Basal in the absence of food

Interested in what your thoughts are on basal. What should it do in the absence of food?

Well, I think it should keep you flat in the absence of food within reason. But if you are fasting for 14 or 18 hours – should basal keep you flat? Your underlying metabolic rate has changed!

You are now in starvation mode! Even a person without diabetes is likely ramping down their insulin production during this time and cranking out glucagon to prevent BG from dropping too much! And yet I’ve definitely had people on other sites basically say “it doesn’t matter if your son didn’t eat anything for 10 hours, basal needs to keep you flat without food.”

And I guess I think that with toddlers especially, the speed at which their bodies transition from “business as usual” to “starvation mode” is a lot quicker than in an adult. At least with our son, we’ve found that the basal rate that needs to keep his BG steady in the absence of food can sometimes be hours of ZERO – but that he is busily producing ketones at an elevated rate at these times. So clearly there are a lot of special cases and exception.

That’s not to say that you should hard-program into your pump a basal rate that forces you to drop like a stone and eat every two hours or rise like crazy… in general whatever you program in should give you leeway to not eat for an hour or two beyond your scheduled time and be fine. But I think for us, the general principle shouldn’t be that basal must absolutely and always keep you flat in the absence of food, but instead that “the optimal basal rate is the one that, on average, requires the minimum number of interventions, tweaks or monitoring to keep BG steady throughout the day.” Which is a little bit different. Does that make sense?

I also think the amount of insulin needed to keep you flat in the absence of food is NOT necessarily the same as the amount of insulin needed to prevent your liver from dumping glucose in the minutes to hour after you start eating… this is just my pet theory of course based on my own kid, but because T1Ds do not produce incretins, they are not producing the ancillary insulin dump prior to food reaching the stomach, which in glucose-normal people immediately works to rein in the sugar dump as soon as food touches our mouths. My guess is that, for instance, the reason @ClaudnDaye has found early extended bolusing and temp basals prior to a meal work so well to rein in spikes is because it’s essentially mimicking this early spike of insulin that normal bodies produce.

Does that make sense?


Yes it does make sense. I would have never thought that keeping flat for 10 hours would be a target, and glad I don’t know anyone who is advocating that. Of course, we skip or postpone meals to test our basal rates from time to time, but honestly, enough is changing day to day even in our teenager, where some days when he postpones a meal he is flat, and another day it doesn’t work that way. Every month or two we postpone a meal, and check if the basal is set approximately right and move on. Glad we don’t have to fiddle with it more.


@TiaG, it suddenly occurs to me that in toddlers, you must be able to see phenomena that are obscured in adults, but uncover some fundamental mechanisms that are true for all, but can be seen in toddlers because they have a lot less “auto-recovery” capacity.

So understanding what happens in toddlers may actually give a much better understanding of why and how the body reacts in general. I had never realized it until I read your post. You probably are getting a finer understanding of many aspects that are obscured to us. Of course, that comes at a price.

1 Like

Basal for MDI’ers is really a tricky thing. In my experience it is almost impossible to titrate it precisely enough to be able to go without eating for a whole day.

Back off a few units and you’re high all day. Add a few units and you’re chasing lows all day. Get it just right and you need a couple of snacks in the absence of meals.


@Chris, you’d be surprised how dogmatic folks can be. I butted heads with someone in the past on the basal rate for the morning for Samson.

In general, we assume Samson is eating breakfast by 7am. He goes to bed at 8 pm and hasn’t usually eaten food since 6pm. His basal rate always keeps him steady at night and in the 1-2 hours after he wakes up. But on the weekends maybe we’ll go to brunch and then that means he doesn’t eat till 10am or even 10:30.

Should we be programming his basal rate so that he doesn’t drop if he waits 3 extra hours to eat – meaning he’s now gone like 16 hours without food?

On another site someone said yes! And when I told them to stop responding to my threads with concern trolling and claims that this basal rate was “unsafe” and contrary to medical advice (which, by the way is silly since it was suggested by my Endo) they told me they wouldn’t because they felt the information was important to convey to other users…

1 Like

During the day we would do an 8 hr fasting test. First 4 hrs are really prep to clear the food/bolus (for the most part) out of the way. Then the next 4 hrs are to see if the BG stays straight. I don’t think too excessive.

Overnight is longer. Maybe 12~14 hrs? Ideally we should be able to have dinner at 6 PM and breakfast at 8 AM and have BG where we want them between based on the dinner food/bolus, the overnight basal and taking into account the DP in the morning.

The difference between school days and weekend days is challenging but doesn’t seem to actually be as big of an issue as one might assume. CGM certainly makes things much easier. But if no CGM and sleeping in on the weekend then there will be a BG check in the morning around 6AM on a sleeping individual.

1 Like

During the day we would do an 8 hr fasting test. First 4 hrs are really prep to clear the food/bolus (for the most part) out of the way. Then the next 4 hrs are to see if the BG stays straight. I don’t think too excessive.

I can see this being useful for an older kiddo. Basal testing like that is probably not that difficult once you have a teenager or even a 6 or 7-year-old but it’s pretty hard in toddlers… they have so many snacks built into their day that it becomes very onerous and disruptive to eliminate them. i think that’s the other problem with that dogma – it’s basically impossible for little little kids to do true basal testing except overnight. We tried a few times and first off, the results never produced optimal BGs and second off, it was a pain in the butt for poor Samson.

Now our basal testing is basically just “delay eating lunch for an hour or two – does he drop, rise or stay steady?” Really that’s all we care about because realistically that’s the main scenario that we encounter. with any frequency. He is never going to go 8 hours without eating during the day unless he’s sick, and then we have a totally different issue to deal with…

1 Like

I will say that those days where we delay a meal, or he doesn’t feel like eating (I know in a teenager nonetheless) I always feel like it is magic when his bg stays level during the fasting time. Feels like we won something important.

1 Like

IMHO the purpose of basal is not necessarily for the event that fasting were to occur during the day. Rather fasting during the day is an extra event thrown in to allow the basal testing. The proper basal rate for time of day is the starting point of the other settings in the pump. All the other settings are determined with the starting point of the basal in place for that time of day.

Overnight fasting is somewhat different in that this is typically an expected and normal situation.


The proper basal rate for time of day is the starting point of the other settings in the pump. All the other settings are determined with the starting point of the basal in place for that time of day.

We definitely use this general logic – for instance I always troubleshoot any highs or lows in the morning before moving on sequentially to the next part of the day. But for us the testing method isn’t taking away the food, it’s just whether or not Samson either a) went high or b) needed a correction. It is trickier to debug but there are small clues you can use to decipher whether it’s basal or bolus that’s the issue – for instance if you shift the meal by an hour or two, does the low or the high persist at the same time or seem to shift with the meal? That kind of thing.

1 Like

I have a special few words I use on occasion for people like this. :wink: It has to do with what they can do and where they can take their unwanted, unwarranted and trolling comments.

1 Like

Add me to this list. The only time we can successfully test is during the night. We’ve tried many times and always had to correct in one way or another during the testing, which resulted in failed test. I have a notebook practically full of “failed test cases.”

1 Like

I am a bit confused on this. Can you clarify this? I would think that since it is only occasionally, you don’t need to change the permanent basal rate. But on the weekend when you know breakfast is going to be late, you would just make it a temp basal for that one particular morning, right?

Was the person posting stating that you should change the basal settings for everyday, just because of the occasional weekend?


1 Like

He was claiming that because his blood sugar would drop in the mornings when he gets a higher basal rate IF he doesn’t eat for 3 hours, that is too high. Basically that the weekends are offering a chance to do basal testing of the morning basal rate and it failed.

So let’s say his basal rate is 0.075 units/hr at night and then at 6am it ramps up to .375 units/hr. Usually that keeps him steady till up to about 9am, but if he doesn’t eat till 10 he will start drifting low and we have to hand him a few saltines. This is a very easy and low-stress process, by the way, but this guy was saying that’s a sign the basal program had failed.

We don’t have two different basal schemes for weekdays versus weekends because our pump doesn’t have that option, unfortunately. We use openAPS to set a high temporary target on the weekends, which effectively gives him al ower basal than is programmed.


Gotcha, that makes sense.

I am guessing the guy would hate me, because I have a “usual” basal profile, but I am making changes to it all the time, percentages up or down for most of the day, based on different days being…different! Some nights, a normal 100% basal works fine, and other nights I go down to 40%. But that doesn’t mean my profile is wrong, it just means it is wrong for that particular night.


Just means you’re doing whatever it takes to use insulin for basal that doesn’t work well for basal insulin… no easy task…