Finally (because I've been waiting for it), Dexcom received FDA approval for both it's 7-Day continuous glucose monitoring system and it's next generation software to go with it. Though it will be a while longer before most of us see it.
The company press release doesn't provide much additional details besides the fact that the new sensors are approved for seven days of use.
But there's also a product announcement webcast from Las Vegas. It's over 39 minutes long, and does not come with any additional slides, though there were slides shown during the presentation.
In the remainder of this long blog post, I've summarized what I picked up from listening to the webcast.
Webcast Summary
In the webcast Andy Rasdal, CEO of Dexcom says that there were no conditions from the FDA on the approval. Dexcom plans to have a limited launch of the new product offerings (presumably both hardware and software) at the end of this quarter with a full launch in the 3rd quarter.
He says that Dexcom has eliminated the use of a shower patch. The needle size has also been reduced, so it should be more comfortable to insert (yay!). Mr. Rasdal claims that Dexcom has listened to their users so features we have asked for will be included. He mentioned that the price is likely to be increased, though the price per day is effectively reduced. (For those of us already getting more than 7 days out of a sensor, it just may end up being more expensive.)
Andy Rasdal also talked about studies that show improved outcomes (over 7 day periods) when using the Dexcom system. This shows a 40% reduction in time spent hypoglycemia, a 33% reduction in hyperglycemia, and a 24% increase in the time spent euglycemic (normal blood sugar levels, also called normoglycemia).
The study looked at folks with less control and found that within a few days they could effectively double the time they spent in euglycemia. These were people for whom it was believed that CGM would not provide any benefits.
The new Dexcom data manager supports downloads of data to the PC for more analysis and for communicating to doctors. I guess we'll see what this means when I can get my hands on it. I still haven't succeeded in cracking the Dexcom data format, and this alone would be a big improvement
Dexcom has 3 peer-reviewed studies published in various journals. During the webcast he referenced a new study on long term outcomes to be published in the June issue of Diabetes Technology and Therapeutics
Dexcom emphasized that this is the second generation of the Dexcom technology with more coming in the future.
Towards the end of the presentation (31 minutes in) he mentioned the new CMS codes for CGM, and the importance of working hard on better insurance reimbursement. He mentioned five letters from Congress that endorsed CGM and looking for better coverage (I wonder whether these could be obtained under Freedom of Information). And that Dexcom has a dedicated reimbursement team, but that reimbursement is still spotty.
Interestingly Andy also talked about the needs of intensive patient monitoring post surgery. This is to move the use of Dexcom technology from just ambulatory to in-hospital use. There is 'compelling clinical evidence' to show intensive management reduced mortality and morbidity by 50%, and not all of these folks (only 13%) had diabetes. I think he mentioned a Dr. Vandenburg who had done these studies.
Dexcom sees this as a large market opportunity. There are over 110,000 critical care beds and Dexcom estimates that over 25% of hospitals have implemented intensive regimes with 12-24 finger sticks per day.And that at $3-$5 per stick, the CGM systems might be more cost effective, as well as less stressful to the patient.
This is also a more focused and effective distribution channel instead of selling to individual patients. That would clearly reduce the cost of sales for Dexcom and presumably increase profitability. Dexcom is also building in the capabilities to interface with any monitoring systems.
They see partnering with other devices makers as being significant in the future. He mentioned not just combining with an insulin pump, but perhaps also with an insulin pen. This would mean several million new users, over the several thousand already using insulin pumps.
My Final Thoughts
I think this will put more competitive pressure on Medtronic Minimed, and may also have implication for the (as yet unapproved) FreeStyle Navigator. I'm excited for the folks at Dexcom, but I'll reserve judgment on the new system until I get one in my hands.
Note: I currently own some shares in Dexcom, I work not to let this ownership influence what I'll say about the company or its products.
...there is no current consensus on how to compare CGM devices, how to define the appropriate level of agreement given the time lag between blood and interstitial fluid levels, or how to display and interpret data. The guideline being co-developed by CLSI and DTS will address these issues by reaching consensus on how CGM data should be presented and compared between different devices and different glucose methodologies.
I've been trying to promote a standard just to represent the data values themselves. This agreement seems to go far beyond this.
It will take some time and hard work to agree on this guideline but I see it as an important development in the acceptance of CGM technology.
How would you feel if your blood sugar went low, and later you found yourself being prosecuted afterwards?
In a nutshell, this is what's happening to Doug Burns (Mr. Universe). Despite support from many folks, the district attorney seems determine to prosecute him.
Update:June 4. The district attorney has dropped this case and Doug Burns will not be prosecuted. Thanks all who participated in phoning the DA involved.
I'm supporting Amy Tenderich's call to action about this. Please read her blog post about the issue. Then call both the telephone numbers listed and leave a message. A long message.
My suggestion is to actually call both numbers and leave messages each and every day until the district attorney drops this case. Thanks Amy, for getting this going.
This isn't a big surprise to me, but a new study that will be published in Diabetes Care shows that children with Type 1 diabetes who watch more TV have poorer control.
My guess would be that's because they lose the opportunity for extra activity, and probably snack some more, I guess I'll have to read the full report when it's available.
The numbers are pretty amazing. The number of hours per day of TV watched, is tied to the average A1c for the children in the study.
Adam Bosworth gave a speech at the 2007 American Medical Association of Informatics (AMIA) Congress.
In the speech he talked about three 'core principles' for future healthcare systems:
Discovery: we should be able find information that we care about
Action: We should have access to personalized services so we can get the excellent health support
Community: We should be able to form a community with those who have similar conditions and from their health practitioners.
His blog post provides the notes from the speech. When I read these, this comment early in the speech struck a note with me:
This vision for the future of health care starts with the premise that consumers should own their own total personal health and wellness data ... and that only consumers, not insurers, not government, not employers, and not even doctors, but only consumers, should have complete control over how it is used. [My emphasis]
As you may already know, I've tried to do a number of things so that I can get access to my data. Despite repeated attempts, Dexcom still does not let me get to the data behind their pretty graphs. And I can't export it to another program.
And for the other diabetes devices, I can see the results as the software maker intends, but I can't export it. The OneTouch software locks the database with a password, as does the AgaMatrix Zero-Click software. OneTouch does have an export facility, but it's the only one of four software packages I've tried that does this.
If I can't take my data out of their system, I can't combine it and manipulate it. I can't easily share it with my healthcare team. It's unusable except within the software package. It's basically trapped there.
Remember Adam Bosworth saw ownership of my data as the premise, or assumption, on which his vision is based. If I can't access it easily, I don't own it.
As a software architect I understand the importance of raw data. You can't build a software system unless you have data that can be shared by all the pieces. This is the main reason for my paper promising data format for blood glucose meters, and my creation of the Diabetes Data wiki.
If you'd like to get involved and help me, I'd love to hear from you.
Since Amy posted her review of this meter, I thought you might be interested in seeing it in action.
So here's a short video of the meter being used to take a reading.
More about this meter next week.
Update: The meter is made by a New England company called AgaMatrix. I don't know how many insurance companies are covering it. I really like the styling overall.
I was talking with someone in support at Dexcom last week. We were discussing some fluctuations in my numbers and how to control these. And he offered to send me this guide that might help.
It arrived in the mail yesterday, and it's 38-pages about how to use CGM effectively to control your blood sugars. It's written by Dr. Steven Edelman (from TCOYD) and Dr. Timothy Bailey (co-author of Using Insulin). Dexcom supported this book by providing a grant, and there are graphs taken from the Dexcom STS system throughout the book.
From reading it, I think that anyone using any CGM system will benefit from reading this short book.
After the Introduction, there are three main sections:
Blood Glucose Levels on the Rise
Blood Glucose Levels on the Way Down
Blood Glucose Levels During the Night
The book has a total of ten scenarios. Each one describes a situation that happened with details about insulin dosage, food eaten, etc., and then shows a graph of what happened to BG levels. You then get to answer a question choosing from 3-4 answers per question.
The book helped me think about how to best interpret what was happening and decide what alternate actions I could take to improve a given situation.
Now here's the best news. You don't have to be a Dexcom customer to get this book. You can get a PDF version here courtesy of the Taking Control of your Diabetes site.
I was just thinking about how my Dexcom CGM is more than just something that buzzes annoyingly when my blood sugar is going high or dropping low.
To me it's a bit more like a time machine.
Ever since I first got diabetes I would hear or read about how "you have to control your blood sugar levels or..." something terrible will happen.
But here's the problem. The something terrible doesn't happen immediately.
Let's say I go to a church potluck (always causes eating problems for me) and overeat. My blood sugars go high and stay high for a few hours. It's not like I see this or there's any immediate issues. And no-one, including me, can really tell anything happened. Not by looking at me.
The only immediate problem is some level of drowsiness. Unless the numbers run high for some time, I really won't be able to see any immediate issues. And then, by the time I do see issues with my eyes, or kidneys, or heart, or something else it's almost too late.
But this is where my time machine comes in. When I'm wearing the Dexcom, I can see those highs. And knowing what they're doing to me, I really can see the future.
If I have a day where my blood glucose goes to, say, 230, and it takes me a few hours to get it down, I feel those highs in a very different way. I almost imagine the possible future problems and (for me) it's an encouragement to get things more under control.
Don't get me wrong. Seeing the future is more of a pain than a benefit. When I have days where my numbers don't get above 180, that feels good. But the swings and the long highs are hard to take.
I'm still learning how to use this information for better control. And I'm still learning to accept the (WAY TOO LOUD) alarms the Dexcom gives. (Can you hear me, Dexcom folks).
But I'm also starting to get somewhat better control. My total daily dosage has gone down a little, and there are less profound swings.
I thank God for this small blessing, and I'm looking forward to the next generation of devices that will keep me healthy until something better comes along. (Hint: it's a four letter word, C**e).
What might this mean? Well here's the problem for those of us with diabetes.
How many diabetes data devices do you use? Glucose meters? Continuous glucose meters? Insulin pumps? Each of these typically collects some amount of data related to your diabetes treatment.
Have any of you got an easy way to collect this data into a single piece of software to that you can do some analysis on it?
Now imagine a future where all of the various pieces of software that come with your devices each exports the data in a common format. What might that mean?
If that format was shared and freely useable by anyone, then it allows different people to create software to collect the data and make logbooks out of it, or analyze it in different ways.
To get device makers to think about this, I've sent e-mails to various companies that are currently working on new blood glucose monitoring devices. Companies like Biosign Technologies who are developing a new wrist-worn monitor; and Nilimed who is working on a new continuous glucose monitor. I'm also trying to interest the Artificial Pancreas project in supporting standard data formats as part of that effort.
Would you like to get involved? If you've got any experience with programming, especially how to use XML and process XML, then you may be able to help.
Leave a comment, or send me an e-mail (mail address on the wiki), and I'll drop you a note.
On Tuesday the Wall Street Journal had a very interesting article on cutting copays(preview only). It's worth looking for it in your local library.
Apparently some employers and health care companies are trying to reduce the long-term costs for those with chronic diseases. And now they believe that if they provide the necessary drugs for maintenance of those diseases with zero copay amounts, then folks might actually take the medicine.
Well duh. I guess that's why the executives of these companies get the big bucks.
According to the Journal "employers and health plans have targeted conditions like diabetes and heart disease in part because chronic illnesses are major drivers of the over rise in health-care costs". The emphasis on diabetes is mine.
Now my guess is that they're really talking about Type 2 diabetes. Of course I still hold out hope that they'll do the same for Type 1 diabetes and actually cover the technology that we need to maintain our blood glucose levels and reduce long-term complications.
Let's call this article a glimmer of light at the end of the tunnel.
Today is the second day of Mix07. There's a lot more people, and much more buzz than yesterday.
Twitter is being used to send updates out to folks, and enable impromptu sessions and meeting among folks.
As a result of twitter I got a chance to meet with Scott Hanselman and talk with him some about CGM devices and the challenges of handling different data formats from these data collection devices. And Scott did a great job summarizing the announcements from day one of Mix.
Mix07 seems to be much more around rich media and media delivery to you via desktop apps, web browsers, or mobile devices. For those of us who are building enterprise applications, the rich media stuff is not immediately of interest. But we're also seeing a lot more tools around developing UIs for Silverlight. And once you're on that platform - the world is yours (or so Microsoft would have you believe).
At lunch I was chatting with the Microsoft Program Manager responsible for the MSDN user interface. She showed me how they've used Starlight to add videos to the front page, and how they've improved the search experience. Nice job, and it sounds as if more is coming soon to MSDN.
My biggest complaint so far? Too many interesting sessions on at the same time: I guess that's a good complaint. Having the videos on the Mix07 site will help, but it's not really the same. And as of right now (late on Tuesday) very few of the videos are actually there.
Tomorrow I'm looking forward to a session on Amazon Web Services and much more on Silverlight.
Update: (August 19, 2007) I'm now using the latest Dexcom, the Dexcom SEVEN. The software that comes with the Dexcom SEVEN (Data Manager 2) has much better support for statistics and it allows you to export the readings in two different formats. This software also works with existing Dexcom 3 systems. Given this solves the problem for me, I'm not doing any more work in trying to understand their crazy data format!
Still no success getting data from my Dexcom STS CGMS. But my recent challenge post about decoding the Dexcom data format seems to have drawn a lot of attention.
Several people have jumped in to have a try at cracking the problem. I sent some e-mails to Dexcom technical support and customer support about this issue. So far, I've received no response. On June 5th I had a conversation with Dexcom which indicated they aren't about to open up the data even in their new version of the software.
I loaded up another Dexcom R2Downloads data file. From looking carefully at the graph in the Dexcom software it seems that there must be at least four types of records in the data stream. These are:
Start of use. With date (MM/DD/YY) and time (HH:MM:SS) values.
Gap in use. With time (HH:MM:SS) value.
Sensor reading. Glucose value, date (MM/DD/YY) and time (HH:MM) values.
Meter reading. Glucose value, date (MM/DD/YY) and time (HH:MM) values.
You can now find on my site two example data files, plus a hand-captured text file containing the values that are probably at the start of one of these data files.
How You Can Help
Maybe you know something about encoding or encryption. These skills may be enough to crack the problem.
If you can't do this, perhaps you can help in an indirect way. And I'll admit this is a little devious. :-)
If you search on Google for Dexcom my blog comes in around the 15th position in the results. I think I might get more attention from Dexcom if my blog came in as the number 1 search term for Dexcom on Google.
Impossible you say? Well, maybe not.
Here's where you come in. Can you post a blog entry on your blog which points to this blog post, and has Dexcom as the word in the link? Maybe write a note explaining that I'm trying to crack this problem for the Dexcom, highlight the Dexcom word and then add http://www.bernardfarrell.com/blog/2007/05/decoding-dexcom-data-format-update.htm as the hyperlink.
This way I'll get the word out about this challenge, plus Dexcom marketing may wake up and realize that this is important to many folks.
Can you tell I'm a little impatient to get to my data? How do you feel about a system that captures data that's important to you, but that doesn't let you get to it? Am I the only one who feels this strongly about this?
Update June 7th 2007: As of today, many people have linked to this post. When you Google search for Dexcom, this post is now about number 8 in the results, which puts me on the first page of Google results for Dexcom. I'm have still to write a letter to Jorge Valdes, VP Engineering, at Dexcom asking him to release the data format so that we can use it to access our data. First I want to hear what they say on the June 11th conference call.
If you'd like to help me, please post about this issue (setting our data free) and link to this blog entry using the word Dexcom, as I've just done here. Thanks.
Note: I currently own shares in Dexcom, I work not to let this ownership influence what I say about the company or its products.
The Diabetes Technology Blog is focused on using technology to life better with diabetes. I review: blood glucose monitors; continuous glucose monitor; blood sugar meters; diabetes software and living with diabetes.
About Me
Name: Bernard Farrell
Location: Massachusetts, United States
I was born in Ireland and now live in the US.
I have had Type 1 diabetes for over 36 years. I struggle with my blood sugar, the same as most people with diabetes.
I wear a Cozmo 1800 insulin pump and a Dexcom SEVEN CGM to track my blood glucose levels.
I'm blessed by God, and every day brings the possibility of a cure.
