So then. It’s official. After three long years, the Dexcom G6, with its new insertion device has been released. Complete with lower profile, calibration free design and a ten day limit on use.
Dexcom have been talking about it for a long while, teasing us with pictures and all sorts, and now, the FDA has approved it, as a class II rather than a class III medical device, which lowers the barrier to entry for everyone else in the future.
But rather than talk about the FDA decision, which is interesting in its own right, let’s look at what they’ve approved and how that might affect the We Are Not Waiting community.
Rather than dig into the detail of the release, which Mike has done over at DiabetesMine very effectively, I want to look at what it means for those of us using the various We Are Not Waiting systems. Note that this is all speculative, as few people have got their hands on a system, and I hope that in some cases, my speculation is wrong.
Firstly, Dexcom have confirmed that the sensor only lasts for ten days. Regardless of whether you start it officially or use it in calibration mode (see Mike’s article for an explanation). However you look at it, that materially changes the cost model for everyone that doesn’t get Dexcom sensors through insurance, which is most of the T1D population globally.
Reading through the various snippets of information about how the sensor is started, it strongly suggests that the over the air communications may have changed. It looks likely that the sensor ID is being used as part of a security code within the protocol, and while the literature states that once the transmitter is paired it remains paired, that doesn’t stop changes to the underlying protocol. If this proves to be true, there’ll need to be a whole new initiative to re-interpret the setup. So much for extending support for the JDRF “Open Protocols Initiative”.
On the other hand, there are plenty of members of the We Are Not Waiting community working within Dexcom now. Can they provide any insight, or maybe even assistance?
So, working under the assumption that there’s a new protocol involved, the question then arises as to whether it’s possible to “unlock” the timer if you unlock the protocol. We know, for example, that in the G5 you can inject calibration data into an in-date transmitter and allow it to manage calibrations. It may be possible to kid the transmitter that a used sensor is in fact a new one, but we won’t know until that’s been properly interpreted.
Secondly, looking at the various literature, it very much appears that Share hasn’t changed. Why is that important? Well, whilst you may not have an offline looping solution without breaking into the protocol, all of the systems can and do loop from the cloud. It will mean that Nightscout can still be run off share data (via the official app) and that all three DIY looping solutions can also access that data.
Given that there’s currently work underway to upload via USB from the new G5 receiver using DIY tools and that the images look like it’s a software upgrade to the same receiver for the G6, then it doesn’t seem too far beyond belief that you’d at least be able to go offline with the G6 using a USB plugged receiver, which is a supported use case in OpenAPS. So for those of us looping with Dexcom data, it all looks as though it’s going to be possible to continue, all-be-it in a slightly restricted fashion.
Coming back to the ten day usage limitation, it’s obvious why it’s there. If you have a system that requires zero calibration from a two hour warm up, then realistically, you can’t let someone extend it, especially when the sign off is for use with external systems that will be making dosing decisions using that data. You also can’t have someone start a zero calibration period 14 days into a sensor’s life! So, whilst the zero calibration approach is great for many people in terms of making usage easier, the restrictions that are required between the FDA and the manufacturers do have a wider effect.
And that’s on the cost. If the rumours are to be believed, Dexcom is looking at setting the price of the G6 in line with that of the G5. If used on label, annually in the UK, we’re talking £2,640. If they really mean they’ll match the pricing that you get on the subscribe and save model, where people are clearly buying the sensors to extend them, we’d be talking £1,920 annually. While I’d like to think that Dexcom would offer the latter, I strongly suspect that this won’t be the case, and that we’re talking the larger of these sums.
Please Dexcom, prove me wrong. You are one of the key providers targeting Time in Range as a key measurement of living with type 1. Surprise us all and make your product far more accessible by reducing the cost to a level where many more people can benefit from it. Otherwise, I can imagine that the G5 will continue to be where most of us go, simply to get the extra life out of the sensors and keep our costs lower.
I’ll be honest, I am looking forward to taking Tylenol without my readings being messed up for hours.