The most award winning
healthcare information source.
TRUSTED FOR FOUR DECADES.
By Julie Crawshaw
A recent study shows that patients who discuss their care with their physicians from the inception of their illnesses greatly improve their odds of getting the kind of medical care they want. Conducted by Joan M. Teno, MD, MS, professor of medicine at Brown University, the study evaluated the decision-making and outcomes of 1494 seriously ill patients who stayed in one of five ICUs for at least 14 days (median stay 35 days) and were enrolled in a larger treatment outcome study. Teno and her fellow researchers interviewed patients, surrogate decision-makers, and physicians about prognosis, communication, and goals of medical care.1
"One of the findings from this study is that the doctors are not talking to the patients and patients’ families are not talking to the doctors," Teno says. "There’s a high rate of people getting care they believe is inconsistent with their treatment preferences."
The researchers used their findings to develop a new conceptual model they called Patient-Focused, Family Centered Medical Care. One of the cornerstones of that model is promoting shared medical decision-making. "There is a series of outcome measures we’ve developed around involving patients in treatment decisions," Teno says.
Elliott Fisher, MD, MPH, another of the study’s researchers, says he believes the reason patients and physicians aren’t talking more is that the medical model in current use is futility-based. "We start raising the questions when everybody recognizes that there’s absolutely no hope," Fisher says. "What we really need to do is start talking about this throughout the patient’s illness, not just when everyone sees that the patient is dying."
When patients run out of treatment options—usually when the toxicity of the treatment outweighs its benefits—they reach a point of futility. Both Teno and Fisher perceive a need to move from a futility-based model of decision-making to one that has the patient taking an active role in treatment decisions from diagnosis forward.
"There are always treatment options," Teno says. "Hypertension is an excellent example. There’s a veritable slew of anti-hypertensive medications with very different costs, but the majority of patients don’t know this because they’re not involved in medication decisions."
The study found that many seriously ill patients see a mismatch between their treatment preferences and the degree to which they believe those preferences are being honored. Fisher points out that correcting this doesn’t necessarily mean that physicians must immediately begin discussing end-of-life measures when a patient is first diagnosed with a potentially fatal illness. "There are ways of having the conversation without really posing that. It’s really about understanding the person’s values and preferences. Sit and listen and understand where people are in dealing with the illness," Fisher says.
Teno concurs, adding that one of the things the medical community does not do well enough is listen to the patient’s concerns. "If we began by listening and then double check by repeating back to the patient what we think we’ve heard, we’d be a lot better off," Teno says.
Following completion of their study, the Brown researchers designed a toolkit they are making available free of charge.2 The toolkit was designed, Teno says, because end-of-life care is different from critical care in which the patient survives. In order to determine what medical decisions are best for patients to live well with a life-defining illness, physicians must:
• Re-engineer advance directives
• Change the culture through education
• Develop measures of and demand quality of care
• Create systems of care that deliver quality medical care
• Develop measures and ways of measuring that work
The overall strategy for the toolkit involves conducting focus groups with dying persons and their loved ones. The information gained is then used to review guidelines for key processes of care. This is followed by reviewing the evidence that indicates that these processes will actually result in quality medical care.
The toolkit emphasizes the importance of family members’ perspective in assessing quality of care.
It uses retrospective family interviews and prospective patient interviews in a core module applicable across settings of care. Tests were run with a population of bereaved family members whose loved ones died in a hospital, nursing home, or while under hospice care. Researchers say the instrument is equally reliable and valid in all settings. Interviewers administered the survey either over the telephone or in person.
Different versions of the toolkit are available. The hospital version of the toolkit is based on a longer instrument, and has been tailored to reflect hospital services. That allows hospital staff to obtain that perspective using a measurement tool geared specifically toward hospital care. To maintain consistency across versions, researchers have retained the numbering from the original instrument—which means that the numbering for the hospital version appears out of sequence.
Additional modules are available that allow users to modify the survey to their own particular needs. Reports resulting from toolkit use come with a resource guide that suggests the next steps to improving the quality of care. If users choose to limit the domains of interest, researchers suggest that a useful survey would include the following four domains:
1. physical comfort and emotional support
2. promote shared decision making,
3. focus on individual, and
4. attend to the emotional and spiritual needs of the family.
Using the instrument results in information on seven different aspects (or domains) of quality of care. Domains are color-coded so that the questions pertaining to a specific domain all share the same color.
Users can choose to focus on one or more specific domains. However, researchers stress the importance of including all of the questions within each domain of interest for valid, reliable results. To maintain the validity and reliability of the instrument, the questions must be asked in the order that they appear.
The survey can be used to assess the quality of care received by an individual patient, though it was designed to be used with groups of people. Another study now underway will produce norms for the United States, which should be available this spring.
One of the most dramatic toolkit illustrations shows a photo of a smiling, apparently happy 94-year-old female patient with acute myocardial infarction, low blood pressure, and who is short of breath at rest. A capition asks: "Is she terminally ill’? Or is it time to terminate the term terminally ill?"
Researchers for another study, conducted by the U.S. Department of Veterans Affairs, concluded that end-of-life conversations should automatically be included in care plans so that critically ill patients can die according to their own values and wishes.3 The study proposes creating a new health care professional working under physicians whose sole job would be to facilitate and document end-of-life conversations. The VA Health Care Network Upstate New York is using this new model.
Other measures VA researchers suggested include multiple programs to train physicians for end-of-life dialogues and to raise community awareness to facilitate such conversations between health professionals, patients, family members, and caregivers.
Daniel R. Tobin, MD, of the VA Health Care Network Upstate New York, Albany Division, was lead author for the study. Tobin and co-author Dale G. Larson, PhD, professor of counseling psychology at California’s Santa Clara University, found that training all health care practitioners in standard communication models for conversations at the end-of-life is essential to providing quality end-of-life care. He and his fellow researchers observed that patient preferences for life-sustaining treatments are often inadequately discussed and documented, and that hospice or home care referrals frequently come too late or not at all. Despite the ability to create a pain free death, many patients still die in unrelieved pain after long hospital stays and intensive care.
Tobin and Larson say they expect the quantity and quality of end-of-life discussions in the future to be closely linked to improving physicians’ communication skills, adopting a patient-centered model of care, focusing on improving the quality of remaining life, and developing clinical models and programs to support such discussions earlier in the health care process.
The VA study found both personal and institutional barriers that presently make meaningful end-of-life conversations difficult in health care settings. Patients with an advanced illness may avoid end-of-life conversations because they can’t deal effectively with their emotional pain and fear. Physicians may avoid end-of-life conversations because they are afraid those conversations may cause pain for patients and their families. Also, many physicians are unfamiliar with advance directive laws and fear that talking about whatever medical alternatives might be available could lead to disagreements with the patient or family members.
Further, the study reports that because patients may receive treatment at a variety of health care delivery sites, responsibility for end-of-life discussions may not be clear. Health workers may assume the end-of-life issue has been handled elsewhere and therefore not ask about it. The fact that there is currently no financial compensation for physicians for psychosocial conversations, including those that concern end-of-life, may deter some physicians from initiating an end-of-life dialogue.
Tobin, who advocates randomized trials to test new approaches and models for enhancing end-of-life conversations, is nonetheless excited about the program at his own facility.
"We need to train all health care practitioners in standard communication models for conversations at the end-of-life," Tobin says. "Here, we’ve also introduced the new role of an advanced illness care coordinator who works in collaboration with the patient, family, and physician to secure dignity and control in the last years of life."
1. Teno J, et al. J Am Geriatrics Soc 2000;48:S70-S74.
2. To obtain the toolkit, contact Joan_Teno@Brown.edu or her assistant Jeff Edmonds at (401) 863-9630.
3. Tobin D, Larson D. JAMA 2000;284:1573-1578.