"News" 8 Sep 2016: PACE trial team analyse main outcome measures according to the original protocol

[snowathlete]

They've not released the recovery data, and that's what will cause them the most trouble. If this was an attempt to get their presentation of pre-specified analyses out before others got the data they'd have included recovery outcomes imo. I think that this is an attempt to avoid some of the blow back from the scientific community that they're expecting when they appeal.

Hard to see this stopping blow back from science when the figures are so damning. I think they didn't do Recovery because those figures are even more damning. This is damage limitation (though a bit futile) in the hope that people see this and therefore the Recovery rates that follow will maybe pass them by - they'll think they've already seen the latest PACE news.

 

[sarah darwins]

How very House of Cards. Not scientists, spin doctors.

 

[user9876]

Also

33 patients (21%) improved with GET yet they claim that 32 of those (22%) recovered. 43 (28%) were in the normal range and 81 (56%) no longer met the oxford criteria.

With SMC
16 (10%) improved but unlike GET and CBT only 11 (7%) recovered. 22 (15%) were in the normal range and 62 (41%) no longer met the oxford criteria.

With APT
15 (9%) improved. 12 or 8% recovered. 25 (16%) were in the normal range and 64 (43%) no longer met oxford.

(Figures on improvement from their recent release the rest from the recovery paper)

So the normal range looks dodgy as does the revised recovery criteria and the Oxford criteria that they revised for the recovery definition,

 

[user9876]

The fact that they are able to do an "analysis" at all helpfully demonstrates that the data is still available and intact, thus preventing them from using the "Gremlins Defence".

Why are they doing this? Because they hope it will help them when they appeal? Because they hope it will help them when an action is brought for contempt of court for not releasing the data? Because they hope it will pre-empt and deflate criticism from real scientists when they finally get to analyse the data? Who knows, but whatever their desperate plan is, the chances are they have shot themselves in the foot with their own incompetence once again.

I wonder if they will say look how open we are being and then appeal. But given a comparison between those that improved and those they claim recovered is so dodgy I think their release strengthens the case for data release.

 

[worldbackwards]

I wonder if they will say look how open we are being and then appeal. But given a comparison between those that improved and those they claim recovered is so dodgy I think their release strengthens the case for data release.

Well, exactly. If they were that open they'd just release the data. It just makes you wonder what else is in the vault.

 

[Sidereal]

Based on this analysis and their strange behaviour / appetite for endless litigation, I think it's safe to assume that recovery rates were in the 0% range.

 

[JohnCB]

Let's not forget to thank Alem Mathees for shaking the PACE tree. We are now seeing why they were so opposed to FoI requests. I doubt this would have been published if it were not for the hard work and eventual success of Alem and those who supported him. I'm looking forward to seeing what else comes tumbling down.

 

[daisybell]

If I'm reading this correctly, the bottom line is that 10% of people improve naturally over time (SMC). If you focus on exercise and tell people this will help, an extra 11% will report some improvement. If you give people CBT and tell them they will feel better, 10% say they do to some extent. Those figures are dreadful considering the power of the placebo effect!

 

[paul80]

is SMC = standard medical care?

 

[Chrisb]

There was somewhere in one thread discussion of the possible "nocebo" effect on those entered in the SMC arm of the trial which might further reduce the alleged improvements.

It will soon be time to start asking the traditional questions.

Who knew what and when did they know it?

My main regret is over the ones who got away.

 

[Simon]

If I'm reading this correctly, the bottom line is that 10% of people improve naturally over time (SMC). If you focus on exercise and tell people this will help, an extra 11% will report some improvement. If you give people CBT and tell them they will feel better, 10% say they do to some extent. Those figures are dreadful considering the power of the placebo effect!

I think strictly speaking, what you are describing is response bias, ie people changing their scores as opposed to really improving. There's now good evidence that the placebo effect doesn't really amount to much - almost nothing doing on objective measures of improvement (across many illnesses), hardly any better on self-reported outcomes.

Placebo interventions for all clinical conditions. - PubMed - NCBI
Is the placebo powerless? Update of a systematic review with 52 new randomized trials comparing placebo with no treatment. - PubMed - NCBI

 

[Sean]

No matter what claims they make, their claims do not apply to patients described by the CCC or even by the IOM report.

It is even worse than that. PACE not only used Oxford as their primary criteria, they only studied mild-moderately affected patients.

So PACE patients were the least sick from the least stringent criteria, and it was still a null result for the long-term follow-up.

Their claims don't apply to anybody. Time to put this farce to the sword.

 

[Yogi]

Another good one by MEMilitant

https://twitter.com/i/web/status/773958071588622336

 

[Kati]

https://twitter.com/i/web/status/773944315450105856

https://twitter.com/i/web/status/773945233633636352

https://twitter.com/i/web/status/773946330037256192

(Use this tweet for RT, with all the hashtags)

https://twitter.com/i/web/status/773953438287880192

https://twitter.com/i/web/status/773961851881349120

https://twitter.com/tomkindlon/status/773963428281090048

 

[Dolphin]

Additionally, they used a modified Fukuda that only required symptoms for 1 week, not 6 months, which they acknowledged might not be accurate.

No matter what claims they make, their claims do not apply to patients described by the CCC or even by the IOM report. The AHRQ Addendum found insufficient evidence of effect for CBT and GET once you eliminated Oxford studies.

Three very important points there.

They have to also supply the recovery and the 2.5 year follow-up figures according to the original published protocol.

Not sure what if anything they promised for the 2.5 year follow-up data.
Though I'm pretty sure they collected employment data but so far we haven't seen any.

 

[Dolphin]

We will see if they appeal. If they do, then they are almost certainly hoping that this controlled release will disarm critics.

It's important that people don't get carried away just yet. I think an independent analysis of PACE will be much more damning, and we should not be content with letting PACE authors interpret their own results, regardless of the protocol used.

Tuller already pointed out how they don't talk about recovery.

They also don't talk about the missing actimetry data(!), and again emphasize subjective outcomes. Their claims about adverse events are still at odds with patient experience. There are many unanswered questions still, and it's likely that every question answered will only make PACE look even worse. They're still not critically evaluating these interventions.

So continue to push for an independent analysis regardless of what happens.

Just to be clear, they didn't use actometers as outcome measures in the end (they did use them at baseline but that's not very interesting).

 

[Sidereal]

There was somewhere in one thread discussion of the possible "nocebo" effect on those entered in the SMC arm of the trial which might further reduce the alleged improvements.

For this and other reasons I don't think SMC and APT were true control conditions. You can't use subjective self-report questionnaires when you have created a situation in which you have patients in two arms of the trial being told to expect full recovery from the most effective treatments known to man (CBT/GET) while the other two arms are given a leaflet, told to do 70% of what they think they can do, effectively told to fuck off and die with an intractable condition.

 

[JohnCB]

Hardly 'latest research' as over five years late:

http://www.wolfson.qmul.ac.uk/current-projects/pace-trial#news

[snip])

OK, I must admit that I am not that clued up on the protocols for publishing research data, but isn't this rather odd in that respect. There is no mention of this being published in a scientific journal and no suggestion that it is a pre-print. Would this sort of thing not usually be submitted as at least a formal letter to a journal or a comment on an online paper? Does it count as science if it is not formally published?

Won't even The Lancet touch this now?

 

[BurnA]

Does it matter what the data says ?

PACE is valueless for one reason: the combination of lack of blinding of treatments and choice of subjective primary endpoint. Neither of these alone need be a fatal design flaw but the combination is.

 

[Simon]

Would this sort of thing not usually be submitted as at least a formal letter to a journal or a comment on an online paper? Does it count as science if it is not formally published?

Good point, I hope the authors will now publish in a journal to set the scientific record straight.