Resource 2
Insulin, Part I
Working Out The Action
Background
I read a book years ago that went into some detail on how insulin works in the body after an injection. It was a new idea to me. There are several different models people use (eg assuming 1/3 each hour for 3 hours).
Then I thought, why not come up with my own model, just for me?
It’s, of course, not everyone’s cup of tea to do so, but if it is…
The following is the method I used.
(See main Resources page for a link to the file itself).
4. Mid-way Time of Action
2. BGL-Time Graph
- BGLs
3. Boxes to help guess the action
5. Rule of Thumb
Step-by-Step
Let’s go through a real-life example step-by-step.
1. BGLs
START: I woke up with high blood sugar in the morning at 7:00am.
INS: I injected 8 units
BGLs: My first BGL was 20.4. I recorded every 30 mins or 1 hour to get a reasonably good graph (CGMs give many more measurements). After 5 hours I stopped measuring as BGL wasn’t falling any more.
2. BGL-Time Graph
Y-AXIS: In the vertical bits, I filled in the numbers 5, 10, 15, 20, 25. It doesn’t need to start at zero (since I’ll never get to that number!)
FILLING IN: I started off filling in 20.4 (my first BGL) at time zero, then continued…
30 mins = 20.3 > 1 hr = 19.1 > just before 2 hrs = 15.7 > 3 hrs = 14.8 > 4 hrs = 11.5 > etc.
LINE OF BEST GUESS/FIT: This is more art than science. I could’ve drawn a more complicated line with a steeper bit (from 1 > 2 hrs) and shallower bit (2 > 3 hrs) then steeper again (3 > 4 hrs).
Here, I decided to keep it simple, with a flat start, then fall, then flat again. It’s best to do multiple tests to get a good feeling for a ‘model’ line.
It’s not vital to test BGL on the exact 30 mins or 1 hour – it just makes it a bit easier for calculations later.
3. Boxes to Help Guess Action
This is the hardest part, so let’s go slowly here.
GRAPH: Using the orange guess-line, I mark where the BGL is at 1 hour (just over 19).
At 2 hours, the BGL is about 16.5
At 3 hours, the BGL is about 14.0
At 4 hours, BGL is at end (11.3).
Achtung! Here comes the maths bit!
TOTAL DROP: about 9 (20.4 at time zero falling to 11.3 at 4 hours).
HOURLY DROP: I already know the 1st 30 mins showed no change, so I write:”1st 30 mins = 0″
At 1 hour, hourly drop in orange line is about 1.5 (20.4 > 19). Of the total drop of 9, 1.5 = about 17%.
At 2 hours, hourly drop is about a further 2.5 (19 > 16.5). Of the total drop of 9, 2.5 = about 28%.
At 3 hours, hourly drop is about a further 2.5 (16.5 > 14.0). (Again, 2.5 = about 28%.)
At 4 hours, hourly drop is about a further 2.5 (14.0 > 11.3). (Again, 2.5 = about 28%.)
I’ve almost gotten my own model worked out.
Whew! That’s a lot of maths (but very interesting for real life).
4. Mid-Way Time of Action
This is my favourite ‘easy’ bit of maths, as it’s a handy number to know for when the insulin is half-done. Is it exactly half-way through the total time of action?
I know that the total drop was 9 in BGL. Half of that is 4.5. So, 20.4 – 4.5 = about 16.0.
Looking at what time 16.0 happens, it’s just over 2 hours (say about 2 and a quarter hours). Which confirms that the insulin is about half done at 4 hours. Good to know.
5. Rule of Thumb
On a subsequent test, I decided there was a steeper action from 30 mins to 2 hours, and came up with an action of
1/8; 3/8; 2/8; 2/8
Again, very approximate.
In Part 2, I’ll take you through the actual usefulness of doing all this calculating.