We doctors aren’t very evidence-based (we just think we are)

I went to a meeting for journalists and broadcasters on Tuesday night, organised by the Guild of Health Writers, dedicated to the subject of dementia. One of the speakers there was Jerome Burne – a London-based health writer of some repute (he won an award from the Medical Journalists Association in January). Jerome in the co-author of a book entitled ’10 Secrets to Health Ageing’ which is to be published in April in the UK. Jerome talked a lot about the general inadequacies of the research in the area, but also highlighted one study which I didn’t know about and found interesting.

The study concerns an amino acid called homocysteine, which previous studies has linked with an increased risk of dementia. Then some researchers in Oxford decided to see if lowering homocysteine levels with B-vitamins might help individuals with ‘mild cognitive impairment’. The results of this study were published in 2010 [1]. Compared to placebo, those taking B-vitamins saw reduced ‘brain atrophy’ (essentially, shrinking degeneration of the brain).

The same research group followed this study up with an analysis revealing improved brain functioning in those who had taken the B-vitamins [2].

Jerome made the point that the first study was largely ignored, and derided in some quarters. He stressed the need for the work to be repeated, but also gave the impression that the work is of potential significance, and I agree.

After the talks, a couple of people took exception to Jerome’s highlighting of the brain atrophy study. One of them was general practitioner Dr Tom Smith, author of the book ‘Reducing your risk of dementia’. Dr Smith said he found it ‘disappointing’ that Jerome highlighted homocysteine as a potential factor in dementia. He asked for there to be research into whether homocysteine and dementia are linked in populations. Actually, there already is such research, and Jerome had already referred to it briefly in his talk.

Dr Smith then went on to highlight the findings of his own research, which has identified two things linked to reduced risk of dementia: light exposure and playing in an orchestra. Yet, these lines of evidence are ‘epidemiological’ in nature, and can only tell us these factors are associated with dementia risk. As evidence goes, they are weak compared to the clinical evidence presented by Jerome earlier (by the way, Jerome was allowed no right of reply to Dr Smith as time ran short due to Dr Smith’s verbosity).

Earlier on, Jerome had made the comment that he believes in evidence-based approaches, but he also wishes doctors would actually do more of it. He’s right about this: doctors often talk about being ‘evidence-based’ in their practice, but the fact remains that much of what we do as doctors is not evidence-based at all.

When Dr Smith got up, he proclaimed to be ‘evidence-based’ in his work. The irony is, of course, that it seems Dr Smith does not appear to actually understand the relevance of different types of evidence at all.

Jerome, during his talk, gave the impression that one of the reason the homocysteine/B-vitamin/brain atrophy research did not get the attention it deserved and has not been repeated is because, well, there’s not much commercial potential here. After all, B-vitamins are non-patentable. In comparison, millions have been spent on drugs that don’t work. While this view may appear cynical, I share it. It also occurs to me, like it does I think to Jerome, that some will ignore or deride research simply on the basis that it is perceived a ‘natural’. As Dr Smith appears to demonstrate, these opinions on not always on the firmest of scientific footings, and appear to be borne more out of bias than a quest for the truth.


1. Smith AD, et al. Homocysteine-lowering by B vitamins slows the rate of accelerated brain atrophy in mild cognitive impairment: a randomized controlled trial. PLoS One 2010;5(9):e12244.

2. de Jager CA, et al. Cognitive and clinical outcomes of homocysteine-lowering B-vitamin treatment in mild cognitive impairment: a randomized controlled trial. Int J Geriatr Psychiatry. 2011 Jul 21 [Epub ahead of print]

14 Responses to We doctors aren’t very evidence-based (we just think we are)

  1. tess 23 February 2012 at 8:51 pm #

    you’ve made a great point with this posting — hope it gets the recognition it deserves!

  2. Lara 24 February 2012 at 12:16 pm #

    I am glad you went to Jerome’s talk. He has been very supportive of evidence-based nutritional approaches in the healthcare arena. Dr Smith’s reaction is not an isolated one. I get similar responses in the hospital. In the nutrition area of course we ‘Nutritional Therapists’ are up against a lot of current media criticism led predominantly by dieticians claiming we are not evidence-based but they are (as their BDA website states). A lot has to happen to get best research and best practice in place, but the research you highlight is taught at universities of BSc(Hons) and MSc degrees of Nutritional Therapists and it strikes me how much of that over here is still largely ignored. Last years ‘Good Research -Good Practice’ conference showed the increasing efforts to move on to find solutions to the RCT ‘gold standard’ dilemma when dealing with multivariate approaches: http://bit.ly/xKcmpd. On the back of Jerome’s work and your highlights this is very much what the Brain Bio Centre in London looks at and leads on their own research in these areas… worth having a look. Thanks for posting this – great news.

  3. Jenny 24 February 2012 at 3:06 pm #

    Well, well well – that explains it all. My Mum, now a very frail 94, has had Alzheimers for almost 20 years. She was an early Nature Cure follower, always interested in diet, grew her own vegetables, read a lot, was a very good pianist – but she didn’t play in an orchestra. I’ve been a professional oboist – played in many orchestras – I hope that’s my ticket to good mental health in my declining years . . .!

  4. Linda 24 February 2012 at 6:00 pm #

    Very interesting and although I’ve not had any experience with doctors’ lack of ability to be more evidence-based when it comes to dementia, I have recently had the experience with their lack of evidence-based abilities in regards to my son’s ongoing ear infection which all doctors seem to think repeated courses of antibiotics will solve (because they’ve read that it would…!?!?) which of course it doesn’t.

    I’m glad there are doctors like you John who are able to see thorough all the rubbish that other doctors do and say and voice a professional opinion that although contrary to most doctors’ opinion, in my opinion is much needed and seems desperately lacking amongst the majority of the medical profession these days. Thank you for your contribution to making us all a little more aware (and healthy) one post at a time!

  5. Monika 24 February 2012 at 7:43 pm #

    Thanks for posting. As a Nutritional Therapy student we are taught about homocysteine and its relevance in inflammatory conditions. The Homocysteine Revolution written in 1997 (!) by Harvard Pathologist Kilmer McCully discusses how he discovered the associations between high homocysteine levels and heart disease and is required introductory background reading on our course.

  6. sooperb 24 February 2012 at 9:18 pm #

    An interesting subject homocysteine. Patrick Holford pulled together some interesting research on the subject of raised homocysteine being a good predictor of heart attacks. Funnily enough in 2009 he posted a page on the web devoted to this which includes a short response on the subject from the same Dr. Smith.

  7. DanC 25 February 2012 at 4:31 am #

    Dr Andrew Weil (Harvard Univ) blew the whistle on homocysteine 15 years ago.

  8. Chris 25 February 2012 at 11:57 pm #

    I think you are right to be cynical about the reasons potentially helpful interventions but ‘low-input solutions’ (like addressing B vitamin deficiencies arising from diets that are suboptimal in B vitamin content) receive lukewarm receptions. Such cynicism is sufficiently well founded to be elevated to the status of ‘scepticism’ – and ‘healthy scepticism’ at that. In other disciplines of science Parkinson and Langley reported in the SGR report that enthusiasm for the profitably expedient solutions triumphs over enthusiasm for ‘low-input solutions’.

    Proponents of the homocysteine theory in re. heart disease were undoubtedly cold-shouldered by the herd. This is perverse and repressive because in the lipid hypothesis the herd willingly subscribes to a hypothesis that presents more holes than a colander. Rather than being founded upon evidence it was founded upon a big fat mistake (misidentification of cholesterol for the deleterious effects of oxidised cholesterol) followed by the further insult of a big fat lie (the presentation of highly selective and misleading ‘evidence’.) Homocysteine threatens to knock the wheels of the cholesterol bandwagon while the bandwagon is party to income streams that go on to contribute to research funding.

    However, I have a potential spanner to cast the way of the works in relation to dementia.

    Is it known or mooted by the proponents of the homocysteine theory of dementia how homocysteine would be conveyed to the brain, then once there is it known or mooted what further interactions explain homocysteine’s part in the advance of dementia? Is it possible the association of homocysteine with atrophy or impairment is indicating B vitamin deficiency, perhaps with relative certainty, while not necessarily indicating certainty that homocysteine has any active involvement? High carb diets are deficient in sources of B vitamins so is it possible high carb diets could be indicated in cause and consequence and could dementia warrant being branded ‘type-3 diabetes’?

  9. ed 26 February 2012 at 1:23 am #

    More information on this topic from a reliable source

  10. Lindy 28 February 2012 at 8:00 pm #

    There are ‘thinking and questioning’ doctors and there are doctors that just ‘go with the flow’ . I know which I prefer!

  11. Gemma 3 March 2012 at 11:36 pm #

    I think that doctors are supposed to be unbiased when it comes to every scientific thoery or study being conducted, but they’re human, and they side with what they think is right, usually based on evidence presented to them, but then if that evidence is wrong, it takes a lot of persuasion to change their mind. That could well be the case. Look at the arguments about the violation of the first law of thermodynamics due to dark matter. If those violations are correct, a lot of people would have to be convinced otherwise.

  12. Hugo Minney 28 May 2012 at 9:45 pm #

    Well done Dr Briffa for seeing the bigger picture, and I do apologise that I’ve taken so long to read this.
    Naturally I’m biassed, since a lot of my work is identifying the symptoms of B12 deficiency, but the AD Smith Oxford study clearly showed not just the association between B12 supplements and lower homocysteine levels and reduced brain shrinkage and improved cognitive ability, but also went a long way to demonstrating the chain of causality. We all know that B12 changes homocysteine to methylmalonic acid (and a lack of B12 causes a build-up of homocysteine which is bad for the heart). What AD Smith et al showed is that the absence of B12 is strongly associated with brain shrinkage in the elderly and that it’s due to another chemical in the brain, and that supplementing elderly people with extra B12 (on top of the amount in their diet) stops this shrinkage and the cognitive decline that follows. Since the population they studied weren’t specifically B12 deficient they were anybody over 65, and the results were statistically significant, they concluded that B12 supplements on average prevent dementia.
    Far from being ignored, NICE (National Institute for Clinical Excellence) took this finding very seriously. In UK we have a payment system called “Quality Outcomes Framework” or QOF, which consists of up to 250 indicators. To get a new indicator in, an old indicator has to be removed. So when NICE introduced a requirement within QOF for GPs to record all dementia patients’ B12 status, they were according this finding the highest possible importance.
    That’s when the conspiracy theory kicks in. The normal cycle for QOF is that in the first year, GPs are required to record status (whether cholesterol, blood glucose, B12). In the second year and for usually around 4 years, they are required to do something to improve the status (eg for blood glucose they are scored on the % of patients with HbA1c below 8.5, 7.5 and 6.5) and the thresholds and % required to get paid get tighter each year. Then when nearly everyone is getting 100% on a score, that score is removed and the next thing that needs improvement is introduced.
    In the case of B12, this hasn’t happened. There’s a lot of money to be made from people’s misery and continued sickness, and very little money (for big pharma of course) to be made from getting people well. Of course for the country, sick people cost and well people contribute, and NHS should be seen as a profit centre (the thing that keeps people at work or gets them back to work quickly) not as a hole you pour money into. So, what happened to B12 in QOF? Well the requirement didn’t change this year – it didn’t follow the usual pattern. So it is highly likely that something has made a case to prevent this progression.
    No of course I can’t prove it – just because I’m paranoid doesn’t mean they AREN’T out to get me!


  1. Still no evidence that the Swedes are killing themselves with low-carb diets | Dr Briffa's Blog - A Good Look at Good Health - 27 February 2013

    [...] heard similar utter nonsense spout from his mouth at a scientific meeting. You can read about this here. In short, he lauded quite useless epidemiological evidence again in his ‘theories’ [...]

  2. B-vitamin therapy looks promising for staving off Alzheimer’s disease | Dr Briffa's Blog - A Good Look at Good Health - 23 May 2013

    [...] year, I reported on research which found that supplementation with the vitamins B6, B12 and folic acid had been found to reduce [...]

Leave a Reply