Navigate Up

UPMC/University of Pittsburgh Schools of the Health Sciences

End Of Life ICU Utilization May Require Re-Evaluation According To University Of Pittsburgh-Led Study

One of Five Terminally Ill Americans Dies In an ICU

PITTSBURGH, April 6, 2004 Although most Americans say they would prefer a low-tech approach to death, the opposite is happening with more than 20 percent of terminally ill patients dying in intensive care, according to an article in the March issue of Critical Care Medicine, the journal of the Society of Critical Care Medicine.

This is the first study to provide national scale to the often discussed end-of-life public health issue, says Derek Angus, M.D., Ch.B., M.P.H., professor of critical care medicine and vice chair of research in the department of critical care medicine at the University of Pittsburgh School of Medicine. Previously there were no reliable estimates of the magnitude of this situation.

In light of their findings, the researchers are concerned that hospitals and intensive care units (ICU) are not prepared to deliver end-of-life care to the population older than 65 years, which is rapidly expanding.

The researchers analyzed hospital discharge data from 1999 for six states (Florida, Massachusetts, New Jersey, New York, Virginia and Washington) and the National Death Index to estimate the use of critical care at the end-of-life.

The researchers found that of the 552,157 deaths they were studying, 38.3 percent occurred in the hospital and 22.4 percent occurred following ICU admission. Nationally projected, this means more than one-half million Americans die annually following ICU admission.

End-of-life ICU use was highest (43 percent) among infants, ranged from 18 percent to 26 percent among older children and adults and fell to 14 percent for people older than 85 years.

Additionally, the researchers found that terminally ill ICU patients had an average length of stay of 12.9 days at a cost of $24,541 and non-ICU hospitalization of the terminally ill had an average stay of 8.9 days at a cost of $8,548.

Nine out of 10 Americans say they would prefer to die at home; however, 20 percent die after receiving the most technologically advanced care available while in an ICU. This contradiction may be explained by the uncertainty surrounding time-of-death predictions, which is particularly true in acute-care settings where advanced technology is used to prolong life.

The number of Americans who are older than age 65 is expected to double by the year 2030, putting a great of deal stress on ICUs. The researchers recommended a system-wide expansion in ICU care for dying patients unless the healthcare system pursues an alternative approach.

Intensive care plays a major role in end-of-life care in the United States, says Dr. Angus, who also is director of the Critical Research, Investigation, and Systems Modeling of Acute Illness (CRISMA) Laboratory at the University of Pittsburgh School of Medicine. We have an enormous social and medical imperative to take very seriously our efforts to make sure critical care is only provided when appropriate and provided in a way that has all the elements of compassion and humanism necessary to treat and manage end-of-life care with dignity.

Dr. Angus recommends improving prediction models, which help reduce ICU admission at the end of life for patients with poor prognosis. He suggests critical care physicians partner with referring physicians to consider risk prevention models before admitting terminally ill patients who cannot benefit from critical care.

Even with excellent prediction models, we will still admit patients at high risk of dying because they also have a good chance of survival, says Dr. Angus. There will still be a large, irreducible number of patients who die in intensive care. This means we have to carefully consider providing and promoting good end-of-life care in the ICU.

Dr. Angus says it is difficult for patients, families and physicians to strike a balance between not wanting to die and not wanting to die badly.

Some groups of patients are less likely to die in the ICU than others, says Dr. Angus. For instance, HIV/AIDS patients and cancer patients use the ICU less often at the end of life, presumably because they have more clearly established guidelines for their end-of-life care. However, other groups who end up in the ICU at the end of life may do so without forewarning or forethought. Better awareness of what ICU care can offer, and better a priori discussions between patients and their healthcare providers might promote more efficient use of intensive care services.

Along with some other recent data demonstrating interventions for improving the quality of end-of-life care in the ICU, the data from Angus, et al provide the basis for a new era in ICU end-of-life care: one in which we move from simple descriptive studies to acknowledging the magnitude of the problem and developing specific, practical interventions aimed at enhancing the quality of care received by dying patients and their loved ones, said Mitchell M. Levy, M.D., author of the accompanying editorial. It is time to move our understanding of the barriers to quality end-of-life care from the bench to the bedside and make our care for those dying in the ICU something that is inevitable for almost 20 percent of Americans more humane, compassionate and of the highest quality. Dr. Levy is professor of medicine at Brown University in Providence, R.I.

Critical Care Medicine is the official journal of the Society of Critical Care Medicine. It is the premier peer-reviewed, scientific publication in critical care medicine. Each issue presents critical care practitioners with clinical breakthroughs that lead to better patient care, the latest news on promising research, and advances in equipment and techniques.

This research was supported by a grant from The Robert Wood Johnson Foundation, Princeton, N.J. In addition to the University of Pittsburgh School of Medicine, researchers from the University of Pittsburgh Graduate School of Public Health, Health Process Management in Doylestown, Pa. and Harborview Medical Center at the University of Washington, Seattle contributed to this study.

UPMC | Affiliated with the University of Pittsburgh Schools of the Health Sciences Supplemental content provided by Healthwise, Incorporated. To learn more, visit

For help in finding a doctor or health service that suits your needs, call the UPMC Referral Service at 412-647-UPMC (8762) or 1-800-533-UPMC (8762). Select option 1.

UPMC is an equal opportunity employer. UPMC policy prohibits discrimination or harassment on the basis of race, color, religion, ancestry, national origin, age, sex, genetics, sexual orientation, gender identity, marital status, familial status, disability, veteran status, or any other legally protected group status. Further, UPMC will continue to support and promote equal employment opportunity, human dignity, and racial, ethnic, and cultural diversity. This policy applies to admissions, employment, and access to and treatment in UPMC programs and activities. This commitment is made by UPMC in accordance with federal, state, and/or local laws and regulations.

Medical information made available on is not intended to be used as a substitute for professional medical advice, diagnosis, or treatment. You should not rely entirely on this information for your health care needs. Ask your own doctor or health care provider any specific medical questions that you have. Further, is not a tool to be used in the case of an emergency. If an emergency arises, you should seek appropriate emergency medical services.

Pittsburgh, PA, USA |