The author, who has lived with type 1 diabetes for more than 50 years, shares his account of a successful appeal of Medicare's policy of refusing to pay for continuous glucose monitoring (CGM).
Although most commercial insurers now cover continuous glucose monitoring (CGM) devices, payment for this technology is routinely denied for an important and growing group with diabetes: Medicare beneficiaries.1,2 This occurs despite the May 2015 statement from leading professional societies of endocrinologists that such denials are “imprudent” for patients on intensive insulin regimens with a history of hypoglycemic events or frequent hospitalization.3 While appealing for CGM coverage is not easy, some persons with type 1 diabetes (T1D) do try. Evidence-Based Diabetes Management asked Dan Patrick, an Ohio Medicare beneficiary, who this spring, became a 50-year Joslin Medalist,4 to share his story.
CMS has taken a profound, deeply troubling, and dangerous, clinically unsupported therapeutic direction3 through a local policy which effectively blocks the reimbursement for a CGM under a Medicare rule. Through an arduous appeal process, I managed to win coverage for my CGM. Here is my story.
This spring, I reached the 50-year mark of living with T1D. For most of that time, I lived without CGM, but that changed more than 3 years ago. In 1999, our family adopted a puppy named Maggie, and for 10 years, she served as my sleeping service dog. Maggie would lick my face to wake me in advance of a dangerous hypoglycemic event, and the necessary corrections were made. In the fall of 2009, Maggie was diagnosed with T1D, which is relatively common in female dogs as they age.5
For the next 3.5 years, I became her caregiver and never really slept to be prepared for hypo events. After Maggie passed, it was time for a different CGM model. Starting in January 2013, I began using a CGM system. I had not yet turned 65 when I started using CGM, so it was covered by commercial insurance.
In 2014, when I reached 48 years as a patient with T1D, I had become unable to discern the onset of hypoglycemic events and was diagnosed with a condition called hypoglycemia unawareness.6 That same year, I turned 65 on May 4, 2014, and became eligible for Medicare effective May 1, 2014. At the time of my appeal, CMS’ policy for CGM coverage was spelled out in the Local Coverage Article: Glucose Monitors—Policy Article—January 2014 (A47238). (This article has since been removed from the CMS website, but a copy is available as an electronic appendix.7) In 2014, and in compliance with Medicare rules, I enrolled. There are clearly defined parameters for enrollment, with financial penalties the patient for noncompliance.8
My Medicare enrollment was also required by my personal healthcare coverage, which changed from a private non-Medicare plan to a Medicare Advantage PPO plan (MAP). At this point, all medical information had to be refiled, even though I was still with the same insurance company and had the same identification number. At first it seemed that I would receive CGM under Medicare without any problems: this particular insurance company went so far as to suggest choices of medical pharmacies to supply the 3 required CGM components, and I chose an in-network pharmacy to reduce costs. A box of 4 CGM sensors arrived every 28 days, starting in May 2014.
The bottom fell out, in August 2014, with the reorder of a CGM transmitter. The reason given when Medicare refused to pay was that the transmitter was not covered under my MAP insurance plan, in compliance with Medicare rule (A47238).7
It’s worth noting that I had no trouble receiving approval from the same insurer for use of a CGM in December 2012, which I began using it in January 2013. The approval process, under my private non-Medicare plan, had only involved processing by my endocrinologist. The necessary CGM components were all covered outside Medicare. What changed, before and after enrolling in Medicare, that a lifesaving medical piece of equipment was not necessary to manage my T1D condition? The answer was, and still is, nothing.
I have been an active viewer of discussions and information regarding various denials by Medicare via the Diabetes Online Community (DOC). Denials for Medicare CGM reimbursement are increasing, creating serious negative impacts on families of persons with T1D. These continued actions have prompted introduction of federal legislation to overrule prior individual CGM Medicare denials.1,2
Members of the DOC have come together, and a Diabetes Patient Advocacy Coalition (DPAC) has been initiated, making it easy to keep policy makers’ attention on people with diabetes.9 While federal legislation was introduced to overrule prior CGM Medicare denials in 2013, there was no action in the 113th Congress (2013-2014).10 The 114th Congress (2015-2016) reintroduced the legislation; the current bills are S.804 in the Senate and H.R. 1427 in the House, with 273 co-sponsors. No action had been taken at press time.11
Given the lack of Congressional action, this became a personal fight for my life. In 2014, after living with T1D for 48 years, I realized that it would be imperative to prepare objective documentation to support an appeal to CMS. At the start of my appeal, there was no real guidance on how to proceed for approval of a CGM.
Denial of coverage for the transmitter came in a letter, which arrived in October 2014 and clearly referenced the specific Local Policy (A47238). It is interesting that the MAP made no reference to any communications with my endocrinologist and considered my initial appeal to be from the patient only. How is it that the endocrinologist was removed from the discussion regarding my life with T1D? The blatant disregard of my medical condition, without consulting the specialist who handles my care, made no sense. I decided to assemble the documentation for review and consult with my endocrinologist for this appeal. This action continued through the first level of appeal with the MAP, and the second level would be with the independent reviewer of Medicare claims, Maximus Federal Services.
Maximus’ mission is very specific: they are appeals experts. Medicare retains Maximus to review files and decide on the accuracy of a Medicare Advantage health plan’s decision—Maximus works for Medicare, not the plan.12 What was frustrating at this stage was that Maximus’ charge was not to look at the facts—which is what my physician and I assembled—but to look at the narrow question of what constituted Medicare policy, and whether the health plan had applied it.
To me, and to all persons with T1D, facts matter. My professional and personal life has a clear focus on measured data, to document my progress in striving to obtain a goal at hand. How can I clearly and conclusively document how CGM achieve a personal outcome of near normal glucose levels with minimized hypoglycemic events? The challenge for all T1Ds is to reduce hypoglycemic events, especially at night. Needless to say, Maximus ruled against allowing coverage, and the next step—Level 3—would be before an administrative law judge.
Interestingly, to address the fears of parents of children with diabetes, a new tool has been introduced by Dexcom: Dexcom G5, which has been approved by the FDA for children as young as age 2.13 Thus, young children have an approved tool to manage their diabetes, but those over age 65 are denied Medicare coverage for CGMs. How absurd. My CGM had already proved to be an assistance in dealing with nighttime hypoglycemic events. The need for evidence to overturn a policy requires a review of the policy to present cogent points against it.
I obtained a printout of Local Policy (A47238). The article reads: Local Coverage Article: Glucose Monitors — Policy Article – January 2014 (A47238). Under the section, Non-Medical Necessary Coverage and Payment Rules, the following line has been added: Continuous glucose monitors (A9276-A9278) are considered precautionary and therefore non-covered under the DME benefit.7
A9276-A9278 are the billing codes for sensors, the transmitter, and the receiver. DME stands for Durable Medical Equipment. The Local Coverage Article applies to 7 states: Illinois, Indiana, Kentucky, Michigan, Minnesota, Ohio, and Wisconsin.
The actions taken by MAP raise serious questions regarding the use of age and geographical factors. The most thought-provoking and compelling question becomes: on what basis is the Local Coverage Article an absolute, offering no ability for a successful appeal for this lifesaving monitoring system? Failed appeals, thus far, outnumber permanent successes.
It is important in an appeal process to clearly define the terms. Medicare did not define or describe CGMs. My choice for a CGM was a Dexcom product. The Dexcom G4 Platinum Continuous Glucose Monitoring System was approved by FDA’s Medical Devices section on October 5, 2012. The FDA wrote the following under: What will it (Dexcom G4 Platinum) accomplish:14
“People with diabetes can use the tracking and trending glucose information to help understand patterns in their glucose levels, and be alerted when glucose values are approaching potentially dangerous high (hyperglycemic) and/or dangerous low (hypoglycemic) levels. When used along with a blood glucose meter to obtain a more accurate reading of actual blood glucose levels, the Dexcom G4 PLATINUM can also help people with diabetes make long-term adjustments to their treatment plan to keep blood glucose levels in a safe range.”
The most troubling word in this short Local Article is “precautionary," which is the basis for Medicare denying CGM coverage as a DME. This is a statutory term, and CMS bases its interpretation on FDA labeling that calls for a self-monitoring blood glucose test with CGM before an insulin dose. But this narrow reading flies in face of evidence and expert opinion. While the term “precautionary” draws a blank via an internet search under medical definitions, it takes on a totally different meaning with requirements when defined as a precautionary principle. This is a precautionary approach, and the objective is to set policy that is not harmful; the burden of proof rests with Medicare.
Can Medicare supply sound evidence with clinical consensus that removing reimbursement coverage for T1D using CGM will not cause any harm to the individual, their family, the community, and the public? CMS cannot. The website Children With Diabetes, offers an excellent product discussion on CGMs that includes a list of more than 40 links that offer support for the importance of CGM, including clinical studies.16
The journal Endocrine Practice reported a study, in December 2014, which found that patients older than 65 years achieve a significant and durable improvement in glycemic control when using CGM, comparable with younger patients. In fact, the reduction in severe hypoglycemia may be of particular benefit to older patients.17 And in May 2015, after a review of the evidence, the American Association of Clinical Endocrinologists and the American of College of Endocrinology issued a joint statement calling it “imprudent” for payers to deny coverage for high-risk patients with intense insulin regimens, a history of hypoglycemia or hospitalizations, or other risk factors.3
The Local Policy declared that a CGM is a non-durable equipment (DME). However, the definition for DME has been, and is defined by Social Security, as “equipment which can withstand repeated use” and is used primarily for a medical purpose.18 How does one reconcile the Social Security rule with this attempt by CMS to totally change the definition of DME? What other historical facts are available that could assist in my appeal?
The Diabetes Control and Complications Trial (DCCT) is the seminal clinical trial that began in 1983. This study settled the question that keeping glucose levels as close to normal as possible with intensive insulin therapy, with glycated hemoglobin (A1C) equal to or less than 7%, would stave off complications when compared with conventional insulin therapy that maintains A1C at or greater than 9%. However, the DCCT found a 2- to 3-fold increase in hypoglycemic events for individuals utilizing an intensive insulin therapy due to a lack of tools.19
Edward Damiano, PhD, professor of biomedical engineering at Boston University, has a succinct discussion on the challenges to manage T1D with insulin in the first-half of a YouTube presentation.20 Insulin is a miraculous substance; while it regulates glucose levels in the blood, it also has a high toxicity index and a narrow therapeutic range. A slight overdose of insulin can prove lethal. For a person with T1D, the greatest fear is having severe hypos at night—this keeps us awake at night. Insulin allows those diagnosed with T1D the ability to survive, but we suffer long-term complications, armed with the knowledge gained in the DCCT study.21
Since CGMs did not exist during the DDCT study, the case can be made that a CGM is a new technical tool for T1Ds to manage our condition. My personal opinion is that I have been successful in decreasing severe hypoglycemia during the night, which is the deadliest period of the day for a T1D. The fear of severe hypoglycemia is documented in the multiple surveys by Children with Diabetes.22 A CGM brings a visual representation of glucose trending levels. One particular finding of the DCCT was that it took up to 3 years to substantiate the success of intensive insulin therapy. Using Dexcom Studio software, I pulled data from my first months of using a Dexcom G4 Platinum CGM, in January 2013. Next, I assembled data from the same period 2 years later, in January 2015, making a comparison guided by the standard set in the DCCT study.
The 4 figures (FIGURES 2-5) below present compelling data to support my appeal for approval for CGM coverage by Medicare. FIGURE 2 shows data from the first months of my use of a Dexcom G4 Platinum CGM. The time frame is for a 90-day period. The patient was only able to obtain a normal glycemic level for 34.2% of the given time period. Non-normal glycemic time was 65.7%.
FIGURE 3 covers a 90-day period 24 months later, beginning in January 2015. I obtained a normal glycemic level for 41.5% of the time period. Non-normal glycemic time was 58.5%. During the time frame presented in FIGURE 3, my A1C was 7.0%. It is interesting to note that a grade of 41.5% is not a passing grade and does not indicate exemplary success in the given outcome.
FIGURE 4 displays an average time in minutes at various glucose levels for a 24-hour day. A given 24-hour day has 1440 minutes. A Dexcom G4 CGM can record a total of 288 reading every 5 minutes in a 24-hour day. The denial of the usage of a CGM would mean total reliance upon blood glucose meter readings of 10 per day. The increase in measured data points by a CGM facilitates a greater ability to initiate micro adjustments to avoid either hyper- or hypoglycemic events.
FIGURE 5 compares the Dexcom glucose readings for 90-day periods at the start of 2013 and 2015. The most telling point is the fact that I obtained a reduction in the percentage of time in hypoglycemic events. The denial of coverage for a CGM by Medicare could lead to a 2- to 3-fold increase in hypoglycemic events as documented in the DCCT, without the use of a CGM.
As my case progressed, another working its way through the system had a direct impact on mine. On May 26, 2015, the United States District Court, Eastern District, Wisconsin, in the case Whitcomb v. Burwell, Case No. 13-cv-990, ruled in favor of Jill A. Whitcomb to have coverage for a CGM. Her appeal began in 2012. She was represented by Debra Parrish and Robert Theine. Even though UnitedHealth continued to fight the ruling and deny Whitcomb payment, her successful resolution at the 5th appeal level was responsible for favorable rulings in many other cases—including my appeal, which ended with a successful resolution at the Medicare Administrative Law Judge, Level 3 appeal.23,24
On July 29, 2015, a Medicare Administrative Law Judge (ALJ) ruled that my CGM is medically reasonable and necessary as part of my diabetic care. Contained within the ruling was precedential information, listed below, which were cornerstone determinations to my successful appeal at Level 3. The ALJ made the ruling retroactive to July 2014. What is the outlook for Medicare coverage in the wake of the Whitcomb case, and mine? Unfortunately, it’s not good. According to Mike Hoskins, a journalist with T1D who writes for DiabetesMine, the outlook for coverage is getting worse, not better.25 Hoskins has encouraged the DOC to be vocal about the situation with Medicare.
Here are 3 key points I have learned about Medicare and CGM coverage since my appeal:
1. The Local Article should not have been used as a reason to deny coverage for CGM. This was a key finding of Whitcomb.24
2. CMS, and lawmakers, must start to take into account the cost of hypoglycemic events among those with T1D in making coverage decisions. A groundbreaking study in The American Journal of Managed Care, found that following an hypoglycemic event the average cost of an inpatient admission was $17,564, the average cost of an emergency department visit was $1387, and the average cost of an outpatient visit was $394.26 Now that measurement of hospital readmission is a cornerstone of value-based reimbursement, Medicare payment for CGM seems, not only clinically necessary, but also imperative.
3. Finally, HHS Secretary, Sylvia Mathews Burwell, did not file an appeal within the statutory deadline with the US District Court in Whitcomb. According to information from her caregiver, Dan Kraft, the federal court’s ruling recognizes that government’s refusal to cover a CGM for the management of Whitcomb’s diabetes was not supported in law or fact. Simply put, the local article is not singular reason for denial for CGM coverage by the MAP and Medicare. The doctor-patient relationship was brought back into the equation.
The T1D community owes a debt of gratitude to Jill Whitcomb, her caregiver, Kraft, and her legal team. She traversed a complete 5-step Medicare appeal and won. The length of time for her appeal is obscene, and such delays in the process need to be examined and changed as quickly as possible.
On the question of payment, Medicare is a secondary payer. This Local Policy can be viewed as an attempt to prevent appeals, and it complicates the process by not reim-bursing the MAP for the coverage of a CGM for any T1Ds. If the MAP knows it will not be paid, it will not want to cover an individual’s claim for CGM coverage. However, there is a serious challenge which plagues T1Ds. We need help to overcome the long-term complications, which is a plague to our morbidity and dramatically shortens our lifespans. Medicare’s attempt to reduce short-term costs will further exasperate the issue. The real cost factors with diabetes are the necessary services and taking care of complications resulting from a lack of knowledge or tools. A CGM is another step in responding to the challenges for a T1D to manage our condition and live a healthier complete life with near-normal glucose levels, and striving to minimize hypoglycemic events and no severe hypoglycemia.
Was my appeal the end of the story? Sadly, no. In March of 2016, I received a notice that once again, the MAP is denying payment for a replacement transmitter. I’ve sent a letter asking how this can happen when I have a ruling from Medicare Administrative Law Judge: my story highlights the need for Congress to act—and act now—to compel CMS to give T1D persons in Medicare access to CGM technology that makes sense clinically and is cost-effective. Our lives depend on it. REFERENCES
1. Collins SM. Medicare’s failure to cover CGM at odds with other health, research agencies. Am J Manag Care. 2015;21(SP11):SP378.
2. CGM Medicare Coverage. JDRF website. http://jdrf.org/take-action/advocacy/cgm-medicarecoverage/.Accessed April 25, 2016.
3. Grunberger G, Bailey T, Camacho PM, et al. Proceedings from the American Association of Clinical Endocrinologists and the American College of Endocrinology Consensus Conference on Glucose Monitoring. Endocr Pract. 2015;21(5):522-533.
4. Smith A. Big questions about diabetes lead to answers from those who have lived with the disease the longest. Am J Manag Care. 2015; 21(SP15):SP517, SP524-SP525.
5. Diabetes in dogs: symptoms and treatment. WebMD website. http://pets.webmd.com/dogs/guide/dog-diabetes-symptoms-treatment. Accessed April 24, 2016.
6. Common terms. American Diabetes Association website. http://www.diabetes.org/diabetesbasics/common-terms/common-terms-f-k.html. Updated April 7, 2014. Accessed April 24, 2016.
7. CMS Local Coverage Article, (A47238). January 1, 2014. PDF available as e-appendix.
8. Medicare eligibility and enrollment. WebMD website. http://www.webmd.com/health-insurance/insurance-basics/medicare-eligibility-and-enrollment. Accessed April 24, 2016.
9. Diabetes Patient Advocacy Coalition website. http://diabetespac.org/. Accessed April 24, 2016.
10. Caffrey MK. Getting CGM covered by Medicare Still on the Agenda. The American Journal of Managed Care website. http://www.ajmc.com/focus-of-the-week/0115/getting-cgm-covered-bymedicare-still-on-the-agenda. Published January 2, 2015. Accessed April 24, 2016.
11. Congress.gov. https://www.congress.gov/search?q=%7B%22source%22%3A%22legislation%22%2C%22search%22%3A%22CGM%22%7D. Accessed April 26, 2016.
12. “Appeals” language from Maximus website. http://www.maximus.com/federal-appeals. Accessed April 25, 2016.
13. FDA approves Dexcom G5 mobile continuous glucose monitoring system [press release]. San Diego, CA: PRNewswire; August 25,2015. http://www.prnewswire.com/news-releases/fdaapproves-dexcom-g5-mobile-continuous-glucose-monitoring-system-300132661.html.
14. US FDA approves the Dexcom G4™ PLATINUM Continuous Glucose Monitor (CGM) [press release]. San Diego, CA: Business Wire; October 8, 2012. http://investor.shareholder.com/ dexcom/releasedetail.cfm?ReleaseID=71175.
15. What is continuous glucose monitoring? Vimeo website. https://vimeo.com/101342418. Accessed February 29, 2016.
16. Continuous glucose monitoring. Children with Diabetes website. http://www.childrenwithdiabetes.com/d_06_e00.htm. Accessed February 29, 2016.
17. Argento N, Nakamura K. Personal real-time continuous glucose monitoring in patients 65 years and older. Endocr Pract. 2014; 20(12):1297-1302. doi:10.4158/EP14017.OR.
18. Program Operations Manual System. https://secure.ssa.gov/poms.nsf/lnx/0600610200. Accessed February 29, 2016.
19. Summary of the Diabetes Control and Complications Trial / Epidemiology of Diabetes Interventions and Complications. NIH website. http://www.ncbi.nlm.nih.gov/projects/gap/cgi-bin/Get- Pdf.cgi?id=phd000390. Updated November 9, 2005. Accessed February 29, 2016.
20. The bionic pancreas. Edward Damiano, PhD. TEDx Sacramento. https://www.youtube.com/watch?v=bZXmfTxd79Q. Published July 14, 2015. Accessed February 29, 2016.
21. The Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression on long-term complications in insulin-dependent diabetes mellitus. N Engl J Med. 1993;329:977-986. doi:10.1056/NEJM199309303291401.
22. Poll results: at what time of day are you most worried about hypoglycemia? Children with Diabetes website. http://www.childrenwithdiabetes.com/poll/poll20141029.htm. Published November 5, 2014. Accessed April 25, 2016.
23. Brown P. Judge: CMS must cover CGM for T1D patient. MedPage Today website. www.medpagetoday.com/PublicHealthPolicy/Medicare/54785. Published November 19, 2015. Accessed April 25, 2016.
24. Caffrey MK. Deadline nears for decision on Medicare CGM coverage. The American Journal of Managed Care website. http://www.ajmc.com/newsroom/deadline-nears-for-decision-onmedicare-cgm-coverage. Published August 25, 2015. Accessed April 24, 2016.
25. Hoskins M. Medicare’s lackluster diabetes coverage: let’s make some noise. DiabetesMine (Healthline) website. http://www.healthline.com/diabetesmine/medicares-lackluster-diabetescoverage-lets-make-some-noise. Published March 24, 2014. Accessed February 29, 2016.
26. Quilliam BJ, Simeone JC, Ozbay B, Kogut SJ. The incidence and costs of hypoglycemia in type 2 diabetes. Am J Manag Care. 2011;17(10):673-680