#74 January 2020 Update on My T1DM Management

This is a monthly update on my glycemic management of type 1 diabetes (T1DM) using Humalog and now Tresiba insulin injections with a ketogenic whole-food diet and resistance and aerobic exercise.

If you feel you might benefit from some individual attention and suggestions for achieving success with blood sugar control for type 1 or type 2 diabetes and/or losing excess body fat, I can assist you with a personal consultation via Skype. See the Coaching page for more info.

In October I did an interview with Ross Wollen on asweetlife.org titled “Interview: Dr. Runyan is the Diabetic Ketogenic Athlete.” You can read it here.

I wrote an article for DietDoctor in July titled “What you need to know about a low-carb diet and your kidneys.“ You can read it here

My books are available for purchase: The Ketogenic Diet for Type 1 Diabetes also available on Amazon in print, and my other book, Conquer Type 2 Diabetes with a Ketogenic Diet, is also available on Amazon in print

Glycemic Results For January 2020

My January 2020 glycemic results were a new record: the lowest monthly blood glucose variability, as measured by standard deviation, since my diagnosis of type 1 diabetes in 1998. I started taking Tresiba on November 22, 2019 in hopes of further reducing my blood glucose variability and I think it has contributed to my improvement.

The graph below shows all the blood glucose measurements and daily insulin dose totals for January 2020.

My blood sugar goals are shown in Tables 2.3 and 2.4 below. I used as my measure of normality, a standard deviation of blood glucose (SDBG) of less than 18 mg/dl (1.0 mmol/l). This value was measured in 434 metabolically healthy subjects in China as reported in this study.

The table below shows my monthly mean body weight, mean total daily insulin dose, mean basal insulin dose, mean bolus insulin dose, and the mean bolus insulin dose for each meal and the bedtime bolus insulin dose which I started using a few months ago to try to improve my fasting blood sugar while keeping the basal insulin dose constant. Diluting my Humalog 5:1 helps to give precise doses with meals and at bedtime.

The graphs below are new and show the individual mealtime and bedtime blood glucose, PRE and POST, values and the corresponding bolus Humalog doses for the month. I think what is remarkable about these doses is how much they vary from day to day despite keeping each meal and exercise session constant. This is currently consuming my thoughts as I search for a method to smooth-out both the blood glucose and insulin doses from day to day. I have in the past and continue to notice a lag effect of blood glucose to changes in insulin doses. When I have to increase insulin doses for higher blood glucose readings, invariably about two days later, I get one or more low blood glucose readings.

The table below shows my blood glucose variability data including the monthly mean blood glucose, the standard deviation of blood glucose (SDBG), the coefficient of variation (COV), the calculated HbA1c, and the percentage of blood glucose values < 74 mg/dl, between 74 and 126 mg/dl, and > 126 mg/dl. As mentioned, these are the lowest monthly SDBG and COV values since my diagnosis of T1DM and technically equivalent to a non-diabetics blood glucose values. I did not take any glucose tablets (or Smarties) this month either.

As you can see for January 2020, I just missed my goal of < 5% of blood glucose values < 74 mg/dl. My total daily insulin dose was higher this month for reasons that are not exactly clear other than I needed more for mildly, but still in range, blood sugars. I am optimist that I can continue to improve my results. I continue daily exercise and a very low-carbohydrate diet. My exercise program is easy 2.5 mile walks with a 20 kg weight-vest twice daily and olympic weightlifting 2 hours daily, for a total of about 4 hours/day. As mentioned in the past, about 1 hour of the weightlifting time is resting between lifts.

Table 2.2 below shows the mean and 95th percentile of the interstitial glucose (IG) and standard deviation of the interstitial glucose (SDIG) of 564 nondiabetic subjects as measured by CGM from the five studies referenced below.

The references for these five studies are shown below.

This month, I plan on working on a mathematical method to predict insulin doses in the hopes of smoothing-out the blood glucose readings and insulin doses. I expect this approach will only be effective after one has established a consistent schedule of meals, exercise, and sleep, i.e. factors known to affect insulin sensitivity.

Till next time….


  1. Brian Lucido

    Your standard deviation is outstanding this month! Like you, I’ve noticed this same thing where after ~2 days of increasing insulin, I sometimes need to dial it back. It’s almost like there is some form of oscillation that is inherent in our insulin needs. I’ve also noticed macro cycles in insulin needs – but unlike you, I’m not controlling for exercise or diet – which can be highly variable seasonally. It seems as if your exercise is very regular, which means you can learn a lot more from your experiments. Congratulations on having better than (in my opinion) non-diabetic blood sugars!


    • Keith Runyan, MD

      Thanks, Brian. I’m think that when blood sugars start to increase, there needs to be some limit on insulin dosage increases and allowing blood sugars to be slightly higher than I would like to possibly prevent the subsequent low blood sugars. That’s what I’m going to work on this month and will see if that helps.


      • Ellenor Malik

        I know that what PID controllers for cruise controls do is progressively increase the throttle if they’re going too slowly (initially giving a light press, then progressively increasing the manifold pressure until ultimate disconnection), and progressively decrease the throttle if they’re going too quickly. If it’s too responsive, it can hunt. I think what’s happening is that you may be responding too fast or too authoritatively and you’ve got hunting going on.

        You say you took no gluco tabs in January – did you take any glucagon or were you truly free of medical-emergency hypoglycemic events? I’m assuming the latter.


      • Keith Runyan, MD

        Thanks for the comment, Ellenor. I like your cruise control analogy. I had the same thoughts about trying to correct high blood sugars completely to my target of 100 mg/dl. I have already come up with a possible mathematical solution which I will need the month of February to test. Secondly, this month I was truly free of symptoms of hypoglycemia or the need to correct low blood sugars with anything other than reducing mealtime bolus doses. Symptoms of hypoglycemia typically appear as the blood glucose falls to 50−55 mg/dl (2.8−3.1 mM). I have never taken glucagon and stopped renewing it due to the cost years ago.


  2. Rick Phillips

    I have my A1C down to about 5.7 unfortunately i must do it with far more insulin, average 65 U per day ranging form 55 to 73 U in January. I suspect i might be a bit larger (220 Lbs). I admire your wonderful numbers.

    I will say that my doctor has suggested i let my Blood sugar float up and place it in rage of 130 – 170 instead of 80-130 so I can lose this stubborn 20 pounds. Rick going a little DKA will help, then we can dial back in he says. Yes I know – but damn I want it all.


  3. Ryan R Connor

    Dr. Runyan – just subscribed to your blog. I have a 12 year old with Type 1. Endo Dr. says he needs carbs for growth and brain development. Where do we start learning more to help him and his daily battle with his highs and lows.
    Ryan Connor
    Warsaw, IN