-----Flatliners Club-----


My thinking about temp basals has changed since I’ve been living with a continuously variable basal system. Before hooking up to my current Loop set-up, I had a hard time believing that a temp basal rate could have any influence in the short term.

My Loop system shows my BG trace from the present time and back several hours. It also shows a projected BG going out for four hours. I’ve watched time and again as Loop makes a temporary basal change. At first I thought that the projected BG line at about two hours and more would change, but not the near-term zero to two hours out. I do often see the near term forecast BG respond to basal rate changes. I found this puzzling since I know my insulin, Apidra, does not peak for 60-90 minutes.

While the insulin does not peak for an hour or more, its onset is 15-20 minutes. How can the onset of insulin action move a strongly trending BG line? The answer, to my understanding, is that it can’t. But what about a lightly trending BG line, whether flat or slowly moving up or down?

If my blood glucose is well balanced, then even a small force, like the insulin action temp-basal onset, can move the direction of near-term BG trend. I think of it like one of those finely tuned balance scales. If you place exact weights on both sides, the scale balances. If you add even a small amount of weight on either side, the scale tips in that direction. So, if your BG trend is nearly balanced, then I think temp basal changes can exert significant influence.

Well, that’s my theory.


As always YDMV. I did not used to nearly as useful a response from suspending my basal until last year when I changed my diet fairly significantly. I think my liver response improved at that time, so temp basals work for me for minor lows. But your quick rise when disconnected would make it harder, I agree.


Ok. Haven’t seen a day this good in a while - just had to post it! TIR: 100% with BG Average 97 with an SD of just 6:

My 14-day graph wasn’t too bad, either - BG Average 105, and an SD of 22. TIR was 90% (0% low).

My challenge now is to lower that average – I’d really like to get it back below 100, but have not been having a lot of luck. Seems like I keep gravitating to a sorta ‘baseline’ of 104-106. Any ideas?


I’ve recently enjoyed seeing my average drop below 100 (5.6) when I cut out some snacking. I’m in the middle of reading Adam Brown’s Bright Spots & Landmines recently released book. One of the habits he credits with helping a lot with his overall control is eliminating eating anything after dinner. (Except for treating hypos.)

Now this is not a new idea and one I’ve recognized the value of for a long time. Actually doing it is another thing for me. Reading about someone else using this to great advantage has stoked my motivation. I suspect that you may be able to get your average down to below 100 if you identify one or two things that could help. Your line above is exactly how you can get to what you want. Do more of that! In any case, your present control is acceptable.

I don’t mean to be a smart-a$$, but you get the idea. By the way, that is an awesome line! I don’t think I’ve ever recorded a 24-hour line with a single digit SD.


Oh, man, I envy you – most nobly! Congratulations on this great day/fortnight.

(SD is the bane of my personal diabetes battle.)


This is one bullet-proof line and stats that no endocrinologist, even a reflexively hypo-phobic one, could criticize. 0% low!


I agree with you about the importance of glucose variability as measured by standard deviation. I rank SD in value as higher than BG average because in order to bring down the average, SD must be brought down first or the risk of hypos goes up.

Good luck with battling to rein in your SD! It’s a worthwhile struggle.


Wow! That’s so cool, @Thas, I thought daily SD of 6 would be outside the realm of possibility for anyone, not just PWD. Great lines overall, and thanks for keeping this thread alive - it’s inspiring and uplifting.


Six! Wow! One of these days you’ll post a standard deviation of 0! :wink:


I’m reading this now (about 2/3rd done) and highly, highly recommend it.


@Terry4, what is your definition of “near term” BGs? I find that any basal changes take at least half an hour to kick in and most of the change hits at one to one and a half hours, which is why I find it very easy to accidentally overshoot and end up high. For me, in treating lows, half an hour is too much of a delay compared to glucose tablets, which take half that time or less and have no lingering effect an hour or more later.

I have used temporary basal rates to decrease the amount of food I need to eat if I’m plummeting and notice it in time (before I’m actually low). I think that’s about the only time I use a 0% temporary rate. I do often use 70-90% temporary rates to deal with increased activity and exercise. I find anything below 50% hard to manage in terms of not ending up high later on.


I’m thinking of BGs in the next 30 minutes. I agree with most of the change occurring in the 60-90 minute timeframe, but I’ve found that even the small force represented by the onset curve is enough to influence the BG line in the near-term. That implies a line that is not changing by more than 1 or 2 mg/dL/5minutes (0.06-0.11 mmol/L). If the glucose trend has much of a slope to it, like > 0.28 mmol/L/5 minutes (5 mg/dL), then the visible effects of the leading edge of a temp basal change are blunted or even wholly obscured.

I think changing basal rates manually is hard to do in response to CGM changes. There are probably people who do that much better than I could.

Slightly off topic but something I found humorous when I heard Stephen Ponder say about insulin pumps in a podcast. He said, “The pump is your minion.” If having a pump means I have a minion, does that mean I am master of all I survey? :wink:


Today I decided to try the temporary basal rate to avoid an impending low. I saw that I was at 4.2 mmol/L (confirmed by meter) at around 11:00 and had 0.5 units of insulin on board (or thereabouts). I ate one glucose tablet for the insulin on board and did a 0% temporary basal rate for 60 minutes. You can see from the below image that, although I did prevent a low (likely because of the glucose tablet), I had a large rise in my blood sugar about an hour after treating the low (likely from the 0% basal rate).

I did some googling and there are several studies that have found that blood glucose rises by 1 mg/dl for every minute of time without insulin delivery for people with Type 1 diabetes. This seems to fit almost exactly with the amount I rise without insulin delivery: today I rose by about 3.5 mmol/L or 63 mg/dl for a 60-minute suspension of insulin delivery (except when exercising, when I can rise significantly more if disconnected from the pump).


That’s a useful scenario to capture and describe, Jen.

A 1 mg/dL per minute rise when disconnected from a pump is a nice round number to remember. So, for mmol/L users like yourself, that’s 1 mmol/L every 18 minutes disconnected or about 3.3 mmol/L per hour.


I’d think of it as rising 0.05 mmol/L every minute. That way it would be easier to decide how many minutes I may want to suspend my pump for (ten minutes to rise 0.5 mmol/L, for example). I’m going to try to use this as a strategy for giving a little bump to BG rather than eating. I’ll continue to eat if I’m low or near-low, though, since the basal change seems to take an hour to kick in. But this could be useful if I’m skirting my low limit and want to bump my blood sugar up by 0.5 or 1 mmol/L.


I thought I would post my past week of using low carb (I’m not keeping track exactly, but likely around 50 grams a day) and pumping with Fiasp. They seem to be the perfect combination, though I do have concerns about Fiasp causing more irritation than other insulin seems to.

My biggest problem is overnight. Some nights I flatline perfectly, but other nights I rise or drop and spend the rest of the night high or low. I can’t figure out what contributes; it’s nothing obvious like food or exercise having any kind of straightforward effect. I do struggle with after-dinner readings, too, so I’m trying to sort those out (it’s hard to tell how much is from food, how much is from basal, or whether it’s just the fact that I change sites in the evening and so that’s the time I start running high if a site is done). It doesn’t help that I don’t wake up for Dexcom alarms, either, so by the time I wake up I’ve usually spent hours high or low. It would be nice if I’d wake up so I could correct the situation and move on.


I am following a low carb diet & using Regular insulin. I am trying really hard to get a flat line but:-

After a meal, my 1 hr & 2 hr readings are great, but by 3 hrs, my BS levels slowly increase to around 140. Once it has risen to 140, it does not come down at all unless I take a Humalog correction.

What should I do to mitigate this? Should I just continue taking Regular & Humalog correction? Or should I mix Regular and Humulin-N?

I’m not on a pump, but I am considering going on one.


Sorry I missed your question here, @the_explorer. I’m glad you reposted as a separate topic and received some help there.


Ok. Been a little while…

Last weekend, I did the “40-mile” route on a Tour de Cure ride on a humid 94-degree, sunny morning. While my cycling performance needs LOTS of improvement (I had to cut a few miles off the ride with a shortcut, went painfully slowly, and still barely completed the ride!), my BG “performance” was quite good! Average for the day was 97, 100% in range and an SD of just 11:

Started the ride with no carbs and, therefore, no bolus insulin on board and a BG of 107. That dropped fairly quickly to 88, but stayed right at 88 for the rest of the ride, with two exceptions - I had around 10g carbs at one rest (which I didn’t “need” but felt like I wanted to be cautious) and I had a huge apple at the last rest stop - which I also didn’t need from a BG-perspective, but I felt like I had “nothing left in the tank” energy-wise, so I ate it. My BG rose to 124, then went right back to… 88.

Last 14 days were somewhat less perfect (not surprisingly, with all my stress lately), but not bad. Average of 104 with an SD of 22 and TIR of 88% – though there’s a 1% ‘low’ in that, and at least part of that was due to a failing sensor:


Nice work on the bicycle ride lines, @Thas. While you ran out of gas with your performance, you knew that it was not because of your blood sugar control. Your 14-day line continues to be excellent. Have you looked at the AGP report yet?