Special Feature
Should All ICUs Have 24/7 In-house Intensivist Coverage?
By David J. Pierson, MD, Editor, Professor, Pulmonary and Critical Care Medicine, Harborview Medical Center, University of Washington, Seattle, is Editor for Critical Care Alert.
One of the hottest topics in critical care these days is whether all ICUs should be staffed around the clock, seven days a week (24/7), by physicians with special training and qualifications in critical care (intensivists). Four recent editorials1-4 offer different perspectives on this topic and highlight its relevance to the readers of this newsletter. In this essay I will try to identify the key issues involved and summarize the pros and cons of 24/7 in-house intensivist coverage. The issues are not as straightforward, nor is the evidence as clear, as one might wish, and key aspects of the matter are beyond the control of individual clinicians working in the ICU.
Origins of the Controversy
The current push for 24/7 intensivist staffing in ICUs has multiple origins, but its most prominent driver in the United States has been the Leapfrog Group. This is a consortium of health care purchasing organizations formed in response to ever-increasing health care costs in order "to trigger giant leaps forward in the safety, quality and affordability of health care by supporting informed healthcare decisions by those who use and pay for health care and promoting high-value health care through incentives and rewards."5 One of Leapfrog's core intended "leaps" is for all ICUs to be staffed by intensivists, although it does not specifically call for the physical presence of an intensivist in the ICU 24/7. However, a 2000 survey found that only 4% of U.S. ICUs met the Leapfrog physician staffing standard,6 and available evidence suggests that while the proportion may be higher today, it is still far short of the goal. In fact, only a minority of critically ill patients receive any care by an intensivist, let alone 24/7 care.2,6 In the United States there are numerous reasons for this, but a big one is that there are not enough qualified intensivists to go around.
ICU Physician Staffing Models
How ICUs are staffed by physicians and how this staffing affects the outcomes of critical care have been the subject of considerable study. Rubenfeld and Angus describe several models for such staffing,7 depending on the degree of involvement of intensivists in the care of patients in a given ICU. These include the following distinctions:
- Admission and discharge decisions and all patient management in the unit are by intensivists (closed ICU) vs admission and discharge decisions by the patient's primary physician, with or without involvement of intensivists in management (open ICU);
- Closed ICU as above, or mandatory intensivist consultation for all patients (high-intensity ICU) vs open ICU as above, with or without intensivist consultation or availability (low-intensity ICU);
- At least 80% of patients managed by intensivist (high-intensity ICU) vs some but < 80% of patients managed by intensivist (intermediate-intensity ICU) vs no intensivist;
- Closed ICU, or open ICU with mandatory intensivist consultation ("no choice" ICU), vs open ICU with consultation of intensivist at discretion of primary physician ("choice" ICU) vs no intensivist available ("no choice" ICU).
Perhaps three-fourths of all critically ill patients in U.S. ICUs are managed under the low-intensity physician staffing model defined above.8 However, most studies evaluating the association between ICU physician staffing and patient outcomes have shown lower hospital mortality and shorter ICU and hospital stays with high-intensity physician staffing.8 A prominent exception, published 2 years ago, examined data from more than 100,000 patients in 123 American ICUs participating in Project Impact, a voluntary critical care database.9 That study found that high-intensity ICU staffing (defined as having at least 95% of all patients cared for exclusively by an intensivist) was associated with higher, rather than lower, severity-adjusted mortality as compared to low-intensity ICUs (defined as 5% or less of critically ill patients receiving exclusively intensivist care). As discussed in a previous special feature in this newsletter (see Critical Care Alert, September 2008, pp. 46-48), there are a number of potential explanations for these findings, including unaccounted-for differences in patient populations, severity and complexity of illness, and processes of care among the Project Impact ICUs.
The majority of studies dealing with ICU staffing have looked at patient outcomes before and after full-time intensivist coverage was introduced to the ICU. Usually when this happens, numerous other things change besides just the addition of intensivists to the unit such as admission and discharge criteria, the use of protocols, and emphasis on evidence-based practice, to name just a few. Typically, multiple aspects of patient care improve when such changes are introduced, and it should be no surprise that better care results in better outcomes. These and other study design issues notwithstanding, though, the weight of available evidence strongly favors the closed-ICU and high-intensity intensivist staffing models.
If care by intensivists is associated with better outcomes, it stands to reason that having those intensivists in the unit 24/7 would be the best possible staffing arrangement. However, here the picture is less clear, and the supporting evidence more equivocal, than with the simple presence or absence of intensivist care.
Pros and Cons of 24/7 In-house Coverage
Table 1 (below) summarizes the reasons that have been advanced for and against implementation of 24/7 in-house intensivist coverage.1-4,8,10 Extrapolating from the evidence in support of intensivist care in general, advocates point out that critical illness does not respect "business hours," and that at least as many patients are admitted to ICUs at night and on weekends as during usual daytime hours. Evidence supporting early, aggressive intervention in sepsis, acute myocardial infarction, and trauma is typically offered in support of having an intensivist at the bedside whenever these or other critical illnesses are diagnosed. A number of studies document increased ICU mortality during nights and weekends, and also among patients initially admitted during these times. In contrast, several reports from units with 24/7 intensivist staffing find no increase in mortality during off-hours. Mortality among patients cared for in academic, teaching-hospital ICUs has been reported to be higher than in non-teaching settings, and adjustments for patient demographics and other factors have led to the conclusion that being managed by housestaff accounts for the difference. All these findings have been cited in support of 24/7 intensivist coverage. The devil, however, is in the details, and none of these arguments is as clear-cut as it first appears. For example, patients admitted at night and on weekends tend to be sicker than those admitted during the day, and teaching hospitals care for more complex, more severely ill patients than nonteaching hospitals.
| Table 1. Pros and Cons of Implementing 24/7 Intensivist Staffing in the ICU. | |
Pro: Potential Reasons for Implementing
|
Con: Potential Reasons for Not Implementing
|
One single-center, before-and-after study found that processes of care in the ICU were improved after 24/7 intensivist coverage was implemented.11 There was better adherence to evidence-based practice guidelines, ICU-related complications decreased, and hospital length of stay was shortened. Staff satisfaction also improved under the new physician staffing paradigm. It may be argued, however, that the improved outcomes associated with better implementation of evidence-based guidelines and systems approaches to reducing ICU infections have also been reported in other studies unrelated to physician staffing, so that it may be possible to achieve these improvements without the 24/7 presence of an intensivist. That is, systems approaches to implementing, for example, early goal-directed therapy for sepsis and lung-protective ventilation for acute lung injury may or may not require physical presence of the intensivist at the bedside.
Critical care is a multidisciplinary activity necessarily involving multiple individuals in the management of an individual patient. Still, several aspects of the shift approach to physician coverage pose challenges to both optimal patient care and efficient staff functioning. Handoffs between shifts have been identified as important sources of medical error, and the potential effects on patient and family satisfaction of multiple switches between managing physicians are important considerations in implementing 24/7 intensivist coverage. Also important are the effects on the intensivists themselves. In academic settings off-hours coverage may fall disproportionately on junior faculty members, interfering not only with work-life balance for these individuals,12 but also with other activities upon which their long-term career prospects depend.4 It is likely that analogous pressures affect more junior members of practice groups in non-academic settings as well. In addition, the prevalence of burnout and other consequences of work stress may have an adverse effect on the attractiveness of critical care medicine as a career choice for residents.
Finally, prominent in any discussion of the possibility of implementing 24/7 intensivist coverage are the problems of money and manpower. Staffing an ICU around the clock with qualified intensivists requires a substantial commitment of resources on the part of the institution, with as many as 5 physicians required to fill one full-time equivalent position 365 days per year.10 This is a big practical problem for the individual institution, albeit a potentially solvable one, but at the regional and national levels the shortage of qualified intensivists renders implementation of 24/7 intensivist staffing for all ICUs an impossibility at least in the short term.
Options for Optimizing ICU Physician Staffing
Table 2 (below) lists 6 different options for staffing a "closed" ICU. All of these are currently used in the greater metropolitan area in which I practice. Of the medical ICUs in the main teaching hospitals of our university system, one relies on the first option and the other uses the fourth listed in the table, with "nocturnalists" (hospitalists, typically recent medicine residency graduates) covering the unit and supervising the residents at night. Other ICUs in these same institutions use other models on the list, as do the community hospitals around us. Some of our senior critical care fellows moonlight in option #3. One of the largest hospitals in the region recently went to 24/7 intensivist coverage, and another relies on ICU telemedicine.
Table 2. Potential Options for ICU Staffing |
1. Current "business model" staffing: intensivists present 12h, on-call 12h
2. Intensivists present in unit 24/7 3. Senior-level critical care trainees staff unit during off-hours (moonlighting) 4. Hospitalists (nocturnalists) staff unit during off-hours 5. Mid-level practitioners (physician assistant, nurse practitioner) staff unit during off-hours 6. Telemedicine (with one or more ICUs covered electronically by off-site intensivist, the "E-ICU") |
Given the unfeasibility of providing an intensivist 24/7 for every ICU, the practical thing to do would seem to be to deploy intensivists where they can do the most good for the largest number of critically ill patients. To this end, several changes have been proposed for the way critical care is organized in the United States.13 Because outcomes for patients with a given condition tend to be better at hospitals that manage more patients with that condition, the first proposal is tiered regionalization. This would involve categorizing hospitals according to the level of critical care they can provide, and systematically transferring high-risk patients to higher levels of care. For example, one study based on simulation of regionalized care for the majority of nonsurgical patients requiring mechanical ventilation found that managing them at a few high-volume centers would confer a substantial survival advantage.14 For regionalization to be effective, however, would require major financial and organizational investment and the will to deploy it.13
A second systems-level organizational change would be more extensive utilization of ICU telemedicine, which is in use in some areas but not currently widespread. Although the evidence that telemedicine improves outcomes is inconsistent, this approach to intensivist staffing is particularly appealing for smaller hospitals and those geographically distant from large population centers.13 Again, though, major changes in health care organization and clinician acceptance would be necessary if ICU telemedicine were to be broadly implemented.
A third approach to improving the quality of critical care and more efficiently utilizing the available intensivist workforce is quality improvement through regional outreach.13 Instead of physically moving patients to higher-expertise hospitals or importing physician expertise via telemedicine, regional outreach involves cooperation among all hospitals in a given region to improve outcomes through benchmarking and quality improvement. Several studies have shown the effectiveness of this strategy in such areas as improving staff performance, increasing adherence to evidence-based guidelines, and reducing ICU-related complications.13 For regional outreach to work, however, requires not only the willingness of different institutions to cooperate, but also the presence of an adequately supported central coordinating authority to facilitate education and data collection.
Nguyen and colleagues have recently described an organizational framework for implementing all three of the above systems approaches.13 Such implementation would require a unified national strategy for improving ICU care. In the meantime, even if one is convinced that it is the best thing to do, deciding to move to 24/7 intensivist coverage involves juggling the benefits and costs at the level of the individual hospital. Adopting 24/7 intensivist coverage for a particular ICU may be possible given enough financial support from the institution and enough trained critical care physicians in the community. However, although some regions of the country may be better stocked with intensivists than others, nationwide there are definitely not enough to go around. The staffing options listed in Table 2 provide possible alternatives, at least until this issue is addressed in a more global way and/or more definitive evidence becomes available.
References
- Burnham EL, et al. Am J Respir Crit Care Med 2010;181:1159-1160.
- Kahn JM, Hall JB. Am J Respir Crit Care Med 2010;181:1160-1161.
- Cartin-Ceba R, Bajwa EK. Am J Respir Crit Care Med 2010;181:1279-1280.
- Jones SF, Gaggar A. Am J Respir Crit Care Med 2010;181:1280-1281.
- The Leapfrog Group. Available at: www.leapfroggroup.org/about_us. Accessed July 9, 2010.
- Angus DC, et al; Committee on Manpower for Pulmonary and Critical Care Societies (COMPACCS). Crit Care Med 2006;34:1016-1024.
- Rubenfeld GD, Angus DC. Ann Intern Med 2008;148:877-879.
- Gajic O, Afessa B. Chest 2009;135:1038-1044.
- Levy MM, et al. Ann Intern Med 2008;148:801-809.
- Arabi Y. Crit Care 2008;12:216; doi:10.1186/cc6905.
- Gajic O, et al. Crit Care Med 2008;36:36-44.
- American Thoracic Society work-life balance initiative. Available at: www.thoracic.org/education/career-development. Accessed July 10, 2010.
- Nguyen YL, et al. Am J Respir Crit Care Med 2010;181:1164-1169.
- Kahn JM, et al. Am J Respir Crit Care Med 2008;177:285-291.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content