Dr. Mehmet Oz Says Taxpayers Paid for PHANTOM Care

Judge's gavel beside cash and scales of justice on a wooden desk — Photo: AVN Photo Lab / Shutterstock

The hard lesson from Minnesota’s Housing Stabilization Services collapse is not simply that fraud occurred; it’s that enrollment controls and program design determine whether fraud stays isolated or metastasizes into a system-wide failure—yet administrative screening outcomes are not the same thing as proven fraud.

The Short Version

One proven case can topple a fragile program: a federal jury convicted a provider of defrauding Minnesota’s Housing Stabilization Services (HSS), and the state shuttered the program amid ballooning costs.
Revalidation screens enrollment risk; it is not a fraud verdict. Minnesota’s mass revalidation removed thousands from the rolls largely for incomplete or deficient paperwork, not adjudicated schemes.
Improper payment and fraud are not interchangeable. National watchdogs warn that error rates capture documentation and eligibility failures alongside deception; they don’t quantify fraud.
The durable fix pairs tight, risk-based provider screening and site checks with fast, evidence-driven adjudication—so essential services aren’t collateral damage.

What actually happened in Minnesota’s HSS program

Minnesota launched Housing Stabilization Services to help people with disabilities and those at risk of homelessness maintain housing—coordination and support services typically delivered outside the clinical setting. Within a few years, the program’s spending exploded far beyond early projections, and state officials ultimately shut it down while federal investigations accelerated. That decision did not occur in a vacuum: a federal jury convicted Sharmaine Meadows on multiple counts of health care fraud tied to HSS, finding that she fabricated services and siphoned millions through claims for work that did not occur. The U.S. Department of Justice summarized the case as a straightforward false-claims scheme that exploited the program’s structure and weak controls.

The scale mismatch was striking. Legislative recaps noted HSS was projected to cost a few million dollars per year at launch; actual payouts reached nine figures as the program matured, with tens of millions more in the subsequent period, even as red flags mounted and criminal cases surfaced. The Meadows conviction is dispositive for her conduct—it establishes that fraud, not mere paperwork error, occurred in HSS. But it does not, standing alone, define a program-wide fraud rate. That distinction matters for what came next.

What revalidation is—and is not

After HSS careened off course, Minnesota required expedited “revalidation” for providers in high‑risk categories. Revalidation is an enrollment and screening function: the state asks providers to reaffirm licensure, location, ownership, exclusions history, and other integrity markers. It is a gatekeeping tool to prevent obviously noncompliant or dubious entities from billing going forward. By the state’s account, roughly 5,583 providers in high‑risk lines were reviewed on deadline. About 2,061 were approved; approximately 3,400 received disenrollment notices, largely for incomplete submissions or administrative deficiencies in their filings, with some later reinstated through appeals.

The crucial point, stressed by Minnesota’s own program integrity office, is that disenrollment from revalidation is not a finding of fraud. It can reflect missing or mismatched documentation, failure to appear ready for an unannounced site visit, unresolvable ownership questions, or other risk indicators that justify cutting off future payments while issues are cured. The agency’s fact-check page is plain: screening is not an investigation or adjudication; administrative disenrollment does not, by itself, establish fraud, waste, or abuse. Independent coverage echoed the same distinction and reported that the state flagged a small subset—dozens, not thousands—for deeper review based on what revalidation revealed.

How real fraud risk coexists with administrative error

Medicaid sits in a well‑documented tension: improper payments—overpayments, eligibility errors, and missing documentation—are prevalent and costly, and fraud lives within that universe but is only a portion of it. The Government Accountability Office (GAO) has, for years, cautioned that national improper payment estimates are not a fraud meter; they cannot be read as the percentage of corrupt outflows. In 2023, Medicaid improper payments were pegged at roughly $50 billion; the figure is a composite of administrative misses and true schemes, and GAO’s guidance is unequivocal that the fraud fraction cannot be directly inferred from that total.

This matters in Minnesota. The Meadows conviction and other HSS charges demonstrate genuine criminal exposure in a high‑risk benefit category; that is the justification for heightened screening and site checks. But a mass failure to successfully revalidate—particularly if driven by incomplete applications or process errors—does not equate to a mass fraud finding. National program‑integrity literature backs the strategy that Minnesota adopted: for high‑risk categories, do on‑site inspections, verify the physical practice and operations, and require periodic re‑enrollment to keep the registry clean. GAO has long recommended such risk‑based screening, including unannounced visits, as a core control against sham providers.

The numbers that bear weight—and the ones that don’t

Where numbers are concerned, two threads deserve confidence. First, the criminal case record: Meadows’s conviction, tied to millions in false HSS claims, is grounded in courtroom scrutiny and federal charging documents. Second, the program outlays: legislative write‑ups track HSS’s growth into the nine‑figure range, a magnitude that, combined with the charging activity around HSS-linked entities, supports the conclusion that Minnesota faced material integrity failures in that line of service.

By contrast, claims that equate a specific share of failed revalidations with a fraud rate demand caution. Minnesota’s own disclosures attribute most disenrollments to incomplete administrative submissions and reiterate that revalidation is not a fraud adjudication—language that directly rebuts any one-to-one mapping from screening failure to criminality. Reporting that only 59 providers—about 1 percent of those revalidated—were formally referred for further review undercuts the narrative that revalidation itself uncovered systemic fraud at a one‑in‑three clip. Screening is a sieve; referral is the beginning of proof.

Design choices that determine outcomes

HSS illustrates how program architecture shapes risk. Non-clinical, community‑based services are susceptible to “phantom billing” because units of service—coordination calls, housing navigation, accompaniment—are hard to verify post hoc if documentation rules are lax and claims systems lack corroboration hooks. Effective countermeasures pair prepayment controls (credential checks tied to licensure databases, exclusion lists, ownership transparency, and site verification) with postpayment analytics that reconcile claimed hours against feasible human capacity and beneficiary context—hospitalizations, overlapping services, or implausible travel. The Meadows case reportedly involved billing patterns that would fail these plausibility tests; strong analytics should flag them upstream.

The institutional response should be proportionate: aggressively prosecute obviously fabricated care and the organizers behind it; fix enrollment defects that let shell providers pass initial screens; and preserve access for legitimate providers by separating curable administrative faults from indicia of deception. GAO’s body of work reinforces this balance: align antifraud operations with a risk framework, target high‑yield controls (like ownership transparency and site visits), and treat improper payment metrics as a steering input, not a verdict on fraud prevalence.

Dr. Oz said CMS asked Minnesota to revalidate providers in 14 high-risk categories, and roughly 30 percent—almost one in three—could not be revalidated.

He explained revalidation as checking whether providers were authorized and eligible to deliver the services they were being…

— Mary Dinan (@dinan_mary) October 30, 2026

What to watch next

Three markers will tell you whether Minnesota is getting this right. First, referral-to-conviction yield in HSS-adjacent cases: are the entities flagged in screening translating into prosecutable fraud or largely administrative cleanups? Second, rebound and stability in legitimate service capacity: are approved providers able to meet need without the perverse incentives that fueled HSS growth? Third, maturation of data-sharing and financial forensics: collaborations that connect claims anomalies to money flows—bank records, beneficial ownership, and tax filings—will separate paperwork sloppiness from organized theft. Sustained progress on those fronts converts a crisis response into lasting program integrity.