Monday, October 30, 2023

CCLA: Xifin Presents on Appeals Rates in Genomics at Calif Clin Lab Assoc

 At last week's California Clinical Laboratory Association meeting, there were some excellent presentations, including from Susan Van Meter (head of ACLA), and from three MAC Medical Directors (Drs. Mumuya and Radjadhyaksha of Noridian, Dr. Bien-Willner of Palmetto-MolDx).

There was also an excellent presentation by Lale White, CEO of XIFIN, on recent data in denials and appeals rates across different insurers. 

See the Xifin white paper listings here:

And the June 2023 Appeals/Denials white paper here:

It's authored by Diana Richard.  


AI Corner: Can ChatGPT Help with RFP's? A Lab Example

Below, I've asked Chat GPT to draft an RFP approach for revenue management services.   This is not an area I work in, and I'm only offering it as a "look and see" for what it might produce.  (I'm in no way vouching for the report, but it's still interesting.  I suspect one could find articles on Google with titles like, "Things to consider when soliciting revenue cycle management.")

Chat GPT 4's quirky Dalle-3 figures



We are running a medium-size clinical laboratory.  We derive about 75% of our revenue from conventional testing (clinical chemistry, etc) and about 25% from genomic testing (primarily cancer tests like EGFR, etc).    

We have diverse payers (Medicare, Medicare Advantage, commercial plans and Medicaid) and we are convinced our revenue capture and receivable days are not at the industry standard (too many denied claims and appeals and too many under paid out of network claims).  We need good help with revenue cycle management in the clinical laboratory space.  

How would we design a survey or request for proposals that is granular enough to assure we hire a high quality vender who really is impactful, and not waste money on weak support services?

ChatGPT 4

Designing a comprehensive Request for Proposals (RFP) for revenue cycle management in the clinical laboratory space requires a combination of clear objectives, detailed requirements, and structured evaluation criteria. Here's a step-by-step guide on how you can structure your RFP:

1. Introduction

  • Background: Briefly describe your laboratory, its size, specialties, and the revenue distribution between conventional and genomic testing.
  • Objective: Clearly mention that you are seeking a vendor for revenue cycle management with expertise in the clinical laboratory space.

2. Scope of Work

  • Revenue Cycle Analysis: Vendors should be able to provide an analysis of your current revenue cycle, identifying gaps and areas of improvement.
  • Claim Management: Emphasize the need for efficient handling of denied claims, appeals, and underpaid out-of-network claims.
  • Payer Relations: Describe your diverse payer mix and express the need for effective communication and negotiation with these payers.
  • Reporting and Metrics: Highlight the importance of regular reporting on key metrics like Days in Accounts Receivable (DAR), Collection Rate, Denial Rate, etc.

3. Detailed Requirements

  • Experience: Ask vendors to provide evidence of their experience in the clinical laboratory space. This could include case studies, references, or success stories.
  • Integration: Your chosen vendor should seamlessly integrate with your existing systems. Ask about their experience with integration, especially with any specific systems you use.
  • Training: Inquire about the training they provide to your staff during the transition phase.
  • Customer Support: Understand their support structure – is it 24/7? What's the average response time?

4. Evaluation Criteria

  • Relevant Experience: Give higher weightage to vendors with significant experience in the clinical laboratory space.
  • Proposed Solution: Evaluate the comprehensiveness and feasibility of the proposed solution.
  • Cost: While cost is essential, it shouldn't be the sole deciding factor. Consider the value the vendor brings in.

References: Reach out to the references provided and gauge the vendor's effectiveness and customer service.

5. Proposal Submission Guidelines

  • Proposal Structure: Define the structure you want the proposals to follow. This could include sections like Executive Summary, Proposed Solution, Case Studies, Cost Breakdown, etc.
  • Submission Deadline: Clearly mention the last date for proposal submission.
  • Contact Information: Provide a point of contact for vendors to reach out in case of queries.

6. Questions for Vendors

  • Can you provide a detailed breakdown of how you've improved revenue cycle metrics for other clinical laboratories?
  • How do you handle denied claims and appeals? What's your success rate?
  • How do you ensure compliance with regulations and payer-specific requirements?
  • Describe a scenario where you faced challenges with a client similar to us and how you overcame them.

7. Terms and Conditions

Clearly mention any contractual terms, payment terms, confidentiality agreements, etc.

8. Conclusion and Next Steps

Thank your vendors for their interest and specify the next steps in the selection process, such as interviews, presentations, or site visits.

Remember, the goal of this RFP is not just to find the most affordable vendor but to select a partner who understands your unique challenges and can provide tangible solutions to improve your revenue cycle management.


At what point should I get my attorney (external counsel) involved.  At RFP generation now?  Or later when we are actually contracting.   

Should he / she review the responses (for whether they are legally clear or legally vague?)


Involving your attorney (external counsel) at various stages of the RFP process can be beneficial for multiple reasons. Here's a breakdown of when and why you might want to involve them:

RFP Generation: It can be beneficial to involve your attorney during the initial stages of drafting the RFP. They can:

  • Ensure that the language used in the RFP is clear and unambiguous.
  • Help draft the "Terms and Conditions" section to protect your interests.
  • Ensure that there aren't any clauses that could unintentionally bind your lab legally.
  • Advise on any regulatory or legal considerations specific to your industry or region.

Reviewing Responses: While primarily the evaluation of responses will be based on technical and financial criteria, having an attorney review them can be advantageous:

  • They can identify any responses that contain legally vague or ambiguous statements.
  • They can provide insights on any potential legal risks associated with a vendor's proposal.
  • They can ensure that vendor responses align with the legal terms and conditions set out in the RFP.

Contracting Phase: This is the most critical stage where you'd want your attorney deeply involved. They will:

  • Draft or review the contract to ensure that all terms are clear and protect your laboratory's interests.
  • Negotiate terms with the selected vendor's legal team.
  • Ensure that all regulatory and compliance requirements are met in the contract.
  • Advise on potential risks and liabilities.

Post-Contractual Issues: If there are any disputes or issues that arise after the contract is signed, your attorney will be instrumental in providing guidance and representing your interests.

Recommendation: While it might seem like an additional cost, involving your attorney early on, especially during the RFP generation and response review phases, can preempt potential legal pitfalls. It's often more cost-effective to prevent issues at the outset than to resolve them later. Furthermore, having legally sound RFPs and contracts not only protects your laboratory but also instills confidence in potential vendors, as they see that you're thorough and professional in your approach.



Are there meaningfully different approaches to RCM for clinical chemistry vs genomics? If so, should I ask for specialized expertise in the two domains rather than just 1 generic domain?


Yes, there are distinct differences between revenue cycle management (RCM) for clinical chemistry and genomics, primarily driven by the nature of the tests, the technologies involved, reimbursement structures, and regulatory considerations. Here are some of the differences:

Nature of Tests:

  • Clinical Chemistry: Involves standard tests that have been in use for many years, such as blood tests, urine tests, and others. These tests are widely accepted and understood by both medical professionals and payers.
  • Genomics: Involves advanced genetic and molecular tests, which might be new, specialized, and not as widely accepted. These tests can be more complex, and their clinical utility might be questioned by some payers.

Reimbursement Structures:

  • Clinical Chemistry: Reimbursement codes and structures for these tests are well-established, with clear billing practices.
  • Genomics: Given the novelty and complexity of some genomic tests, reimbursement can be more challenging. Some tests may not have established reimbursement codes, or payers may be hesitant to cover them without clear evidence of their clinical utility.


  • Clinical Chemistry: These tests are generally less expensive.
  • Genomics: Genomic tests can be significantly more expensive due to the advanced technologies and expertise required.

Regulatory Considerations:

  • Clinical Chemistry: Regulations are well-established.
  • Genomics: The regulatory landscape for genomics is evolving, with new guidelines and considerations emerging as the field advances.

Payer Negotiations:

  • Clinical Chemistry: Given the widespread acceptance and understanding of these tests, negotiations with payers are often straightforward.
  • Genomics: Negotiations can be more challenging. Payers may require more evidence of a test's clinical utility or may not recognize the value of newer tests.

Given these differences, it's advisable to seek RCM vendors with specialized expertise in both domains. 

While there's overlap in the fundamental principles of RCM, the nuances and challenges of each domain can significantly impact revenue capture, claim denials, and payer negotiations.

Recommendation: In your RFP, do specify that you are looking for vendors with expertise in both clinical chemistry and genomics RCM. 

  • Ask them to provide evidence of their experience in each domain, such as case studies, success rates in claim approvals, and expertise in negotiating with payers for both types of tests. This will ensure that you partner with a vendor who understands the unique challenges and opportunities in both areas.

And Then There Were 12: Proliferation of AMA CPT GSP (Tumor) Codes

 From about 2014-2022, genomic sequencing procedures for tumors were so simply many of us had the codes memorized.  There were only 3.   81445 was for solid cancers 5-50 genes, 81450 was for heme cancers 5-50 genes, and 81455 was for any cancer, 51+ genes.   The two small codes were around $600 and the big code was around $3000.   The situation was complicated for records review and public health publications, since in 30 states, MolDx used 81479 (unlisted code, "other test") instead of 81455.  But for AMA CPT, just 3 codes.

This complexified in CY2023, when AMA created an additional code for "RNA only" studies.  (The original codes remained DNA, or DNA+RNA).   To me, this encouraged people to bill the 50 DNA genes on Monday and 50 RNA genes on Tuesday, but let that lie.  

Now for 2024, AMA has raised the total codes to 12.   They've added three more variants for cancer procedures if add-ons like MSI or TMB are used.   They added three variations of codes for liquid biopsy (previously, there was no call-out for LBx tests).   

CMS has proposed all 6 of these codes be crosswalked to 81445 (about $600) for 2024.  But CMS won't issue final prices til late November, so we'll be watching that.    None of these six new codes (for 2024) have a gene count mentioned.

UPDATE:  This crisis of underpricing covered in a subscription article at Genomeweb/360DX by Kelsy Ketchcum on November 8, here.  My updated blog Nov 9, here.

In the 2024 codebook AMA CPT publishes a table of the whole big menagerie. Click to enlarge.

Get the AMA 2024 CPT code book from Amazon, AMA, or other sources.  It's now 1283 pages.

click to enlarge

AI Corner: An Economist Writes a Book You Can Talk to in AI

Tyler Cowen is a creative and prolific economist at George Mason University.  He's written a new book, on the idea, "Greatest Economist of All Time" or GOAT.   (He decides there are several great ones, and no single winner).   

Find a blog about the book here.  And you can load it on a Kindle, but wait, there's more.  He's created a website that has "read" the book, and you can ask any range of complex questions about the book, "to the book," so to speak.  Website here:

I played with this for a while and found it really amazing.  

In healthcare, we worry about language models vaccuming up a lot of crap and nonsense from the internet and regurgitating it.   Here, for the GOAT book, the AI obviously has "learning English" as a large language model, but interrogates one book as a  "walled garden."   Similarly, AI could have one massive internal medicine text book (or several of them) as it's "walled garden" before long. Wolters Kluwer seems to be working on something of that type for its encyclopedic subscription website UpTodate.


Hi! Let’s see how AI can reinvent the book-reading experience. What would you like to know?

What are the two main similarities the author sees between Hayek and Adam Smith?

Based on the text provided, Tyler Cowen draws two key similarities between Hayek and Adam Smith.

Firstly, he notes that both economists inspire in their readers a sense of marvel and curiosity. This suggests they both possess a unique ability to provoke deep thinking and contemplation about the complexities of decentralised systems, such as economies or societies.

Secondly, both Hayek and Smith encourage readers not to assume every politician is of the same quality. They inspire readers to critically evaluate political constitutions, particularly in terms of how they attract and repel different individuals in their quest for public office.

In essence, both Hayek and Smith's works open readers’ eyes to the underlying mechanisms of societal structures, providing a fresh perspective on the conventional understanding of economics and governance. Even though the book doesn't elaborate more on the specific comparisons, we can infer these similarities based on the author's admiration for both economists.

What does the author think was the greatest weakness in the work of Keynes?

Tyler lends significant consideration to Keynes' shortcomings in the third chapter of the book. He asserts that one of the significant weaknesses in the work of Keynes lies in his theoretical economics - Keynes' aptitude as a microeconomist was notably weak, even when likened with leading economists of his time like Jacob Viner.

Further, the author also criticizes Keynes for not valuing intellectual rigor over political positioning. He believes that Keynes was not completely accurate and impartial in his readings and interpretations of his intellectual adversaries' approach; instead, he put his political persuasion ahead of scrupulous accuracy in reasoning.

Lastly, Tyler thinks Keynes had a limitation in his perspective - often viewing it from the British elite perspective, limiting his understanding of other world views. He articulates this when he discusses Keynes's attacks on greed, suggesting they emerged more from his exclusive Cantabridgean perspective and his upper-class lifestyle rather than a broader understanding of economic realities and challenges.
Remember, Tyler also acknowledges his personal bias, having been taught early on that Keynes' General Theory was flawed, possibly reflecting in his critique. But these are some of the key issues he sees with Keynes' work, according to 'GOAT'.

How did writing this survey work change Cowen's viewpoint of the great economists?

Tyler doesn't specifically articulate how writing this book has changed his viewpoint on the great economists. But we can infer from the text that the exercise of diving deeper into their lives and works likely helped him better appreciate their unique contributions and understand their respective philosophies.

It's like making a pot of stew, where each economist is a different ingredient, and as they simmer together, Tyler gets a more nuanced understanding of the flavor each one brings to the overall taste of economic thought. This exercise likely expanded his view, helping him see overlaps, contrasts, and conversations between economists that he might have missed on a less thorough pass.

On a lighter note, he also probably clocked a few more hours in the 'economist fan club' meeting room while working on the book!

Cowen notes he plans an audiobook auto-narrated in his own voice style.

Brief Blog: MEDPAC to Address "Software as a Service," Thurs Nov 2, 2023

MEDPAC is a longstanding official body that advises Congress on Medicare policy issues.  At their Nov 2, 2023, virtual meeting, they will discuss Medicare and "Software as a Service" (330-445 ET).  

>> Register for video access:

The short agenda is described as,

11/02/2023 . 3:30 pm - 4:45 pm ET:

  • Medicare coverage of and payment for software as a medical service: An overview
  • ISSUE: Software is becoming increasingly important and pervasive in healthcare.
  • KEY POINTS: Medicare covers and pays for certain types of medical software that receive approval or clearance by the Food and Drug Administration. The software that we discuss in this session generally stands apart from hardware (i.e., software that is used for one or more medical purposes to diagnose or treat an illness or injury without being part of the medical service).
  • ACTION: Commissioners will provide feedback on the material presented.
Medicare wrestes this topic, reluctant  to pay for licensing fees or per click fees, and generally viewing software as "built in" (such as inside a PET scanner purchase) or "overhead" (aka ignore).

A Few Perspectives
  • In current PFS rulemaking, under its RVU system, CMS declined to accept an AMA RUC valuation of Heartflow-type software (I believe the AMA RUC accepted an invoice-based $1500/service license fee).   
    • This will play out in the final CY2024 PFS rule, to be issued circia 11/01. 
    • Blog.
  • Other examples may include a new Category III code for software (and assistance) that helps rank-order cancer therapy choices for patients who have full multi-hundred gene sequences. 
    • Press release
    • CureMatch, "Patient-specific, assistive, rules-based algorithm for ranking pharmaco-oncologic treatment options based on the patient’s tumor-specific cancer marker information obtained from prior molecular pathology, immunohistochemical, or other pathology results which have been previously interpreted and reported separately."   
  • In another area, some half-dozen PLA codes have been created that include AI or "augmented intelligence" and these have been valued anywhere from around $700 to $5400.
    • Blog.  Scroll down for "slide-based imaging".
  • Note that, just like CMS writes NCD's only for FDA approved tests, CMS is conspicuously  talking about payment for FDA-approved software.

Wednesday, October 25, 2023

Cutting Edge Tests in Popular Media: Time, WSJ re: Path AI, MCED

It might not really grab our attention when we work in biotech or genomics all the time, but it's worth noting how the most cutting-edge testing, things that were beyond the horizon a few years ago, now appear in mainstream media articles.  

WSJ has a new update of the GRAIL Galleri multi cancer early detection test (MCED) by Alex Janin (October 25, 2023).   

TIME has a piece on the Artera AI test, used for prostate cancer, by Andre Esteva, CEO of Artera, with tie-ins to health equity and Cancer Moonshot.  (October 18, 2023).


AI corner:

Halloween Special, Chat GPT rewrites an Edgar Allen Poe story, here.

Tuesday, October 24, 2023

Open Access Articles at CAP TODAY Highlight Alzheimer's, Digital Pathology

CAP Today has two interesting deep dive articles.  One is on Alzheimer's disease, as FDA-cleared tau and amyloid tests gather real-world experience.  Fujirebio and Roche are featured.  The other is a deep dive article on the challenges of digital pathology.  Paige AI, Fuji, are featured venders and experienced pathologists at several institutions are profiled.





Mayo Clinic's experience with Alzheimer's disease assays in 2023 highlights significant advancements in diagnosis and biomarker utilization. Cerebrospinal fluid testing orders have surged since 2020, especially following the approvals of aducanumab and lecanemab. Dr. Joshua Bornhorst emphasized that FDA-approved Alzheimer's biomarkers demonstrate high precision, though challenges like preanalytical protocols and unresolved payment issues persist. 

Novel aspects include the recent incorporation of blood-based biomarkers in proposed diagnostic criteria by the National Institute on Aging and the Alzheimer's Association, expanding diagnostic possibilities. Biomarkers such as Aβ42, p-tau181, and NfL are all playing crucial roles in improving Alzheimer's diagnosis, with different biomarker profiles aiding in identifying disease stages. Mayo Clinic offers a range of CSF biomarker tests, including Roche's and Fujirebio's FDA-cleared assays, with ongoing refinements to enhance accuracy and diagnostic capabilities. 


  • SIDEBAR.  Other current ALZ news include Quanterix introducing a lab test based on J&J pTau217 assets (here, here), and Quest licensing a Diadem Alzheimer blood test based, more unusually, on p53 (herehere).  Diadem's AlzoSure has FDA breakthrough status from January 2022, but it sounds like it would be revalidated as a Quest LDT under the license. 
  • SIDEBAR.  I've heard since August that CMS would release some sort of guidance to MACs about Alzheimer tests (related to the newly released Alzheimer drugs), but I haven't heard of anything concrete.


The article discusses the current state and future prospects of digital pathology, highlighting the slow adoption of this technology in clinical practice in the United States. Despite the initial predictions of rapid adoption, traditional glass slides remain the primary method for diagnosing cancer in the U.S. Only a small percentage of hospitals have fully embraced digital pathology.

The articles also note some positive developments in the field, such as the FDA approval of the first product combining digital pathology with artificial intelligence (AI) for cancer diagnosis. Companies like Paige and Fujifilm are investing in digital pathology and AI technologies, signaling a growing interest in the clinical market.

  • TIP: FDA just updated the public on the rapid pace of its AI approvals of software - here.

Partnerships among digital pathology companies are increasing, facilitating interoperability and streamlined workflows between scanners, image management software, and AI tools. DeciBio Consulting estimates the global clinical digital pathology market to be worth around $270 million in 2023, with significant growth expected.

However, challenges remain, including the lack of clear reimbursement models for digital pathology and concerns about job displacement among pathologists. Some pathologists argue that certain aspects of diagnosis, especially those requiring nuanced interpretation, are irreplaceable by machines.

  • SIDEBAR:  Digital Radiology received a final push when CMS began to cut payments for non-digital radiology (noted in CAP Today).

Despite these challenges, experts believe that digital pathology, when used as a tool alongside AI, can enhance the efficiency and accuracy of pathology practices. Digital pathology is seen as the next frontier in digitization within the medical field, offering the potential to improve patient care while allowing pathologists to focus on more complex cases.


AI corner.

Summaries, AI assisted (GPT3.5).

Halloween Special, Chat GPT rewrites an Edgar Allen Poe story here.

Friday October 27, 2023: Last Date for CLFS Pricing Comments

Friday, October 27, 2023, is the last date for comments on the proposed prices for new lab tests for CY2024.

High on the agenda for comments will likely be the major new genomic sequencing procedure codes, which stakeholders expected to be priced in the $3000 range.   CMS offered only lowball pricing in the $600 range, Crosswalk to 81445.

I noticed today that AMA CPT 2024 codebooks are available overnight on Amazon.

8X017  GSP, with MSI

8X018  GSP, with CNA and MSI

8X019  GSP, with CNA, MSI, TMB, rearrangements

8X020, GSP, blood, DNA/RNA, CNV, rearrangements

8X021, GSP, blood, DNA, CNV, MSI

8X022, GSP, blood, DNA RNA, MSI, TMB, rearrangements

All: CMS proposes crosswalk to 81445, 5-50 tumor genes, $597.

Monday, October 23, 2023

Brief Blog: FDA releases roster of over 150 new AI devices and programs

On October 19, 2023 - the day after former FDA commissioner gave a talk including several AI topics - FDA released a new "accounting" of its authorizations for AI devices.   The news was covered in detail at STAT.

The FDA added 155 new devices, representing decision dates from August 2022 to July 2023.


The FDA updated its list of AI/ML-enabled medical devices, adding 171 new products, with 155 authorized between August 2022 and July 2023. The FDA emphasized the importance of AI/ML in healthcare, noting its potential to derive valuable insights from healthcare data. The list aims to inform the public about these devices and the FDA's work in this area.

The update revealed an expected 30+% increase in 2023 (still, a slower year-on-year growth than before.) Radiology dominated the field with 87% of devices in 2022 and 79% in 2023. The FDA plans to periodically update the webpage list. 

The agency has not yet authorized devices using generative AI or large language models. FDA notes that device summaries from manufacturer do not necessarily reflect the full range and breadth of data evaluated by the FDA in its product review.  


Multiple articles in medical journals and trade journals have noted that AI should be approved and adopted with great care.   In this STAT article, they suggest the transparency follows prior concerns were raised the FDA's approach to AI device regulation.  The FDA noted that many AI device summaries lack essential details about testing demographics and locations, which hinders understanding of biases and weaknesses. The FDA's challenge will lie in in keeping pace with rapidly evolving AI technology and ensuring public understanding of these devices' inner workings.

Friday, October 20, 2023

FDA Webinar on Big LDT Proposal: October 31, 1-2 Eastern (10-11 Pacific)

At the above, read about an FDA webinar on its "Regulation of LDTs" proposal.   The webinar will be one hour (at 10 pacific, 1 eastern) on Tuesday, October 31.   FDA will only answer questions it chooses to discuss, from questions in advance at a provided email by Monday, October 23.   The presentation will eventual be posted (info provided).

It looks like the link is an open Zoom link (no registration).


Webinar - Proposed Rule: Medical Devices; Laboratory Developed Tests

OCTOBER 31, 2023


On October 31, 2023, the U.S. Food and Drug Administration (FDA) will host a webinar to provide information on the proposed rule regarding Laboratory Developed Tests or LDTs.    

During the webinar, the FDA will:

  • Provide an overview of the rulemaking proposal to amend the FDA’s regulations to make explicit that in vitro diagnostic products (IVDs) are devices under the Federal Food, Drug, and Cosmetic Act including when the manufacturer of the IVD is a laboratory. 
  • Describe the proposed phaseout of FDA’s general enforcement discretion approach to LDTs. 
  • Host a Q&A session based on questions that have been submitted prior to the webinar at Questions will not be taken during the live webinar. All questions are due by October 23, 2023, to be considered for the discussion.

Illustration: Dalle 3 via Chat GPT 4.0 subscription app.

Thursday, October 19, 2023

CMS Gives NCD STI Coverage to New Abbott STI PCR Test

In a transmittal released October 19, 2023 (T12318/CR13390), CMS gives coverage to PLA code 0402U, under NCD 210.10, Screening for STI's.   There are a few interesting policy nuances.

What is Code 0402U?

0402U is the Abbot ALINITY (TM) STI assay, which was released by AMA on June 30, 2023, effective October 1, 2023, a few weeks ago.  The code text is:   

  • Infectious agent (sexually transmitted infection), Chlamydia trachomatis, Neisseria gonorrhoeae, Trichomonas vaginalis, Mycoplasma genitalium, 
  • multiplex amplified probe technique, 
  • vaginal, endocervical, or male urine, 
  • each pathogen reported as detected or not detected. 

CMS recently issued a crosswalk price for this code (under placeholder X072U) at $143, crosswalk to 87631, 3-5 respiratory viruses.  0402U has 4 (chlamydia, gonorrhoeae, trich, mycoplasma).  The code is gapfilled from its effective date (October 1) to the date of the new annual fee schedule (January 1), although this NCD release is titled the "April 2024" update.

What Happened?

CMS regulates coding and processing for most NCDs through national (uniform) instructioons.  In CR13390, code 0402U is added to the instructions for NCD 210.10.  These instructions apply to Part A (outpatient), Part B, and Medicare Advantage.

What is NCD 210.10?

Dating back to November 2012, this is one of a number of NCDs that are the Medicare implementation of preventive USPSTF coverage (endorsement).  

The text is fairly long - here.  It covers laboratory screening for chlamydia, gonorrhea, syphilis, and HBV, plus an STI counseling session, but there are bug-specific rules.  

The NCD covers the same screening conditions for chlamydia and gonorrhea. But it has different rules for syphilis, and different rules for hepatitis B.  The NCD doesn't mention mycloplasma.

How does CMS Handle Differing Rules per Bug?

I'm not sure.  The CR13390 contains a link to an online zip file with granular Excel claims processing instructions.  The link worked when I tried it on a second day. 

This zip file holds about a dozen Excels, each for one NCD, and the relevant one is titled:  210.10 Screening STI  and HIBC eag 100323 v2 pmbs 101323.xlsx.  Looks like it was signed-off Oct 13, 2023.

The rules are very complicated, spanning 37 row of Excel.  It was last updated to add PLA 0353U, a Cepheid chlamydia and gonnorhea test.   It is now updated to add 0402U, the Abbott test, to those parts of the NCD that govern chlamydia or gonnorhea.   As I read the Transmittal, the implementation date (required last date to program computer) is 4/1/2024, with an effective date (for claims of the date) 10/1/2023.  A few requests have faster implementation, but not STI Item 12.

For example (click to enlarge) see 0402U added in "red" to lines 24 for chlamydia and 25 for gonorrhea:

PLA Codes and NCDs

Generally, CMS doesn't add LDT tests to NCD's.  Most PLA codes are LDT tests.  However, PLA codes can also represent national IVD kits, as in this case for Abbott.


360DX: Stakeholders Likely Prefer VALID to Immediate FDA Regulation

On September 29, 2023, FDA dropped a regulatory change that intends to rapidly bring all LDTs under FDA regulation using the  current 510K/PMA models.   I wrote a white paper the following week, pointing out that many sources had misquoted the cost to industry as "$5.9B," which FDA published as the "annualized 20 year cost," the full cost being $5.9Bx20 or about $120B.  Much of that, $40B, would be due in the first several years.  In addition, $40B of regulatory cost would be 40,000 person-years (at $1M/person-year) and 400,000 person years (at $100K/person-year).   FDA buried this in table 35 on page 97 of online reference at footnote 34.  

I've read the FDA budget (for the whole huge agency) is about $8B in 2023, including user fees (a bit less than half).   The IVD proposal would add several billion dollars to FDA labor/work/cost in just a couple years, according to the FDA's own projections.   While some of that could be "third party" reviewes, the question then becomes where do you have several billion dollars' worth of IVD-trained regulatory professionals? 

And why regulate LDT's newly in 2023/2024, based wholly on what tests were on the market on the morning of May 28, 1976?   Sounds nutty on its face.

Adam Bonislawski's Article

I can't quote in detail from an excellent subscription article by Adam Bonislawski at 360Dx, but I am eager to point readers to it.   In a nutshell, most of the stakholders that Bonislawski interviewed (including AdvaMed) see the VALID act as preferable.  VALID Act is an A-to-Z reboot of FDA's diagnostic laws.

The article focuses on the current IVD industry, such as AdvaMed, Abbott, etc.  

Is there a middle way?

FDA proposed to rapidly dump the whole world of USA LDTs into itself in a couple years, which I rated as simply impossible in my white paper.  If that's right, then whatever happens, will be very different that the FDA's September 29 proposal.   (As written, the FDA Sept 29 proposal is like a design for a plane that has 10,000 pounds of weight and 1,000 pounds of engine lift.  Whatever else happens, you can conclude safely that it won't fly.  It doesn't matter whether you like the plane or not.)  

The VALID Act is a much-worked-over wholesale revision of FDA law.  While there are many short descriptions on the web, I find it hard to read because of the way it inserts and deletes passages of FDA law.

In my white paper, I wondered if there is a middle way.   FDA raises two key concerns, LDT accuracy and LDT marketing claims.   IF we would requrire "the equivalent of" current New York State rules for LDTs  might solve the first problem.   As to the marketing claims problem, fpr tjat surely there would be a some creative way to solve that short of a$40 billion dollars and 100,000 man-hour solution.

Wednesday, October 18, 2023

Scott Gottlieb Keynote: Why Cell Therapies are So Important (& other topics)


On October 18, 2023, Dr Scott Gottlieb, prior head of the FDA, provided a keynote address for the 13th Annual "Value Based Cancer Care" conference in New York City.  

In the first half hour, Dr. Gottlieb highlighted his upcoming book on the history and future of cell therapies.  Next, an interview by John O'Brien of the National Pharmaceutical Council explored the challenges of reimbursement innovation, regulatory decision-making, and regulation of AI applications.  These topics all emphasized the need for FDA adaptability in an evolving healthcare landscape.


I asked Chat GPT3.5 to write up an essay based my detailed conference notes today; I've edited it.


The 13th annual Value-Based Cancer Care conference kicked off with a powerful keynote address by Dr. Scott Gottlieb, the former Commissioner of the FDA and a healthcare policy expert. The hour included an interview by John O'Brien (National Pharmaceutical Council). 

Dr. Gottlieb gave us a glimpse into his upcoming book, on the history and promise of CAR-T cell therapies. He highlighted situations where the FDA provided "regulatory tailwinds" that encouraged the development of cell therapies.

The journey of cell therapy (CAR-T therapy) in oncology has been marked by decades of incremental progress punctuated by significant breakthroughs. There was constant interaction between our growing knowledge of the cell biology of cancer, and the science and technologies of emerging cell therapies.  Early progress was related to innovations in bone marrow transplantation, and the role of emerging role of T cells.

Dr. Gottlieb touched upon the transformative concept of engineering T cells to combat tumors, moving from passive harvesting and infusion to the idea of creating anti-tumor T cells. This led to the development of chimeric antigen receptor T cell (CAR-T) therapies. 

Gottleib also highlighted the pivotal role of Dr. Carl June at the University of Pennsylvania Dr. June's work in stimulating and replicating T cells outside the body was a key part of bringing CAR-T therapies to healthcare, and the Penn group was equally interested in innovations to industrialize and scale the procedure.  Gottlieb has never forgotten a meeting at Penn where June argued convincingly that the FDA (at the time) was hindering innovation and needed to be more adaptive.

One of the key takeaways from Dr. Gottlieb's talk was the importance of patience and long-term investment in medical research. He emphasized the critical role played specifically by physician-scientists who are committed to their work, often driven by direct patient care. He expressed concerns about the current process for NIH grants, which may fail to support early-stage risk-taking.

Translational work was another crucial aspect Dr. Gottlieb emphasized, highlighting the mundane yet vital challenges of scaling up manufacturing processes. He pointed to the need for a clear regulatory framework, which has been more readily established in oncology (with its depth of cancer cell biology) compared to other medical areas.

Dr. Gottlieb concluded his talk with optimism, noting the significant strides made in cancer research, particularly in genomics and minimal residual disease (MRD) monitoring. He highlighted the importance of thinking differently in regulatory decision-making, citing instances where unconventional approaches led to favorable outcomes.


Following Dr. Gottlieb's presentation, John O'Brien led an insightful interview that touched upon various aspects of healthcare policy, including reimbursement innovation, regulatory flexibility, and the impact of data on healthcare decision-making.

Competitive Bidding for Part B Drugs: Dr. Gottlieb expressed his belief that the entire reimbursement scheme for Part B drugs should be subject to competitive bidding. He argued that Medicare's role as a price taker, combined with its aim to control utilization, raises concerns. He cited the example of the controversial Alzheimer's drug, Aduhelm, where CMS imposed a Coverage with Evidence Development (CED) requirement, effectively denying coverage. He believed that CMS's clinical criteria were flawed and that this approach could have far-reaching consequences.

Challenges of Applying Clinical Criteria: Dr. Gottlieb highlighted a recurring issue where CMS has applied clinical criteria to major coverage decisions and often made incorrect assessments. He gave examples of past cases where CMS's clinical criteria led to suboptimal decisions, such as the use of implantable defibrillators and vascular-delivered valve replacements. He suggested that this approach might be extended to oncology, raising concerns in the field.

Implications of IRA (Innovative Reimbursement Arrangement): Dr. Gottlieb discussed IRA, a mechanism intended to create a synthetic loss of exclusivity for drugs after nine years. He expressed concerns about the impact on small molecular drugs and the potential disincentives for investment. He also noted that the growing elderly population, primarily under Medicare, might face the brunt of these changes. He predicted that reimbursement might be safer for large molecular drugs, and biosimilars could continue their slow price inroads.

Interchangeability and Biosimilars: The conversation touched on the challenges surrounding biosimilar interchangeability. Dr. Gottlieb pointed out the lack of a clear pathway for biosimilars to achieve interchangeability in the market. He suggested that more could be done in terms of interchangeability, emphasizing the importance of legislation and pharmacists' role in switching medications based on an AB rating.

VALID Framework and Regulatory Approaches: The topic of the FDA's recent proposal to regulated LDTs according to the 510(k) and PMA rules.  Dr. Gottlieb discussed the VALID legislative framework, supported last year by the FDA's leadership.  

The Impact of Data and Regulatory Framework for AI

Dr. Gottlieb and John O'Brien also explored the role of data and regulatory frameworks in the era of artificial intelligence (AI) in healthcare:

Walled Garden Approach to AI: Dr. Gottlieb discussed the FDA's confidence in AI applications that operate on "walled garden" data sets [my term- BQ]. These represent AI tools trained on specific and curated datasets with well-defined outcomes, such as recognizing labeled heart rhythms or differentiating benign from malignant conditions on validated MRI scans. FDA regulators are more comfortable with these applications as they involve clear data and outcomes.

Challenges of Language Models (LLMs): Dr. Gottlieb acknowledged the potential of Language Models (LLMs) for patient engagement and natural language processing but raised concerns about their regulatory complexities. He noted that regulators need guidelines on the datasets used to train LLMs, which currently lack standardization. While LLMs offer unique patient interactions, their regulatory oversight remains a challenge.

From VALID to AI: Interesting Projection.  Gottlieb indicated that if the novel VALID laboratory paradigm becomes law, one might use some similar constructs for AI. He explained that the FDA could regulate the firms developing AI tools rather than each individual device, making it important to ensure that the data going in and coming out of AI models are scrutinized, without the need to "rip apart" the technology.  He used the concrete example of an Apple Watch and its heart rhythm features, which the FDA approved based on its accuracy "outside the box," without trying to "take apart" and regulate the entire Apple watch and operating system.


Regarding AI, FDA released an update on its AI approvals on the next day,10/19/23, here. It was covered in detail at STAT as well.


Tuesday, October 17, 2023

Journal Club: Tsimberidou, Future of Tumor Boards? Nat Rev Clin Onc 10/2023

Journal Club

Tsimberidou, A.M., Kahle, M., Vo, H.H. et al.
Molecular tumour boards — current and future considerations for precision oncology.
Nat Rev Clin Oncol (2023). 

I first attended oncology tumor boards as a pathology resident at UCLA 30 years ago, in the early 90s, and continued as an attending at NYU and Northwestern, through 2001.   Early on, tumor boards helped deal with uncertainty: a patient might, or might not, get chemotherapy; might, or might not, depending, get radiation; might, or might not, have a pair of mets ablated.    Moving into the 2010s, oncology also got more complexity, as more and more mutations were associated with drugs (EGFR, ALK, ROS1).   

Today, more than ever, oncology is complexifying rapidly, and a few years from now there may be more proteomics, RNA-omics, methylomes, and other input to drug response.   How are tumor board systems evolving, and how should they evolve if agility and growth is a key parameter?   These are some of hte topics in TSIMERIDOU, Nat Rev Clin Oncol, released October 16, 2023.

See the citation and link above, and see a summary of the paper below.


[Tsimberidou via Chat GPT-3.5]

The field of oncology has seen significant advancements in our understanding of cancer biology and the development of targeted therapies in recent years. However, this progress has also led to increased complexity in the diagnosis and treatment of cancer. Tumors are recognized as highly heterogeneous, with different molecular profiles and responses to treatment. 

Tumor boards were designed to address this complexity by bringing together multidisciplinary teams of experts to review individual cases, discuss treatment options, and make personalized recommendations based on the patient's unique molecular profile and clinical history. These boards play a crucial role in guiding treatment decisions, especially in cases where (1) standard treatments may not be effective or where (2) multiple therapeutic options are available.

However, as the field of oncology continues to advance, tumor boards are facing new challenges. The explosion of molecular data, including genomic sequencing, has provided a wealth of information about tumor characteristics and potential targeted therapies. While this is a positive development, it also presents challenges in terms of interpreting and prioritizing this vast amount of data. Tumor boards must stay up-to-date with the latest research and treatment options to provide the best recommendations for patients.

Another challenge is the timely turnaround of molecular profiling results. While advanced technologies have sped up the process, there can still be delays in obtaining these critical results. Patients often require bridging therapies while waiting for the molecular profile, which can affect the timing and effectiveness of treatment. [*]

Access to matched targeted therapies is another issue. While many promising therapies are in development or available through clinical trials, not all patients have equal access to these treatments. Disparities in access to advanced diagnostics and therapies need to be addressed to ensure that all patients have a fair chance at receiving the most appropriate treatment.

To meet these new challenges, tumor boards are evolving. 

  • They are incorporating artificial intelligence (AI) and machine learning tools to assist in data interpretation and treatment recommendations. These tools can help analyze large datasets and identify potential treatment options more efficiently. 
  • However, it's essential to ensure that AI systems are based on high-quality data and validated through rigorous clinical studies.

Additionally, tumor boards are exploring ways to streamline the process from molecular profiling to treatment initiation. [**] Efforts are being made to reduce turnaround times and improve the availability of targeted therapies, especially through clinical trials. This includes centralizing tumor board functions in some cases to ensure consistent, high-quality care.

In conclusion, while the increasing complexity of oncology presents challenges for tumor boards, they remain a critical component of personalized cancer care. By embracing new technologies, staying informed about the latest research, and addressing disparities in access to care, tumor boards can continue to evolve and play a pivotal role in improving patient outcomes in the era of precision oncology.

Scheiter, 2022, review of tumor board experience in Germany. 
O'Dwyer, 2023, review of NCI MATCH and its lessons.


[*] NCCN guidelines for lung cancer provide one path for patients without genomics (awaiting genomics) and a different guideline path for tumors that do have genomics.

[**] I have seen some interesting presentations and papers from MGH on this topic. Eg Dagogo-Jack 2023,


Real World: What a Rapid-Care Walk-In Clinic & EHR Looks Like Today

In this blog, I share a personal experience - trying a modern walk-in clinic and noting the efficiency of their Electronic Health Record (EHR) systems.

Walk In Care

One significant shift in healthcare is the move towards making basic care more accessible. More and more we see walk-in clinics in pharmacy chains or large retailers. In California, the landscape is shifted. We're witnessing the rise of new, innovative walk-in clinic brands.  I'd guess there are five different ones within 1-2 miles of my house in mid-city Los Angeles, plus Cedars Sinai medical center runs a large free-standing walk-in center as well.    

My first visit to one of the new chains.   

For five weeks, I had been battling a night-time cough, untouched by dextromethorphan, thinking every week just had to be the last and I'd be over it.  So yes, I could have visited my internist, the five mile drive and the wait for an appointment made the prospect less appealing, and I assumed the result would be the classic and common "it's viral, no antibiotics, keep in touch."

My experience at the clinic ("Rapid Doc"; here) was very good. I walked over; check in took two minutes and within ten minutes, I was seeing a board-certified ER doctor in a Zen-like contemporary examination room. The physician was thorough, posing 15 to 20 questions in a well-practiced stream.  The exam matched the complaint and after discussing "it's just viral?" we opted for trial with doxycycline for a possible mycoplasma infection. (And I was 90% better in 48 hours). 

Pretty Cool EHR

Besides the efficent experience, I was struck by the EHR system in place.  I was handed a three-page printout of my EHR. I was pre-programmed to expect verbose, generic, and templated content. But I liked it: the record was detailed and tailored specifically to nuances of my visit and medical history. It was generated within a couple minutes of our 15- minute visit. For those curious, I've "fearlessly" uploaded the EHR PDF, revealing.... that I'm 5'7" and take hydrochlorthiazide for BP.

Friday, October 13, 2023

AMA Releases October PLA Code Applications; Rapid Comment Period

On October 13, 2023, AMA released the PLA code applications for the October (4th Quarter) comment cycle.  If you want to review and comment, you must do so very quickly, the comment period is only open for a few days.   Requesters must state a "material interest" in the code (e.g. working in the same field).  

The codes shown below will be edited by the PLA subcommittee over the next month and then voted on by a teleconference of the CPT panel as a whole.

Find the agenda  of 33 codes here:

I list very brief summaries here, for details, see the full PDF.   The final codes will be announced January 1, these codes will be effective April 1 and priced by CMS in Summer 2024.

  1. Coronary risk MAAA
  2. BluePrint breast cancer classification (luminal, etc)
  3. IntelliSep infection classifier (viral, fungal)
  4. HPV test, DELETE
  5. Precision CHD, coronary risk
  6. KidneyIntelX.DKD, this is the FDA version (FDA suffix added)
  7. NFL Light Chain, "normal or abnormal"
  8. Aventa Fusion, large gene fusion panel
  9. GenetWorx UTI DELETE
  10. ElecSys Phospho Tau & Amyloid, positive or negative
  11. Venient MCED (female multi cancer test)
  12. aiSLE DX, Lupus flare
  13. OncoReveal lung & colon, 22 genes
  14. Venient Cx, 39 analytes, anemia
  15. UNITY Carrier Screen, 5 genes (eg CFTR)
  16. aiSLE Disease Activity, lupus
  17. Venient Cx Cardiovascular, multi analytes
  18. Venient Cx Pancreative, multi analytes
  19. Venient Thyroid, multi analytes
  20. Venient Parathyroid, multi analytes
  21. "", hepatic
  22. "", renal
  23. "", hydroelectolytic
  24. "", Prostate
  25. "", musculoskeletal
  26. "", Compr female
  27. "", Compr male
  28. "", breast
  29. "", colon
  30. "", liver
  31. "", ovarian
  32. "", prostate
  33. "", MCED (male)
Neuro Note

This fall, Gapfill was completed on the Quest amyloid plasma test, 0346U, $96.  Here, there are two new neuro tests (7 & 10) for 2024.

CPT Agenda February 2024 San Diego

The Category I lab tests for February 2024 CPT (San Diego) will be posted around November 9, 2023, for a short comment period.  Registration opens about the same time.

The National PLA Coding Center

Thursday, October 12, 2023

CMS Releases National Part B Payment Files for CY2022; Lab Data

On October 12, 2023, CMS released Part B payment files for CY2022, in two sets.  

  • One is "national" by code series (e.g. lab, radiology, etc).
  • The other set is "carrier" which is basically by state... but not quite. 
    • Each geographic Excel is a MAC contract subdistrict, usually a state.
    • It's hard to find up to date tables that map a state to a CMS contract code number, so with these files, you're guessing which state is file #01112, which state is #03302, etc.

Although a little clumsy to use (because they are either split up by small code series or else by state), the new data demonstrates that CMS finally stopped its $300M a year of "bleeding" due to 81408 fraud.  Separate blog - here.

UPDATE:  There's also a cloud database that (after a learning curve) would usually be a better resource than the Xcel files, and it's also already updated for CY2022 claims data.  Thanks to John Leppard at Washington Analysis for a heads-up to this public link at CMS:

The above is not the most complete database; on a longer delay, CMS released a database with every CPT code for every lab and doctor - but that's about 18 months after the end of a calendar year - here.

All Lab Codes

For 80,000 series codes, plus --U and --M codes, I sorted for only rows with the modifier "total."  

This gives 1668 rows of lab services, and allowed charges (including path codes and COVID codes) of $8.4B, roughly in line with recent years.  (2021 had been about $9B with about $1B in COVID and $0.5B in fraud.)   

2022 COVID was $500M for U0003 and $155M for U0004 and $196M for U0005; spreadsheet MISC. They bring the Lab total to about $8.4B, otherwise $7.5B.

PLA: U and M Codes

U and M codes run up to 0354U, totalling 136 lines (the spreadsheet has no entry for codes with 0 CMS payment, more than half of them in this set.)  

Total spending for these codes was $341M allowable, but it was highly concentrated, with 75% of spending for 136 lines, concentrated in the top 8 lines.  

FDA Tests Score Big Dollars: While I suspect very few of all PLA tests are FDA approved, FDA approved PLA tests garnered 4 of the 5 top tests, and the top 5 tests had 60% of all the money.

Tier 1 CPT Genomic Codes

In the Tier 1 "Genomic" CPT codes - 811, 812, 813, 814, 815xx - spending was $1.5B, shown in  275 lines.   

Again, 75% of the spending was in the top 8 codes.  


One-third of all spending was 81479, unlisted code, racking up $476M.   81479 is also highly concentrated with 975 of spending in 10 codes (90% in 5 codes).  93% of 81479 spending is in MolDx states, 7% in Novitas states.  See more about 81479 below.

After 81479, Cologuard trailed at $273M, and then dollars plummeted again to just $95M for Oncotype and $75M for BRCA. 

Leaving aside Unlisted Code, 5 of the top 7 genomic codes were MAAA codes.   


Compare to my October 2022 report on 2021 spending here.


2022: The Year CMS Finally Stopped Hemorrhaging 81408 Fraud

As documented in trade journal articles, this blog (since September 2020, when I saw 2019 data), and a 2023 OIG report, Medicare lost a billion dollars in daffy claims for CPT code 81408 (rare long sequence genes for orphan diseases, including pediatric).   Add at least half that number again, for adjacent codes used in ridiculous ways (81406, 81407).  

Did I say ridiculous?  Most labs billing 81408 billed every possible Tier 2 (rare) code to the maximum, quarter after quarter.  Imagine a 90 year old nursing home patient, and a claim for 14 rare genes full sequence.   Now imagine thousands of such claims pouring in from a lab that just popped up and nobody's ever heard of, every 90 year old needing the same 14 rare genes.  As my college physical chemistry professor said, it's about as unlikely as all the air molecules in this auditorium rushing to the same corner at once. 

The OIG report last summer said, that CMS MACs assured it that the problem had been solved by 2022.  I was, well, uncertain, since the problem was well-known since public DOJ reports in mid 2019.

But in fact, 81408 payments did plummet from a colossal $283M in 2021, to only $372K in 2022
Finally.   The cutoff date should damn well have been around October 2019, not December 2021, saving the $480M lost in 2020 and 2021, but at least the cut off date finally arrived.  That's still more than the natural rate of $9000/year (in 2017), but a lot better than the $900M spent overall.  Throughout the saga, payments were almost exclusively through the Novitas and FCSO MACs.

81408, Part B, per year

June 2023

Where's it come from?

On October 12, 2023, CMS released Part B payment files for CY2022. There are two sets, one is national Part B payments by CPT code category (like "pathology.")  The other is files on a state by state basis.


Tuesday, October 10, 2023

A New White Paper on FDA LDT Proposal (17pp)

This blog releases a new white paper on FDA's proposal to regulate laboratory-developed tests.   The proposal was released September 29, 2023, and is in open comment until December 4.

My belief that this analysis is worthwhile was reinforced today when I saw that the usually-accurate Dark Report badly misquoted the FDA's proposed regulatory cost to the lab industry (as "$5B in collective new costs to industry" instead of well over $80B, with $50B up front) -- due to misdirection in the way the FDA data was written (for details, read on.)

I initially set up this blog with the first chapter of the White Paper.  I've replaced that with a  400-word summary of my 4000-word white paper, written by Chat GPT 3.5 (w/ minor edits).

The provided white paper by Bruce Quinn discusses the FDA's proposal to regulate laboratory-developed tests (LDTs) in the United States. It presents several key features and messages:

Overview of the FDA Proposal: The FDA introduced a comprehensive plan to regulate LDTs on September 29, 2023. The FDA proposal aims to bring LDTs under the same regulatory framework as other medical devices, including registration, reporting, and labeling requirements.

Enormous Costs: The paper emphasizes the staggering costs associated with the FDA's plan, estimating it to be around $50 billion over the first few years. This financial burden on the industry is seen as practically unattainable and far exceeding the annual profits of major diagnostic companies.

Timeline Challenges: The FDA's proposed timeline for implementation is considered unrealistic. The transition from a relatively unregulated LDT environment to full compliance with FDA regulations, including submission of 510(k) and PMA applications, within a few years is evaluated as logistically impossible.

Health Benefits: The FDA claims that its proposal will result in significant health benefits, primarily by improving the accuracy of diagnoses and reducing errors associated with LDTs. However, the paper questions the validity of these claims and suggests that they are speculative.

Legal Considerations: The paper highlights the legal controversy surrounding the FDA's authority to regulate LDTs. While the FDA asserts its legal authority, it acknowledges that this is a disputed claim, as demonstrated by the multiplex legal arguments presented in the paper.

Alternative Solutions: Instead of implementing a sweeping regulatory overhaul, the paper suggests alternative solutions achieving the same goals at far less cost.  These include:

If you want to ensure the "level playing field" between labs and IVD manufacturers, and prioritize that:

1) You have the VALID Act, although I hope they can find a somewhat simpler version that might still accomplish 95% of the goals.

If you want to prioritize solving the FDA's two stated problems, LDT accuracy and LDT claims:

1)  Address accuracy by requiring reviews of LDTs at roughly the current "New York State" level, and

2) Address concerns about misleading healthcare claims through far less costly means than the FDA's proposed tens of billions of dollars of regulations.

Challenges Ahead: The paper concludes that the FDA's proposal, as outlined, faces significant challenges, including industry opposition, legal disputes, and logistical hurdles. It questions the feasibility of executing the plan in its current form.

In summary, the white paper highlights the immense financial burden and practical challenges associated with the FDA's proposal to regulate LDTs. It questions the speculative health benefits cited by the FDA and suggests that more targeted and cost-effective solutions might be more appropriate. Additionally, it underscores the legal controversy surrounding the FDA's authority in this matter. Overall, the paper presents a critical analysis of the FDA's plan, raising doubts about its feasibility and potential impact.


AI Corner / Humor.  

As a humorous footnote, I provide an AI-generated review of my white paper here.

Webinar Oct 17: Concert Genetics and State Biomarker Legislation

Legislation requiring insurance coverage of biomarkers has passed in a number of states, and is pending in others.

See an October 17, 2023, webinar (11 PT 2 ET) sponsored by Concert Genetics on the status of state biomarker laws.  They've lined up an impressive group of guest speakers.

Find out more here:

Partial text:

A wave of legislation has passed in statehouses across the country requiring insurance coverage of biomarker testing. These bills have wide-ranging implications that are unfolding in real-time as they are enacted into law. They offer both the promise of expanded access to clinically valuable testing for patients as well as challenges and uncertainties related to implementation, compliance, and their ultimate impact on patient care.

In this webinar, panelists from Tempus, Foley Hoag, Blue Cross Blue Shield Association (BCBSA), and Blue Cross and Blue Shield of Louisiana (BCBSLA) will discuss efforts to provide patients with affordable, evidence-based, clinically appropriate testing in the context of this emerging legislation.

Monday, October 9, 2023

New Study of Payor Policies Identifies Wide Disparities in PGx Coverage

In August 2023, the Next Generation Dx conference in Washington had a number of exceptionally worthwhile talks. Two of them will be highlighted here at Discoveries in Health Policy. A future blog will discuss the presentation of Lee Fliescher MD, prior chief medical officer at CMS, who provided insights into TCET, CED, and other CMS topics.

Today, I'd like to highlight one of several excellent talks on pharmacogenomics. This talk, by Annette Taylor MS, PhD of Labcorp, featured research by the STRIPE Collaborative Community’s Study Design Task Force [1], to appear soon as Patel JN et al. Her talk reviewed several major recent papers in PGx, the landmark MolDX PGx policy in 2020 for which she played an advisory role, and similar Medicare policies at Novitas/FCSO, and she reviewed selected early data from Patel.

Patel et al. evaluated 12 payor or LBM policies for their PGx coverage. As shown in the bar chart below, coverage for the 63 strong evidence drug-gene pairs assessed ranged from 2 (Cigna) to 63 (MolDX), suggesting wide disparities currently in the coverage process (Patel et al., submitted 2023).  See a select bar chart below from their data.  I'll cover the range of research findings in more detail when the paper appears.  

  • AMA just published a white paper study of the diverse range of coverage for the AMA CPT digital medicine codes.   My blog here.
  • Translational Genomics, a major supplier of PGx software, just shut down after difficult interactions with FDA.  My blog here.

Click to Enlarge

AI Corner

I asked ChatGPT4 to read and discuss a JPEG of the figure above.  This is the verbatim AI result.


This graphic represents the number of drug-gene pairs covered per policy, with an emphasis on the wide variability across different entities. The entities appear to be health insurance or healthcare-related organizations. The vertical bars represent the number of drug-gene pairs covered by each entity's policy. The color segments of each bar correspond to different criteria for the drug-gene pair inclusion:

Green: Drug-gene pairs covered under "CPIC guidelines only."
Orange: Drug-gene pairs that fall under "FDA table section 1 only."
Blue: Drug-gene pairs that are covered by both "CPIC guidelines and FDA table section 1."
Grey: Drug-gene pairs that have "No CPIC guidelines and are not in the FDA table section 1."

For example, the entity "Aetna" has 10 drug-gene pairs covered under the "FDA table section 1 only" category. Meanwhile, "CMS (MolDx)" covers the most drug-gene pairs overall, with a total of 63, distributed among the different categories. The bar for "Patel et al. To be published" seems to have a special note, possibly indicating a future publication that will cover 43 drug-gene pairs.

The primary takeaway from this graphic is the variability in drug-gene pair coverage policies among different entities and the criteria they use for inclusion.