Publication bias has a good deal of research surrounding it , and it is indeed quite worrying, but what about publication dogma?
Dogma can generally be define as that which is taught, by some kind of authority, often irregardless of the facts at hand.
Now you’d expect that if something was ‘peer reviewed’ by scientists, that that would be analytical of the facts and not overborne with the ‘authority’ of peermanship. Well if you actually dig into journals, especially in controversial subjects such as schizophrenia, challenges to such authority are not met kindly. This is also apparent in other fields such as mathematics and science where papers have been lambasted on for reasons such as ‘not understanding set theory’ and not the actuality of the work itself, valid or not.
Indeed I have had Dr’s say that they will only accept published, peer reviewed studies (which are hard to get, and the data is nigh on impossible to get), where all that would be needed is the anecdote of another person with the same symptoms to say ‘I think I’m one of those’.
Recently I was looking up published data on diet, and every man and his dog has some kind of new fad diet they want to sell you. One of the things I wanted to look up was fat and carbohydrate metabolism, as really that’s key, especially if one aim is to increaese good cholesterol and decrease bad cholesterol and indeed know why one is good and the other is bad. There was quite an interesting paper on carbohydrate metabolism where the author stated in his summary that current advice on a low fat diet was valid and high carbohydrate made no difference. In the detail of the paper his supporting evidence for this was that: a high fat diet results in weight gain and that athaleates often have diets high in carbohydrates. Recommending having a low fat diet is not the same as not recommending a high fat diet!!!! Reading further down the paper there where a number of low GI diets presented, which where all more successfull than the government recomended diets and indeed in one case the government recommended diet, ad-lib (not calorie controlled) resulted in people gaining weight where as in the low GI group the people lost weight. The most notable things about all of those ‘low GI’ diets was that the proportion of fat in relation to carbohrdrate was such that people where eating a lot more fat and a lot less carbohydrate. I think this is a good example of dogma, where the actually facts in the paper don’t match the ‘dogmatic’ summary given as the results.
Perhaps more telling is when an authority on a subject tries to assert their authority:
I’ve been looking up details on thyroid dissfunction and wanted to know how accurate the test actually are, since I found some people challenging them. Really what I wanted to know where the ‘normal’ and ‘abnormal’ ranges, though I didn’t expect I’d actually get any real data on it and it turns out that tests are only asserted on the 95% ‘normal’ range not on the ‘abnormal’ range which could have a much greater overlap (it could also have no overlap at-all if the data is multi-modal as all good indicative data should be, making the standard 95% test being pretty meaningless).