Biomedical articles on MEpedia - purposes and pitfalls

I think that MEpedia's position as a patient driven resource could justify extra caution with the claims being made - some people are likely to be sceptical of it anyway, and any inaccuracies or exaggerated claims are likely to be viewed less generously than they would be elsewhere. Hopefully those involved in editing pages will be open to having others point out that the evidence is not as strong as they claim, or even contradicts some of their claims. Going through and discussing these things can be a lot of work (thanks to all those contributing) but that seems to be the way with these sorts of projects.

For a lot of things I'm already finding MEpedia to be a really valuable collection of links and resources, but I'm sure that there are also going to be a lot of rough patches. All these wikis end up with some nonsense in them, and it's always worth checking the sources that they cite before assuming that anything is true.

 

I realise that I am several years late in asking these questions but:

1. What is the advantage of having a separate MEpedia site rather that adding the same information to Wikipedia where it would potentially get more editors and readers? (I don’t mean this as a criticism, I’m just interested to understand.)

2. Is MEpedia content copied onto Wikipedia? If not, should it be?

FWIW, I am a big fan of Wikipedia, and I have also found MEpedia exceedingly useful – both are particularly useful for finding primary sources. If I had more capacity I would help edit MEpedia (as I do from time to time on Wikipedia) and I hope to be able to contribute in future. I am confident that its accuracy will improve over time with more contributors.

The biggest problem on Wikipedia seems to be when PR companies or other bodies take over certain pages in which they have vested interests. I found this when trying to edit a page about Daniel Tammet (author of Born on a Blue day) after reading Joshua Foa’s (excellent) Moonwalking with Einstein. Documented facts about his life which did not fit comfortably with his narrative/image appeared to be systematically deleted. In the end I gave up.

Sorry, I’ve drifted off topic a bit, but my point is that Wikipedia is excellent for most subjects but can be very unreliable for certain topics. The challenge for MEpedia will be to determine where its strengths and weaknesses are and to adapt accordingly. I am confident it will – assuming there is a good answer to question 1 above.

[Edit - typos]

 

Exactly, this is the dilemma. Nothing has been replicated many times over with ME/CFS. I don't see a problem with referencing to smaller-scale studies as long as the studies were peer reviewed and not fraudulent. A wiki page will anyhow always mostly serve function as a reference to other articles, not as an independent piece of work.

AFAIK the ME/CFS wikipedia page is very tightly moderated and under the same principle that only large-scale and "best" evidence is worth mentioning, so what it will contain under treatment is mainly CBT/GET and that's the end of story. I reckon it has gotten a little better over the years, but there is no chance MEpedia content can be brought under the official wiki page.

 

• Quality: You can only cite meta-studies on Wikipedia. You can't cite individual research articles. We have very few meta studies in this field.
• Notability: The vast majority of individuals (patients, researchers, clinicians) on MEpedia could not have pages on Wikipedia. They wouldn't meet the notability requirement. I think it would only be me, Ian Lipkin, Ron Davis and Laura Hillenbrand +/-. This is also true of books and events that may be notable in our community but are not notable in a broader, public sense. I'm more focused on science, but I know these pages mean a lot and are useful to people.
• Focus: Every general page on MEpedia (e.g., "cardiovascular system," "acteylcholine") is expressly written with this disease in mind and fairly narrowly focused on ME research, although other research is also incorporated. A lot of this content would get cut from Wikipedia for being off-topic and too specific.
• Specificity: A lot of the content on MEpedia is much too specific. I'm not sure a PACE trial page would survive on Wikipedia. We certainly couldn't have individual pages for diagnostic criteria, etc. All of this would likely get moved into the general article and extensively pared down.

I'm pretty sure our license is the same as Wikipedia (share and share alike) so if you want to copy content from MEpedia to Wikipedia, you can of course try. A lot of it would probably get deleted for the reasons mentioned above.

Thanks! And I agree.

 

Thats reasonable

 

I think there are important differences. For Wikipedia there are dedicated moderators who systematically filter material. They are not always the right people but most are OK. Also for Wikipedia every expert in a field is likely to feel a duty to try to help get balance. I have rewritten four or five large Wikipedia pages and taken many hours over it because they deal with my own specific areas of expertise. I think it is unlikely that anyone with the relevant expertise will do that for MEpedia. Wikipedia does run into trouble with contentious topics and the problem for ME is that almost everything is contentious, except the symptoms and epidemiology.

 

Absolutely. None of these theories should be presented on MEpedia as having any firm basis, or even written about as if they were quite likely to turn out right and were supported by encouraging data.

 

Suggestion : Not just size of study, but also number of male subjects and number of female subjects (where the subjects are male or female - this obviously doesn't apply to cells!)

 

Sorry, but I don't think that is true, @Alvin. I said nothing about conclusively proven of course but I am talking about findings that are replicable and have a very plausible interpretation in terms of what is going on. Lack of replication does not hold up medical science. In the cases where new findings over the last thirty years have proven reliably replicable, maybe half a dozen major examples a year, massive replication has usually occurred within weeks or at most months. When a new autoantibody pattern was found for Wegener's granulomatosis or Helicobacter was identified in stomach ulcers routine labs in university hospital throughout the world repeated the test as soon as they had read the paper and got the reagents and found the same thing. So in the great majority of cases everyone can talk about cast iron results by Christmas. There simply is no problem for findings that really stand up.

The reason why findings in ME have not been replicated is mostly that they are not replicable or that other scientists have not thought the original report was going to lead to anything significant enough to be worth replicating.

 

This is in the back of my mind too. At a recent meeting of 'top people' in ME research in London at which three representatives of major government funding agencies were present it was interesting to read the body language. A number of people present enthused about the 'multi-system disease with 9000 papers showing biological abnormalities' approach. The UK funding representatives had polite smiles but I strongly suspected that the enthusiasts were digging a hole for themselves. The meeting achieved a lot simply in terms of networking and a shared appreciation of just how urgent a problem ME is, but I don't think the sales pitch was ideal.

 

But this isn't how Wikis work. It's an open publishing platform. We can't stop people from creating new pages, nor should we.

All of these conversations were had 13-15 years ago re: Wikipedia.

 

You're comparing a 17 year-old project with a 2 year-old one (written largely by 3 people) that has barely begun. Wikipedia didn't start this way and there is no reason why we couldn't eventually have all of these elements. And I don't think it's at all unlikely that people with relevant expertise will eventually contribute.

 

I have read a different story about helicobacter, according to which Marshall et al met skepticism and lack of interest at first because everyone knew stress was the cause of stomach ulcers.

Maybe everyone knows that ME is just a psychological disorder, as proven by exercise and psychotherapy being able to reverse it (tongue in cheek).

I don't think this attitude could prevent progress or replication forever but it could delay it quite a bit.

 

I agree, but we need to be careful not only to weed out the bad but to make sure we don't throw out the good.

There is an old axiom, build it and they will come, if we don't try because of what might happen we won't accomplish our potential.

I partially agree, though the replication crisis shows what happens when things are not replicated and weren't, those things held up fine till the replication was finally attempted recently.
I would love to see everything replicated in triplicate by independent teams, we don't have the money or resources or researchers. If we did then i support this fully. In fact in all of medicine i would like to see money specifically put towards replication. Medical research could even have teams who specialize only in replication. They would likely be more objective since they have little skin in a particular outcome.

But ME/CFS is a niche area, as i just said our resources are minimal and this is not happening for our disease. I would love for this to change.

I suspect the more likely reason is that there are so few researchers working on ME/CFS to do the replication and derivative research. Chicken and the egg, if we say the research is thin so we can't come to conclusions or talk about the results we do have or build on it then it will stay thin and unbuilt on. We need to break this cycle, get more people and more money on our disease. This is likely one of the many goals of MEpedia. I seem to recall @JenB saying she has been contacted by researchers who read technical info on MEpedia.
Should we put disclaimers where needed, you better believe we should. As an aside i personally do not think enteroviruses have much to do with our condition, beyond perhaps a trigger, probably one of many. That said we should look at what we have and we should look at where we want to go next. If the link turns out to be minimal then we should find that out, if it turns out to be a key piece, we benefit from proving that.
I am often reminded of the streetlight effect, i wish we could avoid it.

Well said.

 

Yes we do. Replication costs nothing much because all the preliminary work has already been done by the people producing the entitle report. I think this is a myth. The Japanese PET study is a bit different because the reagents are so expensive but most other findings for ME could be repeated very easily. Most findings have in fact had an attempt at replication which failed.

I absolutely agree that more money is needed for research. But the reason I am raising these cautions is that the scientists I know who would be good to recruit are not going to be encouraged by a wiki that gets things wrong half the time. They will be put off.

 

Then we would have no replication crisis, especially in psychology where replication is even cheaper...
I believe there is funding being given to replicate Mark Davis' results, if we could do that for free lets redeploy the money.

Ok so where replication has failed then we definitely need to include that.

 

What the psychologists call a replication crisis is the fact that every time they try to replicate something it doesn't. That is just an indication of how lousy psychology research is, nothing to do with ME!

Chris Ponting has some money to replicate mark Davis's T cell work but to be honest it is a tiny sum that could probably have been raised without a formal grant. It is good that it has been through peer review and flagged up at committees. But the money is tiny - maybe 1% of what the NIH has recently doled out.

 

This is not the only field lacking in replication. I would like to see large scale hands removed replication

I agree in the grand scheme (though far more then most of our annual incomes) its small but you did state "Replication costs nothing much"
So replication teams that only do replication should also be very cheap, i can live with that, more incentive to support their creation.