Search results

I thought there research was one of the few things that was replicated.

I was thinking the same I'm sure I read something by him in the early 90s but no idea what.

Although I would agree it does make the point that statistical significance doesn't necessarily mean a difference that a patient would notice. Hence statisticians need to be aware of the issue.

It is also a paper that is widely read at universities so perhaps stories attacking poor quality academic research don't go well with the readership.

That address bounced it seems to be observer.readers@observer.co.uk

I assume they will have an ethics sub committee (of 2) turn it into a full trial and merge the results in together or is this something that is only done in Bristol?

I thought this was a shocking comment What they are basically admitting to is very late publication (and possibly design) of the protocol duing the collection of follow up data but making it seem like it was submitted prior to the trial starting. They claim that external reviews looked at it...

But they may see it as keeping in with the establishment in the UK and that as important to their credibility. I often think that those pushing for 'better evidence are really anti industry rather than pro better methodology.

Cochrane has been their last line of defense for sometime so to see it challenged and even Tovey (who would normally back them) being concerned will be worrying to them. What else have they left - the continued support of the Lancet?

The Guardian have a long history of censoring comments made on ME articles that are critical of PACE.

Given the emails around this are public it could be very embarrassing for Cochrane to fail to deal with the issues. Although it would surprise me if the new editor will allow it to pass.

Interesting that the journalist seems to mainly cover politics and sport.

If people find it easy to become and remain thin doesn't this reduce the value of such activities? I wondered if the metabolic factors made it easier for someone to function at a very low weight and for the body not to give vast 'eat now' signals? Hence the genetic factors may be necessary to...

I think in general the Guardian has deteriorated in the last few years - very few articles there these days seem to have any degree of analysis.

So if a medical professional reads information from a trusted source such as Cochrane or a 'good' textbook or a paper in a good journal how much effort should they put into checking it? I think that the publishers need to be held to account for the information they give especially where there...

I put no and explained why the specialist services have no value as there are no treatments but the services push them despite the lack of evidence.

But does the meme that is generated and hence the diagnoses that people are given (by the muddle-headed-physicians) mean that they do need to be considered together - from the perspective that some people will have been given those diagnoses.

The question should be around sampling methods. If for example, you were to sample people as they came out of the supermarket then that would introduce a huge bias as those with fatigue are unlikely to spend the additional time to talk to someone. So how they decide on getting a representative...

I feel quite shocked that someone could write this and an editor could publish it. As a paper it seems completely unaware of the literature in the area. It makes vague unjustified statements such as around safety (blaming GET that is badly done but with no evidence and there is now evidence to...

I've only skimmed the paper but this worried me: When using ML you typically split the data into training, validation and test sets and quote results on the test set not on the entire data. I can't see any information about the relative splits of the training and test data which makes me...