Medicare Hospice Model Improved Quality of Life, Reduced Medicare Spending

November 30, 2023 - The Medicare Care Choices Model (MCCM) improved end-of-life care, increased hospice use, and reduced inpatient admissions for Medicare beneficiaries while simultaneously lowering Medicare spending, a study published in Health Affairs found.

Beneficiaries can receive palliative care through the Medicare hospice benefit, but when choosing that route, they must waive the right to Medicare payment for non-hospice services that treat their terminal condition. Due to this policy, beneficiaries who want to continue conventional treatment of their illness may postpone or avoid hospice enrollment. Hospice and palliative care have been shown to improve care satisfaction and quality of life for terminally ill beneficiaries and their caregivers.

CMS developed the MCCM—tested between January 1, 2016, and December 21, 2021—which allowed eligible Medicare beneficiaries to receive palliative care from hospice providers and conventional treatments at the same time. CMS paid participating hospice providers a per beneficiary per month fee, while beneficiaries continued to receive Medicare fee-for-service coverage for Part A and Part B services related to their illness.

Researchers used Medicare claims and enrollment data from 2013 to 2021 to assess how the MCCM impacted beneficiaries’ end-of-life care and Medicare spending. They evaluated outcomes among beneficiaries with MCCM claims between the model’s start date of January 1, 2016, and the last date of model enrollment, June 30, 2021, who died on or before the model’s end date of December 31, 2021.

Of the 141 hospice providers selected to participate in the MCCM, only 81 remained in the model for the full six years, with even fewer enrolling 50 or more beneficiaries in the model. The study sample included 5,153 beneficiaries and 15,269 matched comparison beneficiaries.

MCCM beneficiaries used more healthcare services, had higher Medicare spending, and had higher Hierarchical Condition Categories (HCC) risk scores before enrollment than potential comparison beneficiaries. The final comparison group was crafted to avoid these types of differences and achieve valid comparisons.

The findings revealed that the MCCM improved end-of-life care for those enrolled in the model. For example, MCCM beneficiaries were 15.3 percentage points less likely to receive an aggressive life-prolonging procedure, surgical procedure, or diagnostic test in the last 30 days of life compared to matched beneficiaries.

MCCM enrollees also were less likely to have multiple acute hospitalizations or intensive care unit stays in the last 30 days of life (21.0 percent versus 36.8 percent) and were less likely to die in an inpatient facility (10.4 percent versus 21.8 percent).

Beneficiaries enrolled in the MCCM had lower Medicare spending than comparison beneficiaries. The average Medicare Part A and B expenditures per person were $46,810 for MCCM enrollees and $56,385 for comparison beneficiaries, marking a 17 percent difference. After accounting for model payments for MCCM services, net Medicare expenditures were 13 percent lower per person for MCCM beneficiaries.

Lower spending within the model was driven by a 38 percent reduction in inpatient expenditures, which also offset the increase in hospice expenditures, researchers noted.

MCCM beneficiaries had 26 percent fewer inpatient admissions and 12 percent fewer emergency department visits than comparison beneficiaries. Meanwhile, those enrolled in MCCM were 17.9 percentage points more likely to use hospice (83.2 percent versus 65.3 percent). MCCM enrollees also entered hospice earlier and spent an additional 22.8 days in hospice than comparison beneficiaries.

Secondary analyses indicated that over half of the reduction in Medicare Parts A and B spending was attributed to MCCM beneficiaries’ more frequent and earlier use of hospice care.

CMS did not expand MCCM after its performance period ended in 2021 due to concerns about participant attrition and generalizability. However, the study results suggest that allowing concurrent hospice care and conventional treatment can improve quality and reduce spending.

In addition, the model’s results informed end-of-life care components of other CMS models, including the Guiding an Improved Dementia Experience (GUIDE) model, the Value-Based Insurance Design (VBID) model, the Accountable Care Organization Realizing Equity, Access, and Community Health (ACO REACH) model, and the Kidney Care Choices model. These models test palliative care services alongside other interventions among a broader set of enrollees.

Editor's note: This article was updated on 12/4/23 to correct an inaccurate statement about the comparison group.