Friday, July 15, 2016

CMS Proposes Pre-Diabetic Preventive Benefit - Raises Concern about Dhealth Delivery of DPP

Last March, CMS announced successful program metrics for a pilot diabetes prevention program (DPP) targeted to patients meeting prediabetic criteria.

At the time, I noted the support of myself and many organizations for DPP services.

However, I also noted, in some detail, that the actual support for the particular Medicare DPP pilot program was thin and that it had some contradictory results.  For example, it reported large cost savings of several thousand dollars per patient, whereas these programs are usually rated as cost-effective, not cost saving.

The project is not published in a peer reviewed journal, and to achieve its metrics, CMS allowed the program evaluator to make some statistical waivers, like waiving the "p<.05" concept that bedevils most scientific publications.

CMS gave the DPP pilot data a "pass," in order to meet larger program goals, whereas other coverage and policy decisions get nitpicked to death by policy gurus at CMS.

CMS has now incorporated the DPP benefit in summer rulemaking, and is requesting public comment.   CMS will include virtual or remote delivery of DPP.   CMS writes,
CDC recognized DPP organizations deliver DPP services in-person or virtually via a telecommunications system or other remote technology. The majority of current DPP organizations provide DPP services in-person, but an emerging body of literature supports the effectiveness of virtual sessions delivered remotely. We propose to allow MDPP suppliers to provide MDPP services via remote technologies.
However, CMS states that:
We recognize that the provision of MDPP services by such virtual methods may introduce additional risks for fraud and abuse, and if this proposal is finalized, we would propose specific policies in future rulemaking to mitigate these risks. We thus seek comment on whether there are quality or program integrity concerns regarding the use of virtual sessions, or whether they offer comparable or higher quality MDPP services when compared to in-person services. We seek comment on strategies to strengthen program integrity and minimize the potential for fraud and abuse in virtual sessions.
More detail, including some in's and out's of the legal implementation tactics, below the break.

Expansion of Diabetes Prevention Program

Last March, CMS announced that a pilot community diabetes prevention program had been successful (here).  See the current rule at page 46413 forward.  In March, and in current rulemaking, CMS certified the program was both clinically effective and cost-saving, thus meeting the two legal factors that allow CMMI to roll out DPP as a permanent national Medicare benefit.

While I am happy the program will go forward, the "analyses" conducted by CMS to justify the program as effective and "cost saving" were shaky.  See my March 2016 discussion (here).

My point in March was that a program or product CMS does not like, will be "pecked to death" by the agency's critical, and sometimes speculatively critical analysis.   In contrast, a program the agency wants to push forward is bulldozed forward however thin or contradictory its data.

One of my main points is that many DPP trials have found these interventions to be cost-effective (e.g. dollars per QALY), but never before as "cost-saving" and never before, saving a couple thousand dollars in a few months, even in patients with only 1 training session.  

CMS assertions that the quickly-done and thinly-documented CMS DPP will be "cost saving" (a pesky legal requirement for program expansion) should trigger skepticism in light of the extensive prior experience and literature.  If anyone actually reads the supporting documents, the flaws are all over the place.  But, as the saying goes, nobody reads anymore.

The DPP program could be adopted either through the unilateral action of the CMMI; or, if the program were an approved preventive service by the USPSTF.   The latter channel is not available, which I explain next.

As buried on page 46,414 of the rulemaking, CMS notes it has authority to add additional preventive services under 1861(ddd) if and only if they are approved as grade A or B by USPSTF.  DPP isn't so approved.

Therefore, CMMI will use its authority to waive any section of the statute (1115A(d)(1)), and waive the requirement for USPSTF A or B grading -- which is otherwise the main pillar of the preventive benefit law (1861(ddd)) that CMS is invoking.

Maybe USPSTF standards are too strict - , the ADA has complained that USPSTF is not doing enough regarding diabetes prevention (here).  On the other hand, at almost the same time the Medicare rule came out, the WSJ ran an article critical of excess hype and alarm over prediabetes and its new industry of diagnostics and prevention, here.   While "prevention is worth a pound of cure," parallel concerns about medicalization and/or commercialization of mild conditions dates back to at least 2002 (here; see also a 2005 article here, a 2008 article here and even a 2012 European Parliament review, here).

Also on page 46,414, CMS says it will also waive the requirement for new preventive benefits to pass through the hard evidence review and public transparency of the NCD process.  CMS says it is avoiding the NCD process because "it would create implementation problems" because the benefit also creates a new "supplier class" (e.g. a DPP delivery center).   Obviously, the NCD process could have commented on the merits of the proposal while the rulemaking created the supplier class.  The NCD process is normally based only on peer-reviewed data, which the "Y" program doesn't have.

In summary, legalistically, CMMI isn't creating a new type of benefit.  It is simply deploying DPP under the existing 1861(ddd) benefit for preventive services, which it can do, by waiving the USPSTF requirement and waiving the NCD requirement found at 1861(ddd).

But wait, there's more.   CMS has huge problems with fraud, and requests public comment on how to handle potential liabilities with the new DPP supplier class.  The Y (one potential supplier) is not likely a liability.  But CMS writes that "non medical personnel will deliver services in a non medical setting."   CMS already views this class of supplier as a "high risk category."  CMS "recognizes the risks of beneficiary inducement or coercion."

Payments to suppliers are heavily based on weight loss (5% from baseline).  There is a $160 bonus for each patient who meets this in the first six month, and hundreds of dollars of maintenance payments (to the supplier) are predicated on maintenance of a 5% weight loss.  CMS notes this is a payment approach similar to commercial programs.

Virtual Delivery of DPP

CMS puts some special effort into discussing virtual delivery.

"CDC recognized DPP organizations deliver DPP services in-person or virtually via a telecommunications system or other remote technology. The majority of current DPP organizations provide DPP services in-person, but an emerging body of literature supports the effectiveness of virtual sessions delivered remotely. We propose to allow MDPP suppliers to provide MDPP services via remote technologies. 
As part of our evaluation of the MDPP expansion, to the extent feasible, we will evaluate the effectiveness of MDPP services, particularly in relation to virtual versus in-person services, and,using the evaluation data, may modify or terminate this component of the expansion as appropriate.  
To permit such evaluation, we are considering specifying the nature of the virtual service and the site of the service in codes included on claims submitted for payment, as well as collecting information on the nature of the virtual service and the site of service at the beneficiary level from MDPP suppliers.  
We seek comment on this approach. Under this last example, MDPP suppliers would be expected to maintain this information as part of the beneficiary level cross walk discussed under the IT Infrastructure and Capabilities section of this proposed rule. 
We plan to monitor administrative claims for virtual services to identify any unusual and/or adverse utilization of the DPP benefit. We seek comment on specific monitoring activities or program integrity safeguards with respect to virtual services, in addition to the time period in which such enhanced monitoring activities should occur."
CMS continues,
MDPP services provided via a telecommunications system or other remote technology will not be part of the current Medicare telehealth benefits and have no impact on how
telehealth services are defined by Medicare. We recognize that the provision of MDPP services by such virtual methods may introduce additional risks for fraud and abuse, and if this proposal is finalized, we would propose specific policies in future rulemaking to mitigate these risks. We thus seek comment on whether there are quality or program integrity concerns regarding the use of virtual sessions, or whether they ffer comparable or higher quality MDPP services when compared to in-person services. We seek comment on strategies to strengthen program integrity and minimize the potential for
fraud and abuse in virtual sessions.