A federal watchdog says Medicare could have avoided $255.1 million in improper hospice payments tied to new enrollees who, on closer review, didn’t clearly qualify for hospice care in the first place. The finding comes from the Department of Health and Human Services’ Office of Inspector General, which examined a sample of hospice claims from fiscal year 2021 and found nearly half fell short of Medicare’s documentation and eligibility standards. For families weighing hospice care for an aging parent or spouse, the audit is a reminder that Medicare’s hospice benefit rests on a specific medical standard, and that the government is actively watching how consistently that standard gets applied at the point of enrollment.
A 100-File Sample Found Problems in Nearly Half
OIG reviewed 100 initial hospice certification periods and found that the documentation behind 45 of them did not meet Medicare’s requirements. For 21 of those periods, the clinical information in the enrollee’s medical records did not support a terminal illness diagnosis, producing $251,067 in payments auditors classified as unallowable. For another 24 periods, the records simply didn’t meet the documentation requirements hospice eligibility calls for, adding $294,432 more.
Extrapolating those error rates across the full population of new hospice enrollees in its sampling frame, OIG estimated that $255.1 million in hospice claim payments could have been avoided during the audit period if stronger eligibility checks had been in place, according to the inspector general’s report, numbered A-06-22-09003.
Free retirement updates: Keep more of a Social Security check and retirement savings with plain-English updates on the changes, deadlines, and costly mistakes retirees miss. Subscribe free.
Why Enrollees Without a Recent Hospital Stay Drew Scrutiny
Medicare’s hospice benefit is built around a specific medical judgment: a physician must certify that a patient has a life expectancy of six months or less if the illness runs its normal course. OIG’s audit zeroed in on a particular slice of new enrollees — those who had no inpatient hospital or emergency room claims in the 18 months before starting hospice care. That absence doesn’t automatically mean an enrollee was ineligible, but auditors treated it as a signal worth checking, since a terminal prognosis often follows some kind of acute medical event or hospital stay. The hospice benefit itself is administered through Medicare Administrative Contractors, regional claims processors that pay hospice providers directly, and OIG’s recommendation centers on giving those contractors formal review procedures for exactly this enrollee profile before payment goes out rather than after.
The audit does not allege that any individual patient or family did something wrong. Hospice election is a decision families and physicians make together, often under difficult circumstances, and the report’s findings are about the documentation hospice providers kept, not about whether any specific patient deserved care. Where the report draws a hard line is on paperwork: if a hospice’s own records don’t show the clinical basis for a terminal diagnosis, or don’t meet Medicare’s eligibility documentation standard, the claim doesn’t hold up on audit regardless of the care that was actually delivered.
How an Initial Certification Period Gets Reviewed
The 100 files OIG examined were all “initial certification periods” — the first stretch of hospice coverage a patient enters once electing the benefit, which under Medicare rules requires a physician’s written certification that the six-month prognosis standard is met. A second physician typically has to concur on that first certification. Auditors pulled the underlying medical records for each sampled period and checked them against Medicare’s own hospice coverage and documentation requirements rather than relying on the certification paperwork alone. That distinction matters: a hospice can have a signed physician certification on file and still fail an OIG review if the clinical notes behind it don’t independently support the terminal diagnosis or don’t meet the separate documentation standard Medicare requires.
Because OIG worked from a statistical sample rather than every hospice claim nationwide, the $255.1 million figure is an estimate built by applying the 45% error rate found in the 100-file sample to the full population of new hospice enrollees in the audit’s sampling frame, not a tally of dollars OIG traced claim by claim. That method is standard for large-scale program-integrity audits, since checking every hospice file in the country for a single fiscal year isn’t practical, but it also means the number represents a projected exposure rather than a confirmed loss.
CMS Agreed, but the Fix Isn’t in Place Yet
CMS concurred with OIG’s recommendation to work with the hospice Medicare Administrative Contractors on developing pre-payment or post-payment review procedures targeted at new enrollees who lack a recent inpatient or emergency room claim. As of the report’s public tracker, that recommendation remains listed as “Open – Unimplemented,” with OIG’s next status update expected by December 17, 2026. Until CMS and the contractors put a specific review process in place, the audit’s core finding is that the government has identified a fixable gap in oversight but has not yet closed it. The $255.1 million figure applies to the fiscal year 2021 claims OIG sampled; the report does not estimate ongoing annual exposure, but the same review gap it describes would apply to any new hospice enrollee whose file lacks a recent hospital or ER claim until CMS finalizes a fix.
Inside the kit
51 state Medicare cost-help packs, the new Part D out-of-pocket cap, the prior-authorization appeal steps and a medication and cost tracker.
The Medicare Cost & Coverage Protection Kit includes a 10-page kit, 51 state Medicare cost-help packs and the new Part D out-of-pocket cap.
Open The Medicare Cost & Coverage Protection Kit.
This article was produced with the assistance of AI and reviewed by The Financial Wire editorial team.



