I am feeling REALLY overwhelmed on this particular Monday morning, so I figured I would give you guys a shot. (And for sharing my angst, you’re welcome.)
This all started over the weekend while I was having a debate with a fellow Minimed 670G user over what the profanity auto mode is thinking and doing. This fellow user, assuming he’s not really a robot (or a Medtronic agent) swears he’s found the “sweet spot” (his words) with auto and is now cruising along “Perfect BG” Highway. I understand people will have different experiences and that he really may have found the secret, but I also understand that in this faceless communication, there is WAY too much context that can’t be shared. The very first piece, of which, is what defines “excellent” control. There are people who are okay with 200s. There are people who are not. They might not share the same definition.
I’m pretty wound up. I’ll try to get to the point. The debate really was over whether or not adjusting your ICR settings would fix the problems in auto mode. My argument is that it’s a global change laid overtop of an uneven basal delivery. His argument is that our basal rates are almost ALWAYS (all of us diabetics) too high. Our ICR settings are almost always too low. Their relationship is very important and our basal rates are really set to clean up the mess we’re making with our light ICR settings. Auto mode is just paving the way for us to get our settings right. (He then went on to call my Auto Mode experiments “worthless”, and I’m still burning, as is evident).
MY argument is that in order to determine whether or not a settings change is appropriate, it’s best to start with establishing a baseline. This is what the experiments are for. To see what auto mode is going to deliver before I start messing with things. And so far, auto mode is a DISASTER and often doesn’t deliver nearly enough. But I decided it was time to get back to basics, and I picked up my copy of “Pumping Insulin” to refresh AND to learn. Now I’m surrounded by everything basal and bolus, super bolus and super micro bolus, OpenAPS and closed loop, artificial pancreas and auto mode… I’m near toppling over.
I don’t even know what the question was. In fact, I’m not sure I had one. What is all of this, and what do they know??? Medtronic is just the first one to get its closed loop system out there, right?? But all of these DIY things?? Are they the same premise? Do THEY work? Is it possible the problem really is ME?
There it is. I DID have a question. It’s loaded… I’ll just warn you.
I’m on MDI but I’ll give you my opinion regardless.
As a general rule with most complex machines you just have to follow the directions, stop trying to overthink it, and just let the machine reach it’s homeostatsis. Tinkering with it implies that you know more about it than the manufacturer, which I don’t think you do.
Relax, let it do its thing, and you’ll be steady at 120. If 120 is too high for you, then the 670g automatic mode is not for you. In that case look into other options, like OpenAPS, Android APS, or other open source options.
I don’t find it implausible that it works well (perfectly?!?) for someone. Calling your experiments worthless is pretty rude. Telling you what YOUR basal and ICR settings should look like is beyond insulting. I think this thread pretty much sums up the frustration of the auto mode for someone with near normal glucose control.
For most people, who have high A1c’s and tolerate 200’s well, I have no doubt the 670 is a godsend. For those used to normal bg control, it seems that the inflexibility of the system can frustrate to no end.
I don’t think the problem is you. I have read enough of your posts to see a sharp, engaged, and proactive mind.
I have heard a lot about the 670G. There are people it might be great for. But the fact is, it has a target of 120. Would you be content if your pump tried to keep your BG at 120, and if you went to 110, it would try to adjust that back up? I would not be happy with that. As an example, this morning my CGM said 81 but 2 consecutive BG tests had me in the 120’s. If my pump was listening to my CGM, I’d be pretty pissed at that! (pardon the terminology)
About the specific question of if your ICR would affect how it works in auto-mode, I do not know how they use ICR for the basal adjustments, and I do not know the algorithm. And most likely neither does that other 670G user, unless he works for the engineering team at Medtronic. They may use it to affect how it adjusts basal, they may not. Who knows?
That’s funny, he’s never met me! How amazing that he can know so much about me.
There are a lot of advantages to having a strong basal foundation.
It’s hard to know exactly what he meant and the context, but I can tell you this about ICR’s - it really depends on so much. That is the problem I find with formulas and this pump in general - assuming your ICR is always the same is a recipe for frustration. Are you always the same for every lunch? I know that I am not.
I can’t offer you much for your debate with this person, but if you want to specify the issue you have with meals or whatever, I can offer my opinion on things. And you may have already posted it, but again, I am not savvy enough to keep track of all the peeps!
In my “experiments” (and really, what is insulin therapy except a never-ending series of experiments?) with auto mode, I found that AIT adjustments had the most obvious effect, and ICR was more ambiguous. I was told that ICR settings could only affect what happens with the bolus wizard, which makes sense. But at the same time, the way the whole system works tends to blur the distinction between basal and bolus. In fact, the term “basal” isn’t used–it’s called the “microbolus” rate. And AIT itself is somewhat detached from the question of how long insulin is active in your body–almost everyone has to alter that setting for auto mode from whatever it was in manual. In my case I bottomed it out at the 2 hour limit, and still wasn’t getting the results I wanted, particularly in the dawn phenomenon period, so I decided to try tweaking ICR at that time since that was the only other parameter I had any control over. Even though I wasn’t taking any wizard-calculated meal boluses at 3am, it seemed like maybe it was making a difference. But it was hard to be sure because any changes you make take a week to get integrated into the algorithm, which makes it hard to determine the results and adjust accordingly. And that was about the point at which I finally decided it was causing me more aggravation than it was helping and I went off of it.
It’s a good 2 cents. What’s funny is when I started on the 670G last June, I had zero expectations. I didn’t tinker with anything, and I did follow the rules. In fact, I never stopped following the rules and never did start tinkering. All I do now is observe. The reason I own an insulin device of which the main attraction is nothing more than little personal science lab is because it never did what it was supposed to… or at least not much of it. Yes, 120 is the target, and in all honesty, I’d be okay with a 120 if it were an easy one. (If you want to make a 670G wearing-diabetic fall out of their chair in laughter, tell them to “relax and let auto take care of it”). But it’s neither easy NOR a 120. It turns out if you’re not willing to bend the rules or tinker quite a bit, this thing really runs you high. It doesn’t just not handle your highs well, it actually sends you high and THEN doesn’t handle your highs well. The only time I consistently saw a 120 was over night, and that would’ve been great IF 1. It were really a 120 (because we’re talking a Medtronic sensor, don’t forget) or 2. It didn’t mean you were probably going to spend your morning at a 285 because what the pump DID over night in order to keep you at your 120 (aka 150) was to withhold insulin, therefore setting you up for a wicked morning BG climb. Wait… but that was just MOST nights. There was a small percentage of the time it actually delivered a “full” amount, even sometimes delivering extra. Without having any idea what amount of insulin you had received over night, it made every morning a cliffhanger. Because the disease doesn’t have enough surprises of its own.
Now this WAS all without tinkering. And this WAS staying in auto at all times with the exception of a sensor change for about a month and a half to about 2 months. So actually tinkering seems to be what’s in order to have any kind of success with this thing, and I don’t much like it. I don’t like the idea of having to change things to change the outcome when that outcome is a result of the pump’s design. I certainly don’t like having to trick my insulin pump into getting more insulin.
I’m not directing this at you, despite all of the “YOUs” strewn across it. I’m just in full vent mode. I appreciate your response, and in fact, I feel like it validated a good part of my frustration. I don’t know more than the manufacturers for sure, and I shouldn’t have to in order to get good results.
I also don’t find it implausible… I mean, people out there win the lottery, right? See, I’m feeling better already…
I know there are people who are going to do fine on this thing, and, yes, there will even be those who do great. I understand that. There’s this very bizarre dynamic at play here in the 670G world, and it’s actually quite ugly. There’s what we call “manual mode shaming” or “auto mode cheerleading”, and although those are terms coined in good humor, they’re real as all get out. There are support groups where people are told they’re not supposed to speak negatively about the pump. If you actually voice some concern over the fact you’re, say, hitting 300 every morning, you’re told you’re not trying hard enough, you’re not washing your hands, you’re not carb counting, you have (only recently) a pronounced dawn phenomenon, and basically, you’re told to pipe down and deal with it. There are thousands of people in one room, and they’re badgered if they’re not spinning a complete success story. This is why I started my own group… I was APPALLED by it. There were parents of young kids in there… there were PREGNANT women. There were grown men who would leave comments like “attention moderators, please don’t take this as anything negative or as criticism of the 670G, but i’m just really concerned over these highs”… I hate to be so cliche as to make an Alice in Wonderland reference here, but a hardcore drug one just seems so harsh in a family joint.
For the most part I’m surrounded by good people in my group. Great people. Most of them have returned to manual mode. Many of them have returned to Dexcom. Some are beating their heads trying to return their pumps. I don’t allow in my room what I witnessed in that other room, that minimization or shaming or dismissing. In that other group when I expressed my concern over my highs of 300 for HOURS despite having not eaten, I was told not to worry “over a few highs”. When one of my members DRANK PANCAKE SYRUP because he was receiving .3 and .4 micro boluses with a blood sugar of a 30, he received a “LOL. I also sometimes overtreat my lows, too.”
Alice in Wonderland. She keeps popping up.
You guys are very nice. I brought drama from another world over here into this upbeat, light-hearted, intellectual exchange. I won’t make a habit of it. This was a tough weekend. I actually blocked the member, and it upset the junk out of me to do it. I have a room of 350 people, and I haven’t had a problem (not a significant one) yet. But here was this guy, again, doing to me everything I created the room to avoid having to be exposed to. Don’t worry… I’m not looking for any kind of confirmation on the decision. I’m good with it. I just hope he doesn’t come here.
I love DIY people. I think I might be ready to hear some of what you guys are DingYs. I didn’t know we COULD do things ourselves. In fact, I thought doing home a1cs was cutting edge and really sticking it to Big Pharma.
You’re absolutely right! The 670g is advertised as “set it and forget it”. I have no idea of it’s vagaries, but it does not seem to live up to its expectations.
You seem to know enough about the pump to see the first problem… it wants to correct you back to a 120, even if you’re blood sugar is great. Now it does seem to use trends via the sensor signal as another piece in determining whether or not you need insulin, but it’s almost guaranteed an 81, even a 91, would probably result in reduced-sized micro boluses. When you look at it as a snapshot, it looks fine, but when you pan out and see the overall picture, it gets pretty crazy. I could spend 4 hours in auto with a blood sugar between an 85 and a 120, and auto would handle that time by giving me less insulin than what I would get in a regular basal rate. From one 30 minute window to the next, it would be fine, but if you add up 4 hours of it, it’s like receiving, say, a 60% temp basal.
I’ll come down from all of this. I just got myself all wound up this weekend, and, of course, I’m still playing pieces of it out in my head. I’m forever the worrier that I’ve done something wrong. It was my belief that each individual setting, basal, ICR, AIT, as well as all of our other numbers, produce the best results when they’re fine-tuned. Auto, with it’s ever-changing and constantly withholding delivery rate, stuns me. I am crawling under a rock for the rest of the week with my “Pumping Insulin” and am going to try to get a handle. I’m sure to have some questions when I come back out.
. [quote=“Eric, post:5, topic:4064”]
I don’t think the problem is you. I have read enough of your posts to see a sharp, engaged, and proactive mind.
[/quote]
You guys are wonderful. It might be because you’ve only read enough of my posts to see the positive but NOT enough to see it all, but I’ll take it.
Interesting… that was part of our debate, and what he, and currently everyone, is saying is that ICR settings affect the SIZE of your micro boluses. Supposedly it is not only a matter of increasing your food bolus but is a matter of unlocking your pump’s ability to deliver bigger micros. (My micros, by the way, are disappointingly small—- it’s like throwing pebbles at an asteroid when a rare .175 is all you’re packing.)
Yes, I do know this. I refer to it as a “basal rate” because it’s what it’s replacing. And because what I’d like to call it has too many bad words and would take to long to write out each time I reference it.
This was actually the experiment I was talking about when I was told by said twerp that my experiments were “worthless”. I have many logs of data that show the only reason I have those stubborn, ridiculous highs in the morning are because auto mode delivered less insulin over the course of the night, but people have a hard time grasping it. So I was doing an experiment, alternating 4 days in auto and 4 days in manual, to show how my only “dawn phenomenon” occurred in auto. It’s not there in manual because it’s not true dawn phenomenon. I don’t know if you had figured that out before you left completely, but it drives me crazy that Medtronic is telling people it’s their “pronounced dawn phenomenon” that’s to blame.
I hate everything about auto mode, and I don’t know what it says about me that I haven’t arrived there yet. I’ve hated auto mode for 9 months. I really need to be done… I’m going to go think about that for a minute…
How are some people so clear on when a person is not worth talking to again, and some are so not? This was the last of about 16 conversations with this guy (probably way more) that really took their toll on me. This one ended with a throbbing headache and an emotional binge on tortilla chips and tasteless gluten free/dairy free cookies (I’m not saying they’re ALL tasteless… these were). I hit a 580 that night. I haven’t seen a number like that in a year. I woke up the next morning and, surrounded by family who promised it was okay, blocked him. Life is so good sometimes, it is, and sometimes it’s just downright silly. :).
It wasn’t just his “perfection”. He had a very mean style. I don’t miss him.
You have nailed it exactly - the problem with this device, and other devices that are soon to come out and do the same thing and try to pretend they are a pancreas.
The whole expectation of endos and pump companies of this thing being an automatic pancreas is idiotic, because all it has is insulin. And guess what?! A real (working) pancreas also has alpha cells, which it uses for lows. Seems so many people forget about the other half of our pancreas’s which are broken - the alpha cells!
A real pancreas doesn’t just cut your basal until you come up from a low! It talks to your liver and says, “Glycogen, please!”
I learned fairly soon after starting with a pump - screwing too much with basal is not the way to go for lows. I can get away with a small cut here or there. And I do turn it off for exercise. But often times, doing a 4 hour adjustment might cause me problems several hours down the road.
Don’t crawl under a rock. Hang out here and chat, post, ask, read. You might get just as much - if not more - from this place as you can from a book, since this place is interactive. Not only can you read here, you can ask, which you can’t do with a book.
In my case the DP is real enough and has gotten worse with age. In Manual I have a pretty impressive basal rate programmed to kick in around 3am that reins it in. The problem with auto wasn’t just that it wasn’t aggressive enough, but that the Guardian sensors, which on the whole I found pretty accurate, consistently registered low overnight, so they just weren’t signaling that the pump needed to start ramping up the doses to get ahead of DP. So I went from mostly getting fasting #s in the 100-140 range, to more like the 160-180 range, and the minging corrections weren’t getting it down to what I consider an acceptable range before midday meal, and on it went.
Interesting that you’ve heard others say adjusting the ICR does affect microboluses–that was what I surmised, but when I asked my Mdt trainer about it, she checked with her overlords and they said Nope, wizard boluses only. Which goes to the fact that the close-mouthed attitude among their support staff about how the algorithm actually works is a big part of the problem.
I’m so stuck on this whole dawn phenomenon thing, it’s taking its toll on my ability to hold a conversation about anything else. Losing friends left and right. (I actually only had one to begin with, but she’s had it).
So I have something else to say about it… then i won’t speak again about it ever in a conversation with you. This is what I think… 1. I’m not arguing that people have dawn phenomenon. Of course we do. Some have it worse than others, and like everything else diabetes, it may change whenever it dang well pleases. 2. Most of us (I don’t know if that’s accurate) have it on lockdown with our manual settings. But the exact thing you just described is the exact thing I experienced and is the exact thing a whole LOT of people are describing, and after doing more testing than I’d like to admit, I have a different explanation.
There’s a far bigger problem than that, at least with what I see… BECAUSE those numbers are registering lower over night, and BECAUSE of its target, it’s often delivering a VERY reduced rate of insulin throughout the night. So it’s not just that it’s not registering the need for insulin quickly enough in order to head it off, it’s actually CAUSING it. I could show you nights on my graph screen that have a solid line of pink dots all the way across, and my SG looks like I’ve flatlined at a 120. It’s lovely. But when you really start looking at those numbers, you realize how many of them were micros of .025, and that line REALLY is a 140 and rising. So that climb I saw (I really have to stop saying “you”) was a result of getting the equivalent of a temp basal. And when i added up all of my micros, because i’m crazy enough to have done that time and time again, i would see i had received a total of 38% of the amount of insulin I would’ve received in manual over the course of say 4 hours. That part is a fact. Now what percentage of that was good because it helped me avoid the low? It’s hard to determine. But again, time and time again, I would see these very profound highs and they followed nights where I would add up a reduced rate. On the flip side, on the mornings with a different distribution, where I received a lot of insulin in the couple of hours leading up to my alarm, my BG would hold steady, even dip. One night, in particular, I received just 19% of my regular basal rate over the course of 4 hours. 4 hours! I need more insulin than that. I also learned that by getting OUT of auto just 2 hours before I actually woke up, my waking number would be lower, and I could avoid that rise. It just meant a full 2 hours of insulin heading into waking…
I know you’re done with auto. For the most part I am, too. I’m just all fired up right now for this exact reason. I alternate my days between auto and manual to show that every morning in manual is flat or drops, then the next morning in auto, I climb, and again, and again… I’m upset because Medtronic is scrambling to make people like this thing and, honestly, probably scrambling to get it to work better, but I keep feeling like part of their strategy is just to hush people’s concerns. And to use wording that might lead to people analyze their results differently. So the word is that some people’s “dawn phenomenon” is too profound for auto mode to handle. In the many months i was in auto before figuring out what was happening, MY dawn phenomenon was too profound for my auto to handle. Then I figured out what was happening with the insulin delivery over night (only most nights, of course), and I corrected it. For me, auto mode was causing a profound morning BG rise that then auto mode could not handle. This is a different problem, and I think Medtronic needs to help people accurately identify their problem before throwing more insulin at them.
Of course, that brings me back again to the ICR setting and whether or not it affects the size of those micro boluses. If my micros were bigger, could they better fight the rise auto mode has caused?? I can’t say they don’t because so far I haven’t been willing to do it. That’s what I got blasted for. Not being willing to crank that setting. But my point is that if we’re really cranking that setting to get auto mode to give us more insulin but along with that setting we get more insulin at every food bolus, then how am I not setting myself up for unpredictable lows?? I never know hour by hour what auto mode has delivered unless I add up my micro boluses so how would I ever know whether or not I really needed that extra insulin for the same food i always eat?? Does that make any sense?? Many people have HUGE crashes in auto, many of whom say they’ve never crashed like this before, and I think this is why it’s happening.
I could complain less, I suppose, and just add micro boluses all day, but I feel like that would’ve been good information to include in their brochure.
That was it. I hope you didn’t even make it this far down the page to be reading this. It’s my next video, and I’m addressing it to Medtronic, and I’m getting myself pumped. People are gonna throw tomatoes.