How Americans die has fundamentally changed with advances in medical technology and the ways diseases are treated. For centuries, death commonly occurred in one’s home with care provided by relatives and community members. Yet since the 1960s, the hospital and intensive care unit have become places of passage as people approach the end.
In this week’s New England Journal of Medicine, we report that home has become the most common place of death among Americans dying of natural causes for the first time since the early 20th century, while deaths in hospitals and nursing facilities have declined. Our analysis of data from the Centers for Disease Control and Prevention and the National Center for Health Statistics also showed striking differences in place of death according to who you are and what you die of: individuals who are nonwhite or those dying from diseases other than cancer are less likely to die at home than those who are white or those who die from cancer.
These findings are encouraging. But they can also have a dark side, illustrated by Margaret Peterson, who helped her husband, Dwight, die at home in Illinois. “My experience was positive in the sense that my husband succeeded in dying in his own damn bed, in his own damn house,” she wrote in response to a query that one of us (H.J.W.) posted on a Facebook group for caregivers. “It was negative in the sense that the medical management we needed, such as subcutaneous or intravenous morphine, was not available. It was terrible.”
The emergence of hospice
Concerned by the inadequacy of care provided to the dying, Cicely Saunders founded the modern hospice movement in England in 1967. The growth of hospice in the United States was likewise fueled by growing disillusionment with the aggressive treatment that cancer patients received even when death was approaching.
Hospice is specialized interdisciplinary care for those with limited life expectancies. It focuses on managing symptoms and maintaining quality of life. It differs from palliative care, which may be administered at any stage of serious illness and in conjunction with conventional treatment.
In the United Kingdom and Europe, hospice is predominately an inpatient service; in the United States, home care is prioritized. End-of-life care in the U.S. was revolutionized by the Medicare hospice benefit, which became permanent in 1986.
Hospice has since been transformed from a social movement into an essential component of the health care system. The rise in home deaths documented in our study is likely a result of greater use of hospice along with broader efforts to de-medicalize and improve end-of life care.
Hospice in the U.S. is a uniquely American creation: an insurance benefit intended to make hospice a cost-neutral service. It was designed based on the needs of people with cancer, a disease that progresses differently from other common causes of death such as heart failure or dementia. It was also created with the expectation that family members would provide the majority of hands-on care.
Hospice does not provide 24-hour care in the home except in short crisis situations. Symptom flare-ups, like a spike in pain or difficulty breathing, may be challenging to manage at home and the day-to-day burdens on caregivers are often significant.
Inpatient hospice facilities, which are able to provide intensive symptom management, are at increased risk of closure due to insufficient funding and strict Medicare regulations.
Improving the system
We are advocates for hospice and are encouraged by the expansion and uptake of this service. But many individuals need the kind of care that the current hospice model cannot provide. Given the rising number of people dying at home, ensuring the availability of services to support them and their caregivers is a matter of urgency.
Only about half of Americans die while receiving hospice services, and home-based non-hospice palliative care is still in its infancy. We need a new movement, one that embraces the best of what hospice pioneers envisioned but that also adapts to the reality of modern health care and society.
Upgrade hospice. By agreeing to cease potentially “curative” care, patients who choose hospice are expected to have reduced expenditures from expensive treatments and hospitalizations. This model of care was unfair from the start, but it is now terribly outdated as the line between “curative” and symptom-easing treatments has become increasingly blurred with the development of new medical technologies. Hospice should be redesigned with quality of care as its priority, not the cost of care. Eligibility criteria for hospice should be based on a patient’s needs, not inaccurate estimations of prognosis or the treatments she or he is willing to forgo.
Change the one-sized fits all approach. Current policies present barriers to wider hospice use for people with non-malignant condition, who may benefit from disease-directed therapies late in their disease course and whose prognosis is less predictable. Racial minorities have lower rates of hospice use, likely due to greater mistrust of the health care system, cultural beliefs and preferences, and greater desire for life-sustaining care regardless of prognosis. The ability to receive disease-directed treatment should be incorporated into new models of hospice care and the viability of disease-specific hospice benefits should be explored.
As our population becomes increasingly diverse, the need for care that incorporates varying cultural backgrounds and preferences will also grow. Geographic disparities and workforce shortages have fueled innovative telehealth hospice programs. We will need other inspired programs to extend high-quality person-centered care to all.
Pay attention to social determinants of health. People with fewer socioeconomic resources are less likely to die at home — presumably because they lack the ability to afford and access services that make dying at home feasible. Paying for private caregivers is beyond the ability of most Americans and the current hospice benefit provides coverage only for limited social services. Many hospices, however, report using internal funds to assist patients with non-reimbursable needs such as food, shelter, and utilities. We must think outside the box and develop programs that meet social needs at the end of life and consider incorporating additional social supports into new care models.
Make quality end-of-life care available in all settings. When asked, most people say they would prefer to die at home, and hospice use increases the likelihood of this occurring. Yet a home death is neither preferable nor possible for everyone. It is good when preferences can be honored and people are able die in the place of their choosing. Yet the reality of serious illness and the capability for caregiving are complicated. Every individual is different, as are their needs and preferences. Indeed, both may change over time. Everyone, regardless of the setting in which they receive end-of-life care, should have access to high-quality symptom management and psychosocial support at the end of life.
Hospice allows many people to experience what some refer to as a “good death” in their homes. The goal of our health system should be to ensure that all Americans have the ability to choose such an opportunity.
One such person was Mary who, in her late 90s, realized she was dying. Her daughter-in-law Nancy described her final moments. “She died on April 24, peacefully and painlessly, about an hour after her hospice aide, Marie, had given her a loving bed bath and shampoo, changed her sheets, and massaged her feet and hands with oil,” Mary wrote to us in an email.
“Mary held out her hand to Marie and said ‘Thank you,’ snuggled down into her bed, and about an hour later was gone.”
Sarah H. Cross, M.P.H., is a doctoral candidate in the Sanford School of Public Policy at Duke University in Durham, N.C., and a former hospice and palliative care social worker. Haider J. Warraich, M.D., is the associate director of the heart failure program at the VA Boston Healthcare System, a cardiologist at Brigham and Women’s Hospital, and an instructor in cardiology at Harvard Medical School.