Trouble in Reimbursement City

We’ve been asking the question, “How soon will insurance providers start covering continuous glucose monitoring (CGM) devices?”  But it seems the more accurate inquiry would be, “Why won’t insurance providers ‘reimburse’ for CGM devices any time soon?”Rejected_stamp

There are some serious barriers in the way, as I’ve experienced personally in the last few weeks, submitting multiple claims for my DexCom supplies — and receiving rejection after complicated rejection.  Each “Explanation of Benefits” form from Blue Shield states a different reason for rejecting my CGM-related claims.  Was the diagnosis info not clear enough?  Was the HCPC (Healthcare Common Produce Code) incorrect or missing?  Well, duh, this is new treatment technology, so no codes exist for it yet!  Which is just the trouble, as it turns out.

To address this issue, the national Medicare Coverage Advisory Committee (MCAC) met in Baltimore on August 30.  Despite the fact that the JDRF and Congress are both pushing hard for coverage soon, this meeting didn’t go so well, according to diabetes industry consultant Kelly Close:

“Unfortunately, we felt like we had walked back in time about, oh, a quarter century.  The committee did not make any decisions that will immediately undercut diabetes patients or the companies that sell goods or services to them. But several comments from Medicare panelists raised doubts about the need for one of the most important tools in diabetes management: home glucose monitoring. Frankly, we thought these battles had been won long ago…  Never mind that the DCCT (completed in 1993), plus dozens of other studies since, have provided overwhelming evidence that near-normal blood sugars reduce the risk of complications…  It appears, however, that this memo has not reached some Medicare officials.”

[insert cry of consternation, plus expletive, here]

David Kliff, who observes the D-industry from a financial perspective on his Diabetic Investor site and newsletter, explains the obstacles to CGM coverage this way:

Healthcutbills_1 “The economic reality of diabetes management centers on two important factors, time and money…  Some companies see the role of the educator and/or physician as being the knowledge enabler. With this approach, the patient merely gathers the information which is then transmitted to their educator or physician for analysis. (But) will the physician and/or educator have the time to analyze all this data and, if they do, how will they be compensated for their time?  Third-party payers are already reluctant to pay for patient education; it’s a wonder that so many companies believe they will pay for data analysis.”

The trouble with all these advanced glucose monitoring tools (and “smart” pumps and the like), Kliff explains, is that they provide almost too much information — too much for the patient to utilize unassisted, anyway.  Think up to 288 glucose readings a day, and complicated trend graphs that require a trained eye to interpret.

In covering the AADE (American Assoc. of Diabetes Educators) conference this summer, Kliff noted: “This is the hidden issue with CGM, as valuable as this data is especially when combined with insulin and food intake, analyzing the data is a time-consuming process. In a clinical setting when patients have the assistance of physicians and educators, this isn’t an issue. However, in the real world all this information can be overwhelming…  Based on the products displayed on the AADE exhibit floor, it’s quite clear that increasing patient knowledge has taken a back seat to gathering even more information.”

So we’ve got two principal barriers here:

1) officials’ lack of understanding about the value of new glucose-monitoring technology to patients’ long-term health (and thus long-term healthcare costs)

2) the costly time required for professionals to analyze CGM data; Who will pay?

So far, it seems only a few “extreme” patient cases have managed to garner coverage for CGM technology  — where the patient suffered severe hypoglycemia unawareness and was landing in the ER every other week.  There, the insurance provider could see an immediate, short-term cost savings in their own interest.  The long-term for us “middle of the road” patients is going to be a much harder sell. 


16 Responses

  1. TorMo
    TorMo September 20, 2006 at 7:41 am | | Reply

    “The trouble with all these advanced is that they provide almost too much information — too much for the patient to utilize unassisted, anyway. Think up to 288 glucose readings a day…”

    By that reasoning, we should take speedometers out of cars. How can drivers be expected to interpret all that information while driving at the same time?

    The point is that you quickly learn to spot trends and unusual readings; you ignore the normal readings and only act on the problemmatical ones.

  2. Living With Diabetes
    Living With Diabetes September 20, 2006 at 7:56 am | | Reply

    CGMS Coverage?

    Diabetes Mine: Trouble in Reimbursement City Diabetes Mine writes about it: We’ve been asking the question, “How soon will insurance providers start covering continuous glucose monitoring (CGM) devices?” But it seems the more accurate inquiry would be,…

  3. CB
    CB September 20, 2006 at 10:21 am | | Reply

    As usual, you’ve raised an important point about the next level of personal D care – and one now having a “chicken and egg” quality to it. However, it’s not a new issue. The issue of “too much” data has always been present as technological advances continue for D care (and, for that matter, almost any type of medical care). The same so-called “concerns” by care providers was made when the first “personal” BG monitors were marketed. Now we have CGMs that provide far more data to users (but how much more information and insight?). As TorMo states in his/her comment, it may not be necessary to individually review all 288 data points/day, but in order to find those critical inflection points, trends and unusual readings, you really need some easy-to-use, accurate software. This is why CGM vendors sell the software to us – it’s a necessary ingredient to make such technology valuable for our care. And an important income stream for them.
    For some time, my professional experience focused on how to create useful insights from reams of data collected by all sorts of devices. It always requires clever, easy-to-use analysis software that can give the CGM user (say, you or me) the ability to “see the forest through the trees”. It’s difficult enough to capture the essence of what our home BG monitors are telling us; how many Ds are able/willing to be disciplined enough to assess between 50 and 70 times as much data PER DAY from a CGM (depending on how often you test yourself)? In part, this is why I developed my own Excel spreadsheet to help me get clear about what my “simple” OneTouch monitor’s data was trying to tell me.
    In addition, our doctors and care-givers must sign-on and agree that reviewing such detailed CGM information about our BG levels is worth their time and effort. This is an open question, given their all too hectic schedules.

  4. Bernard Farrell
    Bernard Farrell September 20, 2006 at 12:01 pm | | Reply


    I’m a little surprised at your assumption that folks with diabetes can’t use this much information. Many folks can read temperature or stock charts without any problems.

    Yes, there will be folks for whom this is too much. These are probably the same folk who don’t believe in multiple daily injections, or pumps, or testing more than a few times a day.

    But for those of us who want the information so we can control things more effectively, this data is CRITICAL. We don’t need to pay someone to interpret the results for us, we’ll do it ourselves.

    I agree with CB’s comment about the need for good software. I think this will only become possible when meter manufacturers agree on a data standard for the readings themselves. This is the theme of my recent post ( ) about microformats.

    Once the data format is standardized, I’d guarantee that folks will start producing software to graph it in useable and useful ways.

  5. AmyT
    AmyT September 20, 2006 at 12:08 pm | | Reply

    Oh my, Bernard, that wasn’t MY assumption; I am quoting various experts in this post.

    As you know, I am an early adopter of CGMS, and am not the least bit overwhelmed by the data.

  6. TorMo
    TorMo September 20, 2006 at 1:05 pm | | Reply

    It was David Kliff who made the remarks I was criticising. It was clear that Amy was just reporting them.

    Incidentally, although I work in IT myself, I’m not sold on the idea that we need software to interpret the results. Entering all the details of what you ate and how much exercise you got seems like too much of an overhead. That’s largely why I gave up on tracking stuff in Excel.

    I think that with a meter that reports results every five minutes, most users will quickly learn to make appropriate adjustments simply through experience and guesstimation. Software might be nice for historical analysis, but for most people, just knowing what’s going on RIGHT NOW will be of primary importance.

  7. Kassie
    Kassie September 20, 2006 at 5:02 pm | | Reply

    Re TorMo’s comment:

    “I think that with a meter that reports results every five minutes, most users will quickly learn to make appropriate adjustments simply through experience and guesstimation. Software might be nice for historical analysis, but for most people, just knowing what’s going on RIGHT NOW will be of primary importance.”

    I think this is what appeals to me – the potential of empowering the person with diabetes to make adjustments that apply to the exact situation that’s occurring.

    How many times have you looked at historical data and tried to remember what exactly was different about that day, week, meal, activity, etc that might have ’caused’ a blood sugar? Historical data is interesting and certainly useful in spotting large trends, but I think people with diabetes can and will benefit from being able to see the road in front of them as they are driving.

  8. Bertie
    Bertie September 20, 2006 at 5:52 pm | | Reply

    Have I missed something here about CGM (Constant Glucose Monitoring) devices? The one I have read about states that you should verify any highs or lows with a finger stick using a manual glucose monitor. Specifically this in regards to the new Medisense inslin pump with a monitoring system. It also recommended doing at least four checks a day using a manual meter. This strikes me as two steps forward and four back?

  9. Becky
    Becky September 20, 2006 at 8:38 pm | | Reply

    I have a comment about the MCAC meeting Kelley Close reported on addressing Medicare’s coverage of CGMS. They were specifically concerned with continuous monitoring of people with type 2 diabetes, since the majority of diabetic Medicare recipients have type 2. It seems from reading the panel questions that the Committee doesn’t know what the heck type 1 diabetes is.

    For example:
    Question 2. Although the largest segments of the Medicare population are >65 and/or have type 2 diabetes, CMS is interested in the evidence base for newer technologies for glucose monitoring, including continuous monitoring of interstitial fluid and their role in type 1 diabetes. How confident are you that the effectiveness of continuous monitoring in patients with Type 1 diabetes is best assessed by changes in HbA1c, concomitant hypoglycemia rates, hypoglycemia-related falls, post-operative morbidity, wound-healing, and weight?

    They also added quality of life to the final version. Unfortunately, some of the panelists were confident that weight and quality of life were good assessments of the effectiveness of continuous monitoring. YIKES. This will definitely be an uphill battle.

    Oh, and I’m officially tired of waiting for Navigator and contacted the Dexcom rep this week. My order is in!

  10. JasonJayhawk
    JasonJayhawk September 21, 2006 at 12:38 am | | Reply

    I’d like to know if the MCAC participants have anyone in their family with Type 1 diabetes. I think it would only be fair to include at least one Type 1 on the committee!

    In twenty years from now, people will look back and think how barbaric life must have been without CGMS, just as we look back twenty years ago on SMBG policies. We just have to keep the pressure on as new technology comes to us… can you imagine testing your glucose 6 or 10 times a day with the Ames Reflectance Meter?!

  11. mcityrk
    mcityrk September 21, 2006 at 2:21 am | | Reply

    Hi All-

    I remember seeing the first comments via David Kliff about how poorly this meeting went and felt that it was a real blow to the prospects for CGM reimbursement in the near term. However, on second thought, it would seem that a Medicare advisory committe is hardly the target audience for companies attempting to get insurance reimbursement primarily for Type 1 CGM. It just happened to be one of the first publicized opportunities for the companies involved to hone their presentation on the advantages to patient care and the resultant cost benefits to insurers. Before you completely dispair, give these companies some time and see if they get a chance to meet with and convert the real movers and shakers in Type 1 diabetes care reimbrsement over the next six to twelve months.

  12. art-sweet
    art-sweet September 21, 2006 at 6:19 am | | Reply

    I think this opens up a broader debate about what it means to provide quality care for people with type I. I would, if not kill, seriously maim, to be able to find an endo/CDE who would:

    - trust that I’m trying and give me positive reinforcement to try harder
    - look at trends, rather than alarmedly pointing out the random 350 that, who the hell knew sushi had so many carbs in it? I corrected. It’s over. Yes, I know high blood sugars are Bad.
    - return my phone calls promptly and not delegate to a nurse who then has to go and ask the doctor and call me back AGAIN
    - I know this is probably pie in the sky, but my ideal endo model would be a scheduled ten minute phone consultation every two or three weeks to review uploaded data. Far more useful than an every three month checkup that lasts half an hour. Pluses: I get to stay on schedule. The doctor gets to stay on schedule. We’re not looking at ancient history.

  13. Jenny
    Jenny September 21, 2006 at 1:10 pm | | Reply

    We have received insurance coverage for the Medtronic CGMS and sensors. I think committees would benefit by knowing that there are quite a few cases already that have obtained coverage for these devices. (I have heard of several others.)

  14. AmyT
    AmyT September 21, 2006 at 3:28 pm | | Reply

    Funny,I hadn’t thought about the single-minded focus on Type 2′s here. But it’s kind of shocking that the committee wasn’t provided an understanding of the need in the Type 1 community…

  15. Kevin
    Kevin September 24, 2006 at 3:27 am | | Reply

    The problem I see is currently these devices must be replaced yearly due to the battery. That is AT LEAST a $500 replacement cost. Then sensors are $35 each and have to be replaced every 3 days (insurance must be able to pay for 3 day changes.)

    That above total is $4770. Take into account say a 20 year old with type 1 going to 60 years old. That is $190,800 in coverage there. There are very few complications that even get near that cost in diagnosis and treatment.

    That is the problem with insurance companies right now. They have to factor in which costs less. Remember, they already lose money on diabetics with insurance supplies, so a CGMS is just another item to costs more money.

    While I think it would be great if they covered CGMS, they have to look at where the benefit to them is, not to mention, they rarely cover first generation devices. Insulin pumps weren’t covered for about 10-15 years after they initially came out. Same applied to home BG meters.

  16. Jane
    Jane March 7, 2008 at 7:16 pm | | Reply

    Too much data is a lame excuse. Why couldn’t the data be analyzed by a computer program?

Leave a Reply