Or more simply put, I won’t be the idiot dad calling home to ask “What’s for dinner?” right in the middle of what is most likely a site change or some form of trouble shooting the insulin pump, etc…

I just thought I would take a minute to share what it’s like to have real-time automated intelligence within the family dynamic of type 1 diabetes. I don’t know too many endocrinologists who would understand why I would want to get a real-time glucose alert (via email as shown above or on my cell phone which is my primary DI interface) when diabetes is managed by trends.

Yet, providers that work with patients in our trials are thrilled to receive automated blood sugar trending reports based on algorithms that define high risk. While these reports are infrequent, they are reliable and improve efficiency. That’s been missing from the world of self-reported patient data for years.

In this new world of patient-centric technologies, we need to realize how many players are on the patient’s team and that the needs of each team member are different. With automated diabetes management systems (ADMS), everyone gets what they ordered and none of what they don’t want or need.

I wonder what’s for dinner?

Bennet’s recent post asking for the Holy Grail in diabetes technology (and I’m totally with you on that cure thing, too) at his blog got me to thinking about a few things this evening. Mainly that his wish list, which is shared by just about everyone in diabetes land, is exactly in line with what we’ve been working on here at Diabetech. I can’t believe it’s already 2007 closing in on 2008! Always pushing to get things done faster but life seems to move ahead at its own pace.

Personally, my experience and my earliest investigations into the pace of innovation and the business models of the incumbents led me to believe that this Holy Grail would be a long time coming if ever and I had some insight and passion to contribute to the mission. So, I drew up a plan, redrew it and redrew it. Like they say in entrepreneurial circles, “If you want to make God laugh, show him your business plan”. Keeping the goal (or Grail in this case) in mind is key as I mentioned in a recent article in Medical Products Outsourcing magazine.

2001

May 21, 2001 - My 27 month old daughter was diagnosed with type 1 diabetes. With a professional background in wireless devices, wireless networks, telco infrastructure, software integration and hospital information systems I had a few ideas about how far behind the times current diabetes technology was and a few ways to make improvements.

2002

During the Spring of 2002, I had a call with the medical device team at Eli Lilly to discuss a comprehensive design for an automated diabetes management system capable of delivering the right information, at the right time to the right person… on a silver platter. Following the call I was told that this device and system were the closest thing to the Holy Grail they had ever heard of.

The notion at the time was that day to day management of diabetes was already a lot of work. Adding tasks for the patient with diabetes and his team would only be seen as extra work and few people would be willing to do anything beyond what they were already busy doing. In the world of insulin dependent diabetes this means multiple daily blood glucose checks, insulin injections or pump management and counting carbohydrates for every meal and snack. Tracking this data and relevant comments can be a lot of work not to mention doing something with this data including trend analysis, trend identification and translation into a change in regimen across a virtual team.

We knew we needed to get as much of this data as possible by automatically connecting to diabetes devices like glucose meters, insulin pumps and maybe even insulin pens via a wireless module since this technique would allow us to automate the collection process while also improving accuracy over any kind of manual self-reporting method (e.g. - web, phone, fax…). We also knew back then that additional sensor data would be interesting for automatic collection of data like temperature, activity levels, heart rate, blood pressure, etc… and that the implementation of the device would require a new package in a form factor that would allow it to not be seen as extra work. Messaging would be a very important part of our system but it would also have a customized dashboard which allows every member of a patient-centric team (including the patient) to tailor the system to be whatever they want it to look like while focusing on being easy and fast to use.

December 20, 2002 - Original clinical trial of the GlucoMON® as part of the world’s first End-to-End Wireless Diabetes Management System. This trial must have been the first if not one of the first examples of how you use glycemic variability to assess risk across a large patient population… automatically! Who knew then that SD and GV would become such a buzz in diabetes care.

2003

In early 2003, Diabetech led an effort to standardize diabetes data formatting and sharing called Diabet-ML. We borrowed the Diabet-ML name from Scott Hanselman - one of the early pioneers in diabetes data management software. Bernard Farrell has recently picked up the torch on this one via his microformats and diabetes data wiki initiatives.

Clinical trials continued throughout the year focused on usability and efficient methods of interaction - simplicity.

2004

In 2004, Diabetech’s GlucoMON helped us to earn the 2004 Niche Patient Monitoring Competitor of the Year Award from industry analyst firm Frost & Sullivan. They loved the automated nature of the GlucoMON and its unique ability to connect the diabetes patient with their remote caregiver(s) in real-time without any extra work by the patient (ie - no cell phone, computer, cables, Internet, or button pushing required).

Also in 2004, we announced GlucoDYNAMIX version 1.0 as a comprehensive hosted software service including wireless airtime, data collection, data storage, analysis and feedback with device support for GlucoMON (incorporating data from the Ultra) and published a paper describing our vision as the Real-Time Virtual-Loop.

Here’s your 24hr time slot report!

2005

2005 was the year we kicked off research into Real-Time Virtual Patients; an investigation of real-time two-way communication on the way to developing personalized predictive algorithms for accelerating the realization of what we call the remote control artificial pancreas.

We also began developing several detailed clinical protocols that we like to call Intensive Management Protocols or IMPs. Given all of the people and the data and the devices and all of the decision points and different ways to present data, request data, etc… we figured out the best (only) way to make sense of it all is to package and test defined configurations. For example, the system works one way for a newly diagnosed child with type 1 and in a very different way for an adult with type 2 and still differently for a teenager with type 1 starting insulin pump therapy.

We also began the design of GeNI - Glucose Nanobiosensor Implant. Our design incorporates the smallest implantable cgm on a chip (1mm x 1mm) with short range wireless communication to the handheld mentor which also includes long range wireless supporting remote connectivity, automation, GPS, software updates, etc…

2006

Several of these protocols (including Pancreatic Islet Cell Transplant Screening & Monitoring with Baylor Research Institute) are currently in progress with patients throughout the US delivering some very exciting outcomes while others are closed with data published recently at the ADA Scientific Sessions. We also began getting more phone calls from large companies like AT&T and the big diabetes companies wanting to collaborate. We also began implementation of a USDA federal grant for our technology to support a regional diabetes care network with Driscoll Children’s Hospital based in Corpus Christi, Texas, serving kids with type 1 and type 2 diabetes throughout South Texas.

Fast forward to 2007

Here we are knocking on the door of 2008 and planning for commercialization based on years of experience involving patient-centric design, development, testing and retesting… The time is coming finally for more of that Grail thing from Diabetech for more people and we are prepared to continue our leadership in helping to make your wishes a reality. If you share Bennet’s wish and mine for the Holy Grail, why not give us your vote in reply to this Monster post and help spread the word about the Diabetech® technology. I always thought I would write a book about this experience but I didn’t think I would start writing it tonight.

Thanks Bennet.

I was at ADA recently and there was a heightened presence of A1c testing as a tool in the diabetes kit this year. Everyone assumes that a measurement of A1c = a measurement of historical glucose which is not entirely correct. Further, hemoglobin makeup has the potential to change every few years so it’s not even measurable on the same scale within the same person unless you get your A1c on a frequent basis.

There was an initiative announced that is trying to finalize an international standard which takes % hemoglobin A1c from any test method and recalculates it into mean blood glucose along the same lines as what you see on your blood glucose meter. The thought being that patients don’t understand the A1c number and that they do understand the meter based scale. A bigger problem is the presumption that you can simply change from presenting patients with a % A1c vs. a formula translation into mean blood glucose. A1c is only a measure of the glucose left over within a small fraction of your hemoglobin A and then only within the 1c and not 1d, etc… assuming that everyone has the same hemoglobin signature…which is entirely not true. The most common forms of A1c testing (DCA2000 in office and laboratory venipunctures) cannot detect these variants and therefore the oh, so often questioning if the A1c result is correct is entirely warranted.

This same attitude of ‘good enough’ has also contributed to physicians basing clinical decisions on poor quality data and when that involves the A1c, for many people the only data point available for determining therapy effectiveness and decisions, we end up with situations that confuse and confound both physician and patient (lack of screening, inaccurate A1c test methods and imprecise analysis).

For patients in our healthcordia programs we share both A1c and its calculated average but they are both just numbers. The real utility however is first screening for the abnormal hemoglobins using our HomeCheck combination hemoglobin screening + A1c laboratory analysis and then identifying the increase or decrease in a highly accurate A1c/MBG.

The timing of checking for average blood glucose is probably a bigger issue since approximately half of the A1c is based on the immediately preceding 30 days. Therefore, unless you check bg’s with your meter around the clock, you still don’t have a good measurement reflecting patient actions and therapy effectiveness given that lifestyle changes usually do not occur 30 days prior to your doctor visit A1c checks. Bottom line is nobody knows what they’re really measuring which is making it hard to understand and patient actions are not getting accurate feedback from which to encourage or reinforce.