Video Preview
🎧 Podcast episode for paid subscribers only. Also available on Apple Podcasts and Spotify.
To listen to paid episodes in Apple or Spotify, link your Substack subscription via the show settings on those platforms (instructions inside the Substack app under Subscriptions → Podcast).
Table of Contents
The press release, decoded
Anomaly detection is a solved problem, sorry
The part where someone drives to the strip mall
Suspension, revocation, and the due process meat grinder
Cat and mouse in Texas
What this means if you build, invest, or sell in program integrity
Abstract
CMS announced 1.6 billion dollars in prevented or recovered fraudulent Medicare lab payments, including 732 million from revoking 157 lab providers, 500 million plus in payment suspensions across 185 labs (out of 600 investigated), 276 million recouped from 442 overpayments, and 127 million tied to 85 law enforcement referrals.
The detection side of this story is genuinely easy now. Claims data plus an LLM writing analysis code gets you 90 percent of the flagging logic in an afternoon. The features are public knowledge and have been for a decade.
The hard 10 percent is everything after the flag: site visits proving 14 labs were never operational, medical review, credible allegation standards, suspension mechanics, revocation authority, and surviving an appeals gauntlet designed to protect legitimate providers.
The Texas cases in the release show the adversarial loop in action: a lab gets denied 1.2 million, shifts billing patterns within weeks to dodge the edits, and gets caught again only because humans kept watching.
Implication for builders and investors: the model is not the product. The defensible asset in fraud is the case file, the evidentiary chain, and the boots. Services layer economics apply, with everything that means for margins and moats.
