Sunday, May 10, 2026

What? CMS Runs RFI on Why Labs Hate Prior Authorization

A big thanks to Vitali Khvatkov at Linked In for highlighting a CMS RFI on 'Prior authorization" with specific attention to the nightmare it creates for clinical laboratories.   (See also a blog by Brendan Keeler).  (See also a CMS announcement and Fact Sheet.)

Find the April 14, 2026, Fed Reg page here:

https://www.federalregister.gov/documents/2026/04/14/2026-07205/medicare-and-medicaid-programs-patient-protection-and-affordable-care-act-interoperability-standards

Go straight to the 173-page proposal here.  April 14, 2026, 91 FR 19890-20062.

https://www.govinfo.gov/content/pkg/FR-2026-04-14/pdf/2026-07205.pdf

Comment by June 15.

The word Laboratory occurs 42 times, mostly in the RFI section, Section III, Part E, page 20014-16. This concerns whether labs have special added burdens when confronted with prior authorization portals and denials.

What CMS Asks For [AI Corner - Claude Opus 4.7]

####

CMS Prior Authorization RFI on Labs: What CMS Said, and What They Didn't

CMS has opened a Request for Information on prior authorization for laboratory tests (bundled with DMEPOS) inside its larger interoperability rule. 

The lab RFI lives at Section III.E of CMS-0062-P, published at 91 FR 19890 on Tuesday, April 14, 2026. The framing matters: CMS is openly acknowledging — in the Federal Register, with citations — that lab prior auth in the commercial and MA worlds is generating real harm. That acknowledgment is the soft underbelly worth pressing on in comments.

Friday, May 8, 2026

AI Guest Column: the Roche-PathAI Acquisition?

Big news that Roche buys PathAI for up to $1B.  I pulled together some twenty viewpoints harvested from Linked In, press releases, a Health Advances podcast, and put Chat GPT into "agentic" mode to review it all.  I specifically asked for some perspective from the earlier Tempus-Paige deal.

As always with this kind of blog, the point is not exactly what stance Chat GPT takes from the material, but rather it's a case study how an AI can attempt to bring business strategy ideas together.  

(See some meta-comments about how the 2000-word article was constructed by AI, as a Sidebar: Building the Article, at bottom.  Also find a Sidebar tie-in with ArteraAI's breast prognostic test win at FDA.)

###

AI CORNER

### 

Roche’s $1B PathAI acquisition signals that digital pathology is moving from promising software to strategic diagnostic infrastructure. Despite narrow reimbursement pathways for routine AI pathology, PathAI gives Roche a platform for workflow efficiency, biopharma services, companion diagnostics, and future computational biomarkers. Tempus’s Paige acquisition followed a different logic: enriching its multimodal oncology data engine. 

Both deals suggest digital pathology’s value may come less from standalone payment than from ecosystem control.

###

Roche buying PathAI is less a bet that digital pathology will suddenly become generously reimbursed, and more a bet that AI pathology will become infrastructure for oncology diagnostics, companion diagnostics, and drug development. PathAI brings Roche an FDA-cleared image-management platform, an open AI ecosystem, enterprise deployments, biopharma trial tools, and a foundation-model strategy. Tempus buying Paige was a different but equally logical move: Tempus wanted pathology data and AI to extend its molecular, clinical, and multimodal oncology engine. In short: Roche bought a workflow/CDx platform company; Tempus bought a data/model/oncology-AI accelerator. Both may have bought the “right” digital pathology company for their own strategic problem.

Roche Buys PathAI: Why This Deal Makes Sense

Thursday, May 7, 2026

H.R. 8500: Ensuring Timely LCDs! The Bill's Language is Posted!

HEADER: 

Old News:
Last week, CAP and some Congressionial offices announced H.R.8500, the "Timely Access to Coverage Decisions Act." Early news -  Here.  

New News:
The legislative language is now posted!  

  • Find the full text of the bill in "Congressional Legalesehere.
  • Find the full contents of the bill described in "Just Plain English" right here below.
    • For students seeking extra credit, the full text in Latin, here.
  • See detailed HR 8500 coverage on subscription at 360Dx here.

###

AI Corner Chat GPT

##

Timely Access to Coverage Decisions Act of 2026 — 

What It Would Do in Plain English

TLDR — The bill would put a clock on the Medicare LCD process. When a MAC receives a formal LCD request or reconsideration request, it would have 60 days to say whether the request is complete. If incomplete, the MAC must identify what is missing. If complete, the MAC must act within one year. The bill also strengthens the public draft-LCD process: open meetings, remote access, expert input, posted meeting records, public comments, written responses, and a requirement that the final LCD be a logical outgrowth of the draft.

##

A new House bill, the Timely Access to Coverage Decisions Act of 2026, would put firmer timelines and procedural guardrails around how Medicare Administrative Contractors handle Local Coverage Determinations — the LCDs that decide whether Medicare will cover many services, tests, procedures, and technologies in a MAC region. The bill was introduced  in April 2026, by Rep. Neal Dunn of Florida, with Reps. Nanette Barragán and Claudia Tenney as cosponsors.

In practical terms, the bill tries to solve a familiar problem: an innovator, physician group, lab, manufacturer, or other stakeholder can ask a MAC for a coverage decision, but the request may sit for a long time with no clear endpoint. This legislation would not guarantee a favorable LCD. But it would require the MAC to say whether the request is complete, identify what is missing if it is not complete, and act within a defined time period if it is complete.


Deep Dive: The New UPMC/Pitt Report on Access to Genomic Testing

UPMC report here.  

See subscription coverage at Genomeweb here. (Alexandra Byrne.)

###

AI CORNER (Chat GPT)

##

UPMC Precision Medicine Report: 

When “Access” Means More than Just the Test 

Precision medicine has spent two decades being described as the future of care. The new UPMC/University of Pittsburgh report offers a more practical update: in many health systems, the future has arrived, but it is still stuck in the plumbing.

The issue is no longer just whether genetic tests exist, or whether they can be sent to a reference lab. The harder question is whether a health system can order the right test, manage payment, return the result, place it usefully in the EHR, interpret it, and act on it.

That is the central message of Operationalizing Precision Medicine: How U.S. Health Systems Are Implementing Genetic Testing, a 22-page report from UPMC’s Center for Connected Medicine, the University of Pittsburgh Institute for Precision Medicine, and KLAS Research. The report is based on interviews with 21 U.S. health system leaders conducted between October 2025 and February 2026. A GenomeWeb/Precision Medicine Online article by Alexandra Byrne highlighted the same theme: programs are maturing, but reimbursement remains a persistent drag.

HUMOR: I Start Using CLAUDE OPUS. Claude Opus Reviews "Me"

Header: Using CLAUDE.   I have been using Chat GPT exclusively for three years.   Recently, friends have introduced me to areas where Claude Opus has advantages.   I asked Claude Opus to review "me."  I haven't altered the result - it's below.



###

Claude Opus 4.7 Researches Bruce Quinn MD

Professional Identity

Bruce Quinn is an expert on health reform, innovation, and United States health policy and payment systems — a pathologist by training who has worked on a wide range of health policy issues ranging from biopharma to genomics. His consulting practice, Bruce Quinn Associates LLC, is based in Southern California, and he is widely recognized as one of the leading independent voices on Medicare policy for diagnostics and precision medicine.

Wednesday, May 6, 2026

Legislation Would Demand Faster LCDs. See H.R. 8500 from Congressman Dunn (R FL).

SEE UPDATE: THE BILL IS POSTED.

CAP and LCD Policy

College of American Pathologists has taken the point position, for years, on improving LCDs at Medicare.  For example, some of their ideas were incorporated in 21st Century Cures, which passed during the Obama-Trump transition.

New News

April 28, CAP announces creation of H.R. 8500, "Timely Access to Coverage Decisions Act" (TACDA).

I'll cut and paste the CAP News below.  You can also scroll  for similar text about "8500" at this CAP page here.  See a media quote from CAP here.

See also a Congressional press release here.

The actual legislation for HR 8500 is not posted online yet. The lead sponsor is Rep. Neal Dunn R FL, and cosponsors are Nanette Barragan D CA and Claudia Tenney R NY.

See also the "CAP Issue Brief" link at bottom.



Tuesday, May 5, 2026

Medicare's 14 Day Rule and 3 Day Rule: Why One is Crazy and the Other Isn't

 .

I learned much about this when I recently asked Chat GPT to write long detailed essays on the history of each rule, 3-day and 14-day

 Since I first worked as a CMS medical director in 2004–2008, I have been familiar with Medicare date-of-service rules and their interaction with hospital billing. For inpatient admissions, Medicare generally bundles related hospital services, and all diagnostic services, furnished in the three days before admission into the inpatient DRG payment. This is the familiar three-day rule. Separately, since 2007, Medicare has applied a special laboratory 14-day rule, under which certain lab tests can be pulled back to the hospital episode based on the date of specimen collection.

  • I recently learned much more about these policies by asking ChatGPT to generate long, detailed essays on the history of both rules: the three-day rule and the 14-day laboratory rule. 

The two rules have very different origins. The three-day rule is longstanding Medicare law, grounded directly in statute. The statute uses relatively straightforward language: diagnostic services furnished within three days before admission, and certain “related” other services furnished by the same hospital or hospital system, are treated as part of the inpatient stay and bundled into the DRG payment. It is plain English and, conceptually, simple English.

click to enlarge

Congr Comer Asks CMS - Does AMA CPT Complexity Support Fraud?

Header: Congressman Comer (KY) writes CMS, raises questions about AMA CPT structure and coding, and fraud and abuse issues.

Comer letter here.   Comer press here.   STAT PLUS (subsc.) here.

###

Could the complexities of US coding systems contribute to healthcare fraud? Certainly something went awry in Medicare - when the Novitas MAC got lost in the forest of  CPT genetic codes and assumed some rare costly genetic codes (81408 and others) would never be needed or expected in Medicare.  And therefore had no edits on them.  Result: One to two billion dollars flowed out through just those unexpected codes (over s i x y e a r s, half under Trump I and half under Biden) before it was stopped.

###

RFK Jr has also raised concerns about the US coding system.

###

AI Corner

Chat GPT 5.5

###

Detailed Summary: 

Comer Letter, Comer Press Release, and STAT Article on CPT Codes, AMA, and Fraud

TLDR

Takeaway Message

The Comer documents mark a notable escalation in the politics of medical coding.

CPT has long been criticized as an AMA-controlled proprietary standard, but Comer places it inside the current Washington frame of fraud, waste, abuse, and federal healthcare cost control. His letter asks CMS not only how it polices improper AMA CPT coding, but whether CMS has the authority to simplify or move away from the current CPT-based system altogether. That makes this more than a routine anti-fraud inquiry. It is potentially a challenge to one of the basic operating systems of American healthcare reimbursement.



####

Monday, May 4, 2026

Chat GPT Discusses the Serena-6 Trial, ESR1 by LBx, and FDA Advisory Committee ODAC

Chat GPT summarizes the data on the recently released SERENA-6 trial and ODAC-FDA advisory board.  OncLive here.  Fierce Biotech here.  Precision Medicine Online (Turna Ray) here.

###



Chat GPT:

SERENA-6 was a Phase III trial in HR-positive, HER2-negative advanced breast cancer testing whether therapy should be changed when an ESR1 mutation first appears in circulating tumor DNA, before conventional radiologic progression. Patients on first-line aromatase inhibitor plus CDK4/6 inhibitor therapy were serially monitored with Guardant360 ctDNA testing. When an ESR1 mutation was detected but imaging had not yet shown progression, patients were randomized.

Sunday, May 3, 2026

RFK Blog 02: Chat GPT Assesses the Seven RFK-Hill Transcripts (April 2026)

 

RFK Jr.’s Seven-Hearing Hill Marathon: April 16–22, 2026

Blog 01 - I asked Chat GPT to find all 7 hearings and identify transcripts. Blog 01 is all the source material for Blog 02.

Blog 02 - Here - I ask  Chat GPT to write about the 7 documents, based on 200,000 words in 7 transcripts (see Blog 01 including ZIP file). Prompt at bottom.

##
See Claude Opus version.  See Google Notebook LM (free) version.
See a comparison of the 5 "humor" sidebars here.

##
RFK Jr.’s Seven-Hearing Hill Marathon: April 16–22, 2026

Robert F. Kennedy Jr.’s late-April swing through Congress was not one hearing but a seven-stop stress test of his second-year leadership of HHS. The hearings ranged across House and Senate authorizers and appropriators: Ways & Means, House Appropriations, Energy & Commerce, Senate Appropriations, Senate Finance, and Senate HELP. The master file lists the seven events, their dates, committees, topics, and video links; despite the file title’s “April 14 to April 22” framing, the seven hearings documented in the source set begin on April 16 and run through April 22.

The result was a compressed portrait of Kennedy’s HHS: ambitious, disruptive, rhetorically potent, and politically combustible. His message was consistent. America is sick; the old health system is broken; HHS has been captured by profit-driven and status-quo institutions; chronic disease, nutrition, fraud, rural hospital decline, drug prices, and bureaucratic friction must be attacked at the root. He repeatedly framed the department’s agenda as a generational correction, including food policy, prior authorization, rural health, faster FDA approvals, ultra-rare disease programs, drug pricing, and fraud enforcement. In his own Ways & Means opening, he said the country was at a “generational turning point,” that children were the “sickest generation in modern history,” and that HHS was replacing policies that fueled chronic disease with policies that put Americans’ health first.

But the hearings also showed the limits of that message. Democrats repeatedly focused on vaccines, measles, CDC changes, NIH and CDC cuts, Medicaid reductions, contraception, grant terminations, and whether Kennedy’s public-health worldview had moved from skepticism into governance. Republicans, by contrast, often treated him as the rare cabinet officer who was willing to say the system was rotten and needed to be rebuilt. The result was a week in which Kennedy’s strengths and weaknesses were almost identical: he was most effective when talking broadly about incentives, chronic disease, rural care, and fraud; he was weakest when cornered on precise yes/no policy commitments, technical details, or scientific claims.

click to enlarge


AI Guest Column: Twenty Years of Lab Date of Service Rules

 

Memorandum on the Medicare Laboratory Date of Service Rule

Executive summary

 

This memorandum was written by Chat GPT in its “deep research” mode, after I gave it brief remarks on some of the key dates to look for (1997, 2001, 2006.)  

I would add an explanation of how outpatient hospital patients  have been handled:

Outpatient Hospital Specimens

1) Before 2006, tests from hospital outpatients to outside labs were generally billed by the outside lab to its local MAC.

2) The 2006 rule abruptly changed this, requiring billing by the hospital only.  This raised the risk its local MAC might not cover, or might inappropriately price, the test run by the distant lab when billed by the local hospital.

3) In 2013, CMS made most routine lab tests bundled to hospital outpatient events (like surgery or E.R.)   However,  CMS stated that molecular pathology tests were payable – if they were billed only by the local hospital to its local MAC and on the patient's biopsy DOS.

4) In 2017, CMS stated that molecular pathology tests could be billed by whichever lab performed them, and, on the date they were performed.   Hospital labs would bill CMS using their local date of test performance, and outside reference labs would bill CMS using their date of test performance, and billing the lab’s own MAC.

The rest of this memo is written by Chat GPT in “deep research” mode.

 

click to enlarge

The operative congressional push for national laboratory uniformity, requested by industry, came from the 1997 Balanced Budget Act, enacted by the U.S. Congress[1]. The required negotiated-rulemaking process began in 1998, produced a proposed rule in March 2000, and culminated in a November 23, 2001 final rule by the CMS.   That 2001 rule, not the 2006 rule, is where CMS first selected specimen collection as the default national date of service for laboratory claims. CMS implemented the policy in 2002 program memoranda, clarified the archived-specimen concept in 2005, and codified the modern DOS framework in 42 CFR 414.510 in December 2006, effective January 1, 2007. Strictly speaking, the “14-day rule” is only one branch of the broader Date of Service policy, not the whole policy. [2]

The strongest modern reform case is not that CMS acted irrationally in 2006. It is that the rule was built for a late-1990s uniformity and anti-unbundling problem, while today’s oncology market relies far more heavily on post-discharge molecular decision support. In 2006 CMS focused on relatively narrow stored-specimen and chemo-sensitivity scenarios and repeatedly emphasized fear of unbundling hospital care. In 2017 CMS itself acknowledged that the existing DOS policy could delay outpatient molecular testing, then created a direct-billing outpatient exception for qualifying molecular pathology tests and Criterion A ADLTs, while expressly declining to extend the same logic to inpatient specimens because of DRG/IPPS ratesetting implications. That means any serious stakeholder effort to change inpatient policy must combine the clinical-access argument with a narrow test list, anti-gaming safeguards, and a credible answer on inpatient payment policy. [3]

RFK Blog 01: Chat GPT Saves Hours: Finding the 7 RFK Jr Hearings

Blog 01 - HERE - I asked Chat GPT to find all 7 hearings and identify transcripts.  HERE.

Zip file of the 7 transcripts at google drive here.  For example, upload the transcripts into your AI and ask it questions (E.g., "Did the meetinsg discuss telehealth?")

Blog 02 - I ask  Chat GPT to write about the 7 documents.
https://www.discoveriesinhealthpolicy.com/2026/05/rfk-blog-02-chat-gpt-assesses-seven.html


###

 Over the past few weeks, the Secretary of Health, RFK Jr, appeared before seven different Hill committes, sessions which were covered in variable levels of detail by the mainstream media and the health journalism media.

I put Chat GPT in "Deep Research" mode and asked it to find all seven hearings, list them by date, and wherever possible provide links to archived video of each whole hearing.   At many sources, like YouTube, those archived hearings will ahve autotranscript text available too.

Here's the result.  We summarize the 7 hearings and locations first, then a deeper dive follows.

Friday, May 1, 2026

CMS Announces Summer CLFS Pricing Meetings + Special PAMA Pricing Meeting

CMS has posted dates for its regular summer meetings for new lab code pricing.  In addition, CMS has planned a September two-day meeting for pricing those codes with missing or anomalous results in the current PAMA pricing survey.

click to enlarge

Public Comment June 10

CMS will hold its public comment meeting for pricing new lab tests for 2027, on June 10, 2026.  That's quicker than the normal late-June date.  Submit your comments by May 29.  

Normally, the summer pricing cycle has about 100 new codes (mostly PLA codes.)  This year, the number may be closer to 200 new codes, becaues of a bumper crop of PLA code applications being handled by AMA CPT right now (April-May 2026).

https://www.govinfo.gov/content/pkg/FR-2026-05-01/pdf/2026-08511.pdf

CMS Experts July 14-15

CMS convenes its expert panel to discuss each new code and recommend a price (or gapfil) for each new code.   You can watch by livestream.

https://www.govinfo.gov/content/pkg/FR-2026-05-01/pdf/2026-08512.pdf

No Meeting: CMS Releases Proposed Prices: 1H September

During 1H September, CMS will release its proposed prices, having taken account of the public meeting and the expert meeting.

Third Meeting: PAMA Pricing Problems, Stpember 15-16

CMS will hold a meeting to discuss how it should price tests that were part of the PAMA data collection process (codes activated by second quarter 2025), but for which no data was submitted in the May-June-July current PAMA data survey.   This is September 15-16, with a registration date of "August 3 to August 31."  CMS will not know what tests had no PAMA data until after the close of the PAMA submisison process, May 1 - July 31. (Read more about the data collection process itself here.)

https://www.govinfo.gov/content/pkg/FR-2026-05-01/pdf/2026-08513.pdf




Wednesday, April 29, 2026

AI Guest Column: Can It Explain the Three-Day Rule (for DRG)

 Can Chat GPT explain gnarly tortuous Medicare policies?   One example is the 3-day-rule.  Services to a patient by a hospital are bundled for 3 days backwards from the admission date.   This includes time in the emergency room, but potentially quite a bit more.  It's  a tricky policy and I asked Chat GPT to research and explain it.  

This pre-admission three-day rule, which bundles services up to 3 days prior to a hospital admission, does not require the legal fiction used in the 14 day post-admission rule.  The latter  conjectures that due to its date of service, a test run in California July 10 was actually run at the local hospital on July 1 (the date of inpatient biopsy, date of service). 

The three day rule we discuss in this article does not require those verbal and legal gymnastics (or fantasies) because it is based on statute  at 1886(a)(4) from OBRA 1990 and subsequent regulation at 42 CFR 412.2(c)(5).  (See also Claims, 100-04, Ch 3, 40.3.)

 

click to enlarge

###

Medicare’s three-day rule sounds obscure, but it shapes hospital billing for labs, imaging, and some outpatient services before admission. This short essay traces its post-DRG history, the 2010 expansion to nondiagnostic services, Condition Code 51, ownership boundaries, and why legal structure—not clinical logic—often drives payment.

###

Medicare’s three-day payment window is one of those rules that sounds like a technical billing footnote but actually teaches a great deal about how Medicare thinks about hospitals. The basic idea is familiar: if a Medicare beneficiary receives outpatient diagnostic services at a hospital—say, a CT scan, laboratory tests, or imaging studies—and is then admitted to that same hospital within the payment window, those outpatient services are not separately paid but are swept into the inpatient DRG claim. Historically, this was not part of the original 1983 DRG/IPPS statute. The modern three-day rule traces mainly to OBRA 1990, which expanded an older same-day policy into a broader preadmission payment window. CMS describes the rule today as applying to services furnished by the admitting hospital, or by an entity wholly owned or wholly operated by that hospital, during the three calendar days before admission for IPPS hospitals, or one day for certain non-IPPS hospitals. (Centers for Medicare & Medicaid Services)

The first core rule is that preadmission diagnostic services are bundled very broadly. This includes clinical laboratory tests and imaging, and the policy does not turn on whether the diagnostic service later proves to be clinically related to the admission. If the patient has a CT scan Tuesday and is admitted Wednesday to the same hospital, the CT scan is effectively wrapped forward into the inpatient stay. The policy is therefore not just a clinical-relatedness doctrine; for diagnostics, it is a payment-boundary rule. It prevents hospitals from carving out preadmission diagnostic workups as separately payable outpatient services immediately before an inpatient DRG stay.

The less familiar twist came in 2010, when Congress addressed nondiagnostic outpatient services. The relevant law was the Preservation of Access to Care for Medicare Beneficiaries and Pension Relief Act of 2010, Pub. L. 111-192, section 102. CMS says this law did not change the diagnostic-service rule, but it did change the handling of admission-related nondiagnostic services. Before 2010, nondiagnostic services were treated as related largely through a diagnosis-code matching approach. After 2010, the rule became broader: outpatient nondiagnostic services furnished during the window are generally treated as related to the admission unless the hospital determines and attests that they are clinically unrelated. CMS later summarized the rule as requiring hospitals to include on the inpatient claim all outpatient diagnostic services and admission-related outpatient nondiagnostic services furnished within the window. (Centers for Medicare & Medicaid Services)

That 2010 change creates a practical distinction. For diagnostics, the rule is essentially automatic. For nondiagnostic services, it is a rebuttable presumption. Suppose a patient receives outpatient physical therapy on Monday for chronic osteoarthritis, then is in a car accident on Wednesday and is admitted to the trauma service. It would make no clinical sense to bundle the arthritis PT into the trauma admission. Under the post-2010 system, the hospital can treat those nondiagnostic services as unrelated, but it must be able to support that determination. Operationally, the hospital uses Condition Code 51, formally an attestation of unrelated outpatient nondiagnostic services. CMS guidance says unrelated nondiagnostic services are not subject to the window, and the hospital should retain documentation supporting the unrelated determination. (Centers for Medicare & Medicaid Services)

This is where the rule becomes operationally interesting. Condition Code 51 is not modifier 51, which is a CPT modifier for multiple procedures. It is a condition code on the institutional claim. In practice, coding or revenue-cycle staff make the determination, often under HIM or compliance protocols, and may escalate gray cases. The claim is then processed with the attestation; CMS does not generally adjudicate the clinical relationship in real time. The system is essentially pay now, audit later. The hospital’s separate outpatient claim survives unless later challenged by a MAC, RAC, or other reviewer. This means the 2010 rule is less like a purely algorithmic claims edit and more like hospital billing jurisprudence: a broad presumption, a self-attested exception, and post-payment audit risk.

Ownership is the other key boundary. The rule clearly applies when the preadmission services are furnished by the same hospital. It also applies when services are furnished by an entity that is wholly owned or wholly operated by the admitting hospital. That phrase matters. A hospital-owned outpatient department or wholly controlled entity can be swept into the rule. But loose affiliations, referral alliances, branding relationships, clinically integrated networks, or academic partnerships generally do not create the same result if they fall short of ownership or operation. Thus, a CT scan at Hospital A followed by admission to Hospital A is bundled; a CT scan at a legally separate Hospital B followed by admission to Hospital A may be separately paid, even if the two institutions cooperate closely. CMS and MAC guidance repeatedly frame the rule around the admitting hospital and wholly owned or wholly operated entities, not around general affiliation. (Centers for Medicare & Medicaid Services)

The policy lesson is that Medicare’s three-day rule follows legal and billing structure, not always clinical reality. It was grafted onto the DRG system after DRGs had already been created, as an anti-unbundling rule. It expanded over time from a same-day concept, to a three-day diagnostic window, to a broader 2010 framework for nondiagnostic services. For laboratories and hospital strategy, the rule is a reminder that inpatient and outpatient payment boundaries are often artificial but financially decisive. A test performed just before admission may be clinically identical to one performed elsewhere, but its payment treatment can turn on timing, ownership, site of service, and claim coding. In that sense, the three-day rule is a small Medicare rule with a large teaching value: it shows how DRG payment, outpatient bundling, hospital ownership, revenue-cycle attestation, and audit risk all intersect in the ordinary life of a hospital bill.

Chat GPT Explains Its Understanding of Where AMA Appendix S (AI Services) Comes From

For a year, AMA has been extensively revising its Appendix S, which is guidance for categories and coding of software-intensive services like AI.

See the current Appendix S here.  See a peer reviewed article on its goals here.  (PMC9719561.)  See an online article by its co-author Dr Frank, November 2025, here.

AMA has multiple bodies - like the Digital Medicine Payment Advisory Group - which can be confusing to outsiders.   I asked Chat GPT 5 to explain how it all works.   So far as I understand them, its feedback is correct.   Let me know if any errors and I will update.

###

Here is Chat GPT's understanding:

The AMA’s digital medicine coding structure appears to have evolved in two stages. 

First Stage

First, in 2017, AMA formed the Digital Medicine Payment Advisory Group (DMPAG) after identifying, in 2016, a gap in expertise around digital medicine. DMPAG was designed as a broad advisory group to AMA, drawing on physicians, industry experts, consultants, regulatory expertise, and other qualified health care professionals. Its work has focused on barriers to digital medicine adoption and on proposed solutions involving coding, payment, coverage, clinical data, use cases, and AI payment frameworks. (American Medical Association)

Insert Appendix S Here

Appendix S itself originated from this broader DMPAG ecosystem. AMA has stated that Richard Frank, MD, PhD, a member of the CPT Editorial Panel and former Siemens Healthineers chief medical officer, co-chaired the AI Working Group of the AMA-convened DMPAG, and that this working group “generated the content for Appendix S.” Appendix S is the CPT appendix that supplies the AI taxonomy—assistive, augmentative, autonomous—to describe the relationship between machine work and physician/QHP work in AI-enabled services. (American Medical Association)

Second Stage - DMCC

The second stage was the creation of the Digital Medicine Coding Committee (DMCC). AMA states that the CPT Editorial Panel created the DMCC in 2024. Its charge is not to own CPT policy independently, but to respond to requests from the CPT Editorial Panel and support the Panel on coding issues involving digital medicine and/or artificial intelligence by providing or obtaining subject-matter expertise. AMA is explicit that the CPT Editorial Panel retains sole authority to create, revise, and update CPT codes, descriptors, and applicable guidelines. (American Medical Association)

[DMCC may seat experts with expertise not found on the AMA Editorial Panel itself - BQ]

Thus, my understanding is that the DMCC comments on, advises on, and helps shape Appendix S revisions, but does not formally “own” Appendix S. The governing authority is the CPT Editorial Panel. The practical drafting process appears to be driven by AMA CPT staff and subject-matter leaders, especially Dr. Frank, with DMCC input, and then routed back through the CPT Editorial Panel process for review and approval.

Current relevant roles are as follows. Christopher Jagmin, MD, is the current chair of the CPT Editorial Panel; Barbara Levy, MD, is listed as vice chair. Richard Frank, MD, PhD, is currently listed as a member of the CPT Editorial Panel and is a leading architect of the Appendix S AI taxonomy. Mark Synovec, MD, is a pathologist, former chair of the CPT Editorial Panel, and currently serves as co-chair of the AMA Digital Medicine Coding Committee with Dr. Frank. (American Medical Association) (American Medical Association)

In practical terms, the quarterly Appendix S revisions appear to be a CPT Editorial Panel-controlled iterative process: draft revisions are circulated, the DMCC and stakeholders provide input, Dr. Frank/Synovec and AMA staff help refine the language, and the material goes back to the CPT Editorial Panel. The DMCC is therefore best understood as the expert coding advisory body, while the CPT Editorial Panel is the formal decision-making authority.

###

This may be a shareable link.