Tuesday, September 10, 2019

New CMS-Funded Study Faults Cologuard Cost Effectiveness and Life-Years-Gained Effectiveness

On September 4, 2019, PLOS One published an open-access study by Naber et al. that is highly critical of the clinical effectiveness (life-years-gained, LYG) and the cost-effectiveness ("worse than any other method") for Cologuard (CPT 81528, $512).

See the paper here.  For trade news at 360DX, here.  360DX discusses potential flaws in the study; I'm not an epidemiologist and I'm discussing the study as-is.  I did notice (see also 360DX) that the costs of a Medicare colonoscopy were significantly underestimated by the authors.  See an Exact press release here; an article in MedTechDive here.




The paper is funded by HHS/CMS grant to the Mitre Corporation, but there are no Mitre authors listed.  Authors are from the Netherlands, Univ MN, Kaiser, RAND, MGH, MSK, etc.

In a key table, the authors find that a Cologuard-managed population would have 30 CRC cases and 8 deaths.   However, an FOBT poulation would have 25 cases and 6 deaths, a FIT population 27 cases and 6 deaths, and a colonoscopy population 9 cases and 2 deaths.   While the Cologuard population would have 79 life years gained per 1000, nearly all the other methods would have more LYG than that.


Authors report that Cologuard could be a dominant cost-effective strategy if cost was reduced to about $10.

CMS Asked About Cost-Effectiveness Rates

The introduction states directly that CMS requested the analysis, asking whether Cologuard was a cost-effective alternative, and if not, at what reimbursement rate or screening interval Cologuard would be cost effective.



____

The paper explicitly describes how the Cologuard rate was set (by crosswalking; 81315+81275+82274), a 2016 rate at circa $500 that was minimally changed by PAMA surveys in 2016 and rate setting for 2018-2020.

____

Note that in addition to the 20-page PDF publication, there are a number of supporting figures online, inline with the web text.
____

See also references to Ladabaum & Mannalithara, ref. 46, Gastroenterology 2016 151:427.

____

Cologuard was the highest-line-item cost in the CMS molecular payments for 2017, table:

click to enlarge

(Note in the above table, while we hear about 200,000 or 300,000 genetic tests, or 5000 new genetic tests per week, just four familiar codes captured over 50% of CMS spending.)

____


Lab tests for CRC in 2017 at Medicare Part B were:

click to enlarge

____

In Statute, at 1861(oo) and (pp)Medicare's add-on line item benefits for preventive prostate testing (which have never been used) include cost effectiveness; for colorectal preventive testing, they do not (here).  When preventive benefits are added after USPSTF approval, cost may be considered.

____

Over two days, Exact share price slipped about 12% (122 to 106), a market cap change of about $1.5B.


____

Note that there are two required levels of compliance with fecal testing.  One is whether the test kit is ever used and returned (the ick factor).   The second is, when you have a positive test, do you advance efficiently to a colonoscopy to investigate (the oh-no colon prep factor and the day-lost factor).