Bad Medical Science

Medical science is a mess as is illustrated by this article. PR

“The case against science is straightforward: much of the scientific literature, perhaps half, may simply be untrue. Afflicted by studies with small sample sizes, tiny effects, invalid exploratory analyses, and flagrant conflicts of interest, together with an obsession for pursuing fashionable trends of dubious importance, science has taken a turn towards darkness.”

The link to Dr. Horton's PDF (Lancet).

There are some other interesting links.

Last edited by

26 Replies

  • PR, I didn't realise it was so prevalent. I hope someone acts soon and starts the clean up.

  • Clutter, this is another article, same place. PR

  • Like the original articles. Worried that sound bites get bandied around and serve to undermine the 50% that's done well. This has to be about solutions not about pointing out the (pretty obvious) problems with entrenched systems. Medicine and biomedicine works in stats which are interpretations that can be skewed depending on what's left in and taken out. Much like hitting article...until you read his suggested solutions it's just a very good description of a big problem!!

  • Thank you for this. It is clearly true, and it is pernicious because most of us have so little scientific knowledge that we believe scientists almost as mediaeval peasants believed the Catholic priests who sold indulgences. There are very few journalists challenging scientific authority and asking awkward questions as they do of politicians. We should feel free to challenge doctors - politely - on whether they are practising evidence-based medicine, just as we are encouraged to ask them in the ward whether they have washed their hands. I think the internet means that a slow Reformation is beginning - giving the punter direct access to research is equivalent to translating the Bible into English.

    Research and practice on the thyroid seems especially poor - I read that the "normal" TSH level was set in the early 1970s by testing 200 people, and I read that no study has yet been done to demonstrate whether this level correlates with an absence of symptoms associated with thyroid illnesses. There are clearly some major flaws in the studies which show no imporovement with the addition of T3 to the treatment of hypothyroids whose symptoms do not resolve on Levo. What are they?

    What do the highly-informed sufferers on this forum think are the major, basic errors or omissions in the thyroid research they have read?

  • My late husband was a research scientist, I worked in a University department that was research orientated and I often did the illustrations for these papers and for hospital based writers as well. To me it makes sense that not every piece of research works out how you wish it to be and in fact things that don't are often equally important in that something positive can still come out of that as it allows them to rethink or even take their research in a different direction. Some of the most important discoveries where probably found by accident. Yes it is true that you need to be producing scientific papers to gain reward or more funding but that doesn't mean it's necessarily flawed. Maybe even the opposite. A scientific paper has to be discussed before your peers, not just your colleagues but could even be worldwide! my husband worked not only in the UK but with others in South Africa, USA, Netherlands, Ireland and probably others I never knew about. So yes they swap ideas and conversely discuss at great length. The article pinpoints research in Medical Journals but getting something in there is a long way down the line and yes the initial trials can be just a small number to start with-it's a big leap from the lab rat to your loved ones so it has to start with small samples as it might not work or could even be harmful. This is where I suspect it could be rushed. Big Pharma often fund the research in many ways, expensive equipment, running costs etc but then I imagine the race to produce the magic pill etc and claw the money back takes over. The doctors haven't a clue because the drug companies come along and spin them a yarn and this is where I feel the system breaks down, where the fast forward button is pressed! But look at the other side of the card and it happened with Ebola, relatives pressed for untested drugs and methods because there was nothing else and if we are being very honest wouldn't we do the same thing? It's the human trials I feel that are found wanting but it doesn't necessary mean the original science is wrong but may not be right for humans.

  • Thyroid science is chock full of non-science.

    Even if an original paper were not guilty of any of the factors that have been criticised above, the translation from a scientific study to clinical use either does not occur at all, or the science is pummelled into submission at the hands of everyone involved. Many of those have their vested interests, desire to cover up ignorance, or a need to simplify to the point of misinformation. Or costs get in the way.

    When I started levothyroxine, I was already aware of one of the papers suggesting that taking it at bed-time was a reasonable alternative. I went with that and still do so. Years later, a few people occasionally get told that is an option. Yet that is about the only positive change in thyroid treatment for the masses that I can think of.

    I wonder what percentage of even the good papers ever reach clinical use?

  • Agree with that. My husband worked very much in the thyroid field. His Ph.D was on TRH and I was horrified to hear that although the contraceptive pill worked they didn't really know why. He told me things about thyroid treatment and its flaws that I'm only hearing about recently. The most important being that the TSH is not something to diagnose with. I heard that over 40 years ago!

  • We could probably count on one hand the number of people here who have ever had TRH measured! (Might be my bad memory, but it certainly isn't commonly done.)

    It looks to me as if some level of compromise in the hypothalamus/TRH and pituitary/TSH mechanisms (including nervous system connections) could be widespread. Quite likely not severe enough to be readily identified and classified as central (or secondary/tertiary) hypothyroidism. But enough to limit the rise in TSH and hence ensure non-diagnosis. (Quite possibly, while the thyroid is functioning well, these people have little or no problem - it might work up to that point.)

  • I think there is still so much we either don't know or don't understand but to me that means we should be more open to suggestions etc. more trials would be a good start! It worries me that more of us feel forced to self medicate. It they were to ask for volunteers for properly conducted trials they would be inundated.

  • I hope I don't and never will even at my advanced age, fall into this category. There are two overwhelming rules that any scientist in any discipline must follow: 1) ask a plausible but untested question, design experiments properly to test it, test it to as near destruction as you can, and abandon it without hesitation if the results don't fit, 2) if you get something wrong and publish it, retract publicly. Sometimes, as in my case as a young scientist I was led astray (luckily not in an important thing) by a diabolically unlikely set of circumstances. Fortunately, I was cautious enough in the published papers to make statements that were true, but didn't really define what I thought more rashly was happening. In short I got away with it. But it is not something to be proud of. And unfortunately lots of researchers simply try to retrieve an impossible situation by refusing to admit their errors. In such a way nonsense gets embedded in people's thinking and it takes a heck of a lot of time to remove it.

  • My prof used to have a saying on his notice board. The man who never made a mistake never name anything. As you say you have to retract. The reason I said no research is useless and finding something is wrong is just as important as finding something is right. To me saying publically don't do this because ..... Is just as important as say this particular thing showed ..... as to help research move on.

  • I don't feel qualified to add anything substantive to the other comments here, but could I direct anyone who is interested in this general issue to read Dr Malcolm Kendrick's new book, Doctoring Data? It is available via his blog,

  • Great book - great blogs.....

  • I agree. Anybody who is interested in this subject should read his book.

  • One other thing today. Matt Ridley in the Times has picked up on the cholesterol/fat/salt story coming out of the States. And Dr Kendrick has been quoted approvingly. But Ridley has stated that as a long-time heretic, being right won't endear him to the establishment (esp in the UK). I think the long-term mistakes in thyroid diagnosis/treatment and the unhappiness caused to so many needs equal airing. Anyone inclined to contact him? Might advance your cause.

  • I think he's already on the case!

  • I think that diogenes meant to contact Matt Ridley.

  • I landed a job as a manager in a major general hospital in the 80s. It was my dream job.


    Firstly the NHS then, and even more now was riddled with politics and self interest. As a general rule consultants spent more time fighting their corner (about things like the size of their office, as well as more important stuff like budget) than they did caring for patients

    Secondly, in common with most hospitals we were involved in lots of studies, they brought money and interest in what is mostly a dull routine job where consequences of mistakes can be horrifying. I was apalleded to realise that doctors are NOT scientists. They have the scantiest understanding of the mechanisms in the body, almost no understanding of biochemistry, and absolutely no training in statistics. What doctors are good at is learning lots and lots of facts - and trotting them out to sound as if they understand what they are saying. It was seriously scary, and I haven't trusted a doctor since!

    This is how we end up with endos who think its fine to rely on a statistical construct ('normal' which just means no worse than 95% of the population) and ignore the clinical signs. Then of course they are bored, because they don't get any interesting cases!

  • The sentence sentence that that really jumped out at me from from this is "doctors are not scientists". I was shocked by this, too, when I became ill.

    They are looking at our test results as part of a flow chart, not as a package of evidence that can be used test theories about what might be wrong with us.

  • Thanks to the alternative media, this is becoming more widely recognised. Consequently more people are taking responsibility for their own health, whilst declining Fluoride, Statins, Antidepressants etc and choosing superior, natural metabolic alternatives.


  • My only comment is: relative versus absolute risk - lies, damn lies and statistics.

  • Firstly doctors these days are in it for the money, secondly the big pharmaceutical companies get away with murder (literally) they decide what tests and trials are going to find all in the glory of greater profit! Your GP might make sympathetic noises then say his (or her) hands are tied, this is blatantly not true and is down to cost not consideration to the patient.

  • Medical research relies on statistical analysis to validate empirical evidence. Experiments are expensive to set up and often for practicality the samples to validate a specific hypotheses are relatively small. Add to that some confounding factors present in the data or worse yet left out of the analysis; therefore a medical scientist may arrive to a conclusion that not necessarily captures they real benefit of a treatment over a placebo. Metrics such as confidence intervals, probability values, signs of the statistical parameters guide the scientist in deciding whether the research results are statistically robust. However, it would be nice if the data and additional exploratory analysis were available to other researchers to check and replicate the findings.

    On the implementation side of the validated hypotheses there are other issues that have been addressed by others already. That is the political and economic interests of the pharmaceutical industry (HCV $1000 pill comes to mind), the publish or perish mentality at university medical centers or the grants hunting to support institutions in an environment of increasing cuts to research. On top of this if we add the complexity and uniqueness of each human being and the difficulties in understanding the medical jargon and body reactions, thus it should not be a surprise that we have been increasingly skeptical about some of the treatments for our medical ailments. The only hope is that there are enough researchers honest in their work and truly concerned about human health and there are otherwise we could be in a much worse situation. But we need to educate ourselves and challenge medical practitioners because our safety is at risk. They may not like it but the outcome for us could be deadly.

  • Just to add, please sign the AllTrials Petition which will oblige all trials to be published, not just the ones that fit the desired outcome. There is considerable pressure to do this, and hopefully it will start to erode this sort of thing.

  • My background is in social research. I think a big problem with medical research is that there is no philosophy of science is included in the process. Nobody actually asks "Is it safe to draw this conclusion from this data? Are we measuring what we think we're measuring?". The latter is really crucial, otherwise you get problems like "a normal TSH", being treated exactly as if it means "having no hypothyroid symptoms", and other big slippages like that. Ironically, before I got ilI was working attached to a clinical trials unit, and seeing how clinical trials trials are actually carried out persuaded me that they are are an incredibly brittle instrument for finding out anything - in so many ways they can be disrupted so that that what comes out the end is hogwash, and commercial trials are well known inside the industry to be completely underhand and sneaky. They design their studies based on the strengths and weaknesses of the drug they want to peddle, and have no transparency or accountability. Nobody even knows how they recruit patients.

    But now that I've been ill and and am looking at at the relationship between research and and what happens in the doctor's surgery it's even more shocking. That channel is even more flawed, because enormous amounts of research do not make it into practise.

  • Percentages are often used to make tiny effects look huge.. eg 1.0 to 2.0 result translated to 100% improvement. Lots of info on how statins are promoted as beneficial in this way ignoring the links to alzheimers etc. We have to question things and invest in taking time to educate ourselves.. especially with thyroid issues it seems.

    We have to realise that pharma is an industry maximising profit and not altruistic, and that doctors get their continuing education from the drug companies and don't have time to look into everything however well meaning and altruistic they intend to be.

You may also like...