Organization is the act of assembling elements into an orderly, functional whole. Management is the ongoing revision and renovation of that careful assembly to cope with change. The concept of “bedside management” is familiar to clinicians who titrate vasopressors or adjust ventilator settings; intensive care unit (ICU) management is itself a form of titration and continuous adjustment. ICU management extends beyond simply implementing policies and procedures, organizing service and teaching rounds, preparing budgets, and complying with regulations. The successful ICU manager must also innovate and facilitate change. Creativity is important, but perseverance may be more essential because of the ways a typical organization will resist change. Knowing how to navigate the obvious and subtle impediments to change is a key skill for the ICU manager.

The already staggering cost of health care continues to escalate, and now represents 16% of the gross domestic product (GDP) in the United States, with estimates that unchecked, it could double again to 31% of GDP in the next 25 years [1]. Hospital costs are roughly a third of total health care costs, and intensive care alone consumes between 4% and 10% of total healthcare costs, or 0.56% to 1.5% of GDP [2,3,4]. One-third of Medicare patients spend part of their hospital stay in the ICU or coronary care unit, at an average unit cost per day of $2,616 (in 2004 US dollars) [5]. Discrepancies exist between the Medicare Provider Analysis and Review File (MedPAR) and the Hospital Cost Report Information System (HCRIS), two federal databases used to assess inpatient and critical care costs in the Medicare population [6]. In fact, critical care days may have decreased by 4.5% between 1995 and 2000 based on HCRIS data, while an increase of 7.2% was seen using MedPAR data, which includes “post/intermediate” billing codes [6]. Nonetheless, the Center for Medicare and Medicaid Services continues to forecast a substantial increase in the rate of growth in volume and intensity of medical services as the leading edge of the “baby boom” generation enters retirement [7]. Physician and nursing shortages [8] and increasing costs will constrain growth of intensive care services, while consumer demand (fueled in part by easy internet access to information) and an aging population with chronic disease will exacerbate existing capacity issues. New, unpredictable risks (e.g., novel bacterial and viral threats, terrorism) require preparedness and the ability to ramp up critical care capacity in a crisis. Meanwhile, attention continues to be focused on preventable medical errors. This confluence of events implies that attention must be paid to the health and well-being of the ICU in addition to addressing the needs of individual patients [9].

The conceptual frameworks [10] and business skills for successful ICU leadership must somehow be acquired, whether in business school or on-the-job. Important characteristics of leaders include self-awareness, self-regulation, motivation, empathy, and social skill [11]. The American College of Physician Executives is one organization that provides information on how to prepare for and succeed in medical management [12]. A formal Masters of Business Administration (MBA) program will typically include courses on accounting, data analysis, ethics, financial analysis, human resource management, information systems, marketing, production/operations management, organizational behavior, organizational planning and strategy, quality improvement, team building, and leadership. Given the difficulty in compressing a multiyear MBA curriculum into a book chapter, we will focus on typical ICU organization patterns, human resource issues, the roles of the ICU director, methods for monitoring clinical ICU care, and ancillary management issues.

In broad terms, there are three common models for ICU organization:

Pronovost et al. [14] conducted a systematic review of articles examining physician staffing patterns and clinical outcomes published through 2001. The model of care in each of 17 studies was classified as low intensity (no intensivist or elective consultation) or high intensity (mandatory critical care consultation or closed ICU). The high-intensity model was associated with lower ICU mortality (pooled mortality risk estimate 0.61) and lower hospital mortality (pooled mortality risk estimate 0.71). Although the literature overwhelmingly favors intensivist staffing models, a recent retrospective analysis of the Project IMPACT database by Levy et al. [15] demonstrated higher odds for hospital mortality in patients managed by critical care physicians. These counterintuitive findings have been challenged as being caused by unmeasured confounders including case mix differences [16] and the role of trainees and part-time academic faculty [17]. The higher risk-adjusted mortality in teaching hospitals where more invasive interventions are performed [18] may also counteract beneficial effects of full-time intensivists.

Case-control studies, where outcomes have been examined before and after implementing a closed model, offer additional insight into the value of intensivists. Patients admitted to closed units tend to be sicker [19,20], as might be expected with tighter triage criteria, although average severity scores are not necessarily higher in closed units [21]. Nursing confidence in physician clinical judgment improves [18], as a closed system allows the nurse to contact one managing physician rather than having to call the pulmonologist for ventilator changes, the nephrologist for fluid and electrolyte issues, and the cardiologist for arrhythmias. (Although, as Marik et al. [17] have pointed out, detrimental “parceling out” of care may occur in an academic setting even when full-time intensivists are present). These efficiencies are generally reflected in shorter ICU and hospital LOS [19]. The effect of dedicated intensivist staffing on ICU LOS remains significant after case-mix is adjusted for risk factors such as patient age, admission severity of illness, pre-ICU length of stay and percentage of patients requiring mechanical ventilation [22].

Staffing patterns, in terms of in-house, overnight coverage, also vary widely [23]. The benefits of around-the-clock (versus business hours) in-house intensivist coverage is uncertain, despite outcome differences documented as a function of ICU admission time and day of week [24,25,26,27,28]. At the hospital level, there is no statistically significant mortality difference based on time of admission for most (77%) diagnoses [29], including acute myocardial infarction, congestive heart failure, pneumonia, stroke, gastrointestinal bleeding, and many surgical conditions. Mortality was higher, though, in patients with ruptured abdominal aortic aneurysms, acute epiglottitis, and pulmonary embolus, when these patients presented on the weekend. This suggests that for at least some conditions, adverse effects occur because of decreased weekend staffing, lack of patient familiarity with cross-coverage, and perhaps less supervision. Around-the-clock intensivist coverage may reduce severity-adjusted mortality [30] but there is debate if the on-site physicians need to be intensivists, especially given the current shortage of specialists [31]. Introduction of continuous on-site intensivists improves processes of care and staff satisfaction, and decreases ICU complications and hospital length of stay [32].

Remote intensive care, using a telemedicine approach, has been proposed as a partial solution to the shortage of intensivists. Using intensivists and physician extenders to provide supplemental monitoring and management of ICU patients between noon and 7 AM, Breslow et al. were able to demonstrate reductions in hospital mortality (RR 0.73), ICU length of stay (3.63 vs. 4.35 days) and lower variable costs per case [33]. Given the critical care shortage of intensivists, tele–ICU systems can potentially permit these specialists to monitor more patients, and those patients who might not otherwise have access to an intensivist. Despite the shortage of data, Leapfrog Group and the University Health System Consortium have encouraged the application of tele-ICU [34]. Results from the first federally funded multicenter evaluation of tele-ICU of approximately 4,000 patients from before and after activation of a tele-ICU did not demonstrate any differences in adjusted hospital and ICU mortality, length of stay or ICU complications with telemedicine intervention [35]. Of interest, improved survival rates were observed in the sicker population while mortality for less severely ill patients was increased. A major limitation of this multicenter trial include limited authority delegated to the tele-ICU by the majority of attending physicians; that is choosing to limit the remote specialists to monitoring rather than direct intervention authority. In addition, the inability to share the ICU electronic medical records with the central facility could have potentially delayed implementation of tele-ICU orders. The mixed outcome benefit of telemedicine for the ICU noted in recent trials [36] may indicate that the actual mechanisms of implementing telemedicine in ICU may play a significant role as to its effectiveness. Understanding and identifying local hospital wide operations including ICU staffing levels, evaluation of standardized care processes if any and availability of computerized order entry capability may help identify which ICUs benefit from tele-ICU.

A hospital’s approach to ICU organization will depend on its patient population, existing professional talent, physical facilities, and economies of scale. Reimbursement for critical care and evaluation/management services typically cannot cover the cost of a dedicated intensivist in smaller units. Triage functions and general management of the unit (as opposed to management of individual patients) cannot be billed to patients, and thus does not generate professional revenue. However, there is ample evidence that hospital investment in physician intensivist services is recouped with better patient flow (reducing the need for additional ICU beds) and lower utilization of pharmacy, laboratory, and radiology services. Simply having an intensive care physician round daily on postoperative patients shortens LOS, reduces complications and lowers total hospital cost in patients undergoing esophageal resection [37] or abdominal aortic surgery [38]. Organizational restructuring of a cardiothoracic unit with an attending physician dedicated to ICU care resulted in reduced pharmacy, radiology, and laboratory utilization, and a per-patient decrease in hospital costs of $2,285 [39]. Pronovost et al. developed a financial model for 6-, 12-, and 18-bed intensive care units for hospitals transitioning over a 1-year period to the Leapfrog Group ICU physician staffing standard. Cost savings ranged from $510,000 to $3.3 million, depending on bed size [40]. Their best-case scenario results could generate up to $13 million in annual savings, while a worst-case scenario imposed net costs of $1.3 million.

Hiring full-time critical care specialists is already a challenge with the growing shortage of intensivists. Critical care work force needs have not been adequately addressed by public
policy [41]. Medicare payments often do not cover the costs of providing critical care [42,43]. Angus et al. predicted in 2000 that supply and demand of intensivists would remain in equilibrium until 2007, but that demand would subsequently grow, producing serious shortfalls by 2020 [44]. The Society of Critical Care Medicine conducted a survey of 731 critical care physicians in 2004. These respondents planned to retire at an average age of 62 years, and to change focus or reduce patient load beginning in their fifties [45]. Nearly 40% of the respondents were already over the age of 45. Their average workweek was 66 hours, with a typical shift of 10 to 12 hours, providing clinical care an average of 48 weeks per year. It is unclear that the next generation of intensivists will continue to work at this level of intensity, or that critical care will be a viable career choice when remuneration is better for specialties with shorter working hours and less stress.

Current Leapfrog Group standards call for in-house intensivist staffing for a minimum of 8 hours, 7 days per week [13,46] or ≥ 2,920 hours per year to cover one ICU, with requirements for off-hours coverage met by an intensivist on beeper call, with an FCCS-certified physician or physician extender immediately available in-house. Hospitalists with FCCS certification can also potentially provide off-hours ICU coverage. In a retrospective study of care provided during after-hours coverage of a pediatric intensive care unit, Tenner et al. found improved survival with hospitalists compared with housestaff [47].

It is helpful to consider the concept of a clinical full-time equivalent (FTE) to represent the amount of work done by one individual working only on direct patient-care tasks in the intensive unit. In reality, some ICU clinicians will also allocate professional time to research, administration, or education; choose to work part time, or spend part of their clinical time on the trauma team, in the pulmonary clinic, or administering anesthesia. A full-time physician working only in the ICU might have grant funding for 0.25 FTE, and another 0.25 stipend for administrative and educational activity, leaving 0.5 FTE for ICU clinical activity.

How many hours will one FTE work in a year? The SCCM respondents’ reports annual work hours from less than 1,000 to more than 4,000, but most commonly 2,000 to 2,500 hours [44]. Since attractive jobs currently offer at least 4 weeks vacation, about 10 paid holidays and at least 5 days of meeting time, we’ll consider annual work to be 45 weeks with 10-hour days, yielding 2,250 hours, in accord with the range reported in the SCCM survey. If in-house coverage for the ICU is around-the-clock, 365 days per year, with 30 minutes overlap at the beginning and end of 12-hour shifts, then annual hours to be covered are 9,490. Thus, 4.2 FTEs would be needed to cover the clinical workload. This workload might be met by five physicians, assuming each worked full time and 0.8 FTE was sufficient to attend to administrative and quality assurance activities. If coverage is only during the daytime (3,650 hours per year) fewer FTEs would be required; although on-call hours must still be staffed.

Staffing calculations must consider intensivist-to-patient staffing ratios, which are not well-defined. In England and Wales, where intensivists staff 80% of ICU’s, the average six-bed general ICU has three consultants committed to the unit, and another three consultants participating in the on-call rotation [48]. A retrospective study from the Mayo Clinic [49] did not find differences in the severity-adjusted mortality rate at daytime intensivist-to-bed ratios between 1:7.5 and 1:15 although ICU length of stay increased at the higher extreme. Larger hospitals with closed units may take advantage of cross-coverage between units, providing daytime care at intensivist to patient ratios of 1:8 to 1:12; and increasing the ratio during off-hours when there are fewer acute interventions or procedures to be accomplished.

The enormous work force requirements and economic burdens of providing round-the-clock critical care staffing has led physician leaders, hospital administrators, and insurance companies to re-examine models of health care delivery. Some medical centers now employ physician extenders on the critical care team as a response to physician shortage at both the attending and house-staff level. Physician extender is a broad term covering mid-level health care providers such as nurse practitioners (NPs) and physician assistants (PAs). Physician assistants must complete an accredited education program, usually 2 years in duration, but often requiring prior college and health care experience. PAs must pass a national examination to obtain a license, and always work under a physician’s supervision. A nurse practitioner is a registered nurse who has completed advanced training and must be licensed in the state where practicing. Following state licensure, NPs may seek national certification from professional nursing boards and/or pursue specialty certification. NPs have more latitude to practice independently.

Driving forces that have accelerated employment of the physician extenders include cutbacks in federal funding for residency training, identifiable patient care needs, and ACGME standards placing strict limits to the number of hours that medical trainees can participate in providing care. Physician extenders can provide safe and cost-effective care as part of a collaborative medical management team in acute care settings and they are well received by patients, nurses, physicians, and administrators. A limited number of studies suggest that introduction of NP/intensivist team-based care is beneficial to patient outcomes, financial outcomes, length of stay, and patient satisfaction [50]. An attending physician/NP team can safely manage former ICU patients admitted to a subacute unit therefore allowing the intensivist/fellow team time to care for higher acuity ICU patients [51]. Decreased overall length of stay and ICU length of stay, lower rates of UTI and skin breakdown, and a shorter time to mobilization have been documented after introduction of an NP team to neuroscience ICUs [52]. NP participation in weaning protocols for mechanical ventilation has been associated with greater reductions in mechanical ventilation days, ICU length of stay, and hospital length of stay when compared to pre-NP participation [53]. NPs and physicians in training had equivalent efficacies in performing required tasks but residents spend more time in nonunit activities (lectures, rounds) and NPs spend more time monitoring patients, talking to families, and collaborating with other health team members [54].

A team-oriented culture characterized by timely communication is associated with a shorter length of ICU stay, greater ability to accommodate the needs of patient families, and a higher quality of technical care [55]. Including PAs on house-staff-directed ICU teams does not appear to affect rates of occupancy, mortality, or complications [56].

Intensive care services are among the most urgent and costly aspects of healthcare in the United States, and national surveys indicate the need to accommodate about 50,000 patients a day [43]. Professional societies are projecting an inability to meet this demand with intensivists, so the role of physician extenders will need to be further examined as a major component of the healthcare delivery model for critically ill patients.

The Joint Commission on Accreditation of Health Care Organizations (JCAHO) requires that an individual be designated as the ICU Director, but actual job descriptions vary. At one
extreme, the medical director may simply approve critical care policies and serve as a resource for questions that cannot be solved by nursing administration. He or she may triage only in times of high census, and may have very little role in the delivery of critical care, other than to his or her own patients. At the other extreme, the medical director may lead the team of intensivists that assumes total responsibility for all patients occupying ICU beds. Nonclinical duties may consume more work effort than clinical responsibilities when committee membership, administrative tasks, budget preparation, educational activities, and the business of running the ICU physician practice are included. When the medical director is heavily involved in day-to-day operations, ICU occupancy rates and number of patients misallocated to ICU beds decline [57]. ICU admission decisions are only part of the triage function. One in six patients experience ICU discharge issues (unexpected medical deterioration, level of care issues, administrative problems [58]) that demand executive resolution.

Larger hospitals typically have multiple intensive care units, each with its own director. The directors or designees may participate in a hospital-wide Critical Care Committee that sets overall policies and procedures. In some units, the medical director may delegate administrative tasks, quality improvement, education, and research to associate medical directors. Typically, the ICU director(s) will have a close working relationship with the nursing unit manager in each unit. Multidisciplinary units will involve interaction with other professionals (pharmacists, dieticians, social workers, clergy, utilization management specialists) and the medical director may have an overall coordinating role. Essential character traits of the successful ICU director include willingness to collaborate, ability to delegate, trust in colleagues, and excellent communication skills.

Tasks performed by the ICU Director can best be divided into strategic versus tactical (Table 208.1). Strategic tasks involve the “big picture”: recognition of patterns and trends, setting priorities, considering alternatives, and implementing change. The ICU Director is often the champion for process improvement projects. Areas deserving of strategic consideration include cost containment, the overall culture of the ICU, quality improvement efforts, education of physicians, nurses and other health professionals, and coping with change driven by ICU, hospital and external factors [59]. Developing a strategic vision and communicating it well are essential roles. Yet, it is equally important to lead by example, particularly when it comes time to drive change, such as implementing electronic medical records or computerized physician order entry.