Psychiatry’s True Blood? Pt 1.

By Neuroskeptic | February 5, 2012 9:32 am

Imagine that there was a blood test that could detect depression. Wouldn’t that be useful?

It depends.

Ridge Diagnostics are a US company who offer such a test. They’ve just published some results of the technology in Molecular Psychiatry. In two samples of patients with major depressive disorder (MDD), they report differences in the “MDDScore”, between the patients and healthy controls.

The MDDScore is an aggregate value, calculated from the levels of 9 metabolites in blood serum. They’re all well-known molecules, including hormones, such as cortisol and prolactin. The novelty is in how they’re put together to make the MDDScore. We’re given equations – but the key variables are not provided, because they’re proprietary:

Long-term Neuroskeptic readers will recall that this “secret ingredients” approach to publishing science was also adopted by another company offering a different depression test.

Anyway, the performance of the test was impressive. In both the pilot and the replication samples, the MDDScore was significantly higher in the depressed people than in the controls. In both cases, the test had a sensitivity of over 91% and a specificity of over 81%, which is pretty good. Ridge Diagnostics are already offering the MDDScore clinically. For $745 a pop.


Although there were two depressed patient groups (n=36 and 34), there was only one set of controls (n=43); both patient samples were compared to it. This means the second, “replication”, test was not fully independent of the first one. If the first finding was a fluke caused by the control group having weird results by chance, for instance, then the second study would just repeat the fluke.

The patients were significantly older, and with a higher BMI, than the controls. They did control for these variables, which is good, but this raises the question of whether these folks differed in other ways, that they didn’t measure, and hence couldn’t control for.

In both samples, the patients had a very significantly higher MDDScore than the controls (p less than 0.0001, both times). But in both cases, the difference in levels of EGF (epidermal growth factor) was almost as strong: p=0.0003 and p less than 0.0001, respectively. Other metabolites weren’t far behind. Testing for EGF would almost certainly be cheaper than getting an MDDScore.

Finally, all these data demonstrate is that the test can distinguish between people with MDD and entirely healthy people. But how often are doctors going to need to do that? More likely, they’ll want to distinguish depression from other things that are often confused with it, such as: bipolar disorder, anxiety disorders, chronic fatigue syndrome, bereavement, “stress”, and all manner of physical illnesses e.g. thyroid problems. Daniel Carlat said last year that

If the test cannot distinguish different psychiatric problems, then the MDDScore is simply a non-specific “biomarker” for emotional difficulties of all stripes, and would be essentially useless.

How disorder-specific is the MDDScore? This paper doesn’t tell us. And to date, this is the only published paper mentioning the MDDScore. The website mentions some conference presentations, but none have yet appeared in a peer reviewed journal.

Ridge Diagnostics have an interesting history. But that’s another story – stay tuned for Part 2.

ResearchBlogging.orgPapakostas, G., Shelton, R., Kinrys, G., Henry, M., Bakow, B., Lipkin, S., Pi, B., Thurmond, L., and Bilello, J. (2011). Assessment of a multi-assay, serum-based biological diagnostic test for major depressive disorder: a Pilot and Replication Study Molecular Psychiatry DOI: 10.1038/mp.2011.166

  • Paul Whiteley

    This whole area of 'metabolomics' is starting to come to the forefront in recent times.

    A similar study looking at schizophrenia reported an AUC=1 based on five serum and one urinary metabolite (using a training and test group design)
    (with a few caveats on ethnicity et al included).

    Obviously a lot more work needs to be done in this area, but even if such tests turn out to be an aid to diagnosis, it might be useful (as well as potentially guiding intervention, etc).

  • Manchester Psychiatry Society

    Thanks for another great post.

    This bit confused me though…

    “In both samples, the patients had a very significantly higher MDDScore than the controls (p less than 0.0001, both times). But in both cases, the difference in levels of EGF (epidermal growth factor) was almost as strong: p=0.0003 and p less than 0.0001, respectively.”

    Isn't the strength of the differences measured in terms of effect size not significance level? When you talk about significant level 'p' I thought this is the chance of making a type 1 error not the extent to which the differences differ. Such that you can have quite a weak difference between variable a and variable b but a very 'strong' significance, p.

  • Anonymous

    Why would you ever need a blood test to decide if someone was depressed relative to “healthy” people? Just ask them and save $750.

  • toodles mcgee

    anon- i was thinking the same thing. but maybe it could be useful for people who have illnesses that present with depression like symptoms – could it help differentiate? or perhaps not? which begs the question – is it identifying “major depression” or symptoms of major depression? or are they one and the same?

  • tal

    The sensitivity and specificity values they report aren't cross-validated in any way. Given that they're using nine predictors, small samples, and, as you point out, the same control group for both depressed samples, the reported values are almost certain to be grossly exaggerated.

  • Neuroskeptic

    toodles mcgee: Well, that's the big question. If it did, it would be very useful, but we'll need to see it tested against people with depression-like symptoms.

  • Bernard Carroll

    The journal Molecular Psychiatry is part of Nature Publishing Group. Doesn’t NPG hold itself out as having the highest of scientific standards? The editor of MP was asleep at the wheel here. Editors are supposed to be the gatekeepers of quality, not the enablers of commercial hyperbole. Whatever possessed the editor, Julio Licinio, to allow these authors to speculate on diagnostic test status with such tiny samples and no clinical comparison group?

  • toodles mcgee

    And do the the test results change when a patient is ” in remission” on antidepressants? It is all quite curious to me.

  • Bernard Carroll

    @toodles: we may have a hard time answering that because the subjects weren't allowed to be on antidepressant drugs when tested while depressed. So they would need to disentangle direct drug actions from changes associated with remission.

    Considering that they did not bestir themselves to match patients and controls on age or sex or BMI, three notable potential metabolomic confounds, I am not persuaded that these authors understand the task ahead of them.

  • Neuroskeptic

    On their website, they say that:

    Will the test be accurate if the patient is already on an anti-depressant?
    Yes. The biomarkers are a combination of state and trait markers with the majority being trait. If a patient is currently being successfully treated with an antidepressant, preliminary studies suggest it is likely they will have an MDDScore that is indicative of a higher than normal likelihood of having MDD.”

    But they don't refer to any specific studies.

  • infinidiv

    While “…this raises the question of whether these folks differed in other ways, that they didn't measure, and hence couldn't control for.” is quite true, it is true of any study. I agree with the general assessment of the study, in particular regarding the 'secret formula', but mentioning potential unknown differences is not a valid reason to reject a study, considering that it may be (and probably is) true for most studies.

    Doesn't change the fact that this article seems more like an advertisement for a product than a true peer-review of a brand-new process… 😉

  • pj

    At $750 a pop it is still going to be cheaper to see a GP or a psychiatrist and get diagnosed the old fashioned way.

    This is of relevance only for basic research into causal mechanisms (or at least it would be if it wasn't proprietary). Of course it'll all be the results of confounding and over-fitting anyway.

  • Kelly Latta

    Good post.

    You did accidently misplaced chronic fatigue syndrome, also known as myalgic encephalomyelitis (Acheson 1959) before physical illness instead of with thyroid problems. Many do not realize the WHO classifies ME and CFS under G93.3 in chapter six with other neurological diseases.

    It's not any cheaper, but Duffy et al 2011 showed that qEEG differentiates ME/CFS from depression. But, blood tests would certainly be cheaper should they have sufficient sensitivity.

    I doubt anyone disagrees that misdiagnosing someone with something that has no cure when in reality they have something for which there is treatment can have grave consequences.

    Plus there is the whole what is normal vs what is not discussion.

  • Neuroskeptic

    infinidiv: You're quite right that unmeasured confounds could be at work in any study… but I'd have said that when the measured ones differ, the chances that the unmeasured ones differ is higher than if they were matched.

    Hard to prove that. But it seems reasonable.

  • moviedoc

    People who suffer from MDD are not “depressed” between episodes. Will the test predict another episode?

    The test could be useful for insurance companies to deny inappropriate claims for treatment of MDD.

  • Petter Häggholm

    “Why would you ever need a blood test to decide if someone was depressed relative to “healthy” people?”

    I can think of a few reasons:


    As other have said, it could help—or at least be a step toward—distinguishing between depression ‘proper’ and disorders with similar presentation, including bipolar disorder, anxiety, and ‘purely physiological’ disorders that cause depressive symptoms—like vitamin B?? deficiency. This could be very important: A bipolar patient with low incidence of mania can look very like a depressive patient, but misdiagnosing bipolar disorder as ‘monopolar’ depression can lead to disastrous treatment decisions.


    In my personal opinion, psychiatric medicine has an enormous problem in drug prescription in that there just aren’t any good ways of predicting what will work. If you’re diagnosed with dysthymia (chronic depression), your doctor will probably make a mental list of all effective antidepressants and pick the one that on average has the best profile of efficacy versus low incidence of side effects. However, this varies for individuals, and you may have to trial-and-error your way through several drugs before you find the antidepressant that works for you.

    If we can find diagnostic markers for depression, perhaps we can start to correlate more specific markers and use them for individualising prescriptions. “Your blood test suggests that an SNRI is more appropriate than an SSRI” would be vastly preferable to “I’m going to put you on these six drugs, one at a time, until we find one that work; meanwhile let’s hope the side effects of the failures don’t make you even worse”.

    In both cases, of course, I regard findings like those claimed in the paper—even if solid, validated, and replicated—as an early stage, a stepping-stone toward usefulness rather than a major goal; but it would be an important achievement nonetheless.

  • omg

    So you can market vaccines to combat depression.

  • Neuroskeptic

    Petter: Absolutely. It could be great. But at the moment we have no published data available on whether it will be useful for those applications.

  • Ivana Fulli MD

    Petter Häggeholm,

    Trust Bernard to know what he tells you about morals and Beware bribing investigators to make money selling unsound test like a new (to me ) scandal :

    Whistleblower Claims Forest Bribed Study's Investigator to Favor

    I cite :”(…)The article for which Pigott was coauthor, the complaint stated, documented that, in contrast to the positive results in STAR*D's published findings, only 108 of STAR*D's 4,041 patients (2.7 percent) had an acute-care remission and neither relapsed nor dropped out during the 12 months of continuing care that followed. The article also documented how STAR*D changed its research outcome measures and analyses, which resulted in an inflation of STAR*D's remission rates by 44.9 percent.(…)”


    Plus,as wrote a British GP in the BMJ for Christmas 2011, we should treat the clients not the lab tests.

    And it is ludicrous to prescribe dangerous drugs on a lab test only- specially when you can prescribe exercice and psychotherapy, changing lodging (a noisy sleeping room gives bad mood in some people and aspies living with a significant other thrives when they have a room all for themselves to let the pressure down when needed, etc..

  • Beth

    Where is part 2?



No brain. No gain.

About Neuroskeptic

Neuroskeptic is a British neuroscientist who takes a skeptical look at his own field, and beyond. His blog offers a look at the latest developments in neuroscience, psychiatry and psychology through a critical lens.


See More

@Neuro_Skeptic on Twitter


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Collapse bottom bar