'Hypos' but not as we know them?

[Essex]

This is better

"Continuous Glucose Profiles in Healthy Subjects under Everyday Life Conditions and after Different Meals"

Fig 2: Mean continuous IG trace obtained in healthy volunteers under everyday life conditions by means of the microdialysis technique. Mean interstitial glucose trace ± 2 SD (corresponding to the 95% range), n = 21; Cap. BG: capillary blood glucose values, used for calibration of the continuous glucose monitoring device.

Divide by 18 to get mmol/mol from mg/dl
(160 = 8.9, 140 = 7.8, 120 = 6.7, 100 = 5.6, 80 = 4.5, 60 = 3.4)

All study participants exhibited normal values for HbA1c (5.0 ± 0.2%), fasting concentrations of insulin (42.4 ± 16.0 pmol/liter), C-peptide (0.59 ± 0.17 nmol/liter), proinsulin (3.5 ± 2.3 pmol/liter), lipids (total cholesterol 4.6 ± 0.7 mmol/liter, high-density lipoprotein cholesterol 1.3 ± 0.3 mmol/liter, low-density lipoprotein cholesterol 2.9 ± 0.7 mmol/liter, triglycerides 0.8 ± 0.3 mmol/liter), creatinine (86.4 ± 12.6 µmol/liter), uric acid (272.2 ± 58.6 µmol/liter), liver enzymes, and hematology parameters.

 

[Deleted member 21371]

Don't over think it.
Everyone is different.
Just eat to whatever your meter tests as.
That's all you can do.
You're engineering with a sample size of one here.
The stats are completely knocked out of the park.
I went down the route of researching every diet, every cure, every thing I could find, and simply picked out the common best bits, and took it from there.
I also considered other things apart from BG, such as cholesterol, blood pressure, weight.......

 

[Essex]

Sure, I do that

but I just wanted a baseline to get a feel of how my values compared ball park figure wise
The means +-2SD give me a feeling of the spread (although only n=24)

And, I just like to know these things 🙂

I had no idea where a jump to 12 after breakfast fitted in to the scheme of things
But that was just one day, on the 20th I had a much smaller jump after weetabix

I thought today that perhaps I need to do more exercise, perhaps with even more emphasis than on diet

 

[Docb]

@Essex. You might be interested in the graph above which is a summary of my data for the last 18 months. I rounded times to the nearest hour, calculated means and standard deviations for each hour and then plotted the mean (blue line) and an upper 95% ( mean +2SD), the red line. Don't have the n's for each hour but there are around 3000 results in total. OK, I've taken a bit of a liberty with the stats because the distributions are in reality highly skewed but it does give a feel for what is going on.

So for me, I average around 8 post breakfast but I expect the odd result in double figures and treat them as part of life's rich pattern and do not react to them!

 

[Essex]

@Docb
Thanks, that is very interesting! But I am a complete geek
I understand that the data doesnt include variation between subjects (plots), nor effects of other variables like what you ate, exercise etc, but the 95% gives an idea of the range of this
But, the important thing is,

So for me, I average around 8 post breakfast but I expect the odd result in double figures and treat them as part of life's rich pattern and do not react to them!

So, today's experiment is to eat the same as yesterday for breakfast and see what happens.
I am so sad but I love this type of thing
Always look on the bright side of life - I may be on the wrong side of the tracks wrt diabetes but I get to make more spreadsheets and experiments.
As my friends at home say, what a tw*t - let's assume * = 'i'

 

[Essex]

Well, exactly same breakfast and routine - completely different results 🙂
I even feel a bit foggy and very hungry now so wont be able to stick to yesterday's food as I am going to have to eat again now. Ho-hum

Date​	02/12/21​	​	​	​
​	​	​	​	​
​	Time​	mg/dL​	mmol/mol​	Food / Activity​
​	​	​	​	​
Waking​	07:18​	95​	5.3​	​
​	​	​	​	​
Breakfast​	07:30​	​	​	2 x Plain 125 g lactose free yoghurt, Tea​
​	​	​	​	​
Post Prandial #1​	08:43​	80​	4.4​	School & dog walk​
Post Prandial #2​	08:50​	92​	5.1​	At computer, feeling very hungry & a bit foggy​

 

[rebrascora]

Did you wash your hands and retest yesterday when you got that 12 reading? It is always a good idea to check a result that looks rogue.... although I appreciate that at this early stage, identifying a rogue reading is not always easy.
Having a tiny smear of residue on your finger from maybe passing the jam jar to someone else at the breakfast table or chopping onions for dinner or from eating a piece of fruit earlier can all impact your reading as well as the possibility of a defective test strip, or it could be that your body is starting to settle down now it has had 2 low carb breakfasts on the trot. This is the main reason why we recommend not taking any decisions off the back of one result. Looking for trends with diabetes is the way to make progress.

 

[Docb]

To continue the geekyness (I think I made that up).....

The problem is that there are a whole string of variables which can affect blood glucose and the amount of carbohydrate you eat is one of them. Just because it is the one thing you can measure and control does not mean that it is the primary control variable. Between shoving it in your mouth and seeing the effect on blood glucose, the carbohydrate has to be converted to glucose (mmmm.... how does that happen and what affects that process); get transferred to the blood stream from the gut (mmmm.... how does that happen and what affects that process); get flushed round the system and be subjected to the vagaries of the pancreas in trying to control it (mmmm.... how does that happen and what affects the process) and finally, what affects the reading you get on your meter (mmmm... sampling errors, measurement errors, contamination errors).

My thought for anybody in this area is always to look for patterns which seem to have consistency over time. You can get 80% of the information by screwing your eyes up and taking a broad view of the numbers. You can geek as much as you like but the other 20% is always going to be a complete mystery unless you can measure some of the variables I have alluded to above.

 

[Deleted member 21371]

I worked with stats for a long time.
I also tested when I was first diagnosed.

BG has too many variables for any meaningful conclusion, apart from "if you constantly go high don't eat it again"
As you have found out, even Weetabix results aren't repeatable.

What you ate before, how fast you ate it this time, did you chew it more, did you soak it in the bowl longer, how fast it cleared your digestive tract, and that's before any attempt your body has with insulin. (Which changes from previous meal to meal).
If you ran upstairs before testing, if your finger is hot, or cold, or you stay in front in the TV and don't move, these will all dramatically change your BG.
I got to the stage were I could change mine by thinking about it.
(Increase, or slow your heartbeat, and release adrenalin or relax).

Some of these responses can be sometimes predicable, but then again.........

 

[Essex]

@rebrascora Yup, I am very clinical about hand washing and even checking that my wife hasnt used the teacloth as a dishcloth to mop up food (why do the Portuguese do that? ). I would have retested for an outlier, but had no idea if a value like that could be an error or not. I also thing that maybe I am more stabilised after a few days on a better diet in terms of lower carbs. But I'm probably wrong

@Docb @travellor. absolutely - so many covariates. I had to move to a stats based approach for my work as there were so many more variables to control and measure for non-homogenous anisotropic composites than for simple metals and because in the marine industry its usually made my hand layup so there are even more variables and loads of variability that cant be controlled. But, I am an (research) engineer rather than a scientist, and definitely an engineer who uses stats (at least I did for my PhD 25 years ago, forgot most of it now) not than a statistician, so I use engineering as the art it is to make reasonable assumptions, look for trends etc rather than take everything on pure stats.
The thing is, I know the human 'system' is infinitely more complex than composites (Drs must be geniuses), but have almost no knowledge of the diabetes 'systems' and so don't know what are common or relative values, or anything to be in a position to be able to say, for example, this variability / factor is generally low so we can ignore it etc
So, I am most defintely not trying to say grumpily 'dont teach grandma to suck egg' which I can now see this post might look like (my psychologist wife taught me to see things like that, for me if its fact then I dont always see the other human interpretations ) but is an attempt to put us all on the same page
I bet I have failed and look like a complete tw*t

What I am trying to say is I can see all the pitfalls of the stats but I'm bumbling along clumsily trying to make sense of it and my way of doing that is through science approaches
I also know that sometimes its best to drop that as I had to for learning languages quite late in life, but that's 'how I am'

Anyway, everyone's input has been of great use and I am very, very grateful to everybody, I am slowly getting the hang of this and I am confident that it will make such a big difference to my life as did all the things I found out about psychology (thats even more complex than biology 😱 ) and gout, and allergies and lactose intolerance (cured my acid reflux and bloating and signalled my liver enzyme and diabetes discrepancies). That last sentence makes me sound like a complete wreck - but its the opposite, natural erosion with age apart I am healthier than I was when I was in my 20's.

Another thing I tend to do is just write/speak as I think - work in progress.

Currently bouncing around 5, after 2 plain yoghurts and tea for brekky and a snack of nuts and seeds at 9:30

geekiness is a word

EDIT: And I think I am forgetting the other part of the puzzle - exercise, when I was younger I was always very, very active, but now I dont do much beyond walking - I know that was one of the most efficacious things with my anxiety and depression so I must not forget to get that going again. Tomorrow I will start jogging again!

 

[Docb]

@Essex. As an aside, did my PhD on anisotropic composites (On the Compatibility of Carbon Fibres with Metal Matrices). That was in the late 1960's and it is amazing to think that the theoretical part of my work, which took 6 months of hard brain graft, I could do today in a couple of days on a phone. Wonder where we will be in 50 years time and what will be thought about our current puny efforts in this area!

 

[Essex]

@Docb
Wow!
But I disagree!
The older the work I find tends to be so much higher and deeper 'quality' as the lack of tools to do fancy stuff meant they had to find real solutions to the underlying problems rather than just crunch numbers with computers and commercial software that gives pretty pictures.
If you got a PhD in the 60's in composites then I take my hat off to you.
Must have been really exciting as it was really cutting edge then.
I know I look at loads of the 60's work and find it amazing.
I look at my PhD and cringe

 

[Docb]

Thanks for that @Essex. The neat thing is that my fundamental conclusion that it was a waste of effort trying to get carbon fibre/metal composites to work in high temperature environments still holds. No matter how clever the analysis or how pretty the pictures are, the basic principles of thermodynamics and stress analysis will rule the roost.

 

[Deleted member 21371]

Thanks for that @Essex. The neat thing is that my fundamental conclusion that it was a waste of effort trying to get carbon fibre/metal composites to work in high temperature environments still holds. No matter how clever the analysis or how pretty the pictures are, the basic principles of thermodynamics and stress analysis will rule the roost.

I always remember something one of my lecturers told me, when I was writing a program in Fortran, on a teleprinter.
He reckoned you would never be able to get a video over a modem connection, just text and still pictures.
The maths showed the bandwidth of a copper cable simply was too low for the bandwidth required for a moving picture.
Which was true.
As you say the basic principles showed it was impossible.
But human beings moved the basic principles.
When the internet came along I started out with a 300bps dial up modem, and analogue transmission.
We then moved up to digital, and video compression.
I'm still on copper cables, and stream about five different videos at a time when everyone is in.
The impossible does happen.

 

[Deleted member 21371]

@rebrascora Yup, I am very clinical about hand washing and even checking that my wife hasnt used the teacloth as a dishcloth to mop up food (why do the Portuguese do that? ). I would have retested for an outlier, but had no idea if a value like that could be an error or not. I also thing that maybe I am more stabilised after a few days on a better diet in terms of lower carbs. But I'm probably wrong

@Docb @travellor. absolutely - so many covariates. I had to move to a stats based approach for my work as there were so many more variables to control and measure for non-homogenous anisotropic composites than for simple metals and because in the marine industry its usually made my hand layup so there are even more variables and loads of variability that cant be controlled. But, I am an (research) engineer rather than a scientist, and definitely an engineer who uses stats (at least I did for my PhD 25 years ago, forgot most of it now) not than a statistician, so I use engineering as the art it is to make reasonable assumptions, look for trends etc rather than take everything on pure stats.
The thing is, I know the human 'system' is infinitely more complex than composites (Drs must be geniuses), but have almost no knowledge of the diabetes 'systems' and so don't know what are common or relative values, or anything to be in a position to be able to say, for example, this variability / factor is generally low so we can ignore it etc
So, I am most defintely not trying to say grumpily 'dont teach grandma to suck egg' which I can now see this post might look like (my psychologist wife taught me to see things like that, for me if its fact then I dont always see the other human interpretations ) but is an attempt to put us all on the same page
I bet I have failed and look like a complete tw*t

What I am trying to say is I can see all the pitfalls of the stats but I'm bumbling along clumsily trying to make sense of it and my way of doing that is through science approaches
I also know that sometimes its best to drop that as I had to for learning languages quite late in life, but that's 'how I am'

Anyway, everyone's input has been of great use and I am very, very grateful to everybody, I am slowly getting the hang of this and I am confident that it will make such a big difference to my life as did all the things I found out about psychology (thats even more complex than biology 😱 ) and gout, and allergies and lactose intolerance (cured my acid reflux and bloating and signalled my liver enzyme and diabetes discrepancies). That last sentence makes me sound like a complete wreck - but its the opposite, natural erosion with age apart I am healthier than I was when I was in my 20's.

Another thing I tend to do is just write/speak as I think - work in progress.

Currently bouncing around 5, after 2 plain yoghurts and tea for brekky and a snack of nuts and seeds at 9:30

geekiness is a word

EDIT: And I think I am forgetting the other part of the puzzle - exercise, when I was younger I was always very, very active, but now I dont do much beyond walking - I know that was one of the most efficacious things with my anxiety and depression so I must not forget to get that going again. Tomorrow I will start jogging again!

If you really want a system for diabetes, chaos theory is probably your best bet.
At least anisotropic composites have some repeatability.

And don't worry how you come over, just keep going and let us know what you discover.
Any results you find will be interesting.
(Just worry about what your wife will be saying if you are still doing this next year though!)

The results of the exercise is going to be another interesting experience.
Some of us find our BG rises, others find it falls, and then you get into the timing, and the length of the change............

 

[Deleted member 21371]

Another annoyance I ran into was the accuracy of a meter, even without any other issues such as contamination.

A meter calibrated to ISO standards says they should provide results 95% of the time,
within +/- 0.83 mmol/L of laboratory results at concentrations of under 4.2 mmol/L
within +/-20%of laboratory results at concentrations of 4.2 mmol/L or more

but to be fair, you would expect a meter to be reasonably repeatable, from the same batch of strips and at the same temperature, at least.

Factors Affecting Blood Glucose Monitoring: Sources of Errors in Measurement

Glucose monitoring has become an integral part of diabetes care but has some limitations in accuracy. Accuracy may be limited due to strip manufacturing variances, strip storage, and aging. They may also be due to limitations on the environment such as ...

www.ncbi.nlm.nih.gov

And the spec for the nutritional values of foods, if you get into recording BG against carb intake, is simply they should be a best guess, based on analysis, or based on the previous guess by someone else on the ingredients, and not intentionally misleading.

Don't take this as trying to put you off, just another engineer that started of looking at my own response when I was diagnosed.

 

[Essex]

Mmmmm.
identical breakfast but no peak this time
Then same morning snacks and it goes up this time.
Then when I give up and eat a pasta salad it goes down!
Back to the drawing board I guess
I think I just need to take a big step backwards and exercise, eat carefully, measure on waking and after breakfast and stop obsessing!

Date​	02/12/21​	​	​	​
​	​	​	​	​
​	Time​	mg/dL​	mmol/mol​	Food / Activity​
​	​	​	​	​
Waking​	07:18​	95​	5.3​	​
​	​	​	​	​
Breakfast​	07:30​	​	​	2 x Plain 125 g lactose free yoghurt, Tea​
​	​	​	​	​
Post Prandial #1​	08:43​	80​	4.4​	School & dog walk​
Post Prandial #2​	08:50​	92​	5.1​	At computer, feeling very hungry & a bit foggy​
Post Prandial #3​	10:08​	99​	5.5​	Nuts and seed 9:30​
​	​	​	​	​
Elevenses​	11:30​	​	​	Sunflower seed toast ham cheese​
​	​	​	​	​
Post Prandial #1​	13:15​	157​	8.7​	At computer​
Post Prandial #2​	13:21​	146​	8.1​	At computer​
​	​	​	​	​
Lunch​	13:30​	​	​	Pasta salad​
​	​	​	​	​
Post Prandial #2​	16:10​	121​	6.7​	At computer​

 

[Docb]

I always remember something one of my lecturers told me, when I was writing a program in Fortran, on a teleprinter.
He reckoned you would never be able to get a video over a modem connection, just text and still pictures.
The maths showed the bandwidth of a copper cable simply was too low for the bandwidth required for a moving picture.
Which was true.
As you say the basic principles showed it was impossible.
But human beings moved the basic principles.
When the internet came along I started out with a 300bps dial up modem, and analogue transmission.
We then moved up to digital, and video compression.
I'm still on copper cables, and stream about five different videos at a time when everyone is in.
The impossible does happen.

Ah, but electron movement down a conductor is not controlled by the laws of thermodynamics or the basic principles underlying stress analysis. Your lecturer was bound by his imperfect understanding of fundamentals. No wonder he got it wrong. 😉

 

[Essex]

Waking 5.5
I went jogging after my breakfast of 2 plain yoghurts and a cuppa - woohoo!
2 hrs later 4.8
No more testing (today)
I am happy with that 🙂

On a geekier note,
Have you two read the book 'The Order of Time' by Carlo Rovelli?
I could only follow it as best I could, feeling like the youngest, smallest kid on a bike ride (it was always me), but it is crazy!
I never knew all this quantum stuff was so old in terms of when it was developed/discovered, and I finally even almost nearly understood what Entropy actually is rather than just mixedupiness (certified correct word there).

 

[Essex]

So, now using Tidepool its much easier for me to get my trends from the data
I have only been testing before and after breakfast mainly since Friday and have been always within range after having two yoghurts for breakfast
In fact for the last 4 weeks, based on 67 readings I have an average BGM of 6.1 mmol/L with a CoV of 29% (apparently, 'research suggests less than 36% is OK)
2 weeks, n = 38, Ave = 6.1, CoV = 26 %
1 week, n = 17, Ave = 5.6, CoV = 23%
The 1 week stats are a bit 'skewed' as they only pretty much include waking and 2hrs after b'fast

From here, this corresponds to an HbA1c value of 36 mmol/mol (5.5%)
"*The eHbA1c calculated here is just an estimation based on the average blood glucose you have entered. It should not replace the HbA1c check carried out by your doctor and the result shown here may differ from the lab value you get. This calculation should not be used to make therapy decisions or changes. You can discuss with your healthcare professional how this approximate measurement could help you in your overall diabetes management."

So, I understand that this value of 36 mmol/mol is not definitive as it is only based on a few 'snapshots' per day but the general trend seems good and the values are not fluctuating wildly either (OK Coeff of Variation values)

I did remember that before the last blood test I was eating a lot of bananas as the only fruit I can really, and I didnt realise that they mattered ('natural' sugar and all that)

I'm also eating smaller portions of less carbs and watching what I eat,

That's it.

Quite happy with that

Thanks so much for helping me start on my diabetes education