For patients with advanced cancer who are nearing the end of life, the likelihood that they will spend their final days in a costly hospital intensive-care unit or at home depends largely on where they live and the hospital at which they seek care.
Where you live = how you die
'Geography is destiny' for end-of-life care, which often is costly, aggressive
In other words, geography is destiny, according to David Goodman, professor of pediatrics and health policy at Dartmouth Medical School, Hanover, N.H., and director of the Center for Health Policy Research at the Dartmouth Institute for Health Policy and Clinical Practice.
Goodman also is the co-principal investigator of the Dartmouth Atlas of Health Care, a long-running project that uses Medicare data to examine variations in the way healthcare is provided across the country.
And in its first report on the quality of end-of-life cancer care for Medicare beneficiaries, released Nov. 16, Dartmouth Atlas researchers found little evidence that treatments are aligned with patients' wishes.
The majority of cancer patients would much prefer to receive end-of-life care that allows for the highest quality of life and that enables them to live comfortably, without pain, supported by palliative or hospice care, said Goodman, the report's lead author. “That does not mean all patients want that, but on average, as many as 80% of patients at the end of life strongly want to be in a more homelike setting,” he said.
But far from reflecting those preferences, the report—which examined data from 235,821 Medicare patients with advanced cancer who died between 2003 and 2007—found significant variations in end-of-life care from region to region and from hospital to hospital. Roughly 29% of advanced-stage cancer patients across the country died in a hospital during the three-year period, and that number reached as high as 46.7% in the borough of Manhattan in New York City and 41.1% in Newark, N.J.
That contrasts sharply with Cincinnati, Ohio, and Fort Lauderdale, Fla., which had rates of 17.8% and 19.6%, respectively. And in Mason City, Iowa, the rate of in-hospital cancer deaths was only 7%.
The disparities persisted across other measures as well, including intensive-care admissions and chemotherapy administered during the last two weeks of life. About 70% of cancer patients in Detroit were hospitalized at least once during their last month of life, compared with 46.3% in San Angelo, Texas, according to the report. And cancer patients were far less likely to receive life-sustaining treatments such as endotracheal intubation or CPR in the last month of life if they lived in Minneapolis or Seattle than if they lived in Manhattan or Los Angeles.
Why is there so much variation, especially when most cancer patients say they would prefer to die at home? There are several answers, Goodman said during a media briefing. First, and perhaps most important, physicians often assume they know patients' wishes for end-of-life care and those critically important, honest conversations about alternate paths such as hospice and palliative care just don't happen. Also, cultural norms often push physicians to press ahead no matter what, he said.
“We are very reluctant to share information that a cure is unlikely,” Goodman said.
When those conversations do occur, and patients are given accurate information about their disease progression and the prospects of a cure or remission, they make better decisions, said Elliott Fisher, director of the Center for Population Health at the Dartmouth Institute for Health Policy and Clinical Practice, co-principal investigator of the Dartmouth Atlas and one of the authors of the report.
Recent research has pointed to the ability of palliative care not only to enhance quality of life, but also, in some cases, to prolong it. In an August 2010 study published in the New England Journal of Medicine, researchers determined that patients with metastatic non-small-cell lung cancer who received palliative care actually lived longer than patients who received standard curative treatments.
“It's very important to understand that the treatments we are providing for cancer patients are often harmful,” Fisher said. “They're being performed on the chance that they will provide benefits, and they can lengthen life in some and shorten it in others. Therefore it is critical that patients make choices with full understanding of their condition and what it will be like to experience that treatment.”
The communities that have been most successful in providing patients with information about their health status and offering alternate care options have several things in common, Fisher said. They usually have community-level consensus about the importance of end-of-life preferences, they train physicians to have those conversations with patients, and they have systems in place supporting palliative and hospice care.
In La Crosse, Wis., for instance, a region often cited for high-quality end-of-life care, 90% of all adults have a written advanced directive detailing their wishes, said Bernard Hammes, director of medical humanities at La Crosse-based Gundersen Lutheran, an integrated health system that includes a 252-bed hospital and more than 60 outpatient clinics.
And those instructions and patient goals, which are placed in the electronic medical record, are nearly always consistent with the care provided in the region, Hammes said.
Hammes acknowledged that one of the main reasons La Crosse has been able to provide such good end-of-life treatment is because almost all of the care in the region is provided by two integrated health systems. “We're connected in ways that systems in many other regions are not, and I know that has made it much easier for us,” he added.
Still, Hammes expressed hope that new models of care delivery, such as patient-centered medical homes and accountable care organizations, will provide a more coordinated approach that facilitates better end-of-life care. “Right now, palliative care is not valued in how we pay for healthcare,” he said. “But as physicians are increasingly paid for quality and not just for procedures, we can start to define those end-of-life outcomes that really add value to patient care.”
Palliative care is associated with lower healthcare costs. A 2009 study published in the Archives of Internal Medicine found that among patients with advanced cancer, end-of-life conversations with physicians resulted in significantly lower costs for treatment. In addition, researchers found that higher-cost care was associated with worse quality of death.
And in a 2008 study in the same journal, researchers found an adjusted net savings of nearly $5,000 in direct costs per admission for palliative care patients who died in the hospital, when compared with patients who received the usual course of care.
But despite cost savings, reimbursement remains a major obstacle, said R. Sean Morrison, president of the American Academy of Hospice and Palliative Medicine and a professor of geriatrics and palliative medicine at Mount Sinai School of Medicine in New York. In-hospital palliative care is reimbursed through Medicare Part B physician billing, so the hospital's palliative care team does not receive payment for end-of-life services, Morrison said.
The Medicare hospice benefit covers the physician and team's time, as well as home health and medications, but to select that benefit, patients must give up curative treatments and two physicians must certify that the patient has less than six months to live.
“It's limiting to so many patients,” Morrison said, adding that when it comes to in-hospital palliative care, “the business model has been one of efficiency rather than reimbursement. There's not a lot of revenue, but it's a quality issue and something we do because it's the right thing to do.”
The CMS' Center for Medicare and Medicaid Innovation, which officially launched Nov. 16, does have the potential to affect some pretty big changes in the payment system, said Helen Darling, president and chief executive officer of the Washington-based National Business Group on Health, and co-chair of the National Quality Forum's Workgroup on Palliative Care and End-of-Life Care.
The center was established by the Patient Protection and Affordable Care Act of 2010 to develop and evaluate new healthcare delivery and payment systems. With a 10-year budget of $10 billion, the center is charged with testing the viability of medical homes, telehealth services, patient self-management and many other models of care. “It's exciting that there will be an emphasis on finding ways to solve problems instead of paying for a dysfunctional system,” Darling said.
Yet another issue facing many physicians treating patients at the end of life is one of language, said Paul Glare, chief of pain and palliative care service at 434-bed Memorial Sloan-Kettering Cancer Center, New York, a hospital with a successful palliative care program and low rates of ICU cancer deaths. Physicians feel tremendous pressure to avoid saying that “hope is lost” or that they are “giving up,” and that's a mistake, he said.
“This word hope, a lot of people see it simplistically, like a military approach to living longer and treating illness harder,” Glare explained. “But I think we have to realize when it is time to change direction. Having hope also means having options and other pathways. It doesn't mean dying in the hospital ICU and having chemotherapy in the final weeks of life. People want to have choices.”
One effective way to address that issue is to integrate palliative care with more aggressive, life-prolonging treatments from the point of diagnosis, said Morrison of the American Academy of Hospice and Palliative Medicine.
In a 2001 report on improving palliative care for cancer patients, the Institute of Medicine advocated a coordinated approach that integrates palliative care and symptom control with curative treatments across the continuum of care, calling it the “ideal mix.”
Some level of that kind of integration is the norm for the majority of the roughly 1,500 U.S. hospitals that have palliative care departments, Morrison said, and he pointed to recent studies that have shown improved pain management and lower rates of depression when palliative care is offered early on.
“If we were to fully integrate that kind of care into the healthcare system, we'd go a long way toward reducing the variation you see in the report,” Morrison said.
At 241-bed Mercy Medical Center-North Iowa in Mason City, palliative care staff members credit their region's success to a variety of factors including supportive administration, a long history of focusing on end-of-life care, and a holistic approach to treating patients. And, according to W. David Clark, a palliative medicine physician at the hospital, people in the region are trusting and appreciate the plain, straight talk that comes with palliative care.
Mercy Medical Center-North Iowa's palliative consulting service and fellowship began officially in 2006, and staff members see patients in the hospital, nursing homes and in other settings, said Debra Groath, a nurse practitioner in palliative medicine. “When an oncologist says, ‘We have nothing left to offer you,' we tell the patients, ‘Yes, we do. We have comfort,' ” Groath said.
Once patients with life-limiting illnesses are identified as candidates for palliative care—usually by a hospitalist or an oncologist—a 60- to 90-minute goal-setting meeting takes place. The palliative care program uses a multidisciplinary team of nurses, physicians, chaplains, social workers, pharmacists and dietitians from other departments, and they make collaborative decisions based on everyone's feedback, said Anne Zook, Mercy Medical Center-North Iowa's palliative medicine nurse coordinator.
“We take the time to really listen and match the treatment to patients' goals,” Zook said.
For now, questions still persist about the reasons for the variations seen in the report, about what the correct level of palliative treatment is, and about whether care really matches what patients want, said Morrison of the American Academy of Hospice and Palliative Medicine.
“We don't know what the definition of ‘right” is yet,' he said. “Dartmouth has identified the problem, and it's our job to understand it and to fix it. It's a right of patients and their families to have treatments that match their goals, and that's what modern palliative care is trying to do.”
Send us a letter
Have an opinion about this story? Click here to submit a Letter to the Editor, and we may publish it in print.