You set these figures yourself, somewhere (it’s a while since I used Clarity). I recall an endo with what I interpreted to be eyesight problems saying my TIR was wonderful on looking at Dexcom data without, apparently, noticing that I had set the TIR (IRC I set it to 80-170, but I can’t remember).
The “default” in Clarity is 70-180, but your endo can override this.
For a long time the only TIR range I saw reported was the 70-180 one. For myself I don’t like the low limit of 70 because that’s where I start experiencing low BG effects; I prefer 80 because it gives me a shot at fixing them. That’s the alarm settings of course, not the stats, but having stats for 80-180 shows me how often I need to do a fixup. All the same I give my endo the 70-140 and 70-180 figures, but he seems to just want the graphs (which, truth to tell, reveal the same information).
I do also use HbA1c, which I measure by fingersticks. I’ve taken to doing this once a month, I don’t think they are that accurate but the number from my CGM isn’t either; I use a G7 and for the first couple of days it isn’t that accurate. I will typically calibrate after a couple of days but all the same it seems to read consistently high until I calibrate.
All that said I’ve been running a closed loop since July 1. “Closed Loop” means, in the new terminology, I don’t enter any carbs or do any boluses. I just let the algo (AndroidAPS beta) take care of everything. I haven’t been trying to make it work, indeed, I been somewhat cavalier about my food intake and I’ve been running maybe 20mg/dL higher than I want on average.
So, no bolus, no carb entry, a diet which doesn’t involve staples (bread, rice, tatties) but does involve fruit (citrus, berries, but I avoid apples and pears). Result? Amazingly good considering I’m not even trying, not quite as good as when I was trying, but that was a lot of effort:
HbA1c (Walmart Relion fingerstick): 5.7 (down from a reported 6.0 on June 14 with a lab result from the same day of 5.8, I don’t think the Relions are that accurate).
HbA1c from AndroidAPS (calculation basis unknown to me): 6.6% (it was, IRC, around 6.3% on June 14).
TIR (AndroidAPS) over 7 days: 70-180 0%(low)/92%(in range)/8%(high),
70-140 0%(low)/52%(in range)/48%(high).
The 30 day figures are broadly similar, slight better (1-3%) for “high”, but the highs seemed to be higher, IRC.
So I’m certainly with the T1DExchange conclusions that @MsCris posted:
By combining TIR with A1C, you can get a more complete picture of your glycemic management to make informed decisions about your personal T1D treatment.
The HbA1c blood test factors out errors in the CGM and gives us, and our endos, a measurement that can be tracked over time. The TIR suffers from those errors, that can’t be fixed, but it gives us another metric which I find useful for day to day management. I hate going low; it is completely disabling for me. I hate going high (significantly about 200) because that is also disabling; I become irrational and angry. TIR helps me manage the swings.
Will I stick with a full closed loop? Possibly, possibly not. The algo I’m using (DynamicISF in AndroidAPS) seems to cope extremely well with carb meals. I don’t normally eat those but there is little choice in a restaurant; seeing the sugars is certainly whack-a-mole, the diabetic equivalent of seeing the fnords.
Rather than mad bolusing I suspect I might end up sticking with temp basals to handle high fat and/or protein intake.
TIR has only partial value in assessing daily management. You also need to look at deviating, average bg,and time in the 5 quanderants of TIR to assess management, A1C is useless since on event the morning of a test, blood type, recent physical activity, or last night bg will impact the measurement using the decades old science.
Here is some good info on TIR.
Time In Range PDF
Diatribe, Time in Range
Stats on time in range is more like a report card to me, since I have 24x7 real time CGM data, with alerts and smart pump to steer me back in range.
But my endo will review reports from Tandem pump/cgm and often sees trends I miss, and may suggest tweaks to pump settings.
It would be different for those wo CGM and relying on different data such as finger bg check or just A1C.
Yup, my xdrip alarms are set to 70 for low and 127 for high.
Much better to be at 68 one quarter of the time than 178 all of the time.
If I was to design a blood sugar measure, I think it would take a range for healthy non-diabetics as the actual range and have a score that increases exponentially as the blood sugars get further away. As to how that exponential increase would be configured, that’s debatable.
Yes! The TITR, 70-140mg/dL, that several endocrinologists were advocating at recent international conferences, is pretty close. I am so glad to see this shift, personally! Above that is when macrovascular damage occurs.
And I totally agree with you here, as data show that non-diabetics regularly go down to 60/65 mg/dL when fasting, and are fine: