Scott Hanselman

It's WAY too early to call this Insulin Pump an Artificial Pancreas

September 30, 2013 Comment on this post [41] Posted in Diabetes
Sponsored By

app_6_steps_621

The diabetic internet and lots of mainstream news agencies are abuzz about the new insulin pump from Medtronic. Poorly written news articles that are effectively regurgitations of the Medtronic Press Release have exciting headlines like this:

Other news outlets have slightly better headlines like

But then ruin it with vague subtitles that are missing important context:

  • FDA approved the company’s automated insulin delivery system.

This is Step 1, possibly Step 0.

TO BE CLEAR. This new Medtronic 530G pump is NOT an artificial pancreas. It is an insulin pump, similar to the very model I'm wearing right now. It is paired with a revision of Medtronic's CGM (Continuous Glucose Meter) system and it does one new thing.

This new pump will turn off if you ignore its alarm that you may be having a low blood sugar.

Read it again, I'll wait.

Note the JDRF chart above describing the steps we need to towards a true artificial pancreas. This new 530G from Medtronic is arguably Step 1 in this 6 step process. It's the first step of the first generation.

But wait, doesn't your pump just handle things for you? You don't have to stick your fingers anymore, right? Wrong.

Let's stop and level set for a moment. Here's a generalization of your day if you're not diabetic.

image

Here's what a Type 1 diabetic (like me) does:

image

If I get this new pump that news outlets are incorrectly calling an artificial pancreas will anything in this cycle change? No.

There's NOTHING automatic here. I want to make that clear. Today's insulin pumps are NOT automatic. I set them manually, I tell them what to do manually. Yes, they "automatically deliver insulin as I sleep" but only because I told it to. If I eat and do nothing, I WILL get high blood sugar and today's insulin pumps will do exactly NOTHING about it.

If I only make decisions about insulin dosage based on my CGM then I WILL eventually get in trouble because today's CGMs are demonstrably less accurate than finger sticks. And, here's the kicker, finger sticks aren't even that accurate either.

Even more insidious is the issue of lag time. Medtronic's last generation of CGM lagged by 20 to 30 minutes BEHIND a finger stick. That meant I was getting "real time values" that in fact represented my blood sugar in the past. It's hard to make reliable altitude changes in your plane if your altimeter shows your altitude a half hour ago.

The Medtronic Press Release says that this new Enlite Sensor is 31% more accurate. I hope so. I personally continue to use a Medtronic 522 pump (this new one is the 530G) but I have given up on Medtronic's CGM in favor of a Dexcom G4. I am thrilled with it. The G4 has about a 5 minute lag time and is astonishingly accurate.

NOTE: I have no personal or investment relationship with either Dexcom or Medtronic. I am not a doctor or a scientist. I write this blog post with the expertise of someone who has been a Type 1 Diabetic for 20 years, a user of a Medtronic Pump for 15 years, a user of a Medtronic CGM for 4 years, and more recently a user of a Dexcom G4 for a year. My most recent A1C test was 5.5 putting my blood sugars at near non-diabetic levels on average. TL;DR - I'm a very good diabetic who uses the best available technology to keep me alive as long as possible.

I am extremely disappointed in the lack of research, due diligence and basic medical common sense in these articles. If you are a Type 1 Diabetic or have someone in your life who is, do the research and the reading and please spread the word so people can make informed decisions.

Related Reading

About Scott

Scott Hanselman is a former professor, former Chief Architect in finance, now speaker, consultant, father, diabetic, and Microsoft employee. He is a failed stand-up comic, a cornrower, and a book author.

facebook twitter subscribe
About   Newsletter
Hosting By
Hosted in an Azure App Service
September 30, 2013 5:03
Thank you for this.

It's also worth noting that the "one new feature" that's being touted as 'artificial pancreas technology' because of some specific category that the FDA filed the approval under (https://twitter.com/thedamon/status/384420935685910529) has already been available in other countries for a couple of years.
September 30, 2013 5:13
Thanks for this Scott. As usual the mainstream media regurgitates press releases and make claims that don't stand up to rudimentary investigation. It's like reporting on the cure-seriously? the number of breakthroughs that are reported and yet in 99.9% of the reports they're nowhere near trialling anything on humans. Huge dose of salt required when reading anything in the media.
September 30, 2013 5:35
Scott, thank you for reminding me to not take my health for granted. Sorry you have to deal with this.
September 30, 2013 5:36
Well said. It's a nice step forward, but not anywhere near where I'd hoped we'd be by now.
Kim
September 30, 2013 5:56
Exactly what I thought when I got the email from medtronic and went looking for some real info. This is barely different than what's already out. My cosmo pump had features that my medtronic one still doesn't have due to some patents.(That's what the medtronic rep told me anyways when I was switching)
September 30, 2013 7:18
I've been Type 1 for 18 years (dx at 12) and I watched the entire video thinking, "my God, how does it do it all?" and then I realized, I do this too. Excellent post about Medtronic and excellent video.
September 30, 2013 7:18
Sorry, "how does HE do it all"
September 30, 2013 8:49
In the UK we've had this pump for a while and when recently deciding which pump to get my daughter I never let its so called artificial pancreas status sway me towards it. Instead we went for an Animas Vibe.

From people/reports/blogs it seems Medtronic's CGM is way slower to react in comparison to Dexcom G4 and that is a key reason for the low glucose insulin off on the Veo. It also seems that Medtronic CGM wearers rarely get more than a week out of their sensors, which is a big thing for us Brits who generally have to self fund CGM sensors.

In comparison a Dexcom G4's lag time is a lot less, it's (allegedly) more accurate and more important for me it seems easy to get over 2 weeks on a sensor, I know many who regularly get 5 weeks. Hence the reason for the Animas Vibe.

I'll wait until we get to the second generation phase before I get too excited. Having said this I believe any pump is a great step forward in comparison to the multiple daily injections my daughter was on back in June.
September 30, 2013 8:55
Thank you, I am not or know any type 1 diabetic. But I really enjoy reading your posts. I feel less ignorant.
September 30, 2013 13:42
As I grow older I often find that if I know any subject in detail whether development, rock climbing or various historical periods I always find the press produces utter rubbish that often astounds me in its simplicity and inaccuracy.

Seems a variation of the Dunning-Kruger effecrt

http://en.wikipedia.org/wiki/Dunning%E2%80%93Kruger_effect
September 30, 2013 14:11
Hi Scott,
i'm a type 1 diabetic since 4 years (now 4 months with a pump, this thing is good!).
Because of that a colleague sends me today the following article:
https://myglu.org/articles/bionic-bill-day-1-falling-in-love
I would love to test this piece of technology..
Keep up the good work!
September 30, 2013 18:01
As so many others have said, thank you for this!!! I've been seeing all the hype around this supposed "artificial pancreas" and tried to explain to friends and family that this really does-not-affect-my-day-to-day-life-with-diabetes. It's just one more feature in a diabetes management system. The day I wake up and don't have to think about every single speck of food/drink that enters my body, how much exercise I'm doing, how much insulin I need to take, if I have too much insulin already in my system, etc...then I just might think I've received an artificial pancreas.
September 30, 2013 19:14
Scott ... That was very personal; it must be hard to share. Thank you for sharing; it was educational and moving.
September 30, 2013 19:57
Thanks for an excellent post! I too was annoyed by the inaccurate early MM press and a single feature that is really just not very useful. My daughter wears the new Dexcom and we are finding it to be a huge help in her control.
September 30, 2013 20:36
I noticed on your site that you are raising money for ADA and I was curious if you have looked into the organization's goals compared with goals of JDRF. From what I can tell, and I'm just a diabetes mom, JDRF is funding the big research around the world with the focus of curing diabetes and advancing technologies that meanwhile make life easier, while ADA is focused more on education and advocacy for type 2s. My husband and I are sending our dollars to research. You may have your own reasons and I'm sure both organizations do good.

I do volunteer for JDRF in Nebraska, and spread the word where I can! http://jdrf.org/about-jdrf/fact-sheets/research-funding-facts/

Really appreciate your great posts.
September 30, 2013 21:18
Ha! A fellow diabetic and I were having this same discussion this morning and that was the question: What does this new 'artificial pancreas' do that my current pump doesn't already do. And yep, the answer was 'This new pump will turn off if you ignore its alarm that you may be having a low blood sugar'.

I won't even wear a CGM until they get the calibration down to a few checks every day...
September 30, 2013 21:26
Thank you Scott. I had the same reaction you shared when I first read the news and my co-workers have heard me rant over this type of reporting and the snail's pace of development in the world of diabetes. As an example, when Pancreum won the DiabetesMine Design Challenge back in 2011, I was very excited about the new design and potential for what may come with it but over 2 years later, I don't see much regarding progress (I have had conversations with the creator). I understand the process of developing a new piece of hardware and the software involved to control it, especially when it comes to the effects it can have on human life. Add in FDA trials and by the time you release, your idea is out of date. I've been a proponent and a critic of insulin pumps and associated devices for some time now. A proponent because the devices are a huge help in my quality of life and a critic due to the almost negligible improvements over the years. I was a latecomer to the world of insulin pumps and my first one was in January 2006; an Animas pump. I received my next pump in 2010 which is the Medtronic Paradigm with the CGM. I love the integration of the devices, however, as a software engineer, it's obvious the potential to get much closer to the level of an artificial pancreas and yet, the biggest breakthrough I've seen from Animas in 8 years is a color display and integration of the meter and the insulin pump. Now, in the last 3 years with Medtronic, it's auto-shutoff and an improved sensor.

With the number of dollars spent on research in the diabetes space and the cost of these devices and aftermarket supplies ($1B insulin pump market), I find it difficult to understand the lack of advancement. Not to downplay the hard work of those involved but, and this may be ignorance and frustration on my part, I feel as though stem cell research and/or gene therapy will lead to an answer before an artificial pancreas (me being extreme).
October 01, 2013 1:26
That video while educational was actually quite moving. Thanks for sharing and informing people.

I had no idea that diabetics had to go through that process and pain.

I'm actually surprised that in 2013 there isn't a better way to deal with this. Forgive my ignorance if that is a stupid thing to say.

October 01, 2013 1:58
Hi Scott, this is Amanda Sheldon from Medtronic. Thanks for your feedback on the use of the term artificial pancreas. We do not want to be misleading. We decided to use the term artificial pancreas device system with threshold suspend because of the new category created by the FDA called “OZO: Artificial Pancreas Device System, Threshold Suspend.” We understand that this is just a first step, but we believe it is a critical one. We know that people with diabetes have been waiting a long time for an artificial pancreas, and it is our vision to bring a fully automated artificial pancreas to market which is why we invest so heavily in research and development and clinical trials. I’d be happy to discuss this with you further if you have time.
October 01, 2013 2:04
Hi Scott,
My wife has been diagnosed less than a year ago as type one. It was a big change in our lives. But if one good thing came out of it, it was that I realised how much I love her and how much I took her for granted. I wish she could stop destroying her pancreas and I would happily give her mine.

Thanks for the info.
October 01, 2013 5:08
Definitely shared to all my friends.

Too much hype, and so many people take the "Cliffs Notes" version that the media puts out as cold, hard facts, without checking for themselves.

Appreciate this post.
October 01, 2013 16:40
Scott, thank you for this post. I knew right away exactly what Medtronic was trying to get through FDA approval. I have been following its progress for a while now. While I am sorely disappointed in the portrayal of this device as an "artificial pancreas," I am really excited that it finally passed FDA approval. The problem the FDA had with approving this device (remember it has been commercially available in Europe, known as the Veo for several years now) was that they were afraid it would tend to increase a person's A1c. The thought was that if you hold everything else constant, but remove most of your lows, your A1c would go up. Since the FDA approved guidelines for A1c numbers years ago, they didn't want to approve a device that was known to "raise" A1c levels. Medtronic was fighting this notion, arguing it should be up to the patient and their medical team to decide how best to manage their own diabetes. It looks like they finally won their fight. What everyone should be excited about with the release of the 530G is not the pump itself, but the fact that there is now a precedent at the FDA level to approve these types of devices that make decisions on their own without user input. A baby step in technology, but a huge leap in terms of the FDA approval process. However, I am going to contact my Medtronic rep and give him an earful about how this 530G was advertised in the media.
October 01, 2013 18:26
Hello Scott, this is somewhat related. I was curious if you've been following this resistant starch series over at freetheanimal.com? Some experimenters claim to be regulating glucose levels -- I thought I'd seen some Type 1 diabetics commenting but at a quick glance I could find what I was thinking of. Anyway, it seems like something you'd like to look into if you haven't already.



Kind Regards,
Dave
October 01, 2013 21:42
Hi Scott,
I was lucky enough to be part of the study (been type 1 since 1984). I've used a medtronic sensor in the past, but the insertion device (the blue tube-thing held at an angle) was brutal. And having to do that every three days wasn't appealing.

I was pleased with the Enlite sensor insertion device, a much easier process. Also, the 'tape' provided with the sensors kept it on for the full 6 days.

My A1C improved over the course of the study period. I attribute it to the fact that having my blood sugar number so easily accessible kept me from 'forgetting' I was a diabetic. True, there is a delay, but it's better than not having a sensor. I'll have to look into the G4, thanks for that.

The alarm woke me up a few times for lows, but I caught it before it stopped the basal delivery.

I agree the headlines are misleading. But hopefully future studies get me closer to having a pancrease in my pocket..
October 02, 2013 23:25
Scott: Excellent video content! I had always wanted to do something similar, but my wife said that no one wanted to see me insert things into my body. :) Thank you for explaining it the way it really is for a pump wearer. I'm on the Medtronic 522 pump alone (my PPO does not pay for any wearable CGM device, only the pump, but I'm glad about that!) I've been a type 1 diabetic since I was 12 (42 years), but have only been on the pump the last 12 years. I use the 9mm canula, yours look way shorter than mine. But I have more fat than you.
October 03, 2013 11:39
Hi Scott,

...why nobody uses some hardware ie dexcom G4, Omnipod, polar actocity tracker?

a seperat cgm sensor like the G4

a small pump like the omnipod

controlled by only 1 controll-device - maybe a smartphone

.and. include some activity sensors in one of the devises you have to wear.



This device could to everything except count carbs...
October 04, 2013 1:49
My son was dx at 13 three years ago. We use nano pen needles for humalog and lantus. We also have a dexcom g4 which lasts about 10 days and is very accurate, but he does not wear it very much of the time. We have a1cs in the 7s. He does not want to pump. Its tough to admit we use mdi in the DOC because you are treated like you not up to snuff and looked down on. Oddly enough, we used to use syringes until I saw and video by a doc with t1 (Stephen Ponder) and then read an article proving that the nano needle is actually more effective of a delivery method than the syringe. Of course, the pump is even finer control. But we feel quality of life would be sacrificed in his case. This of course is a personal choice from person to person and what stage of life they are in, etc. Anyway, not sure if this applies here but I wanted to share how the nano needle is more effective than syringe, because it is a common misconception among people that the syringe is more effective. Thanks for showing how the pump and meter work. It was interesting. I agree with what you said on what a burden that would be on a child. Heres the study on the nanos. http://insulinnation.com/faster-insulin-short-and-sweet/#.UVSIsjeC2oU. I suspect the AP will be cumbersome at first, just like the first bg meter. I hope eventually the cgm will be like a chip inserted inside the body that transmits info to a unit or ones phone, just like a heart pacemaker, but hopefully much smaller. Anyway, if research continues to be funded, one hopes tech will become much more sophisticated.
October 08, 2013 14:34
Bad Science (http://amzn.com/B002RI9ORI) is a great book for understanding how the press got this story so wrong. Great that you're pointing out the errors.
October 08, 2013 18:47
Hey Scott, I'm a little late on the bandwagon here, but thank you for saying and illustrating what I've been thinking ever since Medtronic announced this news. Artificial pancreas this ain't--is it convenient to have low glucose suspend? I suppose, but it's not enough to make me go for Medtronic.
October 10, 2013 19:56
Wow, Scott, thank you for this. I was actually browsing for .Net posts for my team as I have been a follower for years. I had no idea you were T1D. This is timely for me as my husband and I just bought a Medtronic Minimed Veo for our two and a half year old son who was diagnosed less than a year ago. We will of course have to operate the pump for him and test it on ourselves first in the training. I am actually leaving work in 15 minutes to attend our first CGM training. I had to close my office door at the end where you speak of children going through this as I just could not stop myself from breaking down. Anyway, thank you so much for this and for encouraging awareness and supporting organizations like the ADA and JDRF. It always makes me happy to learn of such successful and talented individuals with T1. It gives me hope that this terrible disease will not impact my son's potential.
October 14, 2013 23:46
Andrew - I've been T1 for 21 years and my A1C is 5.5%. It's possible. I plan on dying of old age. Thanks for your comment!
October 23, 2013 23:01
I agree with Scott and many others. Though it is huge that the FDA approved this new device, it also gives me concern that companies are intentionally lagging in their design. How long will Medtronic wait for its next version? Until they have reaped $billions from this 530? That is typical of big pharma. Same thing is happening in advancements in insulin. How long has it been since a truly new insulin has been actually commercialized? Long time - because big pharma wants to squeeze every last dollar it can out of their current patents before they expire. I had one Novolog sales rep (who is a close friend) slip the other day that he hopes they don't find a cure before he retires in 10 years bc he is making too much money with the current version of insulin.
November 06, 2013 20:41
Scott -

Thanks for publishing this. I've been getting calls from Medtronic of late about moving to the new pump/sensor system and it's nice to have a independent view.

One question: In your experience does the improved accuracy of the Dexcom outweigh the added hassle of having to lug around another chunk of tech on your belt? I understand that you've been using the Dexcom for over a year now, and wondered if you still thought well of it.

FYI I was diagnosed almost 50 years ago, and while the changes in tech are impressive, it's still a pain to do the diabetic gear dance. In the old days, they used glass syringes with needles that had to be re-sharpened(!) and urine tests for blood sugar, and there have been big changes since then, but the gear hassles are still with us.
November 16, 2013 0:36
Scott with an A1C that low do you ever experience extreme lows? My A1C's are typically around 7 b/c that's where I've been told they should be. A 5.5 blows me away! I haven't even considered trying to drive them lower. Don't worry, I'm not taking your blog as medical advice - I know you're not a doctor :) But it does give me something to talk to my doctor about.

Btw, I've been following your blog for a few years as a developer. I've spent the past couple of hours digging through your posts on diabetes and am quite literally motivated to manage my bg better. Thanks for that!

-k
November 19, 2013 21:34
Kelly - I have some lows, but I'm always prepped. I hit 50 sometimes, but I use Apples or Apple juice to get back up.
November 21, 2013 21:19
Scott, great article and I enjoyed your video! As a fellow 20 year insulin dependent diabetic now age 46 I can tell you without question we have similar mindsets when it comes to big Pharma. I tried the Mimimed/Medtronic pump for almost 6 months roughly 10 years ago and can state on record that the tubing and insertion hazards and the constant reminder hanging somewhere on your body was a deal breaker for me. I went back to shot therapy and regular glucose testing as this has worked very well for my personal goals. A1c's have always been 5.5 - 6.0 range so I consider myself medically sound according to all the guidelines of modern medicine. As I've stated since diagnosis in 1992, money is in the medicine,not the cure. Keep up the good work!

Seriously if Medtronic would just release the super secret wireless pump that has supposedly been in development for the last 4 years they could actually own the business landscape if the device were capable of handling more than a 3 day supply of insulin, that's the roadblock of all these companies, OMNI-POD has a wireless unit but same deal, unit change out every 3 days or so and it's just too dang large on your body and for me it's simply "NOT GOOD ENOUGH".
November 22, 2013 14:01
It may be way too early to call this (DTU scientists goes open source with artificial pancreas (via Google translate)) an artificial pancreas - I am not a diabetic so I don't know that much about it. But you may like the fact that this project is open source.
November 22, 2013 14:31
Hi, I recently proposed <a href="http://area51.stackexchange.com/proposals/62091/living-with-diabetes>this Stack Exchange site</a>, and thought you and your readers might be interested.

(Please delete if you feel this is spam)
November 23, 2013 1:08
Thank you, Scott.

At the age of 45 I know diabetes (Type I myself for the last 25 years). Let me tell you about the injections I give...

7 AM = 15 to 20 units of R
12 Noon = 15 to 20 units of R
5 PM = 15 to 20 units of R

If at any time the blood sugar appears over 300 I up the dose.

At night, I give 15 to 20 units of N.

I noticed sunlight helps a lot!
I noticed exercise helps a lot!
I noticed that proper vegan foods help a lot too!
I noticed a proper family helps at times too.

I keep praying for a termination to the disease upon myself, but I sometimes feel weak in my prayers, but I keep going nonetheless. Thank you for your video. I enjoyed it greatly. I keep my faith in my/our Father. I hold no desire to use a pump, but I enjoyed the video greatly. Again, thank you.
November 23, 2013 1:09
I just wanted to add my FaceBook ID to the previous post, in case anyone wants to contact me.
December 10, 2013 4:37
scott -


Don't know if you've seen this, but a PhD. just published a comparison of the Medtronic and Dexcom systems, which makes for very interesting reading. He seems to agree with you that the Dexcom comes out on top accuracy-wise. But I'll admit I'm still kind of hung up on the ease of using the Medtronic system, given the fewer numbers of tech one has to hang on one's belt, and the easy communication between the medtronic pumps and sensors.

His post is here: http://www.tudiabetes.org/group/dexcomusers/forum/topics/report-analysis-medtronic-enlite-vs-dexcom-g4

and his report is here: http://api.ning.com/files/aoq71LIlZOBimTCXqcAyB2ecBcsOPFzbkpgald3cdssaiyWnRB2KxhZtduGDNw4Af-B2vrWHwI7C2IoIY4FmPqxLxxAe39dm/CGMreport.pdf

Sorry, but I can't figure out how to insert links. You'll have to copy and paste the URLs.

Comments are closed.

Disclaimer: The opinions expressed herein are my own personal opinions and do not represent my employer's view in any way.