Thursday, January 23, 2020

Medicare Trivia: As Jurisdiction E Re-bids, Can It Go to a Non-MolDx MAC?

This blog is probably near a peak level of Medicare trivia, but it's interesting from the perspective of the MolDx policy system for lab tests, and the heavy concentration of Medicare's mopath spending in California.

It looks like none of the non-MolDx MACs could successfully take on Jurisdiction E, which is being re-bid right now, and take it out of the MolDx policy system.  Reasoning follows.

Jurisdiction E (California, Hawaii, Nevada) Is Up For Re-Bid

Jurisdiction E, one of the largest MACs, especially from the perspective of molecular testing, is up for re-bid.

The federal Request for Proposals (RFP) for Jurisdiction E posted on August 20, 2019, with responses due November 4, 2019.   See website here, solicitation number  75FCMC19R0023.  The anticipated award date for the new contract was posted as August 4, 2020.
(Click through the documents at the bottom of that webpage, 5 at a time, to see the most interesting document, the 150 page statement of work.  On my PC, I had to rename as a PDF-type file to open it.)  
See my cloud copy of the statement of work here.

Pp. 47-50 describe the medical director duties; there should be 3 FTE medical directors for this MAC.

CMS Caps Contractor at 26% of Workload; Contractor Group at 40%

In 2017/2018, the NGS MAC protested the limit on total share of MAC workload that can be held by any one MAC.   It lost.  Here.  According to the case, CMS caps workload of any one corporate entity to 26% of national MAC workload, and a group of companies shall not hold more than 40% of MAC workload.  My understanding is that Palmetto + CGS are a group of companies; and Novitas + FCSO are a group of companies.

Do The Math

Using a table included in the GAO case, I've rearranged the contractors by group.

There are three non-MolDx contractors, NGS MAC, FCSO, and Novitas.  Here's my math.  FCSO can win California, which has 9.1% of CMS workload, but then the FCSO+Novitas group will have 32%+9% = 41% of CMS workload, more than 40%.  So FCSO can't win California, I don't think.

Novitas can't win California, because it already has 25% of CMS workload, basically at its 26% limit.

NGS MAC has 20% of CMS workload, so it can't add 9.1% for California, which would give it 29%, over the 26% limit.  However, NGS MAC could spin off into two shell companies under one parent company, and then either NGS MAC company (with 7.8% and 12% respectively) could add the 9% JE MAC. 


So, taking this as a sort of puzzle, the only way I see that a non-MolDx MAC could win JE is by NGS MAC splitting into two companies (possibly under a parent company, something discussed in the GAO legal case cited) in which case it could win JE.  If that happen, JE might become a non-MolDx MAC.   Check back in August 2020 for the result.


See the MolDx federal "Statement of Work" here.

Wednesday, January 22, 2020

Very Brief Blog: Quest Acquires the Blueprint Genetics Laboratory

On January 22, 2020, multiple news outlets reporting that Quest has acquired Blueprint Genetics, a Helsinki-based genetics laboratory with a US center in Seattle.

Financial terms were not disclosed.

See stores at MedCityNews here, at MedTech Dive here.  Quest's press release here.

MedCityNews writes,
Secaucus, New Jersey-based Quest Diagnostics said Wednesday that it would acquire Helsinki-based Blueprint Genetics. Financial terms of the all-cash deal were not disclosed. Blueprint – not to be confused with Blueprint Medicines, a U.S.-based drugmaker (...) – makes tests for 200 genetic panels and 3,900 single genes, across 14 medical specialties. It has been expanding its U.S. presence as well, with a new hub in Seattle. It focuses on gene variant interpretation using next-generation sequencing and bioinformatics technology.
See Blueprint Genetics website here.  See Blueprint newsroom for information on their recent new tests and initiatives, here.

In October 2019, Genomeweb ran a subscription article on Blueprint's new lab in Seattle (here).  The firm had raised €23M from investors and had €15M in global CY2018 revenue, expecting that to double in CY2019.

Medicare Part B and Quest's 2017 Genomics Payments (Spreadsheet)

Medicare paid claims data for Quest and other providers for CY2017 is available at CMS (here). 

I've put an Excel spreadsheet in the cloud for 2017 Part B Quest paid claims for CPT codes 812xx 813xx 814xx (most of genomics) here.   

Part B 2017 payments to Quest tallied about $2.8M dollars allowed.  Total CMS genomic payments in 2017 were circa $500M, so Quest garnered about 0.5%.

Quest's billings are somewhat complicated; for example, they billed Medicare for 81206 (BCR ABL) from about 15 different locations. 

The two largest volume codes were 81206 (BCR ABL minor breakpoint), 2375 services for $522,000; and 81207 (BCR ABL major breakpoint), 1393 services for $220,000.   Quest also garnered around $500,000 for different codes under HLA genotyping.  BCR-ABL and HLA were about half of Part B payments to Quest in the genomics code series.

Quest also garnered about $440,00 for Level II codes (tiers 2,3,4), and almost nothing for unlisted code 81479.

(sample screenshot of spreadsheet)

Very Brief Blog: CMS To Cover Acupuncture as Non-Opioid Therapy for Back Pain

Last January, CMS unveiled a plan to review acupuncture again (it is a service that has been non-covered by NCD for decades).   See original blog here.

On January 21, 2020, CMS released a final decision covering acupuncture.  The final decision provides coverage to anyone with " chronic back pain" (not further defined, but not due to "systematic cause" like metastatic disease).   12 visits in 90 days are covered.   An additional 8 visits can be covered if the therapy is working.  A capped number of 20 visits per year are covered.

Two notable points - first, the decision was late; it was due in October.  Second, the draft decision in July had placed acupuncture under limited coverage with evidence development (coverage only in trials).

  • Read the NCD in full, here.

Opioid Policy

The NCD specifically cites Trump administration policy on the opioid crisis and the need for alternative therapies.  CMS also cites coverage of acupuncture to one degree or another by the VA and some private insurers.

CMS writes, "We believe that in light of the relative safety of the procedure and the grave consequences of the opioid crisis in the United States,  there is sufficient rationale to provide this nonpharmacologic treatment to appropriate beneficiaries with chronic low back pain."

Uses of acupuncture for other conditions than lower back pain remain non-covered.

Coding and Payment

There are existing AMA CPT codes for acupuncture - 97810, 97811 for first and second 15 minutes.  They are priced at around $28 and $38 respectively (higher in cities).  Up til now, those two codes been categorized as nationally non-payable (non-processable) by CMS. 

Historically, it often takes over six months for CMS to issue claims processing instructions on new NCDs.  I would expect CMS will remove these codes from "rejection" status faster in this case.

CMS could elect to create and use other codes (G-codes) for the NCD.   I recall at 2019 recent AMA CPT meeting there were some extensive changes, additions, or Category III codes under debate in this area discussed.  I didn't track if they passed or not. 


Individuals with only a state healthcare acupuncture credential can't enroll for payment from Medicare.  Acupuncture has to be administered by (or under direct supervision of) someone who can enroll in Medicare, e.g. a doctor or nurse practitioner.

Normally CMS has pretty vague rules about who can perform services incident-to a physician (e.g. "appropriate" personnel.)   Here, the incident-to person must be a state licensed acupuncturist.  (See the NCD for full details and see the Q&A public comments section). 

A chiropractor under Medicare law can only be paid for spine manipulation, but he/she could be paid for acupuncture if he/she was also (in parallel) a licensed acupuncturist and under the supervision of a physician.

Coverage with Evidence Development

The draft NCD last summer proposed coverage with evidence development; as far as I saw, there is no mention of the proposal in the final decision, it's just gone and replaced.

Tuesday, January 21, 2020

Very Brief Blog: CMS Somewhat Cleaned Up NCCI Genomic Edits for CY2020

Just over a year ago, in November 2018, CMS released a few  sometimes-bizarre new rules for coding of panel testing and especially genetic test panels (see original blog and links here and here).  This led to ongoing letters and meetings between ACLA and others for many months in CY2019.  See e.g. a subscription article on the topic in Dark Report in April 2019 here.

Without much fanfare, CMS released some revisions to its edits in November 2019 for Calendar 2020.   The revisions are in Section F (Molecular Pathology) of Chapter 10 of the National Correct Coding Edits Policy Manual.   (Download via link "NCCI Policy Manual Effective January 1, 2020" at this website here.)

ACLA has already publicly complained, in December 2019, about some of the NCCI language.  See online here.

You should read the ACLA letter for yourself.    I have a few comments below that may be a "glass half full" view of some of the edits, at least in Sections 7,8,9.   ACLA thanks CMS for some of the edits I'll be highlighting next, while complaining that elsewhere in the NCCI manuals there remains adverse language.  In particular, there are arguments that NCCI language would cause mass overuse of unlisted codes.

Mostly Favorable Small Revisions to Genetic Edits

Whereas Section 7 originally said that Tier 1 or Tier 2 MoPath codes "shall not" be reported with a genomic sequencing procedure (e.g. panel), it now says that those codes "should not, in general, be reported" with a GSP.  If procedures are reported together, they should both be medically reasonable and necessary.

Whereas Section 8 originally said that NGS procedures must be reported by a single CPT code (e.g. GSP) or 81479 (unlisted procedure), the text now says that if no single code accurately describes the NGS procedure, it MAY be reporting with 81479 or MAY be reported with multiple individual CPT codes "when medically reasonable and necessary."

Whereas Section 9 blocked some pairs of MoPath codes from being reported together, it now says that code-to-code edits are codes that "should not, IN GENERAL, be reported together."   Again, it is restated for the third time that if codes are reported together they must both be reasonable and necessary.

On the whole, improved.

NCCI labels changes in red, so it's easier to see additions than deletions. I didn't note any other changes that seemed significant, but you should check the manual yourself for your own area of interest. 

I've put a direct cloud link to a January 2020 version of the Pathology policy chapter in the cloud here.  However, the CMS webpage linked above should be checked for any possible updates.

Friday, January 10, 2020

Very Brief Blog: Coalition Files Citizen Petition, Urges FDA Stop Actions Against LDT PGx Labs

According to an article on January 10, 2019, in Genomeweb, a "Coalition to Preserve Access to Pharmacogenomics" has filed a federal Citizen Petition "asking the US Food and Drug Administration to stop its oversight attempts to limit these services.

  • See Genomeweb article here.
  • See Citizen Petition here.
    • The petition is signed by Hyman Phelps attorneys Gibbs and Javitt.
    • Dated January 9, 2020.

Nerd Note Deluxe!

The FDA published a draft 20pp guidance on how it handles Citizen's Petitions in October, 2018  here.  (Although primarily in regard to malicious submissions intended to delay competitors).  The corresponding Federal Register announcement was 83 FR 49935, October 3, 2019, here.  The agency received 21 comments (here, bottom).  For my money the most interesting was a 6 p comment from the Federal Trade Commission (here).

Thursday, January 9, 2020

Very Brief Blog: CareDx/Natera Court Case Cites MolDx Article

On January 8, 2020, Genomeweb ran an article about an ongoing legal dispute between CareDx and Natera regarding donor-DNA renal transplant tests - here.

I don't have any comment on the case, but the news article at Genomeweb includes a quotation that, "Medicare specifically rejected the very arguments CareDx advanced in its complaint when it finalized its Local Coverage Decision (LCD) for Natera's Prospera donor-derived cell-free DNA test."

How do you find that? 

  1. Go to Palmetto's "LCDs, NCDs, Coverage Articles" webpage - here.
  2. Click on "Future Effective LCDs."  Here.
  3. Look for L38041, MolDx: Prospera.   Here.
  4. Scroll to the bottom of the webpage of the online LCD text and see:
  5. A57821 - Response to Comments - Prospera.  Here.
  6. The CareDx comment is #4.
I've also put a cloud copy of the LCD here and a cloud copy of the article here.

You can also go to CMS Coverage Database and enter the document numbers in the small top-right box "Document ID" - here.

Tuesday, January 7, 2020

MolDx Posts Final LCDs for NGS Profiling of Solid Cancers and Myeloid Cancers

MolDx has finalized complex and important LCDs for comprehensive genomic profiling in cancer. Each of the two LCDs also has an accompanying coding & billing article, and a lengthy article which  is a response to public comments.

L38045 covers solid cancers, and L38047 covers myeloid cancers.  Expect other MACs in the MolDx policy system to roll out matching LCDs soon (WPS MAC, CGS MAC, Noridian MAC.)

One way to find these LCDs is to (first) go to the Palmetto MAC "LCDs, NCDs, Articles" page and (second) click on "future effective" LCDs.  Here.  Note, however, that after February 10, these become "effective" LCDs and will move to the "effective LCDs" link.

The Two LCDs - Each With Two Articles

When you open the two LCDs - L38045 and L38047 - near the bottom of each LCD is a link to its coding article and a link to its public comment Q&A article.   For L38045, see coding article A57831 and Q&A article A57830.  For L38047, these are found as A57387 and A57836.

You can find AMP's comments on the articles under the June 20 date at AMP's public policy webpage here.

MolDx doesn't use AMA CPT code 81455 (50 or more tumor genes) but wants large cancer tests coded as 81479 (molecular unlisted code).

You'll Also Want Article A57757

Note that you'll also want article A57757, which defines primary and recurrent tumors. Medicare's NGS National Coverage Decision 90.2 states that NGS tests can be provided only in advanced cancers and only one of the same test, per primary cancer.  MolDx defines a non-covered test as a second test on the same specimen on the same day, and a different primary cancer is a cancer that is "different enough to respond differently to therapy."  (To my eye, that suggests any tumor that stops responding to a prior effective therapy is different enough to escape the one-test-per-primary-cancer rule).   You may also want Article A54795 - which defines hotspot tests versus CGP or comprehensive genomic sequencing tests.  Find them using the document ID search box at CMS Coverage Database (here).

Grandfathered Tests Require TA's - and Fast

MolDx is also going to require that grandfathered NGS cancer tests, which might not have undergone detailed tech assessments in prior years, must have an approved TA on file as of February, 2020.  See article here.

Tech Assessment Documents

Some of the requirements for NGS tests are found not in LCDs, not in coding & billing articles, but rather in MolDx website technical assessment documents.  Entry point from the MolDx home page, here.  The TA assessments for NGS tests are exhaustive - MolDx doesn't let you by with just a CLIA certificate on file.

Cloud Copies of Documents Cited

I've put a Zip file with articles and LCDs cited in this blog, in the cloud - here.

Nerd Notes

L38045 states point blank that: "NGS is not a specific test but a sequencing methodology utilized to capture genomic information."   This is practically verbatim what stakeholders have been telling CMS regarding its NCD 90.2 - stakeholders like ACLA, CAP, AMP, 21st Century Coalition, NCCN, and others.  CMS seems to (??) view NGS as a "test" (sic), which is either "covered" or "not covered" in certain situations.  Thus, NGS is "not covered" in patients with cancer that is not advanced cancer, a nonsensical situation that means that even FDA approved HIV genotyping tests are "not covered" in these early cancer patients.   In this regard, the existing CMS NCD simply makes no sense at all.

The multi-level implementation of the LCD, with an LCD, an attached article, collateral articles for definitions, and then separate MolDx website documents like Tech Assessment spreadsheets with gene lists, means it would be very hard for a RAC or auditor to know if a given test was appropriate covered or not on any given day of service.  You'd need contemporary copies of all these documents, a real handful, for a particular date of service in the past.

Anthem Executive to Lead CMMI

Anthem executive Brad Smith has been appointed to head the Center for Innovation (CMMI) at CMS, announced January 6, 2020.

During the Trump administration, CMMI was headed by Adam Boehler until October 2019.   Boehler moved from CMS to head the U.S. International Development Finance Corporation (here).  Prior to CMS, Boehler had started Landmark Health, which aimed to assist Medicare Advantage and other plans in managing chronic disease better.

Long-time CMS executive Amy Bassano had been acting director of CMMI since last fall.

Interestingly,  Smith was CEO of Aspire Health (founded by former Senator Bill Frist), which provided home services to patients with complex and chronic illnesses, not too different from Boehler's background with Landmark.  Smith, a Harvard graduate, was a Rhodes scholar.  According to one news source, the head of CMMI oversees 600 employees.

Trade journal stories include:

  • Healthcare Dive here.
  • Modern Healthcare here.
  • The Tennessean here.
  • CMS press release here.

While CMMI is commonly described as having bipartisan support, in fact there were Hill hearings in 2016 that were highly confrontational and concerns that CMMI's authorities might exceed statutory limits.  For example, CMMI is allowed to set up "demonstration programs" by law (in the ACA), but there is no written restriction on the size, scope, duration, or content of the demo programs.   CMMI's statutory authorization is just a few sentences long and has been viewed as self-implementing.

Friday, January 3, 2020

Books and Movies List - 2019

Off topic from my regular business blog, my fifth annual book list, 2019 version.

In the cloud as a PDF here.

Internet version here.

Best wishes in 2020 !!

Bruce Quinn
Bruce Quinn Associates LLC

Thursday, January 2, 2020

CMS Updates PAMA Page for New Law: Reporting Year Is Delayed, But Year 4 at 15% Cut Is Not

CMS has updated its PAMA REGULATIONS page to account for new law passed in late December, section 105(a)(2) of the Further Consolidated Appropriations Act of 2020 (FCAA).  See PL 116-94 (HR 1865) online here, here.  It's Division N, Title 1, Subtitle A, Section 105.  The statutory language is clipped here.  It's cryptic since it inserts and deletes bits and pieces of phrasing into existing law not directly shown in PL116-94 (see SSA 1834A, here.)

FCAA does the following:
  • Leave the payment data period for the next three-year PAMA cycle the same, 1H2019.
    • Claims paid during 1H2019 for active codes in 1H2019 will be part of the next PAMA data collection.
  • Delay the data collection from 1Q2020 to 1Q2021.
  • Delay the implementation of new prices to January 2022.
CMS notes that another time cycle is not changed, in its view.  This time cycle provided 10% pay cuts (at the max) in Year 1,2,3 under PAMA, and 15% pay cuts (at the max) in Year 4,5,6.   Since 2020 is Year 3, pay cuts are limited to 10%, but since 2021 is still year 4, pay cuts will be 15%.

Whether lab tests get a 15% cut in CY2021 depends on whether the 2020 price is still 15% higher than the price median found based on 2017 data.

CMS updated the top part of its PAMA page, here.  See table (click to enlarge) clipped below.

click to enlarge
By my informal tally, comparing 2020 current prices to the median of 2017 reported private payer prices, about 206 CLFS CPT codes will have a 15% cut in CY2021 if CMS's update is correct.   Another group of about 390 will have a 10-15% cut and finally about 130 will have a 0-9% cut.

I've put my personal best-effort spreadsheet in the cloud here.  This Excel has two tabs; one sorted by percent cut 2021 (my prediction), and one sorted by CPT.

Recall there were some huge anomalies in PAMA reporting pricing; for example, 81341 TRB gene rearrangement had a CPT price of $68 in 2017 but a PAMA median price target of one penny.   81435 Lynch Panel had a CMS 2017 price of $802, but a PAMA median price target of $37.99.