Cracked (2 page)

So what precisely did Spitzer do to try and set things right? How was he going to make psychiatric diagnosis more reliable and scientific? His answer was simple. The
DSM
needed to be altered in three major ways:

• • Many existing disorders would be deleted from
  DSM-II
  .
  
• • The definitions of each disorder in the old
  DSM
  would be expanded and made more specific for
  DSM-III
  .
  
• • A new checklist would be developed for
  DSM-III
  to improve the reliability of diagnosis.
  

Let's briefly look at each of these alterations more closely. The first involved Spitzer deleting some of the more unpopular and controversial mental disorders. These included some of the disorders introduced into psychiatry by psychoanalysis. In the 1970s, psychoanalysis had fallen out of vogue in psychiatry, along with many disorders it had introduced to the previous
DSM
. One of the most controversial of these disorders was homosexuality. Indeed, in the
DSM-II
homosexuality was listed as a mental disease. It was described as a “sexual deviation” and was located in the same category as pedophilia.
6

While some psychiatrists felt it was wrong to brand homosexuality an illness, the main push to remove the disorder largely came from outside pressure groups including the Gay Rights Movement. These groups asked why a normal and natural human sexual preference had been included in the
DSM
as a mental disease, especially when there was absolutely no scientific evidence to justify its inclusion. Surely it was prejudice rather than science that had placed homosexuality on the list?

Many in the psychiatric community were not so sure, but the APA, perhaps sensing the change in public mood, decided to consult the wider psychiatric community for their views. So at the APA convention in 1973, all the attending members were asked to vote on what they believed: was homosexuality a mental disorder or not? The vote was closer than expected: 5,854 psychiatrists voted to take homosexuality out of the
DSM
, while 3,810 voted to keep it in. And because the “outers” were in the majority, homosexuality ceased to be a mental disorder in 1974 and was therefore not included in Spitzer's
DSM-III
. It was politics and not science that had removed the disorder from this list. As we continue, it is worth holding that thought in mind.

To turn now to Spitzer's second alteration, this involved making the definitions of each mental disorder more specific and detailed. The idea was that if each disorder could be defined more precisely, psychiatrists would be less likely to misunderstand the disorders and therefore misapply them to patients.

The problem with the earlier
DSM-II
, Spitzer argued, was that its definitions of disorders were too open to interpretation. So, for example, in
DSM-II
“depressive neurosis” was defined in a single sentence:
This disorder is manifested by an excessive reaction of depression due to an internal conflict or to an identifiable event such as the loss of a love object or cherished possession
.
7
Spitzer believed that such vague definitions explained why psychiatrists regularly gave different diagnoses to the same patient. If a word in the dictionary were poorly defined, people would not know how to use it properly. The same was the case with psychiatric diagnoses. If the definitions of mental disorders were too vague and inexact, different psychiatrists would apply them in different ways, making poor diagnostic reliability inevitable.

This imprecision was why, as Spitzer said, for the
DSM-II
“There are no diagnostic categories for which reliability [is] uniformly high … [and why] the level of reliability is no better than fair for psychosis and schizophrenia and is poor for the remaining categories.”
8
Spitzer's hope was that by sharpening the definitions there would be less scope for personal interpretation, which in turn would mean diagnostic reliability would rise.

Finally, to help further improve diagnostic reliability, Spitzer's team made a third and major alteration: they created criteria for each disorder that a patient had to meet in order to warrant the diagnosis. So while, for example, there are nine symptoms associated with depression, it was somehow decided that a patient would need to have at least
five
of them for a period of at least
two weeks
to qualify for receiving the diagnosis of depression.

The only problem was on what grounds did Spitzer's team decide that if you have five symptoms for two weeks, you suffer from a depressive disorder? Why did they choose five symptoms for two weeks instead of six symptoms for three weeks? Or, for that matter, three symptoms for five weeks? What was the science that justified putting the line where Spitzer's team chose to draw it? In an interview in 2010, the psychiatrist Daniel Carlat asked Spitzer this very question.

Carlat: How did you decide on five criteria as being your minimum threshold for depression?

Spitzer: It was just a consensus. We would ask clinicians and researchers, “How many symptoms do you think patients ought to have before you would give them the diagnosis of depression,” and we came up with the arbitrary number of five.

Carlat: But why did you choose five and not four? Or why didn't you choose six?

Spitzer: Because four just seemed like not enough. And six seemed like too much. [Spitzer smiled mischievously.]

Carlat: But weren't there any studies done to establish the threshold?

Spitzer: We did reviews of the literature, and in some cases we received funding from NIMH to do field trials … [However] when you do field trials in depression and other disorders, there is no sharp dividing line where you can confidently say, “This is the perfect number of symptoms needed to make a diagnosis” … It would be nice if we had a biological gold standard, but that doesn't exist, because we don't understand the neurobiology of depression.

I suspect that by now some of you may be scratching your heads. Wasn't the whole point of Spitzer's reform to make psychiatric diagnosis a little more scientifically rigorous? But what, you may ask, is rigorous about a committee drawing arbitrary lines between mental disorder and normality? And what is scientific about asking the psychiatric community to vote
on whether existing disorders should be removed from the
DSM
? In other words, in the name of making psychiatric diagnosis more scientific, had Spitzer's team continued to make use of the unscientific procedures that had dogged the construction of earlier manuals?

As important as this question is, I'll refrain from answering it right now, because there is a more crucial question to be addressed first: Did Spitzer's reforms actually work? Did they solve the reliability problem? If you went to see two different psychiatrists independently today, would they be likely to both assign you the same diagnosis?

In an interview for
The New Yorker
in 2005, a journalist called Alix Spiegel asked Spitzer that very question. His answer was unequivocal. “To say that we've solved the reliability problem is just not true,” said Spitzer. “It's been improved. But if you're in a situation with a general clinician, it's certainly not very good. There's still a real problem, and it's not clear how to solve the problem.”
9
Here Spitzer admits something upon which many within the profession agree: that diagnostic reliability, despite the reforms, is still woefully low.

According to a study published in the journal
Psychiatry
in 2007 that asked a group of psychiatrists whether they thought psychiatric diagnosis was now reliable, a full 86 percent said that reliability was still poor.
10
It was not only their clinical experience that led them to this conclusion, but also presumably their familiarity with existing research, including work undertaken by Spitzer himself to find out whether his reforms had worked. Its conclusions were not reassuring.

You'll remember I said that before Spitzer's
DSM-III
, two psychiatrists would give different diagnoses to the same patient 32 percent to 42 percent of the time. Spitzer found that after his reforms, psychiatrists were now disagreeing around 33 percent to 46 percent of the time—results indicating the very opposite of diagnostic improvement.
11
And these disappointing figures are consistent with other, more recent studies that also implied reliability is still poor. For example, another study published in 2006 showed that reliability actually has not improved in thirty years.
12

4

An obvious question for the British reader is whether poor diagnostic reliability is a problem in the UK After all, in the UK we have, alongside the
DSM
, the
International Classification of Diseases
(the
ICD
). Perhaps the
ICD
leads to greater reliability than the
DSM
? Although this is a reasonable question to ask, when we take the research en masse, it actually shows that using the
ICD
leads to no greater diagnostic reliability than using the
DSM
.
13
This partly explains why in countries like Britain where the
ICD
is used alongside the
DSM
, many mental health researchers and professionals often prefer the
DSM
.
14
In fact, the National Institute for Clinical Excellence (the body that sets the clinical guidelines for the National Health Service in the UK) now recommends the use of the
DSM
over the
ICD
for disorders including depression.
15
Also, in my own experience working in the NHS, the
DSM
is very much an influential manual. But even if you wanted to dispute its precise impact, as an article in the
British Journal of Psychiatry
put it, in Britain “we'd still not avoid all the problems that beset the
DSM
.”

Both manuals were developed and classify mental disorders in pretty much the same way. According to the
DSM
, “The many consultations between the developers of the
DSM-IV
and the
ICD-10
… were enormously useful in increasing the congruence and reducing meaningless differences in working between the two systems.”
16
Herbert Pardes also confirmed this to me when recounting that “the
DSM
worked very closely with the
ICD
to get worldwide cooperation between diagnostic categories.” In other words, diagnostic reliability is a problem for American psychiatry—whichever manual you employ, the reliability rates are broadly the same.

This leads me to one final point about the reliability problem that would be perilous to overlook: What would happen if some day reliability rates in psychiatry were to dramatically improve? This question is important because it reveals a more fundamental problem for psychiatry that it has yet to solve: even if every psychiatrist on the globe independently diagnosed the same patient with the same disorder (for example, with “social anxiety disorder”), this would still not
prove
that social anxiety disorder actually exists in nature, that it is actually a discrete, identifiable, biological disease or malfunction of the brain. It needs much more than mere agreement to prove that. It requires hard evidence. After all, many experts once agreed the world was flat, but that did not make it so.

Unless our sciences can test whether what we agree on is objectively the case, agreement counts for nothing from a scientific standpoint. So even if psychiatrists reach high diagnostic agreement at some future point, this would not
prove
that the mental disorders with which they diagnose patients actually exist as valid disease entities. There need to be other procedures to establish that. So the issue is: Are there other procedures? And if so, what exactly are they?

This question is so central to the entire psychiatric enterprise that I decided to ask Robert Spitzer myself.

THE DSM—A GREAT WORK … OF FICTION?

O
n a sunny May morning in 2012, I catch the train from New York City. As we leave Penn Station the train slowly shunts and rattles under the Hudson River before emerging onto the wasteland of industrial New Jersey. After passing for about thirty minutes through a bleak landscape of gnarled bog land and abandoned warehouses, signs of plusher suburbia begin to break through. As the train gains pace with each passing mile, the outside scene grows steadily more affluent—the houses get bigger, the cars shinier, and the landscape lusher, until, about fifty minutes later, we terminate at the pristine dénouement of Princeton University.

I am traveling to Princeton this early May morning because three years earlier Dr. Robert Spitzer had moved out here from nearby West Chester, Pennsylvania. His wife had taken a job at a local research laboratory, and Spitzer, now in his late seventies, had decided to embark upon one last adventure. They had chosen a large and comfortable house in the historic, leafy suburbs just northeast of the university, and as my taxi pulled up outside it was clear they had chosen well.

“Come on in,” said Spitzer, dressed in shorts, sandals, and a loose sports top, as he led me into the living area. “You wanna stay for lunch?”

Still reeling from my mountainous American breakfast, I struggled to say, “Sure, that'd be nice.”

“Before we do that,” said Spitzer, to my great relief, “how about we first sit down so I can tell you what you want to know.”

Once we had settled comfortably in our chairs, the first question I had for Spitzer concerned one of the other major changes he introduced into the
DSM
. What I did not mention in the last chapter is that a further change Spitzer introduced into the
DSM
, alongside creating a new checklist system and sharpening the definitions for each disorder, is that he introduced more than eighty new disorders, effectively expanding the
DSM
from 182 disorders (
DSM-II
) to 265 (
DSM-III
). “So what,” I asked Spitzer, “was the rationale for this huge expansion?”

“The disorders we included weren't really new to the field,” answered Spitzer confidently. “They were mainly diagnoses that clinicians used in practice but which weren't recognized by the
DSM
or the
ICD
. There were many examples, borderline personality disorder was one, and so was post-traumatic stress disorder. There were no categories for these disorders prior to
DSM-III
. By including them we gave them professional recognition.”

“So presumably,” I asked, “these disorders had been discovered in a biological sense? That's why they were included, right?”

“‘No, not at all,” Spitzer said matter-of-factly. “There are only a handful of mental disorders in the
DSM
known to have a clear biological cause. These are known as the organic disorders [like epilepsy, Alzheimer's, and Huntington's disease]. These are few and far between.”

“So, let me get this clear,” I pressed, “there are no discovered biological causes for many of the remaining mental disorders in the
DSM
?”

“Not for
many
, for
any
!
No
biological markers have been identified.”

“Well, it is important to hear you say this,” I said to Spitzer, “because this is something most people simply don't know. I did not know it when I started out training as a psychotherapist. Most of my patients do not know it today. And I suspect for many people reading this interview it will come as a surprise too.” [If you're one of the surprised or skeptical, I'll inspect this claim more thoroughly in later chapters.]

“So if there are no known biological causes,” I continued, “on what grounds do mental disorders make it into the
DSM
? What other evidence supports their inclusion?”

“Psychiatry is unable to depend on biological markers to justify including disorders in the
DSM
. So we look for other things—behavioral, psychological. We have other procedures.”

Before I discuss these “other procedures,” let me explain why you are probably surprised to hear that biological research did not guide the
DSM
's expansion. This may sound strange to you because we all expect psychiatry to work much like the rest of modern, mainstream medicine. In mainstream medicine, a name will only be given to a disease
after
its pathological roots have been identified in the body, such as in an organ, tissue, cells, etc. With few exceptions, that is how general medicine operates: once you have discovered the physical origins of a problem, only then do you give it a name, such as cystic fibrosis, cancer, or Crohn's disease.

But the surprising truth about psychiatry is that it largely operates in a completely opposite way. Rather, psychiatry first
names
a so-called mental disorder
before
it has identified any pathological basis in the body. So even when there is no biological evidence that a mental disorder exists, that disorder can still enter the
DSM
and become part of our medical culture.

Of course, understanding that psychiatry operates differently does not in and of itself mean that psychiatry's procedures are necessarily wrong. The only way to decide this is to assess whether psychiatry's alternative procedures are scientifically valid. To find out whether this is the case, I asked Spitzer to take me through the procedures his taskforce followed when deciding whether or not to include a new disorder. For example, if the findings of biology did not help the Taskforce to determine what disorders to include in
DSM-III
, then what on earth did?

“I guess our general principle,” answered Spitzer candidly, “was that if a large enough number of clinicians felt that a diagnostic concept was important in their work, then we were likely to add it as a new category. That was essentially it. It became a question of how much consensus there was to recognize and include a particular disorder.”

“So it was agreement that determined what went into the
DSM
?”

“That was essentially how it went, right.”

What sprang to mind at Spitzer's revelation was the point I made in the previous chapter about agreement not constituting proof. If a group of respected theologians all agree that God exists, this does not
prove
that God exists. All it proves is that these theologians believe it. So in what sense is psychiatric agreement different? Why, when a committee of psychiatrists agree that a collection of behaviors and feelings point to the existence of a mental disorder, should the rest of us accept they've got it right?

Perhaps in the absence of having biological evidence to convince us, they can produce other kinds of evidence to assure us that their agreements were justified. In other words, what was the evidence leading the taskforce to agree that a new disorder should be included in the
DSM
?

2

Before coming back to Spitzer for an answer, let me first put this question to the psychologist Professor Paula J. Caplan, currently a Fellow at Harvard University's Kennedy School and former consultant to two
DSM
committees. I interviewed Paula from my home in London in late April 2012, precisely because she has extensively assessed the evidence that guided many of the decisions the
DSM
Taskforce made.

One of the disorders she has focused on closely was called Masochistic Personality Disorder. Spitzer's taskforce wanted to include this new disorder in the
DSM
for people who displayed “masochistic traits,” such as those thought to invite harsh treatment from others, or those leading people to seek out pain for enjoyment.

A crucial reason why Paula J. Caplan and other critics objected to these traits being called symptoms of a psychiatric disorder was because these traits were also said to be typical of women who were victims of violence. So it was thought this diagnosis was very dangerous, not only because it could be used in courts of law to suggest that female victims of violence were in fact bringing violence upon themselves (because they had a “masochistic personality disorder”) but also because it could be used to let perpetrators of violence off the hook—they were simply doing what these women supposedly wanted.

After much energized opposition from Caplan and other psychologists, the committee finally decided to rename “Masochistic Personality Disorder” to “Self-Defeating Personality Disorder”—or the neat SDPD. But the critics then argued that this change in name still implied that there was something “self-defeating” in these victims, something compelling them in some way to invite abuse upon themselves.

“So the change in name was not really a victory at all,” said Caplan to me, “since by renaming the disorder as SDPD nothing really had changed. The renamed disorder could still be used to claim that women victims of abuse, well, kind of asked for it.”

When Caplan made this point to Spitzer, he remained unmoved. In fact, his desire to keep SDPD was so strong it would have been understandable had the critics retreated. But they didn't. Rather, at the last hour Caplan devised a simple plan: “I decided to scrutinize thoroughly the very research used to justify including SDPD in the
DSM
.” And here is what she found.

First, she discovered only two pieces of research—a remarkably small number, by anyone's standards. But as surprising as this discovery was, when Caplan actually looked at the research she became incredulous. “It was so methodologically flawed,” said Caplan, “that it would fail an undergraduate examination. In fact, it was so full of basic errors,” Caplan continued, “that I actually decided to use it on an undergraduate exam in which I asked students to point out every conceivable methodological error, because his study had so many.”

For example, in Spitzer's research a group of psychiatrists at only one university, who already accepted that SDPD existed, were shown some old case studies. All then unanimously agreed the patients in them had SDPD. Caplan pointed out that just because some psychiatrists at one hospital diagnosed their patients with SDPD was not proof that the disorder actually exists. Again, just because a group of people believes that an object in the sky is a UFO does not
prove
the object is a UFO. All it proves is that certain people believe it is. Likewise, as Caplan said, “All Spitzer's research proves is that a group of psychiatrists working in the same institution gave the same label—rightly or wrongly—to a given set of behaviors.”
17
It
proves
nothing more than that.

“But if you think that first piece of research was weak,” continued Caplan, “then consider the second piece. This involved sending out a questionnaire to a select number of members of the American Psychiatric Association. The questionnaire asked them whether the diagnosis SDPD should be included in the
DSM
. If they voted yes, then they were asked to describe what they thought the characteristics of SDPD were. If they voted no, then they were asked to return the questionnaire, blank, without any clinical data. This meant that the only data gathered about the characteristics of SDPD was data obtained from people who believed in the existence of SDPD in the first place.”

So how many psychiatrists believed SDPD to exist? How many voted yes? An official report showed that only 11 percent of those who returned the questionnaire described what they thought the characteristics of SDPD were.
18
So essentially, only 11 percent voted yes, which is surely not a representative sample of the psychiatric community. But what made matters worse was that the questionnaire was also sent to many psychiatrists who already supported the diagnosis and who were deliberately screened into the study. And these psychiatrists, we can assume, made up a portion of this 11 percent.
19

Caplan has therefore convincingly argued that neither piece of research justifies creating a new mental disorder. But that did not stop Spitzer, as she said, from proudly reporting in the
DSM
that the nature of SDPD was defined by examining the “data” from a single questionnaire—a questionnaire Spitzer claimed had been “distributed to several thousand members of the APA.”

“Spitzer does not report the methodological flaws in his research,” said Caplan incredulously, “and instead leads us to believe the creation of this disorder was based upon widespread scientific consultation and study.”

3

The reason I have discussed at length the case of SPDP is because it forces us to ask whether other disorders were included in the
DSM
on the basis of equally poor scientific evidence. Was this just an isolated example, or is it quite representative? To find out, I decided to read to Spitzer the following quotation, which claims that the research was not just poor for SDPD but for most of the mental disorders Spitzer's team included. This quote comes from one of the leading lights on Spitzer's taskforce, Dr. Theodore Millon. Here is what he said about the
DSM
's construction:

There was very little systematic research, and much of the research that existed was really a hodgepodge—scattered, inconsistent, and ambiguous. I think the majority of us recognized that the amount of good, solid science upon which we were making our decisions was pretty modest.
20

Once I'd read this quote to Spitzer, I asked him whether he agreed with Millon's statement. After a short and somewhat uncomfortable silence, Spitzer responded in a way I didn't expect.

“Well, it's true that for many of the disorders that were added there wasn't a tremendous amount of research, and certainly there wasn't research on the particular way that we defined these disorders. In the case of Millon's quote, I think he is mainly referring to the personality disorders … But again, it is certainly true that the amount of research validating data on most psychiatric disorders is very limited indeed.”