[morganherlocker]

Type I diabetic here. Assuming current tech stays where it is (not saying it will), this could easily tack 10 years on to my lifespan. For many who watch their diabetes less closely (something I cannot fault anyone for), this could add 20-30 years.

For anyone who does not know, type I diabetes is not something you can just follow a doctor's direction on and be ok. Even if you follow your doctor perfectly, there can still be serious complications, and type I diabetics with the best control are actually more likely to die from severe low blood sugars.

The reason for this is that the optimum blood glucose level is around 100. <70 and you start to be severely mentally impaired, making it difficult at times to seek treatment (finding and eating sugar, in a nutshell). On the flip side, if you are lax on insulin, your blood sugar might hover around 250 for months, and you will feel close to normal. Having a blood glucose this high on a long term basis will have long term effects that are what kill most diabetics in the long run. A low blood sugar, however, can be fatal within minutes to hours.

Either way, a continuous feedback mechanism would help tight control diabetics, and diabetics who do the minimum. Tight controllers could get faster feedback about when they are going into the serious danger zone without having to initiate any action (checking blood sugar), and lax diabetics would get a constant reminder of how they are letting there life slip away (which they normally would rarely see, since they hardly ever check their blood sugar anyway).

I have to say though, I am still a bit skeptical for a few reasons:

- One, I have been told about this sort of miracle technology ever since I was diagnosed 15 years ago.

- Two, the medical complex locks down their tech and extracts the maximum value out. There is not a single glucose device on the market that lets you extract the data out of your glucose monitor and crunch the data how you want. I have worked on hacking these devices to extract data and the legal verbiage around these activities has strongly discouraged me from releasing anything. Previous continuous glucose monitoring systems. These companies would prefer you rot in the dark, than to lose one bit of profit.

- Three, if one of these devices is not 100% perfect, it gets shot down and banned from the market. This is probably a combination of profit-motivated industry and caution-motivated government. A great example of this is a continuous glucose monitoring, non-invasive watch that came out ~ a decade ago. It was on the market for several years, before being banned. I, like just about every person in the thread I linked, would pay $10k+ for one of these, despite the reduced accuracy over traditional devices. Entrepreneurs in the health industry take note.

[1] [http://www.diabetesdaily.com/forum/testing-blood-sugar/61908...]

[MrFoof]

>type I diabetics with the best control are actually more likely to die from severe low blood sugars.

I'm very much in this crowd. Type 1, AIC of 6.3, LDL cholesterol around 100, BP of 110/72-75. This is with completely manual testing and subcutaneous insulin injections.

The problem of keeping your blood sugar towards an A1C of 6 is you have insulin reactions. Quite often, as you're being rather aggressive in keeping on top of your blood sugar. Worse, is I don't show or feel any real physiological symptoms until I'm at 60 mg/dL or below, which is getting fairly dangerous. I can be as low as 25-30 mg/dL and still be conscious and functioning. My tipoff is realizing that I either feel tired or that I can't think straight. It's hard to realize you're not thinking straight when you can't think straight, and have the cognizance to then test and get some carbs in you.

I've had a few close calls. One where I was driving a supercar north on Route 24 in Boston rush-hour traffic. I subconsciously took an exit and drove 5 miles into a suburb, and managed to not hit anything and the police officer told me I mostly obeyed traffic laws, aside from weaving about (but was completely incoherent) -- I was at 21 mg/dL when the EMTs tested me.

I once didn't compensate for alcohol, passed out on my couch, and when I finally came to in a pool of my own sweat, it took me 3 hours to traverse 30 feet to the kitchen to attempt to drink (and wear at least half) of a half-gallon of orange juice. When I finally tested 30 minutes after consuming an entire carton of OJ, I was barely in the 40s -- I have no idea how low I was, and I was lucky I ever regained consciousness.

[trekky1700]

Just curious, how old are you and how long have you had it? I've had it for 11 years this February, and while my A1C is pretty terrible (high), the idea that I might not be able to detect lows eventually is pretty terrifying.

[MrFoof]

About 13 years. My big tipoff is feeling tired. I'm a night owl and stream on Twitch after a long day of consulting -- I'm NEVER tired before 1 or 2AM.

The most severe problems were the result of a new doctor in their residency changing my long-acting insulin to an intermediate that was also a 30% mix of fast acting, in an attempt to drive my A1C to under 6. This created a serious problem of wanting to give myself insulin for a meal, my blood sugar only being around 90-100 mg/dL, and then being unable to decouple the fast-acting from the intermediate since it was pre-mixed. I hemmed and hawed, but the attending stood by the decision to change to 70/30. I had to change 10 years of regimen instantly, which was fraught with peril.

Two serious episodes later, and a waiting room at a prestigious teaching hospital hearing someone scream, "Your stupid decision has nearly killed me twice", and I'm now back on the long-acting with insulin reactions being rather infrequent (twice a month), and nowhere as severe (catching it around 65-70 mg/dL). My A1C has trended up slightly (6.1-6.2 to 6.3-6.4), but I'd rather that than death by overdose.

[bigtunacan]

Not speaking for the parent, but I'm a 35 year old T1D who has been diagnosed for only about 7 years and I am already nearly asymptomatic when I have low blood sugars. The risk of becoming asymptomatic for low blood sugars increases over time, but it is completely inconsistent. Some diabetics have always been asymptomatic for low blood sugars, others will go their whole lives and always have easy to detect symptoms. I have a friend who is in his mid forties and is also a T1D; he was diagnosed about 30 years ago and he still consistently gets the shakes when he is any lower than 70.

[coldpie]

I was diagnosed when I was 12, about 13 years ago, and I can usually still detect when I'm under 60 or 70. I'm one of the tight-control types that the OP mentioned (6.4 A1c), so this happens a few times a week. I usually notice it when I can't think, and instead wander over the same two thoughts back and forth. Suddenly I realize I'm thinking in the tiniest of circles and go have some yogurt or something.

The first thing my endocrinologist asks when I see him (literally, before he even asks how I'm doing today) is if I can still feel lows. I think he'd get me on a CGM pretty quick if I reported I couldn't.

I'm super interested in this contact lense solution.

[RA_Fisher]

In my experience (Type 1 partner), it's repeated and severe lows, say like 4 or 5 in a week, that will give you hypo-unawareness. It's definitely possible to experience that and still have a high A1C, but I don't see it causing hypo-unawareness. I'd highly recommend a pump to get more control.

[bigtunacan]

That is simply misinformation. The problem is you can not extrapolate your personal experience out to the rest of the population. I'm virtually hypo-unaware and I have low blood sugars maybe 2 or 3 times in a month.

[RA_Fisher]

I didn't say it was the only way to get hypo-unawareness, it's my experience.

[Bjuukia]

Do you just check LDL or do you also check trigs and hdl?

Weird that you don't see any symptoms at such a low level. My vision would completely black out except for maybe half a centimeter blurry hole... everyday. It took hours to recover. I was always munching on snacks to prevent low blood sugar but it never helped. Thankfully I'm alright now.

[bigtunacan]

Another type 1 diabetic chiming in here.

It is not weird not feeling it until such low levels at all. Some diabetics never feel low blood sugars; the longer you have diabetes the more common it is to experience this phenomena. Additionally how you "feel" a low blood sugar varies over your lifetime so the "tells" change. This means we may be feeling different, but have no idea that it is because of a low blood sugar since we haven't felt this new warning symptom before.

When I first was diagnosed with diabetes I could tell quite early if my blood sugar was low. If my blood sugar dipped to just 70 I would start shaking like a leaf. Now I don't have any warning signs at least until I hit around 55 and sometimes not until the low 30s; when I have symptoms at all. I have twice had incidents where I fell because my blood sugar had dipped so low with no physical warning signs. Both times my vision went black for a second and I fell over. I don't know if this was me actually "passing out", but in both instances I recovered almost immediately and immediately consumed large quantities of juice to get my bgl back to normal as quick as possible.

When I do have a tell it is difficult to even notice now; the best way I can even describe it is that I have a very sudden sense of dread as a feel my mind sort of shifting away from me. By this point I'm so low as to be in serious danger zone. I try like crazy to stay out as far away from that as possible by testing as frequently as possible, but your fingers can only take so much; and even if they could take some more the insurance carriers won't cover more test strips anyway.

I think this Google product is a great idea and I would love to see something like this actually work, but like the parent of this thread; I'll believe it when I see it. I've been reading about all sorts of wonderful continual monitoring solutions for BGLs for years and have seen nothing come of it.

[bigtunacan]

Your last point listed here is such a big deal too. It shouldn't matter if these things are 100% effective; they need to be released and patients need to be informed that manual testing still needs to be done. If for no other reason than to help reduce the number of type 1 diabetics who die in their sleep from low blood sugars.

For anyone here not familiar with this phenomenon; it is referred to as "Dead in Bed Syndrome" This is where your blood sugar suddenly drops while you are sleeping and you don't notice the drop because you are sleeping; you then progress into a coma and then die.

To this day I almost never am able to sleep straight through the night because of the sheer dread I have that this will happen to me (In fact I'm posting this at about 3:30am where I am at, because once again I'm up in the middle of the night).

In well controlled diabetics about 55% of all severe low blood sugars occur while you are sleeping. And in type 1 diabetics about 6% of us will die in our sleep by age 40.

http://www.ncbi.nlm.nih.gov/pubmed/1826245 http://www.diabetes.co.uk/diabetes-complications/dead-in-bed... http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2551657/

[Someone]

I found that 6% unbelievably high. A check confirms that. http://www.ncbi.nlm.nih.gov/pubmed/10097898:

"It appears that such deaths occur in 6% of all deaths in diabetic patients below age 40 years."

So, it's not that 6% of diabetics will die in their sleep before they are 40, but of those that die before they are 40, 6% die I their sleep.

That makes the risk per hour of dying in your sleep about a tenth of that of dying while awake. I would guess that is relatively large compared to non-diabetics, but probably nowhere as freaky as you portray it it be.

Further evidence: http://www.ncbi.nlm.nih.gov/pubmed/8542738 shows 16 such deaths in Norway in a ten year period, with 224 other deaths (during the day, or at night, but not matching the criteria for death in bed syndrome)

Also, http://www.fhi.no/eway/default.aspx?pid=240&trg=List_6673&Ma... shows about 20 type I diabetics per 100.000 Norwegians. With 5.000.000 Norwegians, that gives about 10.000 type I diabetics, of which about two die each year of this cause.

[bigtunacan]

Thanks for the link. I've seen the statistic listed several other places and it was shown as 6% of type 1 diabetics under age 40.

The importance of someway to reliably provide continuous monitoring of BGLs is still critical and could help prevent a lot of deaths as well as contribute to the overall well being of diabetics.

[coldpie]

Yep. I actually have two alarms on my phone, one for wakeup, and one for 2 hours after I go to sleep. I've been doing this for years. Most nights I just solve the math problems unconsciously and go back to sleep. But some nights, I can't solve the math problems or otherwise feel low and I know it's time to check the BG.

[bigtunacan]

I really should set an alarm for a middle of the night check as well, but so far shear dread seems to have been pretty effective. :(

[joeconway]

You may be interested to know that Bayer's Contour USB allows you to pull the information off in an SQLite DB

http://www.bayerdiabetes.co.uk/sections/products-for-my-diab...

[morganherlocker]

Almost all meters allow you to extract data over USB, but the data usually comes out in an obfuscated proprietary format. Do you know which format this exports to? Scott Hanselman had a blog post about this a few years ago (can't seem to find the link).

IMO, every meter should be required to export to CSV at a bare minimum. I will look into the one you linked, but for most users, they really need something that can be easily converted to an xls. The current state of things is that they only export to some supremely terrible proprietary program that is really only meant for an endocrinologist to use with the help of an IT team.

I should be able to easily view my health data and crunch it how I see fit. Decent analytics around my glucose data would be extremely valuable, but the current state of affairs is abysmal.

[joeconway]

I really feel your pain. The reason why I love this meter so much is that it doesn't require a £50 USB cable (The usb is built into the meter) the proprietry software is a standalone jar on the USB device which is amazingly quite good. But best of all when you plug the USB device in it mounts as removable storage with a db.sqlite file which is easily accessible and a simple .dump reveals the sensible DB schema. It's really refreshing to be able to get my data and do what I want with it.

[morganherlocker]

Thanks for the tip. I will buy this and try it out, open sourcing any scripts I can write that will assist in the extraction. Do you know of any existing tools?

[pjscott]

If it's a sqlite database with a reasonable schema, you can probably get a surprising amount of mileage out of a few simple SQL queries and this command-line tool:

https://www.sqlite.org/sqlite.html

In particular, the CSV output mode could be handy for getting query results into other programs.

[omh]

Seconding the recommendation. I have this meter for the same reason.

You need to use the Bayer software to talk to it (I'd recommend actually installing the software rather than using the copy from the meter itself). But then the readings are stored in a standard sqlite database without any obfuscation etc. It uses some sort of hash so you can't actually modify things, but for just getting the raw numbers I didn't have any trouble.

[selimthegrim]

Decent analytics exist (disclosure: I am web lead for a company [1] in precisely this device space) but some problems we are facing are situations like the following:

Device manufacturers hiding the ability to decrypt their (proprietary binary blob) file format in the same DLL as their driver, forcing us to install it on the end user's machine, and refusing to release the source code to us either so we could verify it was thread safe and run it on the server or possibly use Emscripten/asm.js on it to be able to package it cross platform in the browser.

In addition a plurality of our userbase is on IE 8 and below. I don't even want to think about their XP situation.

That being said, we have a good start at excellent analytics capability. Feel free to shoot me an email (HN handle at gmail) if you're a device manufacturer, clinician, patient, Scott Hanselman (we're in Portland too! And we use .NET!), or just plain doggone interested in the space or just working on the problem.

[1] http://www.diabetespartner.com. Any feedback on the design is of course welcome. We do in fact support the Bayer Contour mentioned above.

[aberkowitz]

The blog post you are talking about can be found here[0]. Despite an active push by Scott for a standardized data format and communication protocol, little has been done in almost 5 years.

As a person with diabetes for 14 years, I feel frustrated about the roadblocks to accessing my data. I am currently investigating reverse engineering the one touch verioiq.

[0] http://channel9.msdn.com/coding4fun/articles/SweetSpot-The-B...

[Shivetya]

I would like to add to your comment.

My mother has been full on diabetic for thirty plus years. She uses a pump now as it does offer some convenience over just using needles. It still is not easy.

The problem I see with this technology is that her biggest threat is at night. She can go to bed with perfectly fine blood sugar levels, just like any other night, and wake up with a level in the forties. What worked good for her? Besides an attentive spouse a dog works wonders too.

Still I haven't seen an accurate blood sugar monitor, there are variances depending on where they are placed on the body and even at times the same place may not yield similar numbers each week. The same goes for where her pump connects, some weeks its a great spot, another week a similar spot isn't quite the same.

Top it off with, her comment about asking three doctors in day and getting four opinions. The flip side of low blood sugar is silly high numbers. Going to the hospital with 300, or worse 600 plus, and having to debate the emergency room staff that you know whats happening isn't fun. Having to have your level tested three or four times because they cannot believe your functional at those levels is a sign of how much even much of the medical world has problems with this disease.

I am all for new technologies but I believe your chances lie in first not ignoring the diagnoses, second making sure to take the medicine assigned; especially your insulin; third, enlisting your family and even friends in emergency treatment aids, fourth being purely your genetics, and finally knowing when to doubt a medical professional and how to engage them intelligently.

[aarondf]

Type 1 diabetic here too.

I too would pay $10k+ for one of these for precisely the reason you mentioned above: adding years to my life.

[trekky1700]

I would too, I would get a job at Google in their lab if it meant being able to take one home early.

[matznerd]

Switch to a ketogenic diet.

[awolf]

I'm going to go ahead and down vote you here as well. I'm a type one diabetic on a ketogenic diet. Continuous glucose monitoring is still extremely valuable to my health.

[esailija]

I am type 1 on ketogenic diet as well and my sugars are virtually always in healthy person's range (or actually always if the healthy person eats candy or white bread :P).

So I am curious why you consider it extremely valuable with such stable sugars in the first place? I mean I can see it could be nice to have for the stats but that's not worth the obtrusiveness at all.

[samstave]

Is this to say though that the diet is not a good idea?

[venomsnake]

When you give advice that could screw pretty well someone's life, probably you should prove the advice is beneficial or not harmful at least.

[meritt]

Keto is effective for Type-2 which is caused by people not having a proper diet in the first place.

[morganherlocker]

How does this help type I diabetes? I see nothing about this diet that does anything beneficial over alternatives. Type II maybe, but please try to research a bit more before throwing in a dictate.

[aarondf]

I understand that you are trying to be helpful, but it came off as completely insensitive, FYI.

[samstave]

Sometimes statements are statements. People needlessly attach emotion to the written word when there is no need.

Happens millions of times a day on email...

[ShawnBird]

It really comes down to phrasing. If they had said "Have you tried a ketogenic diet?" Or "Some people have found a ketogenic diet to be helpful with type 1 diabetes." it would be fine. Instead it came off as an order or command: "Switch to a ketogenic diet."

One of the things about written communication is we only have your phrasing and word choice to inform us of your tone and meaning. It is vitally important to use proper phrasing to avoid being misunderstood. It is entirely possible that their comment was innocuous but it came off as rude.

[jemfinch]

From Paul Graham's "How to Disagree"[0]:

DH2. Responding to Tone.

The next level up we start to see responses to the writing, rather than the writer. The lowest form of these is to disagree with the author's tone. E.g.

    I can't believe the author dismisses intelligent design in such a cavalier fashion.

Though better than attacking the author, this is still a weak form of disagreement. It matters much more whether the author is wrong or right than what his tone is. Especially since tone is so hard to judge. Someone who has a chip on their shoulder about some topic might be offended by a tone that to other readers seemed neutral.

So if the worst thing you can say about something is to criticize its tone, you're not saying much. Is the author flippant, but correct? Better that than grave and wrong. And if the author is incorrect somewhere, say where.

[0] http://www.paulgraham.com/disagree.html

[samstave]

>"Instead it came off as an order or command: "Switch to a ketogenic diet.""

What if I told you;

It is up to you to be able to not ever take anything anyone else says as an "Order or Command"

[erkkie]

I'm expecting sensor tech to explode in the coming years, including noninvasive cgm. Imagine high level endurance athletes on a full biometrics panel during a race, including glycose and blood pressure (among other things).

[morganherlocker]

If pro athletes are where the demand lies, I am fine with that. Unfortunately, most education around diabetes has conflated type I with type II, which has caused and overemphasis on high blood sugar control, rather than low blood sugar management. Pro athletes might have the same problem as type I diabetics, which could lead to a converging need.

[erkkie]

Getting sensors out on a non medical market (first?) might also be more feasible enabling interested people to "self medicate"

[namank]

Impossible. Anything with "blood glucose" on the labelling or even a hint of it will need FDA approval.

[erkkie]

Interesting, I didn't realize that, wonder if/how that differs in the EU

[hedwall]

What scale is that glucose level measured in? I ask because my partner is diabetic but and those levels are supposed to be between 6 and 12 mmol / L. Clearly 100 is not expressed as mmol/L.

[mortov]

US uses mg/dl while most [but not all] other places use mmol/l.

Conversion is mg/dl -> mmol/l divide by 18 approx.

See http://www.soc-bdr.org/rds/authors/unit_tables_conversions_a...

Note, this is only for glucose; there are different factors depending on what is measured, e.g. cholesterol is different.

[hedwall]

Thanks.

[namank]

Sympathize with you!

How would reduced accuracy really help you? I'm imagining that it could be worse since it could introduce anxiety...

[rosser]

I think what he's saying is that an accuracy of, say, 10 mg/dL on a continuous basis is better than an effortful (requires you to test — and, perhaps more importantly for Type 1 diabetics, to be able to test) accuracy of 1 mg/dL. That's still going to tell you with enough accuracy whether you need to get some carbs into your system, or whether you've put too many in already.

Please correct me if I've misconstrued.

[Someone]

I think that would depend on the type of errors the device makes. If every 1s measurement is of by independent errors, you can simply do a running average and improve accuracy that way. If the error is fixed, you can determine it and compensate.

If, on he other hand, errors between close in time measurements are correlated, that won't work. You might get "things look fine" measurements for an hour while your glucose level is dangerously low or vice versa.

Given that the first two kinds of errors can be worked around, I would guess this device to produce the third kind of error. You don't need 1s intervals in glucose reading to improve the life of diabetics; 10 minutes or even an hour would be fine, too.

[mdup]

I guess most meters have a display accurracy of 1 mg/dL, but that's just on display. If you check your blood twice in a row, you're likely to get at least a 5 mg/dL difference. I'm not even sure this is your meter's problem, but rather your blood having not 0.01 mg/dL-consistent sugar across all of your body.

Nevertheless 10 mg/dL is just the right accuracy needed to control type I diabetes. I don't tend to attach much importance to the last digit anyway.

[morganherlocker]

Reduced accuracy would be offset by frequency of feedback. The current model requires active input; you have to go and prick your finger. With a passive input, you could get a general idea all the time, which is especially important considering that a low blood sugar can sometimes affect your mental capacity.

[kathleenj]

Looks like Microsoft is/was also working on something very similar. Anyone know more details?

"Functional Contact Lens Monitors Blood Sugar Without Needles"

http://research.microsoft.com/apps/video/dl.aspx?id=150832

[psbp]

Did you acknowledge my explanation further down that this is Babak's research and that Microsoft only contributed/funded his research, and then post this up higher for some sort of MSFT brownie points?

[phaker]

She posted here first, before she responded to your post (and likely before she read it).

[thrownaway2424]

404. Did it really disappear in the last 32 minutes?

[kathleenj]

Still works for me. Try this link http://research.microsoft.com/en-us/collaboration/stories/fu...

Or the PDF. http://research.microsoft.com/en-us/collaboration/stories/nu...

Looks like the same person Babak is behind both these projects.

[simik]

Have you tried to control your diabetes with a ketogenic diet? Dr. Bernstein, a type 1 diabetic since 1946, is successfully managing his condition with it, and wrote a book about it.

http://www.diabetes-book.com/book/mylife.shtml

[venomsnake]

Dr Bernstein claiming to have type I diabetes since 1946. There was not proof on his site - no epicrisis, there was no sources chapter in his book ...

[matznerd]

I was going to say this! Ketogenic diet is the best and most stable way of eating for pretty much anyone. Fat is a much better source of energy than glucose.

[ceejayoz]

Rule #1 on diet/health claims: anyone that claims something to be the "best for pretty much anyone" is virtually guaranteed to be bullshitting.