Got questions about navigating life with diabetes? Ask D’Mine! Our weekly advice column, that is, hosted by veteran type 1 and diabetes author Wil Dubois. This week, Wil is tackling a question about fluctuating glucose levels and how closed loop systems (you know, so-called “Artificial Pancreas” technology) might be a way to get a handle on blood sugar woes. Read on…

{Got your own questions? Email us at}


Nancy, D-mom from Wisconsin, writes: Continuous glucose monitoring devices use interstitial fluid to measure glucose. It is known that interstitial glucose has a delay relative to blood glucose of up to 15 minutes. In addition, there is an accuracy problem with all measurements, blood glucose or interstitial glucose. If a diabetic is relatively stable, maybe I can see that a closed loop system using them would work. However, my child and others that I know are not stable. Would you please comment on this?

Wil@Ask D’Mine answers: You bet! Actually, I can make several comments, and my first is: Don’t kid yourself. There’s no such thing as a “relatively stable” person with diabetes, at least not when it comes to blood sugar, and maybe not when it comes to anything else, either. But this lack of blood sugar stability was a big shock to the research community in the early days of continuous glucose monitoring (CGM). They now could see that we PWDs (people with diabetes) were all over the place — much more so than anyone had ever realized before, as no PWD in history had ever been given enough test strips to reveal the inherent chaos under our skins! So, on the bright side, your child is not an outlier.

But the good news here is that the closed loop development crowd knows this fact, and the algorithms in development are up to the task of riding this bucking bronco of real-life unstable blood sugar much better than you’d suspect. More on that in a bit.

Of course, your point on accuracy is dead-on. No matter what consumer grade technology you use to test blood glucose, the results are questionable. That said, the improvement in CGM accuracy over the last decade is nothing short of mind-boggling, and that accuracy has evolved at a rate that puts our long-revered fingerstick tests to shame. It’s not for no reason that the FDA has cleared some CGMs for use to replace fingersticks. They are that good, or, depending on how you look at it, fingersticks are that bad. Either way, I think it’s safe to say that modern CGMs are nearly as good as fingersticks in any single-point measurement.

But it doesn’t end there. Because a fingerstick, even if it were by some miracle or stroke of luck 100% lab-grade accurate, still doesn’t tell you much, does it? A perfect—but isolated—measurement of sugar in a population of people who have inherently unstable sugar isn’t really all that helpful. The power of CGM is being able to view not what your blood sugar is, but what it’s doing. Is it going up? Or down? Down fast? Down really, really fast?

Fingersticks are a photograph. Sure, maybe a beautiful, framed fine art black-and-white print, but  just a still photograph nonetheless. CGM, on the other hand, is more along the lines of “Diabetes: The Movie.” Even the early CGMs, which had rather piss-poor accuracy, still provided superior situational awareness for CGM pioneers like me, and that awareness of the tides and surges of our blood sugar is the best tool we have to attempt to stabilize our unstable diabetes.

As to the delay between interstitial glucose (IG) and capillary glucose, it doesn’t actually matter that much. True, the IG has, as you said, a more or less 15-minute delay when compared to glucose measurements from a fingertip (which is still delayed compared to glucose in the brain, where it matters). But to understand why it doesn’t matter we need to talk about trains. Yep. Like choo-choo trains.

The engine of a choo-choo train gets to any given point on the track before the caboose does, but they both travel the same path. Likewise, even though capillary glucose will be ahead of IG when the sugar is changing rapidly, the IG behind is on a more-or-less predictable path along the same track. Actually, come to think of it, for us PWDs, maybe a roller coaster would have been a better visual than a choo-choo train, but you get the idea. Anyway, the folks working on closed loop systems also know this, and have designed their software to account for the “lag” in IG.

And actually, our current partially-closed loop systems work surprisingly well, especially given that they use only one hormone—insulin—either restricting or increasing its flow to attempt to keep blood sugar variations in check. Early on, before the current systems, my money was on dual-hormone pumps: Systems that could deliver insulin to lower glucose as needed, and glucagon to raise it, also as needed.

That made logical sense to me: a hormone and a counter-regulatory hormone for perfect homeostasis. That’s what the body does in non-diabetic folks. But the stable glucagon needed for such systems has been slow in coming, giving the insulin-only systems a chance to prove themselves—which they are doing marvelously. And like the rest of this D-tech, they’ll only get better and better over time. Will they reach perfection? I very much doubt it. But the day when the closed loop far outstrips the efforts of the most diligent and dedicated PWD using conventional tools is not far off at all. I think that your child, and all the other “not-stable” children (and adults) out there will do much better than you can imagine as this tech matures. 

And ironically, the more unstable a person’s blood glucose is, the more I think they will benefit from a closed loop. After all, if you somehow managed to be relatively stable, you wouldn’t necessarily need a closed loop, now would you?


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. Bottom Line: You still need the guidance and care of a licensed medical professional.