Time In Range - useful for Libre / CGM wearers

Status
Not open for further replies.
everydayupsanddowns

everydayupsanddowns

Administrator
Staff member
Relationship to Diabetes
Type 1
Pronouns
He/Him
I've been meaning to post this for ages, but kept forgetting!

At the ADA this year the international consensus paper on TIR (time in range) was published by a panel of experts with a huge amount of knowledge and expertise in supporting people to use continuous data to help manage their diabetes.

Different targets are suggested for general T1/T2 users, and those who are pregnant or who may be older or infirm.

Hope people find this helpful and that it gives folks something to aim for. Particularly useful to recognise that you have have a reasonable chunk of readings about 10mmol/L and still get a great HbA1c. Conversely, if you are getting 80% in range, but 10% of the other results are hypos, it shows there's still work to be done.

I've tried to attach a graph that shows a summary of the suggested targets. Or you can read the full paper here: https://care.diabetesjournals.org/content/early/2019/06/07/dci19-0028

Time-In-Range_800.jpg
 
Last edited:
For most people with T1/T2 the ideal TIR would be:

More than 70% of time spent between 3.9mmol/L and 10mmol/L

Less than 4% of time spent below 3.9mmol/L including...
less than 1% of time spent below 3.0mmol/L

and

Less than 25% of time spent above 10mmol/L including...
less than 5% of time spent above 13.9mmol/L
 
That is a really useful chart and gives people a much more realistic idea of what is acceptable.... and makes me feel a bit better as I have had an unusually high spell over the weekend due to stress and was concerned that it might impact on my next HbA1c.
There was a recent post on the forum from a gentleman who was getting grief from his practice nurse for not keeping his readings within a very rigid 4-7 mmol range. I will see if I can find his post and link this to it as I am sure he would find this info supportive.
 
That’s really useful. I usually set my parameters at 3.9 to 9.0. When I alter the top one to 10, it suddenly looks a whole lot better.
One thing I’ve found since the Reader update, my 'below 3.9' percentage is now within the 4% (just!) It always used to be around 10%, because the Libre always read lower than my meter at the bottom end, and would often show me crawling along the top of the red all night. Now it doesn’t, and I haven’t changed insulin/diet/exercise. Looking forward to not having 'that' conversation about hypos with my nurse. (My HbA1c has always been higher than the libre predicts, too, so it’ll be interesting to see if it’s more in line next time, if the reader algorithm is now more accurate.)
 
Robin said:
That’s really useful. I usually set my parameters at 3.9 to 9.0. When I alter the top one to 10, it suddenly looks a whole lot better.
Click to expand...

I did have mine set to something a bit tighter than that (I think 4.5-7.8 or something) but when I started using the app on my phone I set it to 4-10 (I think I read or heard of this TIR advice somewhere) and (though it hasn't really changed my behaviour much) I must admit I feel more relaxed. And it appears I'm actually within the guidelines. (Interesting that they're much more concerned about avoiding hypos than being above target, though I think that matches what I've always been advised.)

(I'd still like to aim for a tighter range than that, but at least for now I'll keep it as 3.9-10.0.)
 
I keep my regular checks at 3.9-9.0, (because the advice for fingersticks has been no higher than 9 after meals for a while), but now I occasionally set to 3.9-10 to see the difference.
 
Benny G said:
I usually use xdrip+ rather than the Libre reader. I don't think the accuracy is 100% at the top and bottom of the range, so I set the alarms for 3.5 at the bottom and 10.5 at the top. (Or else I would never get a full night's sleep.)
To see how I have fared over the last 3 months with your suggested 'in range' values I set the limits to 4 low and 9 high.
Result: In 71% / High 18% / Low 9% (Under 3 - 1%)
Estimated hba1c 5.9
Still some work to do. (And if I don't get those lows sorted out by my next appointment, my Consultant will tan my hide)
View attachment 11963 View attachment 11962
Click to expand...

Looks like your main excursions below target are only happening occasionally, and overnight.

This is exactly what I get too and it’s very frustrating. It seems to be down to overnight variation in insulin requirement. There are clinical studies in paediatric populations to show that overnight insulin need can vary by up to 300%(!)

For me it seems that perhaps 10% of days I just need much less insulin overnight.

Your range is impressively narrow though, so it looks like you could raise your median by very little and dodge those occasional overnight lows?

Hybrid closed loop of some sort is the other solution, of course...

upload_2019-7-17_9-0-19.jpeg
 
Robin said:
That’s really useful. I usually set my parameters at 3.9 to 9.0. When I alter the top one to 10, it suddenly looks a whole lot better.
One thing I’ve found since the Reader update, my 'below 3.9' percentage is now within the 4% (just!) It always used to be around 10%, because the Libre always read lower than my meter at the bottom end, and would often show me crawling along the top of the red all night. Now it doesn’t, and I haven’t changed insulin/diet/exercise. Looking forward to not having 'that' conversation about hypos with my nurse. (My HbA1c has always been higher than the libre predicts, too, so it’ll be interesting to see if it’s more in line next time, if the reader algorithm is now more accurate.)
Click to expand...
I had the same problems with the Libre when I was using it before, but now it seems much more accurate and has me settled above 4 through the night - before it was often well below. I have much greater trust in the current version 🙂 About to finish my first sensor today and I have 91% in range 😱 One thing I have noticed is that, no matter how hard I try to pre-bolus for things, my levels generally spike up (briefly) above 10 after eating - of course, if I wait too long then I'm hypoing before eating! Just have to accept it 🙄
 
Benny G said:
That photograph shows 6 mins of mild hypo during 30 days. WOW. Well done.
Nighttime hypos have been a personal demon for many years. The Libre helps track the lows in the night, (which do not occur every night.) The alarm on my smartwatch is fine in the daytime, but often does not wake me up through the night. I might have to use my phone overnight for a better alarm.
More playing with the Levemir dosage and timing to try and fix the overnight fast.
I have to say it again, 6 minutes of mild hypo per month. Wow. That is a goal to aim for.
Click to expand...

Ah thanks, but it’s not quite so good as that. The numbers are averaged per day, so some days during the 30 days I might have had a couple of exercise-related or gardening/DIY-based double-dip hypos when insulin sensitivity changed or the weather was particularly warm. Most days I would have been above 4 all day though.

When I’m wearing sensors and running Smartguard I don’t have hypos at night (when I have much reduced awareness).

The majority of those dips and minutes are between 3.5 and 3.9 though.

Without the alarms and Smartguard I have more low-level dips below 4, but I’ve been lucky to have been able to run sensors for several weeks back-to-back rather than a few weeks on / few weeks off.
 
The chart is very useful @everydayupsanddowns . As others have said it gives a realistic target to aim for,( and an excuse to celebrate when it is better than that).

Add in the more informed users and training available and we are on a winner. The other advantage of the Libre is that it provides the glucose monitoring for those on MDI as well.
 
Just followed a link from another thread and found this one, which I somehow missed in July - thanks for the graph, Mike, that's really interesting. Certainly good to know that my 70%+ in range is what I should be aiming for, and though I don't have the figures to divide my highs or lows I'm pretty sure I'm spending less than 5% of my time over 13.9 - but considerably less than 25% of my time over 10 as I'm spending about 8-15% of my time below 3.9 🙄

Northerner said:
One thing I have noticed is that, no matter how hard I try to pre-bolus for things, my levels generally spike up (briefly) above 10 after eating - of course, if I wait too long then I'm hypoing before eating! Just have to accept it 🙄
Click to expand...
Snap!

Btw, has anyone else noticed that the numbers don't add up properly to 100% - eg in the left hand graph, if the 25% over 10 includes the 5% over 13.9, and the 4% below 3.9 includes the 1% below 3 (as the notes say they do) then the figures add up to 99%. I'm intrigued to know what my blood sugar is supposed to be doing for the remaining 1% of my time 😱:D
 
TheClockworkDodo said:
Btw, has anyone else noticed that the numbers don't add up properly to 100% - eg in the left hand graph, if the 25% over 10 includes the 5% over 13.9, and the 4% below 3.9 includes the 1% below 3 (as the notes say they do) then the figures add up to 99%. I'm intrigued to know what my blood sugar is supposed to be doing for the remaining 1% of my time 😱:D
Click to expand...

My guess was that they found improvement between 5% and 4% (I saw earlier versions of the consensus with 5% at the low end), but didn’t want to put 71% for the main section, so went for the tidier-looking ‘more than 70%’ (which simplistically means at least 71%) 🙂

I think the consensus suggests that there would be benefit in attempting to halve your time below 4 - even if that increases your time above 10. I’m not sure whether your lows are powered by corrections like some of mine are (a risk of continuous data?). Sometimes I find it helpful to aim corrections at 9mmol/L rather than mid-range, to try to reduce this risk.
 
Last edited:
Interesting post.

I am a little "skeptical" about how accurate Libre really is when the blood test meters we think of as being accurate are them selves not necessarily so but hey we have to start with something 🙂

Just looked at my 90 day Libre info, I run mine 4.5-9.5 so 57% in that bit, which I guess is good and only 1% at 13.0 (which I HATE being that high or anything over 10, but that's just me) however <3.9 25% LOL which for me, I am happy about, no sure the experts would be but hey ho 🙂
 
everydayupsanddowns said:
I think the consensus suggests that there would be benefit in attempting to halve your time below 4 - even if that increases your time above 10. I’m not sure whether your lows are powered by corrections like some of mine are (a risk of continuous data?). Sometimes I find it helpful to aim corrections at 9mmol/L rather than mid-range, to try to reduce this risk.
Click to expand...
No, unfortunately it's not corrections which lead to my hypos - I hardly ever do corrections because I know full well that whatever I do my blood sugar's going to plummet anyway! I eat, I spike, I plummet - simple as that. The only way I've found to stop the hypo is to keep checking the Libre and pick exactly the right moment in the downward trend to eat a biscuit ... (and that doesn't always work, sometimes I unexpectedly need two biscuits). Decreasing the bolus doesn't stop the hypos, it just means I spike higher first. Even not eating (or injecting) at all doesn't stop the hypos - if anything it means I have more of them!

I'm pretty sure I have Reactive Hypoglycaemia or something of that sort - which means of course that I'm pretty sure my pancreas is still producing some insulin - but if - as I assume - my diabetes is secondary to my ME, I don't think that's too surprising.
 
everydayupsanddowns said:
I've been meaning to post this for ages, but kept forgetting!

At the ADA this year the international consensus paper on TIR (time in range) was published by a panel of experts with a huge amount of knowledge and expertise in supporting people to use continuous data to help manage their diabetes.

Different targets are suggested for general T1/T2 users, and those who are pregnant or who may be older or infirm.

Hope people find this helpful and that it gives folks something to aim for. Particularly useful to recognise that you have have a reasonable chunk of readings about 10mmol/L and still get a great HbA1c. Conversely, if you are getting 80% in range, but 10% of the other results are hypos, it shows there's still work to be done.

I've tried to attach a graph that shows a summary of the suggested targets. Or you can read the full paper here: https://care.diabetesjournals.org/content/early/2019/06/07/dci19-0028
Click to expand...
That is very interesting thank you
 
Annemarie said:
That is very interesting thank you
Click to expand...

Thanks @Annemarie

I think what is really interesting is that you don't have to strive for ‘perfect’ numbers to achieve an HbA1c that should reduce your risk of long term complications. It’s more important to try to reduce ‘glucose instability’ (big swings from low to high) and you can have quite a lot of ‘slightly high’ numbers while still retaining fairly optimal outcomes.
 
everydayupsanddowns said:
For most people with T1/T2 the ideal TIR would be:

More than 70% of time spent between 3.9mmol/L and 10mmol/L

Less than 4% of time spent below 3.9mmol/L including...
less than 1% of time spent below 3.0mmol/L

and

Less than 25% of time spent above 10mmol/L including...
less than 5% of time spent above 13.9mmol/L
Click to expand...
Hi Mike. I found this very useful, and the link to the article and graph in the initial post. Thanks, Nick
 
Gosh, that’s tight for gestational, mine were nowhere near in that range! Mind you I wasn’t really given much advise 25 years ago. Probably why my babies just kept getting more and more mahoosive !!
 
EmmaL76 said:
Probably why my babies just kept getting more and more mahoosive !!
Click to expand...
Probably. That seems to be something they're trying to reduce. I agree the limits are really tight (part of why rtCGM is recommended, though I think Libre 2 is also being offered which feels not quite right to me). I presume closed loop will be the normal offer if and when it can be afforded.
 
Status
Not open for further replies.
Back
Top