mike@work
Well-Known Member
In my opinion accuracy IS important, and you put some of the reasons very nicely here...
-
Guest - w'd love to know what you think about the forum! Take the 2025 Survey »
Freestyle libre 2 with-Optional-Real-Time-Alarms
- Thread starter O_DP_T1
- Start Date
In my opinion accuracy IS important, and you put some of the reasons very nicely here...
I've had as much of a problem with the Libre's precision as I have with BG meters; they're wrong sometimes and always will be. Although I've never had a full day of incorrect results like what @Mel dCP posted.
I was considering getting a MiaoMiao, for the simple reason that I'd like alerts if going high or low and the Libre 2 would give me this functionality. I'll probably still get a MM as I'll still have to fork out for a new reader for the Libre 2. Plus, looking at Twitter, Abbott are being very non-committal about when they're going to release this outside of Germany - could be ages away.
I was considering getting a MiaoMiao, for the simple reason that I'd like alerts if going high or low and the Libre 2 would give me this functionality. I'll probably still get a MM as I'll still have to fork out for a new reader for the Libre 2. Plus, looking at Twitter, Abbott are being very non-committal about when they're going to release this outside of Germany - could be ages away.
mike@work
Well-Known Member
If we are to trust Mr. Google, it seems that some people get good (accurate) readings with the Libre, and some doesn't. I unfortunately belong to the group, that very, very often get wrong results - and yes, I don't mean it occasionally give false readings. A for me not compatible(?) sensor, gives faulty readings more or less all the time. I don't mind if the margin of errors goes up to say 1.5 mmol/l, but more than that means - you get a reading of for example 7.5, what the BG-meter shows is then usually around 9.5 or more, for me - not good... I believe in my BS-meter, it corresponds to how I feel.
This is true of all CGM. Have you tried others to see whether you are compatible. Some people simply aren't.
mike@work
Well-Known Member
Njaah, I don't know if I dare to say it aloud here, but I get my sensors and monitor for free. I'm not able to get another brand yet, but I have asked for a Dexcom if/when it will be available...
I guess you're right then, some people have accurate results, some don't. Guess it's because it's not directly measuring blood glucose
A don't know how they will power the sensor for 2 weeks so that it can give a continuous data stream.
Two way communication is not hard to do over their system and is already done to activate a sensor. Hence I am thinking the sensor may just transmit a hartbeat say every 15 minutes along with having very limit trigger surport, with a single value raw data pocket sent when a trigger fires. (or just send a packet when the BG value has changed a lot)
Two way communication is not hard to do over their system and is already done to activate a sensor. Hence I am thinking the sensor may just transmit a hartbeat say every 15 minutes along with having very limit trigger surport, with a single value raw data pocket sent when a trigger fires. (or just send a packet when the BG value has changed a lot)
LooperCat
Expert
- Messages
- 5,222
- Type of diabetes
- Type 1
- Treatment type
- Other
That sensor I pictured was spot on once I’d calibrated it - I was driving a fair bit that day so did several blood tests, and each of them was within 0.5mmol of the readings Spike/MiaoMiao was giving me, so I was confident I could trust the readings on my watch over the scanner results. I’ve found that when a sensor is out, it’s consistently out, so with my add-ons, I can still use it safely. In a way, I’m saving the NHS some money by continuing to use a sensor even when it’s reading a bit off, rather than replacing it early and needing to renew my prescription sooner. If that makes sense.
From the posts in my Spike Facebook group, it seems that the Libre 2 will just ping the scanner every minute with a signal that says it’s in range or not - and the alarm goes off to prompt you to scan if it’s out of range. I wonder if the range is factory set, or user set? Not much use to me personally if a “high alarm” doesn’t go off until you hit 12mmol, for example. Mine chimes at 7, which is where I want it. It just seems like a very halfar5ed “upgrade” to me, when they could have built so much more functionality into it. It’s still not going to be CGM. And I’m sure most users would happily accept a bit of an increase in sensor size to give them all the bells and whistles.
From the posts in my Spike Facebook group, it seems that the Libre 2 will just ping the scanner every minute with a signal that says it’s in range or not - and the alarm goes off to prompt you to scan if it’s out of range. I wonder if the range is factory set, or user set? Not much use to me personally if a “high alarm” doesn’t go off until you hit 12mmol, for example. Mine chimes at 7, which is where I want it. It just seems like a very halfar5ed “upgrade” to me, when they could have built so much more functionality into it. It’s still not going to be CGM. And I’m sure most users would happily accept a bit of an increase in sensor size to give them all the bells and whistles.
LooperCat
Expert
- Messages
- 5,222
- Type of diabetes
- Type 1
- Treatment type
- Other
I use Spike, which is the iPhone collector app for MiaoMiao. It’ll alarm if you are rising or falling quickly, as well as if you hit your upper and lower target numbers. You can set it to alarm at whatever rate of change you like. Mine is set at a change rate of 0.6mmol/5mins and it shouts “glucose is rising/falling fast” at me. All the thresholds are set by the user.
Edit - I’ve also set mine to shout whether it’s high, low, urgent high or urgent low.
From the presentation poster, it looks like you can set the range, see pic below, there's an option for user to change.
While's it's not going to have all the bells and whistles xdrip and spike have, I think that's one of the reasons libre has caught on so much: many people like the simplicity of it.
Everybody is speculating about power and frequency of comms. Abbott have clearly stated that there will be comms every minute from the sensor to the reader. So we might as well stop it with the well how's it going to have the power conversation. They've clearly stated that it does. They just been cagey about its content.
The question is therefore not if it can do it but what is it transmitting?
Until we get our hands on it we just don't know.
It's being discussed on jamorham's gitter page (he's the lead developer on xdrip+), so it will be interesting to see their take on it as time goes by.
https://gitter.im/jamorham/xDrip-plus
https://gitter.im/jamorham/xDrip-plus
And in Gitter. And as you might expect, nobody knows as Abbott have not confirmed anything. We'll have to wait until it become available or there is a leak.
Meant to reply to this earlier but forgot. As well as the highly customisable alerts, the android app xDrip+ has a quite sophisticated "predictive simulation" function, which I've found very useful, although, like any prediction, it needs to be used for a while to get used to it's quirks.
What happens is that you input your own parameters for insulin/carb ratios etc as per pic below, customisable for certain times of day.
Once that's done, each time you enter a bolus or carb amount, as well as showing insulin on board, which is very useful on it's own, it'll also extend the live cgm trace into a prediction of where that shot/carbs might take you, see pic below, the purple dots extending out from the blue live dots is the prediction.
The red dots extend out when it identifies a trend, and if it intersects with your low level, it'll say low predicted in x minutes, so gives you a heads up to keep an eye on it.
Like I say, it takes a bit of getting used to, it can be a bit squirrely at times, it'll sometimes radically rethink the prediction as additional 5 min readings come in, but all in all, I've gotten a lot of useful information out of it which I've used to make bolusing or dextro decisions to tweak things when it looks like it might start trending out of range. It's like a bolus wizard on steroids.
A few months back, a libre sensor totally died on me, so I was blind for about 8 hrs till I got home and got a new on. I carried on entering bolus and carb amounts during the dead time, because I still wanted to keep an eye on iob. The app continued to do the predictions. I was testing a lot more and the simulation was surprisingly accurate - wouldn't go as far as saying it was cgm without a sensor but it was close.
Alison54321
Well-Known Member
- Messages
- 1,221
- Type of diabetes
- Type 1
- Treatment type
- Insulin
Yes, I think it's clever. I also think one of the reasons it's clever is because it becomes harder for CCGs,to justify not funding it,because the alarms make it easier for people to stay in range. I saw a tweet this morning showing that research done with a CGM shows that the more people with T2D stay in range, the less complications they have. That almost certainly applies to T1Ds as well, so if they can improve it's ability to help people stay in range it gets harder, and harder not to fund it.
I expect more research to be done on this, and increasing evidence that staying in range is important
Before Freestyle Libre, with CGM, there was no way of knowing how much time was spent in range, and if that now becomes a more important parameter than HbA1c, which it really ought to be, then things start to move forward.
Wow, thanks for such a comprehensive reply - it looks fantastic! I'll deffo be buying an MM next pay day
