PACE trial TSC and TMG minutes released

BruceInOz said:
For two people with the same heart rate change, the person with the greater power will have higher score.
Click to expand...
And by the same token, for two people exerting the same power, the person with the lower heart rate change will score higher, which presumably implies being fitter.

Edit: Sorry, just realised you already said that!
 
Barry said:
And by the same token, for two people exerting the same power, the person with the lower heart rate change will score higher, which presumably implies being fitter.

Edit: Sorry, just realised you already said that!
Click to expand...
Exerting the same power sounds ominous and kind of ironic!
 
The problem then with the version they have used is that it is not adjusted for age, sex or BMI, which means it is only really useful for looking at change in fitness within the individual who has been measured, and not across a group.

Also, it is supposed to be a submaximal test, where did they get the HRmax data from? They should have at least said what they used for that. [Thanks @Trish (see below)]

But that was my main criticism with how they analysed all their measures. They should have looked at mean (or rather, median) changes, not changes in group means.
 
Last edited:
%HRR = [(highest measured HR - resting HR) / (predicted max HR - resting HR)] x 100
Predicted max HR = (220 - age in years) for males; (206 - 0.88xage) for females.

So a sample calculation for a 50 year old female whose resting HR is 70 and whose max heart rate during/after the step test is, say 150, would go like this:
Predicted max HR = 206 - 0.88x50 = 162
% HRR = (150 - 70) / (162 - 70) = 80/92 = 0.87 = 87%

A sample calculation for a 30 year old man with the same resting HR and max HR would be:
Predicted max HR = 220 - 30 = 190
% HRR = (150 - 70) / (190 - 70) = 80 / 120 = 0.67 = 67%

This has therefore taken into account age and gender. Whether it does so in a logical way is open to question.

I guess it does, because the younger man in this instance, though he has the same measured heart rates as the older woman, is not getting so close to his max heart rate as a result if this exercise, and is therefore fitter. This is shows up by a reduced divisor, so a larger fitness measure.

Although it doesn't use BMI, it uses body mass, which is the relevant thing in a step test because it involves liftng that mass up and down the steps.

Or am I talking rubbish?
 
I've done a sample in Excel:
STEPtest_sample.png

The problem is that the equation assumes that if you are younger, you are automatically fitter. That's not going to be the case if you are ill. Surely?

If you self-pace and go slower, you get a much lower score.

Also, the results I'm getting don't bear any relation to those quoted in the mediation paper graph. :/
 
You only have to read the Petrella paper properly to realise that this is never going to work anyway. They studied *healthy* older adults. As soon as they put non-healthy people into their models, they got garbage out - the graphs clearly show they lost all correlation. This was never going to be a good idea. It makes me so frustrated that these important studies seem to be thrown together with so little background work. :mad:
 
Interesting too that Resting HR affects the score.

I notice that now I've got ME I have a higher resting HR than I used to have.
Also if I go somewhere (eg going for a test like this) my HR stays higher again just from being out and about.
So, I wonder, how they ensure they have a sound Resting HR measure.

And putting a higher resting HR into this equation would decrease the %HRR score. Hmmmm
 
Also isn't there something that says with ME our HR doesn't reach our max properly?

So do they test lactic acid accumulation or something to test for inefficient aerobic respiration?

I might not be quite "getting" this whole calculation here.
 
Maybe we could ask some of our fitter/ less badly affected members or friends to try the step test themselves and either do the calculation or give us their data so we can get some realistic figures.

Those calculations were based on numbers we invented and may not be realistic. I'm tempted to try it myself, but I know it would set off a relapse unless I spend a whole day doing it in stages and did nothing else. Not a sacrifice I'm prepared to make for the sake of this conversation.
 
Trish said:
Maybe we could ask some of our fitter/ less badly affected members or friends to try the step test themselves and either do the calculation or give us their data so we can get some realistic figures.

Those calculations were based on numbers we invented and may not be realistic. I'm tempted to try it myself, but I know it would set off a relapse unless I spend a whole day doing it in stages and did nothing else. Not a sacrifice I'm prepared to make for the sake of this conversation.
Click to expand...

All you have to do is use the data from the Petrella paper itself, which I did in my excel version. It simply doesn't work. You don't get anything like the numbers shown in the Mediation paper, and if you actually play with the James equation, it doesn't seem to give any consistent reflection of "fitness" as such. Different ages give very different results for the same level of performance on the test.

For example, if you use data for a fit man in their 30s, and then only change the age, you get a much worse fitness score. That can't be right. If their performance is the same at an older age, they should get a better score.
 
Lucibee said:
Also, the results I'm getting don't bear any relation to those quoted in the mediation paper graph. :/
Click to expand...

I've not looked into what you're saying, but it would be great if you could explain it a bit more before I tried!

I thought that they'd given very little info on their fitness data, so how are you getting independent results of any kind? Thanks.
 
This may be a really stupid comment and if it is please ignore.

One thing that really stands out with my HR monitor is that depending on my level of PEM doing the same task on different day gives a very different HR reading. Even if the routine on each day is pretty much identical. PEM definitely seems to make a difference.

If what I see is accurate and seen by other ME patients then you would need to somehow adjust for levels of PEM (no idea how you could even do that) to make sense of the readings of two different sessions with the same patient.
 
Lucibee said:
All you have to do is use the data from the Petrella paper itself, which I did in my excel version. It simply doesn't work. You don't get anything like the numbers shown in the Mediation paper, and if you actually play with the James equation, it doesn't seem to give any consistent reflection of "fitness" as such. Different ages give very different results for the same level of performance on the test.

For example, if you use data for a fit man in their 30s, and then only change the age, you get a much worse fitness score. That can't be right. If their performance is the same at an older age, they should get a better score.
Click to expand...

Yeah my carer who is 25 years older than me is much fitter. I've measured their HR for fun when I've had my equipment out and it's always better than mine.
 
Yes Invisible Woman I get worse HR values if in PEM.

Although I wonder if anyone could recommend a precise and accurate wrist HR monitor for continuous use in day ? My cheap £30 one seems to show some v high rate which I'm not sure are accurate or precise.

I'd love to be able to use an objective measure so that I can stop tipping into that poisoned PEM state?
 
Esther12 said:
I've not looked into what you're saying, but it would be great if you could explain it a bit more before I tried!

I thought that they'd given very little info on their fitness data, so how are you getting independent results of any kind? Thanks.
Click to expand...

Just the results from the Mediation paper. Here is the graph:
fitness_Pace.png
They were getting scores of between 1.5 and 2, whereas the James equation gives scores <1.0 (unless you are young and fit), and the VO2max estimate from Petrella et al will give scores of between 20 and 30.

And as you all say, if you don't take PEM into account, it's all pretty meaningless anyway. Had they repeated the step test again 24-48 h later, they would have clearly seen those with ME (but also broken them in the process). The only value of doing the step test, if done properly, is diagnostic - it's yet another case where it is not appropriate as a repeated measures instrument.

All these fitness tests will do (and that goes for the 6-min WT too, I would suppose) is measure how good at pacing someone is, and not necessarily their level of fitness, because anyone who is being sensible won't be pushing themselves. Utterly counterproductive.

Actigraphy would have been much more useful, had they ever worked out how to collect it, analyse it and use it properly.
 
Last edited:
Robert 1973 said:
@Lucibee I feel you’re on to something although I’ve not managed to follow all the details. I think these issues with the step test need a wider airing and I’ve not seen them discussed elsewhere. Would you be willing to try putting it all together in an article?
Click to expand...

I'm not sure whether we have enough information to do that yet. But I can make a start, if you think that would be useful. I'll put together something for a blog - then it will make it easier to discuss if everything's all in one place.

Our main problem is that the only place the step test results are mentioned is in the mediation paper. I think @Graham has already tried to get the data from this, and was rebuffed with the usual "vexatious" response.
 
Lucibee said:
I'm not sure whether we have enough information to do that yet. But I can make a start, if you think that would be useful. I'll put together something for a blog - then it will make it easier to discuss if everything's all in one place.
Click to expand...

Thanks @Lucibee, I would be interested too. I have lost track of the thread on this. Are we talking about the step test in PACE? I have even lost track of which study we are in! Four sentences either from you or one of the others to get the rest of us get our heads around the punch line here would be nice.
 
It would be interesting, @Lucibee. I have no idea whether it would be useful.
I guess it's all part of filling out the picture of the incompetence of the PACE researchers to choose appropriate and validated 'objective' measures.

I think actometers worn for a week (or preferably throughout the trial) would capture a much clearer picture of both the patients compliance with the therapy and physical response both in terms of PEM episodes and longer term activity levels than any one off test like the 6 minute walk or step test because of the effect of PEM muddying the picture.

And maybe patients were primed in advance of the follow up tests to be well rested so they could do them - and they were allowed not to do them if they weren't well enough. Actometers would have a better chance of avoiding these distorting factors.
 
You must log in or register to reply here.
Back
Top Bottom