• Guest, the forum is undergoing some upgrades and so the usual themes will be unavailable for a few days. In the meantime, you can use the forum like normal. We'd love to know what you think about the forum! Take the 2025 Survey »

Home blood prick tests - a waste of time?

For weeks now, I have been testing twice a day on 2 fingers (same hand) before meals and within a minute or less of each other. Consistently I have been getting randomly different results. I have been recording the readings and there is no pattern whatsoever to explain the sometimes very large discrepancy (varying from nil difference - only once - to more than 2.4, with an average of about 0.8) between the two results. The highest has been 8.9 - but not often. The lowest 4.6 - again not often. Most readings in the 5.8 to 7.6 range, but apart from once, the readings from each finger have always been different.
Has anyone else experienced this? How should I interpret the results? Why does it happen? and am I wasting my time with a device that spews out totally random numbers?
 
Hi

The thing with glucometers is that there is an allowable inaccuracy of 15%, 95% of the time. This is how it works in practice.

Say you have a true blood glucose value of 6.0mmol/l.

15% of 6.0 is 0.9. So - given a true BG value of 6mmol/l, if the glucometer produces a test result somewhere in the range 5.1 to 6.9 (95% of the time) it's sufficiently accurate.

5% (one in 20) readings might well be outside this range, and that's allowable too.

The implication of this as far as I'm concerned is not to get hung up on a single reading, or differences between almost simultaneous readings. I've had a reading of 2.2 (after a meal) in the last couple of weeks, but it was very obviously just a rogue result, I was not hypo no matter what the meter said.

The glucometer is great if you build up a bigger picture over time of how your system responds to and deals with particular foods, usually repeating a few meals and tests if needed. The inaccuracies average themselves out, and in a few weeks you should have a much better idea of things you might want to limit eating. That's why I would usually never do single snapshot tests to "see how high you go" - it's a one-off and subject to built-in inaccuracy, and tells you very little useful information.

If you've been getting "most readings in the 5.8 to 7.6 range" over the weeks - that looks like you're on the right road*. It's the pattern over the weeks and months that's more informative than any single test.

*I checked, and I think you say you are only doing pre-meal readings. Do you have any BG information from post-meal readings? that's what shows how well you coped with the food, by comparing the second result with the pre-meal.
 
Last edited:
To find out how well your body is dealing with your carb intake, you usually take a reading upon the first mouthful of food, then another reading 2 hours after. If your body has dealt with the carb intake then the Second reading should not be greater than 2 mmol/ls after your first reading.

When I test my blood sugars I don't use the same finger twice, and I only test once, I don’t keep repeating the test one after the other.

Glucometers are calibrated to an accuracy of within 15% Of a lab reading. I believe that’s the standard allowable error. So you are always going to get a variance Within those parameters.

Blood sugars are constantly in flux, even when you have nothing to eat there is a slight variance. The body is always adjusting and compensating to keep our blood sugars in homeostasis. It is a highly complex process. So I would suggest no two reading will be the same.

If you do suspect your meter is inaccurate you may wish to go to a pharmacy and check your readings against one of their devices. I live in Canada. Our pharmacies can do a blood glucose reading for you. I’m thinking you can do this in the UK too . I may be corrected on this. Edited to correct an autocorrection.
 
ajbod said:
Random testing is useless, and a waste of strips
Click to expand...
In the US, a random or a fasting bg of 5.6 or over are both taken to indicate pre-diabetes. The WHO & NHS set the bar higher. So a random test is not imo totally useless, but a fasting test recorded every morning before eating or drinking anything except water can be very useful, as are tests around meals. In my experience, the timing of tests after meals is a bit problematic, as 2 hours after eating I can see a reading lower than before the meal, because the food has not yet arrived in my blood stream. I can see a post-prandial rise eg 4 hours after eating. I therefore try to consume a very low carb/keto diet, sometimes record my carb intake for the day on a spreadsheet, and don't often try to check around meals.

For some people, testing fasting first thing in the morning is problematic, because of the Dawn Phenomenon. However, the great Richard K. Bernstein says in his classic book "the Diabetes Solution" that the DP is likely caused by a surge of growth hormone, which is less marked as we age. So at 82, I am not often troubled by the DP! There are advantages in ageing.

Unless using a cgm, we can't know whether our bg is regularly rising while we sleep, which would of course affect our A1c numbers. Cgms are expensive and probably even less accurate than finger prick tests, but it could be worthwhile to invest in a sensor for just 2 weeks to see in real time your bg rising and falling.

I find testing my fasting bg hugely motivating.

There is an alternative to the A1c test: the Fructosamine test.
 
There are a number of medical papers on this subject. May I suggest you Google “bmj results may vary”. And read the advice given to doctors. Specifically, the British Medical Journal published a tool which doctors can use to compare variation in laboratory tests including HbA1c and blood glucose. The above Google will show a link to the BMJ article and thus to that tool. You can enter HbA1c and your results and then see just how much, or how little!, they mean. The tool also shows results of testing blood glucose levels in the laboratory. This with expensive equipment, large blood samples and experienced operators. Consider if your 25p strip and single drop of blood taken by an amateur can be even as accurate as this.! The only way to increase your accuracy is to take Multiple readings and then use an Average.
 
A result of 42mmol actually indicates a wide range of actual level, including within the normal nondiabetic range:
 
Bill_St said:
Sorry to give a complex reply but too many people (and medical professionals) think they can rely upon a simple Number. They should, imho, explain much more View attachment 69442
Click to expand...
The blood test results do naturally show significant variation. But BG levels do, quite normally, show wide variation due to many factors, including recent diet and exercise. But do remember the above variation is large sample controlled laboratory test. Your home test will be much wider
 
I test my blood on waking, before and 2 hours after meals and log it on Diabetes:M. Estimated hbac1 is pretty accurate, within .2 mmol. I’m self fund a cgm as I’m on insulin, so can obviously see if my blood sugars rising and take action.But doesn’t happen too often as I am on low carb diet and have good blood sugar control.
 
Make sure you wash your hands and all the soaps cleaned off
 
Just a small correction on your US fasting figures a Ac1 (HbAc1) range for Prediabetes in the US is 5.7% to 6.5% (39- <48 mmol/mol )
 
Melgar said:
Just a small correction on your US fasting figures a Ac1 (HbAc1) range for Prediabetes in the US is 5.7% to 6.5% (39- <48 mmol/mol )
Click to expand...
I was not citing the US A1c numbers. Fasting before testing is not required for an A1c test because that is supposed roughly to give an average of the last 3 months' numbers (though it seems that in fact the more recent weeks count for more). I was referring to the fasting test taken when nothing has been eaten or drunk thay day, except water. 5.6 was my UK translation of the US figure (100-125mgdL).

In any case, the US A1c numbers you quote confirm my contention that numbers qualifying in the UK and EU as non-diabetic are considered pre-diabetic in the US. I prefer to follow the US rules, as I believe much if not most research into diabetic complications has been done in the US and shows problems already arising at their pre-diabetic level. My guru, Dr Richard K. Bernstein encourages us to aim for much lower bgs, equivalent to the bgs he sees in "normal" people. Personally, I have not succeeded in getting my bg so low. It is not for want of trying!
 
Em50 said:
For weeks now, I have been testing twice a day on 2 fingers (same hand) before meals and within a minute or less of each other. Consistently I have been getting randomly different results
Click to expand...
As well as testing before meals, have you also been testing eg 2 hours after meals? It would be interesting to see if the post prandial rise is consistent eg 2 mmo/L, though of course the rise and its timing will vary according to how many g carbs / fat / protein are in the meal, and also any alcohol will have an effect.

I agree with others that single readings should not be taken too seriously. (This applies even to single A1c tests.) Have you tried testing with 3 strips from the same drop of blood? I have seen wildly different readings doing this. Still, imo testing does help - to keep us honest and deter carb creep and also to give a general idea of what is going on and what adjustments we may need to make.

Yes, the inaccuracy of finger-prick tests, and indeed of most if not all medical laboratory tests, is annoying and disheartening, but imo we are all the same so much the better off now we can test and experiment at home. It was not always so.
 
As a nerdy type, I finger prick tested on waking up, before every meal, 90 minutes after every meal and before bed. I recorder and weighed all food and was able to find patterns. The Lifestyle clinic (NHS in South Wales) helped with video sessions on sugar, metabolism and so on. It turns out I had something called Dawn Syndrome - high BG in the morning, I tried eating something carb based before bed and this was quickly resolved. By mapping every single meal I could see patterns and spikes, then I started excluding things - mostly simple carbs, sugar, bread, flour products, wheat/oats, pasta, potatoes and rice. I am now firmly in remission, and still test every morning and follow a low carb diet. It's not for everyone, but I felt in control of the whole process, I can now tolerate a few potatoes without spiking bloods.

This strict regime is not for everyone, I also used an app called mySugr (free for basic stuff) this app calculates an average HbA1c score but it requires quite a few readings in every 7 day period.
 
so just to clarify - something carb based before bed reduced your Dawn Syndrome measurements of BG on waking ?

Just asking because although i am also firmly in remission i still seem to have high BG on waking adn i wonder what the mechanism is that it addresses

Thanks
 

That’s really interesting and something that has worried me, hence, I posted a new topic to ask ‘how high non-diabetic blood sugars rise after a meal’, and during, for that matter. Your information now helps me to understand that, but I’d still be interested to know the answer to my recent topic. I’ll leave it there but thanks.
 
You must log in or register to reply here.
Menu
Log in
Register
Cookies are required to use this site. You must accept them to continue using the site. Learn More.…