The future of virtual medicine: what the CMS MDPP ruling means for patients and digital health

The Centers for Medicare and Medicaid Services will not yet reimburse for telehealth delivery of the Medicare Diabetes Prevention Program (MDPP). What does this mean now and going forward?

tl;dr

  • The Centers for Medicare and Medicaid Services (CMS) announced it will only reimburse its diabetes prevention program (DPP) when it’s delivered in-person (and will not reimburse for “virtual,” digital health versions of the DPP).
  • Multiple media outlets reported that the ruling was because “self-reported weight loss is not reliable for the purposes of performance payment.” But that’s not the whole story.
  • CMS has not fundamentally rejected remote monitoring technologies. Rather, it intends to test virtual DPP models more fully in another pilot. We believe that the delay hurts patients and chills innovation.
  • We suggest an alternative framework—one already used by FDA—that provisionally approves significant “breakthroughs” while data collection continues. This would allow CMS to balance risk while providing early access to potentially high-value interventions.5

This month, CMS published 1,200 pages of payment and agency rule changes. The rules held disappointing news for digital health and people living with prediabetes: CMS will only reimburse the Medicare Diabetes Prevention Program (MDPP) in cases where patients can travel to an in-person service provider.

Under this rule, “virtual” programs will not be reimbursed by CMS. Patients who do not live near a provider, those whose mobility is reduced, and those who face economic barriers to travel—much of the population that relies on Medicare and Medicaid—are excluded from a program which is demonstrated to improve health and save money.

It’s tempting to view this story through the simple narrative of a large bureaucracy moving too slowly. But there’s more to this ruling than first meets the eye, including a likely “next chapter” that digital health innovators can learn from.

A bit of history: What is the MDPP and why does it matter?

Before the CMS MDPP was conceived, the Centers for Disease Control and Prevention (CDC) formally established what it calls the National Diabetes Prevention Program (National DPP). The CDC’s National DPP is based on academic research that was further supported by a large-scale NIH-funded clinical trial showing a reduced incidence of diabetes for patients who followed the DPP’s intensive lifestyle intervention.1 Today, CDC’s National DPP is a resource that connects individuals and healthcare professionals with lifestyle change programs and education that follow CDC’s evidence-based guidelines.

However, the National DPP is ultimately just a recommendation from the CDC, which, as an agency, does not provide reimbursement or deliver healthcare services. That’s where CMS comes in.

The National DPP delivers a combination of lower costs and improved patient outcomes around one of the most prevalent and expensive chronic conditions in America—and in March of 2016, the CMS Innovation Center certified a pilot program to test a new reimbursement model for the DPP.2

The final ruling on the MDPP pilot is, in the big picture, a moment to celebrate. We’ve crossed a regulatory threshold in three cases. It’s the first:

  • National expansion of a CMS innovation project
  • CMS reimbursement that’s pegged to an outcome (weight loss)
  • Reimbursement for a medical benefit delivered in a community health context

The potential fourth, however—reimbursement for virtual programs delivered by digital health companies—was a bridge too far (for now).

A measure of caution about weight measurement

The original CDC National DPP established weight loss as the outcome metric that tracks individual patient progress and which predicts reduced incidence of diabetes. CMS followed the CDC precedent and stuck with weight loss as the outcome measure for the MDPP pilot.

But this is also where things got sticky.

Multiple media outlets reported that CMS ruled against fully-virtual versions of the MDPP because “self-reported weight loss is not reliable for the purposes of performance payment.3 This is perplexing for two reasons.

First, CMS itself notes in the Federal Register that it seeks to be consistent with the CDC National DPP; however, in the case of remote monitoring of weight, CMS has flatly contradicted CDC’s evidence-based standards—standards which recognize fully virtual DPP providers, including remote monitoring of weight.

Secondly, there’s some inconsistency in CMS’s application of its “reliability” standards. Whereas CMS won’t accept “self-reported” weight (even, presumably, from the connected medical devices used by many digital health DPP providers), it instructs MDPP suppliers to accept patients’ self-reported, unverified eligibility status when enrolling in the program.

That’s a head-scratcher—doubly so in light of growing coverage for virtual DPP programs by private insurers.

Of greater concern, some have interpreted this move as CMS closing the door altogether on remote monitoring of weight (and perhaps, down the road, other biomarkers). FDA-cleared, connected bathroom scales are substantially different than weight-loss diaries (the analogue alternative) and therefore merit consideration as something other than traditional “self-reporting” of data. It’s reasonable to believe that CMS will ultimately join CDC and come around to a similar view.

Beneath the surface of the final rule (Hint: It’s not about connected bathroom scales)

CMS is not getting out of the virtual DPP business. On the contrary, CMS intends to explore fully virtual DPP options.

Beyond its concerns over patients self-reporting weight, CMS noted what is likely the real sticking point—and pointed to the path forward: Fully virtual programs were not, in CMS’s view, a part of the pilot. In short, CMS wants more evidence that virtual programs are just as effective as in-person programs.

Folks both inside and outside virtual DPP providers do not, however, universally agree that additional evidence should be necessary. On one side are those who believe that evidence for virtual interventions is lacking. One expert we spoke with pre-emptively offered the view that, “Digital health needs to embrace the need for evidence in a manner similar to the drug industry.”

We at Rock Health couldn’t agree more.

On the other hand, numerous studies by multiple virtual DPP providers and academics as well as meta analyses4 of the growing body of published data on this topic offer evidence that virtual DPP programs are effective.

Neither the study authors nor other experts we spoke with suggested that “we’re done” studying this topic, and the phrase “more research is needed” was liberally applied to many aspects of the DPP universe. At the same time, we’re left wondering—how much is “enough” to at least get started with a reimbursement regime that connects a needy and potentially underserved population with a flexible, cost-effective intervention?

A framework to build on?

This ruling strikes a balance between slowly accumulating evidence and rapidly scaling valuable, cost-saving innovations. The dollars and health outcomes at stake are tremendous.

Does it have to be this way, though? Perhaps not.

We’d like to see something akin to FDA’s “breakthrough designation” for new drug development for CMS. Under FDA’s “breakthrough designation,” companies can go to market early with promising but not-yet-final evidence of the efficacy of a drug with dramatic potential. In the CMS context, such a program could provide for reimbursement of high-value, low-risk incremental changes (like taking a well-characterized in-person program like the DPP and delivering it virtually).

As with FDA, ongoing data gathering would be required, but both patients and innovators would benefit. Moreover, we’d suggest a step further than FDA’s precedent: Continuous, ongoing performance reporting to CMS.

Digital health service providers, in our view, ought to be able to provide ongoing performance data to CMS with little effort. In the bargain, CMS would gain the ability to maintain and perhaps continuously raise outcome-driven performance standards, moving beyond “fee for service.” In addition, by allowing CMS to set performance targets (and reimbursement) based on industry performance data, such a regime has the potential to speed innovation to market on the front-end while spurring free-market driven competition in the long run.5

CMS’s willingness to move forward with a pilot for virtual MDPP providers is both a disappointing delay and, because it is a first for the agency, exciting progress for digital health. In the meantime, private payers continue to adopt reimbursement for DPP services, providing a path forward for innovators. In the initial MDPP pilot, the industry learned a great deal about how to develop evidence and engage CMS, and we’re looking forward to learning from the upcoming pilot as well.

We’re always looking to learn. If you have a perspective on virtual DPP delivery and reimbursement that could help us serve the digital health community, please reach out!


Footnotes

1 The trial also studied the association between metformin use and reduced incidence of diabetes.
2 The CMS Innovation Center was established through Section 3021 of the Affordable Care Act (Obamacare) which amends Section 1115A of the Social Security Act. It enables CMS to develop innovative, cost-effective payment models like the MDPP. In order for CMS to certify a new payment models it must demonstrate lower costs, improved outcomes, or both.
3 See Section (IV) “Eligibility throughout the MDPP services period,” Part (2)(A)
4 A sample of independent meta analyses on DPP programs includes:

  • Meta analysis in 2012 of 28 studies of “real world” DPP-based interventions has concluded they work, even when virtual.
  • Meta analysis in 2014 of 25 studies found a similar but somewhat smaller effect on weight loss (and noted that variability among programs could be reduced).
  • And similar meta analysis of only virtual programs published in 2016 came to similar conclusions as the 2012 study.

5 CMS is almost certainly limited by law from unilaterally implementing such a program. Perhaps we’ll convince Congress to take up the cause in the next round of healthcare reform?