Jeffrey Groeger, MD, FCCM
Urgent Care Service,
Memorial Sloan Kettering Cancer Center
New York, NY, USA

In 1991, the Society of Critical Care Medicine (SCCM) conducted a national survey of critical care units in the United States. The National Survey of Critical Care Resources (NSCCR) was designed to provide data on the resources, staffing policies and patients in each of these units during a typical day. The goal was to develop a comprehensive database of information about critical care resources.

The Society determined it was time to update the data gathered in 1991 and has now analyzed data from a new survey.  In addition to the resources, staffing and clinical information obtained in 1991, the Society also gathered information on critical care policies, practices and models of care used in hospitals within the United States.

In late 2003, surveys were mailed to 5,700 administrators and clinicians throughout the United States. The response rate was 13%. Six hundred eighty-two of the American Hospital Association’s (AHA) 4,430 hospitals submitted information.

Allow me to highlight some of the exciting conclusions of this survey. As was the case in 1991, the majority of respondents were nurses (71%), and the vast majority of the units represented were found in non-government, non-for-profit hospitals. The largest number of responses came from combined medical-surgical-cardiac units that represented the single intensive care unit of the responding hospital. These units represented approximately 9,700 fully operational intensive care unit (ICU) beds, and we have information on the clinical characteristics of about 7,700. Thirteen percent of the patients were younger than age 21, an additional 39% were between the ages of 21 and 64, and the remaining 48% were older than age 65. Respiratory insufficiency-failure was the predominant indication for ICU admission (19%), followed by postoperative management (14%). Thirty-eight percent of all patients in the responding units were being mechanically ventilated with 21% of them on an intravenous vasoactive or inotropic agent.

In total, 682 units supplied information on the type of management model utilized in patient care.  Drs. Martin Strosberg and James Lambrinos of the Graduate College of Union University presented this information at the Society’s annual meeting in January 2005 as evidence that as the medical director’s involvement in ICU management increases, ICUs are managed in a more successful manner.  In addition, variables such as presence and availability of the medical director, as well as medical director involvement in functions such as admissions and unit conflicts, also were positively correlated with successful ICU management.

Critical care medicine is no longer a field of anecdotal management, but rather one of peer-reviewed, evidence-based practice plans. The survey solicited information on whether general guidelines existed for tight glucose control, venous thrombosis prophylaxis, transfusion thresholds, aseptic precautions to decrease catheter site infection, and others. It solicited from ventilated patients whether they received daily spontaneous breathing trials, low tidal volumes for acute respiratory distress syndrome (ARDS), and even protocols for the maintenance of the elevation of the head of bed to prevent hospital-acquired pneumonia. 

The most prevalent guidelines were for aseptic precautions to decrease venous catheter site infection (79% of units) and in ventilated patients to keep the head of the bed elevated (71% of units). Where these guidelines existed there was a wide range as to whether they were occasionally followed, never followed or always followed.

More details from the survey are summarized under the following headings:

Operational Model. Though there is evidence that the best practice model of an integrated team of dedicated experts continues to be adopted and gain recognition, intensivists remain unavailable to most ICUs the majority of the time, and fewer than half of critical care units have a full-time medical director capable of supporting the integrated team. Most critically ill and injured patients still do not have access to the model of care clinically demonstrated to provide optimal outcomes.

Workforce. Staff shortages continue to be a significant problem, and there is evidence that bed closings due to the nursing shortage have increased since the 1991 NSCCR. In fact, nurse shortages account for more bed closings than all other causes combined.

Bed Capacity and Occupancy. Total ICU bed numbers in the United States have not changed significantly since the 1991 survey, but bed organization has and there is increased access to support beds. Despite such changes, ICUs face increasingly serious problems. Overcrowding and emergency vehicle diversion continue to plague ICUs as patients are turned away from the emergency department (ED) because full ICU beds prevent the transfer of current patients out of the ED.

Technology in the ICU. The lack of set standards remains the key problem affecting the use of technology in ICUs. It further complicates the question of how best to use available technology to improve outcomes and increase management efficiency.

Patient Population. While the share of patients older than age 65 years has not changed dramatically since the 1991 survey, there has been a significant increase in patients older than age 84. Also, data on primary diagnosis upon admission and current treatments provide some evidence supporting an increase in the severity of illness among ICU patients.

More information is included in the book Critical Care Units: A Descriptive Analysis available from SCCM. Look forward to seeing these data in monograph format as well as in peer-reviewed manuscripts.

To order a copy of Critical Care Units: A Descriptive Analysis call SCCM Customer Service at +1 847 827-6888.