Hi,
I'm not new to the forum.
Hypothetically, if I had blood drawn at the same time and blood was sent to both NHS and Medichecks (for example), is it reasonable to expect the percentage through the ranges to be pretty similar (within a few %)?
I know each lab has their own kit and their own ranges, but can % from labs be compared?
I assume they should be pretty similar otherwise there's no real way to compare.
Thanks!
in my experience ... no .
if using test assays with widely differing ranges , ie. one very narrow , one much wider, then the %'s could appear very skewed.
This is not perfect as 'evidence' as the blood was taken 45 minutes apart after i'd walked home from GP test ,,, but :
~~~~~11am (GP veinous ) ~~~~~~~~~11.45am. (NHS fingerprick/ postal)
tsh........2.65 [0.57 -3.6] .................................2.9 [0.27 -4.2].............................................
fT4 .....15.7 [7.9 -14] 128% .......................22.1 [12 -22] 101% ........... fT3 4.0 [3.1 - 6.8]
Thank you!
Your TSH is within what I'd call a margin-of-error (68.6% at 11, 66.9% at 11:45 = 1.7% difference).
But you FT4 is significantly different.
I noticed that the ranges of my tests are varying and my spreadsheet only has one set of ranges so it's time for something more sophisticated! I'm the author of the blood test calculators (thyroid.chingkerrs.online) and I'm thinking of writing a proper blood result tracker (honing my database skills that I'm studying).
yes ,, i know what you mean ,, been there and bought the t- shirt here : healthunlocked.com/thyroidu...
the only way i can figure of getting all result on one graph is to convert all to % ,,, but it seems we can't even rely on that , or at least not if one of the ranges involved is significantly narrower than the others .
TSH assays have become more standardised over recent decade , but fT4 / fT3 assays are still a disorganised shambles, and endo's don't care enough about them to push the assay manufacturers to sort it out.... so we are left with a mess that makes it more or less impossible to produce a nice graph ... or even to know with any certainty what our levels actually are ... the best we can do is look at fT4 / fT3 results as a 'ball park' measure . i know my fT4 on that day was "very top end / bit over ". but there seems no point trying to interpret them more precisely than that until we have access to more standardised testing assays . .. ideally i would always use the same lab / assay ,,, but it's not always possible ,
Maybe the T4:T3 ratio is usable, but it's only one aspect.
The image is a sketch of how I envisage graphing results over time. The pink boxes represent the range from the lab and the black dots are the result.
I'd not draw lines between results as that is a lie - a straight line implies too much - who knows what happened between those two times. (sorry too much detail, I'm getting into dev mode 
)
In the past, we used often to see TSH ranges of 0.5 up to as high as 7 or 8. The reasons for that being:
They did not rigorously exclude people with thyroid issues when setting the ranges.
They did not take steps to avoid any impact from things like antibodies interfering with the tests.
By a combination of seeing enhanced tests from the manufacturers and more considered range setting, the top ends of TSH ranges dropped all round. Often now under 5.
Someone who would have had a result of, say, 2 years ago, would quite likely still have a result of 2.
But their percentage through the range would have changed from 20% (0.5 to 8.0) to 33.3% (0.5 to 5.0).
That is a case where the issue had been over-reading on the tests.
If you had two labs, and they tested FT4, and the results were 17 (12-22) and 15 (10-18.3) - it would be so easy! You could just multiply the results by 5/6 (or 6/5 to go the other way).
But when you see ranges like 7-14 - almost to the point of there being no overlap at all, it is less persuasive. Or when you see 10-24 - one range entirely including another range.
If we had results for people at some particular reference points, say 20, 50 and 80% through the ranges, tested at two labs, I think we might get somewhere.
But the somewhat simplistic calculation of percentage through range is, unfortunately, the best we've got. We simply do not have any other tools for assessing tests from different labs.
And on other analytes, it gets even worse. Ferritin is terrible. Age a day and your percentage could jump huigely at one lab - let alone acorss labs.
helvella - Iron Panel
A short article about explaining what is meant by "iron panel" in relation to blood tests including some reference interval (range) information.
Last updated 18/07/2024
helvella.blogspot.com/p/hel...
I cannot think of a better approach than your pink diagram. Just needs (as always) to be taken as a general indication not having pinpoint accuracy.
yes agreed , line between points is hugely misleading ... any given point could be on its way up or down .
i suspect ratio has the same 'comparison' issues as using individual T4/T3 results does , i thought about it once,
people were comparing ratio's achieved by mulitplying / dividing pmol numbers from different assays , but how can it not matter what the fT4 [range] is when
14 on a [12-22] is low fT4 .... but
14 on a [7.9-14] is high FT4.
so we can't reliably use the number '14' to work out a ratio and then compare it to a ratio from a different assay .... the resulting ratio would be significantly different ... this is maybe only a problem at the extremes ...... with FT4 ranges that are not much different from each other , the difference would be smallish ,but the 7.9-14 ish assay is a significant issue because it is so much lower and narrower than most other fT4 ranges that it skews things quite a bit .
Helvella and i had a brief discussion about the effect ranges has on ratio, in the replies on this post : healthunlocked.com/thyroidu....
some other stuff i have about ranges is here if it's of any interest : healthunlocked.com/thyroidu....
Plot-able graph for test results
Results in on blood test.....help needed please
Help interpreting private blood test results
Ft4 range different labs
Blood results - advice 
