Bipolar Diagnosis: Spectrum or Yes/No?

Table of Contents

What’s with this “spectrum” talk? The DSM is the official rule book of diagnoses, right?

Bottom line:  The “bipolar spectrum” view used to be regarded as radical, breaking away from the DSM.  But it’s not radical anymore. It’s practically mainstream now.  Both ways of looking at diagnosis have value. Use both.

“Spectrum” is not radical

First of all, let’s get this straight:  the Harvard-associated Bipolar program has been using a spectrum approach to diagnosis since 2004.  They say “Do you have bipolar disorder, or not?” is the wrong question. The right question is “how bipolar are you?” .  Note that in their Bipolarity Index they give tremendous weight to family history and age of onset of depression. These factors are markers of bipolar disorder and yet they are not in the DSM, even the new DSM. So anyone who’s insisting the DSM is the only way to approach bipolar diagnosis is hanging on to an old system. Let me show you…

Just consider the genetics: there are 226 genes associated with bipolar disorder.Nurnburger Imagine how many combinations are possible.  If each gene had only two variations — many have far more — that would be 35,000 different versions. Okay, say half of those are variations are not really different becaus they lead to a single common pathway.  That’s still at least 17,000 variations?

You get the point: this is not like Huntington’s Disease, another psychiatric illness, where you either have it or you don’t, because there’s just a single gene involved. This is completely different.  How many of the bipolar genes do you have, and what are they saying?How bipolar are you? 

For historical reasons we still have a diagnostic system that insists: you either have bipolar disorder, or you do not. There is no middle ground.  You can’t be “a little bipolar”. You can only have the whole thing, or none at all. That just doesn’t match the way this illness works.  Very few mood specialists would disagree.

Why hasn’t the whole diagnostic process been changed to reflect this new understanding? Because the DSM is a very political, very conservative system. (Think about the legal system, which uses it. Judges would not take kindly to “a little bipolar”.  They need a yes or no).

Nevertheless, more and more research papers are now referring to the “Bipolar Spectrum Disorder” (example ). The  NIMH bipolar research group is called the Bipolar Spectrum Division. Their director said in a recent important journal:

categories will remain somewhat arbitrary because they will be imposed on fully continuous, smooth distributions”  [italics mine]Liebenluft

In other words, a diagnostic system based on categories is not mapping reality. Reality is continuous — a spectrum from absent to fully present. Thus insistence on the yes/no “dichotomous” view is getting harder to justify. As one of Barcelona’s great bipolar researchers said:

“Dichotomies are useful for education, communication, simplification. Unfortunately, simplicity is useful, but untrue; whereas complexity is true, but useless.” Vieta

In the case of bipolar disorders,  the DSM is useful, to a point. It’s just not “true” (although people certainly act and talk and write as though it is). What’s true is complexity, “bipolarity”.

More evidence that the DSM-only view is changing

When the DSM-5 got underway, there was an effort to bring into it a “spectrum” view. But that failed almost completely. Here are interesting notes about that early process.

But since the arrival of the DSM-5, movement away from yes/no “categorical” system has continued. Here are two striking changes in that direction.  First, the chairman of the DSM-5 wrote that “unipolar” and “bipolar”

“might be better represented as an affective disorders continuum, with variable expressions of bipolarity representing dimensions of underlying pathophysiologic processes.”Kupfer

Dimensions is the jargon term for “spectrum”.  In other words, the guy in charge of the DSM doesn’t buy the DSM’s categories anymore.

Secondly, the entire NIMH has said the DSM system is not matching well with findings from research.  For example, even though schizophrenia and bipolar disorder are, in the DSM, completely different illnesses; in reality they share a lot of symptoms. Both can cause auditory hallucinations. Both can cause delusions. And both share a lot of genes. So, maybe they really aren’t two separate conditions. Maybe they’re two extremes on a continuum. That’s what the NIMH now says. And they’ve come up with a new way of dividing mental phenomena using a “spectrum” system (the Research Domain Criteria; RDOC).

So why didn’t the DSM-5 become a spectrum-based system? The best rationale I ever heard, from colleague Jack Katzow: “maybe it’s not a good idea to completely change the entire psychiatric diagnostic system too many times in one century.”

Okay, so let’s just agree at this point to put the “spectrum” system up there alongside the DSM system. (That’s what the Harvard-associated mood clinic did 10 years ago). Each patient should be viewed through both lenses. Some people’s problems will make sense through one lens, some through the other.  Keep your eyes on the prize: is the patient getting better?  Is it time to try another strategy, based on another view?

Double depression, bipolarity-style

Here’s one more example of how a rigid division between “unipolar” (plain depression) and “bipolar” is interfering with understanding what’s really going on.

If a person has “dysthymia”, a persistent depression, and then has episodes of Major Depression on top of that, this has been called “double depression”.  The term is used informally; it’s not in the DSM. No one quibbles about the logic of these two superimposed conditions.  Double depression is recognized to be very frequent, an unfortunate consequence of a rough childhood plus adult hard life, plus a genetic dose of Major Depression on top of that.

But what if someone had that persistent dysthymia and instead had a bipolar cycling pattern as well?  That too could be called double depression. No one ever talks about this.

Yet watch what happens when you apply this logic. First, take a typical diagram of Bipolar II cycling:


Then, lower the baseline mood from an average of 5 to 3, representing an unfortunate persistent depression:


And now superimpose the same Bipolar II cycling pattern:

double depression.2

Oh, that’s interesting:  it doesn’t look “bipolar” anymore, does it? Now it just looks like recurrent severe depressions, right? But that’s not called bipolar disorder. That’s called Major Depression, Recurrent.  Do you see how the DSM diagnoses could be confusing everyone?

Perhaps what matters is “cyclicity, not polarity” (to quote Fred Goodwin, former head of the U.S. National Institutes of Mental Health and co-author of the bipolar bible) .


If we can’t entirely abandon the DSM approach (and there are some good reasons not to), at least we should add a spectrum-based approach and use both at the same time to better map the problems we’re trying to understand.

Get Smarter About Mental Health

Our Brain Bulletin decodes mental health updates for you.

It’s free.