What hospice gets right that the rest of healthcare doesn't

Margaret Osei had been trying to get her father's pain managed for eleven months before he enrolled in hospice. Kwame was 79, had metastatic prostate cancer that had spread to his spine, and was in the kind of pain that made him grip the arms of his recliner when he shifted his weight. His oncologist prescribed oxycodone. The prescription required prior authorization. The prior authorization took four days. The pharmacy didn't stock the dose. A different pharmacy did, but it was out of network. Margaret, who works as a paralegal in Baltimore and was managing her father's care from her apartment twenty minutes away, spent parts of three days on the phone sorting out a prescription for a 79-year-old man with cancer in his spine who was gripping his chair to change positions.

When Kwame's oncologist finally said there was nothing more to offer in terms of treatment and referred him to hospice, Margaret braced for more of the same. Instead, a nurse named Darlene arrived at Kwame's house two days later with a kit. Inside the kit were medications — liquid morphine, lorazepam, haloperidol, a few others — already dispensed, already labeled, already there. No prior authorization. No pharmacy run. No phone calls. The medications were in the house because Darlene and the hospice team had looked at Kwame's diagnosis and his trajectory and anticipated what he would need, and they had made sure it was there before he needed it.

"I kept waiting for the catch," Margaret says. "I kept waiting for someone to call and say there was a problem with the insurance or we needed a different form. It never happened. They just took care of him."

Kwame died at home six weeks later, in his recliner, with Margaret and his sister beside him. His pain was controlled. He was not in the hospital. He had not spent his last weeks in a facility he didn't recognize, surrounded by people he didn't know. He died the way he had said, for years, that he wanted to die — at home, with family, without a lot of machines.

The hospice team made that possible. Not because they had access to some special technology or some extraordinary clinical skill that the rest of medicine lacks. Because they were organized around a different question.

The question that changes everything

Most of American healthcare is organized around a single question: what can we do? What treatment is available, what intervention is indicated, what does the standard of care require? The question is not wrong. For a 40-year-old with a ruptured appendix, it is exactly the right question. For a 79-year-old with cancer in his spine and a prognosis measured in weeks, it produces a different result: a cascade of interventions, each individually defensible, that collectively fail to serve the person they are meant to help.

Hospice is organized around a different question: what does this person need to live well in the time they have? It sounds like a small shift. It is not. It reorganizes everything — who is on the care team, what counts as a good outcome, what the system is willing to pay for, how the family is treated, what happens at 2 a.m. when something goes wrong.

The hospice model was formalized in the United States in 1982, when Congress created the Medicare Hospice Benefit as part of the Tax Equity and Fiscal Responsibility Act. The benefit was modeled on the work of Dame Cicely Saunders, the British physician who founded the modern hospice movement and who argued, with considerable force, that dying patients deserved the same quality of attention as any other patients — and that the attention they deserved was different in kind from curative medicine. Saunders coined the concept of "total pain," meaning the physical, emotional, social, and spiritual dimensions of suffering that a dying person experiences and that a purely biomedical approach cannot address.

The Medicare Hospice Benefit pays a per diem rate — currently around $215 a day for routine home care — that covers essentially everything a patient needs: nursing visits, aide visits, social work, chaplaincy, medications related to the terminal diagnosis, medical equipment, and 24-hour on-call access to clinical support. The family caregiver is explicitly included in the model; hospice is one of the only Medicare-covered services that formally defines the family as part of the unit of care, not just as bystanders to it.

In exchange for this coverage, the patient agrees to forgo curative treatment for the terminal diagnosis. This is the trade that many families resist, and the resistance is understandable — it feels like giving up, like closing a door. What it actually does is open a different door, one that the rest of the healthcare system keeps locked.

What the per diem buys

The per diem model is worth examining closely, because it is the structural mechanism that makes hospice behave differently from the rest of healthcare.

In standard fee-for-service Medicare, providers are paid for each discrete service they deliver. An office visit is a billable event. A lab draw is a billable event. A phone call is not a billable event, which is why phone calls are hard to get and short when you get them. The incentive structure rewards volume and procedure. It does not reward the 45-minute conversation with a family member who is frightened and needs someone to explain what is happening. It does not reward the nurse who drives to the house at 10 p.m. to assess whether a change in breathing is something to worry about. It does not reward anticipation — the act of putting medications in the house before the patient needs them, because the team has looked at the trajectory and knows what is coming.

The hospice per diem pays for all of those things, because it pays for the outcome — a patient who is comfortable, at home, with their needs met — rather than for the individual services that produce the outcome. The hospice organization has a financial incentive to be efficient, which means anticipating needs rather than reacting to crises, because a crisis that sends a patient to the emergency department is expensive and represents a failure of the model. The incentive structure, for once, is aligned with what the patient actually needs.

This is why the comfort kit arrives before the patient needs it. This is why the nurse calls on Tuesday not because anything is wrong but because she wants to know how the weekend went. This is why the social worker sits with the family and asks not just about the patient's symptoms but about how everyone is holding up, whether anyone is sleeping, whether there are financial pressures that are making the caregiving harder. These are not acts of exceptional compassion. They are what the model is designed to produce.

Families who have experienced both standard hospital-based care and hospice describe the difference in almost identical terms: in the hospital, they felt like they were managing the system on behalf of their loved one; in hospice, they felt like the system was managing itself on behalf of their loved one. The labor of coordination — the phone calls, the prior authorizations, the pharmacy runs, the explaining and re-explaining to each new provider who the patient is and what they need — largely disappeared. Someone else was holding the thread.

The team that shows up whole

Thomas Reinholt is 61 and lives in Duluth, Minnesota. His wife, Petra, was diagnosed with glioblastoma in March 2024 and died the following October. She was 58. They had been married for 31 years.

Thomas had never interacted with hospice before Petra's diagnosis. He had a vague idea that it was for people who were very close to death, that it was somehow about giving up, that it was a service you called when there was nothing left to do. What he found, when Petra enrolled in June 2024, was something he had not expected: a team of people who seemed to understand that his wife was a person, not a diagnosis.

The hospice team included a nurse who visited twice a week, an aide who came three mornings a week to help Petra bathe and dress, a social worker named Carla who met with Thomas separately from Petra to ask how he was doing, and a chaplain named Father Brendan who was, Thomas notes, not Catholic and did not assume that Thomas was either. Father Brendan came because Petra had indicated on her intake paperwork that spiritual support was important to her. He came and sat with her and talked about what she believed and what she was afraid of and what she hoped for, and he came back the following week and did it again.

"Nobody in the hospital ever asked her what she was afraid of," Thomas says. "They asked about her symptoms. They asked about her pain level on a scale of one to ten. Nobody asked what she was afraid of."

The social worker, Carla, helped Thomas and Petra have a conversation they had been avoiding: what Petra wanted her death to look like, specifically. Where she wanted to be. Who she wanted present. What she did not want — the resuscitation attempt, the ambulance, the ICU. Carla did not push them toward any particular answer. She asked questions and took notes and made sure that what Petra said was documented in a way that would be honored. She also helped Thomas understand what to expect as Petra's condition changed — not in a clinical, detached way, but in a way that prepared him for what was coming so that he was not blindsided by it.

This is the interdisciplinary team model that hospice has practiced for forty years and that the rest of healthcare has talked about practicing for forty years without actually doing it. The difference is not philosophical. It is structural. In standard care, the physician is the center of the model, and other disciplines — social work, chaplaincy, nutrition, pharmacy — orbit the physician and are consulted when the physician decides to consult them. In hospice, the team meets weekly to discuss every patient, and every discipline has standing to raise concerns and shape the care plan. The social worker does not need a physician referral to go sit with a frightened family. The chaplain does not need an order to come back next week. The aide who notices that the patient seems more confused than last Tuesday raises it at the team meeting, and the nurse adjusts the assessment accordingly.

The result is a form of care that is attentive in ways that standard medicine is not — attentive to the whole person, attentive to the family, attentive to the things that matter to the patient that have nothing to do with vital signs.

The 24-hour phone call

There is a specific feature of hospice care that family caregivers mention more than almost any other, and it is not the medications or the aide visits or the social work. It is the phone number.

Every hospice patient has access to a clinical on-call line, 24 hours a day, seven days a week. When something happens at 2 a.m. — the breathing changes, the patient becomes agitated, the caregiver doesn't know whether what they're seeing is normal or an emergency — there is a number to call and a nurse who answers and who will either talk the caregiver through what they're seeing or come to the house.

This sounds like a minimum standard of care. It is not available in most of American healthcare. A patient with congestive heart failure who wakes at 2 a.m. with worsening shortness of breath has, in most cases, two options: wait until the office opens in the morning, or call 911. The 911 call leads to an ambulance, an emergency department, a hospitalization, and a bill. The waiting leads to anxiety and sometimes to a crisis that could have been managed at home if someone had been available to help manage it.

The hospice on-call line exists because the per diem model makes it financially rational to staff it. A call that keeps a patient at home costs the hospice organization the nurse's time. A crisis that sends the patient to the emergency department costs the hospice organization the emergency department bill, which the hospice is responsible for covering if the visit is related to the terminal diagnosis. The incentive to be available at 2 a.m. is built into the payment structure.

But the effect on families is not primarily financial. The effect is that someone is there. Not in the house — though the nurse will come to the house if needed — but reachable, present, available to say: yes, that breathing pattern is normal at this stage, here is what you can do, call us back in an hour if it changes. Or: that sounds like it needs an assessment, I'm going to come over. The availability transforms the experience of caregiving in the final weeks of a person's life from one of isolated vigilance to one of supported vigilance. The caregiver is still watching. But they are not watching alone.

Margaret Osei called the hospice line three times during the six weeks her father was enrolled. Once at 11 p.m. when his pain seemed worse than usual and she wasn't sure whether to give the extra dose of morphine she'd been told she could give. Once at 4 a.m. when his breathing changed and she was frightened. Once the morning he died, when she needed someone to tell her what to do next. Each time, someone answered. Each time, she got what she needed.

"I had spent eleven months feeling like I was fighting the healthcare system to get my father basic care," she says. "For six weeks, I felt like the healthcare system was on my side. It was the same disease. It was the same insurance. It was a completely different experience."

What doesn't travel, and why

If hospice works this well — and the evidence that it does is substantial, with studies consistently showing that hospice patients have better pain control, fewer unwanted hospitalizations, and higher family satisfaction than comparable patients who don't enroll — the obvious question is why the rest of healthcare doesn't work the same way.

The answer is not that the rest of healthcare doesn't know about hospice. The hospice model has been studied, written about, and held up as an example for four decades. Palliative care, which applies hospice principles to patients who are not yet at end of life, has grown significantly in hospitals over the past twenty years. The interdisciplinary team model, the focus on goals of care, the attention to symptom management and family support — these ideas are not secret. They are taught in medical schools and nursing programs and social work programs. They are the subject of conferences and journal articles and quality improvement initiatives.

They are not, in most places, how care is actually delivered. The reason is structural, not attitudinal.

The fee-for-service payment model that governs most of American healthcare does not pay for the things that make hospice work. It does not pay for the weekly team meeting. It does not pay for the social worker who meets with the family separately from the patient. It does not pay for the chaplain. It does not pay for the on-call nurse at 2 a.m. It does not pay for the comfort kit assembled in advance. It pays for procedures and visits and tests, and so procedures and visits and tests are what get delivered, whether or not they are what the patient needs.

Palliative care programs in hospitals have made real progress, but they operate within the same payment constraints. A palliative care consultation is a billable event. The follow-up conversation with the family is harder to bill for. The ongoing relationship — the team that knows the patient over weeks and months and adjusts the care plan as the situation evolves — is nearly impossible to sustain in a system that pays per encounter.

There is also a cultural dimension that is harder to name but impossible to ignore. American medicine is organized around the belief that more intervention is better, that the goal is always to extend life, and that a physician who recommends comfort-focused care rather than aggressive treatment is somehow failing the patient. This belief is not universal among physicians — many doctors understand, deeply, that there are situations where aggressive treatment causes harm rather than benefit — but it is embedded in the incentive structure, the liability environment, and the expectations of patients and families who have been taught that more medicine means better medicine.

Hospice challenges that belief directly. It says, explicitly, that there are situations where the most medically appropriate thing to do is to stop trying to cure the disease and start trying to support the person. That reframing is threatening to a system built on the premise that giving up is failure. The hospice model has survived and grown because it is Medicare-funded and because the evidence for its effectiveness is strong. But it has not transformed the rest of the system, because the rest of the system is not organized to be transformed by evidence alone.

The enrollment problem

There is one thing hospice gets wrong, or at least one thing it has failed to solve: people enroll too late.

The median length of hospice enrollment in the United States is about 18 days. The Medicare Hospice Benefit allows enrollment for six months, with extensions available if the patient continues to meet eligibility criteria. The gap between what is available and what patients actually use is enormous, and it represents an enormous amount of suffering that could have been prevented.

The reasons for late enrollment are multiple and overlapping. Physicians are reluctant to have the conversation that leads to a hospice referral, partly because prognostication is genuinely difficult and partly because recommending hospice feels, to some physicians, like admitting defeat. Patients and families resist the conversation because hospice has been framed, culturally and sometimes by the healthcare system itself, as giving up hope. The eligibility criteria — a prognosis of six months or less if the disease runs its natural course — require a physician to make a prediction that medicine is not always good at making.

The result is that families like Margaret Osei's spend months fighting a system that is not designed to help them, and then spend six weeks in a system that is, and then the six weeks are over. The care that could have been available for months is compressed into days or weeks at the very end. The social worker who could have helped the family prepare for what was coming meets them when there is almost no time left to prepare. The chaplain who could have helped the patient work through what they believed and what they feared arrives when the patient can barely speak.

This is not a failure of hospice. It is a failure of the system that surrounds hospice — the system that delays the conversation, that frames the referral as a last resort rather than a care option, that leaves families to fight for eleven months before offering them six weeks of something that actually works.

What families can do with this

The practical implication for families is uncomfortable but important: the hospice conversation needs to happen earlier than feels natural, and families often need to be the ones who initiate it.

When a family member has a serious, life-limiting illness — advanced cancer, end-stage heart failure, advanced dementia, ALS, any condition where the trajectory is toward decline rather than recovery — it is worth asking the physician directly: would my family member qualify for hospice? If not now, when? What would need to change for a hospice referral to be appropriate? These questions are not giving up. They are gathering information about a care option that may become relevant and that takes time to access and benefit from.

It is also worth knowing that palliative care — which is not the same as hospice and does not require forgoing curative treatment — is available earlier in the disease trajectory and can provide some of the same benefits: symptom management, goals-of-care conversations, family support, coordination across the care team. Asking for a palliative care consultation is not a step toward death. It is a step toward care that takes the whole person seriously.

And it is worth knowing, for the families who are already in the middle of it, that the experience Margaret Osei described — the prior authorizations, the pharmacy runs, the phone calls, the feeling of fighting the system on behalf of someone you love — is not inevitable. It is the product of a payment model and a care model that were not designed with the patient's actual needs at the center. Hospice was. It is not perfect, and it is not available to everyone, and it comes too late for too many people. But it exists, it is covered by Medicare and most Medicaid programs, and it is, in the experience of the families who have used it, something close to what healthcare should feel like.

Kwame Osei died in his recliner with his daughter beside him. His pain was controlled. The nurse had called that morning to check in. The comfort kit was on the kitchen counter. Nobody asked Margaret to fill out a form.

That is not a small thing. In the context of American healthcare, it is close to a miracle. It should not be.