In a study that surprised its researchers, National Cancer Institute-designated hospitals as well as academic centers provided their dying oncology patients the same low level of palliative, comfort care or hospice services as community hospitals.
And overall, the amount of aggressive and futile care, such as chemotherapy in the last two weeks of life, was disturbingly higher than expected across all hospital types, and not according to established quality measures, with wide variation among hospitals within each type.
Those are some conclusions from the Dartmouth Atlas Project's, latest research in which investigators looked at cancer care provided to 215,000 Medicare beneficiaries with advanced metastatic disease at 4,400 facilities. The period examined was between 2003 to 2007.
Overall 30.2% of these terminal patients died in the hospital instead of at home or in a hospice facility, 65% were hospitalized at some point during their last month of life, and 24.7% were treated in the intensive care unit during the last month of life.
It would have been nice if the data had revealed some distinctions in care practice—for example, that a patient with a terminal diagnosis who wanted more aggressive care could seek a particular type of hospital, said Nancy Morden, MD, lead author of the paper published Monday in the journal, Health Affairs. But that was not what the study found.
"It's surprising, and a bit frustrating, and I think most importantly, sad for the patients, because they can not go to some public data repository and know what kind of care they're going to get at a particular hospital or a type of hospital," Morden says. "There's very little pattern to it, and we found that it was a hard to sort out."
Morden explains that researchers anticipated that NCI-designated cancer centers in general "would do a better job, would be more in line with quality, [and] they would be more in tune with patient preferences. But we didn't find that."
The facilities were divided into four types:
- 21 National Comprehensive Cancer Network (NCCN) hospitals,
- 22 non-NCCN NCI centers
- 161 academic medical centers
- 4,240 community hospitals
They were also grouped by size: fewer than 150 beds, 150-300 beds, and greater than 300 beds; and by for-profit versus not-for-profit status.
There were some differences in the level of aggressive care between the hospital types, although the range within each type was wide.
For example, while across all hospitals, 9.3% of cancer patients underwent a life-prolonging procedure (such as insertion of a feeding tube, use of a ventilator, or initiation of cardiopulmonary resuscitation) during the last month of life, 13.1% of patients at non-NCCN NCI cancer centers did so, 12.6% of patients at academic hospitals, 10.3% at NCCN centers, and 8.9% at community hospitals.
Chemotherapy at the end of life varied between types of hospitals as well, with 6.3% undergoing drug treatments at community hospitals, 6% at NCCN centers, 5.4% at non-NCCN NCI hospitals and 5.3% at academic hospitals.
Said another way, Morden explains, comparing the patients treated at NCCN hospitals, patients at community hospitals were 36% more likely to receive chemotherapy during the last 14 days of life.
Community vs. Academic Hospitals
"I don't know exactly what drives community hospitals to be much more likely to give patients end-of-life chemotherapy at literally, within two weeks of death. But the reimbursement structures are different at academic centers versus community hospitals, and at academic centers, the physicians tend to be salaried so these incentives may be blunted."
While overall, 24.7% of all dying cancer patients spent time in the intensive care unit during the last month of life, there was not that much difference in ICU utilization between the hospital groups. For example, 26% of academic hospitals, 26.3% of non-NCCN NCI centers, 23.3% of NCCN cancer centers and 24.6% of community hospitals put dying patients in the ICU.
"The ICU is a rescue center, and I don't think you should send anyone there unless you have some hope that you can rescue them, and you can not rescue people with metastatic cancer" at this late stage.
"In an ideal setting, people would get the care they want, and we would figure out a process to assess preferences and deliver preference-sensitive care in all settings," she says. "It wouldn't be the patient's burden to figure out where they're going to get what they need."
The report follows the Dartmouth Atlas Project's voluminous report from 2010, which first pointed to wide regional disparity among the nation's hospitals in aggressiveness of care during a terminal cancer patient's last few weeks of life.
Medicare Fee-for-Service Model to Blame?
Morden noted that the Medicare fee-for-service payment system may be behind much of the trend for more aggressive end-of-life care. "I don't necessarily think that on average, at an institution that has tendencies toward more aggressive intense care, that there are individual physicians running around with this in mind, specifically. But the incentives, consciously or subconsciously, all push people to do more because we get paid for what we do."
Oncologists, she says, make "the overwhelming majority of their income from chemotherapy as opposed to sitting with patients and consulting, giving advice about treatment decisions."
The Dartmouth report also found that at for-profit hospitals, 6.6% of patients received chemotherapy in the last 14 days of life but at not-for profit hospitals the rate was 6%. Hospitals with 150 beds or more provided 6.5% of dying cancer patients with chemotherapy, while at hospitals with fewer than 150 beds, 5.4% of patients received drug regimens.
Morden says that the Dartmouth group hopes to update its numbers to reflect more recent Medicare data, to see if more physicians and hospital-teams are getting better at following quality measures for dying patients.
Morden adds that it's up to policymakers to bring care practices into alignment with a patient's preferences in an honest way, because giving chemotherapy to someone who is going to be dying quickly is giving them false hope.
"I don't think we're doing them any service. There's a fair amount of evidence to say that it's painful, and toxic. We are dramatically decreasing quality and we may even be shortening life."