Wednesday, May 18, 2022

CMS Releases Proposed Gapfill Prices for 2022; Comment Period 60 Days

Gapfill Proposals Released

Every summer, CMS reviews new lab codes and prices them by an existing price ("crosswalk") or a de novo price ("gapfill.")   MACs reviewed codes sent to gapfill last fall, this past spring.   Now, on May 11, 2022, CMS has released the "proposed prices."   CMS takes comment for 60 days, which it will share with the MACs.    Finally, CMS will release final prices around September.   This triggers a final appeal period which occurs between the lab and CMS lab price staff.

Now you know everything I know.

Where to Find It

Here's the webpage for lab information at CMS:

https://www.cms.gov/Medicare/Medicare-Fee-for-Service-Payment/ClinicalLabFeeSched/Laboratory_Public_Meetings

Here's the link to the Gapfill Prices, which are an Excel inside a Zip.

https://www.cms.gov/files/zip/2022-clfs-gapfill-prelim-recommendations.zip

Comments may be submitted to clfs_inquiries@cms.hhs.gov by by July 11, 2022.

BASIC DATA

There are 36 codes in the gapfill process.  CMS calculates the pricing based on the 57 zones of the Clin Lab Fee Schedule, which date back to old geographic regions in 1984 that don't really exist anymore.  CMS calculates the final price as the "median" of the 57, or the 28th price from low to high.  MolDx has more than 28 zones, so the uniform MolDx price will always be the median, regardless of minority MACs propose.

The second tab of the spreadsheet is "rationale."

Two prices are near the top of the historical range (0267U $6739 and 0265U $5475.)  These are PRAXIS WGS plus optical genome analysis, and PRAXIS WGS alone.   After that, 4 prices are between $1000 and $1500 and five prices are between $2200 and $3200.

ADVANCED GAPFILL-OLOGY

The NGS MAC appears to always agree with the MolDx MACs.   I tested this by subtracting the CT price from the CA price and this always equaled zero.

As a crude measure of how much the MACs agree or disagree with each other, which means how much the Novitas-FCSO MACs disagree with (MOLDX + NGS MAC), I did the following.  For each row, I calculated the standard deviation and then the standard deviation as a percent of the median (the proposed price).

Using this "percentilized SD," for 13 codes the variation was >25%, for 5 codes >40%.   These high variation codes were 0272U (65%), 0274U (54%), 81560, 0018M, 0257U (all 42%.)  code 0272U was $608 in the MolDx MACs (controlling the median) but over twice as high at $1511 at the Novitas MAC.  

Gapfillology in the Cloud

I've posted an open access copy of my gapfill math in the cloud here. Note there are several tabs.