Information Commissioner's Office orders release of PACE trial data

[Valentijn]

Oh, that's good! It explains both why they wouldn't want to release the data (if they fudged it) AND why they'd be so freaked a patient would recognize their own data (which the patient should already know anyway).

They did some things which patients might not be very happy with. If the patient didn't give a self-rated Global Clinical Impression score (very much improved, much improved, minimally improved, etc), it was substituted with the score from an unblinded clinic doctor [1].

This is somewhat interesting, because in the protocol they claimed that they would deal with the lack of therapist blinding by only using self-rated or objective outcome measurements [2]. Hence 22 patients were probably not at all expecting to have a practitioner decide how much their pet therapy was helping them.

[1] PACE Recovery paper, last paragraph of page 3.
[2] PACE Protocol, 2nd paragraph under "Assessments and Procedures" on page 11

 

[Sidereal]

I find the patient reports of harms associated with GET very difficult to reconcile with the PACE narrative that GET is a very safe treatment.

The reported physical function scores are not those of mildly sick patients that might be able to do GET without risking deterioration.

So where did all the deterioration go? Did it magically disappear?

That's where the fraud lies I suspect. A career ender if it was discovered - hiding severe adverse effects is serious matter.

Yeah this issue was recently brought to the forefront in the BMJ reanalysis of the infamous Study 329 on paroxetine where the authors of the reanalysis actually went back and looked at the original patient-level reports of adverse events and recoded them into more appropriate categories than the ones reported in the original publication. They showed that the original coding by the investigators was obfuscation which underestimated the actual incidence of suicidality because it lumped it together with other far less serious adverse events into one category of "emotional lability". It is possible to hide things in plain sight in published papers.

It cannot be that there were no harms from GET because at baseline PACE patients scored about 40 on the SF-36 PF which is about what patients with MS and ALS score, and about 10 points (out of 100) above class III congestive heart failure, so they weren't mildly ill, contrary to some claims in our community. Mild as far as ME goes (vs. housebound or bedridden), sure, but not mild compared to disability seen in other diseases.

 

[Sidereal]

I suspect that patients either did activity substitution where they reduced other things they were doing or didn't really follow the advice. However, the was someone in the latest MEA survey who said the deteriorated after GET in the PACE trial.

Of course Wessely claimed that there was a high level of compliance since they monitored sessions with therapists. But they did nothing to check what patients actually did

A couple of first-hand reports of deterioration have also emerged on Facebook.

 

[worldbackwards]

If you haven't already signed the petition and spread the word, please do it now.

I'm amusing myself by thinking of our "highly orchestrated campaigns", which appear to be Sasha asking people to sign a petition and put the odd comment on an article.

 

[searcher]

I saw one person comment that she deteriorated and the therapist said it was because she "moved house." I suspect there are a lot of examples like this. If anyone knows participants who have stories about their experience in the trial, we would love to have participants anonymously share their stories at http://stories.meaction.net/submit-your-story/.

 

[SOC]

They did some things which patients might not be very happy with. If the patient didn't give a self-rated Global Clinical Impression score (very much improved, much improved, minimally improved, etc), it was substituted with the score from an unblinded clinic doctor [1].

This is somewhat interesting, because in the protocol they claimed that they would deal with the lack of therapist blinding by only using self-rated or objective outcome measurements [2]. Hence 22 patients were probably not at all expecting to have a practitioner decide how much their pet therapy was helping them.

[1] PACE Recovery paper, last paragraph of page 3.
[2] PACE Protocol, 2nd paragraph under "Assessments and Procedures" on page 11

This so-called research never ceases to amaze (and horrify) me.

 

[Sasha]

I'm amusing myself by thinking of our "highly orchestrated campaigns", which appear to be Sasha asking people to sign a petition and put the odd comment on an article.

It was me all along?

 

[Gijs]

I would like to know 1. if these patiënts have objective orthostatic intollerance, 2. how low their Vo2max were before and after GET.

 

[SOC]

It was me all along?

What, you didn't know you're the leader of an organized campaign of ME militants?

Odd isn't it, how much they know about us that we don't know about ourselves? You'd think it was them living in our bodies instead of us.

 

[Valentijn]

The reported physical function scores are not those of mildly sick patients that might be able to do GET without risking deterioration.

So where did all the deterioration go? Did it magically disappear?

Yes, it magically disappeared during the unblinding of Serious Adverse Events (SAEs), to determine if they were Serious Adverse Reactions (SARs) to the treatments. GET had 13 SAEs compared to 7 for SMC, which turned into 2 SARs for GET compared to 2 for SMC, after unblinding the assessor regarding the patient's treatment arm [1].

And a great many adverse reactions were hidden by the categorization of SAEs versus Non-Serious Adverse Events (NSAEs). An adverse event was only considered serious if it resulted in severe deterioration lasting a couple months or resulted in hospitalization. Most deterioration would have been lumped in under NSAEs, which could include basically anything from feeling a bit tired to weeks of being completely bedbound. 89-96% of participants in each trial arm reported at least 1 NSAE, with an average of about 6 per patient[2].

[1] Table 5 of http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4065570/
[2] Table 1 of http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4065570/

 

[Sean]

I suspect that patients either did activity substitution where they reduced other things they were doing or didn't really follow the advice. However, the was someone in the latest MEA survey who said the deteriorated after GET in the PACE trial.

Of course Wessely claimed that there was a high level of compliance since they monitored sessions with therapists. But they did nothing to check what patients actually did

Actometers would have gone a long way to doing that. Except of course they dropped them halfway through the trial.

 

[worldbackwards]

It was me all along?

I hear the political parties are sniffing around you for the 2020 election.

 

[Large Donner]

I just took a look at the Commissioner's report following your link, @Sasha. What is extraordinary to see is the amount of expensive legal research that went into denying this FOI claim. There are citations to case law from Australia, for heaven's sake! It must have cost the University a fortune to fund this.

Yes, what's doubly amazing about this is they had previously denied a FOI release on the grounds that the few hundreds of pounds it would cost to process the info would exceed the FOI guideline amount even though they must have had all the required data sitting right there.

 

[Kyla]

Yes, it magically disappeared during the unblinding of Serious Adverse Events (SAEs), to determine if they were Serious Adverse Reactions (SARs) to the treatments. GET had 13 SAEs compared to 7 for SMC, which turned into 2 SARs for GET compared to 2 for SMC, after unblinding the assessor regarding the patient's treatment arm [1].

And a great many adverse reactions were hidden by the categorization of SAEs versus Non-Serious Adverse Events (NSAEs). An adverse event was only considered serious if it resulted in severe deterioration lasting a couple months or resulted in hospitalization. Most deterioration would have been lumped in under NSAEs, which could include basically anything from feeling a bit tired to weeks of being completely bedbound. 89-96% of participants in each trial arm reported at least 1 NSAE, with an average of about 6 per patient[2].

[1] Table 5 of http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4065570/
[2] Table 1 of http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4065570/

Eep!

@Valentijn would be great if you posted these comments to either Coyne or Keith Laws blogpost. I don't think harms have even made it into the discussion anywhere yet. Coyne seems to be reading comments and sometimes replying.

 

[Large Donner]

I don't understand something here, why does it matter if you see your own data? Surely the only issue is if you can recognize other's data.

Perhaps they are worried that trial participants see data and start to wonder if all of their scores have been misrepresented during the actual trial.

 

[Large Donner]

I'm amusing myself by thinking of our "highly orchestrated campaigns", which appear to be Sasha asking people to sign a petition and put the odd comment on an article.

Yes and we didn't have £5 million, the DWP, the SMC, top government psych manipulators and the insurance lobby at our disposal.

Just imagine the audacity of a bunch of sick people, their doctors and advocates asking for the release of data from a study that gets published in the Lancet after peer review without the data actually appearing, all the objective measures not being used, the recovery definition changed after commencement of the trial to the level one could have been declared recovered and sick enough to enter the trial at the same time.

Never mind they used a criteria now advised by the IOM to be retired.

The 100 or so highly orchestrated people they claim to be at odds with the general PWME population turns out to be 8500 in nine days on a petition to retract the PACE trial, whilst the Uni who hold the data are so afraid to release it because they fear it could bring the Uni into disrepute.

You couldn't make this stuff up!!

 

[Sidereal]

Eep!

@Valentijn would be great if you posted these comments to either Coyne or Keith Laws blogpost. I don't think harms have even made it into the discussion anywhere yet. Coyne seems to be reading comments and sometimes replying.

Or better yet, email them.

 

[worldbackwards]

I don't think harms have even made it into the discussion anywhere yet.

This has been bothering me as well. Surely this is right at the heart of matter about the problem we have with these treatments. We need to understand why PACE is so at odds with our day to day experience on this issue. Not that I couldn't hazard a guess.

 

[Esther12]

The PACE trial was quite different to how CBT and GET are often provided. It would not surpris me if there were genuinely few harms related to CBT and GET within the PACE trial.

 

[jimells]

Dear aforementioned visitors. I hope you like our little community. Feel free to join and engage us directly to discuss any concerns you might have. Contrary to what you might have heard, we don't bite.

But our bark is quite frightening, apparently.