UnitedHealth and Dexcom’s unprecedented pilot: CGM and coaching in 10,000 type 2s! Ramping over next 6-9 months; could save “billions” – November 30, 2017

Executive Highlights

  • UnitedHealth Group’s Investor Conference announced plans for an unprecedented pilot to use Dexcom CGM in ~10,000 people with type 2 diabetes! Wow – what a victory for the company and for the field. Dexcom’s EVP Steve Pacelli said separately at the Piper Jaffray conference that the program will ramp in the next six to nine months, followed by a broader rollout over the next several years. We assume it will use G5, but aren’t positive.
  • Clinicians will make house calls to help participants set up their CGM, coaches will counsel patients remotely on lifestyle and medication, and PCPs will also receive glucose data. Meal and (possibly) activity data will also be tracked, providing coaches the ability to recommend dietary changes, walking after meals, etc. We’ll be fascinated to see what kind of outcomes this program can deliver and whether it can scale. Presumably United saw strong potential for cost savings and positive ROI, based on early programs with Dexcom. One area could be reducing medications; as well, some patients may need more or intensified medications that could lead to lower costs longer term.
  • Details were not shared on location and source of patients, outcomes, financial terms, devices to be used, etc.
  • United’s EVP of R&D Dr. Deneen Vojta is optimistic that the program could “literally” save the system “billions of dollars and many, many lives.” This is the biggest ever private US payer vote of confidence in CGM in type 2 diabetes – at least that we can recall hearing. She’s a big deal and very smart and we love hearing this sentiment from her. We expect positive health and cost outcomes in this pilot could lead to wider adoption and reimbursement of CGM in type 2 diabetes.

At UnitedHealth Group’s 2017 Investor Conference, EVP of R&D Dr. Deneen Vojta announced an unprecedented partnership with Dexcom to pilot CGM in “nearly 10,000” type 2 patients “in the next few months.” Wow!

Dr. Vojta shared that clinicians will make house calls to help participants set up their CGM, patients will be able to track their own data on a smartphone, United will receive glucose data wirelessly so that its own coaches can counsel patients, and patients’ PCPs will be looped in as well. Dr. Vojta said she is extremely optimistic about this project moving forward, suggesting that “in total, this literally could save the system billions of dollars and many, many lives.” This is the biggest ever private US payer vote of confidence in CGM in type 2 diabetes – at least that we can recall.

The next day, at the Piper Jaffray Healthcare Conference, Dexcom’s Mr. Steve Pacelli more specifically said that the initial 10,000-patient study will probably ramp up in the next six to nine months, and then roll out more broadly over the next several years – “you should assume this is kind of a first step partnership here.” He added the goal of the program is to get data – including meal and activity (he thinks, which would make sense, because United already has a partnership with Fitbit in the Motion program) – to patients and clinicians in order to facilitate behavior change and the modification of drug regimens. He also cited some early work with United, where CGM+coaching has been able to “get these people to get up and walk around the block after a meal,” reduce A1c, reduce medications, and confer “huge” cost savings. We have not seen these preliminary data but certainly hope to in 2018.  

Additional details on the program, such as device to be used (G5? G6? Verily gen one?), location, outcomes tracked, and financial terms were not discussed. Since CGM will be used, we hope it can help validate the benefits of improving time in-range and reducing hypoglycemia, independent of A1c. As for finances, Mr. Pacelli commented that “we’ve always said the business models will be different” and that “UHC (United Healthcare) has been very public about constantly trying to take on more risk” – this implies that there is a risk-sharing, outcomes-based component to payment, in line with Dexcom’s previous comments.

This is clearly an enormous victory for Dexcom and potential reimbursement for CGM in type 2 diabetes. If the pilot goes well, it could serve as a strong demonstration project on the power of combining real-time glucose data with coaching in type 2 diabetes. In addition to United, it might even encourage other payers and providers to use CGM in type 2.

This also brings another win for Dexcom as it seeks to push into type 2 diabetes, following Medicare coverage (for type 1 and type 2) and the Verily partnership.

Scroll down for more, including transcripts from both conferences and our questions. Dexcom told us it is not yet ready to go into more detail, but we expect to hear more at JPM in January and in the 4Q17 call.

  • When characterizing the program, Dr. Vojta gave the example of a coach recommending that the patient take a walk after a meal or modify diet slightly and tracking the effects on blood glucose, or even working with the patient to optimize medication dose and timing. We wholeheartedly agree – CGM truly excels at giving this kind of real-time feedback for any patient with glucose fluctuations. We love the coaching philosophy, as presented, because it avoids the normative and simply encourages and facilitates “n of 1” experimentation to push toward personal lifestyle and medication strategies that work for the individual.
  • One unanswerable question at this point is: Can it scale? 10,000 is huge for a device pilot or trial, but considering that United is the largest private payer in the US, we have to imagine there are plenty of coaches on board that can handle the data, do house calls, etc.
  • We’ll be fascinated to see how payers like UnitedHealth work with patients’ own providers – since these are United’s own coaches, how will PCPs be looped in? Will they be overwhelmed or excited? What incentives and responsibility lines are created in this type of pilot? How will Dexcom vs. UnitedHealth vs. providers take on responsibility for medical advice?
  • Is this a positive for Medtronic and Abbott as they seek to push CGM into type 2 diabetes? What kind of pilots or clinical trials are possible with all three CGMs in type 2 diabetes?
  • We’re noticing more high-profile payer/industry pilots with diabetes technology – most recently, Blue Cross Blue Shield and Onduo and Innovation Health (Aetna), Sanofi, One Drop, and Common Sensing. It is outstanding to see payers finally buying into diabetes technology and more service-oriented models that combine humans with connected devices. One open question is pilot size – small scale experiments allow stakeholders to stay nimble and invest heavily only in the solutions that seem to work, though this strength is also their weakness; scale is always a question mark, because that which works in a small pilot may not necessarily translate to large-scale implementation. How do they strike a balance?


Relevant Transcript from UnitedHealth Group Annual Investor Conference

Dr. Deneen Vojta (EVP, R&D, UnitedHealth Group): We are also unlocking the power of data and technology that now allows patients to see directly how nutrition and exercise are powerful medications; how nutrition and exercise directly affect their need for pharmaceuticals.

People living with diabetes need to check their glucose level frequently, because if blood sugar levels get too high or too low, this leads to serious health concerns. The most current testing method includes taking blood from a finger stick every day, often multiple times a day.

UnitedHealth Group is now partnering with Dexcom to advance a smarter, less invasive way of continually monitoring blood glucose throughout the day with a sensor that goes just beneath the skin on the abdomen. In the next few months, Optum plans on enrolling nearly 10,000 participants in a pilot of this approach.

House call clinicians will visit participants in their home and teach them how to set up and use their continuous glucose monitor. Participants will use their smartphones to see real-time glucose reading from the sensor and whether those levels are rising or falling. It will be dramatically clear to people.

Additionally, UnitedHealth Group will receive all this information wirelessly, so that our coaches can counsel participants on how best to use their information. The coaches, for example, may recommend that a participant take a walk after a meal and see how that exercise directly influences their glucose level or they may recommend a slight modification to their diet and see how that stabilizes their glucose throughout the day, and they will work with participants to optimize their medication. So when is the best time of day to take a daily medication?

All that data will also be fed back to the participants' primary care physician, so they have a more holistic view of the patient and can create a better care plan. This represents a digital therapeutic. We are using data and exercise and potentially reducing the need for higher level pharmaceuticals.

In total, this literally could save the system billions of dollars and many, many lives.

Relevant Transcript from Piper Jaffray Healthcare Conference

Mr. Matthew Oliver O'Brien (Sr. Research Analyst, Piper Jaffray Companies): I guess, just for starters, given it's obviously timely, what's going on with UnitedHealth? Can you just give us a little bit of an update about the partnership that they talked about at their Analyst Meeting? And how should we think about that evolving?

Mr. Steven Pacelli (EVP of Strategy & Corporate Development, Dexcom): For those of you that didn't catch it yesterday, what Matt's referring to is United Healthcare during their Annual Investor Conference was talking about some novel ideas around treatment for type 2 diabetes, obviously, an enormous disease state. We've long spoken about how Dexcom anticipates participating in a number of different fashions within the type 2 space. You should assume this is kind of a first step partnership there.

We've always said the business models will be different. This is going to be more of a diagnostic, program-type solution, where we'll combine Dexcom CGM data with fitness data, I think, carbohydrate data, meal data and really the goal of these programs will be to get the patients the data, get the clinicians the data, and really try to help the patient modify their behavior, potentially help the physician modify their drug regimens. If you can prevent that patient from going on to next most expensive drug or you can actually – we've seen examples in some of our early work with UHC, where we're able to, using the data generated by the CGM and some coaching, get these people to get up and walk around the block after a meal, for example. We can actually reduce the A1c. We can actually reduce some of the drugs that they're on and the cost savings are just huge to the system. So pretty exciting stuff. Still very early. I think they said in the keynote remarks that we're going to start with an initial 10,000-patient study and that will happen over the next probably six to nine months; start to generate that data and then, assuming all goes well, continue to roll it out over the next several years.

Mr. Oliver O'Brien: Have you any sense for how many patients that they have with diabetes? They have, what, 20 million, 40 million lives?

Mr. Pacelli: I mean, tons, right? I don't know the specific number. UHC has been very public about constantly trying to take on more risk, quite frankly. I think they would love to take on more and more of the diabetes, particularly in the type 2 space.

Close Concerns Questions

Q: Where will this pilot take place? Are patients recruited through employers? Health plans? Providers? Aside from type 2 diabetes, are there other inclusion/exclusion criteria?

Q: How will CGM be used – 24/7? Every month? Every three months?

Q: How scalable is this program, which includes house calls, coaching, and PCP engagement? How many patients per coach?

Q: Is the coaching model more reactive (i.e. intervene when someone is trending out of range) or more general education/Q&A-based? What’s the coaching interface – chatbot/text, video-based, etc.? 

Q: What are the financial terms of the pilot? Is Dexcom paid by the sensor, by the outcome, or some of each? Will providers be reimbursed for reviewing data or engaging with patients remotely? How will UnitedHealth take on risk here?

Q: Which outcomes will be tracked, and what threshold cost savings/health improvements will Optum/UnitedHealth be looking for in deciding whether or not to expand the program? 

Q: Have there been smaller Dexcom-UnitedHealth pilots? If so, what were the results from those? What gave UnitedHealth the confidence to move ahead?

Q: How many coaches does Optum have on staff? How many will be deployed in this pilot? Are they diabetes specialists, or are they trained in multiple chronic diseases?

Q: Would UnitedHealth consider doing a similar pilot in prediabetes?

Q: Which Dexcom device is being used? Will Optum leverage other technologies in this pilot (i.e. dose capture, insulin titration, etc.)?

Q: Is this a positive for Abbott and Medtronic?

Q: To what extent is this a threat for the diabetes drug manufacturers? How much will CGM dramatically reduce medications in type 2 diabetes?

Q: What has been learned from the Fitbit partnership in the Motion Program? Could UnitedHealth financially incentivize patients to wear CGM or keep their glucose in range (e.g., premium rebates, $3/day, etc.)?

Q: Will providers get decision support and analytics from UnitedHealth Group or Dexcom? Where is the line between coaches/insurers managing patients’ diabetes medications vs. providers?

Q: How many lives with diabetes does UnitedHealth cover?


-- by Brian Levine, Adam Brown, and Kelly Close