it is now and society is willing to pay for this service, there is little doubt that there will be a lack of trained intensivists by 2025. I will argue that even if these assumptions are true, and it is unlikely that they are, intensivists should focus our efforts in other areas to improve outcomes for the critically ill and let others argue for strategies to increase our numbers. First, the models that estimate the number of intensivists required to optimize health outcomes contain significant uncertainties and assumptions. Second, critical care workforce projections rely on models of future care that are not financially sustainable for the United States. Third, specialty-specific models do not address the inherent trade-offs involved in training one type of physician at the expense of others. Finally, strategies to reduce demand for critical care services and increase quality of care without intensivists are available and present clear opportunities for leadership by intensivists.

The first physician workforce model by Lee and Jones in 1933 used fairly crude expert estimates of how many physicians were required to serve a population. Since then, innumerable commissions, government agencies, professional societies, and academics have built increasingly sophisticated models to estimate the supply and demand for physicians. In 1974, Jeffers and colleagues pointedly observed that workforce models share a major limitation: they have trouble distinguishing the need for medical services (that optimize health outcome) from the wants (that consumers and clinicians desire) and the demand (that can be paid for). Although they incorporate projections about demographic and chronic disease changes, various care delivery and reimbursement models, the use of nonphysician providers, and physician work patterns, all workforce models fundamentally assume that the use of physicians in the baseline year is equal to the need for these services. In other words, we find out how many doctors we need by measuring how many we use.

We do this because it is hard to come by evidence about which types of clinicians provide the best care. There is a considerable database addressing the question of whether intensivists improve outcome. To intensivists outside the United States where ICUs are generally staffed by intensivists, these data are reassuring but irrelevant. To many policy makers in the United States, the data, such as they are, are sufficient to make recommendations. To the critic, they are a compilation of before-after studies in academic centers and recent cohort studies showing no benefit., This fundamental uncertainty about the role of intensivist staffing would stymy any effort to predict optimal physician numbers except that every other field has the same problem.

It might be reasonable to use current intensivist staffing as a starting point to measure needs if we knew that the current assumptions were biased in only one direction. If the current number of intensivists were either adequate or low because most ICUs lacked intensivist staffing, we would know that any predicted gaps based on these numbers would likely be biased toward being underestimates. If every patient who were billed for an ICU bed in the United States required an intensivist, current intensivist staffing would be woefully inadequate and meeting the demand of an inexorably aging population would require training or importing vast numbers of intensivists. For example, to ensure that two-thirds, not even 100%, of patients admitted to an ICU bed in 2000 were seen by an intensivist would have required an additional 1,200 intensivists. This means 15 years ago, before any demographic aging bulge, there had already been a gap in critical care physicians equal to 65% of the then-current workforce.

Unfortunately, we also have a growing body of literature showing that the United States has more ICU beds than do comparator countries, that the use of these ICU beds varies considerably, and that the availability of an ICU bed as well as its reimbursement determines its use.,,,,, So, although it is possible that every patient who is billed for an ICU bed deserves an intensivist, it is not clear whether each of these patients needs an ICU bed. Therefore, to use terminology from Jeffers and colleagues, the intensivist workforce models are giving us a picture of future wants that may be considerably greater than either needs or demands. Intensivists are likely to counter that much of this desire for intensive care comes from families, surgeons, oncologists, emergency room physicians, and the perceived safety of alternate locations in some hospitals for individual patients. This may be true but it is an explanation, not a justification for formulating these predictions.

Intensivists have struggled with the triage criteria for ICU admission at both the low and high end of severity of illness for over 30 years. Many patients receive futile therapy and aggressive treatment near the end of life that might be avoided with better advance care planning. Similarly, although intensivists likely improve quality of care, it is obvious that we are not the only way to do this. Some ICUs without access to intensivists who work in selected environments seem to deliver high-quality care. With appropriate implementation it is likely that in some settings, hospitalists, advance practice nurses, decision support, protocols, and attention to quality improvement can substitute for intensivist staffing. The extent to which reducing ICU demand and improving quality without intensivists can reduce the need for more intensivists simply has not been incorporated into the workforce models. If the United States had the same number of ICU beds as does Canada or France, how many intensivists would we need? If 50% of ICU patients could be effectively cared for without intensivists, what effect would this have on projected demands? It is telling that these analyses have not been part of the workforce debate.

Let us assume that the additional handovers, exposure to resistant bacteria, unnecessary arterial lines, and noise are not harmful to the patient who is admitted to an ICU but could have been cared for on a ward. Let us assume that the demand for intensive care cannot be reduced and that there is no way to substitute for intensivists without patients dying. Then we just have to pay for this care. And pay we will. Per-capita Medicare expenditures double between age 70 and 95, and this segment of the population is projected to grow from 14.9% of the population in 2015 to 21% of the population in 15 years. These are the same years when patients experience much critical illness., Unless the recent and unexplained pause in Medicare cost growth is sustained, the United States is willing to completely reengineer health-care funding, or health-care workers are willing to take significant pay cuts, an unsupportable percentage of the gross domestic product will be devoted to intensive care. It simply may not matter whether society decides it wants this much intensive care; it will not be able to pay the bill.

To demonstrate the challenges in considering workforce models in isolation and their variability over time, consider the recent Association of American Medical Colleges workforce report, which presents a worst-case scenario with a gap of 15,500 medical subspecialists in 2025. The Committee on Manpower for Pulmonary and Critical Care Societies report, generated by a consortium of critical care professional societies, predicts a 2025 gap of approximately 3,500 intensivists using the current, not optimal, staffing model. Either intensivists are responsible for 22% of the predicted gap in all 12 medical subspecialties, which seems unlikely, or these 2 models do not agree. By now, it should be unsurprising to readers whether workforce models disagree or are wrong. In 1993, four major physician workforce models were released. While they disagreed on tactics, they agreed unanimously that there was going to be a physician glut and that there were too many specialists. Twenty years later, the models from the Association of American Medical Colleges predict equal and significant shortages of both generalists and specialists.

The uncertainty and variability in the models do not relieve us of the responsibility of addressing a complex problem. However, the urgent catalyst for this discussion is demographics and it is not a “critical care” problem, but a public health problem. Will we need more intensivists? Almost certainly, but the solutions to the policy questions posed by changing demographics are politically sensitive for guilds that represent professional as well as patient interests. How many more intensivists? If limited graduate medical education dollars means we must choose between investing in training intensivists or cardiologists, or between intensivists and family doctors, whose model should drive that decision? What if restrictive licensing, immigration, or enhanced reimbursement arrangements are needed to recruit intensivists to less desirable locations? Are we prepared to acknowledge that hospitalists or telemedicine delivers equivalent care to an intensivist?, Are we prepared to advocate for a critical look at ICU use and strategies to reduce its use?

Given the vagaries and uncertainties in these models, the potential for bias, and the significance of the problem, I think it unwise and unseemly for intensivists and our leaders to focus on training more intensivists. We should let national workforce panels do this for us. This will allow us to focus on areas within our expertise and without potential conflicts of interest. We can inform these panels with critical deliberations on how to reduce the demand for ICU beds to parallel countries and regions of the United States with lower use. We need to perform some simple comparative effectiveness evaluations of models of care that include regionalization, telemedicine, and the use of nonintensivists. We need to advocate for funding to measure accurately and improve the quality of intensive care regardless of the training of the clinician providing that care. We need to ensure that there will be an adequately trained workforce of nurses, respiratory therapists, pharmacists, and physiotherapists for our future patients. We need to make critical care training and practice more desirable and sustainable. Most important, if the dire predictions about access, quality, and cost across the spectrum of health care are true, we cannot be seen as protecting our turf when the house is burning down.

References

Kahn J.M. .Rubenfeld G.D. . The myth of the workforce crisis: why the United States does not need more intensivist physicians. Am J Respir Crit Care Med. 2015;191:128-134 [PubMed]journal. [CrossRef] [PubMed]
 
Lave J.R. .Lave L.B. .Leinhardt S. . Medical manpower models: need, demand and supply. Inquiry. 1975;12:97-125 [PubMed]journal. [PubMed]
 
Jeffers J.R. .Bognanno M.F. .Bartlett J.C. . On the demand versus need for medical services and the concept of “shortage.”. Am J Public Health. 1971;61:46-63 [PubMed]journal. [CrossRef] [PubMed]
 
Wilcox M.E. .Chong C.A. .Niven D.J. .et al Do intensivist staffing patterns influence hospital mortality following ICU admission? A systematic review and meta-analyses. Crit Care Med. 2013;41:2253-2274 [PubMed]journal. [CrossRef] [PubMed]
 
The Leapfrog Group. The Leapfrog Group for Patient Safety.http://www.leapfroggroup.org/Hospitals/SurveyInfo/leapfrog_safety_practices/icu_physician_staffing. Accessed January 28, 2016.
 
Yoo EJ, Edwards JD, Dean ML, Dudley RA. Multidisciplinary critical care and intensivist staffing: results of a statewide survey and association with mortality [Epub ahead of print May 12, 2014].J Intensive Care Med.pii:0885066614534605.
 
Levy M.M. .Rapoport J. .Lemeshow S. .Chalfin D.B. .Phillips G. .Danis M. . Association between critical care physician management and patient mortality in the intensive care unit. Ann Int Med. 2008;148:- [PubMed]journal
 
Smetana G.W. .Landon B.E. .Bindman A.B. .et al A comparison of outcomes resulting from generalist vs specialist care for a single discrete medical condition—A systematic review and methodologic critique. Arch Intern Med. 2007;167:10-20 [PubMed]journal. [CrossRef] [PubMed]
 
Health Resources and Services Administration. Report to Congress. The critical care workforce: a study of the supply and demand for critical care physicians.http://bhpr.hrsa.gov/healthworkforce/reports/studycriticalcarephys.pdf. Accessed August 1, 2015.
 
Wunsch H. .Linde-Zwirble W.T. .Harrison D.A. .Barnato A.E. .Rowan K.M. .Angus D.C. . Use of intensive care services during terminal hospitalizations in England and the United States. Am J Resp Crit Care Med. 2009;180:875-880 [PubMed]journal. [CrossRef] [PubMed]
 
Wunsch H. .Angus D.C. .Harrison D.A. .et al Variation in critical care services across North America and Western Europe. Crit Care Med. 2008;36:2787-2793 [PubMed]journal. [CrossRef] [PubMed]
 
Stelfox H.T. .Hemmelgarn B.R. .Bagshaw S.M. .et al Intensive care unit bed availability and outcomes for hospitalized patients with sudden clinical deterioration. Arch Intern Med. 2012;172:467-474 [PubMed]journal. [CrossRef] [PubMed]
 
Chen L.M. .Kennedy E.H. .Sales A. .Hofer T.P. . Use of health IT for higher-value critical care. N Engl J Med. 2013;368:594-597 [PubMed]journal. [CrossRef] [PubMed]
 
Safavi K.C. .Dharmarajan K. .Kim N. .et al Variation exists in rates of admission to intensive care units for heart failure patients across hospitals in the United States. Circulation. 2013;127:923-929 [PubMed]journal. [CrossRef] [PubMed]
 
Marshall M.F. .Schwenzer K.J. .Orsina M. .Fletcher J.C. .Durbin C.G. Jr.. Influence of political power, medical provincialism, and economic incentives on the rationing of surgical intensive care unit beds. Crit Care Med. 1992;20:387-394 [PubMed]journal. [CrossRef] [PubMed]
 
Wagner D.P. .Knaus W.A. .Draper E.A. .Zimmerman J.E. . Identification of low-risk monitor patients within a medical-surgical intensive care unit. Med Care. 1983;21:425-434 [PubMed]journal. [CrossRef] [PubMed]
 
Huynh T.N. .Kleerup E.C. .Wiley J.F. .et al The frequency and cost of treatment perceived to be futile in critical care. JAMA Intern Med. 2013;173:1887-1894 [PubMed]journal. [CrossRef] [PubMed]
 
Neuman P. .Cubanski J. .Damico A. . Medicare per capita spending by age and service: new data highlights oldest beneficiaries. Health Aff (Millwood). 2015;34:335-339 [PubMed]journal. [CrossRef] [PubMed]
 
Angus D.C. .Linde-Zwirble W.T. .Lidicker J. .Clermont G. .Carcillo J. .Pinsky M.R. . Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med. 2001;29:1303-1310 [PubMed]journal. [CrossRef] [PubMed]
 
Rubenfeld G.D. .Caldwell E. .Peabody E. .et al Incidence and outcomes of acute lung injury. N Engl J Med. 2005;353:1685-1693 [PubMed]journal. [CrossRef] [PubMed]
 
Association of American Medical Colleges. The complexities of physician supply and demand: projections from 2013 to 2025.https://www.aamc.org/download/426242/data/ihsreportdownload.pdf?cm_mmc=AAMC-_-ScientificAffairs-_-PDF-_-ihsreport. Accessed August 1, 2015.
 
Abt Associates, Inc. Future needs in pulmonary and critical care medicine.http://www.abtassociates.com/reports/chest2.pdf. Accessed August 1, 2015.
 
Rivo M.L. .Jackson D.M. .Clare F.L. . Comparing physician workforce reform recommendations. JAMA. 1993;270:1083-1084 [PubMed]journal. [CrossRef] [PubMed]
 
Wise K.R. .Akopov V.A. .Williams B.R. Jr..Ido M.S. .Leeper K.V. Jr..Dressler D.D. . Hospitalists and intensivists in the medical ICU: a prospective observational study comparing mortality and length of stay between two staffing models. J Hosp Med. 2012;7:183-189 [PubMed]journal. [CrossRef] [PubMed]
 
Wilcox M.E. .Adhikari N.K. . The effect of telemedicine in critically ill patients: systematic review and meta-analysis. Crit Care. 2012;16:R127- [PubMed]journal. [CrossRef] [PubMed]
 

Figures

Tables

References

Kahn J.M. .Rubenfeld G.D. . The myth of the workforce crisis: why the United States does not need more intensivist physicians. Am J Respir Crit Care Med. 2015;191:128-134 [PubMed]journal. [CrossRef] [PubMed]
 
Lave J.R. .Lave L.B. .Leinhardt S. . Medical manpower models: need, demand and supply. Inquiry. 1975;12:97-125 [PubMed]journal. [PubMed]
 
Jeffers J.R. .Bognanno M.F. .Bartlett J.C. . On the demand versus need for medical services and the concept of “shortage.”. Am J Public Health. 1971;61:46-63 [PubMed]journal. [CrossRef] [PubMed]
 
Wilcox M.E. .Chong C.A. .Niven D.J. .et al Do intensivist staffing patterns influence hospital mortality following ICU admission? A systematic review and meta-analyses. Crit Care Med. 2013;41:2253-2274 [PubMed]journal. [CrossRef] [PubMed]
 
The Leapfrog Group. The Leapfrog Group for Patient Safety.http://www.leapfroggroup.org/Hospitals/SurveyInfo/leapfrog_safety_practices/icu_physician_staffing. Accessed January 28, 2016.
 
Yoo EJ, Edwards JD, Dean ML, Dudley RA. Multidisciplinary critical care and intensivist staffing: results of a statewide survey and association with mortality [Epub ahead of print May 12, 2014].J Intensive Care Med.pii:0885066614534605.
 
Levy M.M. .Rapoport J. .Lemeshow S. .Chalfin D.B. .Phillips G. .Danis M. . Association between critical care physician management and patient mortality in the intensive care unit. Ann Int Med. 2008;148:- [PubMed]journal
 
Smetana G.W. .Landon B.E. .Bindman A.B. .et al A comparison of outcomes resulting from generalist vs specialist care for a single discrete medical condition—A systematic review and methodologic critique. Arch Intern Med. 2007;167:10-20 [PubMed]journal. [CrossRef] [PubMed]
 
Health Resources and Services Administration. Report to Congress. The critical care workforce: a study of the supply and demand for critical care physicians.http://bhpr.hrsa.gov/healthworkforce/reports/studycriticalcarephys.pdf. Accessed August 1, 2015.
 
Wunsch H. .Linde-Zwirble W.T. .Harrison D.A. .Barnato A.E. .Rowan K.M. .Angus D.C. . Use of intensive care services during terminal hospitalizations in England and the United States. Am J Resp Crit Care Med. 2009;180:875-880 [PubMed]journal. [CrossRef] [PubMed]
 
Wunsch H. .Angus D.C. .Harrison D.A. .et al Variation in critical care services across North America and Western Europe. Crit Care Med. 2008;36:2787-2793 [PubMed]journal. [CrossRef] [PubMed]
 
Stelfox H.T. .Hemmelgarn B.R. .Bagshaw S.M. .et al Intensive care unit bed availability and outcomes for hospitalized patients with sudden clinical deterioration. Arch Intern Med. 2012;172:467-474 [PubMed]journal. [CrossRef] [PubMed]
 
Chen L.M. .Kennedy E.H. .Sales A. .Hofer T.P. . Use of health IT for higher-value critical care. N Engl J Med. 2013;368:594-597 [PubMed]journal. [CrossRef] [PubMed]
 
Safavi K.C. .Dharmarajan K. .Kim N. .et al Variation exists in rates of admission to intensive care units for heart failure patients across hospitals in the United States. Circulation. 2013;127:923-929 [PubMed]journal. [CrossRef] [PubMed]
 
Marshall M.F. .Schwenzer K.J. .Orsina M. .Fletcher J.C. .Durbin C.G. Jr.. Influence of political power, medical provincialism, and economic incentives on the rationing of surgical intensive care unit beds. Crit Care Med. 1992;20:387-394 [PubMed]journal. [CrossRef] [PubMed]
 
Wagner D.P. .Knaus W.A. .Draper E.A. .Zimmerman J.E. . Identification of low-risk monitor patients within a medical-surgical intensive care unit. Med Care. 1983;21:425-434 [PubMed]journal. [CrossRef] [PubMed]
 
Huynh T.N. .Kleerup E.C. .Wiley J.F. .et al The frequency and cost of treatment perceived to be futile in critical care. JAMA Intern Med. 2013;173:1887-1894 [PubMed]journal. [CrossRef] [PubMed]
 
Neuman P. .Cubanski J. .Damico A. . Medicare per capita spending by age and service: new data highlights oldest beneficiaries. Health Aff (Millwood). 2015;34:335-339 [PubMed]journal. [CrossRef] [PubMed]
 
Angus D.C. .Linde-Zwirble W.T. .Lidicker J. .Clermont G. .Carcillo J. .Pinsky M.R. . Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med. 2001;29:1303-1310 [PubMed]journal. [CrossRef] [PubMed]
 
Rubenfeld G.D. .Caldwell E. .Peabody E. .et al Incidence and outcomes of acute lung injury. N Engl J Med. 2005;353:1685-1693 [PubMed]journal. [CrossRef] [PubMed]
 
Association of American Medical Colleges. The complexities of physician supply and demand: projections from 2013 to 2025.https://www.aamc.org/download/426242/data/ihsreportdownload.pdf?cm_mmc=AAMC-_-ScientificAffairs-_-PDF-_-ihsreport. Accessed August 1, 2015.
 
Abt Associates, Inc. Future needs in pulmonary and critical care medicine.http://www.abtassociates.com/reports/chest2.pdf. Accessed August 1, 2015.
 
Rivo M.L. .Jackson D.M. .Clare F.L. . Comparing physician workforce reform recommendations. JAMA. 1993;270:1083-1084 [PubMed]journal. [CrossRef] [PubMed]
 
Wise K.R. .Akopov V.A. .Williams B.R. Jr..Ido M.S. .Leeper K.V. Jr..Dressler D.D. . Hospitalists and intensivists in the medical ICU: a prospective observational study comparing mortality and length of stay between two staffing models. J Hosp Med. 2012;7:183-189 [PubMed]journal. [CrossRef] [PubMed]
 
Wilcox M.E. .Adhikari N.K. . The effect of telemedicine in critically ill patients: systematic review and meta-analysis. Crit Care. 2012;16:R127- [PubMed]journal. [CrossRef] [PubMed]
 
NOTE:
Citing articles are presented as examples only. In non-demo SCM6 implementation, integration with CrossRef’s "Cited By" API will populate this tab (http://www.crossref.org/citedby.html).

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging & repositioning the boxes below.

Find Similar Articles
CHEST Journal Articles
PubMed Articles
  • CHEST Journal
    Print ISSN: 0012-3692
    Online ISSN: 1931-3543