Claims Integrity

Pre-payment screening for public health programs. Sovereign, explainable, built to survive appeal.

Oction LabsSovereign AI Infrastructure

OCTION LABS

The Thesis

The money is already gone by the time anyone reads the claim.

OCTION LABS01 / 13

OCTION LABS

The Official Number

Medicare and Medicaid combined

over $100 billion

Medicaid

$50 billion

Medicare fee-for-service

$31 billion

Medicare Advantage, Part C

$17 billion

Sources: HHS estimates reported in GAO-24-107487. These programs accounted for 43 percent of all government-wide improper payments that year.

OCTION LABS02 / 13

OCTION LABS

A Distinction We Do Not Blur

GAO defines improper payments as payments "either made in an incorrect amount or should not have been made at all." That is broader than fraud. It includes documentation failures, eligibility errors and coding mistakes alongside deliberate abuse.

Anyone who tells a program integrity director that the whole hundred billion is fraud has told them something they know to be untrue. The opportunity is real without the exaggeration.

OCTION LABS03 / 13

OCTION LABS

Why Pay And Chase Loses

Claims clear, then a fraction are sampled for audit

Every claim is evaluated at the gateway before payment

Recovery proceeds years later through appeals and litigation

The overwhelming majority pass untouched, in the payment path

Money owed by a dissolved entity is money not recovered

A small minority are held, with the reason attached

Reviewable volume is a rounding error against what was paid

Volume is not the constraint, because nothing is read by hand

OCTION LABS04 / 13

OCTION LABS

A Claim Does Not Look Wrong On Its Own

It looks wrong next to everything else. The signal is in the relationships around the claim, not the fields inside it.

OCTION LABS05 / 13

OCTION LABS

What Screening Actually Catches

Impossible aggregates

All same-day claims for one rendering provider, against the hours in a day

Eligibility contradictions

Date of service against the eligibility record, including date of death

Identity movement

One beneficiary identifier across unrelated providers, states and a compressed window

Referral topology

Provider clusters collapsing onto shared addresses or ownership

Peer distribution

Complexity and volume against the same specialty, setting and population

OCTION LABS06 / 13

OCTION LABS

The Case That Decides Whether A Program Keeps It

A large facility billing like a large facility is billing correctly. A screener that holds it generates work, delays legitimate payments, and gets switched off within a quarter. High volume alone is not a finding.

OCTION LABS07 / 13

OCTION LABS

You Are Withholding A Payment

Every flag has a sentence

Not a probability. A statement of which signals fired and what was concluded, readable by an adjudicator and a provider

The evidence is retained

Claim as submitted, comparison records, signal outputs and timestamp, held as one record so the decision can be reconstructed months later

Humans decide

The system holds and explains. It does not deny. A person with authority makes the adverse determination

OCTION LABS08 / 13

OCTION LABS

Architecture

Ingest

Claims evaluated at the gateway, on your infrastructure, at program volume. No beneficiary record leaves the environment

Memory

Providers, beneficiaries, facilities and referral paths held as connected records rather than separate tables

Retrieve

Each signal produces a stated finding with the records behind it, not a contribution to an opaque score

Act

Held claims route to the right reviewer with evidence attached, and the decision record is retained for appeal

OCTION LABS09 / 13

OCTION LABS

Why It Runs On Your Infrastructure

Protected health information for an entire covered population, moving at gateway volume. The deployment model is not a preference here. It is the only one that clears review.

OCTION LABS10 / 13

OCTION LABS

Engagement

Quoted against scope. We do not price on a share of recoveries.

What sets the number

Claim volume through the gateway, programs and lines of business in scope, which systems connect, the state of historical data, and the review workflow holds route into

First step

Retrospective evaluation against a defined window of already-adjudicated claims, inside your environment

OCTION LABS11 / 13

OCTION LABS

What We Do Not Claim

We publish no recovery figures and no detection rates, because we have not run this against your claims. The retrospective evaluation exists so those numbers come from your data rather than from our marketing.

Any vendor quoting you a detection percentage before touching your environment is quoting you someone else's program.

OCTION LABS12 / 13

OCTION LABS

Screen before it clears.

Let us scope a retrospective evaluation against your own adjudicated claims.

OCTION LABS13 / 13