Thursday, April 29, 2021

CMS Updates Some PAMA Lab Pricing Reporting Guidelines

Section 216 of the PAMA law of 2014 requires labs to report private payor pricing every 3 years to CMS, which CMS then uses to set a triennial Medicare lab fee schedule.  

The first cycle involved a claims period in 1H2016, reporting in 1Q2017, CMS publications in September 2017, and a 3-year fee schedule for 2018, 2019, 2020.   Due to some quirks and updates to PAMA law, the next three year cycle is atypical.  The claims period is still 1H2019, but data will be reported in 1Q2022 (not 1Q2020), for price publication in September 2022 and a fee schedule for 2023, 2024, 2025.

On April 20, 2021, CMS updated some documents on its website for PAMA lab pricing policy.   

  • The main webpage is here.
  • At least two documents have been updated:
  • Summary of Private-Payor Rate-Based CLFS, 10 page PDF, here.
  • FAQ, 14 page PDF, here.
While these two PDFs show new dates of April 20, 2021, there isn't any guide to new information.  I can't tell easily if they've merely updated dates (per legislative changes) or if they've updated definitions or instructions in a material way.  [Note, I found a 2017 FAQ online, but that version was much longer at 18 pages, so probably too different to benefit from a redline.]


The regulations upon which the guidelines and FAQs are based are found at 42 CFR 414.500ff, here.

By request of Congress, MedPAC is preparing a report on CMS and PAMA rules, with a public preview in April 2021 and final public report in June 2021.   CMS summer rulemaking might make some changes to PAMA rules which would take effect on January 1, 2022, in time for the 1Q2022 reporting period.

___

What Will Happen to Codes Newer than the 2019 Claims Period?

Several hundred lab codes (mostly PLA codes) will have been produced between the closure of the 2019 PAMA claims period and the 2023 fee schedule.  In the last PAMA pricing period, all lab tests that didn't have PAMA payor data (from claims in 1H2016) were crosswalked or gapfilled in August 2017 (about 100 codes) to produce prices for the 2018 fee schedule.

One reading of PAMA regulations at 42 CFR 414.508 is that new codes get crosswalked or gapfilled in 2018, 2019, 2020, 2021, but only those with payment data from 1H2019 will get repriced via PAMA in fall 2022 for CY2023.  Other codes (such as a code new in 2H2019 or 1H2020) will stay at their same price indefinitely, until they do get PAMA data (perhaps in 1H2024).  

Another reading of 414.508 is that codes without PAMA data get priced by the method in 414.507, which states that codes without PAMA data get crosswalked or gapfilled for the next 3-year schedule (e.g. 507 sends codes without new PAMA data right back to method 508, which is "CW/GF" pricing).  

I think the simplest reading of 414.508 is that codes crosswalked in 2020 or 2021 stay at that price, and that CMS responds to their lack for 2019 PAMA data as meaning they are set aside until the next PAMA cycle when they do have PAMA data.  But it's a bit of a brain teaser, semantically, as 414.507 and 414.508 point back and forth at each other like two mirrors facing each other.  

In addition, most codes are crosswalked, and a 2020 code crosswalked to a code WITH 2019 PAMA data, could be repriced in 2023 because its CW reference, the code with 2019 PAMA data, is being repriced for CY2023.*
___

Nerd note.  A person can redline two PDFs in a business version of Adobe Acrobat, but it's kind of expensive software.  If you google "redline 2 PDFs" you'll get some cloud websites that allow the redline action between two PDFs.  E.g. https://tools.pdfforge.org/compare-pdf   

___
* Let's say Code 01U is priced at $100 in 2018, and gets new PAMA data in 2019, repricing it at $80.  This repricing will occur in 2023, due to the extended PAMA schedule we are in.  

Code 02U is new in 2020, and crosswalked to 01U at $100.   I believe in 2023, even though 02U has "no data" from any 2019 survey, its price will fall to $80 because its referent crosswalk (01U) is changing.  In my experience, CMS has always viewed crosswalks as "living" crosswalks, subject to change if the crosswalk target code changes.  You can't necessarily read this lability from the wording of the 414.508 regulations, but it's the way CMS has behaved.

Wednesday, April 28, 2021

Read 3 CMS New Tech Reviews: (1) AI for Head CT (Pass), (2) AI for Pulm. Emb. (Proposed), (3) Pneumonia Pathogen Panel (Fail)

 As I noted in an earlier blog, on April 28, 2022, CMS released its initial assessment of 22 technologies proposed for FY2022 hospital inpatient new tech payments (here).

Two Technologies for Hospital Inpatient "AI"

Of note, one of the proposals today is for hospital inpatient use of artificial intelligence software in radiology.  This is the AIDOC BRIEFCASE FOR PE system by AIDOC.  

Last winter, CMS granted add on payments to a system for hospital inpatient use of artificial intelligence in stroke management.   This is the ContaCT software from VIZ.AI.

I've put both of these inpatient AI technologies in a zip file; link at bottom.   Both of these AI systems pre-read CT scans in queue and can highlight those that are 90% likely to be positive for first-in-line review.  This, in turn, should speed therapy for the stroke or for the pulmonary embolism.

CMS's Critique of Biofire Outcomes Data

In addition, and making a contrast, adjacent to the December 2020 "ContaCT decision" I noticed that CMS had published a negative analysis and negative decision for the Biofire Pneumonia Pathogen Panel.   

CMS criticized the Biofire panel has having inferred or speculated benefits rather than confirmed ones.   This is a frequent criticism of diagnostic tests, and for anyone planning diagnostics outcomes data or clinical utility arguments for CMS review, this example may be worth a read.  And interestingly CMS nixed extra payments for the Biofire pneumonia panel in 2H2020, at the height of the COVID epidemic, even knowing that a significant portion of the most complex cases likely had coinfections that would be hard to culture in patients on broad spectrum antibiotics.

Zip file of the three items, HERE.

See CMS announcement re a "AI CHALLENGE" contest, April 28, here.



CMS Releases Inpatient Proposals for FY2022; Includes AI Hospital Software as a New Tech Add-On

The evening of April 27, 2020, CMS released its proposed hospital rule changes for FY2022.

  • See early coverage at RevCycle here.  AHA here.
    • Reports 22 technologies applying for add-on payments; public comment requested.
      • See p. 287ff.
      • See discussion of "alternative new tech pathway," p.292ff.
      • See mention of Council on Technology and Innovation p. 298ff.
      • See recordings of the December New Tech applicants' meeting here.
      • Proposal to extend New Tech payments, p. 305ff, related to "Covid year" only
        • New Tech payments usually expire in a mechanical way in 2-3 years, depending on the product's date of FDA approval.
      • CMS Considers New Tech "AI" Technology
      • New applications begin with "AIDOC" artificial intelligence software for pulmonary embolism, p. 313ff.  CMS writes, "Briefcase for PE is the only FDA cleared technology that uses computer-aided triage and notification to rapidly detect PE and shorten time to notification of the radiologist."
      • Recall that CMS discussed difficulties with RVU pricing of AI technologies in December 2020 (here).  CR12071 p. 11 here.   New Tech payments are based on real or expected invoices which raises less issues than RVU pricing.  See discussion of the CMS philosophy to handling software claims and payments, p. 316, which refers also back to a 2021 discussion (cited to: 85 FR 58628 in contact of CONTACT device 58625ff).  CMS has also puzzled over subscription-based costs (cited to: 85 FR 58630).  Applicant asserts clinical outcomes will improve because notification of PE drops from 64 to 4 minutes.   The discussion refers to a "golden hour" for reducing mortality and LOS (p. 321). Overall, by p. 326-8, CMS raises a number of questions about AI technology in this context, for public comment.
      • Forbes discusses AI revolution in healthcare today, April 28, here.  Healthcare Dive discusses that AI devices are pre-eminent among recent FDA breakthrough designations here.  Fenn et al. discuss machine learning in ER and ICUs, here.
    • Current add-on payments proposed for continuation for 14 technologies.
      • Table, p. 304ff.
  • CMS Fact Sheet here.
  • CMS Press Release here.
  • The Federal Register preview text (inspection text) is online at 1914 pages here.  
    • Until I update it, the pagination in this blog reflects the Inspection Copy version (1914 pp).
    • The published Federal Register version at 86 FR 25070-25642 appeared May 10 here (721 pp).
    • Comments to June 28, 2021.
Transparency Walkback

CMS is proposing to pull back some of its notorious reporting of private negotiated contractual rates, it looks like mostly those involving Medicare Advantage negotiations.   See HealthCareDive here, with links therein to AHA etc.  

Although not including this week's CMS rule, see a very recent HealthLeadersMedia article on the transparency topic and its fallout, here.  Late healthcare economist Uwe Reinhardt famously discussed chaotic and disparate hospital pricing, in exactly today's terms, as long ago as 2006 (Health Affairs here.)

Nothing About SEP-1

Nothing about CMS's increasingly debate-ridden SEP-1 measure.  It's in place for FY2023 (p. 25583), FY2024 (p. 25584), FY2025 (p. 25586), FY2026 (p. 25587).  

In 2020, IDSA sharply criticized SEP-1 and called officially for changes.   Herehere.  

In April 2021,  negative paper by Barbash et al. with Op Ed by Klompas.  Herehere.  (For much additional literature see the bibliographies of these two items.)

In May 2021, see a guidance article by Yealy et al. and a fierce Op Ed by Harvard's Jeremey Faust (herehere).  

Faust notes the emerging body of negative publications about CMS SEP-1 may make it nearly impossible for NQF to re-affirm this measure when it comes up for review and vote this year.  He also notes that even Congress has weighed in on the SEP-1 issue and asked CMS to act (here, p. 144).

____________________________

DRG "Grade Inflation"

In other hospital news, OIG reported that hospitals are trending to higher severity levels in DRG reporting.  Hospitals have base DRG codes (for example, for pneumonia or a cholecystectomy) and payments can double or more for comorbidities reported by additional ICD10 codes.  February 26, 2021 article at RevCycle, here.  OIG here.

CMS disagreed with OIG's insinuations, possibly correctly.  Yes, DRGs payments may be doubled if one finds and adds one "major complication" code to the ICD10 list.  That's actually correct coding, on the part of the hospital, if the patient has that condition, and the payment windfall that follows is simply automated by the DRG programming system.   


Monday, April 26, 2021

Very Brief Blog: A flurry of new policy items in precision oncology

Monday April 26, 2021 - A small flurry of public policy documents or articles relevant to precision medicine.

FDA Reviewing a Set of ImmunoOncology Drugs in Advisory Meetings

This week the FDA is holding panels on a substantial cohort of oncology drugs which each have indications in the "limbo" between somewhat aging accelerated approvals and conversion to full approvals.  

Recall that accelerated approvals may be based on single-arm trials and surrogate biomarkers (like tumor response), as long as there is unmet need and the drug seems promising.   

FDA Places Article on Accelerated Approvals in the NEJM

Timed to the FDA meetings, see an article in NEJM by Julia Beaver and Richard Pazdur; Pazdur is the longtime head of Oncology at FDA.   Here.  The article is called " 'Dangling' Accelerated Approvals in Oncology."   

A table includes 3 "dangling" approvals for atezolizumab, 4 for pembrolizumab, 2 for nivolumab, and 1 for durvalumab.   

Authors note that a failure of the follow-up confirmatory trial need not be fatal, but the circumstances need to be evaluated carefully when that occurs.   In addition, the evolving status of "unmet need" for that indication needs to be evaluated as well.   

I consider this a precision medicine issue; most accelerated approvals are in oncology and of that, many in immunooncology.   The status and performance of immuno-oncology biomarkers (TMB, PDL1, MSI, and others) remains very much in evolution (see section ImmunoOncology Biomarkers at bottom.)  

ICER Releases Stringent Suggestions to Improve "Accelerated Approval"

Separately, see a brand new 40-page white paper by ICER on "Strengthening the Accelerated Approval Pathway," which mostly takes the form of describing holes and weaknesses in the accelerated approval pathway.   Here.   

Again, it's a precision oncology issue in as much as many accelerated approvals are in oncology and often involve biomarkers or surrogate outcomes.  Topics covered in the 40-page review include:

  • Manage uncertainty better:  (1) Better Surrogate endpoint use; (2) Threshold for "meaningful change' in surrogate endpoint [e.g. 30% change in the endpoint, not 8%]; (3) Issues with lack of randomization (one arm studies).  
  • Process issues include: (1) delayed confirmatory studies, (2) Inconsistent decisions [one drug to the next]; (3) high prices of drugs with only conditional/accelerated approval.
  • Proposals for FDA's adoption include:  Use standard templates for info [promotes comparability and uniform processes in review]; greater use of RCTs instead of single-arms; robust requirements to finish confirmatory trials; sunset rules.
  • Payment ideas include:   Mandatory rebates for drugs sold under "accelerated approval," or sale only at "marginal cost price" or under "outcomes based contracts."   (Note that laws for payment issues are generate separate from laws for FDA approval issues.)
Legislation Tightens Drug Definitions

At the White House, Biden signs two bills relative to drug policy, one of which (S.415) tweaks the definition of "active moiety" which relates to new chemical entities (and their privileges) in FDA law.  Here.  How to get bills passed today?  This one had a "unanimous consent" action in the Senate in March and quick voice vote in the House.

Combining Endpoints in New Ways - Surrogate, Process, Outcome

Any time we measure biomarkers, we measure against some endpoint, so how rigorous is the endpoint?  

Regarding endpoints, we're use to composite clinical endpoints like "stroke or death."   What about combining process or other endpoints?   See an essay at Endpoints News by Camarero that recommends combining oncology objective response rate and duration-of-response (ORR/DOR) into a composite endpoint that would have its own rules and acceptance as a benefit.  Here.   
  • (I recall seeing a similar theme in a paper on better antibiotics outcome measures, combining two different process measures at once, see Evans et al., 2015, combining the "DOOR" and "RADAR" rankings into one antibiotics outcome measure - here. See also op ed by Molina & Cisneros).

ImmunoOncology Biomarkers

Graphic:  PubMed Hits for "Immunoonology + Biomarkers," 2010-2020.  See an open-access 2020/2021 reviews, here, here.








Thursday, April 22, 2021

MolDx Finalizes Broad Coverage of LBx cfDNA for Transplant Rejection

Several years ago, the MolDx program provided coverage of the CareDx test AlloSure, which detects leakage of donor organ DNA to recognize transplant rejection.

On April 22, 2021, MolDx released a revised policy that's notable for several reasons.  

First, it's not a policy for one branded test, but for a family of tests with a common indication.  MolDx calls these "foundational" policies (I think of them as "categorical" or "umbrella" policies).  

Second, the scope isn't limited to liquid biopsy cfDNA tests, but includes gene expression tests as well.   Thus, either technology may be covered under this LCD, as long as they serve the common purpose of organ transplant rejection diagnostics.  This is an expansion of scope; the original LCD was proposed as "Liquid Biopsies for Organ Transplant Rejection."  

Consistent with the expansion of scope, the final LCD is heavily redlined (some parts are about 80% redline, 20% original).   This is in contrast to quite a few LCDs, which have minimal changes from draft to final.

The LCD has 37 references.  The policy took nearly a year to finish; the release date in the draft is listed as 4/23/2020.

  • See the new MolDx policy here.
    • L38568; effective 6/6/2021.
    • Helpfully, we already have a billing and coding article, A58010, here.
    • There's a Q&A of comments submitted, A58675, here.  
    • >> Some interesting comments in there.
  • See a single cloud zip file with the draft, final LCDs, and the articles and the redline here.
  • See an article on the policy's release, at Genomeweb, here.
  • See CareDx press release here.
    • CareDx did not move much today, but its share price of $76 is near-high, and about double the average price ($20-35) the last couple years.  
    • The stock tripled from $21 on March 1, 2020, to $68 on March 1, 2021.

The final policy was extensively redlined.


Huang et al. show, "What Would the VALID ACT (FDA) Really Cost Labs?"

Congress has the chance to consider the VALID ACT, which would extend FDA regulation to many lab-developed tests.   There hasn't been much known about how much this would cost on per-lab basis.

An international team - with experts from MGH to Melbourne and Antwerp, along with Friends of Cancer Research - tackles the problem.   Huang et al. report that VALID could cost an active molecular pathology lab from $638,000 to $1.2M (potentially raising US health costs by as much as $1B over a 3-year period).  This modeled the added expense for the U.S. count of 886 cancer treatment centers.

Here are some links:

  • See an MGH press release here.
  • See the original article released ahead-of-print at Journal of Clinical Oncology Practice.
    • Link here.
    • Subscription.
I had the chance to be author 4 of 8.  The article points out that VALID doesn't lead directly to cost projections; you have to interpose a series of assumptions, and the FDA could have wide latitude to produce a less expensive or more expensive regulatory program.





Monday, April 19, 2021

Full Length Unofficial Transcript of Brooks-LaSure Senate Hearing (April 13)

 On April 13, 2021, the Senate Finance Committee held a three-hour hearing on two Biden nominees to senior health policy positions.   These were Chiquita Brooks-LaSure for administrator of CMS, Andrea Palm for Deputy Secretary of HHS.

Here, I link to a full length (unofficial) transcript of the hearings.

  • Unofficial Transcript here.
    • As Google Doc.
    • 64pp, 24,000 words
    • As PDF here.
  • Senate Website (Streaming video) here.
  • Trade Press and Law Firm Summaries


For the Becerra hearings a few weeks ago, I collated all the links and documents in one blog here.  For Becerra, I provided a computer-generated auto-transcript, which is about 95% accurate.

In 2017, in the opening months of the Trump administration, the Senate hearings for Seema Verma were very interesting and shed lights on a range of new directions she intended to take the agency in - here.



Thursday, April 15, 2021

Two Genetics Bills on the Hill: Genetic Counselors Reimbursement; Genetic Testing for Cancer Risk in Medicare

Two bills are percolating in the new Congress, on what may be familiar topics.

In the longstanding battle for payments and coverage for genetic counseling for Medicare patients, such as women who are BRCA-positive, a bill is re-introduced.   See Repr. Brian Higgins' (D-NY) introduction of HR 2144, April 2021, to provide Medicare coverage for beneficiaries who require genetic counseling services.  Trade journal here, legislation here.  [Full text pending as of 4/15.]

Medicare covers BRCA (or HBOC) and LYNCH testing for Medicare beneficiaries who are both at high cancer risk by family history AND ALSO have a personal history of cancer.  Remarkably, some NCCN guidelines are written in two columns, one for patients with personal history, one for others, making it easy for Medicare medical directors to see which is which.  Getting ready for introduction is the Reducing Hereditary Cancer Act of 2021, which would eliminate this problem.  See a two-pager about the proposed legislation at FORCE, Facing Our Risk of Cancer Empowered.  Here.


and




Brief Note: FDA Withdraws Offer to Deregulate 83 Device Categories

 In January, HHS released a proposed regulation that would "deregulate" 83 categories of devices currently reviewed by FDA.   I heard it was released by HHS staff over the objection of the FDA staff.  

But its comment period ran 60 days, and according to one news article, many of the comments were against the deregulation.   (For example, devices that are not cleared and approved by FDA may not fall under some reimbursement rules.)   

Supporters of the deregulation included Roche and Mayo, both in the context of telepathology or virtual pathology devices.

FDA has posted that, having reviewed the comments, it is withdrawing the proposal.  See the inspection copy of the withdrawal, here.   See a trade journal article on the withdrawal, here.  See final publication, 86 FR 20174 (4pp), 4/16/2021, here.

The withdrawal is a bit harsh, accusing the Trump administration of having issued the January proposed rule with notice or approval of FDA, and being scientifically flawed and unsound.  While they remark that the rule attracted many negative comments, they don't discuss the comments.  Instead, they primarily withdraw the proposal due to finding it flawed and unsound.



Parallel?

HHS is currently re-reviewing the MCIT proposal, to cover breakthrough devices.  The context is different, in that MCIT was a finished and finalized proposal.   The HHS put it on hold, and reopened it for a 60 day delay opening with an extra 30 day comment period.  That means they had to have significant concerns (to make that work for themselves).  My hunch is, they can or could do whatever they want on May 15 when the 60 day delay is up and they publish a final position on MCIT.  Our hope is that they'll finalize MCIT "again" more or less the way it was finalized the first time.  But if they wanted to say it was "flawed and unsound," well, they're holding the pen, so to speak.   


AMA Has Busy PLA Agenda; Takes Comments til April 22

AMA takes PLA code applications quarterly and provides a brief public comment period.   See the PLA website here and the current PLA proposal PDF here.   

Assuming this year works the same as past years, all codes coming into the May CPT meeting (whether Category I or PLA) will quickly show up at the CMS pricing meetings this summer.

I tally about 30 new codes, with in addition several withdrawals or edits.  According to a separate calendar webpage, public can comment on the 30 codes until April 22.

I wrote a PLA White Paper in February.  For April 1, there were only 8 new PLA codes (here), and for January 1, only 13.   But for July 1, expect basically all 30 of these new proposals to become actual new PLA codes.


___


Praxis, based in Atlanta, applies for 4 genomics codes.  These are:  Optical genome mapping (DNA), whole genome sequencing, whole transcriptome, and whole genome + optical mapping.   For a description of their optical genome mapping, here.  The technology was featured September 2020 in Genomeweb, here.  Praxis has 50 tests registered with MolDx (here).  The several tests I sampled had a MolDx "Non Covered" remark, but I didn't try to check all 50 entries.




Other chromosomal mapping technologies include microarray and low-pass NGS (here).  The low-pass NGS has 3 labs with PLA codes (Mayo, Perkin-Elmer, and NYGC), plus a proposal for a Cat I low pass NGS code on the May 2021 CPT agenda.   See also a study of the Bionano optical cytogenetics method here.

Saturday, April 10, 2021

Novitas Leaves iRhythm Price at Just Half the Old Price; Investor Call Announced

Update Monday April 12.   iRhythm shares fell rapidly by -35% from about $135 to about $85 on Monday.   iRhythm issued a press release saying they may withdraw from the Medicare marketplace here.  See a typical follow-on story by MedTechDive here.

Share price is down from a high of $250 in January 2021, although it's not much different today than exactly 1 year ago.
_________

As of April 9, iRhythm had a share price of $132 and a market cap of $3.9B, both values down by half from the beginning of the year.  

The cause was a transition from longstanding Category III CPT codes, paid at several hundred dollars, to new Category I codes which CMS was unable to price in time for CY2021.   

Seeing the new CPT codes in January, the main MAC paying for this service (Novitas) dropped its price from $264 to $97, down 70%.  iRhythm issued a rapid press release on January 29 (here) which was soon followed by headlines like "iRhythm Stock Tanks" (here).
  • I noted the impending problem in a December blog, and summarized headlines when the event happened in January (here, here).
In a surprise move on Saturday, April 10, Novitas posted new rates which bring the main iRhythm payment up to about $140, but that's still just  half of the 2015-2020 pricing when the service was a locally-priced, Category III code.  

See Novitas pricing page here.  

Also on Saturday, April 10, iRhythm issued a rapid press release that it would host an investor call on Monday, April 12, here.

A class action lawsuit was announced along the way, but that's a fairly common event (here).  In separate news, an investor called Hill-Rom was reportedly trying to exit a once-expected acquisition of supplier Bardy for $375M (March 1, here; also here.)



____

For several years, iRhythm (and other venders using the code, which were much smaller) was by far the most successful AMA Category III code in CMS utilization.   The transition to RVU pricing is stormy in digital health (my earlier December articles, blog and blog).   More than a decade ago, CardioNet also had a stormy transition between local MAC pricing ($1200, as I recall, then abruptly cut substantially) and CMS RVU pricing (circa $700, as I recall.)  (Some Cardionet detailed historical entry points about MAC pricing here, and here; not vetted for accuracy.  But these are easily available on the internet to CMS medical directors; the first is a qui tam case against CardioNet, the second a shortseller's thesis against IRTC.)  Also available to CMS medical directors, in the past months, a consultancy called MCDA submitted several detailed documents to CMS about RVU pricing of iRhythm codes (entry point here).

The new iRhythm price at Novitas is retroactive to January 1, 2021.  The MAC is not required to, and does not, provide any justification for its price points.  Because RVUs are expected to provide roughly break-even values (this much labor, this much supplies, this much amortized capital equipment, this much malpractice), Novitas may be considering that IRTC's 2020 revenue was $265M while it stated its COGS as $70M.  Typically RVU indirect costs are less than 100% (less than COGS).  The PFS/RVU process does not include value (as in a white paper essay by Brian Blase.)

CMS Challenged by Pricing Multiple DHealth Codes, Not Just iRhythm

Similar story on retinal AI-based testing.  See an article on the winding path of valuation and value-based medicine for AI-retinopathy technology - here.   That one was from the CEO of Retina-AI-Health.  And see a detailed essay on the CMS AI reimbursement topic from Brian Blase, quite critical of CMS RVU pricing methods, here.  

Like the iRhythm code, CMS also decided to leave the main retinopathy-AI code "contractor priced" in December 2019; see rulemaking here and search for "contractor pricing for CPT code 92229."   
At that time, CMS wrote, "...AI applications are emerging, we recognize that issues involving the use of AI are complex. While we agree that the costs for AI applications should be accounted for in payment, AI applications are not well accounted for in our PE methodology."   
More on CMS and iRhythm 

For CMS December policymaking decision regarding local pricing for iRhythm, see the same link and just search for "93241."

CMS has a new (Fall 2020) office called, "Technology Coding and Pricing Group" which may be involved in the CMS decision-making for 2022, during 2021.

___

Original version of this blog April 10, also posted at Linked In.

I've put a ZIP of most of the cited links (Fed Reg, Qui Tam, MCDA, Kerrisdale short thesis, Blase, etc) in the cloud here.   Copy of the April 10 Novitas page here.  See also a December 2020 Bloomberg article here,on the growing crowd of short-sellers like Kerrisdale at that time.

See an NEJM article this week on whether remote monitoring is overused.

Historically, most MACs haven't published their prices for locally priced codes.  Palmetto recently began systematically releasing prices on locally-priced molecular tests (here).




Friday, April 9, 2021

Very Brief Blog: United Healthcare to Require DEX Z-Codes for Medicare Advantage; June 2021

According to a blog by Diana Richard on the XIFIN website, dated April 7, United Healthcare (UHC) will require Medicare Advantage claims to include Z-codes, effective June 1, 2021. Read it at XIFIN - here.


There have been news stories about UHC requiring various types of registry codes for lab tests dating back to mid-2020 (see a subscription story at Dark Report here.)

This is the first major sale of DEX Z-codes to a commercial payer that I've read about, although it appears to apply only to United's Medicare Advantage contracts and not its overall book of business as a payer.  According to Kaiser, Medicare Advantage has 24M enrollees in 2020, about 40% of Medicare.  According to United, they have 6.5M Medicare Advantage patients in 2020.  That gives United about 27% of the MA beneficiaries (or about 10% of all Medicare benes, including the FFS ones.)   

Palmetto GBA (which is related via holding companies to South Carolina BCBS) took over the DEX or Z-Code assets last summer, as was announced at the time when the DEX website was rebranded.  (Here, here).   There are currently about ~14,000 entries in the online DEX test library - here.   Some locally-set prices (prices that are not on a national fee schedule) are newly available in DEX (here).

______

The Z-code system was originated by McKesson circa 2010, prior to the modernization of the CPT genomic coding system.   It was spun off independent of McKesson as Change Healthcare, which in turn is now owned by United, but along the way the Z-code business silo was acquired by Palmetto.   In March 2021, AHA opposed the union of Change and United/Optum in a filing with the Department of Justice (here).






Very Brief Blog: Senate Hearing for Head of CMS, on April 15: Brooks-Lasure

 It's been over a month since Chiquita Brooks-Lasure was announced as the expected nominee for head of CMS.   Since then, Secretary of Health Becerra has had hearings, and has been confirmed.  (Entry point here.)

Senate Finance Committee has now set the Brooks-Lasure hearing for Thursday, April 15, and set up the web page for it.

Brief announcement at Inside Health Policy here.   See the shell webpage for the April 15 hearing at SFC, here.



Wednesday, April 7, 2021

Brief Blog: MedPAC Presents on PAMA (April 2): Posts Deck; Transcript to follow

On April 2, 2021, MedPAC gave a half-hour presentation on PAMA policy and its recommendations for potential changes.  The full report will appear in a June 2021 as a report to Congress.  

_____ 

Stakeholders have complained that in 2017, CMS used rules that selectively under-reported test payments to hospital-based outreach labs and to physician office labs.   In fairness to CMS, the agency was navigating rules from Congress about a lab's revenue sources (Part A vs Part B), and had instructions from Congress (and advice from AMA) not to overburden small physician labs with reporting.   Since then, likely in response to legal actions, CMS has broadened the basket of hospital labs required to report in the next PAMA cycle.

MedPAC used an outside contractor, RTI, to review two topics.  One, could one selectively sample labs and get equally accurate pricing?   The answer is obviously "yes," you don't have to count each one of billions of claims to pin down what the median is, within a percent or so.   And, if you sampled more hospital and physician claims, the median price which sets the next CMS fee schedule, would go up.  

  • Key data showed that independent labs (and mostly big ones) contributed 90% of PAMA data, while hospital and physicians labs were 51% of the market (29%+22%).
  • Hospital and physician lab payments were 45% and 53% higher, respectively, than independent lab prices.

They also considered just going straight to commercial claims databases and not bothering labs directly for data at all, but they seemed lukewarm to this idea.  

In discussion, one panelist wondered if other pricing methods like "competitive bidding" would be better.  That, of course, would be a wholly different legal and policy approach with a lot of complexities (see footnote).   A panelist noted that point of care (local) tests had values, like rapid care and avoided return visit, that may justify marginally higher costs and prices.   A MedPAC staffer noted that the MedPAC was thinking about how to handle all sorts of high priced items with new approaches, whether drugs, DME, or genomics.

From last autumn, 9/2020, see a news article about the MedPAC report process and see a letter at that time from ACLA about the MedPAC report.  Here, here.

Sources

  • Web page for the meeting here.
  • Deck presented, here.
  • Official MedPAC Transcript here.
  • Genomeweb coverage here.
  • Inside Health Policy here (Subscription).
  • I posted a 3 minute video about the meeting at YouTube - here.
  • I discussed some of the background issues in a blog in March, here.






___

There's an academic article in 2014 about the 2007/2008 effort to push the lab fee schedule into a competitive bidding framework - Kautter & Pope, here (fee).  The lab competitive bidding originated in the Medicare Modernization Act of 2003.  After various legal battles, the competitive bidding effort was tanked by a judge in April 2008 - here, here.

I've seen the hospital/independent lab difference personally, when one year I had a PSA from a hospital system lab and BCBS paid let's say circa $80 for it and another year they paid "Labcorp-Quest" if I recall circa $25.

Tuesday, April 6, 2021

Very Brief Blog: CMS Updates COVID Patient FAQ Sheet

CMS has updated its 17 page data sheet on characteristics of COVID patients in Medicare.

See PDF here:

https://www.cms.gov/files/document/medicare-covid-19-data-snapshot-fact-sheet.pdf

Webpage for project, here:

https://www.cms.gov/research-statistics-data-systems/preliminary-medicare-covid-19-data-snapshot

See the March 24 press release here.  Data runs through January 2021.

https://www.cms.gov/newsroom/press-releases/cms-updates-medicare-covid-19-data-snapshot

###

So far, 700,000 Medicare patients have been severe enough to hospitalize, of which 18% expired (126,000):


For inpatients, about half had less than 8 day stay, about half, longer than that.  1 in 3 admitted patients, 30% ,were 11 days or longer.  As shown above, only 36% were discharged to home.

Very Brief Blog: Quick Tour of CMS Online Hospital Data

From time to time I need to research CMS data files on hospital performance, and it's usually a bit of a trick to find it again.  A brief outline here.

  • First, CMS has a simple consumer facing website for quickly looking up data at a local hospital or two.  It's called Hospital Compare - here and here.
  • If you want to navigate big Excel files, you'll go the professional site, data.cms.gov.   This gives you categorical entry points like doctors, home health, hospitals, etc.  Find it here.
  • I clicked on the Hospital data website and I went to the bottom of the page to show 50 (rather than 5) sources once.  Here.

For example, if you're interested in hospital acquired infection (HAI) data, scroll down till you find "Healthcare Associated Infections - Hospital" as a title.  This gives you a 175,000 line Excel dataset.

https://data.cms.gov/provider-data/sites/default/files/resources/05819383a875a1ebce5ed5f28755558b_1607630718/Healthcare_Associated_Infections-Hospital.csv


The 32 megabyte 175,000 line Excel looks like this:


For example, for Southeast Alabama Medical Center in Dothan, AL, their MRSA data is no different than national and their CDIFF data is "better than national."

For more info on "hospital acquired conditions" (HAC) rules and metrics or specifically a subset which are "hospital acquired infections" (HAI), some websites are here:


(For example, an inpatient hospital acquired skin ulcer or hip fracture would be HAC's that are not HAI's.)

___


If you enter the website a certain way, it allows you to go to a "beta" new version of cms.data.gov, but when I tried that, it terrified me and it seemed like data categories and files would become much harder to find that the current hierarchy.   The system they have now is kind of boring and tedious, but the new system "beta" seemed like a confusing form of modernization.






Sunday, April 4, 2021

Very Brief Blog: Spring PLA Codes Due April 8; New Final PLA Codes Released April 2

For anyone planning to get in a Spring Quarter PLA code, the deadline is Thursday, April 8.  See AMA here.

For those who submitted PLA codes in early January 2021, AMA released the results on April 2, 2021.  Find the PDF of code names here.

Codes 0248U through 0254U were released - a fairly small crop (7 codes).   3 of the 7 are from one applicant.

These include:

  • 0248U, "3D Predict Glioma" cell culture test from Kiyatec;
  • 0249U, Theralink 32 phosphoprotein test in breast cancer;
  • 0250U, PGDx elio(TM) 505 gene tumor test, I assume the same as the FDA-cleared test in spring 2020;
  • 0251U, Hepcidin ELISAm "IntrinsicDx;"
  • 0252U, Products of conception aneuploidy test, Igenomix;
  • 0253U, ERA endometrial receptivity test, Igenomix;
  • 0254U, SMART-PGT-A preimplanation cytogenomics test, Igenomix.



Conference Bonanza: McDermott, AI & Dig Path, ISPOR Short Courses, More

Flurry of conferences going by...

McDermott Plus; Diagnostics Forum (April 6)

McDermott Plus Consulting holds its sixth annual Diagnostics Forum, Tuesday, April 6, 12-330 ET.  

Website here, four-page PDF agenda here.  There are a wide range of top speakers both from industry and government; CMS speakers including Tamara Syrek-Jensen, head of coverage at CMS, and Dr Gabriel Bien-Wilner, senior medical director for MolDx.  If I'm reading it right, this virtual meeting may be complimentary.

Cambridge Healthtech - AI and Digital Pathology (Sept 30-Oct 1)

This conference will run September 30-October 1, "a comprehensive program for all aspects of digital pathology and AI."   See the website here.  Early bird registration as low as $399 academic, $899 commercial.

ISPOR - Numerous 4 Hour Short Courses (Monthly)

ISPOR, the international society for pharmacoeconomics and outcomes research, has a bonanza of short courses which are virtually available.  Typically, they are two, two-hour sessions, giving you a lot more depth than an hour's overview but not requiring a full day invested.  

For short courses, here.  March 31-April 1 was Value Assessment of Medical Devices; April 28-29 is Intro to Patient Reported Outcomes (PROs); May 24-25 is Tools for Real World Analysis, and so on.  I sampled one and registration (as a member) was $255.

For all ISPOR events, here.

Just Past: Medtech, Diagnostics, Value & Access (Just Past)

While this European-focused conference on value and market access just went by, March 30-31, you might find the speakers and topics of interest.  See more conferences from this sponsor here.

https://www.evidencelifescience.com/meddevma21-agenda









Thursday, April 1, 2021

Very Brief Blog: Telemedicine and Perfect Competition, Podcast and A Few Links

 Everyone knows that there has been an explosion in telemedicine over the past year and there is much speculation on how it will continue and in what form.  (E.g. for one entry point see Paul Sonnier's essay in The Hill in early March.)

There's an interesting essay on economic and industry-structure repercussions which appeared last fall in Medical Economics under the title, "Will Telemedicine Create Perfect Competition?"  It's by economist Adam Block and health policy expert Michael Adelberg, who heads the health strategy practice at Faegre Drinker.  (Bios here and here).  



The article is newly complemented by a podcast at Health Assurance, a new podcast series.  See the 30-minute podcast here.   

(See index of interesting podcasts. The December 6 podcast authors have a 2020 book, Taneja, Klasko, Maney:  Unhealthcare.   Good podcasts also with Caroline Savello of COLOR, Toyin Ajayi of CityBlock, and Jenny Schneider of Livongo.)

_____


Perfect Competition?

Perfect competition is an abstraction in economics where there is no friction between buyers and sellers and the exchange of goods, and profits rapid fall to minimal levels as prices approach marginal cost.  The absence of perfect competition gives profits, which formally are called economic rents or monopoly (oligopoly) power. 

Businesses tend to want to avoid perfect competition; creating barriers to entry of all types, whether brand loyalty, difficult credentialing rules, etc.  

There's a book that was influential on me a decade ago, In Defense of Monopoly (U Mich, McKenzie, Lee).  Without foreseeable profits, there won't be future cash flow to pay for today's risk and investments, and innovation is badly impaired.   It's hard to invest for innovation (and risk) if your product will also and quickly be sold at the marginal cost of production.   

But there are also a lot of natural ways for prices to exceed marginal cost.  Let's say a dry cleaned shirt costs $5, and in my city there's enough business for a dry cleaner every 2 miles.  Let's say it costs $1 for me to drive 2 miles.  My dry cleaner can charge $5.99 for a shirt (with 99 cents profit) before it makes sense for me to drive 2 miles to get a shirt cleaned for $5.00.   It's these inherent frictions that we may rarely think about, that give even a near-commodity business some return on investment.   Take them away, and the economics go into alternate paradigms we aren't used to.  This in turn affects industry structure and vertical integration (see. e.g. Coase's work on "the nature of the firm.")





For the Record: Links to Becerra Hearings, Documents, Auto Transcripts

Early in the Trump administration, there were hearings for the initial Secretary of Health, Tom Price, and for the CMS administrator, Seema Verma.  The Price hearings weren't particular informative of future policy but I found the Verma hearings extremely interesting and worthy of study, at the time (here, here.) 

We've now had the HHS Secretary Becerra confirmation hearings - I don't know how telling they will prove to be, but I provide links below.  In the near future, we'll have hearings for proposed CMS Administrator  Chiquita Brooks-LaSure (here) and I'll provide links when they occur.



For Your File; Some Resources for Becerra Hearings

Becerra had hearings at the Senate Finance Committee, and at the Senate HELP Committee (Health, Education, Labor, Pensions).

See the HELP webpage here - February 23, 2021.

https://www.help.senate.gov/hearings/nomination-of-xavier-becerra-to-serve-as-secretary-of-health-and-human-services

There is a 3h 20m video streaming here.  There is a link to a simple two-page opening statement here.

See the SFC webpage here - February 24, 2021.

https://www.finance.senate.gov/hearings/hearing-to-consider-the-nomination-of-xavier-becerra-of-california-to-be-secretary-of-health-and-human-services

There is a 3h 18m video streaming here.  

There are more document links.

  • Sen. Wyden's 2-page opening statement here.
  • Sen. Crapo's 3-page opening statement here.
  • Becerra's 2-page opening statement here.
  • Of more interest, a 95-page PDF of "questions for the record," here.
Auto Transcripts For the File

I did not see easily available transcripts on the hearings.   I made auto transcripts which are not ideal but which do provide a written record of the content with about 98% accuracy.

I've put all the PDFs and the Word document auto transcripts in a cloud zip file HERE.  

Note that SFC video stops, for me, around minute 30, and so the resulting auto-recording and auto-transcription is in two parts, Part 1 and Part 2, as will be clear in the Zip file.  

Regarding auto-transcripts, the best service I know is otter.ai.   Regarding rapid online human transcripts, my favorite is Rev.com.



Very Brief Blog: Prior Auth Saga Continues; Open Information Webpages at United Healthcare

The saga of prior authorization continues - a colleague pointed out that United Healthcare has a number of open-access webpages that describe its evolving rules for providers.

See UHC "Genetic and molecular lab testing updates" - here:

https://www.uhcprovider.com/en/resource-library/news/2021-network-bulletin-featured-articles/0421-genetic-molecular-lab-testing.html?cid=em-providernews-apr21nb-apr21

See UHC "Genetic and molecular lab testing Notification/Prior authorization" - here:

https://www.uhcprovider.com/en/prior-auth-advance-notification/genetic-molecular-lab.html




Very Brief Blog: Comment til April 23 on CAP Recommendations for Immuno-Oncology PD-L1 Tests

We've read about multiple efforts (!) to harmonize PDL1 testing and reporting, and we've read much about other biomarkers such as tumor mutational burden (in the news this week; here, here).   '

The College of American Pathologists (CAP) has a webpage dedicated to upcoming and recently finished CAP guidelines - here.  

One of the newest is an in-progress guideline for PD-L1 testing - here.  The PD-L1 webpage states, "The primary goal of this guideline is to develop evidence-based recommendations for the testing of immunotherapy/immunomodulatory biomarkers including PD-L1 and TMB in patients with non-small cell lung cancer (NSCLC)."   


The webpage provides you with 8 "key pre analytical questions" and creates an open comment period on proposed guideline conclusions which is open to April 23.   There's a link to a 7-page bibliography, and a link to a PDF of 6 proposed summary recommendations.  You can click on a comment box which collects basic information on the commenter (e.g. profession, such as pathologist) and allows step by step comment on each of the recommendations.

Comments are open until April 23.

At the work in progress webpage here, ten topics are listed: MSI testing, glioma evaluation, monoclonal gammopathies workup, HPV lesion terminology (ano genital), PDL1, whole slide imaging, IHC validation (in general), amyloidosis protocols, HPV testing rules for H&N cancer, TKI testing in lung cancer.  

##

Read about Friends of Cancer Research efforts to harmonize mismatch repair reporting (here) and TMB reporting (here).