Tuesday, 6 December 2011

The Network of Mental Illness

A provocative but problematic paper just out offers a new perspective on psychiatric symptoms.

The basic idea is that rather than psychiatric disorders being entities, they are just bundles of symptoms which cause each other:
...symptoms are unlikely to be merely passive psychometric indicators of latent conditions; rather, they indicate properties with autonomous causal relevance. That is, when symptoms arise, they can cause other symptoms on their own. For instance, among the symptoms of MDE we find sleep deprivation and concentration problems, while GAD (generalized anxiety disorder) comprises irritability and fatigue. It is feasible that comorbidity between MDE and GAD arises from causal chains of directly related symptoms; e.g., sleep deprivation (MDE)→fatigue (MDE)→concentration problems (GAD)→irritability (GAD).
The authors seem to have mixed up their labels in the middle there, but you see the drift.

This symptom-based approach stands in contrast to the idea that psychiatric illnesses are underlying things which lead to some symptoms. So it's a challenge to the notion of underlying biological dysfunction (except maybe for specific symptoms) but it's equally incompatible with any theory of underlying psychological causes - there's no room for Freudian unconscious "complexes" here.

So there's something very straightforward and un-mysterious about this model, which will either make it attractive or suspect, depending on whether you think human life is mysterious or not.

What's the evidence? First, the authors do an analysis of the DSM-IV diagnostic manual in terms of symptoms. They take every symptom which is mentioned in at least one diagnosis. They found 439 symptoms in total, over 201 disorders, with many symptoms, such as insomnia, shared between lots of different "disorders".

They then used network analysis to create a kind of graph where the "distance" between the nodes (symptoms) is based on the number of shared diagnoses. They found that while some symptoms are unique to just one disorder, there's a core of highly shared symptoms which form a "giant component"

It's a very clever approach but I wonder what it really tells us. The DSM-IV is not data about mental illness. It's data about what we think about mental illness. Actually, it's not even that: it's data about what a particular set of people, at a particular time, were able to agree upon.

DSM-V is coming soon, and before that we had DSM's I, II and III. What about them? Do they have a different network structure? I'd have thought they would, but we don't know.

We've already seen the kinds of politics that lie behind the decision to include or exclude a diagnosis in the DSM. In the upcoming DSM-V they're seriously proposing to add a new diagnosis ("TDDD"), purely in order to stop people getting another diagnosis (childhood "bipolar").

There is a lot of symptom overlap between TDDD and bipolar disorder. Because one was designed for the purpose of diverting patients from the other. But that doesn't tell us anything about real people with real symptoms. This is an extreme example and to be fair to the authors they do acknowledge some of these problems with the DSM, but still.

The authors then show that the symptomatic closeness between DSM-IV disorders predicts the rates of comorbidity between those disorders, as measured in the American population survey the NCS-R. This is true even of disorders which don't share a common symptom but which are connected indirectly by a mutual friendship, as it were.

Finally they show that a statistical model based on interacting symptoms can predict the prevalence of depression (10% per year according to the NCS-R survey) and GAD (3% per year). It does so much better than a random model in which symptoms randomly interact.

However, I'm not convinced that all these show us that the symptom-network approach is the best model to explain the occurence of these disorders. It only shows us that it's a model that works better than a crazy random model. I'm also not sure that being able to model the NCS-R data is even a good thing, since these data are themselves of questionable validity.

But it's a genuinely interesting approach and well worth following up.

ResearchBlogging.orgBorsboom D, Cramer AO, Schmittmann VD, Epskamp S, and Waldorp LJ (2011). The small world of psychopathology. PloS one, 6 (11) PMID: 22114671


pj said...

"The authors then show that the symptomatic closeness between DSM-IV disorders predicts the rates of comorbidity between those disorders"

Isn't that tautologous. The symptomatic closeness is the number of diagnostic symptoms that overlap between the conditions. Comorbidity is just the rate at which the diagnostic symptoms for each condition happen to co-occur in the same person. It is hardly a surprise that those conditions where these diagnostic symptoms are most similar are most likely to occur together, that naturally follows from the overlap in symptoms.

pj said...

Also, I'm not entirely clear what this study adds that going through the DSM-IV counting mentions of symptoms wouldn't have done?

That insomnia, psychomotor agitation and retardation, depression, irritability, and anxiety are found widely in mental disorders is not, in itself, particularly surprising or interesting - particularly when you consider that, apart from insomnia, between them they cover most affective and behavioural symptoms.

I'm also not that impressed with any analysis of 'mental illness' that regards major depression as a single diagnostic node (or maybe two) with 7 diagnoses in the mood disorders category altogether, but lists 19 sexual and gender identity disorders and 8 sleeping disorders, 19 substance misuse disorders, 4 types of vascular dementia, and 3 disorders of childhood continence.

Anonymous said...

Clinically, I think it's interesting as it suggests a useful approach to treatment - ie treat the symptoms not the putative disorder/dysfunction. Although, as a psychologist I'd argue we've actually been doing this for a very long time...

pj said...

"Clinically, I think it's interesting as it suggests a useful approach to treatment - ie treat the symptoms not the putative disorder/dysfunction. Although, as a psychologist I'd argue we've actually been doing this for a very long time..."

It does fit in well with a CBT style of formulation, however, if you think you can treat an acutely psychotic patient with sleep hygiene and behavioural activation you're in for a shock!

Neuroskeptic said...

I guess it depends what you define as a "symptom". If you define a negative self-image (say) as a symptom then a CBT model of depression would resemble this.

But if you say, that's not a symptom, it's what causes the symptoms, then that falls apart.

And to go with that example, I don't think it is a symptom, because it's not something that manifests itself in the same way that (in the case of depression) crying, insomnia, anhedonia etc. manifest themselves.

It can take some "digging" to uncover that.

And of course if you take this to it's logical conclusion you could say that everything about a person's life is a symptom and then this model would be true by definition because it would merely say that one thing in life leads to another.

Anonymous said...

you may well already be aware of this, but Borsboom & co. have a BBS article about this : http://sites.google.com/site/borsboomdenny/CramerEtAl2010.pdf?attredirects=0

Neuroskeptic said...

Hey, no I wasn't. Thanks for that.

The more I think about this the more skeptical I get though. The problem is, the network approach is all well and good but the way they've implemented it assumes that the only things at work are the symptoms that appear in the published criteria.

Yet clearly there are loads of important things which affect mental health that aren't in these criteria. Take for instance pregnancy, we all know about postpartum depression and psychosis, whether it's psychosocial or biological in origin.

But "pregnancy" or "being postpartum" is not a DSM-IV symptom so it's not in the network.

You could add it to the network but as soon as you do, you've opened the door to all kinds of other things and the whole thing becomes unworkably broad.

The only way around it would be to say that pregnancy isn't a node, but it affects the connections between nodes, or has a kind of one-shot influence on the network... but again that's opening the doors to all kinds of other influences and the model becomes infinitely complex.

Neuroskeptic said...

Because the exciting idea here is the symptom-based model.

The network approach itself is just a mathematical tool. You could express any biological or psychological theory of disease as a network if you wanted, it's just a way of formalizing it. And it might well be a good idea to do that, to make your theories more rigorous. But I don't see how a pure symptom-based network can account for much.

Gwyn said...

It is not much different than going through a medical symptom checker: sore throat (599 results), aches (1,146 results) -- WebMD.

The point being that our body's ability to express malfunction, change or variation is limited. The brain obviously has distinct and yet reasonably limited ways in which it can express malfunction, change or variation as well. However as a social organ, symptom content varies with societal inputs.

Hypnopompic sleep disorder used to involve hallucinations of having small green hag women sit on the sufferer's chest. Now the hallucinations involve small green aliens instead. Anorexics once starved for close communion with God, and now starve for fear of fat.

And given that asthma was present in the DSM I, your observations that the symptom network is not stable from one edition to the next would be true as well.

kaney said...

Everyone has trouble sleeping from time to time, but when does not being able to sleep become a sleep disorder? The answer to this question is difficult because so many factors can affect the amount of sleep a person gets that are not related to specific disorders. Things like health conditions, stress, and sleep habits are common reasons for a person not getting the rest they need.

intramax weight loss