New Tricks for Tweaking Dosing Formulas for Your Insulin Pump

Kelly Close is the brillant mind behind Close Concerns, a consultancy dedicated to “the business of diabetes.” In addition to having type 1 diabetes for more than 20 years, she is also one of the thought leaders in the diabetes community. One of Close Concern’s main objectives is to go every ― and I mean every ― diabetes conference and report back on the latest findings. Today, Kelly shares some new nuggets she discovered at this year’s AADE conference.
A Guest Post by Kelly Close, diabetes industry consultant
During our time at the AADE (American Association of Diabetes Educators) conference this year in Atlanta, we were impressed (and unsurprised ― it’s like this every year) to see educators traversing the convention center hallways with equal enthusiasm and energy on the last day of the meeting as on the first day of the conference. It always blows us away that some of the final sessions are as packed as the ones on the earlier days. Needless to say, as a field, we are lucky to have such dedicated individuals working to advance and improve patient care. We attended a number of thought-provoking and moving talks throughout the meeting, but especially a session on revising pump settings for type 1 patients! I thought my pump settings were relatively right (I always put off those “skip a meal” exercises that my ― and Amy’s! ― fabulous endo, Dr. Nancy Bohannon.
In fact, this new set of recommendations for pump users was, if technical, really, really useful for me. Here is the summary below — we discovered that this will be published in an article in a fancy scientific journal next year, but if you want to get a head start, talk to your educator or doctor about refining your settings. I changed mine actually quite significantly upon advice of an amazing educator helping me with this:
- We moved my “total daily dose” of insulin from near 40 units (obviously it varies a lot every day) to closer to 30 units (I fought and fought this “it’s not nearly enough!” and I’m doing much better on 30 units)
- We moved the “basal” part of “total daily insulin dose” from about half the total to closer to 40% of the total
- We moved my “insulin correction factor” from 50 to 40
- We moved my “insulin sensitivity factor” from 15 to 11
These were very big changes to make and obviously stuff like this shouldn’t be done without conferring with your healthcare team. For me, as noted, I was convinced that I would be sky-high by reducing my total daily dose; actually, I have found that I wake up at normal levels rather than low, and that I have less hypoglycemia overall, so fewer “high” rebounds, and less “correction” insulin. While I take more insulin for food, it’s less insulin overall — also since I’m trying to watch those carbs!
This is just an example of putting into practice some incredible learnings. See below for our summary of Dr. King’s talk! I would also be remiss if I didn’t say that my CGM has helped me enormously. It helps me day-in and day-out and I have gone from using it occasionally a couple of years back to 24/7 use now — my insurance, luckily, pays for most of it, and I know that these days, I’d find life very difficult without it.
Insulin Dosing Formulas from Pump-Treated Type 1 Patients
Dr. Allen King of the Diabetes Care Center in Salinas, California, gave a fascinating talk discussing better ways to dose insulin with pumps. For a long time, there have been clinical ‘rule of thumb’ formulas that help in setting up insulin dosing for pump patients – these are for TTD (total daily dose), TBD (total basal dose), ICR (insulin to carb ratio), and CF (correction factor). Dr. King has published four studies in which he’s taken care to set patients up perfectly on the pump and then derive updated formulas from his participants.
The commonly used formulas for initiating pump therapy have historically been: 1) TTD (total daily dose) to weight in pounds = 0.27; 2) TBD (total basal dose) = 0.5 * TDD; 3) ICR (insulin to carb ratio) = 450/TDD; and 4) CF (correction factor) = 1700/TDD. These coefficients (0.27, 0.5, 450 and 1,700) were studied by Dr. King using CGM. The revised ratios he recommends, following the studies where he looks at ratios of the patients who have the least variability, are TTD to weight = 0.2, TBD = 0.4*TDD, ICR = 300/TDD, and CF = 1500/TDD. The new coefficients are certainly different than traditional formulas, leading to more bolus and less basal and different ratios for the different factors. Typically, Dr. King said, patients have too little bolus because of safety, and too much basal because bolus is easier to adjust and there is increasing insulin resistance with excess basal. Excess insulin creates insulin resistance. At the end of the day, of course, patients still need to be vigilant — specifically, he recommended actually blousing 30 minutes before eating, testing basal doses by skipping meals, and being getting education on counting carbs — but it seems the new formulas can help greatly in developing better control. As always, changes should be done with your healthcare team.
- The total daily dose (TDD) is the dose of insulin that patients generally need depending on weight; of course this varies each day depending on carbs, exercise, stress, and other factors.
- The total basal dose (TBD) is the total daily basal dose that achieves fasting glucose targets, corrects nocturnal hyperglycemia, and should cause no hypoglycemia if a meal is missed or delayed.
- Insulin to carb ratio (ICR) – a number that represents the number of grams of carbohydrates consumed, which when treated by one unit of insulin brings blood glucose back to baseline within two to four hours.
- The correction factor (CF) – a number that represents the drop in blood glucose per unit of insulin administered.
- The commonly used formulas for initiating pump therapy are: TBD = 0.5 * TDD (TDD = total daily dose), ICR = 450/TDD, and CF = 1700/TDD. These coefficients (0.5, 450 and 1,700) were studied by Dr. King using CGM.
- Four pump studies including a total 61 patients were carried out with CGM, (Medtronic CGMS Gold). Diets were structured to be isocaloric; there was a staggered daily meal omission and a detailed diary. Mean basal glucose was 115 +/- 14 mg/dl, with only 0.46% of the time <70 mg/dl. Careful analysis of the studies resulted in new coefficients for the classic rule of thumb formulas. For some patients, these will be major shifts.
- Rounding the results to one significant figure, the group revised the ‘rule of thumb’ formulas as follows:
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
- Effectively, these rules lead to more bolus (more toward 60% of total) and less basal (more toward 40%). Dr. King said that typically patients have too little bolus because of safety, and too much basal because bolus is easier to adjust and there is increasing insulin resistance with excess basal. Excess insulin creates insulin resistance.
- The revised coefficients have their place, but it is also recommended to reinforce carb counting, better compliance with pre-meal dosing, patient education on underbolusing, and to increase the frequency of evaluation of basal dosing (through skipping meals).
We know it is fantastic if you have a great educator who can work with you on the right stats to use. One sobering statistic we learned this year at AADE – less than 10% of CDEs in a major survey just released are under 35. Not under 30! Not 25! But 35! Not that 35 is exactly middle-aged — but it’s not far! This doesn’t bode well for those who will need diabetes educators in 20 years, that is for sure. We hope each educator walked away from this meeting understanding how powerful and influential their actions and voices can be — and we hope any company not engaging them routinely as advisors will start to do so — they won’t be sorry.
Last! I have been really helped by my diabetes healthcare team (thank you Dr. Bohannon! Thank you Gloria!) and CGM as I have sought to modify my pump settings. If you are interested in trying CGM, enter the diaTribe lottery for $500 to spend at the DexCom store any way you’d like — a new system, an upgrade, or new sensors! Go to www.diaTribe.us/diabetesmine to sign up! And best of luck with your pump settings!
Wow. Thank you for all that valuable info, Kelly.
Explore posts in the same categories: Diabetes Essentials, OmniPod-ing, Products
















Wow…These are SO far off from the settings I use. As a comparison:
TDD=19.2 vs. 23 (my current setting)
TBD=9.6 vs. 16(!!!)
ICR=23 vs. between 18 and 21
CF=78 vs. 128(!!!)
I’m too scared to make any changes, though. This is just too radical of a change for me, especially since I’m really insulin sensitive and I’m doing well with my current settings (last A1C 5.6)
Posted by: Elizabeth Joy | August 20th, 2009 at 12:58 pmInteresting… I had an endo who kept thinking I should have higher basal rates but I resisted!
My mean TBD/TDD over the past 10 days is 0.42.
My mean TDD is actually 26 vs recommended 33, and mean TBD is actually 10.9 vs. recommended 13.2.
However, these suggestions would also tell me to change my ICR from 17 to 9 which would drop me to the floor. The CR of 45 recommended by this method is accurate only if I am not exercising at all (like when I was immobile after getting injured). Currently I use 70.
I wonder what other physiological parameters these factors could relate to, like metabolic rate etc.
-Anne
Posted by: Anne Findlay | August 20th, 2009 at 4:28 pmIt’s interesting to see that my current settings are almost completely aligned with the new recommendation.
For a while I’ve been feeling that my settings are not right, as they don’t conform with the “standard”. Now I feel much better.
Regards,
Henry
Posted by: Henry | August 20th, 2009 at 5:04 pmAuthor of D and The Guy
http://www.dandtheguy.com
Wow. Reading this makes me never, ever want to try a pump. I don’t need another headache in my life. Injections are simple, straightforward, and portable.
Posted by: Lauren K | August 21st, 2009 at 12:35 amI’m an insulin dependent T2, producing very little insulin. C-peptide of 0.5. Here is how I adjust my pump:
Adjust the basal so that the numbers are in line with my goals, and do not drift if I skip a meal. Insulin/carb and correction factors found by trial and error.
30 day stats:
Ave 94
Standard Deviation 12
26% bolus
2% correction
71% basal
The above numbers are low on bolus, due to my eating a very moderate amount of carbs.
I don’t see how anything derived from a formula can be anything but your first day’s settings, beyond that its all tweaking, basal first.
Having your glucose remain as constant as possible when a meal is missed or delayed is a very useful thing. Combine that with good carb counting skills, and a food scale, and you can be very successful.
-Lloyd
Posted by: Lloyd | August 21st, 2009 at 12:38 amThese ratios work out to suggest that, on a perfect day without correction insulin, everyone — regardless of weight — eats 180 grams of carbs.
We start with:
———–
TDD = 0.2 * weight
Basal = TDD * 0.4
————
Since TDD – basal = non-basal insulin (no corrections in an ideal world), we get
non-basal insulin
= TDD – basal
= (0.2 * weight) – (TDD * 0.4)
= (0.2 * weight) – (0.2 * weight * 0.4)
= (0.2 * weight) – (0.08 * weight)
= weight * (0.2 – 0.08)
= weight * 0.12
So, let’s get the carb factor:
300 / TDD
= 300 / (weight * 0.2)
= 1500 / weight
So, if all of our non-basal insulin is going to carbs, it means that:
carb intake
= carb ratio * carb insulin
= carb ratio * non-basal insulin (ideally speaking)
= (1500 / weight) * (weight * 0.12)
= 1500 * 0.12
= 180
Therefore, everyone eats 180 grams of carbs no matter what they weigh.
Heh. Math.
Posted by: Anon | August 21st, 2009 at 6:43 amAs the caretaker of a child with Type 1 (diagnosed at 8 and now 12 years of age), I find all formulas to be quite dangerous. Formulas can only be used as a general, ball park check. For first two years prior to puberty, my niece used 36 percent basal, the rest bolus. Now, in puberty, she is using 60 percent basal, 40 percent bolus. She is 85 pounds. Her Insulin Sensitivity Factor is 75 days, 110 midnight to 6am. This has been checked time and time again and verified by her Endo. I have John Walsh’s charts as a very loose reference (mostly to check basal per hour) and they are way off. Her lowest basal is .95. Her highest basal (which can go higher due to growth spurts), anywhere from 1.75 to 2.7 during the 6pm thru midnight hours. We get night basals almost every night and she wears the Dexcom 7 Plus cgms. Current Insulin to carb ratios are 1 to 7 breakfast, 1 to 13 lunch and dinner. All have been checked time and time again and are verified. We do fasting basals and skip meals to check basals all the time. The one thing you can check with absolute certainty are basal patterns. Unfortunately in teens these change very frequently. Very. Each person should do their own basal testing to determine basal need. Then they can figure out the rest. Forumulas such as these do not and cannot work for each person. Each person is an individual and needs may vary tremendously.
Posted by: Jan | August 21st, 2009 at 10:09 amyes, it is DEFINITELY true everyone’s mileage varies on this! I found the changes really helpful though – even though my A1c has been under 7 for a long time, it wasn’t without a lot of hypoglycemia, so I think I’m nearly a much “higher quality” A1c, which isn’t really a term we hear a lot about. But wow to be at 5.6 Elizabeth – that is amazing! I wouldn’t change a thing either if I had few lows! And Lloyd that’s exactly what the educator said about my numbers – we had to change the basal first, which I hadn’t really realized, so I had long been “tweaking” the other formulas without getting the right basals. Thankfully I’m there now. That’s fascinating, Jen, about the 180 carbs – I think mine varies a ton, but usually around 160, so not too far off. Jan, if I implied anything BUT these are starting points, I’m so sorry. I think their point is the new formulas may be better starting points. I’ve also heard from a lot of diabetic friends that their current settings are the SAME as when they began pumping – probably many people could use tweaks, but the HCP teams don’t always spend time on it. As for a pump vs injections – I don’t think I could ever go back! I know all the math sounds insane, but you really only do it once and then tweak (but I hadn’t tweaked enough, that’s for sure!) And last, even though the reason my insurance pays for pumps and CGM is because I am in much better health with them, one of the nice bonuses about pumps and CGM is also that it is fewer pokes! Thanks Amy for the chance to guest post!
Posted by: kelly close | August 22nd, 2009 at 10:52 pmYes, Jan, these are definitely BALLPARK FIGURES FOR ADULTS. THAT’S WHAT THEY ARE SUPPOSED TO BE!. AND Anon no, Anon, NOT everyone is supposed to eat 180 carbs everyday (no corrections in an ideal world !!!! ???? !!!! ARE YOU CRAZY ?????????? ).
Posted by: T1 D | August 23rd, 2009 at 3:54 pmI eat 24 grm of carbs every day, not every meal BUT EVERY DAY and the correction NEW numbers are about right FOR ME.
I am the Mom of a 12 year old Type 1 girl and puberty is very difficult and not consistent at all. SOME if the CDE’s get caught into the formulas too much, although each family is different in their level of involvement and care.
Posted by: Cathy | August 23rd, 2009 at 7:28 pmi tried this, and I have been amazed at the drop in insulin I’ve had to take every day. the correction levels were totally adjusted for me, and I haven’t been above 160 since doing this! Thanks Kelly!
Posted by: landileigh | August 27th, 2009 at 11:46 am