Discussion in 'Latest ME/CFS Research' started by Dolphin, May 12, 2010.
this may be rather late, but they may find the following helpful:
Dr. Peter Rowe and Dr. Chris Snell talk about the PACE trial at an FDA meeting:
ETA: I hadn't watched that yet. Their full talks without any commentary can be found here in Panel 3:
Tom Kindlon's harms paper (has a section on the PACE trial; this is a peer-reviewed publication):
interview with Tom: http://phoenixrising.me/archives/15567
video on changed targets in the PACE trial:
enitre series: https://www.youtube.com/user/MEAnalysis/videos?flow=grid&sort=da&view=0
website with more detailed info: http://evaluatingpace.phoenixrising.me/homepageanim.html
Nunez paper showing that when symptoms other than fatigue are checked, one might find declines:
Information on a FOI request regarding PACE:
Original FOI request:
http://www.informationtribunal.gov.uk/DBFiles/Decision/i1020/20130522 Decision EA20120229.pdf
Make sure they know about the International ME Consensus criteria.. there's a link to it on the front page (on the right in you scroll down) on the SA ME/CFS society website. That probably has stuff in that about GET and CBT and cautions etc and could be used.
That really sucks. At least thou they will have to publish that data as it will look very bad on them if they dont within the next couple of years and they must realise that now (fortunately their freedom of the card they played will not keep holding out.
The card they tried to play about how it would affect "academic freedom "by being forced to publish didnt work for them (if it had, they would of thought they could forever get away with never publishing the info). They see now they HAVE to publish info in a reasonable time.
Thank you Robert Courtney for taking that to court.
Exactly ... from what I understand, all of their other excuses have been dismissed, and the "we're gonna publish, honest guv" is all that's left. And it won't hold up forever
I wonder how they are going to spin the whole thing, it must be difficult to keep making bad results look good, I hope its giving them stress.
I wonder if they will put out a completely different thing in their favour (a new study or a new spin on a different old one) at the same time they put out the spun poor results, to try to get attention away from them.
This request went all the way to the tribunal and even though the Tribuanl ruled in favour of Queen Mary University London withholding the information, it was only on the basis that publication of that information in a journal was imminent:
Hats of to Robert Courtney who made the initial request and perservered with it too:
His point was that the initial 2011 Lancet PACE Trial paper gave the improvement rates, but that to properly assess the effectiveness of the trial it was important to know how many patients deteriorated by the same measure.
And the ruling also contained and important precedent in favour of disclosure, by explicitly rejecting the argument of the 'academic freedom' to choose what, where and when to publish as suits the interests of the academics:
Hopefully we will see the those deterioration results published shortly.
edit: the best news is that situations like this should not arise in future. The MRC has said that trials funded by them from 2012 onwards are expected to release ALL trial results within 12 months of trial completion. PACE will have taken around 2 years to release these deterioration results.
Thank you, Robert Courtney.
When the Nijmegen BPS group had bad actometer results but good questionnaire results in a set of three studies they 1) didn't mention actometers in the initial papers 2) published the actometer data a couple years later for all three studies in one paper and 3) spun the results to say that actual physical functioning isn't relevant to CFS patients because it's not associated with reporting less fatigue.
I think we'll see something similar from the PACE group regarding deterioration rates. Hopefully it'll get more attention than the ludicrous Nijmegen spin got though.
4) They speculated that hidden increases in activity actually occurred halfway through the trial after the baseline actometer measurements but were converted into increased social activities before the followup actometer measurements.
being the key word here.
Thanks for that. Looks like Rob Courtney did a good job, and it was just that they are now planning to release the data that led to the appeal being denied.
lol at the way the actometer results were spun.
Personally, I'm not expecting CBT and GET to have particularly bad deterioration rates from PACE. The sessions were all being recorded and assessed as part of the trail, so I think that the therapists would be less likely than normal to cause problems. Also, APT looked pretty rubbish to me, with the 70% rule and the various forms and diaries that needed to be done quite possibly making people feel worse too. It would not surprise me if APT did worse than CBT/GET... although the fact that the data has been slow to be released does make me suspect that CBT/GET could come out less well.
But then, all three of those treatments involved much more therapist time than SMC alone, and this would be likely to bias towards more positive questionnaire results.
It is hard with PACE as i) these researchers have shown that they cannot be trusted and are happy to spin data in order to sell their treatments and expertise, ii) it is genuinely difficult to measure the impact of behavioural interventions on symptoms.
Good work Robert Courtney! If you happen to visit Phoenix Rising and read this thread, I just want you to know that I think you still practically succeeded anyway, because I doubt PACE would have published this data on their own so at least we get to see it. I doubt the AllTrials movement would have bought their excuses either, and as Simon suggested earlier, the result sets important precedents and the stonewalling we have gotten used to is becoming less accepted.
All the PACE-related FOI requests I am aware of (eg https://www.whatdotheyknow.com and Co-Cure) are similar to this, about accessing unpublished data which should have been published already or was promised to be published in the past. A few may be expecting too much, but where are all the FOIs which supposedly amount to vexatious harassment and choking up White's time (according to news coverage on PACE)? Is this FOI request what White personally regards as vexatious harassment and one example of what he claims to be wasting 1/4 of his time on?
Well, I guess FOI requests would be "vexatious" i.e. annoying if one does not like the implications of the data being released or does not like losing control over the context it is interpreted in. As one of Tom Kindlon's posts on Co-Cure exemplifies, even simple questions about the recovery paper are being turned into official FOI requests. But it would appear that people such as Paul Smallcombe et al are doing all the work responding, not Peter White et al.
Good point. From what I recall the CBT and GET treatments were actually both pretty similar, and had a strong pacing element in them - being told to back off if symptoms intensify, rather than sticking to a scheduled increase in activity?
Agreed. I doubt the deterioration rates will be shocking. The CBT and GET groups did have a pacing element, but were encouraged to tolerate more temporary symptom exacerbations than APT. Everyone also received SMC and material which contained several key elements of pacing. The vast majority of people in the trial failed to improve or "recover" due to CBT and GET, so will need to fall back on simple autonomous pacing to prevent relapses.
final paragraph from
I set up a thread on this at: http://forums.phoenixrising.me/inde...-unpublished-data-from-a-rct-pace-trial.23716
Peter White letter in the BMJ.
Of course, there have been lots of other freedom of information requests which would not break confidentiality e.g. the data for the original primary outcome measures, deterioration rates, recovery rates (as in original protocol), etc. and he has released virtually nothing following such requests.
Also, no researchers may request the data in which case important data might not be released.
Or researchers who do request the data may only do minimal analyses that still don't reveal interesting information within the data.
Only the most repulsive of quacks would try to wriggle out of releasing data in the manner they laid out in their registered protocol Peter.
No they're not, which is why government statistics departments who request datasets from other departments specify which variables they need for their analysis. Any that they don't need, or can't adequately justify needing, are stripped out. Such data requests are considered on a case-by-case basis by someone whose job it is to analyse at what level of aggregation of variables confidentiality might be breached.
God forbid that somebody applies some common sense, let alone expertise, to this problem. Just take locality out, for a start. Who wants that?
This disgusts me.
You can also try a Google Site Search
Separate names with a comma.