UnitedHealthcare's Medtronic Deal Sparks Furor, but a Year Later, Innovation Continues
When UnitedHealthcare (UHC) announced in May 2016 that it would only cover insulin pumps made by Medtronic for most adults, patient advocates decried the news as both a major loss for those affected and, quite possibly, a harbinger of a world where exclusivity pacts stifled competitive innovation. Investors, it seemed, were in full agreement. Shares of smaller pump maker Tandem Diabetes Care immediately dropped 20% when news of the deal became public.1
Such predictions, however, may prove overly dramatic. The past 16 months have seen Medtronic sign an outcomes-based payment agreement with another insurer, but no insurer has followed UHC’s lead and signed exclusivity pacts with Medtronic or any other pump maker. If UHC patients are protesting the policy, they have been less visible in recent months. The exclusivity deal helped inspire a campaign by JDRF (formerly the Juvenile Diabetes Research Foundation),2 but the early media coverage has waned.
As for the deal’s impact on UHC patients, outsiders are forced to guess. UHC has declined to disclose any figures related to the policy: how many of its policyholders use insulin pumps, what pumps they chose before the transition, or how many have switched to Medtronic after preferred pumps broke or were out of warranty. An estimated 1.25 million3 (0.4%) of the nation’s 325 million people4 have type 1 diabetes (T1D). Estimates of the percentage of American patients with T1D who use a pump vary, but the JDRF currently reports the figure at 40%.5 Roughly 75% of all pump users have a Medtronic pump. The vast majority have T1D, although a very small number have type 2 diabetes.6
Still, it’s hard to get a precise handle on the number of patients affected by the UHC decision, since the policy excludes Medicare Advantage patients, those younger than 18, and enrollees of UHC Sierra Life and Life Commercial.7 Based on UHC’s combined commercial and Medicaid population of 36.5 million, the number of people who might be asked to switch pumps is likely between 10,000 and 12,000.8
Eventually, nearly all adult Medicaid and commercial UHC patients who refuse to buy other pump brands out-of-pocket will have to switch, either to a Medtronic pump or to insulin injections. Those who used their benefits to buy a non-Medtronic pump before the policy took effect can still get coverage for supplies so long as the pump keeps working and remains under warranty. Warranties typically last 4 years, so the transition probably has yet to affect more than one-third of UHC patients who prefer non-Medtronic pumps. Such numbers only tell part of the story. It is impossible to know the strength of patient preferences overall and for individual patients.
Michele Hynes, who is covered by UHC, switched from Medtronic to Animas more than 4 years ago because the Animas pump was waterproof (as some Medtronic models now are) and because it had a remote control. “As a woman, I often don’t have access to pockets if I am wearing a skirt or dress. As such, in these cases, I generally clip my pump to my bra. It is SO convenient to be able to dose from my meter and not have to leave the table or grab my pump out of my bra,” she wrote to Evidence-Based Diabetes Management™ (EBDM™) in an email.
An even bigger deal for Hynes is the sensor. The 41-year-old mother of 2 young children wants an integrated system that automatically delivers insulin based on readings from a continuous glucose monitor (CGM), but she wants a system that works with a Dexcom sensor. If she moved to a Medtronic pump, the only way to automate delivery would be to move to a Medtronic sensor as well.
“In general, I have found that anyone who has used Dexcom is less than impressed with Medtronic sensors. And as such, I am hesitant to switch—to the point I am still using an out-of-warranty Animas pump,” she wrote. “I will probably concede defeat and order a new pump in the next 2 months.”
Looking beyond patient preferences, the other important metrics for evaluating the policy are, obviously, health outcomes, total cost savings, and how any savings are distributed among UHC, premium payers, and pump-using patients. UHC, again, declined to disclose any details.
“Patient safety, service, and cost were key considerations in our decision-making process to partner with Medtronic,” wrote UHC spokeswoman Kristen A. Hellmer in an emailed response to EBDM™ to requests for specific numbers. “We have finalized a value-based agreement, which rewards improved outcomes for UnitedHealthcare members on insulin pumps and places greater focus on quality rather than the volume of care delivered.”
Some observers have argued that patients might suffer worse outcomes after being pushed away from first-choice machines and toward devices they found less intuitive or less suited to their needs,9 but there’s no trial data connecting current Medtronic pumps to inferior outcomes. There is, on the other hand, solid evidence tying Medtronic’s most advanced pump system to superior outcomes.
Medtronic’s MiniMed 670G, the first system to receive the FDA designation of “artificial pancreas,”10 became available to US patients in June. It combines a pump with a CGM and software to automate many dosing decisions. Even in manual mode, it automatically suspends insulin delivery when blood glucose levels fall dangerously low (and resume it when they rise). In auto mode, the system uses CGM readings to adjust insulin delivery levels every 60 seconds. It’s nowhere near perfect, but trial data strongly suggest that the automation—which is unique to Medtronic’s system—significantly improves outcomes. Among 124 patients in a 3-month trial, auto mode reduced the following: average glycated hemoglobin (A1C) from 7.4% to 6.9%, dangerous hypoglycemia by 40%, and reduced time spent with excessive blood glucose by 11%.11
“Although we are in favor of more, rather than less, patient choice, there’s good reason to believe that this UnitedHealthcare policy doesn’t change that much negatively. It’s likely that most UnitedHealthcare pumpers already use Medtronic, that those pumpers will be among the earlier ones to upgrade, and that outcomes will be as good or better with most that are in line to shift from another brand. Medtronic has shown that this product will improve outcomes, that’s what matters to payers in this era of value-based healthcare, and there’s good reason to think that UHC’s policy will deliver on outcomes,” said diaTribe Foundation founder Kelly L. Close, who advised patients who don’t want Medtronic to raise their voices, especially on social media.
“Assuming it’s true that Medtronic systems improve outcomes, the question would then become, which is more important: patient choice or better outcomes? The math would say better outcomes, but I also doubt that deep down, Medtronic or UnitedHealthcare wants this controversy,” Close said in an interview. There is no definitive proof that the 670G produces superior results in real-world settings.
Obviously, Medtronic has not had time to complete a postmarketing study required by the FDA. Until it does, the only indicator of efficacy comes via anecdotes from early adopters and testers, which have generally been positive. Close spends much of her time talking to people about medical devices for T1D, and she says she rarely hears a negative review from anyone who has lived with the 670G. Even the people who are nervous about surrendering dosing control to an algorithm, people who are predisposed to be cautious about turning control over to a system, or even to dislike the system, usually end up telling her that it doses them better than they can dose themselves.
Medtronic, moreover, appears confident that the 670G will improve real-world outcomes and save money for payers. In June, Medtronic announced a deal with Aetna that ties payment rates for its systems to the results those systems produce in patients who switch over from insulin injections.12 The companies declined to disclose the financial details of the deal, but Medtronic said that the achievement of A1C targets will be the first metric used to determine payments. Other metrics, such as reductions of hypoglycemic episodes, could be added later.13
“This agreement is an important first step as we look to broadening our partnership to facilitate patient access to the most advanced diabetes management solutions across the care continuum that not only ensure outcomes, but lower the overall cost of care for this chronic and burdensome disease,” said Suzanne Winter, vice president of the company’s Americas Diabetes Group, in the press release that announced the deal.14
“We recognize the important role that health plans make in coverage, affordability, and choice for type 1 diabetes therapies,” wrote JDRF spokesman Christopher Rucas in an email about the group’s opposition to exclusivity deals. “We have put together a campaign we’re calling Coverage2Control, which will call on insurance companies to provide members with diabetes 3 things to give them control.”
Rucas outlined the elements:
1. Predictable and reasonable out-of-pocket costs for insulin and diabetes management tools.
2. Freedom to choose an insulin pump that makes sense for the individual, which means no exclusive agreements with pump makers that limit choice.
3. Coverage of all life-saving devices, including artificial pancreas technology.
Rucas said the JDRF is “engaging directly with health plans, but also wants to provide the T1D community with opportunities to take action.” The JDRF and other patient advocates believe that widespread coverage for all FDA-approved pumps benefits patients not only by letting them choose whatever current machine works best for them, but also by spreading revenue enough to encourage competition and innovation. There is a widespread belief among both patient advocates and financial analysts that at least 1 smaller player, possibly more, could soon drop out of the market.
Tandem Diabetes Care lost nearly $24 million in the first quarter of 2017 as pump shipments fell 30% to 2816.15 The company lost more than $83 million in 2016 and more than $72 million in 2015.16 Johnson & Johnson has said it is considering selling its diabetes care business, including Animas, a pump-making subsidiary that’s based in suburban Philadelphia.17 This comes after the Animas deal won FDA approval in December 2016 for the OneTouch Vibe insulin pump, the first to integrate with the Dexcom G5 CGM.18 Also in the mix is Insulet, which has struggled to convince Medicare to offer reimbursement for its tubeless Omnipod pump.19 And in August, the remaining patients using pumps from Roche Diabetes Care were turned over to Medtronic for ongoing support after Roche announced plans to stop selling its Accu-Chek pumps in the United States.20
Such a situation would not be historically unique. The number of pump makers has waxed and waned considerably over the past 40 years. New entrants have kept coming in with new technology, trying to dethrone MiniMed, the dominant player purchased by Medtronic for $3.7 billion in 2001.21 None have succeeded, but their efforts have helped fuel a steady improvement in pump technology. Pumps have gotten considerably smaller, more discrete, smarter, easier to use, and more reliable since they first appeared in the 1970s.22 There is reason to believe that more players will enter the market with new ideas. For example, Bigfoot Biomedical is performing clinical trials on a system that includes a pump that has no controls on its body because it is controlled through a smart phone app.23
Even if fewer new companies enter the market and the pace of traditional pump development slows, patients will see differences. Today’s models are good enough that observers like Close see little need for dramatic hardware improvement, regardless of how competitive the pump market is. Mechanical improvement will continue, she thinks, but more of the value will come from the brains of the pump.
“Insulin is an incredibly dangerous drug, and a machine will always figure out dosing better than patients or parents or partners. Automation is where we’re going, and it’s all about the algorithms and who can make the easiest-to-use systems—take away steps, make things easier to use and smarter. That could come from the largest pump manufacturer, but it doesn’t necessarily have to. It can come from another company or academic or even hobbyist who wants to write it,” said Close, who gave the interview while testing a system that used software written by 1 individual to control her insulin.
“I can’t wait to try the 670G, and I can’t wait until it’s widely available. For now, having a smart algorithm in a box that costs $126 and keeps me between 80 and 120 all night every night is pretty excellent. Medtronic raised the bar with its new pump and the outcomes that they have proven (mainly less hypoglcyemia)—and they shouldn’t, and won’t, take their eye off the ball because the field will keep raising the bar ever higher.”