Wednesday, July 31, 2019

Very Brief Blog: ACLA vs PAMA: It's Alive!!

A 3-judge panel in Washington has revived the ACLA's lawsuit against CMS and PAMA regulations, remanding the case back to the lower court for further decisions.

In October 2018, the case was decided negatively in federal court on the grounds that while ACLA had standing as a litigant, PAMA law shielded CMS price setting from judicial review.  Here, here.

Oral arguments on appeal were heard in April 2019 (herehere).    In parallel, a few weeks ago, a bill to delay and revamp the PAMA process was introduced in Congress (here).

On July 31, Appeals Court ruled that whenever a Congressional bar to judicial review is encountered, it should be interpreted as narrowly as possible.   While the court acknowledged that (as the lower court had done) the judicial review prohibition written into PAMA might be interpreted to apply both to the PAMA rulemaking process and the PAMA pricing process, this would not be the narrowest interpretation.
My sense reading the case (as a non-attorney) was that the Appeals Court was also skeptical of applying the bar to judicial review to notice-and-comment rulemaking, because then, a mischievous agency could run roughshod over notice and comment rulemaking, making all kinds of mistakes, and there would be no recourse to court.
Net-net, the Appeals Court decided that the notice-and-rulemaking phase of PAMA pricing was separable from the actual triennial pricing process, and that the former could be judicially reviewed, although Congress had barred the latter from review.

  • See coverage of the decision at 360Dx here.
  • ACLA press release here.
  • CAP had supported the original case with a brief and issued press release on the win.

Source Document

I've put the full 22 page opinion - which has a very clear overview of PAMA, the issues, and some snappy turns of phrase - in the cloud here.   Written by Judge Cornelia Pillard.  Judge notes that since the case was originally filed in December 2017 (here), CMS has considerably expanded the range of hospital reporting, which was the main point of contention originally.  However, the impact of that is irrelevant to the appellate decision, which simply sends the case back to lower court for the consideration of the facts that was avoided the first time due to the original position on judicial review.

Tuesday, July 30, 2019

Two Offbeat Policy Quirks in Outpatient Rule: Heartflow and Non-Opioid Management

On July 29, 2019, CMS released annual proposed rulemaking for the hospital outpatient and ASC setting, the physician/lab setting, and the ESRD/DME setting (here).  The overwhelming focus of attention was on new proposals that hospitals must publicly post all of their contracted rates with each payer by name (e.g. here). 

There were two quirky zones in the hospital outpatient rulemaking that I'll highlight briefly.  One involves a proposed 50% cut in payment for the digital cardiac technology HeartFlow.  The other involves a lack of movement on CMS's part on better payment status for Exparel, a delayed release anesthetic for surgical pain.

HeartFlow - APC From $1500 to $750?

I summarized HeartFlow's initial battle for payment recognition, and the pathway of CMS rulemaking, in early 2018 here.  (That blog, nerdy in itself, at the bottom links to an even nerdier level of detail and links.)

In brief, CMS initially classed the service as non-payable.  In November 2017 final rulemaking, CMS reversed this non-payment status and placed Heartflow's main code in a $1500 New Technology payment tier.   This was based on invoices or information provided by the public, including the manufacturer.  A few weeks after the November 2017 CMS publication, HeartFlow snagged a $240M funding round.

Now, CMS shows it did not forget this particular story.  CMS found 840 claims from hospitals for HeartFlow in CY2018 (see p 189 ff).   After charge deflation via the cost-to-charge ratio and other math, CMS found a mean hospital cost of $750, so this week it announces plans to put HeartFlow's code in a lower-priced APC, for $750.

How many Heartflow claims came to CMS in CY2018?  This rulemaking states there were 840 claims in the hospital outpatient billing setting.  Sometime between September and November 2019, CMS will release CY2018 data for freestanding Part B claims - here.

Exparel: CMS Pushes Back on Special Payment Status in Outpatient Surgery

In this last couple years, there was been a huge growth of interest in ways to avoid opioid prescriptions.  One of the means is non-opioid pain relief.  For one example, CMS just released a plan to cover acupuncture for lower back pain at least in clinical registry settings (here).

For a couple years, advocates have been urging CMS to cover Exparel with separate payment status in outpatient surgeries.   Exparel is an extended-release formulation of bupivicaine that may reduce the need for post-op opioids. 

In the proposed rule (p 87 ff), CMS discusses why Exparel has been packaged into surgical costs with no extra payment.  CMS states that data analysis showed declining use in the ASC setting, so CMS has made Exparel separately payable there.  CMS does not see declining use in the OPPS setting, and CMS declines to change Exparel's OPPS payment status, although it will be accepting public comment.

I do not have a dog in this fight, but CMS's logic seems fuzzy.  Even flat utilization of Exparel does not show that the pricing policy isn't negative; for example, separately-priced Exparel might have 5x or 10x utilization while helping of patients - you don't know from looking at the current data as the status quo.  (On the other hand, CMS staff may be tracking the occasionally mixed literature, here).

At any case, in responding to the opioid crisis, CMS seemed to bend pretty readily on improving acupuncture coverage, while erecting a wall with regards to Exparel coverage, within a few weeks of each other.

While it's anecdotal, in a July 8 NPR interview, Johns Hopkins professor Travis Rieder describes being in-hospital for severe trauma and postsurgical pain, and that non-opioid formulations were heavily rationed by the hospital due to cost and lack of separate reimbursement.

Monday, July 29, 2019

July 29, 2019: CMS Releases CY2020 PFS, OPPS, and ESRD/DME Rules

On July 20, 2019, several weeks later than normal, CMS issued its annual proposed rulemaking for the PFS (Part B) setting, the hospital outpatient and ASC setting (OPPS), and the annual rulemaking for ESRD and DME.

Below I clip the CMS consolidated brief press release, which contains links to each rule-specific press release and each rule-specific FAQ sheet.    The short press release focuses on Trump administration programs for price transparency, which is just one part of each bundle of policies.

Actual Rulemaking Links
  • OPPS Here .  Fed Reg version August 9, here.  84 Fed Reg 39398-39644
  • PFS Here.   Fed Reg version, August 14, here.  84 Fed Reg 40482-41263.
  • ESRD/DME Here.  Fed Reg version, AAugust 6, here.  84 Fed Reg 38330-38421.
  • The three rules and three fact sheets in a Cloud zip file (50 mb) here.
Hospital Negotiated Rates by Payer

The "big deal" in these rules is in the OPPS rule, requiring hospitals to post negotiated rates by payer by procedure.  There's much ink spilled on this one all over the trade press; for one article on legal aspects here.

The OPPS rules contain some detailed proposed tweaks to Date of Service rules (p. 667-689).  These include leaving both ADLTs and human molecular pathology tests unbundled, but leaving only the ADLTs exempt from date of service rules.  For example, on first rading, it looks like an Oncotype-type mRNA-based test would be billable, but only by the hospital within 14 days, unless, it gets ADLT status, in which case, it would be billing by the lab.  Whew.  (P. 685-686).

Outpatient Surgery Prior Authorization
Generally, Part A and B fee for service program's don't have prior authorization (outside of a few DME categories.)  CMS proposed Prior Authorization for just a few categories of hospital outpatient services that is apparently sees as high risk of lack of necessity.  These are (a) blepharoplasty (eyelid surgery), (b) botulinum toxin injections, (c) panniculectomy ("tummy tuck" if nonmedical), (d) rhinoplasty, (e) vein ablation (varicose veins; covered if medically necessary.)   Remark "these services are most often considered cosmetic...only covered [in] very rare circumstances."  They state, for example, "panniculectomy services were combination with procedures performed on the patient's chest region" (p. 39603ff).  See more on Prior Auth rulemaking in a section at the bottom of this blog.

For general entry points in 2018/2019 to a lot of people very aggravated by prior authorization, here.

CMS proposed changes to how DME rates for novel equipment are gapfilled and how these may be revised downward, pp. 146-179; new rules about DME prior authorization (184-208, esp. 203-208), including an acknowledgement of current "overlapping rules and guidance;" and new rules for diabetic strip competitive bidding, 249ff, including changes triggered by the Balance Budget Act of 2018 (253ff; Section 50414 of BBA).

60 Days
Public Comment periods run 60 days, to about September 27, but CMS must finalize the rules by first half of October to meet its November 1 publication deadline (being 60 days before 1/1/2020).

CMS Announces Landmark Price Transparency Proposals that Benefit ConsumersAdditional proposed rules Advance Agency’s Patients Over Paperwork and Meaningful Measures Initiatives, Strengthen Quality Incentives, and Build on Commitment to Improving the Lives of Patients with Kidney Disease 

Today, the Centers for Medicare & Medicaid Services (CMS) is announcing three important Medicare proposed payment rules that support the transformation of the healthcare system and deliver on providing quality care for patients.

Calendar Year (CY) 2020 Medicare Outpatient Prospective Payment System (OPPS) and Ambulatory Surgical Center (ASC) Payment System Proposed Rule
This proposed rule contains historic changes as a result of President Trump’s recent Executive Order on price and quality transparency that lays the foundation for a patient-driven healthcare system. CMS is putting forward price transparency requirements that will increase competition among all hospitals by requiring them to make pricing information publicly available. Each year, CMS issues a proposed rule to update CY Medicare payment policies and rates under the OPPS and the ASC Payment System.

For more information, see the following links:

Calendar Year (CY) 2020 Medicare Physician Fee Schedule (PFS) and Quality Payment Program (QPP) Proposed Rule
This proposed rule updates payment policies for clinicians paid for the services they provide to Medicare beneficiaries under the PFS and QPP. The proposed policy changes will ensure clinicians spend more time providing high-value care for patients instead of on cumbersome paperwork. As part of CMS’s annual changes to the PFS and QPP, the agency is reducing burden, rewarding clinicians for the time they spend with patients, removing unnecessary measures they have to report, and making it easier for them to be on the path towards value-based care. This proposed rule builds on the Trump Administration’s efforts to establish a patient-driven healthcare system that focuses on better health outcomes.

For more information, see the following links:

Calendar Year (CY) 2020 Medicare End Stage Renal Disease (ESRD) and Durable Medical Equipment Prosthetics, Orthotics, and Supplies (DMEPOS) Proposed Rule

The ESRD and DMEPOS CY 2020 proposed payment rule would update payment policies and rates under the ESRD Prospective Payment System (PPS) for renal dialysis services furnished to beneficiaries and the DMEPOS Competitive Bidding Program (CBP) and Fee Schedule Amounts. This proposed rule puts forth updates to the acute kidney injury (AKI) dialysis payment rate for renal dialysis services furnished by ESRD facilities to individuals with AKI. The proposal would also make changes to the ESRD Quality Incentive Program (QIP), introduce a methodology for calculating fee schedule payment amounts for new DMEPOS items and services, adjustments the fee schedule amounts established using supplier or commercial prices if such prices decrease within five years of establishing the initial fee schedule amounts, and also revises existing policies related to the competitive bidding program for DMEPOS.

For more information, see the following links:


Prior Author Additional...

CMS discusses how Prior Auth for surgeries would work (p. 695ff).  Authorization would not be binding on the contractor if later medical review found the services unnecessary.  Getting prior auth would, however, be binding on the surgeon/hospital for any of the listed CPT codes.  Surgeons would be exempt if they have a 90% pass rate.  "CMS or its contractor will review a prior authorization request for compliance with applicable Medicare coverage, coding, and payment rules (new regs proposed at 42 CFR 419.82)."  This suggest MACs might have to have LCDs or articles published for all the codes, to furnish a review standard.  CMS simply states that prior auth cases must provide all documentation necessary to determine compliance on a case by case basis - it would also be hard to submit this without a published guidance.  CMS would give itself 10 days to turn around P.A.  documents.  Upon a denial, a provider may resubmit.   Associated services (e.g. anesthesiology) would also be deniable.  

Data for which US surgeons were top providers of these services would be available by name and address (sorted by CPT code) in a public Medicare database updated through 2017 (here).  CMS makes this download easy, by providing all the relevant CPT codes in Table 38.   AMA and CMS data for 2017 indicate about 2400 cases of the basic panniculectomy code 15830.  However, when I checked that 2017 Part B data for codes 15830, 15847, 15877, there were only a handful of surgeons providing each, collectively having done about 100 cases for 15830, and typically 10-20 per surgeon (this database doesn't show surgeons with <10 cases of any given code).   I may be missing something, but otherwise the 2300 cases must be largely provided by surgeons with <10 Medicare cases per surgeon.

Exact Buys Genomic Health; Market Cap Shifts $1B

Bloomberg reported on Saturday July 27 that Exact Sciences was expected to buy Genomic Health (here).

That was followed on Monday by:

  • Press release, Exact Sciences, here.
  • Article at MedTech Dive, here.
  • Article at Endpoints, here.  Biospace here.
  • Discussion of early stock movements at Yahoo Finance, here.
  • Note, Exact later trended to a value recovery (here), but the dip is still interesting.
As shown in the stock charts above, Exact Sciences was fairly stable last week and dipped about -8% (market cap decrease about -$1B) on Monday July 29.   

Genomic Health began rising last week around July 23/24, scoring a bump of about +20% in a week (market cap increase around +$500M), and +4% on Monday July 29.

Exact shareholders will own about 90% of the new combined company.   The purchase price is only a +5% premium compared to Friday's price for GHDX, but it's about a +20% premium over GHDX's 30-day valuation.


In other recent funding news, in the liquid biopsy space, Freenome raised $160M in mid-July, and Thrive raised $110M in May.


Interestingly, on the same day, Dark Daily ran an article that FIT may be as good as colonoscopy (here) based on a March 2019 article and Op Ed in Annals of Internal Medicine.

FIT is 82274 $17 or Medicare screening code G0328 ($18; 690,000 uses in CY2017, $14M).  Exact in CY2017 had 230,000 Part B uses for $119M (81528).

Friday, July 26, 2019

AMA's Sudden Unexpected New PLA Code Rules: Once a PLA, Always a PLA?

The AMA holds 3 CPT editorial panel meetings a year, and AMA publishes a summary of actions about a month after each meeting (here).   The most recent summary is from May 2019's meeting in Chicago (here).

At that latter link, see Tabs 53, 54.

In these summaries, AMA panelists voted that if a PLA code (the product represented by the PLA code) is later approved for full-fledged AMA Category I status, it will remain in the PLA appendix but will be tagged with an arrow symbol.   A secondary policy is that if a code is pending as a Category I code, you can't create an interim short term PLA code for it.  You have to just sit tight til the Category I code is published.

click to enlarge

Corresponding text will be published in the 2020 code book, which appears soon, in September/October 2019.


I tallied the latest 103 PLA codes (including those under consideration for Q32019 now), and estimated that about 25 represent MAAA codes.  These are proprietary tests which, with wide acceptance, would be (or would always have been) eligible to be Category I MAAA codes as numbered as such (815xx).  Now, that won't be possibly if they've already become PLA codes.

So far as I know, an Administrative MAAA code can still be elevated to a Cat I (815XX numbered) code.

Not How Category III Codes Are Treated

PLA codes are rapidly issued codes, as are Category III codes.  However, when a product (service) in Category III codes meets criteria for Category I, the AMA elevates the code to a newly number Category I status.  It does not simply add an arrow-sign to the Category III code

But Wait, There's More: September Policymaking re PLA Revisions

The September 2019 AMA meeting in Seattle will also have some PLA policymaking.  Agenda item deals with new policies for when PLA revisions are allowed versus when a PLA code needs to be deleted and restarted as a new code.

click to enlarge

Finally - Any Barrier Between PLA and Category I Codes Likely a Leaky Barrier

The barrier to be erected between PLA World and Category I numbering seems to be somewhat leaky. 

For example, I tallied about 17 codes of 103 that I estimated could appear in the future as Category I codes as non-branded codes.  For example, let's say one lab gets an early PLA code for gene XYZ.  A couple years later, XYZ is a gene with wide use for several cancer drugs.  So AMP or CAP might create XYZ as a category I code.   It would be a code everyone could use and the original PLA applicant might or might not choose to delete his own code if pricing or coverage were better for the new general-usage XYZ code.

Very Brief Blog: AMA CPT Posts Consolidated Code Proposals for Sept CPT Meeting

I've posted information on AMA comment periods for PLA codes and for regular lab codes, in past days on a rolling basis, since each comment cycle has a rapid and different start and stop date.

On July 26, 2019, AMA posted a consolidated code proposal list for the September 2019 Seattle meeting.
  • Note that pathology issues are tabs 29-44 & 80, request info by July 29 to comment by August 5.    
  • For other general CPT codes, for which there are about 40-50 tabs, request by September 5, and file your comments by September 12.   
  • AMA sends out info if you are a directly involved stakeholder (they assess this) in any particular code.
AMA CPT agenda PDF here.

Very Brief Blog: Some New Resources on LDT/IVD Legal Reform

Much has been written in the past several years on potential changes in the ways FDA handles LDTs and IVDs.   Currently, one of the top-ranked articles at 360DX (a sister publication to Genomeweb) is Adam Bonislawski's interview with ARUP's Jonathan Genzen.  (Here.)    The interview is subscription access, but in this blog I note that Genzen has written several open-access articles that present his extensive policy research.

Genzen 2019
Regulation of Laboratory-Developed Tests: A Clinical Laboratory Perspective.
Am J Clin Pathol 152:122-131, August 2019.
See open access article here.

Genzen et al. 2017
Laboratory Developed-Tests: A Legislative and Regulatory Review.
Clin Chem 63:1575-1584.  July 2017.
See open access article here.

While Genzen 2017 is obviously not quite as timely as Genzen 2019, the 2017 article includes some additional length and background, so the 2019 paper rests on top of it.

Genzen also participated in a subscription-access 2019 paper, Mohlman et al. (2019) Reliability and validity of proposed risk-stratification methods for LDTs.  Lab Med 50:194-201.  Here.


A couple other resources

This blog today is far from comprehensive but two other papers of interest:

Although not open access, an older article that preserves the CAP 2011 era perspective is here:
Vance (2011) CAP Proposal for Oversight of LDTs.  Arch Pathol Lab Med 135:1432-5.

See also:
Thompson BM et al. (2016) Understanding the FDA's jurisdiction over LDTs, Divisions between FDA- and CLIA-regulated activities.  Clin Lab Med 36:575-85.


VALID Act (aka FDA Reform 2019)

The most recent Hill proposal is the VALID act which incorporated FDA proposals to the Hill in Summer 2018. 

See IDSA's position here, CAP's collated recent positions here, a January 2019 review from McDermott Will & Emery here, AACC's collated recent positions here.

Numerous organizations recommended Congress move forward with VALID in May 2019, e.g. here.  ACLA press release here.  


HHS Comment on VALID vs FDA Comment on DAIA

In August 2018, HHS issued a 59 page PDF of comments on the then-current reform bill, DAIA.  See blog here, and 59 page document here.  The 59-page document got a lot of attention and was a big push on the journey between DAIA and VALID.

In April 2019, HHS issued an 18-page PDF of comments on VALID.   Here.   It updates the HHS viewpoint on legislative reform into 2019, and I think this 18 page document in 2019 got a lot less attention than the 59 page document in 2018.


Tuesday, July 23, 2019

Very Brief Blog: AMA Posts Proposed PLA and CPT Codes for Labs

On July 23, 2019, AMA posted 3Q2019 proposed PLA codes for public comment.

Stakeholders with a direct interest in a proposed PLA code can apply to AMA for a copy of the application and make comment.  The timetable is rapid, with comments due July 28.  AMA will hold a public audio conference on the codes on August 14, 2019.

  • See PLA home page here.  See links such as REGISTER to listen in on audio meeting, which also would require a confidentiality form.
  • See PDF of new PLA applications in cloud here.
  • I tally about 25 new codes; there are also several code deletions.

Close observers will note that this PLA cycle posting date and comments due date is a few days later than the original calendar for 2019.  If my memory serves, based on older calendars, these codes would have showed up at the September all-CPT Seattle meeting, but the current calendar shows they will be voted by CPT Panel now via an audio conference August 14.

The next PLA application date will be October 10, 2019, which codes should be posted for comment October 17.

Regular CPT Codes for Labs

As noted in an earlier blog, here, AMA has also posted the regular CPT proposals for lab codes for the Seattle September 2019 meeting.  There are about 17 agenda You must request any tabs for comment by July 29, and submit your comment by August 5. 

Non-lab codes for the September Seattle meeting should post about July 26 and will have a later comment deadline.

Sunday, July 21, 2019

Very Brief Blog: MEDPAC Posts Annual CMS Data; Consultancy Says Med Adv Could Hit 70%

MEDPAC Data Book

In CMS news, this past week MEDPAC, the independent advisory body for Medicare policy, posted its 2019 "Medicare Data Book," a 206-page PDF.   Find it online here.

This is separate from the annual MEDPAC report to congress, which was released in June (here) and including topics such as Part D and Part B drug payment reforms and recommendations.  That one weighed in at 503 pp.

Medicare Advantage Growth Curve

In other news, Forbes reported on a new LEK consultancy viewpoint that Medicare Advantage plans could reach 70% by 2030, just ten years in the future.  Current Med-Adv share is 35% and steadily rising.   See Forbes here, LEK here.   Forbes also ran an article this week on the popularity of Medicare Advantage and how it plays into "Medicare For All" debates - here

Friday, July 19, 2019

Myriad Settles OIG Investigation for $9M

On July 19, 2018, it was reported that Myriad entered an OIG settlement for $9.1M.   Myriad remarked it felt the charges were incorrect, but that the settlement would avoid a protracted litigation.
  • The issue was initially reported in March 2018.  Here.
  • Genomeweb today, July 19, here.
  • Genomeweb updated, July 22, here.
  • Lexology article by attorney Scott Gallisdorfer, August 27, here.
  • Myriad brief filing at SEC, here.
  • Myriad stock chart, here.
  • Myriad payments by code, online at CMS, 2014-2017 multi-tab Excel in cloud here.
Myriad's stock popped from about $27 to about $29, about 7%, reflecting a market cap bump of about $140M (current market cap circa $2B.)   This suggests that consensus expectations during the period of uncertainty were more negative than the actual resolution was.

While Myriad's SEC filing is very brief (7 sentences, 138 words), a few notes.
  • The case was a qui tam case (e.g. filed by an individual with knowledge of Myriad billing on behalf of the government.)
  • Qui tam filing started in October 2017.
  • This led to an OIG subpoena, in February 2018.
  • It's about Medicare billing for hereditary cancer (not e.g. about Assurex or Prolaris.)
  • Company believes key allegations were false.
  • DOJ declined to intervene in the case.
  • Settled for $9.1M.
    • (In qui tam cases where it is stated publicly, I have seen statements that whistleblowers may net up to 10-20%. In this case it's not stated.)
    • ($9.1M would otherwise have bought 18,000 law firm hours at $500/hr).
    • (A settlement is largely closed-doors, courtroom litigation far more public.)
  • Filed in District of South Carolina (not Utah where Myriad is located or North Dakota where Noridian processes its Medicare claims).
    • (Some screen shots from the District of SC settlement 3:17-cv-02945-JFA at bottom).

click to enlarge

Screen Shots USDC SC 3:17-cv-029145-JFA (or here).

Online documents include, among others, Dr. Jeter's original 24-page complaint about coding choices 10/31/2017.

click to enlarge

click to enlarge
Did CMS Have Another Route to Effectuate Its Concerns?

Having worked in CMS Part B, it seems there was another route called sampling and extrapolated recoupment.  MAC could have sampled 50 claims from Myriad, randomly from one year of claims, found an error rate (say, each paid $2000 instead of $1000), and extrapolated that error rate in a recoupment across the year of claims.  This is a routine internal maneuver by the MAC and quite rapid.  (It leads into the routine appeals channels upwards towards an ALJ). 

It's unclear why this administrative packet of rules wasn't applied, given the level of speed and control a MAC would have had, and given the stated concerns of Dr Jeter in her court documents.

CMS Requires MACs to Refer "Fraud" to Authorities

When MAC staff find evidence of potential fraud, the staff shall refer to a Zone Integrity Program Contracting (ZPIC) (see Program Integrity manual, 3.6, here.)  This is the only instruction to MAC staff, there is no additional instruction recommending that the MAC staff stop and file a qui tam suit for themselves.  Dr. Jeter's 24-page case uses the term "fraud/fraudulent/defraud" 14 times.



Myriad's stock popped from $27 to $29 on Friday, but settled back near $27 on Monday afternoon.

See also Kessman case re billing Medicare; 81432; here.  Article on Myriad billing, FOIA, the SIRF foundation; here

Thursday, July 18, 2019

Treasury Defines Unusual Category of Quasi-Preventive Service for High Deductible Plans

We've had high deductible plans for years now, and in order to have a large health savings account (shielding as much as $8000 per year from your taxes), the IRS enforces some firm rules on the plan.

Specifically, the plan MUST require a high deductible, meaning the plan is handcuffed and CAN'T pay for services before the deductible is met.  (With exceptions for limited types of preventive care, like mammograms, which have no copay or deductible).

The Trump Administration just issued a new I.R.S. advisory which creates a new category of what I have to call "quasi preventive" services, which include statins, insulin, and SSRI drugs.   The new ruling allows a high-deductible plan to pay for this limited palette of services, from day 1 each year, without applying a deductible or copay to them.   This is at the option of the plan; it becomes a point of flexibility in plan design.

The fixed palette of new "quasi preventive" (meaning no copay, no deductible) services are shown in the following table.   Treasury says it hopes to update the list every 5-10 years to ensure stability of the category.

click to enlarge

Source Materials
  • New IRS Policy 2019-45, here.
  • Trade press at Health Leaders, here.   Dive Healthcare here.  MedCity here.
  • Trade press at Forbes, here.  American Hospital Association here.
  • Treasury press release here.
  • IRS press release here.
  • Sens. Thune, Carper (R-SD, D-Del) thank administration, here.

Policy 2019-45 contains a detailed review of previous and parallel policy and how the new one fits in.


It's a little confusing to call these Preventive Benefits.  The point is to create plan design flexibility to allow the high deductible plan to pay for statins or insulin from day 1, if it wants to, rather than being handcuffed to putting these medical services into debiting the patient's HSA.  And what are the things a plan can pay, in fact, has to pay, from day 1: preventive benefits, like mammography.

The concept of preventive benefits has always had some gray areas.  Mammography or a pap smear in a person with no signs and symptoms is clearly preventive.   On the other hand, smoking cessation therapy has been classified as preventive, I guess preventing emphysema and lung cancer, but it could also be therapy, e.g. treatment of nicotine addiction.  No one has argued that dialysis is preventive, but if you have ESRD, it's preventive of going into a coma and dying.  After a car accident, a tourniquet is preventive of bleeding to death...  But I've also used the example that insulin injections may prevent a diabetic from going into a seizure or coma, but insulin medication "isn't preventive."  Here, per the IRS, insulin is on a list of elective preventive services.   It will be a little bit confusing but the IRS medical services above are a special set of services specific only to the world of HSA debit decisions.  Don't look for treatments like statins or insulin to show up on a USPSTF list of "preventive" services rather than treatments.

IRS says the list will be for services that are low cost, likely to prevent exacerbation of a condition, and highly cost-efficient.


Update.  If you found this topic interesting, you might enjoy a July 2019 Health Affairs Blog about updates to IRS policy for FSA/HSA, here.

Wednesday, July 17, 2019

Very Brief Blog: This Year's Proposed Gapfill Pricing: How Late Is It?

Medicare posted 2019 Gapfill August 21, the latest ever.  Here.

Original July 17 Blog:

Very Brief Blog: This Year's Proposed Gapfill Pricing: How Late Is It?

Each year, CMS has the option of pricing new lab codes through the crosswalk process of the gapfill process.  In the gapfill process, CMS determines in the fall these new codes don't have a good crosswalk pricing solution, and gives them to its MAC contractors to price in the following calendar year. 

MAC proposed prices are supposed to come out by late spring, receive public comment, and then final gapfill prices can be revised by the MACs for publication in the fall.

This year, the proposed gapfill prices are later than ever.  Here are some dates.

2015:  29 gapfill codes, proposed May 26, final September 25.

2016:  15 gapfill codes, proposed June 10, final September 30.

2017:  (No gapfill codes??)

2018:  15 gapfill codes, proposed June 13, final October 19.

2019:  18 codes under gapfill (list here).  No proposed pricing released as of ...July 17, 2019.

This year's gapfill pricing is over one month late, relative to recent benchmarks.

Links to blogs on various past years' gapfill processes, here.

CMS's own rules require MACs to submit contractor-specific amounts "on or before March 31," and since the only explicit thing CMS must do is calculate medians, the delay is perplexing (see 81 FR 41084.)  MACs are supposed to submit revised amounts by September 1, but they can't do that if the comment period overrides September 1 due to late release of proposed prices.

CMS states in regulation it will post gapfill proposals by April 30 (42 CFR 414.509) and take public comments for 60 days.   Thus, public comments will run this year til at least late September of 2019.

CMS must provide an explanation of "rationale" for crosswalk decisions under statute (p. 41086), and promised to provide an explanation of gapfilled amounts after January 1, 2017 (p. 41086, bottom middle column and 42 CFR 414.506(d)(4).  )

click to enlarge

Tuesday, July 16, 2019

Very Brief Blog: AMA Posts CPT Lab Codes for September Meeting

AMA holds 3 CPT meetings per year and posts the proposed codes in advance for public comment.   Lab codes are posted about 2 months ahead of the meeting, so comments are in before several AMA subcommittees meet.   Other CPT codes are posted about a six weeks before the meeting.

On July 16, 2019, AMA posted for comment the CPT lab codes for the Seattle meeting September 26-28.    (By July 19, AMA should also post current pending PLA code applications, too.)
  • See the AMA CPT PDF (with comment instructions) here.   
You can get a copy of the application form for comment, but you must meet AMA criteria as being an interested stakeholder directly affected by the code proposal.

I very briefly summarize the codes below.  See PDF for more details.  Note that in some cases, different applicants may submit for codes with related descriptors; whether that has happened to any of these, I'm not sure.

Tab 29
Four Tier 1 codes (SF3B1, etc)

Tab 30
Tier 1 NTRK

Tab 31
SF3B1 common variants

Tab 32
Tier 2 to Tier 1 for TP53

Tab 33
Tier 2 to Tier 1 for MPL

Tab 34
Tier 2 to Tier 1 for JAK2

Tab 35
Tier 2 to Tier 1 for IGH-BLC2

Tab 36
Tier 2 to Tier 1 for CCD1-IGH

Tab 37
GSP for Epilepsy Panel

Tab 38
GSP to remove RNA from 81445 and add new code

Tab 39
MAAA for Melanoma recurrence

Tab 40
Wound pathogens by PCR

Tab 41
Infectious agents UTI's by STI (sexually transmitted)

Tab 42
PCR, Nail pathogens

Tab 43
Antibiotic resistance testing

Tab 44
PLA Instructions.  Instructions on when to revise code text and when to use new code.

Tab 80
PCR for non-sexual UTI pathogens

Monday, July 15, 2019

Very Brief Blog: CMS Releases Physician and Lab Claims Data for CY2017

Each year, CMS releases CPT code based claims data for Part B in three formats:
  • National Part B data
  • State Specific Part B Data
  • Part B Data for each physician & lab by CPT code
National Part B data tends to come out in late summer or in the fall.  So does State-specific data, which is really MAC data broken out into state-sized spreadsheets (Noridian Oregon, Noridian Washington, Noridian Idaho, etc.)   For example, CY2018 data will come out in 2H2019.

Physician and lab data is released on a slower schedule, with a delay of about 18 months.  CMS has just released claims data for CY2017 by physician or lab. 

The home page for this annual data is here:

And the CY2017 data can be found here:

Click through via "interactive dataset" and "view data" and you'll reach a giant cloud database.   The key thing is to use the search panels at the right of the cloud table (for example, search by lab name, and get every CPT code and volume it was paid; search by CPT, and see every physician or lab paid for that CPT).   Use the Export button to export as an Excel spreadsheet. 

MoPath Codes in Cloud

I've put a 1500 line spreadsheet (150 kb) of all the labs billing all the CPT codes 811XX to 81599, in the cloud here.  Have fun!  (It would also be manageable to download all the lab codes into Excel - 8XXXX - about 800,000 lines and 100 MB.)

Of course, you could download anything of interest - cataract operations, prostatectomies, psychotherapy visits, etc.  

First Snapshot: 81162

Going back to the CMS cloud data, within a few seconds, I pulled all labs in the database billing BRCA 1&2 (81162) in CY2017 and downloaded in Excel:

CMS Payments for 81162 (by Lab) CY2017

Entities at Lower End of Net Revenue in CY2017

In CY2017, Tempus Labs did not seem to have much CMS revenue yet, NPI 1740733708, with 26 payments for CPT 81445 (5-50 tumor genes) at $602 allowed (charge $3800), and 18 payments for 88342 (immunohistochem) at about $100 each.

Nanthealth Labs, NPI 1295026938, had payments for 21-23 units each of 81235 (EGFR), 81401 (Mopath level 2), and 81479 (Mopath procedure), with allowed charges of $332, $129, $144, respectively.

Billing by Unlisted Code 81479

CMS paid $115M under Unlisted Code 81479, and 95% of the top 99% of payments went to MolDx states.

Of this $115M, $32M or 28% went to Assurex, for 14,733 tests at $2,180 allowed payment per test.

click to enlarge
Billing by Foundation Medicine in CY2017

Recall that in CY2017, FMI did not yet have its NCD for coverage (which was proposed only in November 2017), and it had never gotten much Medicare payments in his home MAC in Massachusetts.

FMI opened a satellite lab in North Carolina, where the Palmetto MAC paid it for 1,687 services at $3416 each (charge $5800).  Net-net, FMI garnered only $700,000 on its home court in Massachusetts, but at the Palmetto MAC, in NC, it earned $6.2M:

click to enlarge

FMI was acquired by Roche in June 2018 for a total company value of $5.4B (here), or a multiple of 777 times its CY2017 Medicare Part B revenue.

Use of Somatic Panel Testing Still Modest in CY2017

Although 2017 was the year CMS unveiled its proposed NCD for somatic panel testing by NGS, usage was still pretty small in 2017. 

CMS paid for 1,302 cases of 81445 (5-50 genes) and 4,338 cases of 81455 (51+ genes), along with another 1,600 cases of FMI Foundation One paid by MolDx as 81479.   Still, these tallied only $3.5M of direct payments for 81445/81455, and another $5.7M of FMI payments as 81479.

My point is, this all totals less than $10M out of the roughly $500M CMS spend on molecular diagnostics in CY2017.  For comparison, in 2017 CMS paid Exact Sciences $118M for 231,000 cases of ColoGuard, at an average payment of $502.

Click to enlarge / 81445, 81455

Lab Industry in General: The Full Set of all 80,000-series Codes

I downloaded all the 8xxxx code date for 2017, an 800,000 line, 80MB file my laptop just barely handles and slowly.

But some big-data trivia.

I tallied $6.85B in payments.  About 30 claim lines (provider x CPT) paid over $10M.  The highest was Exact Sciences, paid $118.6M.  Among the top ten pay lines, most were specialty genomics (e.g. Genomic Health), but Inform Diagnostics garnered position 7 with $20.5M for 88305 (surg path) and LapCorp got position 9 with $16M for 84443, thyroid.  LabCorp also got positions 13 and 14 for Vitamin D and for Lipids, respectively, at about $13M each.

Just 40 of the 860,000 total paylines (0.005%) accounted for 10% of all payments.

click to enlarge (Top pay lines, entity X code)
Individual Providers and the 80,000-series Codes

The first individual provider on the pay line chart was a Memphis pathologist at line 132 with $3.9M for 87507, digestive tract pathogen.  The second individual provider was line 189 for 80307, testing for presence of drug, $2.9M.   The highest individual pay line for surg path (and it might represent some kind of group practice?) was line 257 with a Baltimore pathologist and $2.2M for 88305 x 34,000 units (surg path).


Sorting the 860,000 lines by self-reported gender, lines 1-475990 are male, and lines 475991-778211 are female, a ratio of about 60:40.  But the average male pay line was $3,586 and the average female pay line was $2476. 


Six entities garnered between $1M and $4M for immunohistochemistry 88432 (first slide of case).  Interestingly, only 3 entities billed over $1M for 88341 (additional IHC slides), with only 3-7 average slides per case.   Only a handful of higher-billing providers of 88341 billed >10 slides per bene, but one provider billed ~5000 slides for 26 bene's (~200 slides per bene) using 88341, which is difficult to understand.

About ten providers had $100,000-$500,000 in payments for 88344 (multiplex IHC slides).

All About Arsenic Testing

Something - old Columbo episodes? - triggered my interest in arsenic testing (81275).

If you are a Medicare beneficiary, there is about 1 chance in 2000 you will get an arsenic test in any given year.  Read on!

CMS paid for 21,276 units of arsenic testing at about 100 labs, and almost never more than once per patient.  Payment per test is $26.   Arsenic testing is highly concentrated, with LabCorp NC receiving $104,000, Genova Dx receiving $85,000, and Unilab receiving $27,000.  These are 19%, 16%, and 5%, or 40% of all US testing.   Total payments were $544,092. 

LabCorp garnered $154,000 from arsenic testing and Quest $108,000, together holding almost exactly 50% of the arsenic marketplace in the over-65 population.   For this test, LabCorp billed from 13 locations and Quest from 17 locations.  68% of LabCorp billings were from NC.  No Quest location had more than 20% of testing.

81275 and Old Lace

National Part B data files are here.   State level aka "carrier level" data is here.

Very Brief Blog: This Year's Part B and Outpatient Annual Rulemaking: How Late Are They?

Each summer, CMS releases two large packages of proposed rules and policies and pricing for the coming year: the "Physician Fee Schedule Rule" (which includes Part B policies like lab tests) and the "Outpatient Rule" (which includes surgical and non-surgical hospital outpatient services and ASCs.)

Having tracked these rules for 15 years, I usually say they come out July 1, plus or minus five days.  

Let's reset our expectations to mid-July.  In the last four years, these rules have come out between July 8 and 13.  This year, 2019, it's July 15 and they're not out yet, so it's going to be even later than the recent norm. 

When the rules are issued, there are 60 days public comment, and the rules are finalized by November 1.   This means that the later the rules come out in July, the further into September the comment deadline runs, and the fewer weeks CMS has left to turnaround the revised and final rules during October.  

Here are recent dates for first release "inspection copy" rules:
  • July 8, 2015 (for CY2016).  Here.
  • July 8, 2016 (for CY2017).  Here.
  • July 13, 2017 (for CY2018).  Here.
  • July 12, 2018 (for CY2019).  Here.
Where To Look It Up Yourself

Federal Register has a dedicated webpage for each agency, the one for CMS is here.  This webpage shows both recent published Federal Register rules and the latest "inspection copy" rules, a typed version which appears ahead of the typeset Federal Register version. 


Often it is rumored CMS releases rules on Friday nights.  While CMS does definitely save some items for Friday night releases, only the July 8, 2016 rule was  Friday night, of the four examples here.

Very Brief Blog: Dark Daily Highlights Pathologist's Role in Nixing Theranos

In a July 15, 2019, article, the open access site Dark Daily, related to trade journal Dark Report, highlights the role of pathologist Adam Clapper in bringing down the Theranos juggernaut several years ago.

The pathologist, who at the time ran Pathology Blawg, was skeptical of a favorable New Yorker article published in December 2014.
  • See the July 2019 Dark Daily article here.
  • The Dark article builds on an April 2019 Missouri newspaper article here.

The "Discoveries" Theranos Webpage: December 2014 to December 2016

Beginning at the same time, in December 2014, this blog Discoveries began a single multi hundred link blog on Theranos.  That blog page was continuously updated for several years to track the rise and fall of the Theranos story - and it's still online here.

The timeline is remarkable.  I googled backwards to provide internet links on Theranos starting in 2006.  But the main saga runs from December 2014 (the New Yorker article), to October 2015 (the landmark WSJ article), to December 2016, when it was clear that Theranos was kaput.  I stopped updating the website once it was clear that Theranos was doomed.


The Discoveries blog runs 10,000 words and circa 500 links.

Wednesday, July 10, 2019

Very Brief Blog: CMS Posts Agenda for July 22-23, 2019, Lab Advisory Panel

Each summer, CMS hosts a panel of experts to discuss how it should price new laboratory codes, and the panel may be asked to discuss additional topics as well.

This year, the Clinical Diagnostic Laboratory Test (CDLT) panel will meet at CMS on July 22-23.
  • Home page for the panel is here.
  • Specific agenda for July 22-23 is here.
    • The panel will discussed about 90 codes, arranged thematically (e.g. microbiology, NGS, etc).
    • In addition to new codes, CMS asks the panel to discuss automated chemistry panel pricing.
Registration closed July 1 but the panel will be webcast (see agenda for links).

If you want a preview of what these panels are like, how the debate sounds, see my online transcripts from July 2018 summer advisory panel, here.

Foreign Nationals

CMS doesn't always make its special rules for foreign nationals clear.  It added a page of instructions to this agenda, which I've clipped on a separate webpage here.  Foreign nationals have to go through a special process starting about two weeks before a CMS site visit.