Image: utah788/Getty Images
Bringing Precision to Medicare Cancer Care
New model for classifying cancer patients could lead to better end-of-life care
By Greta Friar
Medicare policies governing cancer patients’ end-of-life care are based on generalized statistics—such as average survival time and treatment costs—that often fail to reflect the variety of experiences across patient subpopulations, as well as among individual patients, according to a new study led by Harvard Medical School researchers, published in the July issue of the journal Health Affairs.
The study results reveal that the dominant end-of-life care settings for patients with lung cancer—home, inpatient facility, hospice or intensive care–were linked to different survival times, expenses, number of hospitalizations and duration of palliative care.
Failure to account for such critical nuances across patient populations could lead to ill-informed policies that are misaligned with patients’ actual prognoses and needs, the research team says.
“Medicare policies for cancer care ought to be designed with diversity in mind. There is no average patient,” said study senior investigator Laura Hatfield, an associate professor of health care policy at Harvard Medical School.
To rectify the problem, the researchers developed a model that accounts for variation in the clinical course of different patient populations across various clinical settings.
To do so, the investigators analyzed Medicare claims from 1995-2009 for more than 14,000 patients diagnosed with extensive stage small-cell lung cancer, a usually terminal condition with a short survival time. Researchers compared the time patients spent in each care setting from diagnosis to death relative to patients’ overall survival time to identify patterns in care.
The researchers say they hope that results from models like theirs, which classify patients with the same diagnosis into smaller groups with similar characteristics, can be used to inform tailored health care coverage options and, ultimately, better individualized care.
For example, although the average survival time for patients with small-cell lung cancer is eight months to a year, many patients have much shorter life expectancies, the analysis showed. Patients who spent most of their time in inpatient and intensive care unit settings had an average survival time of one month, while the average survival time of the hospice group was around four months. Only patients who spent the most time at home had an average survival time of 10 months. Because the at-home group was the largest, two-thirds of the patients, when all the numbers were averaged, the group’s longer survival time masked the shorter survival of the others.
The authors caution that their findings do not indicate that health care settings necessarily played a role in different survival times. Rather, they said, patients with the worst prognoses likely ended up needing acute care because their condition was rapidly changing and severe enough to require hospitalization. The purpose of classifying patients this way was not to inform clinicians’ recommendations for individual patient care, but to give policymakers better information about variation in patients’ outcomes and health care needs.
“If a patient has only one month to live, then policies should ensure that their care includes more rapid decision making and advanced care planning,” Hatfield said.
Surprisingly, the researchers observed, although the averages suggest that patients with small-cell lung cancer tend to spend time in acute care, and their final days in hospice, many patients spent time in only one of those settings. In fact, two-thirds of the inpatient care group never received hospice care, with the majority dying either at a hospital or in a nursing facility. Meanwhile, the data showed that the hospice group spent relatively little time in the hospital and only a quarter of patients in this group received chemotherapy after their diagnosis.
In recent years, there has been a push to get more patients in hospice care due to concerns that patients are receiving aggressive, burdensome treatments. Past studies have shown that overly aggressive care leads to poorer quality of life, lower satisfaction among family members and more expenses for both patients and the health care system. In light of this, the researchers say, it might make sense for patients with small-cell lung cancer who are on acute-care trajectories to receive earlier and more comprehensive palliative counseling that ensures they are informed about the option of hospice and about the limitations of curative treatments.
Yet, hospice care may not always be a realistic goal for patients in the ICU group, Hatfield cautioned. These patients may have required aggressive care to manage their symptoms, or they may have opted for aggressive hospital treatment over palliative care. Medicare does not allow patients to receive hospice care until after they have given up potentially life-extending treatments such as chemotherapy.
Patients with small-cell lung cancer whose tumors respond well to chemotherapy face an even more agonizing dilemma, as many hospice providers do not offer chemotherapy even for symptom relief. The researchers suggest that one way around this could be hospital-based expanded concurrent palliative options, rather than trying to get patients into hospice care.
Data-driven models that account for differences in care, treatment and outcomes could serve as compelling evidence to insurers to rethink their policies, the research team says.
“We are applying sophisticated methods to tell the stories of cancer patients,” Hatfield says. “We’re using data to serve a very human end.”
This work was supported by the National Cancer Institute (grant R21-AG047175), the Marshall J. Seidman Fellowship in Health Policy and the National Institute of Mental Health (grant T32MH019733).
source: Harvard University – Harvard Medical School