• Welcome to Phoenix Rising!

    Created in 2008, Phoenix Rising is the largest and oldest forum dedicated to furthering the understanding of and finding treatments for complex chronic illnesses such as chronic fatigue syndrome (ME/CFS), fibromyalgia (FM), long COVID, postural orthostatic tachycardia syndrome (POTS), mast cell activation syndrome (MCAS), and allied diseases.

    To become a member, simply click the Register button at the top right.

(Not that important) Paper lists (& explains) 27 types of publication/reporting bias

Dolphin

Senior Member
Messages
17,567
Many people will be aware of the concept of publication bias.
At its most basic level, this is where results are published if they suit the authors and/or funders and not published if they do not [suit the authors/funders].

However there are in fact lots of variations on this theme, and in the methods that could lead to people not coming to the right conclusion because of the way the data was presented (or not presented as the case might be).

I happened to come across of the following paper which looks at this issue in depth. Fortunately one does not need to read a lot of the paper to read the bit I'm highlighting.

http://bit.ly/eigZRY i.e. http://onlinelibrary.wiley.com/o/cochrane/clsysrev/articles/CD008965/pdf_fs.html

Citation: Jefferson T, Jones MA, Doshi P, Del Mar CB, Heneghan CJ, Hama R, Thompson MJ. Neuraminidase inhibitors for preventing and treating influenza in healthy adults and children - a review of unpublished data. Cochrane Database of Systematic Reviews 2011, Issue 1. Art. No.: CD008965. DOI: 10.1002/14651858.CD008965.

The basic premise for this review is that the authors have become aware that not all the data has been published and they are investigating the issue through various means, checking for different types of bias.

In tables 1 and 2, it tries to explain the different types of bias that the authors are checking for.
There are 27 types in total.
They are explained in fairly basic English I think in the sense that one does not need to know biology to understand the issues being raised.
Unfortunately I think they could probably have given slightly more detailed descriptions/explanations in some cases. However there are papers they link to which give more detail for anyone interested.

Generally, publication bias is associated with pharmaceutical companies.

However is no reason why there may not be other types of bias.

I think it would be interesting to look at the issue in relation to some nonpharmacological interventions for ME/CFS.
 
Messages
13,774
Ta. I mentioned an anti-depressants paper to you which was on a similar theme. Apparently the FDA requires pharmaceutical companies to log trials with them, and these logs can then be compared to the data that eventually gets published. I don't think there's any similar system for non-pharmacological work. I'd assume the same affects would occur, but I don't think there's be any evidence of this (the perfect crime).
 

Dolphin

Senior Member
Messages
17,567
Ta. I mentioned an anti-depressants paper to you which was on a similar theme. Apparently the FDA requires pharmaceutical companies to log trials with them, and these logs can then be compared to the data that eventually gets published. I don't think there's any similar system for non-pharmacological work. I'd assume the same affects would occur, but I don't think there's be any evidence of this (the perfect crime).
Yes, what your describe is what I was thinking about and the sort of area the paper discussed (it is looking at unpublished data from other sources like you mention - they are looking at all sorts of data and sources to try to piece it all together).
They may not be formal trials of CBT or GET that was not published but data where they followed consecutive people who attended a clinic say. Also long-term follow up from trials. And biases in what got presented and what didn't. The paper gives 27 methods - I didn't think too much how it might relate to ME/CFS but it seems quite possible. It can be handy to have the jargon and something to reference when complaining about some problem one thinks either is happening or might happen/be happening/have happened.
 

Dolphin

Senior Member
Messages
17,567
Article is free; anyone have problems accessing it?

Somebody sent me this:
this is "pay for" access; do you have a link that is free?
Anyone else have problems? Please try both links as the long link didn't work for me there but the bit.ly link did.
If people have problems with both links I'll get another link.
 

Dolphin

Senior Member
Messages
17,567
Or put 10.1002/14651858.CD008965 into your browser or search engine - the top result gives it to me.
 
Messages
13,774
Somebody told me they had problems with both links.

How about:
http://bit.ly/grbMlJ i.e.
http://onlinelibrary.wiley.com/o/cochrane/clsysrev/articles/CD008965/frame.html and click on pdf along the left hand side.

This is the only one I tried, but it opened fine. Ta. (I've not read it yet though... taking some time off from this sort of reading!)

re "They may not be formal trials of CBT or GET that was not published but data where they followed consecutive people who attended a clinic say. Also long-term follow up from trials."

It does seem amazing that we don't have access to this sort of fata. But these sort of pharmaceutical reviews are only possible because (I think) researchers are now required to log their work before they get their results or publish them. I doubt there's any equivalent widespread and systematic mechanism for psychological interventions, so these treatments are given the benefit of the doubt even though the same sort of biases are likely to be occuring. I wonder if this sort data collection is something we could work towards achieving? - It sounds like a reasonable proposal to me. It's not really a specifically CFS issue though, so I don't know what organisation would be likely to take it up.