Insurers overturned 95% of appealed skilled-nursing care denials
Summary
HHS's inspector general -- the department's internal watchdog -- collected June 2024 prior-authorization data directly from the 19 largest Medicare Advantage insurers and found they denied 12% of requests to admit patients to a skilled nursing facility (SNF) after a hospital stay, then overturned 95% of those denials once a patient or doctor appealed. Nursing home residents were denied initial admission at nearly four times the rate of every other enrollee. A companion report covering the same insurers in the same month found the opposite pattern for costlier settings: only 36% of long-term-care-hospital denials and 43% of rehab-facility denials were overturned on appeal. Skilled nursing -- the cheapest of the three settings -- is the one insurers got wrong most consistently the first time.
A capitated payment, and an incentive to say no
MAOs are paid the same amount per enrollee whether that enrollee uses a little care or a lot, which OIG's own report notes⧉ creates "a potential incentive for insurers to deny enrollees' access to services in an attempt to increase profits." The data breaks along exactly that line: SNF requests processed under for-profit MAO contracts were denied 13% of the time, versus 8% under nonprofit contracts⧉, and Molina Healthcare -- the insurer with the single highest denial rate at 23.4% -- operated only for-profit Medicare Advantage contracts, as did the three largest MAOs by enrollment: UnitedHealth Group, Humana, and CVS Health (Aetna's parent), which together cover nearly 20 million people and denied SNF requests at some of the highest rates in the review.
View data as table
| Molina Healthcare | 23.4% |
|---|---|
| Kaiser Foundation Health Plan | 14% |
| Humana | 13.5% |
| CVS Health (Aetna) | 13.5% |
| UnitedHealth Group | 12.9% |
| Devoted Health | 11.1% |
| Elevance Health | 10.8% |
| Centene | 10.6% |
| BCBS of Michigan | 9.4% |
The insurer that denies the most, then reverses the most
UnitedHealth Group alone received 42% of all SNF appeals among the 19 insurers and overturned 99.7% of them⧉ -- the single biggest driver of the overall 95% rate. Every insurer that received at least 50 appeals overturned more than 90% of them; smaller insurers, with far fewer appeals to review, still overturned 59.5% combined. A UnitedHealth subsidiary called naviHealth does much of the initial reviewing: it processes half of all SNF requests among these 19 insurers -- including, for a fee, requests for rival insurers BCBS of Michigan, Highmark Health, and Humana -- and denied 14% of them, a higher rate than MAOs' own internal reviewers (11%) or other outside contractors (9%)⧉. On appeal, naviHealth's denials were overturned 97% of the time.
View data as table
| Elevance Health | 100% |
|---|---|
| UnitedHealth Group | 99.7% |
| CVS Health (Aetna) | 98.2% |
| The Cigna Group | 96% |
| Humana | 92.1% |
| BCBS of Michigan | 90.7% |
The people already in a nursing home lost most often
The starkest gap in the report is by patient, not insurer. MAOs denied 40% of SNF requests from people already living in a nursing home for more than 100 days, versus 11% for every other enrollee⧉ -- despite this group submitting only about 3% of all SNF requests. asked five large MAOs why, and three said their contractor naviHealth had written guidance instructing reviewers to weigh why a patient lives in a nursing facility⧉, on the theory those reasons often limit how much daily therapy a resident can meaningfully do. Patient advocacy and nursing-home industry groups told that reasoning is wrong on its face -- a person's address shouldn't determine whether they qualify for care a doctor ordered. Under that guidance, naviHealth denied 47% of nursing-home residents' SNF requests, versus 13% for its other enrollees.
View data as table
| Long-stay nursing home residents | 40% |
|---|---|
| All other enrollees | 11% |
Skilled nursing was the cheapest call insurers got wrong the most
A companion OIG report⧉, built from the same 19 insurers' June 2024 data, reviewed the two costlier post-acute settings: long-term care hospitals (LTCHs) and inpatient rehabilitation facilities (IRFs). Insurers denied a much larger share of those requests up front -- 65% for LTCHs and 54% for IRFs, against 12% for SNFs -- but overturned far fewer on appeal: 36% for LTCHs and 43% for IRFs, against SNF's 95%⧉. The setting insurers reversed themselves on almost every time was also, per Original Medicare's own 2023 payment data cited in that report, the cheapest of the three: $16,000 average per SNF stay, against $24,000 for inpatient rehab and $49,000 for a long-term care hospital⧉ -- and nearly three times what Original Medicare pays for home health, the lower-cost alternative a denied SNF patient is often steered toward instead. A purely cost-driven reviewer would get the cheap call right more often, not less.
View data as table
| Skilled nursing facility (SNF) | 95% |
|---|---|
| Inpatient rehabilitation facility (IRF) | 43% |
| Long-term care hospital (LTCH) | 36% |
View data as table
| Long-term care hospital | 49,000 |
|---|---|
| Inpatient rehabilitation facility | 24,000 |
| Skilled nursing facility | 16,000 |
| Home health | 6,000 |
What CMS owes, and hasn't committed to
- A 95% overturn rate is not a close call -- it is a signal the first decision was usually wrong. 's own standard is that MAOs "are expected to implement critical program controls ... while ensuring that enrollees can get necessary and appropriate care"; a coin flip that comes up 'approve' 95% of the time on review is not that control working, it is that control failing quietly, on delay, for the 82% of denied patients who never filed an appeal at all.
- The system's own math should have produced the opposite pattern. Skilled nursing is the cheapest of the three post-acute settings insurers prior-authorize, yet it had by far the highest overturn rate; the two costlier settings, where a wrong denial saves an insurer more money, were overturned on appeal far less often.
- has not committed to fixing it. made three specific recommendations -- fix the initial-review breakdowns driving the 95% overturn rate, explain the spread between insurers, and explain the four-times gap for nursing home residents -- and CMS did not concur or nonconcur with any of them⧉. Its response to the first recommendation cited only existing annual audits, without saying whether those audits will specifically target SNF denials; its responses to the other two pointed to a newly launched MA data-collection pilot and its existing Part C audit framework, respectively -- again without committing to target either specific gap. has asked to name concrete plans in its forthcoming Final Management Decision; as of this report, no date is attached.
Findings are from two companion data briefs published June 8, 2026: OEI-09-24-00331⧉ (skilled nursing facilities) and OEI-09-24-00330⧉ (long-term care hospitals and inpatient rehabilitation facilities). Both cover only June 2024 prior-authorization data from the 19 largest MAO parent companies (86% of Medicare Advantage enrollment) and only requests those insurers processed directly or through a contractor -- states explicitly that the data alone cannot determine how many of the denials were medically inappropriate, only that the overturn-rate pattern raises the question. The $16,000/$24,000/$49,000/$6,000 cost figures are Original Medicare 2023 averages cited by from MedPAC, not what MAOs themselves pay -- notes MAO-specific payment amounts by setting are not publicly available.
Sources(2) ▾
- HHS Office of Inspector General, Office of Evaluation and Inspections, Medicare Advantage Organizations Overturned Nearly All Appealed Prior Authorization Denials for Skilled Nursing Facility Admission, Raising Concerns About Initial Denials (OEI-09-24-00331) (2026-06-08) — Data brief analyzing June 2024 prior-authorization request-, denial-, and appeal-level data collected directly from the 19 largest Medicare Advantage organizations (86% of MA enrollment) for skilled nursing facility (SNF) admission. Fetched directly from oig.hhs.gov and read in full, including all exhibits and appendices summarized in the findings. oig.hhs.gov · original document
- HHS Office of Inspector General, Office of Evaluation and Inspections, The Three Largest Medicare Advantage Organizations Denied Requests for Long-Term Acute Care and Inpatient Rehabilitation at Some of the Highest Rates (OEI-09-24-00330) (2026-06-08) — Companion data brief to doc-oig-snf, released the same day from the same June 2024 MAO dataset, covering long-term care hospital (LTCH) and inpatient rehabilitation facility (IRF) admission requests instead of SNF -- used here for the cross-setting comparison (SNF vs. LTCH vs. IRF denial and overturn rates) and the MedPAC-sourced cost-per-stay table. Fetched directly from oig.hhs.gov and read in full. oig.hhs.gov · original document
Comments
Always open. Logged-in readers can annotate paragraphs in place.
In June 2024, the 19 largest Medicare Advantage organizations (MAOs) -- the private insurers pays a fixed monthly amount per enrollee to run Medicare coverage -- denied 12% of requests to admit patients to a skilled nursing facility (SNF)⧉ for short-term rehab after a hospital stay. That is a routine business decision under prior authorization⧉, the pre-approval step MAOs require before certain care gets covered. What 's inspector general found next is the story: when a patient or their doctor appealed, the same insurers overturned 95% of those denials -- 2,313 of 2,445 appeals -- admitting the care had been medically necessary all along.