Better, if Not Cheaper, Care
This piece by Ezekiel J. Emanuel was originally posted in The New York Times Opininoater section.
It is conventional wisdom that end-of-life care is an increasingly huge proportion of health care spending. I’ve often heard it said that people spend more on health care in the year before they die than they do in the entire rest of their lives. If we don’t address these costs, the story goes, we can never control health care inflation.
Wrong. Here are the real numbers. The roughly 6 percent of Medicare patients who die each year do make up a large proportion of Medicare costs: 27 to 30 percent. But this figure has not changed significantly in decades. And the total number of Americans, not just older people, who die every year — less than 1 percent of the population — account for much less of total health care spending, just 10 to 12 percent.
The more important issue is that just because we spend a lot on end-of-life care does not mean we can save a lot. We do know that costs for dying patients vary widely among hospitals, which suggests that we can do better. And yet no one can reliably say what specific changes would significantly lower costs. There is no body of well-conducted research studies that has proved how to save 5, 10, much less 20 percent.
Recent studies find that hospice may reduce costs in the last year of life for cancer patients by 10 to 20 percent. But they find no savings from hospice care for patients who die of other conditions, like emphysema or heart failure. No one is sure why hospice care doesn’t save more. It may be because patients are enrolled in hospice care too late, or because hospice services themselves are labor-intensive and not cheap.
Even if we can never save a dime, however, there are good reasons to think about changing end-of-life care practices. While end-of-life care has improved considerably over the last 30 years, many Americans still die in hospitals when they would rather die at home. Nearly 20 percent of deaths occur in an intensive care unit or immediately after discharge, and too many patients experience symptoms like pain that are controllable with appropriate palliative care.
Here are four things the health care system should do to try to improve care for the dying, even if they won’t save money.
First, all doctors and nurses should be trained in how to talk to patients and families about end-of-life care. When I was starting out, I was lucky enough to be able to witness how a great oncologist communicated with patients and their families when it was clear they were going to die, but I received no formal training whatsoever. It is hard to improve care for the dying if health professionals don’t know how to talk about it. Fortunately, there are excellent communication techniques and training programs available — they don’t have to be invented from scratch.
A related intervention — an idea that never actually was in the Affordable Care Act but inspired the death panel accusation — is that physicians should be paid a one-time fee to talk with patients about their preferences for end-of-life care. Even if physicians are well trained in communication, these conversations take time and are emotionally draining. This should be recognized through compensation.
Third, every hospital should be required to have palliative care services available both in the hospital and at the homes of dying patients who are discharged. Over 40 percent of hospitals with more than 50 beds do not have palliative care services. And we don’t know how many actually have palliative care services once patients are sent home. These services should be delivered by trained experts in diagnosing and managing common symptoms of the dying, like pain, nausea, insomnia, shortness of breath, fatigue and depression.
Finally, we need to revise eligibility for hospice care. Right now doctors must certify that patients have six months or less to live and patients must agree to forgo life-sustaining treatments. The decision about whether to put a patient in hospice care should not be based on unreliable predictions about how long he has left to live but rather on his needs for specialized care, like morphine infusions.
These changes could be made in at least two ways. The Joint Commission — the nonprofit group that certifies health care organizations — could make training physicians and nurses to talk about end-of-life care and having palliative care available a requirement for hospital accreditation. Alternatively, Medicare, private insurers and, after 2014, state exchanges could require hospitals to provide communication training and palliative services as a condition for payment.
Unfortunately, there is no evidence that these interventions will save money. And I can’t definitively prove they will make the care of dying patients better. But doing nothing to try to help the dying when the rest of the health care system is improving care is not an option.