Tuesday, June 23, 2015

Type 1 Diabetes care: affordability or technology?

We may hold very different opinions on the political and tax issues surrounding universal health care. But there's no doubt that, if you have Type 1 Diabetes, the way your local health care system works has a major impact on your life.

I have recently read Epic Measures: One Doctor, Seven Billion Patients, a (slightly hagiographic) book that chronicles the life, so far (he is not dead yet), of Dr Christopher Murray, the driving force behind the landmark Global Burden of Disease studies. The GBD is the main "product" of the Institute for Health Metrics and Evaluation (funded by the Bill and Melinda Gates Foundation). The GBD was considered so significant and groundbreaking that the highly respected "The Lancet" devoted a triple issue to its initial release in 2012.
The IHME has developed some fabulous visualization tools, accessible to all. 

But let's go back to Type 1 Diabetes... A data lover such as I could not resist the urge of toying a bit with diabetes related data. And, to some extent, what I found surprised me somewhat. A lot of the coolest diabetes technologies are developed in the USA, it is basically impossible to visit any English speaking forum or group on the Internet without being bombarded by JDRF press releases, walks, fundraisers, etc.. There seem to a be constant flow of money and bright minds towards US based projects...

But what does it mean in terms of public health? Our local health systems is definitely not flashy. We don't have "superstars" writing books or actively working on their PR. In many ways, we seem a bit backwards: CGM adoption, for example, remains low. However, because I have spent a few years in the system, I also know that the combination of affordable care, extremely strict regulations and highly trained doctors and nurses has advantages. Is that a bias? What can the IHME tell us?

Type 1 Diabetes: Belgian and USA mortality compared.

The chart is huge (you may want to roll your own versions on the IHME web site) but even at the small scale of this blog, some trends are clearly visible. Until the age of 9, mortality in both countries is extremely low and virtually identical. However, as soon as one crosses the 10 years mark, the death rates start to diverge significantly. At the 40 years old mark, we essentially are looking at two different worlds (in this chart, the cutoff is the maximum death rate in Belgium, in order to keep matching scales).

Here are the unadjusted charts for both countries.


Striking, isn't it?

Some differences could probably be attributed to the size of the county. In case of emergency, it should be easier to get to an hospital in Belgium than to be air-lifted from a remote part of West Virginia. Lets look at the data from a bigger country whose free health system is, thanks mostly to the Daily Mail, supposed to be an awful mess. You guessed it, I am talking about UK and its famous NHS.


Well, it seems to match (or even beat) the Belgian results. Beyond a certain age, your are up to 4 times less likely to die of direct diabetes complications in UK than in the US.


Canada seems to match the European rates until late middle age and then catches up with USA rates.
(I know very little about the Canadian health care system)


Are the numbers valid? Public health data is notoriously hard to interpret in some cases. However, I believe the data reflects the reality for the following reasons
  • its origin is IHME, an organization that has been recognized for finally delivering a reasonably accurate global view of the Global Burden of Disease.The methodology has been extensively documented.
  • the data matches other individual studies (possibly not a totally independent factor if those studies were among the sources used by IHME)
  • all the countries involved in this comparison have extremely high standard in terms of census, birth and death certificates, cause of death identification, reporting, etc...
  • none of the countries have any incentive to paint a better picture of their diabetes death and complication rates. In any case, the GBD looks at all death and disability factors. If a country wanted to embellish the picture of its diabetes death and complication rates, it would have to sacrifice another 'bin' and increase its mortality rate elsewhere. 
What do the number seem to indicate? Let's examine a few possible factors...
  1. technology impact: while I'd give the advantage to the US on that point - they always seem to be the first to develop and use cool gizmos - it certainly doesn't improve their stats. Or, if it does improve them, the starting situation is even bleaker than what we see here.
  2. fundraising, superstars: things like the JDRF walk, fundraiser for treatments, diabetes walk and other assimilated activities always leave me, as a European, in a state of mild puzzlement. The concept is a bit foreign to us: if we need something... well we simply walk, cycle or drive to the next hospital or doctor and get it. On top of that, we don't really have many doctors advertising their wonderful diabetes method and, since their books aren't translated in our 20+ languages, the average Joe doesn't benefit from their advice. Finally, while we do have our share of highly competent researchers, we don't really have flashy ones. There's probably less money available for research, possibly diminishing the incentive to act like a rock star. Still, all the goodies we are missing, do not seem to have a negative impact on our care. If their lack has a negative impact, it only means our results would be better.
  3. affordability of care: that's certainly a point where Europe (and to a lesser extent Canada) can't complain. The financial burden for the treatment of Type 1 Diabetes is either nil or extremely moderate. Our access to diabetologists, specialized nurses, psychologists, diabetes centers is essentially unlimited. In fact, one a kid has been diagnosed with T1D, avoiding care is actually more difficult than getting it: should you be the idiot who doesn't care about his kid, expect your diabetes center to worry and social services becoming involved... (things are of course not perfect: borrowing to buy a house, driving and insuring a car are still areas where T1Ds suffer discrimination). Could it be the reason behind the stats?
I can't find any other explanation... While I am a big technology lover,  while I am sometimes irritated by the total lack of interest of our diabetes team for new gizmos, I am forced to admit that, from a public health point of view, they are doing a good job.

As far as the USA is concerned, while I am a big supporter of the "a CGM for every diabetic" philosophy, I am not sure that a "CGM for seniors" plan is the best way to spend a limited amount of money. 

Making basic modern T1D care affordable to everyone everywhere without any hassle or meaningful financial burden is probably what would have the biggest public health impact.

Well, that's my opinion. But feel free to play with IHME GBD data and come up to another conclusion.

Thursday, June 11, 2015

My position on releasing a third party open source Libre app.

I keep getting messages (about twice a week on average, sometimes more) about releasing or eventually helping develop an eventual open source Libre application (or even a closed source commercial one). At this point, I will neither do it, nor provide direct explicit assistance.

Let's recap the technical side: Abbott uses a custom but simple TI chip. What they are doing is conceptually not far from what a sample humidity/temp sensor would do, except for the fact that they are "driving" their sensor. The data provided is, leaving aside the standard signal processing done on the TI chip, as close as we'll ever get to pure raw data and therefore quite different in nature from the cooked secondary raw data the Dexcom provides. That means that a clean application will have to replicate every process Abbott applies to the data and every behavior deriving implicitly from any model they might use.

I posted this a while ago. This is the result of one of several pseudo-CGM runs I logged and interpreted compared to the official interpretation of the minute by minute data delivered by the sensor.
"Abbot" is the official interpretation of the data.
"public" was a direct interpretation based on what was publicly available at the time.
"private" was what I would call a method based on 'constrained linear extrapolation'.
"private experimental" was a method based on a more sophisticated model, not unlike what is found in the literature.

This looks nice and has consistently looked nice when the Libre worked well. It did blow on occasions, especially when the Libre itself was asking for a few minutes of respite. It also blew on one occasion on what seemed to be a NFC read occurring during a sample write to the FRAM. It blew consistently in temperature change situations (but so does the Libre on occasions).

So, why not release it and bask in the 'glory' of having released the first Libre application?

Several reasons.

  1. technically, on a custom micro-controller, I will never be able to interpret every 'flag' with certainty. We are talking bits here and bits potentially have wildly different meanings in different locations. While a "sampling complete" posted flag is fairly standard, there is always the potential that one rarer flag will be missed or misinterpreted.
  2. any interpretative model makes assumptions: published derivative based models are very sensitive to noise in the signal for example. Imagine that a single data point is a bit funky, such as data point 3 in the sequence 142 - 152 - 146 - 168. BG is clearly rising but, for some reason, point 3 was under measured. If you use simple numeric differentiation, you'll end up with +10, -6, +22. Relying on +10 -6 may lead you to decide the situation is stable or rising slightly. Relying on +10 and +22 can lead you to predict a very sharp increase. Data point selection matters: you can't miss an error because of a technical issue (point 1) and you can't use an unvalidated model just because it happens to work most of the time.
  3. it would be in the hands of real people  - now, let's be politically incorrect here. Real people means everyone including people on the extreme left of the median in terms of cognitive abilities. (yes, I weaseled out of that one as those are unlikely to understand what I just said and be vexed by it). Blowing up from time to time on concerned and involved people who understand the limitations is OK. Blowing up on others can be catastrophic, especially if you don't blow up 99% of the time and instill fake confidence over time.
  4. Abbott is a big aggressive company. You only need to read some specialized boards to realize there is a wide gamut of opinions even among Abbott's employees (or people claiming to be). I've had a few contacts, direct and indirect with them, mostly on the confidentiality issues and have seen/read uninformed lies, informed lies and open, honest truths. The Abbott patents are so long and wide ranging you can't be sure lifting your right finger while scanning your ISIG is legal. Some documents of the patent disputes between Dexcom and Abbott are available on the net and the details discussed just blew me away. You get the feeling that you need more highly specialized lawyers to argue for months about irrelevant points than you need MSP 430 developers, MDs to develop a sensor from scratch. Releasing anything some lawyer inside Abbott thinks is protected IP _could_ mean trouble. That's why I gave links to published papers and patents on the blog. To some extent, they help. Going into deeper details might be an issue. Dexcom has been, on the whole, extremely tolerant of the Nightscout project. Abbott may behave totally differently.
  5. I don't personally care about a phone application, I care about a full CGM application and convenient data collection adds another layer of obstacles.
This being said, investigating the Libre is fun and, if you take the time to dig deeper into the details, you learn a tremendous amount of useful things!

Tuesday, June 9, 2015

Batteries, greed, ingenuity and ...

I am not going away just yet

Thanks to those of you who asked why this blog slowed down. I really do appreciate knowing some of my readers care. Some difficult circumstances have limited the time I could devote to the blog. Don't worry - if you care at all - I am not shutting it down. I currently have a few projects going on, including attempting to get a non-AP Dexcom to work at a level close to a well performing Libre sensor, running different prediction algorithms retrospectively on 18 months of my son's data, the question of dosing or not dosing based on CGM data and, a possible final Libre summary. I've also been a bit busy with the ton of side issues every caregiver for a difficult teen unfortunately knows too well. Diabetes sucks at so many levels, direct and indirect, that even if you get good results, one just feels like being stuck in an endless dark tunnel.

But today I feel like commenting a bit on a "cool hack".

Batteries and greed...

Dexcom, a company that offers very good products and an excellent customer service, isn't beyond milking its customers for the benefit of its main shareholders. Well, that is capitalism and that - we are told - drives innovation...

The area where Dexcom's greed is probably the most obvious is in what I would call the "transmitter tax". The Dexcom transmitter that you clip on your sensor needs electricity to power the sensing and process the data it collects. That electricity comes from two small watch batteries embedded in the transmitter. So far, so good.

But of course, the batteries aren't user replaceable and you have to buy a new transmitter when they run out of juice. It is not totally unreasonable since an IP68 design with user replaceable batteries could be slightly bulkier. There is also the risk that a kid could swallow a loose battery. Swallowed batteries are ugly. Still, Chinese manufacturers manage to include batteries in their cheap toys and comply with regulations. So why not Dexcom? But I digress...

The answer is, in part, because Dexcom charges a ton of money for a transmitter. The clearest indicator that greed plays a major role in that situation is that the price varies a lot from buyer to buyer. The replacement transmitter, functionally equivalent to a watch battery swap, can cost anything from $330 to $900 depending of the buying party.

Fun fact: two batteries weigh 1.5 grs. Assuming you pay $700 for your replacement transmitter, you are paying those batteries

  • 12 times their price in pure gold.
  • 7.4 times their price in pure platinum.
  • 0.000017 times the price of Californium 252
Yeah, I just wanted to be positive with the last item. It can always be worse.

Batteries and ingenuity...

That great battery robbery hasn't remained unnoticed and has attracted the diabetic community hacker's attention. User Joern (joern's post) went as far as taking x-rays of the transmitter to devise an ingenious battery replacement strategy. Extremely impressive in terms of DIY project. I did enjoy the progress reports last year. And, since last week, this evolution has been doing the rounds. A dremel, a few tools will save you anything from $308 to $898 if you can tolerate a Frankentransmitter. Ain't that great?

Batteries and...

The hack isn't easy: people have tried removing the batteries and ended up shorting the transmitter. But ingenuity came to the rescue again! If one could not remove the batteries, why not use their sides as contacts? What a great idea! Time consuming because the batteries are effectively armored, but so cool. But wait a minute - why are batteries effectively armored? Could there be a reason? As a matter of fact, there is...

The content of batteries is typically highly toxic.

That's the reason why you are asked to dispose of your batteries properly and why most industrialized countries offer secure recycling services. To be honest, it used to be a lot worse than it is now. A recent watch battery "only" contains silver and zinc. Just a few years ago, it used to contain mercury (see Renata's advertisement), a metal whose terrible toxicity has been established. You may have heard of the Minamata disaster. Or maybe, since we are talking about hackers, you may have read the story of that electronics inclined hacker or even that would-be hacker-murderer?

Did any hacker check the type of battery in the Dexcom's transmitter? Maybe they did. Maybe not. While watch batteries containing mercury are definitely on their way out in the EU and most US states, "silver oxide" batteries have a shelf life of around 5 years. Mercury free silver oxide batteries, first announced as a major breakthroough in 2005, started taking a big share of the market in 2010 - 2011. But batteries containing mercury batteries can still be found... Has Dexcom stated their products were 100% mercury free? Even if Dexcom's intention is to deliver mercury free products, do they check what their supplier provides? Hard to be sure. Especially since they don't seem to have a very strict quality control process on their current receiver and batteries...

That means that, in the worst case scenario, by reducing a battery to dust with a Dremel and a diamond disk, you are filling your house, lab or shed with mercury tainted toxic dust...

In the best case scenario, it is only silver and zinc. Not a big deal. Or is it?

Let's try to see the balance of the whole process.

On the financial side we have the following condition to satisfy

amount saved > tools purchased + hours of work * hourly rate

That one might be easy, depending on what your job is and the tools you already own.

On the health side we have the following condition to satisfy

health cost * ( HbA1c on meter - HbA1c on Dexcom) * one patient 
((health cost inhaled mercury * 0.1) + (health cost inhaled zinc * 1)) * size of family

In other words, running your Dexcom instead of using a BG Meter has a potential health benefit linked to the reduction of your HbA1c. Is that benefit bigger than the health cost associated with the contamination of your home and your family lungs to zinc dust (100% certain) or mercury dust (let's say there is a 10% probability)?

I don't know how to resolve that inequality, and I don't think hackers pulverizing batteries do either.

But that is a question worth asking if one values rational decisions.

And it is worrying it wasn't even asked.