Too Many Americans Die Alone, in Pain, Attached to Machines

Ads

Efforts to prolong people's lives too often merely prolong dying, say researchers, citing findings from the largest clinical study ever conducted in the United States with patients near death.

The eight-year study, involving nearly 10,000 seriously ill patients in five leading medical institutions, was reported in the November 22/29, 1995 issue of JAMA (The Journal of The American Medical Association).

The principal problem, the researchers say, is woefully inadequate communication between medical professionals and patients near the end of life. Doctors don't clearly convey patients' chances for survival; and patients and families don't discuss their wishes soon -- or often -- enough.

"There is a tragic mismatch between the health care many seriously ill and dying people want and what they get," said study director Dr. William Knaus, critical care specialist, professor, and department chair at the University of Virginia Medical School. "They find themselves in hospital environments that are geared to doing certain high-tech, invasive procedures extraordinarily well. Such treatments can preserve life for conditions that would have previously resulted more quickly in death. These same techniques, however, can also prolong dying and make it a painful and mechanical process. The problem is that the system does not know when or how to stop."

These new findings are particularly alarming because, in one part of the study, patients and families were provided with a nurse-advocate specially trained to encourage communication among patients, their family, and doctors. The results were contrary to outside experts' assumptions about what would make it easier to curtail aggressive medical care, increase the attention given to pain control and other symptoms, and cut hospital costs.

The JAMA article summarizes findings from the Study to Understand Prognoses and Preferences for Outcomes and Risks of Treatments (SUPPORT). The $28 million SUPPORT study, funded by The Robert Wood Johnson Foundation, is the largest, most rigorous study ever conducted with patients facing the end of life.

Study patients were from Beth Israel Hospital, Boston, Mass.; MetroHealth Medical Center, Cleveland, Ohio; Duke University Hospital, Durham, N.C.; Marshfield Clinic/St. Joseph's Hospital, Marshfield, Wisc.; and the University of California at Los Angeles Hospital. Researchers at George Washington University Medical Center in Washington, D.C. -- led by Dr. Knaus (formerly of that institution) and Dr. Joanne Lynn, professor of Health Care Sciences and Medicine and the director of the Center to Improve Care of the Dying at GWU -- coordinated the project and interpreted the results.

"These hospitals, the patients, and their families, were willing to take on a highly controversial topic and to let us look at their processes in minute detail," Dr. Knaus said. "We believe many other U.S. hospitals have similar problems."

A new Robert Wood Johnson Foundation-funded survey, conducted by researchers at Harvard University last month, underscores the viewpoint of Dr. Knaus. The national survey asked about the experience of families of patients who died recently either in a major medical center (like those in the SUPPORT study) or in another kind of hospital. Preliminary data indicate that the problems documented in SUPPORT are probably national in scope.

SUPPORT had two phases. Phase one involved 4,301 hospitalized patients who, on average, could be expected to survive no more than six months. Patients had one of nine life-threatening conditions, including cancers, congestive heart failure, and chronic liver disease. Deaths from these nine causes account for more than one-third of adult U.S. deaths annually. In phase one, which took place between 1989 and 1991, researchers documented the kind of care provided to these seriously ill patients and how closely their wishes were followed.

Phase one produced the following important findings:

Doctors didn't know their patients' wishes. For example, if patients did not want cardiopulmonary resuscitation (CPR), doctors didn't know about it half the time (47%).
46% of do-not-resuscitate orders were not written until a day or two before death, suggesting little advance planning in terminal illnesses. "We don't decide to let patients die in peace until almost the last moment," Dr. Lynn said. "This is hard on patients, their families, and the health professionals who care for them."
38% of patients spent at least 10 days in intensive care, where aggressive treatment is the norm.
Pain was common. Half the patients able to communicate in the last three days of life said they were in severe pain.

Phase two -- a subsequent two-year clinical trial -- conducted from January 1992 to January 1994 involved 4,804 seriously ill patients. Patients were enrolled either in a "control group" that received customary care or an "experimental group," in which doctors and families had special help of three kinds:

A nurse whose job was to talk to patients and families about their concerns and to keep lines of communication open and clear between them and the hospital and physician staff;
Computer-generated reports estimating a patient's probability of surviving six months and how disabled they would be in the next two months; and
Detailed, written instructions on the patient and the family's wishes regarding treatment -- including pain control and "heroic measures" like resuscitation -- and how much they wanted to discuss their treatment with their doctor.

Phase two developed these interventions based on the kinds of information that doctors said they needed regarding patients' medical status and wishes for treatment. "We were stunned to find that it didn't make a bit of difference," Dr. Knaus said. "The tools that experts thought would work didn't."

The three-part intervention did not affect the amount of time patients spent in the intensive care unit, in a coma or on an artificial breathing machine -- a ventilator -- before dying. Reports of pain increased. The costs of care did not change.

"In Phase two of SUPPORT, improved information, enhanced conversation, and an explicit effort to encourage use of outcomes and preferences in decision-making were completely ineffectual," the research report states," despite the fact that the study had enough power to detect small effects."

"The reasons for this are ingrained in our society," Dr. Lynn said. "Physicians are taught to save lives, that death is a failure. Patients and families have come to expect miracles in every case. It's easier for everyone -- professionals and patients alike -- to follow the usual path of aggressive treatment, even when it's clear that it is leading nowhere. No one wants to give up too soon. That's one reason why everyone is so reluctant to discuss dying."

"Improving care of seriously ill hospitalized adults represents a serious and ongoing national challenge," Dr. Lynn said. "We need to create measures to assess quality of care while dying, and change financial incentives to promote better care while dying. Suffering while dying must become a bad outcome in the health care system. Everyone has a vision for living. Our society also needs to create a vision for living well while dying."

The study's authors suggest establishing end-of-life practice guidelines, teaching medical students and physicians that technological approaches to dying can harm patients, teaching palliative care techniques that enhance life at its end, encouraging hospitals to examine patterns of care for dying patients, and emphasizing the new responsibilities that doctors and hospitals must assume when death is unavoidable.

"SUPPORT did not produce results that show us exactly how to improve care for the dying," Dr. Lynn said. "But physicians need to learn to discuss how to live well while dying; focus on care for the family, not just the disease; and take responsibility for the future of the patient, not just the minute-by-minute medical crises. Patients and families should collaborate with physicians and create a plan for dying.

"The public needs to insist on a higher standard of care while dying," Dr. Lynn said. "Talk to almost anyone you know and they will have a horror story of a grandparent or aunt or neighbor who died suffering needlessly. This study is the scientific proof that validates such stories. We can do much, much better. We must insist on it."

Source: The Robert Wood Johnson Foundation, College Road, Post Office Box 2316, Princeton, New Jersey 08543-2316 USA, Telephone: 609-452-8701

Similar of Too Many Americans Die Alone, in Pain, Attached to Machines