Licensing posture

ARKA operates on a first-party AIIE standard with no third-party criteria license in the live data path. Scoring draws on peer-reviewed clinical literature, the CMS Appropriate Use Criteria Program structure established by PAMA §218(b), and the RAND/UCLA Appropriateness Method for the 1–9 scale semantics — not on embedded or transcribed proprietary criteria tables.

ARKA's AIIE engine is designed to operate independently of any single criteria licensor — drawing on peer-reviewed clinical literature, CMS Appropriate Use Criteria Program standards (PAMA §218(b)), and first-party machine-learning validation.

Methodology detail: AIIE Methodology & Evidence Basis · Evidence Methodology Dossier

CMS Appropriate Use Criteria Program — current status

The posture below reflects the latest human- or system-confirmed snapshot from CMS and Federal Register sources. Automated monitoring may propose changes; an admin confirms reinstatement or activation with a citation.

What this means for buyers

• No CDSM consultation is required on Medicare fee-for-service claims while the program is paused.
• ARKA already meets the program's clinical-standard intent via the first-party AIIE engine and peer-reviewed evidence register.
• If CMS reinstates the mandate, ARKA is positioned to re-qualify as a CDSM-aligned mechanism at point of order.

Honest regulatory band

Three dated facts. None of the business case depends on any of them.

• WISeR (1 January 2026) — CMS's Wasteful and Inappropriate Service Reduction model pairs AI with licensed clinical review on the denial path for select services.
• CMS-0057-F (8 February 2024) — Impacted payers publish machine-executable prior-authorisation rules; ARKA executes whichever rules apply rather than authoring its own.
• Imaging AUC (1 January 2024) — The programme ARKA was originally aligned to was withdrawn, and the post-mortem is why ARKA no longer treats a regulation as demand.

Congress's own proposed correction to the imaging AUC programme removes the real-time claims-reporting requirement and keeps consultation at the point of care. That is the failure mode ARKA rules out by construction rather than by promise. It is not a reason to buy ARKA.

What we will not touch

The two concepts share a word and nothing else. Beneficiary assignment is set by programme rule, not by us. Contribution measurement is the product. The commitments below are generated from the never-do catalog — the same artefact CI checksums — with the named enforcement mechanism on each row.

1. 01

   ARKA does not compute, influence, or recommend changes to beneficiary attribution or to risk-adjustment coding. Attribution and risk scores are read-only inputs supplied by your organisation. We measure avoided utilisation against the attribution and risk model you give us, and we publish the specification of how we use them.

Coding firewall — the note is invisible to measurement surfaces

Whether a considered-and-deferred note exists is invisible to measurement, comparison, attribution, incentive and sales surfaces. It is a clinical artefact and nothing else. The commitments below are generated from the never-do catalog — the same artefact CI checksums — with the named enforcement mechanism on each row.

1. 01

   No ARKA surface, artefact, metric, export, contract clause or piece of marketing may reference evaluation-and-management coding, billing level, code selection or reimbursement in connection with the considered-and-deferred note — and the build fails if one does.

   contractual

   Objection: The write-back is a coding-intensity tool, and you will teach my clinicians to document for level.

   lint · CHECK-CF-1 in scripts/lint-coding-firewall.ts; docs/coding-firewall.json computed disjointness; Clause 10 in docs/arka-aj/DEPLOYMENT_AGREEMENT_CLAUSES.md

   Proves: No write-back-touching module or copy may carry banned coding lexicon (CPT, E/M, level of service, MDM, data/risk element, complexity, documentation improvement, capture, uplift, reimbursement); clinical-note and billing-relevant field sets are a computed set-difference

2. 02

   Whether a considered-and-deferred note was written may not enter the attribution ledger, the peer-comparison packet, the contribution export, or any incentive computation. The write-back module graph is unreachable from attribution, ledger and incentive, and the build proves it.

   contractual

   Objection: Clinicians who write the note will look better on the measure than clinicians who do not — you built a documentation incentive by accident.

   lint · CHECK-CF-2 / CHECK-CF-3 in scripts/lint-scope-boundary.ts; CHECK-CF-4 / CHECK-CF-5 in scripts/lint-coding-firewall.ts and scripts/lint-tcoc.ts; Clause 10 in docs/arka-aj/DEPLOYMENT_AGREEMENT_CLAUSES.md

Appropriateness is not coverage

We do not author coverage criteria and our clinical engine cannot read them. Appropriateness cites evidence; coverage cites the specific policy paragraph. Where they disagree we report the disagreement rather than resolving it.

1. 01

   A payer-authored or utilisation-management source may inform a coverage determination and may never inform clinical appropriateness. The two field sets are disjoint and the build proves it.

   contractual

   Objection: Your clinical score is the payer's criteria with a white coat on.

   lint · CHECK-APPR-1 in scripts/lint-tcoc.ts; CHECK-FW-1..3 in lint:sources / lint:coverage / lint:scope; the disjoint-input proof in docs/appropriateness-firewall.json; Clause 9 in docs/arka-aj/DEPLOYMENT_AGREEMENT_CLAUSES.md

   Proves: No payer-authored or utilisation-management source may inform a clinical appropriateness rating. Verdict A (score.ts and the knowledge matrix) and Verdict B (lib/coverage) are computed from disjoint field sets in docs/appropriateness-firewall.json; the clinical engine cannot reach lib/coverage, lib/pa, lib/davinci or lib/ins even transitively; no ModalityRating carries a payer_um / payer_utilization anchor.

Enforcement artefact: docs/appropriateness-firewall.json. The two field sets are a computed set-difference; the build fails if they overlap.

How to challenge a contribution figure

A vendor publishing how to challenge its own numbers is unusual. It is the reason a physician group will accept the module at all.

1. Generation. A figure is computed against a locked pre-period. For 14 days it is provisional and cannot enter any pay export.
2. Dispute. A clinician files a dispute with a reason. The record moves to disputed and is excluded from export until resolved.
3. Resolution. Resolutions are appended, never overwritten. Only after the window has closed and no open dispute remains is the figure exportable.
4. Attenuation. Where the interval includes zero, the exported figure is attenuated toward zero rather than the point estimate, and the attenuation is stated. Paying on a figure statistically indistinguishable from no change is refused.

There is no override that exports a provisional or disputed figure. The pressure to use one arrives at payroll deadline; the type system refuses it.

Policy-trigger watchlist

Section 25 item 39. Each row is a dated answer — yes, no, or unchecked — with a Source Register id and a retrieval date. Legislative triggers also carry a staged history (introduced → … → enacted); a recent stage move shortens the re-check interval (CHECK-WL-4). A recorded negative is kept because it was tested. Re-check intervals are enforced by CHECK-WL-2 in the lint; CHECK-WL-3 fails if any row is still unchecked. This page prints the stored retrieval date only.

Official references

Federal and CMS primary sources for the AUC program and related interoperability rules. All links open in a new tab.

Licensing posture

ARKA operates on a first-party AIIE standard with no third-party criteria license in the live data path. Scoring draws on peer-reviewed clinical literature, the CMS Appropriate Use Criteria Program structure established by PAMA §218(b), and the RAND/UCLA Appropriateness Method for the 1–9 scale semantics — not on embedded or transcribed proprietary criteria tables.

ARKA's AIIE engine is designed to operate independently of any single criteria licensor — drawing on peer-reviewed clinical literature, CMS Appropriate Use Criteria Program standards (PAMA §218(b)), and first-party machine-learning validation.

Methodology detail: AIIE Methodology & Evidence Basis · Evidence Methodology Dossier

CMS Appropriate Use Criteria Program — current status

The posture below reflects the latest human- or system-confirmed snapshot from CMS and Federal Register sources. Automated monitoring may propose changes; an admin confirms reinstatement or activation with a citation.

What this means for buyers

• No CDSM consultation is required on Medicare fee-for-service claims while the program is paused.
• ARKA already meets the program's clinical-standard intent via the first-party AIIE engine and peer-reviewed evidence register.
• If CMS reinstates the mandate, ARKA is positioned to re-qualify as a CDSM-aligned mechanism at point of order.

Honest regulatory band

Three dated facts. None of the business case depends on any of them.

• WISeR (1 January 2026) — CMS's Wasteful and Inappropriate Service Reduction model pairs AI with licensed clinical review on the denial path for select services.
• CMS-0057-F (8 February 2024) — Impacted payers publish machine-executable prior-authorisation rules; ARKA executes whichever rules apply rather than authoring its own.
• Imaging AUC (1 January 2024) — The programme ARKA was originally aligned to was withdrawn, and the post-mortem is why ARKA no longer treats a regulation as demand.

Congress's own proposed correction to the imaging AUC programme removes the real-time claims-reporting requirement and keeps consultation at the point of care. That is the failure mode ARKA rules out by construction rather than by promise. It is not a reason to buy ARKA.

What we will not touch

The two concepts share a word and nothing else. Beneficiary assignment is set by programme rule, not by us. Contribution measurement is the product. The commitments below are generated from the never-do catalog — the same artefact CI checksums — with the named enforcement mechanism on each row.

1. 01

   ARKA does not compute, influence, or recommend changes to beneficiary attribution or to risk-adjustment coding. Attribution and risk scores are read-only inputs supplied by your organisation. We measure avoided utilisation against the attribution and risk model you give us, and we publish the specification of how we use them.

Coding firewall — the note is invisible to measurement surfaces

Whether a considered-and-deferred note exists is invisible to measurement, comparison, attribution, incentive and sales surfaces. It is a clinical artefact and nothing else. The commitments below are generated from the never-do catalog — the same artefact CI checksums — with the named enforcement mechanism on each row.

1. 01

   No ARKA surface, artefact, metric, export, contract clause or piece of marketing may reference evaluation-and-management coding, billing level, code selection or reimbursement in connection with the considered-and-deferred note — and the build fails if one does.

   contractual

   Objection: The write-back is a coding-intensity tool, and you will teach my clinicians to document for level.

   lint · CHECK-CF-1 in scripts/lint-coding-firewall.ts; docs/coding-firewall.json computed disjointness; Clause 10 in docs/arka-aj/DEPLOYMENT_AGREEMENT_CLAUSES.md

   Proves: No write-back-touching module or copy may carry banned coding lexicon (CPT, E/M, level of service, MDM, data/risk element, complexity, documentation improvement, capture, uplift, reimbursement); clinical-note and billing-relevant field sets are a computed set-difference

2. 02

   Whether a considered-and-deferred note was written may not enter the attribution ledger, the peer-comparison packet, the contribution export, or any incentive computation. The write-back module graph is unreachable from attribution, ledger and incentive, and the build proves it.

   contractual

   Objection: Clinicians who write the note will look better on the measure than clinicians who do not — you built a documentation incentive by accident.

   lint · CHECK-CF-2 / CHECK-CF-3 in scripts/lint-scope-boundary.ts; CHECK-CF-4 / CHECK-CF-5 in scripts/lint-coding-firewall.ts and scripts/lint-tcoc.ts; Clause 10 in docs/arka-aj/DEPLOYMENT_AGREEMENT_CLAUSES.md

Appropriateness is not coverage

We do not author coverage criteria and our clinical engine cannot read them. Appropriateness cites evidence; coverage cites the specific policy paragraph. Where they disagree we report the disagreement rather than resolving it.

1. 01

   A payer-authored or utilisation-management source may inform a coverage determination and may never inform clinical appropriateness. The two field sets are disjoint and the build proves it.

   contractual

   Objection: Your clinical score is the payer's criteria with a white coat on.

   lint · CHECK-APPR-1 in scripts/lint-tcoc.ts; CHECK-FW-1..3 in lint:sources / lint:coverage / lint:scope; the disjoint-input proof in docs/appropriateness-firewall.json; Clause 9 in docs/arka-aj/DEPLOYMENT_AGREEMENT_CLAUSES.md

   Proves: No payer-authored or utilisation-management source may inform a clinical appropriateness rating. Verdict A (score.ts and the knowledge matrix) and Verdict B (lib/coverage) are computed from disjoint field sets in docs/appropriateness-firewall.json; the clinical engine cannot reach lib/coverage, lib/pa, lib/davinci or lib/ins even transitively; no ModalityRating carries a payer_um / payer_utilization anchor.

Enforcement artefact: docs/appropriateness-firewall.json. The two field sets are a computed set-difference; the build fails if they overlap.

How to challenge a contribution figure

A vendor publishing how to challenge its own numbers is unusual. It is the reason a physician group will accept the module at all.

1. Generation. A figure is computed against a locked pre-period. For 14 days it is provisional and cannot enter any pay export.
2. Dispute. A clinician files a dispute with a reason. The record moves to disputed and is excluded from export until resolved.
3. Resolution. Resolutions are appended, never overwritten. Only after the window has closed and no open dispute remains is the figure exportable.
4. Attenuation. Where the interval includes zero, the exported figure is attenuated toward zero rather than the point estimate, and the attenuation is stated. Paying on a figure statistically indistinguishable from no change is refused.

There is no override that exports a provisional or disputed figure. The pressure to use one arrives at payroll deadline; the type system refuses it.

Policy-trigger watchlist

Section 25 item 39. Each row is a dated answer — yes, no, or unchecked — with a Source Register id and a retrieval date. Legislative triggers also carry a staged history (introduced → … → enacted); a recent stage move shortens the re-check interval (CHECK-WL-4). A recorded negative is kept because it was tested. Re-check intervals are enforced by CHECK-WL-2 in the lint; CHECK-WL-3 fails if any row is still unchecked. This page prints the stored retrieval date only.

Official references

Federal and CMS primary sources for the AUC program and related interoperability rules. All links open in a new tab.