Diabetes Management: System, not Tech

This post has been kicking around in the draft folder for quite some time now (since NICE published in August actually and I put together a post reflecting on the evidence base), another round of online debate finally pushed me to finish it :-)

The 64 million dollar question: 
On social media, how come, if you ask the question "How's CGM?" on most T1D forums, the overwhelming response is "Amazing", "Wow", "Life-changing" and so on.

But when NICE (the UK's National Institute for Health and Care Excellence) search for evidence to answer - essentially - the same question, they come back with something very different?

In the current Consultation Document for Integrated sensor-augmented pump therapy systems for managing blood glucose levels in people with type 1 diabetes (the MiniMed Paradigm Veo system and the Vibe and G4 PLATINUM CGM system), pending final publication in January 2016, NICE are lukewarm to CGM.

So how did we get from Wow to Huh..?

NICE looked at the evidence, listened to a panel of experts and patient representatives. Surely they'd have picked up the positive vibe... (no pun intended...)?

Here's one possibility, which may go some way to answering it (or may not, but it's certainly not the full story): it's about the system, not the technology.

There are two important differences between the patient populations in the two scenarios above.

Post a CGM question in social media, you are - I think - most likely to hear from those at each end of the spectrum (those making hay on the system and those banging their heads on the wall because of wonky sensors etc...). Across the systems available though, the message I pick up is overwhelmingly positive: better management, fewer hypos, kids bouncing out of bed because they've had a steady night.

In theory, no such pre-selection  is present in clinical trial populations. There's a huge body of peer-reviewed data out there evaluating CGM use. I can't say I've scientifically looked through everything and I'm able to produce a systematic review of a group of matched randomised trials, but my gut feeling is that the results are, on the whole, positive, but not in the "wow" category. As two examples, in different T1D populations, see here and here.

What's not clear in most of the studies I've read is the level of support and training - upfront and continuing throughout the study - that subjects receive. Surely it's not just about how to setup the technology, insert a sensor and calibrate?

I come back to Emily Coles' presentation at the Diabetes Hands Foundation and the message of completing the mission. Any PWD needs the tools to keep the glucose plane in the air. That may well include a funky CGM system, but if you're not even given the keys, how far are you going to get?

In my little head, the system includes not just human support (clinical, educator, carer and community), but also associated technologies, to visualise and identify trends and issues, aggregating data from your pump, CGM, meal details, exercise and, even emotions. This is big data. That technology is emerging - for example, Tidepool and MySugr.

I think it's also about pushing for better design (not the same as more functionality). Janki's 640g is great for us. It's reasonably clever too (e.g. SmartGuard). But for something that you interact with (for us) dozens of times per day, it can be harder work than it needs to be:

Why can't I see the predicted (say 30 minutes time) SG value on the screen and use that to help me decide what to administer for a correction bolus? I have to work it out in my head (or on a little Google Sheet on my phone :-) ). The 640g has this value in its head all the time (for SmartGuard).

Why all the button pushes? I use ISIG data a lot. When I go to the ISIG menu, why doesn't the pump realise that I'm likely to want to know what the current and recent ISIG data is? Why do I have to scroll through from midnight each and every time?

Now these might sound minor (they are individually), but incremental gains, played out lots of times, can reap big rewards (a slightly more accurate correction, PWD more likely to interact with the available data in the cases above).

To be fair to NICE, although their consultation document does not recommend the Vibe + G4 ("robust evidence is needed to show the clinical effectiveness of using the technology in practice"), for the Veo system, it mentions the need for "a structured education programme" and setting "appropriate targets" [that last one would need some further guidance to prevent possible misuse].

So I think it's about what you are enabled to do with CGM data.
To get the most out of a CGM, can you use it to tweak (or overhaul) your basal rates, deal with temp basals, change your carb ratios and bolus profiles, correction factors, nail your insulin lifetime, etc?

It's not an intelligence test, although it might seem like a stamina test sometimes... I think it is about being supported and guided. That's possibly more expensive than the CGM technology itself, but its value in keeping the T1D plane in the air is immeasurable in the here and now for PWD, and - I'll wager, NICE - is good value in the long-term costs to the healthcare system.

Let's start talking about the care package, the system, for people with diabetes, not individual pieces of technology in isolation.