About 1.72 million Medicare beneficiaries received hospice care in the United States last year — roughly half of every Medicare death. Yet the median patient enters hospice with only 17 days left to live, and the industry that delivers that care has changed almost beyond recognition in a generation. Here is what the data says about hospice in America, drawn from CMS, NHPCO, MedPAC, and peer-reviewed research.
The big picture
Hospice is no longer a niche service. In 2023, the most recent year with complete data from the National Hospice and Palliative Care Organization (NHPCO), about 1.72 million Medicare beneficiaries received hospice care, and 51.7% of all Medicare decedents used the benefit before death. That share has more than doubled since 2000, when fewer than a quarter of Medicare deaths involved hospice.
The money has scaled accordingly. Medicare spent roughly $26 billion on hospice in 2023, making it one of the fastest-growing line items in the program. There are now more than 6,800 Medicare-certified hospice providers nationwide — up from about 2,300 in 2000, and a number we maintain in our own directories, county by county.
For most American families navigating a terminal illness today, hospice is no longer the exception. It is the default.
Length of stay: how late is too late?
The single statistic that gets cited most often in the field, by clinicians and policy analysts alike, is the median length of stay. It sits at 17 days.
That number is easy to misread. The mean length of stay is more than five times longer (about 92 days), pulled up by a small share of patients with predictably slow-declining conditions like advanced dementia who stay enrolled for many months. But the median tells the story clinicians find most frustrating: about 1 in 4 patients enroll in hospice during their final week of life. Half are dead within 17 days.
The Medicare hospice benefit covers up to six months of care, with extensions possible when a physician recertifies the prognosis. Almost no families come close to using that much time. Research consistently finds the same three reasons: the attending physician does not raise hospice early enough, the family resists what feels like “giving up,” or the hospital discharge planner only mentions it in the last 48 hours of an inpatient stay. Even with rising enrollment, late referral remains the biggest gap between what the benefit can do and what families actually receive.
What people die of in hospice
The headline shift over the past two decades is the rise of dementia. Cancer was the dominant hospice diagnosis through the 1990s — the benefit was designed for it. Today, Alzheimer's disease and other forms of dementia are the single largest category of hospice patients, at roughly 20% of admissions.
- Cancer: about 30% of patients
- Alzheimer's and other dementias: about 20%
- Cardiovascular disease: about 16%
- Respiratory illness (COPD, etc.): about 10%
- Stroke, kidney disease, ALS, and other non-cancer: the remainder
The average age at admission is about 80 years, and the largest single age bracket is 85 and older. Hospice in America is overwhelmingly geriatric care. The Medicare benefit's prognosis-based eligibility (life expectancy of six months or less if the disease runs its normal course) shapes both who qualifies and how late they arrive.
Where hospice happens
Most hospice care is not delivered in a hospice facility. About three-quarters of all care days happen wherever the patient calls home — a private residence, an assisted-living community, or a nursing home. Patients move between settings as their condition changes; general inpatient (GIP) care in a hospital or hospice inpatient unit is reserved for symptom crises that cannot be managed elsewhere.
- Routine home care (the patient's residence): about 52% of all patient-days
- Nursing facility: about 28%
- Assisted living: about 14%
- Inpatient (GIP, respite, or general): about 6%
The preference for home care is strong and well-documented: roughly 70% of Americans, when asked, say they want to die at home. The Medicare hospice benefit is one of the few policy levers that makes that possible at scale.
The shift to for-profit ownership
Of every structural change in the field, this is the largest and the most consequential.
In 1992, around 5% of Medicare-certified hospices were for-profit. By 2024, that share was approximately 73%. The Medicare Payment Advisory Commission (MedPAC) reports that hospice margins are unusually high for healthcare — about 16% on a Medicare-aggregate basis in 2022, several times the margin most other post-acute providers operate on.
That margin has driven a wave of private-equity acquisitions. The ten largest hospice chains now treat roughly a third of all hospice patients in the U.S. The consolidation isn't neutral: research consistently finds that for-profit hospices spend less per patient-day on nursing and physician visits and enroll a higher share of long-stay dementia patients, where the daily Medicare payment is most profitable. CMS has responded with a Hospice Special Focus Program, launched in 2024, that identifies and intensively monitors the worst-performing 1% of providers.
Family caregiver satisfaction
How families experience hospice is measured directly through the CMS CAHPS Hospice Survey — a standardized federal questionnaire mailed to a sample of family caregivers two to twelve months after the patient's death.
The aggregate numbers are remarkably high for U.S. healthcare:
- About 81% of caregivers rate hospice 9 or 10 out of 10 (top-box rating).
- About 84% say they would definitely recommend the hospice to friends and family.
- Satisfaction with emotional and spiritual support is the single highest-rated component — roughly 89% say the team provided the right amount.
- The component caregivers rate lowest is help with training to care for the patient at home (about 76% say they always got the training they needed).
The pattern is consistent: families overwhelmingly say the hospice team handled the emotional and clinical work well. Where hospice falls short is in equipping family caregivers to physically deliver the day-to-day care.
Who gets hospice — and who doesn't
Hospice utilization varies sharply by race, geography, and Medicaid status. Among Medicare decedents:
- White patients: about 53% use hospice
- Black patients: about 38%
- Hispanic patients: about 43%
- Asian and Pacific Islander patients: about 38%
The reasons are well-studied: longer-running historical distrust of the medical system, language and cultural framing of end-of-life decisions, and lower physician referral rates in communities of color. Rural patients face a different problem — rural counties have roughly 40% fewer hospice providers per capita than urban ones, and average travel time per home visit is much longer, which limits how many patients each team can carry.
This is one of the few areas where the data is unambiguous about an inequity that matters at the end of life.
The bereavement benefit (often forgotten)
Less well known: the Medicare hospice benefit includes thirteen months of bereavement support for the patient's family after death — counseling, support groups, anniversary contacts, sometimes a memorial. It is a Medicare Condition of Participation; hospices that fail to offer it can lose their certification.
And yet, fewer than half of eligible families use any of it. The reasons are mundane: the family isn't aware it's included, the offerings are buried in a sympathy mailing, or the surviving spouse simply doesn't feel up to a phone call. For agencies that take the bereavement program seriously, it's often the component that families remember most.
What's changing in 2025–2026
A few trends to watch:
- Telehealth visits have stabilized at about 30% of routine contacts after the COVID surge. CMS extended the telehealth recertification waiver into 2025; expect permanent rules within 18 months.
- The Hospice Outcomes & Patient Evaluation (HOPE) tool replaced the older Hospice Item Set as the federal data-collection standard in 2025. Expect richer quality reporting and, eventually, an updated star-rating methodology.
- Hospice fraud enforcement is intensifying, especially in California, Nevada, Arizona, and Texas, after a wave of new-provider applications in 2019–2022 that turned out to include hundreds of paper hospices. CMS has imposed moratoriums on new Medicare hospice certification in those four states.
- Per-beneficiary payment reform is on the MedPAC docket. The current routine-home-care rate is widely seen as overpaying long-stay dementia care relative to the cost of providing it.
How to use any of this
Statistics are useful as orientation. The decision in front of a family is always local: which providers serve our city, what their patient surveys say, who answers the phone at 2 a.m. when symptoms escalate.
That's what our county directories are for. Each one pulls live data from the CMS Provider Data Catalog, adds Medicare CAHPS family-caregiver ratings, layers in Google review counts for additional signal, and surfaces the providers Medicare reports as permanently closed so you aren't calling dead phone numbers. Browse by county or by city, compare up to three providers side by side, and check the “Top Rated” section for the providers families have rated 4 stars or better on the federal survey.
Primary sources: NHPCO Facts and Figures (2024); MedPAC Report to the Congress: Medicare Payment Policy (March 2024); CMS Provider Data Catalog — Hospice General Information & CAHPS Hospice Survey; peer-reviewed research summarized in Kaiser Family Foundation and JAMA Internal Medicine.
To put any of these numbers into local context, browse Medicare-certified hospice providers in your area — Alameda County, Contra Costa County, Fresno County, Kern County, Los Angeles County, Orange County, Riverside County, Sacramento County, San Bernardino County, San Diego County, San Joaquin County, San Mateo County, Santa Clara County, and Ventura County.