Hospice in Crisis

Total Medicare hospice payments reached $25.7 billion in 2023, up from roughly $7.6 billion in 2010. Hospice utilization reached 51.7 percent among Medicare decedents in 2023, the highest rate since 2019. The growth is driven by longer lengths of stay, not sicker patients. For-profit hospice providers now account for more than 77 percent of all hospices nationwide. Average length of stay among decedents was 95.3 days in 2022, up from 92.1 days in 2021. At the 90th percentile, length of stay reached 275 days.

The combination of per-diem payment, open-ended benefit eligibility, and weak oversight has created an environment where the financial incentive to maximize days overwhelms the clinical purpose of comfort-focused end-of-life care. Regulators, researchers, and legitimate hospice providers have raised alarms about fraud concentrated in specific geographic markets, length-of-stay manipulation, and the degradation of a benefit that was designed to provide compassionate care to the dying.

Hospice Benefit Design

#

The Medicare hospice benefit provides comprehensive palliative care for beneficiaries with a terminal illness and a life expectancy of six months or less if the disease runs its normal course. Election requires the beneficiary to waive curative treatment for the terminal condition. The benefit covers physician services, nursing care, medical equipment, drugs for symptom control, aide and homemaker services, therapy, counseling, and short-term inpatient care.

Payment occurs through per-diem rates for four levels of care. Routine home care, the most common level, pays a flat daily rate for care delivered in the patient’s home or nursing facility. Continuous home care pays a higher rate for periods of crisis requiring continuous nursing presence. Inpatient respite care covers short-term inpatient stays to relieve caregivers. General inpatient care pays the highest rate for acute symptom management that cannot be provided at home.

The hospice aggregate cap limits total Medicare payments per beneficiary. For fiscal year 2025, the cap is $34,465. If a hospice’s average payment per beneficiary exceeds the cap, the hospice must return the overpayment to Medicare. The cap is calculated at the provider level, meaning hospices with many short-stay patients can offset long-stay patients without triggering cap liability.

The per-diem structure creates inherent incentive problems. A hospice receives the same daily payment regardless of the intensity of services provided on any given day. Patients in the early days after election typically require intensive assessment, care planning, and intervention. Patients in stable periods require less intensive services. Patients in the final days before death require intensive symptom management. The per-diem flattens this variation, creating higher margins during stable periods when service intensity is low.

The For-Profit Transformation

#

The ownership composition of the hospice industry has shifted dramatically. More than three-quarters of all hospice providers were for-profit in 2022, with the remaining share divided between nonprofit and government-owned organizations. Much of the for-profit growth between 2021 and 2022 concentrated in California and Texas, two of the four states CMS has identified as fraud hotbeds requiring enhanced oversight.

Length of stay differs systematically by ownership type. MedPAC data show average length of stay of 187 days for for-profit hospices compared to 130 days for nonprofit hospices. These differences contribute to margin variation. Hospice profitability is closely related to length of stay because per-diem payment remains constant while service intensity varies over the course of enrollment.

The business model that generates margin through length-of-stay optimization creates tension with the clinical purpose of hospice. Legitimate hospice providers argue that longer stays can reflect better access to care, earlier referral, and appropriate service for patients with slow disease trajectories. Critics argue that length-of-stay patterns, particularly among for-profit providers in fraud-concentrated markets, reflect patient selection and retention practices designed to maximize days rather than clinical need.

Private equity has entered the hospice market, attracted by the combination of predictable per-diem revenue, growth in the aging population, and potential for roll-up acquisitions. PE-owned hospices have reported higher profits but lower spending on direct patient care compared to nonprofit hospices. The quality implications of PE ownership in hospice parallel concerns in nursing homes and other healthcare settings: whether financial engineering that extracts returns compromises the care mission.

The Fraud Landscape

#

The Department of Justice and HHS Office of Inspector General have intensified enforcement against hospice fraud, with prosecutions concentrated in California, Nevada, Arizona, and Texas. Common fraud patterns include enrolling patients who do not meet the six-month prognosis requirement, maintaining patients on service beyond the point where they remain eligible, billing for services not provided, and providing kickbacks for patient referrals.

Length-of-stay manipulation is the dominant fraud mechanism. A hospice that enrolls patients who are not truly terminally ill or that maintains patients on service after their condition stabilizes can collect months or years of per-diem payments. The patients may receive little actual care during stable periods because their service needs are minimal, but the daily payment continues.

The California state auditor issued a report noting that growth in the number of hospice agencies in Los Angeles County vastly outpaced the need for hospice services. Red flags for fraudulent activity include a surge in hospice numbers without corresponding increases in eligible patients. CMS has designated California, Nevada, Arizona, and Texas for enhanced oversight, requiring new hospices and those undergoing change of ownership to undergo additional scrutiny.

Live discharge patterns signal potential inappropriate enrollment. A patient discharged from hospice alive either improved (suggesting they may not have met eligibility criteria at enrollment), chose to pursue curative treatment, or moved to another hospice. High rates of live discharge can indicate that hospices are enrolling patients who do not meet the terminal prognosis requirement.

General inpatient care billing has drawn regulatory scrutiny. GIP pays the highest per-diem rate and is intended for acute symptom management that cannot be provided in the home setting. Some hospices have billed GIP for patients whose conditions did not warrant the intensive level of care, generating higher payment for services that could have been provided as routine home care.

OIG enforcement actions in 2024 involved approximately $143.8 million in alleged fraudulent activity. The 2023 figure was $198.1 million. These amounts understate total fraud because they represent only cases that reached enforcement action, not undetected improper billing.

MedPAC’s Reform Recommendations

#

MedPAC has recommended several changes to hospice payment intended to address the incentive problems inherent in the current design.

The two-tiered payment proposal would establish higher payment rates in the first days of enrollment, when care intensity is highest, and lower rates for subsequent days, when stable patients require less intensive services. This structure would reduce the financial incentive to extend length of stay because the marginal revenue from additional days would decline. Hospices serving patients with genuinely long disease trajectories would see reduced payment in later periods, but hospices manipulating length of stay would lose the financial advantage of maintaining patients on service indefinitely.

MedPAC has recommended a 20 percent reduction to the aggregate cap for several years, arguing that the cap level is too high to effectively constrain spending growth. Reducing the cap would increase financial pressure on hospices with high average lengths of stay.

Concurrent care would allow hospice beneficiaries to continue receiving some curative treatment while also receiving hospice services. Under current rules, electing hospice requires waiving curative treatment for the terminal condition. This requirement creates a barrier to timely hospice election: patients and families delay enrollment hoping for improvement, then elect hospice only in the final days of life. Concurrent care could improve the appropriateness of hospice timing by removing the either-or choice between curative and palliative approaches.

The oversight gap is structural. Hospice is among the least-surveilled Medicare benefit categories. Survey frequency is low relative to other settings, complaint investigation backlogs exist, and the enforcement gap between identifying problematic providers and removing them from the program can extend for years. The Hospice Special Focus Program, mandated by the HOSPICE Act to identify and monitor poor-performing hospices, has been slow to implement.

The Structural Benefit Design Question

#

The per-diem payment model may be fundamentally broken for a benefit with open-ended eligibility. A fixed daily payment for an indefinite benefit period creates inherent incentive to maximize days. No matter how much CMS adjusts rates, adds oversight, or prosecutes fraud, the structural incentive remains.

Alternative payment structures have been discussed. Episodic payment, similar to home health’s 30-day periods, would establish defined care episodes with case-mix adjustment. Bundled payment could create accountability for outcomes across the hospice enrollment period. Value-based payment could tie payment to quality measures and appropriate utilization patterns. Each alternative has implementation challenges, and none has advanced to serious policy consideration.

The MA hospice carve-out has been a persistent concern. Medicare Advantage plans are responsible for all other Medicare benefits but not for hospice, which remains carved out to fee-for-service Medicare. This creates fragmented accountability: the MA plan manages the beneficiary’s care until hospice election, then the hospice operates independently. MedPAC recommended eliminating the carve-out years ago, but Congress has not acted.

CMS tested an MA hospice carve-in through the Value-Based Insurance Design model, allowing some MA plans to cover hospice benefits. The evaluation found modest participation: over 23,000 beneficiaries received hospice through their MA plan in 2023. However, CMS announced in March 2024 that the hospice component of the MA-VBID model would sunset in December 2024, citing implementation challenges.

For FIDE SNPs serving dual eligibles, hospice coordination creates complexity. Dual eligibles in hospice may have Medicaid covering room and board in a nursing facility while Medicare covers hospice services. The care coordination requirements of FIDE SNPs must accommodate this split responsibility.

The hospice benefit was designed to provide compassionate, comfort-focused care at the end of life. For patients who receive legitimate hospice services from ethical providers, the benefit accomplishes its purpose. The question is whether the payment structure that has enabled fraud growth and quality variation can be reformed without undermining the benefit for patients who need it.

Related Reading

#

MCR-04_10 Medicare Fraud, Waste, and Abuse: The $60 Billion Structural Problem MCR-12_05 Home Care and PACE Organizations: HHVBP, AHEAD, and the LTSS Policy Moment