Happy Saturday, and welcome back to our weekly advice column, Ask D'Mine! hosted by veteran type 1, diabetes author and educator Wil Dubois.

Just last week, Wil looked at how long we PWDs (people with diabetes) could generally survive without insulin in our systems. Today, he explores how long a bolus dose of this BG-lowering elixir might last in your system once given. Hint: it's not cut-and-dry!

{Got your own questions? Email us at AskDMine@diabetesmine.com}


Amanda, type 1 from Oregon, writes: Wil, First off, I love your advice and really appreciate everything you do! I've had a Medtronic pump for about 4 years now and I have a question about the active insulin time. I have mine set at 4 hours and my endocrinologist says that's pretty standard for most of his patients. However, sometimes I think this is too long. If I test and have a blood sugar of, say 150 and it has been 3 hours since I gave myself insulin and ate, my pump says that I should take 0 units because I still have insulin on board. Sometimes I'll go ahead and give myself a unit to bring it down because I just think my blood sugar is probably not going to come down considerably in the next hour, and may actually go up. I actually have a CGM now, so I can now see better when I have peaked after a meal, and whether I am coming down or going up still. My question is, is there a good way to figure out what your actual active insulin time should be? I'm sure it varies from person to person. I think I may need a shorter time but I can't think of a good way to test this.

Wil@Ask D'Mine answers: Thank you! Actually, there are several good ways you can figure out your active insulin time, each variations of the same theme—but first a little background. For those of you not in the know, active insulin time (also called duration of action) is a measure of how long insulin has the ability to lower blood sugar in your body. It matters for any insulin user, but especially for a pumper, because it's one of the metrics the pump uses to calculate how much insulin to give us under various circumstances.

Innovation 2015

Now, it needs to be said upfront that all the frickin' pumps are different in how they track and deal with this active insulin—and almost no one properly understands how their own particular pump works in this regard. Some pumps track insulin action in a curvilinear way, others in a linear fashion. Some deduct the active insulin only from corrections. Others from meals and corrections. Still others only depending on how much below target you are. I don't have time to get into the details on each pump today, plus it gives me a headache every time I think about it, but I promise to spend a whole column on it... someday. Maybe in 2017.

Anyway, the current batch of Med-T pumps like yours let you set active insulin time anywhere between two and eight hours. Why? Is it because no two persons with diabetes have the same active insulin time? While that is true, I suspect the real reason is that no two endocrinologists can agree on what the active insulin time should be.

Like your doc, I find that four hours works great for most of my pumpers, but for what it's worth, the leading endo in our state is insistent that active insulin should be set for 3 hours in all patients. This has led to a several-year-long passive-aggressive "battle" between us. She re-sets all my pumpers' active insulin time every time she sees them, then I change it back again the next time I see them. But other notables, such as CDE John Walsh of Pumping Insulin fame, go the opposite direction and argue active insulin time should be set at six hours. WTF???

And actually, your pump ships from the factory with its default active insulin time set to six hours, because, according to the pump's manual, this "most closely matches the published scientific data."

Really? Published where, pray tell?

Maybe they were looking at this famous graph:

insulin curve(1)

That one comes from a Novo Nordisk study. And while it shows that in a test tube, fast-acting insulin still has some ability to function after six hours, I'm not personally convinced that in your body it really has enough punch left to have any realistic effect towards the end of its run. That last couple of hours looks pretty flat to my eye. But don't take my word for it. Just read the fine print. In the prescriber's info sheet for Novolog, it states quite clearly that in an FDA approval study for that drug Novo found the duration of action of Novolog is three to five hours. That's a lot of variation in a pretty small sample size of 22 type 1 adults. Or in Novo-speak: the action of insulin "may vary considerably in different individuals." Oh. But it gets even better. Guess what? The famous duration of action graph? It's the average data from those 22 folks.

Of course, to his credit, Walsh is quite correctly worried about the risk of insulin stacking causing lows in some brands of pumps, so he's erring on the side of caution. Still, why rely on published data about other people when it is within your power to easily figure out your very own personal active insulin time?

Here's all you have to do: First, download your pump and CGM data. Next, fix yourself a double of your favorite drink that gets you in a relaxed mood. You are going to spend some considerable time with string theory. Not to worry, this isn't particle physics! I'm talking about the little strings on the computer that show you the drunken wanderings of your blood sugars throughout the days and weeks before your download.

Now, I want you to pay specific attention to trace lines after meals. In a perfect world (Ha!) your trace line will curve smoothly upwards after a meal, gently crest, then return to its baseline in about four hours, leaving behind a bell curve that would make any statistician proud.

Don't hold your breath for that.

If your correction factor or insulin-to-carb ratios are wrong, set a little weakly, or if you are off in your carb counting skills, you might not be getting enough insulin and you won't return to baseline. But for our purposes, that won't matter. Pay attention to the tail of the trace, before it reaches the next bolus or meal. Does it flatten out? Or is it still dropping? If it's still coasting downhill, it's still doing its thing and a has not reached the end of its active time. If it's flattening out, the insulin has run its course. Look at the number of hours between the bolus and the flattening out of the curve's tail to get your active insulin time. How simple was that?

Oh. But, if it's starting to rise a little like you reported to me sometimes happens, then things just got more complicated. Because a rise three hours downstream of a meal is unlikely to be caused by the meal—unless the meal was something with a ton of fat in it like a family sized, double-meat, extra-cheese pizza that you ate all by yourself. (Hey, it happens.) If it isn't caused by the meal, then it must be caused by something else.

Can you guess what?Active Insulin

Well, in the absence food, flu, fleeing from feral flamingos, fending off a fennec fox, or fighting flying fish... sorry... I got carried away... In the absence of all the things that we know can cause blood sugar to go up, the only remaining culprit is too little basal insulin. A rise three hours or more downstream of a typical meal usually signals a deficit in basal insulin during that time period.

By the way, Walsh himself recommends testing the insulin action time in a different way. He advocates running what he calls a "clear out" period with no bolus of any kind for five hours and no food for three hours—citing research that says most carbs only raise BGL for one to two-and-a-half hours—that leaves you above 250 mg/dL. Presumably, two hours into the no-bolus zone, you are to eat a small amount of carbs to raise your BGL. Anyway, the idea is to be at a "clean" 250 with no bolus or carbs in play. That sets the stage. Then you take a correction and wait and see what happens. The beauty of this approach is that you can kill two birds with one stone: you are testing your active insulin time and your correction factor at the same time. Walsh says to do this test three times to "ensure accuracy" and has all the "ask your doc first, blah, blah, blah" disclaimers attached.

Gary Scheiner, in the re-issue of his seminal book, Think Like a Pancreas, offers us another alternative. He suggests that you "label your insulin with radioactive dye and see how long it takes for your body to stop glowing."

It should be pointed out that Gary has a pretty damn fine sense of humor.

His back up keep-it-simple plan is "check your blood sugar every thirty minutes after giving a correction bolus and then see how long it takes for the blood sugar to stop dropping." He advises no food, boluses, or exercise until you flatline (so to speak); as to remove any variables that might throw off your analysis.

Well, I've run out of time, Amanda (in much less than three hours!), but is it possible your active insulin time is less than what's "standard" for "most" of your endo's patients? Damn right, it could be! But only time will tell. My advice is to take Walsh's advice. The "ask your doc first, blah, blah, blah" advice that is—and then run any of the tests above to find your very own active insulin time.

This is not a medical advice column. We are PWDs freely and openly sharing the wisdom of our collected experiences — our been-there-done-that knowledge from the trenches. But we are not MDs, RNs, NPs, PAs, CDEs, or partridges in pear trees. Bottom line: we are only a small part of your total prescription. You still need the professional advice, treatment, and care of a licensed medical professional.

Disclaimer: Content created by the Diabetes Mine team. For more details click here.


This content is created for Diabetes Mine, a consumer health blog focused on the diabetes community. The content is not medically reviewed and doesn't adhere to Healthline's editorial guidelines. For more information about Healthline's partnership with Diabetes Mine, please click here.