Sunday, July 31, 2022

New Email Service for This Blog (August 1, 2022)

Several hundred readers get DISCOVERIES by email.  

We've moved from the old service ("feedburner") to the new one ("FEEDBLITZ.")  

Effective Monday, August 1, interval blogs should be summarized and emailed on Mondays and Thursdays. 


  • Some readers might find the new emails go to spam/trash and please use whatever settings worked for the prior service to untrashify them.
  • Every email includes an Unsubscribe button at bottom.
  • Note that on my blog, on a desktop browser, a link for subscription is shown at the top right. 
  • Subscribe link -


For news aggregators or newsreaders (I use Feedly), you may be able to use the basic DIHP blog link or this special feed link:

Pacific Edge Bio Issues Press Release about Novitas Cancer LCD

A few weeks ago (June 9), I issued a blog about a novel new LCD for cancer biomarkers at Novitas - here.   The LCD primarily acts to refer coverage of cancer biomarkers to outside guidelines and databases.  About the same time (June 12), I issued a blog about "Medicare test coverage without any articles."  (For example, someone would ask me, "Company A says its test is covered, and it bills under Novitas, but I can't find any statement anywhere.")  Here.  Of course, searches are always imperfect, and the only answer with a null search is that, "the rules were hard to find."  

Usually, when I get these requests and can't find a MAC published article, I submit a FOIA request to the MAC for any internal documents or correspondence about covering and pricing the test of interest.

Press Release about LCD

Here we have a press release (August 1, 2022) from Pacific Edge Bio, about how the new Novitas LCD would impact its coverage.  Bottom line: They say the impact might be negative.  Here.   

Here are some procedural comments, by Pacific Edge, about its current experience with the US LCD revision process at Novitas:

...Having consulted with our US-based advisers and industry experts, Pacific Edge believes the proposed changes are unlikely to survive the ongoing review process in their current form. The consensus view we received was that the proposed changes to the LCD are contrary to US legal requirements and precedent....

What does Pacific Edge think of the LCD and LCA?

Pacific Edge believes the proposed LCD and LCA will not survive in their current form because:

  • They do not meet the statutory requirements for an LCD.
  • They do not comply with CMS’ requirements for the development of LCDs.
  • Their new approach of using knowledge bases as a determinant for CMS coverage is not well supported within the draft LCD.
  • The proposed LCD and LCA are insufficient to allow stakeholders to understand whether tests are covered and would inappropriately restrict coverage for tests.

What is Pacific Edge doing to change the Novitas proposal?

Pacific Edge has presented in person to Novitas and has made a detailed submission comprehensively setting out its concerns and the drawbacks of this new approach. It is also sharing insight with other biotech companies that could be similarly affected by the new approach to make their concerns known to Novitas before the end of the consultation period, and engaging with industry advocates.

If the draft LCD was first published in June, why did Pacific Edge only notify the NZX and ASX and call a trading halt on Friday 29 July 2022?

In June the draft proposals did not appear to affect Pacific Edge since there was no indication that coverage for Cxbladder under its existing LCD was changing. Additionally, Cxbladder was not explicitly mentioned in the proposed LCD.....(continues)

On July 28, 2022, Novitas posted a comment period extension (here) with the following elaboration:

Posted: 7/28/2022 NOVITAS

Note Posted on MCD 07/28/2022: 

As communicated on our website on 07/25/2022, the Comment Period End Date has been extended for this Proposed LCD to 09/06/2022. The original Comment Period End Date was 07/23/2022, which displayed on the CMS MCD from 6/9/2022 through 07/28/2022. The extension is due to changes being made to the related draft article. 

Please refer to the Related Local Coverage Documents section at the bottom of this Proposed LCD for changes made to the draft article (DA59125, Billing and Coding: Genetic Testing for Oncology) The changes are indicated on the document note at the top of the document.


I've been meaning to write a blog (or vlog) about precedents for using guidelines for LCDs.   

For example, the NGS MAC article A56199 for BRCA testing simply points, in a short statement, to the current NCCN guideline.  The MOLDX and NOVITAS MACs have LCDs for pharmacogenetics primarily simply point to the FDA guidance on PGx genes and the CPIC guidance and reviews on PGx genes (L38337).   Then, there is this new proposal for Novitas which ties cancer biomarker coverage to NCCN (and a couple other) guidelines.   MOLDX recently finalized a hereditary cancer testing LCD L38966 which has a heavy emphasis on a reference to outside guidelines, as well.

Other times, efforts to tie LCD coverage to guidelines has flopped and quickly.  In May 2016, Noridian posted an LCD that simply tied breast cancer gene expression tests (eg Oncotype, Mammaprint) to "current ASCO guidelines."  This LCD was proposed and withdrawn in a time cycle of just a few weeks.  Here.  

CMS guides Contractors to weight guidelines in their LCDs, along with RCT evidence.   CMS has made national decisions that block coverage of future products that CMS has never reviewed, as discussed recently by Selkoe here.


Friday, July 29, 2022

Medicare Debates: (1) Software Costs in Outpatient Services; (2) Use of Age & Sex in Computer Algorithms

This month, two significant CMS positions on software and algorithms appear in proposed rules, which are open for comment for 60 days.  

One focuses on software in the Hospital Outpatient (APC) setting.  The other focuses on use of parameters like age, sex, race in "computer algorithms" and decision support, to avoid "discrimination."


The non discrimination (and computer algorithm rulemaking) is here and will be officially published in the Federal Register on August 4; comment to October 3. 

SEE THE FINAL PAGINATED VERSION, FED REG, AUG 4, 2022. 87 FR 47824-920, comments to October 3.  Here. 

Press release here July 25. See comment at Health Affairs here and Foley Hoag here.  The rule, and the comments, also touch on the back-and-forth history of some of this regulation which stems from ACA 1557.   I provided a little back story in my July 26 blog here.  It does things like amplify Section 1557 by stating that sex discrimination is defined to include gender identity factors, so that sex discrimination extends to conditions like gender surgery.   

The hospital outpatient annual rule is here (87 FR 44502, 7/26/22) and comment open to September 13.  Focus in this blog on software policy.



Regarding non-discrimination and algorithms, CMS proposes that under Section 1557 of the Affordable Care Act, algorithms that use gender, sex, age, or other conditions in a discriminatory way are not allowed.  

It's not entirely clear to me (either from the text of the rule or the early commentaries cited above) how this impacts AI or other clinical algorithms that depend on sex and age as major drivers of their accuracy.   

Discussion focuses on situations where there have been discriminatory algorithms such as race-weighted algorithms for kidney disease.   However, it is common Bayesian statistics to weight predictions by baseline case states, such as the rise in cancer with age.

The discussion is at page 176ff of the draft rule PDF, and the regulation itself is very concise at 92.210 (page 305).  "A covered entity must not discriminate on the basis of race, color, national origin, sex, age, or disability in its health programs and activities through the use of clinical algorithms in its decision-making."

For example, they call out the ICU critical care SOFA score as potentially racially biased because Black ICU patients have higher average scores.   The causes of the higher scores are important, but it's not clear to me that the "index" SOFA is racially biased.   Black patients also have almost 3X higher triple negative breast cancer, (here), but does that mean Her2neu histochemistry is racially biased?   

See an LA Times article on a California state investigation on use of race and causes of racial bias in hospital protocols, by Noah Goldberg, here.

# # # # #


We were talking about the nondiscrimination and equity rule (algorithm section).  Shifting gears.  

The regular annual hospital outpatient proposed rule is published and paginated in the Federal Register 7/26/2022, 87 FR 44502 forward.  Here.   Pricing specific to the Heartflow product is again discussed, page 44569ff.   However, in this blog, I would point to the later, and lengthy, general discussion of hospital outpatient software pricing, at 44684ff.  This section is titled, "OPPS Payment for Software as a Service."   

Note there are some potential differences between physician office (PFS RVU) pricing and hospital outpatient (APC) pricing.  Office services are paid by piecemeal CPT codes, although CMS has an option to bundle software costs as overhead vs more specific recognition of costs.  CPT codes may contain specified components,e.g. "injection procedure, includes ultrasound if performed.")

Hospital services are paid in broader payment categories (e.g. Radiology Level 1, Radiology Level 2, etc) where different services and often their components are bundled in on APC.   To give a simple example, a PET scan and its radiotracer are paid separately in the physician office setting, but the tracer is secondary to (not paid separately) the PET scan in the hospital outpatient setting.

Turning to software as a service, CMS discussion extends from page 44684 to '89, in pretty fine print, and I'm not sure I've absorbed it fully.   They are describing what they see as problems for software payment, and in particular focus on imaging software that occurs later than the original CT scan or other imaging, and involves an interpretation technology different than that paid for by the original clinical interpretation.  (E.g. very sophisticated next-level software ordered for heart vessel blood flow aka Heartflow company).   CMS is focused on seeking comment on its alternatives for this very important topic.

They are dubious of licensing and per click software, "empirical research has shown that pay per use may lead to overuse of AI technology."  I suspect somewhere down the line, they don't want a $1M MRI machine with software converted overnight to a $1M MRI machine plus a $1M software package to allow it to be turned on and used.  

In July 2022, AMA announced codes for software used in pathology for whole slide imaging, as Category III codes that could be "added" to existing surgical pathology codes.   It's unclear whether, in December, CMS will view this as payable in the APC setting or bundled to the primary surgical pathology code.  (And in fact, I understand that primary surgical pathology codes for hospital outpatients are now already usually bundled to APC fees to the hospital for the surgery).  Much to be written on these topics in the coming months.  


Though not related to software, some readers may enjoy discussion of payment for Category B investigational trial devices, page 44683ff of the outpatient rule.   The outpatient rule on software also has a quick call-out to AI and racial or ethnic nondiscrimination, which is covered in more detail in the separate nondiscrimation rulemaking.  

Thursday, July 28, 2022

Humor: Medicare Part B Proposed Rule

A paragraph that made me smile, reading the July 2022 physician fee schedule proposed rule for CY2023.

Click to enlarge.

Medicare Releases Detailed Database for 2020 Part B Claims Payments

Each year, sometime between June and October, CMS releases a detailed database of all physician paid claims for the calendar year, 18 months prior.  Accordingly, in July 2022, CMS has released this granular data for CY2020.

Find the database here.  It's a cloud database (click VIEW DATA) which can be sorted by doctor name, NPI, specialty, state, CPT code, or more, and any combination of the above.  For example, you could get all the labs that billed Medicare Part B for 81162 BRCA testing in 2020, or all the providers who did cataract surgery.

For example, if you look up CPT codes 81410, 81411 for aortic genetics, 40% of the US services were paid to one lab, in Lafayette, CO.  One can download all the lab codes by every lab and provider (quite a large file) or all the genetic codes 81000-81600, which are quite manageable in Excel.  

The whole database is about 1M lines.  Just pulling down genetic CPT codes (basically 812xx, 813xx, 814xx, 815xx), the data is 3400 lines.  Dollar volume is $1.365M.  Note, however, this series of codes omits a few large PLA codes like Foundation Medicine.   Still, within the domain of codes I checked, Exact Sciences Cologuard is dominant at $208M (15%).   Amazingly, given the huge number of CPT codes in recent years, 4 of the top 6 codes are nonspecific codes (81479, 81599), paid to CareDx, Genomic Health, Guardant, Caris, and Castle.   All the unlisted code payments are in MolDx states.

In the 3400 line database, the top 6 codes are about 40% of all payments.

With the caveat that PLA codes are missing, see the two-page Excel in the cloud here.

CPT code 81408 was almost unknown in the Medicare population through 2017, picked up somewhat in 2018 and drastically in 2019 (approaching $300M).   In 2020, payments are down to $206M.   Commonly, labs that bill 81408 for Medicare patients bill in units of 2, but one lab billed 3.8 uses of this genetic test in the elderly per patient.

Nearly all of the 81408 billing was in Novitas or First Coast jurisdictions (94%).   However, in 2020, there's a California lab that was paid $11.5M for 81408, billing 2 claims per Medicare patient.  This is a little surprising because MolDx usually has a "no pay unless approved" policy for genetics.  I couldn't identify the named lab in question in the MolDx DEX directory, which seemed odd, but sometimes there are DBA doing-business-as factors that make name tracing difficult.

HHS Directory Again Shows Emails

Back in May 2022, I may have been one of the first to publicly comment, that HHS had stripped emails and phone numbers from its longstanding employee web directory.

As of July 28, 2022, the emails and telephones seem to be back.  Note that you search for a name, then get a listing of that name (or matching names), and when you "click" on the last name, you get a new page with more information.

Technical Updates August 1: Repairs to Email Push Service for "Discoveries" Blogs

Back in 2017, my webmaster and I set up a link on Discoveries in Health Policy (upper right) that lets you either subscribe (1) via a blog reader, OR, (2) by email feed.  Google has dropping its years-old Feedburner services (herehere.)   Email feeds to my blog, for several hundred subscribers, may have stopped 7/22/2022.

I believe I've set up an account with the Feedblitz company to manage email lists.

I've added a Feedblitz Subscribe widget at the top right of the blog, and I have set the default for "Mon and Thurs" emails.   The go-live edition is Monday, August 1.

I've migrated already active Feedburner emails to Feedblitz.  If you get a posting you don't want, just click unsubscribe from Feedblitz.   If you are missing the updates on Monday and Thursday, check if Feedblitz triggers "spam."  And you want to manually update or subscribe, here's a link.

Rather than email feeds, I aggregate blogs and news sources in the reader FEEDLY.  Read about that here.

Wednesday, July 27, 2022

Brief Blog: OIG Issues Advisory Against Specimen Collection Compensation

The interaction between a laboratory's specimen collection reimbursement to providers, and, anti-kickback laws, is a complex one.  I am not an attorney, but I have always heard that "fair market value" was one of the key policy considerations.

A new OIG advisory opinion has been posted (OIG 22-09) which raises some important concerns or subtleties around fair market value and other parts of the safe harbor concepts.   

Find the original opinion here:

See discussions at:

  • National Law Review here.
  • Dark Report (Subscription) here.
  • McDermott Will & Emery here.
  • BMD law firm here.
  • Bass Berry here.


Not directly related, see a current news story about a lab in trouble for "sham marketing arrangements," here.

Brief Blog: ASCO Publishes Biomarker Guidance for Advanced Breast Cancer Therapies

 Even more than it used to be, Medicare LCDs are appearing which closely tag coverage to guidelines.   The NGS MAC tags BRCA sequencing coverage to "the most recent version of the NCCN guideline" for that.   MolDx recently released an LCD for hereditary cancer risk testing (such as BRCA or Lynch syndromes), which attributes current coverage policy to outside guidelines.   MolDx and Novitas attribute pharmacogenetic testing coverage to external guidelines from FDA or a PGx organization called CPIC.  Finally, Novitas MAC recently released a proposed LCD for oncology that essentially locks coverage to current guidelines from NCCN or two other databases.

So what's in guidelines?  See the most recent ASCO guideline for biomarkers for systemic therapy selection in advanced breast cancer.  It was released June 27, 2022 as Henry et al.  Find it here:

I'll quote the abstract further below.  If you have a NEW test this year (say, a protein or RNA expression test) to select chemotherapy, it won't be in the guideline, which may count against you at coverage time where the LCDs very specifically point to guidelines as the authorities for coverage.

ASCO also updated guidelines in June 2022 for risk biomarkers for adjuvant therapy (here).


Henry et al:

Candidates for a regimen with a phosphatidylinositol 3-kinase inhibitor and hormonal therapy should undergo testing for PIK3CA mutations using next-generation sequencing of tumor tissue or circulating tumor DNA (ctDNA) in plasma to determine eligibility for alpelisib plus fulvestrant. 

     If no mutation is found in ctDNA, testing in tumor tissue, if available, should be used. 

Patients who are candidates for poly (ADP-ribose) polymerase (PARP) inhibitor therapy should undergo testing for germline BRCA1 and BRCA2 pathogenic or likely pathogenic mutations to determine eligibility for a PARP inhibitor. 

There is insufficient evidence for or against testing for a germline PALB2 pathogenic variant to determine eligibility for PARP inhibitor therapy in the metastatic setting. 

Candidates for immune checkpoint inhibitor therapy should undergo testing for expression of programmed cell death ligand-1 in the tumor and immune cells to determine eligibility for treatment with pembrolizumab plus chemotherapy. 

Candidates for an immune checkpoint inhibitor should also undergo testing for deficient mismatch repair/microsatellite instability-high to determine eligibility for dostarlimab-gxly or pembrolizumab, as well as testing for tumor mutational burden. 

Clinicians may test for NTRK fusions to determine eligibility for TRK inhibitors. 

There are insufficient data to recommend routine testing of tumors for ESR1 mutations [estrogen receptor], for homologous recombination deficiency, or for TROP2 expression to guide MBC therapy selection.

 There are insufficient data to recommend routine use of ctDNA or circulating tumor cells to monitor response to therapy among patients with MBC.

Journal Club: 5 New Major Dx Articles in Major Journals (JAMA, Annals Internal Med)

July issues of JAMA and Annals of Internal Medicine each brought out several articles with new insights for diagnostics.


Since Fall 2021, JAMA has published a surprisingly extensive series of articles on diagnostics and diagnostics excellence.  Newly, we have:

Sarma et al. (2022) Achieving Diagnostic Excellence for Cancer: Symptom Detection as a Partner to Screening.  JAMA July 18 online.  Here.

Sarma et al. describe that only a limited number of cancers have widely endorsed screening tests (cervical, breast, colon, and more recently and less successfully, lung).   Yet even in these cancers only a minority are detected by population screening.  They describe programs in UK and elsewhere looking at early symptom work-up and intervention.  

Reyna et al.  (2022)  Rethinking Algorithm Performance Metrics for Artificial Intelligence in Diagnostic Medicine.  JAMA 328:329.  Here.

Reyna et al. describe cases where special metrics have been weighted to maximize test usefulness during test comparisons.  For example, a sepsis detection test had different weights for penalties when wrong diagnoses were false positives vs false negatives.  This goes beyond classic simple metrics (sensitivity, specificity, area under curve) and merges evaluation with specific use cases.  Another example was in cardiology AI diagnoses, where "false" diagnoses that led to the same workup pathway as correct ones, had smaller penalties.  (Related, see Persson on machine learning ICU algorithms for sepsis, here, and another "rich content" evaluation of algorithms by Cerrato et al. here.)

Turning to Annals of Internal Medicine:

Burke et al. (2022)  The Challenge of Variants of Unknown Significance.  AIM 175:994.  Here 

The authors summarize, "the high prevalence of rare and novel variants in the human genome points to VUSs as an ongoing challenge. Additional strategies can help mitigate the potential harms of VUSs, including testing protocols that limit identification or reporting of VUSs, subclassification of VUSs according to the likelihood of pathogenicity, routine family-based evaluation of variants, and enhanced counseling efforts. All involve tradeoffs, and the appropriate balance of measures is likely to vary for different test uses and clinical settings. Cross-specialty deliberation and public input could contribute to systematic and broadly supported policies for managing VUSs."

Lehmann et al. (2022)  Ethical Considerations in Precision Medicine and Genetic Testing in Internal Medicine Practice: A Position Paper From the American College of Physicians.  AIM here.

The authors summarize, "Whether or not to use genetic tests or adopt technologies such as genome sequencing in clinical care and, if so, when and how to do so needs careful consideration. This paper focuses on the use of precision medicine and genetics in the practice of general internal medicine, where patients may be at increased risk for a heritable condition, may need a drug associated with a known pharmacogenomic variant, or may have a cancer diagnosis for which genetic information may guide treatment decisions.  This position paper is intended to complement and provide more specificity to the guidance outlined in the American College of Physicians (ACP) Ethics Manual, which identifies a number of issues associated with precision medicine..."

Lee et al. (2022)  QUAPAS: An Adaptation of the QUADAS-2 Tool to Assess Prognostic Accuracy Studies.  AIM 175:1010.

One of the world experts in diagnostics tests and theory is Patrick Bossuyt of the University of Amsterdam (here).  Here, the Dutch team adapts a major diagnostic test guideline, QUADAS (QUADAS-2) for prognostic test accuracy: QUAPAS with a "p".   

       QUADAS-2. see here, and here.  See also QUADAS-C, for "Comparisons" of diagnostic tests, here.

(See also my blog on MolDx's newly released checklist for prognostic tests, here.)

There are a lot of issues with conventional low-voltage evaluation metrics for prognostic tests.  

  • First, the tests may often have continuous score or several classes of output (low, medium, high), making familiar binary sensitivity and specificity impossible to calculate.  
  • Plus, Sens/Spec and the related "Area Under Curve" AUC are blind to prevalence, which makes it impossible to calculate positive predictive value and negative predictive value which are crucial for clinical test performance.  
  • In addition, while Sens and Spec are often reported to decimal points ("Our sensitivity is 93.6%") this is nonsensical because small shifts in either prevalence or patient spectrum (along a continuum) will shift Sens/Spec metrics by quite a bit.  
    • Which renders the expression of population test summary metrics with decimal points nonsensical.  
  • And, Sens/Spec are expressed for whole tested populations, where many patients may be either strong positives or strong negatives, but clinically it is the "gray zone" patients where test performance may be basically a coin flip, that are actually the most important patients with the most need for accurate testing.  
    • For an entry point on spectrum effects, see the Usher-Smith 2016 review paper, BMJ 2016:353:i3139, here.
    • Another difficulty is pilot data for a biomarker expressed with excitement as a "p value" between groups - Group High Risk is 8 units, Group Low Risk is 7 units, p<.05 - but the test is not useful because despite scoring a p value on the average, 80% of the patient test scores in Group A and B actually overlap.  

Tuesday, July 26, 2022

Brief Blog; July 18, 2022, Natera Annouces New MolDx Coverage in Bladder Cancer

Let me briefly note a July 18, 2022, press release from Natera, announcing it has obtained bladder cancer coverage for its Signatera liquid biopsy test, from the MolDx program.   Natera notes that it believes this is the first expansion in coverage under a MolDx LCD for minimal residual disease (MRD), which was published in December 2021.

Workers in this space should note that the coverage in the second half of 2022 appears to be based principally on 2019 and 2021 publications, which presumably Natera got to MolDx quickly.  One study had 68 patients with over 600 plasma samples.  The other study had nearly 600 patients with biosamples from an RCT.  In short, Moldx coverage is not fast, not cheap, and not easy, even for one of the most experienced companies.   

I'll put the link to the Natera press below, and quote the press release after the break.


HHS Releases New Proposed Rule for Healthcare Non-discrimination; Deja Vu 2016

On July 25, 2022, HHS released a new proposed rule for non-discrimination in healthcare settings, including in relationship to sexual orientation and gender identity.  First the facts and links, then, more about the back-story.

Facts and Links

See the HHS press release here:

See the new proposed rule here (308pp): 

SEE THE FINAL PAGINATED VERSION, FED REG, AUG 4, 2022. 87 FR 47824-920, comments to October 3.  Here.

The press release quotes Executive Orders under Biden such as one on his first day in office on gender discrimination (here).   In general, the new regulations are framed by the Biden administration as amplifying and specifying rights that are already created by the Affordable Care Act at its Section 1557 (HHS web page and fact sheets here.)   An extensive section discussed the legislative and regulatory history, the prior publication [Trump administration] of CY2020, and recent court cases such as definitions of sex including gender identity.

See news coverage at Healthcare Dive here.    Foley Hoag here.  McDermott here.   (McDermott back story on LGBT in 2020 here).  AAMC here.  See general comments on the topic at WSJ here.  NEJM on COVID and equity-related issues here.  Article on tracking biological sex in health records here.

An interesting section (p 176ff) bans clinical algorithms that cause discrimination in health care services based on inputs including sex and age, as well as race.  See more in my July 29 blog.  For example, they call out the ICU critical care SOFA score as potentially racially biased because Black ICU patients have higher average scores.   The causes of the higher scores are important, but it's not clear to me that the "index" SOFA is racially biased.   Black patients also have almost 3X higher triple negative breast cancer, (here), but does that mean Her2neu histochemistry is racially biased?   


Back Story

There's more back story.  

In 2016, the Obama administration released a "double hitter" proposed rule, that covered gender and other discrimination in hospitals in part, and also, proposed a new requirement for hospital antibiotic stewardship programs (ASP).   81FR39447.

In 2019, the 2016 proposed rule(s) were set to expire uncompleted.  Due to effective communications from stakeholders, the sunset of the 2016 proposed rule was delayed just before the sunset occurred (7/11/2019, 84FR27069).   Then, a few months later, CMS issued final regulations in September 2019 regarding the hospital antibiotic stewardship programs (ASP).  84FR51732.   HHS generally took the position that the LGBT aspects of the regulations from 2016 were redundant to other federal laws, to the extent they were needed at all.  The ASP rules were finalized more or less as proposed in 2016.


By coincidence, CMS just released final instructions to hospitals on the 2019 ASP rules this month, in July 2022, here.



AAMC writes on equity issues in healthcare education in Stat, here.  WSJ responds here.  

See a recent health policy journal special issue on race/racism in modern healthcare here.

Wednesday, July 20, 2022

CMS Updates Guidance for Hospitals for Antibiotic Stewardship: Add ASP Guidance for Labs, from CDC

This blog discusses several new or recent resources.

  1. CMS releases implementation guidance (22pp) on the topic of required Antibiotic Stewardship Programs for hospitals.
    1. Hospitals are required to have ASP since the implementation of March 2020 regulations in the 600 page quality operations manual.
  2. Keep in mind;  CDC recently released its 2020 guidance, Key Activities and Roles for Microbiology Laboratory Staff in Antibiotic Stewardship Programs. (5 pp).
    1. CMS repeatedly refers ASP programs to authoritative ASP guidelines from outside bodies, so in effect they're incorporating this 5 page CDC guide by reference.
  3. More from CDC:
    1. See their 2022 publication, Curren et al.,  on Diagnostic Stewardship for antibiotic resistance and sepsis.  It's based on a December 2019 workshop.  Note it includes the role of diagnostics in ASP, as well as, sometimes the role of diagnostics is to not to use them (don't overuse urine culture).
    2. See also the interesting CDC publication, ASP in outpatient settings, "Guide for Payers."



A few months before the COVID pandemic, CMS updated national requirements for antibiotic stewardship programs in hospitals, finalizing the relevant regulations in the Federal Register (84FR51732 here).  After that publication in September 2019, a few months later, CMS included the regulations themselves in the State Operations Manual, which serves both as a guide to hospital duties and as a handbook for inspectors.  Here.  However, CMS hadn't yet done what it normally does, to provide an amplified practical implementation guide of the regulations themselves.

Now in July 2022 has CMS published a 22-page handbook for how hospitals can implement the ASP in the 2019 regulations.  Find the instructions to hospitals and inspectors online here:

Paper Clip This, to That:

The CMS guidance hardly mentions the word "laboratory" which I found puzzling.  However, the guidance frequently refers to implementing external, authoritative guidelines, and fortunately we have just that from CDC.  See the August 2020 edition of "Key Activities and Roles for Microbiology Laboratory Staff in Antibiotic Stewardship Programs."  It's a critical 5 page addition to the 22 page CMS implementation guidance just publishedHERE.  

But Wait, There's More

Pair this with some important documents from CDC on diagnostics and ASP and sepsis.

And, finally published in 2022, see Curren et al.  This is the result of a December 2019 (pre pandemic!) workshop held at CDC on diagnostics and antibiotic resistance and sepsis, meet report by Curren et al.  Find it in Clin Infect Disease, here.

Although no longer new, there's a 30 page booklet from CDC on how "payers" (!) can improve outpatient antibiotic prescribing.  Here.   Recall studies that overprescribing of Abx in the outpatient setting has been found highest in urgent care centers, middle level in physician offices, and lower in "retail healthcare" e.g. CVS clinics.


I assume that soon the 22 page CMS update on ASP in hospitals, will be integrated into the master 600 page hospital handbook here.

Court Invalidates Patents About cfDNA and Kidney Rejection; Keeps High Bar for Any Dx Patent

 While it didn't change further this week, from June 2021 to June 2022, CareDx's stock is down from around $90 to $20.   In the past few days, a patent court ruled against CareDx.  I am not an attorney, and I'm just quoting from available news stories, to which links are provided.

For a number of years, patent courts have held that biological diagnostic tests are often not patentable, (#1)  if they describe a natural phenomenon that is generally understood (dying transplant kidneys release DNA) and (#2)  if the further bioscience is not itself novel enough (austensibly, here, detecting cfDNA is just understood by everyone who "works in the art.")  One judge dissented strongly.   Case here

Hence, federal appeals court rules in CareDx vs Natera that some patents issued for the cfDNA rejection detection are not patent-eligible.  This concurred with the lower court.   While the case title is CareDx v Natera, Eurofins is also noted on the cover page.  Stanford University is a joint plaintiff with CareDx. 

  • See discussion by the always-lucid Kevin Noonan at PatentDocs here.
  • See PatentlyO here.
  • 360Dx here.
  • See  Reuters here.
  • See Bloomberg Law here.
    • The principles being applied date to Athena v Mayo, at which link see detailed review.
  • See separately Natera v CareDx in May 2022 over Panorama (NIPS) patents, here.

Tuesday, July 19, 2022

Very Brief Blog: MEDPAC Releases Annual US & CMS Health Spending Data Book

Every summer, CMS releases a 200-page book on Medicare spending, with a few comparisons to national healthcare spending

It's out.   This year's version is 208 pages and find it here:

Analyses are usually through 2020, sometimes through 2019.

Note that this year there's a special section on Laboratory Spending, 2005-2020.  It's just one page (PDF 206, numbered 195).   Lab spending rose from $7B to $B 2005-2012, then dipped in 2014-2016 (and that's before PAMA) in part I think due to bundling of many hospital lab tests to underlying hospital services.  Typically hospital labs are 1/3 per year ($3B out of $9B).  Frankly, the biggest lesson I see for lab spending is that it's been relatively flat from $8.6B in 2010 to $9.2B in 2020.      Figure clipped at bottom.

Very Brief Blog: Avalere Launches Series on CMS Policy, Medicare NCDs, Etc.

 On June 16, 2022, Avalere launched a new video and podcast series called "Avalere Health Essential Voice," which will cover topics like CMS and Medicare coverage decisions.   Here's an entry point for the first video, on CMS NCDs and CED.   Scan around the web page for related links and more videos in the series.

Sunday, July 17, 2022

AMA Posts New Proposals for PLA Codes, For Comment

 AMA CPT has published new PLA code submissions for public comment.   The codes will be discussed at a teleconference around August 8 and stakeholders can review the list and request information on a code that they want to comment on.  They now go through ZENDESK portals, which I think is how CPT Cat I codes have been handled too.  

See instructions for how to comment on the PDF.

Web page for PLA here.  I tally 16 proposals.  New proposals in PDF:

CMS Release CY2023 Hospital Outpatient Proposed Rule

On Friday, July 15, after market close, CMS released the proposed hospital outpatient rule for CY2023.

The rule in typescript format is here:

The typeset version will appear July 29.  The typescript version weighs in at 866 pages.

Data files and addenda are an important part of the process, and are found here.  For example, CMS classifies all CPT Category III codes with a hospital pricing level.

The inevitable press release promising great improvements to U.S. medical care through this rulemaking is here and the more detailed fact sheet that summarizes the rule is here.


I haven't reviewed in detail but I noted some changes to organ acquisition charges, which can include lab testing for organ matches (regulations at 42 CFR 413.416, 413.420).   I noted some updates to regulations about drug costs for non-opioid pain management drugs, 416.174.

CMS has a new regulation for HCPCS codes for IDE studies (419.47).  

There is the now-perennial payment classification and reclassification discussion for the Heartflow technology 0503T (typescript page 158ff, or search 0503T).  This has to do with payment rationales for machine learning (or AI) which makes it worth tracking.  As CMS writes,

We noted that HeartFlow was one of the first procedures utilizing artificial intelligence to be separately payable in the OPPS, and providers were learning how to accurately report their charges to Medicare when billing for artificial intelligence services (85 FR 85943). This especially appeared to be the case for allocating the cost of staff resources between the HeartFlow procedure and the coronary CT imaging services. Therefore, we decided it would be appropriate to use our equitable adjustment authority [etc]

CMS remarks that the adjusted claims value (from hospitals to CMS) for Heartflow comes in at $827, which falls between APCs at $498 and $961.   "Since $827 is closer to $961" they classify it at $961.

More broadly than the payment level alone, there is an important later discussion of OPPS Payment for Software as a Service (typescript page 449 ff, final rule, 44502ff) which seems to be evolving some nuances in viewpoint between PFS RVU payments and OPPS APC payments for this important area of medicine.  CMS provides several options for new outpatient software policies, and asks for public comment.  I discuss more about new CMS ideas about algorithms in a July 29 blog here.

There is a section on payment policy for Category B Investigational Device trials (Fed Reg 44683ff).

Congress Says: HHS Needs to Speed Up USPSTF

This past week, see a press release from the office of Representative Eshoo (D-California) on the need to speed up processes at USPSTF.   USPSTF decisions control required preventive benefits for commercial health plans, and are a key gateway to new preventive benefits at Medicare as well.

Here's the link:

I've clipped the text of the letter to HHS after the break.  The letter describes a little bit of slippage in USPSTF funding.  There's some history there.  I recall a Trump HHS budget in 2018 that would have cut USPSTF in half (not enacted).  In 2017, there was a proposal to dissolve AHRQ as part of HHS and move it inside NIH (see Health Affairs here). similarly drastic cuts in a House budget in 2015 here.

There wasn't much press coverage that I could find; the McGuire Woods consulting firm mentioned the letter here.  Compare the 2017 House bill HR 529, "USPSTF Transparency and Accountability Act," which garnered a couple dozen cosponsors, here.  

Wednesday, July 13, 2022

Is Low-Her2 Breast Cancer the Breakthrough Moment for Pathology AI?

The field of artificial intelligence/machine learning applications has been growing robustly over the past five years in both pathology and radiology (see my March 2022 blog here).   A few months ago, FDA released validation documents for its 2021 Paige clearance (here).  One area that's very interesting has been predicting estrogen receptor status from H&E slides (here, here).  But I've been unsure how much that call varies with known worse- and better-differentiated cancers on H&E (which pathologists already see) or how its utility compares with a standard of care estrogen receptor immunohistochemistry (which is standard of care and required in quality metrics plus being pivotal to many drug labels.)

But the new issue in 2022 has been measurement of low-Her2 cases.   Publications back in March 2022 validated that trastuzumab-deruxtecan (Enhertu  - get it?) was a successful drug in conventional, Her2-positive cases (see Cortes, NEJM 2022 here).  But the big news out of ASCO this year was the big success of Enhertu also in very low Her2 cases, which would conventionally fall below Her2-drug guidelines.  See Modi et al. in NEJM (387:9) here, and op ed by Hurvitz (387:75) here.   NCCN has already acted to endorse Enhertu for low-Her2 cases, see June 23 notice here.

But wait, there's more.  How can we define the low Her2 cases?

Pair the NCCN guideline and Modi clinical trial with an excellent cover story and deep dive article in June 2022 CAP TODAY by Karen Titus - here.  

Like the Hurvitz op-ed in NEJM, this article notes that immunohistochemistry for Her2 was always designed to pick up the Her2-2+ and 3+ cases, and not sift through the 0's and 1's precisely.  When you turn to 0's and 1's, pathologist interobserver reliability falls through the floor (Titus article).   See also a one-hour expert panel just posted on YouTube by CAP Today here.

PAIGE and Others See AI for Accurate Low-Her2

PAIGE and others see AI as providing a solution, and potentially a rapid one, for reading low-Her2 cases accurately.   

On July 11, Molika Ashford at Precision Oncology News provided an excellent article (subscription),  on where companies are positioning themselves around the cutting edge of this technology.   Owkin is another company in this space.  

As noted by Ashford, in June, PAIGE already announced "CE-IVD and UK Conformity Assessed designations for a new, artificial intelligence-driven digital pathology assay called HER2Complete, intended to identify patients with breast cancers that express HER2, but at low enough levels that they might otherwise be missed by standard-of-care tests."  

AI Studies Even Easier, Faster, than FFPE RNA Studies

For years, follow-on studies have used FFPE (paraffin blocks) from trials to study new mutations or new RNA signatures.   But AI/ML studies are even more rapid and non-invasive of the archive, because only digitized slide images are required, and no original banked material (including slides) are affected in any way.*   This allows research studies to progress very quickly (although it also potentially lowers barriers to entry for competitors).

Digital Pathology Reimbursement?

As I noted recently, on July 1, 2022, AMA released a long set of Category III codes as add-on codes for pathology case digital imaging (with a remark, if used clinical and not primarily for teaching or research).  With the exception of just one or two PLA codes, other reimbursement codes and more important, coverage policies, for digital pathology are works in progress.  


The FDA-cleared application for PAIGE Prostate slide software is basically scanning for missed neoplastic pathology.   Reading a current opinion column by an experienced prostate pathologist just reminded me that this may be a quite important use case (here).   See an open-access one hour video by Digital Pathology Alliance and Pathology Innovation Consortium PICC alliance on PAIGE's FDA documents, YouTube post June 22, here


*  "No slides were damaged in the making of this FDA regulatory submission."

Tuesday, July 12, 2022

Very Brief Blog: VALID Act and "Interstate Commerce"

Recall that in 1776, the Lee Resolution took the position that "these united colonies are and of right ought to be free and independent states," a statement that the Declaration of Independence was intended to explain.  Right away, it was already perplexing, because they were "united colonies" (united as one thing) and yet "independent states" in the plural, not a nation or "an independent state."   

Various authorities were assigned from these Independent States to the single federal government under the Constitution, such as making international treaties and resolving conflicts between the states including interstate commerce.   The powers allowable under "interstate commerce" broadened in the 1930s.

But I've read that the first version of CLIA (CLIA 67) applied only to laboratories that were "interstate" laboratories, and to hospitals received CMS dollars.   Later, CLIA was quietly expanded by legislation to apply to all human medical testing.  

So what about the VALID ACT, which has been incorporated into the Senate version of current FDA legislation?   It allows the FDA to regulate LDT diagnostic tests as "interstate commerce."   Critically, section 587AA(b) says:
"Any in vitro clinical test that is offered, including by making available for clinical use in the United States, is DEEMED TO BE AN ACT that constitutes introduction into interstate commerce for the purpose of enforcing the requirements of this Act."
You probably see where I'm heading.   If VALID passes, surely some local hospital will want to take this to the Supreme Court, with the argument that offering a local LDT test to its own patients on its own campus is not a reasonable interpretation of what the Founding Fathers would have viewed at the time as active "interstate commerce."   (Look for someone to quote from a 1785 dictionary on the meaning of "interstate.")  That is, you would argue to the Supreme Court that just because somebody pens a definition that a local in-house LDT is "interstate commerce," that doesn't mean the SCOTUS has to agree with the author of the bill.

In the 40,000 words of the VALID ACT, the term "interstate commerce" occurs 23 times.

Monday, July 11, 2022

Very Brief Blog: Example of "Retrospective" Studies Lower Rated

There are several versions of a classic "evidence pyramid" that put retrospective studies much lower than prospective ones.*

If you've like a current example, here's a new ESMO (European Society for Medical Oncology) consensus paper, Trapani et al, that carefully puts "retrospective" data at the bottom of a four-level ranking.  Trapani, Annals of Oncology, Risk-adapted modulation through de-intensification of cancer treatments."  Trapani fig. 1:

Find it here:


* I've always been annoyed by calling these "levels of evidence" since they're only one aspect of that.  An RCT can have a lot of bias, bad endpoints, etc, and shouldn't automatically be ranked on top as "high level of evidence."  Then, there are other types of data where a prospective and retrospective study are intrinsically identical to each other in terms of data (such as observational one year transplant survival when you have the whole cohort and each is followed-up to 1 year.)

Very Brief Blog: CMS Posts Agenda for Lab Pricing Advisory Panel, & Posts Video from June 23 Public Meeting

Brief updates.

CMS held the annual public meeting for new Clinical Lab Fee Schedule codes, on June 23, 2022.  CMS has now posted the video of the two sessions.  Find the video at this webpage, "Recording CY2023...."

CMS will host a two-day meeting of its laboratory expert advisory panel on July 18-19, 2022.  The agenda is posted now, which includes links for video observers.

Thursday, July 7, 2022

CMS Posts Physician Fee Schedule Annual Proposed Rule for CY2023 (July 7, 2022)

 As occurs every summer, CMS posts the multi hundred page Physician Proposed Rule.  This year, it appeared on July 7, 2022.

This triggers a 60 day public comment period, and final decisions around November 1, effective January 1.

Find the Proposed Rule CY2023 here:

CMS initially publishes a "typescript" as above.  (See, "Download the published PDF."   The typeset rule will appear on July 29.  The "inspection copy" or "typescript" weighs in at 2066 pages.

See also a Fact Sheet summary, which is a lot shorter than 2066 pages.  But it's lengthy, at 4600 words.

There is also a press release, which typically includes some somewhat gassy claims about "expanding access to high quality care."

Early Press

Comment at RevCycle here.   Bloomberg Law here. Ophthalmology comment (angry) here.  Similarly at AMA here.  Healthleaders here.  MedPage here.  ASC Review here.  Healthcare Innovation here.  AHA here.  IRhythm press release here

One of the main themes is that CMS proposes circa 4% price cuts (RVU cuts) across the board, the result of financial and legislative rules, not individual decisions at CMS.  

One of the themes is better mental health integration in Medicare, see a separate article on that this week at Health Affairs here.

What's The New Stuff ?

Colonoscopy at age 45

Regulation 410.37 forbids screening colonoscopy payments before age 50.  USPSTF now recommends age 45.  The proposal aligns the regulation to age 45.

Pathology Quality Measure - MMR MSI

I haven't reviewed the whole rule yet but I note a pathology quality measure, page 1650, sponsored by CAP, for the percentage of colorectal cancer reports that include testing for MMR or MSI or both.

Lab Specimen Collection Pricing

There's a section that proposes to update some lab specimen collection rules and payments, page 539 ff to 575 (36pp!).

PAMA Rule Tweaks

Some tweaks to PAMA rules at 532 ff.  They seem to mostly be "conforming" (if Congress modifies PAMA by a year, CMS then modifies its PAMA regulation by a year in a mirror fashion).  

Copayments and Colon Cancer Screening

There have been some legislative changes in recent years that affect copayments when a copay-free screening colonoscopy encounters a polyp (a diagnostic polyp).   Notwithstanding those changes, some of which were in 2021, apparently a screening colonoscopy after a positive Cologuard test is still being considered a procedure "for cause" with a copay.  

CMS proposes to view this differently.  A colonoscopy screening episode will be viewed as "incomplete" if it stops at Cologuard, and "complete" if it encompasses both the Cologuard and the colonoscopy, both of which will be under the preventive services, and copay-free, umbrella.  Page 576 forward.


CMS proposes to nix an outdated NCD for ambulatory EEG.  The disappearance of the NCD, will mean the service is regulated by local LCDs.  Page 593-6.  

iRhythm Valuation Saga

Years-long battles over the RVU valuation of iRhythm patch (codes series 9324X) are settling down on terms not too unfavorable to iRhythm (page 195 forward).  It's an odd area where one stakeholder complained publicly to CMS that pricing was double or triple too high last year (eek).  

Digital Medicine

But there doesn't seem to be much discussion of machine learning or artificial intelligence, which was a big topic last year.   But one digital health topic, remote monitoring services, gets a lengthy treatment at 402 ff, proposing to make codes 98980/81 "non  payable by Medicare" (p. 406) with replacement G-codes. 

NCD Watch: CMS Opens NCD on Cochlear Implants

 CMS has confusing benefits for hearing loss.   CMS by law does not cover "hearing aids" for deafness.  However, some years ago, CMS ruled that a cochlear implant falls outside the definition of hearing aid.   (Apparently, the deceased originalists who wrote the Medicare statute in 1965 would have taken that meaning.)

Anyway, CMS does have a benefit, under an NCD, for cochlear implants.   CMS is considering whether to make the benefit more liberal.   CMS has published a public request to do so.  Right now, CMS covers cochlear implants when hearing scores are <40% (but can cover 40-60% patients under clinical trials).  CMS would raise the regular eligibility requirement from 40% of hearing to 60% of hearing.

  • See the CMS website for NCD activities under current public comment here.
    • Note that public comment regarding whether to revise amyloid PET scan benefits is open, until July 15.
  • See the proposed decision memo for cochlear implants (CAG 00107R) here.
    • Public comment is open for 30 days.
    • See the 9-page letter request here.
  • Note that WHETHER TO ACCEPT THE REQUEST has been under review since October 2020.
    • (Date of 9-page letter).* 

The NCD was opened on March 1 with a pre-decisional opportunity for public comment on the request letter.


The initial version of this blog was incorrect and stated there wasn't a pre-decisional public comment.

NCD Watch: CMS Declines Request to Expand Colorectal Screening Benefits

Medicare has a statutory benefit for conventional stool screening and colonoscopy, but the same section of law allows Medicare to expand to other technologies through the NCD process.   Using this process, CMS has progressively added fecal immuno testing (FIT), ColoGuard, and a benefit for blood-based screening tests if 74% sensitive and 90% specific.  (No such test has been FDA approved, but qualified tests from Guardant and others are expected before long).

One gap is the use of CT colonography (once called "virtual colonoscopy," which now sounds like a colonic Zoom call).   A decade ago, CMS reviewed CTC and nixed it as a benefit.

On April 5, 2022, American College of Radiology (ACR) formally requested that CMS reconsider the status of CTC.   Now, ACR issues a press release that CMS declined the request.

Because CTC is endorsed by the USPSTF for CRC screening, it must be covered by private insurance (see June 29 posting below).   However, Medicare only needs to cover benefits that are electively taken up for an NCD review, and pass it.

  • Read the initial April 6 press from ACR about its proposal - here.
  • See the 15 page April 5 ACR letter to CMS, PDF in the cloud here.
    • ACR checked the right boxes - it's co-branded with support groups like Colorectal Cancer Alliance and Black Women's Health Imperative.
    • Colorectal cancer deaths are 40% higher for Blacks, in part due to lower uptake of fecal-based screening methods (here).
  • See the June 29 turn-down from CMS, in an ACR press release here.

Tuesday, July 5, 2022

MolDx Updates Several Checklists for Tech Assessments

The MolDx program conducts detailed technology assessments for molecular tests, and have templates to guide this assessment in many test categories.   Where instructions are not specific, MolDx expects "industry norms" to be followed such as validation guidelines for that type of test from AMP or other authorities.

On July 1, 2022, MolDx updated 3 of the tech assessment guidelines:

GEN CQ 003 - Tech Assessment CHECKLIST (hint- it's now "Version 9")

MRS PF 020  - Molecular Tests for Risk Stratification (this is "NEW" altogether)

PGX PF 007   - Tech Assessment for Pharmacogenomics (major update)

The Tech Assessment Checklist is a flow-chart based form that guides you to what you need to do with questions like "Is this an NGS test?"   It's adapted to include the brand new MRS Risk Stratification checklist as a landing point, for example. 

The new MRS checklist is detailed, and requires detailed performance criteria, such as sensitivity, specificity, AUC, PPV, NPV, "excess biopsies avoided" (where relevant), etc.   

This actually lays out a point I've mentioned to startup clients recently.  MolDx won't ask if your test measures "6" in positive cases and "7" in negative cases, and p<.05 between the two.  Rather, MolDx will ask what the PPV is, the biopsies avoided (or other outcomes).   A test can easily have a p<.05 value between Group A and Group B yet have very little value in distinguishing them because 80% of the cases in A and B might overlap, despite the small difference in the average.   A different test might also be p<.05 between the groups but have a cutpoint that can separate the cases with little overlap.   

(See the fine paper on Spectrum Effect by Juliet Usher-Smith, BMJ 2016.)   

Risk stratification is a common purpose (such as the Oncotype Dx test for low and high risk breast cancer), and previously MolDx had provided little instruction as how to show the test's real validity.   This should help, and also provides a guide for test developers during the clinical study and data write-up process.    

I'd note, however, that risk stratification scores may not be a good fit to concepts like the requested "sensitivity" and "specificity."   (For example, the Oncotype DX results in low scores with a 5% or less chance of recurrence, and high scores with a 20% or more chance of recurrence, but these aren't really "sens/spec" types of statistics.)  Then again, AUC can also be misleading, since it's based on gold standard for truth, and it's not adjusted for relative frequency (base rate) and assumes no in-between cases other than positive and negative.   

On the special needs of prognostic (risk stratification) tests, see new article by Lee et al. (and the Dutch Bossuyt team), QUAPAS (with a P): Adaptation of the QUADAS 2 Tool to Assess Prognostic Accuracy Studies, Ann Intern Med, July 2022 here.  They view the shift to prognostic test assessment as important enough as to issue "QUAPAS" with a P for prognostic tests, different that QUADAS with a D for diagnostic tests.  The authors write, "Studies to evaluate prognostic tests are longitudinal, which introduces sources of bias different from those for diagnostic accuracy studies. At present, systematic reviews of prognostic tests often use the QUADAS-2 (Quality Assessment of Diagnostic Accuracy Studies 2) tool to assess risk of bias and applicability of included studies because no equivalent instrument exists for prognostic accuracy studies."

QUAPAS (Quality Assessment of Prognostic Accuracy Studies) is an adaptation of QUADAS-2 for prognostic accuracy studies.


The updated PGX checklist is a big upgrade, and is much more detailed.  MolDx began using markedly upgraded tech assessment forms in recent months, such as for infectious pathogen molecular tests a few months ago.