Utterly shameless diagnostic brain imaging neurobollocks

The impish Dr Amen, with (presumably) some fans. Wonder if he's planning on scanning their brains?

The impish Dr Amen, with (presumably) some fans.

Christian Jarrett’s article in Psychology Today (“Your 5-Step Self-Defence Program Against Neuro-Nonsense”) that I linked to in my previous post briefly mentioned diagnostic SPECT (Single-Photon Emission Computed Tomography) imaging, and the paper he cites (Chandler and Chatterjee, 2011 – open access – yay!) goes into some more detail about this particular brand of neurobollocks. After looking into it myself a little bit, I was absolutely stunned that some doctors a) had the gall to to continue offering these diagnostic methods and treatments to people, and b) that they were allowed to continue doing so by medical regulators (I guess that’s the USA’s free-market, laissez-faire approach to medical care – gotta love it!).

So, a little background. A common theme in neuroimaging research is attempting to identify differences in the brains between different groups of people. Male/female, musicians/non-musicians, schizophrenics/non-schizophrenics, whatever. This is all very interesting and worthwhile – if we can identify a common brain pattern or structure that is different in say, schizophrenia then we may be able to devise better treatments that target it. Voxel-based morphometry is one common method of looking at differences and assesses the relative size/shape of different brain structures, but there are lots of other methods that either look at anatomical or functional differences. The important point here is that you always need a group of people of each type that you want to compare. People’s brains vary pretty widely in the size and shape of the sulci and gyri, and in the amount/location of functional activity that you see in a given task. What is being sought in these studies is a reliable difference that is consistent across the group, and remains as a statistical effect, once all the (random) individual variation has been taken into account. Many people’s brains are a bit funny-looking, but 99 times out of 100 it’s just normal individual variation.

What this means is that developing imaging-based biomarkers that reliably indicate particular states or conditions in an individual is fraught with difficulty. Some progress is apparently being made in this endeavour, but a lot of people are still very sceptical about the idea, and rightly so.

However, Dr Amen, (of the eponymous ‘Amen Clinics‘) is certainly not one of those people. He’s been using diagnostic SPECT imaging for years, and has clinics in six major US cities. Dr Amen uses SPECT imaging to diagnose ADHD, anxiety, depression, addiction, autism, ‘marital conflict'(?!), and for generally healthy people who want to lose weight, or just ‘optimise’ their brain. How does taking a SPECT image of their brain help with this? It’s not really clear, but unsurprisingly, Dr Amen has a slick-looking brain-training-like program (‘The Amen Solution’) that promises all the usual bollocks, and an online store filled with his (many!) books, DVDs, etc. plus (of course) his own brand of dietary supplements. So, basically, I’m guessing that the results of the SPECT scans in his clinics typically indicate that the patient needs to complete a course of his brain-training, or supplements, or both.

Just to be clear – there’s absolutely no way that a SPECT scan of an individual can show up anything useful in diagnosing these disorders. Dr Amen is charging people thousands of dollars, and injecting them with radioactive substances, for absolutely no sound medical reason. 

Why use SPECT? The spatial resolution of the images is crap, and no-one uses it for clinical or research work that much anymore. However, it’s the cheapest of the 3D imaging techniques that can show functional activation. MRI scanners are expensive to buy and maintain, and need fairly large facilities. PET scanners use short-lived radio-isotopes, meaning you have to synthesise them yourself and use them immediately – you need a team of chemists and a well-kitted out laboratory. SPECT uses off-the-shelf isotopes with long half-lives; much, much cheaper and easier.

Dr Amen is a funny one, to be honest. Unlike the Brain Balance Center charlatans , he at least has some medical training and certification, and he publishes some research papers, occasionally even in peer-reviewed journals. He may perhaps be a true believer in what he’s doing, which makes him less mendacious but unfortunately also more stupid. Of course, where one quack leads others will follow, and copycat clinics have sprung up, also using SPECT, and claiming similar things; Dr SPECT scan, and CereScan are two of them. Dr Amen’s slick, media-savvy approach (he seems to be a regular on those dreadful audience-whooping, self-help TV shows they have in the US) marks him out as a serious operator though. Because he has some appropriate training, and has published some research, the Amen Clinics only gets a 9 out of 10 on the Neurobollocks rating scale; I know, I know, I’m just too generous sometimes, it’s a problem.

You can read more about the radiant Dr Amen and his techniques here, and there’s also an interesting Q & A with him over on QuackWatch.


16 responses to “Utterly shameless diagnostic brain imaging neurobollocks

  1. It’s extremely concerning. According to Amen’s site, the total radiation exposure from 2 scans (he scans everyone twice) is about 600 mrem or 6 mSv.

    Although the risks associated with low-dose radiation are hard to measure and have to be extrapolated from other data, this is believed to represent a lifetime cancer risk of over 1 in 2000 people (http://en.wikipedia.org/wiki/Technetium-99m#Radiation_side-effects)

    Given that Amen claims to have done 80,000 scans (40,000 people with 2 scans each….)… that is not good news.

  2. Aha! Neuroskeptic may have found the antidote! Also in true American fashion: the class action lawsuit. All we gotta do is find anyone who ever got cancer who had a SPECT scan in their past and sue the pants off the SPECT pushers for causing the cancer. Better yet, perhaps, try to show that those who have had a SPECT scan have a higher incidence of ADHD, etc. than the general population. (Of course there’s a selection bias. You’re thinking like a scientist, not a money-grabbing lawyer or the typical juror.) Sometimes you have to fight bad stats with bad stats.

  3. Good to see a balanced article in Psychology Today. But there’s so much nonsense on that site ( e. g. http://ronmurp.net/2012/10/08/whats-up-doc-heaven-apparently/ ) that makes even these examples of neurobollocks benign I wonder why reputable scientists use it.

    • Agreed – there’s some awful stuff in Psychology Today. They skirt too close to the edge of the motivational, self-help crowd a lot of the time. I guess it helps sell magazines…

  4. Aie, there you go – neuroscience as a chick magnet. It’s all about reproduction, isn’t it. Good scam.

  5. We just went to conference with some very smart and top neuro grad students in Chicago. They are well off of imaging as anything useful anymore. They do imaging research only because it’s popular and funded. Ho hum.

    I think PT bloggers are told to be positive and up beat. It’s like TED, pretty much worthless.

    Getting hot chicks is one thing. Furthering a real career is much more mundane.

  6. Seems like the AMA should perform a circumSPECT to rid us of his neurobollocks!

  7. Pingback: Feminist-inspired neurobollocks from Daniel Amen | NeuroBollocks

  8. Pingback: Neuroleadership – lots of old-fashioned psychology, very little neuroscience | NeuroBollocks

  9. I think it’s worth mentioning that brain scans in general are not useless; qEEG’s in particular have been showing some promise.

    The fact is that ADHD does not seem to be tied to one single genetic variation in the dopamine transporter gene, but rather CAN emerge from several different gene polymorphisms. Some people have a genetic variation in a dopamine receptor gene (making it less vigorous in it’s response to dopamine), some people in a norepinephrine gene (i.e ADRA2A), some in serotonin genes, some in the acetylcholinergic system, etc. It seems that different deficits can cause similar symptoms, but respond radically different to medications.

    As such brain scans offer the chance to narrow down the dysfunctional brain areas; the challenge is as you said establishing statistically signifigant abnormalities that are consistent across patient sub-groups, and then establishing which medication lines up best with each abnormality.

    I would agree that doctors should not be dispensing brain scans to ADHD patients as part of standard treatment, but I definitely think ADHD participation in brain scan studies (in which your scans will help researchers makes connections) is worthwhile.

  10. Here’s a Mayo clinic page citing the usefulness of SPECT scans in diagnosing brain disorders in including dementia, seizures, and epilepsy:


    Here is a clinical study asserting the effectiveness of SPECT scans in diagnosing ADHD:


    Here’s another showing a correlation between brain activity as shown in SPECT scans and OCD:


    SPECT scans and anxiety:


    SPECT scans and depression:


    SPECT scans and PTSD related to TBI:


    Here’s one reviewing multiple SPECT studies of the effectiveness of mental health medication:


    and there are a bunch that use SPECT scans to study the serotonin reuptake system in general, like this one:


    So, I guess I’m wondering what you are talking about? Are you a doctor?

    I would advise anyone looking for information on a medical issue or a technology to look for peer-reviewed research, with stated methodology, data and results, or articles that reference peer-reviewed research, not sites like this.

    • Most of the articles you cite are not relevant to the points I made in my original piece. Let’s take them one-by-one.

      1. The Mayo clinic page. Yes, SPECT might be helpful in diagnosing very serious neurological problems such as seizures, head injuries, or aneurysms. This is very different from using it to diagnose psychiatric problems like anxiety or depression, of which the Mayo clinic page makes no mention.

      2. A paper showing that some kinds of SPECT are better than others in a retrospective study of ADHD. The best kind of SPECT is only 54% accurate in this large group of subjects, so just above chance. Not great.

      3. A research paper showing some differences in cerebral blood-flow in a large cohort of OCD patients. Makes no claim about diagnostic usefulness.

      4. As for 3. above, but in depressed patients.

      5. A study on hypobaric oxygen therapy for TBI patients, where SPECT was used as an outcome measure. Hardly relevant for lots of reasons. Again, no claims about diagnosis using SPECT.

      6. A review of studies using PET and SPECT to assess effects of anti-psychotics. Again, nothing about using SPECT for diagnosis.

      7. A study using SPECT to look at 5HT receptor changes with anti-depressant medication. Guess what? Nothing about diagnosis.

      I never said SPECT wasn’t a useful research tool – it clearly is, and can give us some good information about what happens in groups of patients, compared to a normal group, or about what happens in drug action, compared to a placebo.

      My point was that SPECT (in fact, any brain imaging technology) is not useful for diagnosis of an individual patient. No-one can look at a SPECT scan of an individual and say “this person is depressed” (or whatever). That’s just not how it works. There’s too much individual variation. You need groups of subjects to find clear differences. Sometimes very large groups, because the differences are very small, compared to the amount of individual variation.

      Amen is conning people by claiming to be able to diagnose disorders on the basis of a single brain scan, in a single individual. It’s just not possible.

      >So, I guess I’m wondering what you are talking about? Are you a doctor?

      Hopefully I’ve explained it more clearly. I’m not a medical doctor, but I have a degree in psychology, a PhD in cognitive and brain sciences, and 10+ years experience working as a neuroscience and brain-imaging researcher. I know what I’m talking about.

      >I would advise anyone looking for information on a medical issue or a
      >technology to look for peer-reviewed research, with stated methodology,
      >data and results, or articles that reference peer-reviewed research, not sites
      >like this.

      I would absolutely advise the same. Unfortunately you won’t find hardly any peer-reviewed research claiming that SPECT is useful for diagnosis of psychiatric disorders, because it’s not.

  11. I would really like Neurobollocks to respond to the above comment (Maxine Twothousand’s). Interesting counter-argument.

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s