Study examines trends for Medicare patients at EOL

Findings indicate more time is spent in the ICU

A new study from the Dartmouth Atlas Project seems to indicate the "report card" for Medicare patients at the end of their lives is a mixed bag of pluses and minuses. On the positive side, the study, "Trends and Variations in End-of-Life Care for Medicare Beneficiaries with Severe Chronic Illness," showed that Medicare beneficiaries with severe chronic illness spent fewer days in the hospital at the end of life in 2007 than they did in 2003, and that they were less likely to die in a hospital and more likely to receive hospice care. On the other hand, they were more likely to be treated by 10 or more doctors in the last six months of life in 2007 (36.1%) than they were in 2003 (30.8%), and the average number of intensive care days increased to 3.8 from 3.5.

"The fact that these patients are spending less time in the hospital is connected with the fact that they are spending more time in hospice," notes lead author David Goodman, MD, MS, the co-principal investigator of Dartmouth Atlas of Health Care, professor of pediatrics and of health policy, and director, Center for Health Policy Research, at the Dartmouth Institute for Health Policy and Clinical Practice. "But the fact remains that these are the two domains where patients receive higher-intensity care, and there were more ICU days. And it's not just the ones left in the hospital who are sicker patients; there are more numbers of ICU days across the entire population of those who died, so there is a real increase of ICU care in this population. Patients certainly spend less time on general hospital wards, but they spend more time on the ICU."

Goodman says the medical profession is uncertain as to why this is, and adds, "This is not true for every hospital. It's fascinating that there are some hospitals where their change is in parallel with this study, while others have defied this general trend."

In other words, he continues, it is not the "destiny" of any specific facility to have these patients spend more time in the ICU. "I think one of the important factors that tend to shape the local experience is how local healthcare systems invest — and what they invest in," says Goodman. "Places that make relatively greater investment into ICU units can be providing valuable care for certain patients, but they can have unintended consequences as well."

Patients prefer less intensive care

One of those "unintended consequences," says Goodman, is lower patient satisfaction at a difficult time in their lives. "For this report, one of the major findings is that patients near the end of life on average strongly prefer to spend as much time as they can in a home-like environment," he says. "Many spend time in the ICU when it is not their preference — but that preference is not often elicited or legitimized by healthcare systems that have tremendous resources available for curative care and then assume the patient wants that applied to them, even in situations where the likelihood of their returning to the life they once knew is nil."

In earlier research, notes Goodman, "We studied the relationship of intensity of care and the HCAPS rating, and patients' perception of hospital experience and quality of care, and it really showed a negative correlation — patients were most unhappy in places where care was of the highest intensity. Also, many research studies look at what happens in patients at the end of life in terms of decision quality — whether their preferences are followed by caregivers, even when those preferences are clearly articulated." What did he glean from that research? "We have a long way to go," he says.

Intensity of care has a lot of physical consequences as well, says Goodman. "Sometimes a patient will be in the ICU and we will not only save their life, but create the opportunity for that patient to acquire an HAI, or the clinician may give the patient more cyto-toxic chemotherapy, which can lead to pneumonia. These are fragile patients, and sometimes intensity can tip them over the edge."

Trends must be reversed

Goodman says that for things to improve in hospitals, health systems and providers must "unlearn" certain assumptions. "In places that have grown their population of physicians and sub-specialists, that is the capacity that gets used," he explains.

"I am a trained physician," he continues. "The classic way we think of ourselves is that our job is to gather as much information as we can about the patient and their condition; we have knowledge of treatments, we learn about the patient's condition, and we make a recommendation."

But that common role of making recommendations does not work today, says Goodman, and it won't in the future — particularly for very sick patients. "It assumes we understand all of the care options, and often we don't," he asserts. "Oncologists, for example, are very much focused on curative care, but they won't be experts on palliative care. They may not understand what the patients' values are; studies have shown that doctors often use their own value sets. That approach is well intentioned, but it misses the mark."

What's more, Goodman continues, there is no correlation between intensity of care and measures of technical quality. "When you spend many more days in a hospital ICU, you see many different doctors, but the quality of care tends to be lower," he observes. "We think that's because care becomes disordered. There are more handoffs, and more chances for missed communication."

In addition, he says, electronic medical records do not solve that problem. "EMRs rarely extend to full care, especially in chronic care facilities," Goodman asserts.

Reduce spending, improve quality

Another implication of the study's findings is that "providers can look for insights into potential savings they can achieve through improved care of chronic illness that allows patients to remain safely out of the hospital," asserted Risa Lavizzo-Mourey, MD, MBA, president and CEO of the Robert Wood Johnson Foundation, a long-time funder of the Dartmouth Atlas Project, in a statement released to accompany the study's publication.

Goodman agrees. "We shouldn't be surprised we're spending more money on healthcare than any other developed country in the world, but there might be opportunities to do a better job with less money, particularly when some patients get expensive care they do not want," he says. "There are opportunities for accomplishing greater efficiencies thoughtfully. I'm not talking surgery with dull tools, but crafting our models of care and reimbursements so we can deliver higher quality for less cost."

End-of-life care, he continues, is one place where we know if patients get palliative care services early in the care of chronic illness, they have a much better experience. "They generally have a lower intensity of care, which saves money, and at least in cancer care, there are studies that show they actually live longer," says Goodman.

[For additional information, contact: David Goodman, MD, MS, Co-Principal Investigator, Dartmouth Atlas of Health Care, Dartmouth Medical School, HB 7251, Hanover NH 03755. Phone: (603) 653-0815. Fax: (603) 653-0822. E-mail:]