This article is more than 5 years old. Information may no longer be current.
Earlier hospice, avoidance of ICU improve family members’ perceptions of end-of-life care
Click Here to Manage Email Alerts
Factors including early hospice enrollment, avoidance of ICU admission within 30 days of death and death occurring outside the hospital appeared associated with favorable perceptions of end-of-life care among family members of older patients who died of cancer, according to the results of an interview-based study published in JAMA.
“Patients with advanced-stage cancer receive aggressive medical care at the end of life, despite increasing evidence that high-intensity treatments may not be associated with better patient quality of life, outcomes, or caregiver bereavement,” Alexi A. Wright, MD, MPH, assistant professor of medicine at Harvard Medical School and staff physician at Dana-Farber Cancer Institute, and colleagues wrote. “Few studies have examined whether these aggressive end-of-life care measures reflect patients’ preferences or bereaved family members’ perceptions and expectations of the quality of end-of-life care.”
Alexi A. Wright, MD, MPH
Wright and colleagues sought to observe whether aggressive end-of-life care aligned with patient goal attainment and influenced perceptions of quality among bereaved family members.
Thus, they interviewed 1,146 family members of 1,146 Medicare beneficiaries (median age, 76 years; interquartile range [IQR], 65-87) with advanced-stage lung or colorectal cancer who had been enrolled in a multiregional, prospective observational study and who died by the end of 2011. The interviews occurred a median 144.5 days after death (IQR, 85-551).
Bereaved family members included spouses or partners (n = 498), children (n = 474), other relatives (n = 113) and friends (n = 61). Fifty-eight percent reported always accompanying patients to physician appointments and 60% reported always discussing important medical decisions regarding cancer. Family members reported a median of 7 days (IQR, 6-7) of contact with patients during their last week of life.
The researchers evaluated claims-based quality measures of aggressive end-of-life care, including ICU admission in the last month of life, chemotherapy receipt within 2 weeks of death, utilization of hospice (no hospice vs. ≤ 3 days of hospice services) and death during hospitalization.
A family member-reported quality rating of “excellent” for end-of-life care served as the primary outcome. Patient goal attainment — defined as end-of-life care that corresponded with patients’ wishes, such as death occurring in a preferred location — served as a secondary endpoint.
More than half of the bereaved family members interviewed reported excellent end-of-life care (51.3%; n = 588).
Family members appeared more likely to report excellent end-of-life care when the patient received more than 3 days of hospice care (58.8% vs. 43.1%; adjusted difference, 16.5%; 95% CI, 10.7-22.4).
In contrast, fewer family remembers reported excellent end-of-life care when the patient was admitted to the ICU within 30 days of death (45% vs. 52.3%; adjusted difference, –9.4%; 95% CI, –18.2 to –0.6).
Family members of patients who died while hospitalized also appeared less likely to define patient’s end-of-life care as excellent (42.2% vs. 57.4%; adjusted difference, –17%; 95% CI, –22.9 to –11.1).
Patients who did not receive hospice care, or who received hospice care for 3 or fewer days, appeared less likely to have died in their preferred location than those who received hospice care for more than 3 days (40% vs. 72.8%; adjusted difference, –34.4%; 95% CI, –41.7 to –27).
Further, a greater proportion of family members of patients who received chemotherapy within 2 weeks of death reported the patient did not die in their preferred location than family members of patients who did not receive chemotherapy (42.1% vs. 57.5%; –17.7; 95% CI, –34 to –1.4).
The researchers acknowledged limitations of their study. Because the study population included older patients who primarily died prior to 2008, the results might not be generalizable to younger patients or patients who have died more recently.
Further, because Medicare served as the primary insurance provider for these patients, the results might not be generalizable to privately insured or uninsured patients.
These data emphasize the importance of generating care plans prior to end of life, Atul Gawande, MD, MPH, professor of health policy and management at Harvard T.H. Chan School of Public Health, professor of surgery at Harvard Medical School, and general and endocrine surgeon at Brigham and Women’s Hospital, wrote in an accompanying editorial.
“There is no intrinsic reason that the quality of care for the dying should be poorer in hospitals than elsewhere,” Gawande wrote. “The common missing element, whether patients die in the hospital or at home in hospice, likely remains a lack of early discussion and planning of care around their goals and priorities for the quality of the life they are leading as they experience serious illness.”
Further, medical education should be focused more clearly on end-of-life care across disciplines, not just among those who plan to specialize in palliative care, Gawande wrote.
“Such support is not a responsibility to be outsourced to a specialized few,” Gawande wrote. “Palliative care clinicians, geriatricians and others have been doing essential work to develop, articulate and deploy the skills required to serve the well-being of the sick and the frail. … There are reasonably well-established best practices for communicating prognosis, planning goals of care and managing symptoms. However, training in using them has been absent from medical education in most specialties.” – by Cameron Kelsall
Disclosure: Wright reports no relevant financial disclosures. Please see the full study for a list of all other researchers’ relevant financial disclosures. Gawande reports royalties and compensation from numerous publishers and media outlets for writing and medicine-related media.
Perspective
Back to Top
Supriya Gupta Mohile
The researchers interviewed caregivers shortly after the deaths of their loved ones, asking about the patients’ treatment preferences, late-treatment symptoms, end-of-life wishes and quality of care. Results illustrated that family members were more likely to report receiving excellent care when patients received timely hospice services and were less likely to report receiving excellent care when patients were admitted to an ICU shortly before death. Further, family members of patients who died while hospitalized — rather than at home or in their preferred end-of-life setting — reported displeasure more frequently. The most robust correlates of excellent quality were receipt of timely hospice services and death outside the hospital.
It is important that physicians communicate the natural history of cancer appropriately to patients and caregivers. For the most part, advanced lung and colorectal cancers are incurable and life-limiting. Studies have shown that many patients are not aware of the terminal nature of their diseases. It is important for physicians to discuss this openly and honestly with patients, so that they have the opportunity to discuss preferences for end-of-life care with their families and loved ones.
Often physicians are aware that a patient’s time is limited if he or she is not responding well to treatment, but instead of communicating this, they tend to recommend a treatment interval, with the option to resume treatment later. This approach prevents patients from accepting what might be inevitable and keeps them hoping for treatment in the future, oblivious to its futility.
In this study, 6.5% of patients received chemotherapy in their last 2 weeks of life, but twice as many (13.1%) spent time in an ICU in their last month of life. More than half died in the hospital without timely hospice care. Open and early communication would allow for earlier transitions to nonaggressive care options.
It would be interesting to know how many of these patients and caregivers had the opportunity to interact with multidisciplinary palliative or geriatric oncology specialists prior to death. These care teams are often very effective in helping patients through difficult conversations as they transition from cure to care.
The sample used in this study were older and had many comorbidities, so uncertainty exists regarding the risks and benefits of treatment in general. Incorporating patient preferences at every step of the decision-making process could help aid in planning care decisions that would lead to important quality measures — such as death in a preferred setting — and lead survivors to a feeling that their loved one received the best care possible.
Of course, some patients and families struggle with accepting terminal illness, making it hard to engage in difficult conversations about prognosis and preferences. This reluctance may be the cause of the increased ICU admission and lack of hospice care. It is difficult to know if such preferences are truly informed when oncologists, for whatever reason, often do not engage their patients in open communication on difficult topics. We must surmise that these situations can lead to survivors reporting that they feel less satisfied with the care their loved ones received because their loved one died.
This investigation adds significantly to the literature and strengthens the argument for open communication and engagement with palliative experts earlier in the process of care for advanced cancer. It further provides insights into correlates of satisfaction ratings, which will become increasingly important with mandated public reporting of quality metrics. Future research should explore patient and family personality characteristics, backgrounds and dynamics in the acceptance of cancer as a terminal condition. Researchers should also explore care delivery models that examine access to palliative care, geriatric oncology, psychology services and social workers who can advocate for patients and families.
-Supriya Gupta Mohile, MD, MS
University of Rochester Medical Center
-Paul Duberstein, PhD
University of Rochester Medical Center
Disclosure: Mohile and Duberstein report no relevant financial disclosures.
Click Here to Manage Email Alerts