DSM-5: Forensic applications (Part II of II)

Courts cling to DSM as "bible"

As alluded to yesterday, in Part I, mental health professionals know not to take the DSM (or the ICD, for that matter) too seriously. It's just convenient fiction, or at best "useful constructs," mainly used to attain insurance reimbursement.

Only, there's this curious phenomenon: In the legal system, where the consequences of error can be grave, DSM diagnoses have taken on a mantra of grand truth. Increasingly, I find myself being asked during court testimony about some nit-picky little criterion or another (such as the six-month specifier for pedophilia) as if it is sacred gospel, rather than the arbitrary creation of some idiosyncratic back-room committee.

One bold colleague, when asked on the witness stand to confirm that the DSM is indeed "the bible of psychiatry," answers with a resounding "YES!" But, he adds, "Bible is Greek for 'book,' and the DSM's are a collection of books or chapters submitted by sundry subcommittees and approved or not based on politics. As with the Christian Bible, some known books (like the Book of Thomas) did not make the cut."

I don't recommend that tactic unless you are well grounded in theological studies. I myself cannot state under oath that the DSM is "the bible," when the attorney is really seeking to have me confirm its status as a learned treatise, that is, sufficiently authoritative that it should be relied upon in court. It may be the only game in town, but it's hardly known for its empirical fidelity. The text's assortment of vague generalities are not even referenced, so we don't know where they came from. If you are going to testify about a specific mental condition, such as delusional disorder, I recommend relying on empirical research from reliable sources that you can cite. 

Turning now to specific changes in the DSM-5 of most potential relevance to forensic work....

The good news is that some of the more outlandish proposals -- such as parental alienation syndrome and hebephilia -- got a resounding thumbs-down. So, here's my first-glance summary of what's new and different. 

Sexual paraphilias

An attempt by an ambitious minority to add a slew of new sexual disorders fell flat. So, you won’t find hebephilia, paraphilic coercive disorder or hypersexuality in the DSM-5. They didn’t even make the appendix for "conditions for further study" (which is populated by such non-starters as caffeine use disorder, internet gaming disorder, and the more worrisome attenuated psychosis syndrome).

These defeats are a big blow for the civil commitment industry, which lobbied for them to replace the shady "not otherwise specified" diagnoses being used to justify indefinite detention of offenders who don't have legitimate mental illnesses.

The section does, however, contain a few pesky little wording changes that may come into play in forensic cases. Each  disorder except pedophilia in the paraphilias chapter now has two remission qualifiers. If the person has not been impaired for five years, the disorder can be said to be "in full remission." This is a nod to the reality that sexual kinks often come and go over time. But there's a catch: The remission must be while the person was "in an uncontrolled environment." Otherwise, a new remission specifier of "in a controlled environment" can be applied. I anticipate that government evaluators in sexually violent predator trials may use this language to argue that a prisoner whose predicate offense was decades in the past is still disordered and at risk today, despite no objective evidence of such.

Another important change is in the text accompanying sexual sadism disorder, which now reads more like it was written for adversarial deployment. There are now two types of sadists -- "admitting individuals" and deniers. For deniers, the fact of having "inflicted pain or suffering on multiple victims on separate occasions" may be sufficient for a diagnosis. As a "general rule," the text instructs, recurrent can be interpreted to mean "three or more victims on separate occasions."

As discussed yesterday in Part I, the DSM-5 does not provide citations to empirical research to back up its recommendations. This is especially problematic in the case of sexual sadism, because even most chronic rapists are not necessarily aroused by a victim's suffering; rather, the victim's suffering fails to inhibit their arousal as it would for other men. The fact of inflicting pain or suffering also says nothing about what is going on in the mind of the inflicter, and three is just an arbitrary number pulled from a hat. These new guidelines will only complicate a problematic diagnosis with abysmally poor reliability and no predictive validity.

Antisocial personality disorder

Early buzz was that this pejorative label -- which can be applied to essentially any chronic offender -- would be revised to more closely align it with the even more pejorative and controversial construct of psychopathy. But the APA abandoned all proposed personality disorder changes (including a radical move to drop half of them altogether and to place the rest of them on a dimensional spectrum), so this diagnosis remains unchanged.

The real news here comes from the field trials. In regard to reliability, antisocial personality disorder came in at the bottom of the barrel, down there with the new mixed anxiety-depressive disorder with a kappa reliability rating of only 0.2. Historically, kappas below 0.4 have been considered poor. Although DSM-5 chief statistician Helena Kraemer is arguing that lower kappas should be deemed "acceptable," a 0.2 essentially means that even trained professionals cannot agree on whether a given individual has a disorder. This makes antisocial personality disorder far too unreliable for use in court.

Speaking of empirically dubious disorders, intermittent explosive disorder got a change worth noting. Whereas the aggressive outbursts at the core of this disorder used to require physical aggression, now "verbal aggression" suffices. If you've ever reviewed psychiatric hospital charts, you know that this is how hospital technicians chart episodes of disquiet among patients. For example, I recently saw a chart notation that "John Doe was verbally aggressive" stemming from an incident in which the involuntarily hospitalized Mr. Doe muttered profanities at hospital orderlies who had barged into his room while he was sleeping and confiscated the gauze pads he was using for an acute injury. In short, look for upticks of this disorder wherever the powerless are concentrated.

Posttraumatic stress disorder

Psychologist Richard Samuels checks his DSM
"bible" during testimony in Jodi Arias murder trial

PTSD got some significant tweaking in the DSM-5, mostly in directions that could increase its prevalence. The requirement of experiencing “fear, helplessness or horror” in reaction to the trauma was eliminated. There are now four "symptom clusters" rather than three. A new symptom of "reckless or self-destructive behavior" has been added, and the symptom of irritable behavior or angry outbursts has some added language, "typically expressed as verbal or physical aggression toward people or objects" and "with little or no provocation" (have fun explaining that one in court!).

In clinical practice, these changes won’t much matter. As Greenberg noted, "Mostly we’re content to find a label that matches people in some vague way and then get on with the business of helping them figure out what's going on in their lives that landed them in our offices." However, in court the devil is in the details. Difference between an "and" or an "or," or a three-month versus a six-month time specifier, can be critical. Unfortunately, there are no side-to-side charts with the changes from DSM-IV to DSM-5 highlighted or crossed out. The biggest benefactor of all this tweaking will be psychological test companies, whose psychometric tests for PTSD will have to be revamped. So get out your pocketbooks now.

Intellectual functioning and the death penalty

Last but not least, changes to the developmental disabilities section could make more criminals eligible for execution. Under the U.S. Supreme Court's Atkins standard, an IQ score of below 70 had been like a magic line in the sand, below which one becomes ineligible for capital punishment. However, the DSM-5's intellectual developmental disorder (renamed from mental retardation) drops IQ scores in favor of the more subjective construct of adaptive functioning, or the ability to live independently in the world.

"There are a lot of courts that are hostile to the basic legal doctrine the Atkins case established," death penalty lawyer David Dow told Reuters. "When you replace a test that is one part objective, one part subjective with a solely subjective test, it becomes easier for courts that are hostile to the constitutional principle of Atkins to evade that criterion."

"We believe that we are providing the courts with a more fine-grained means to consider adaptive functioning more comprehensively and more meaningfully," countered James Harris, of the DSM-5 work group.

Other specified or unspecified disorder

As I just mentioned, the devil is in the details. When a person does not meet minimum criteria for a diagnosis, clinicians can choose between the new categories of other specified disorder and unspecified disorder (the listed example being the unwieldy "other specified depressive disorder, depressive episode with insufficient symptoms"). These quick-and-dirty options are meant for use in the emergency department, where clinicians have little time and not much background information to go on. But the DSM-5 authors open the door for forensic misuse by stating their desire for "maximum flexibility for diagnosis." How's this for a loophole large enough to drive a Mack truck through:

"When the clinician is not able to further specify and describe the clinical presentation, the unspecified diagnosis can be given. This is left entirely up to clinical judgment."

Look to shady evaluators to misuse these "other" and "unspecified" labels to create nonexistent disorders for forensic use. That won't be anything new; it's essentially the same phenomenon we now see in sexually violent predator proceedings with the deployment of the DSM-IV-TR classifier "paraphilia not otherwise specified (NOS)," which these new categories replace. Such improper diagnosis may be legal, but that doesn't make it ethical.

Forensic caveat

One welcome change in the new manual is that the old cautionary statement about use of the DSM in forensic contexts gets more prominent play. Rather than being buried in the introduction, it's got its own little page in the DSM-5:

"... In most situations the clinical diagnosis of a DSM-5 mental disorder ... does not imply that an individual with such a condition meets legal criteria for the presence of a mental disorder or a specified legal standard...."

But when push comes to shove, judges and juries are going to do what they want to do, forensic cautions or no. As Texas lawyer Susan Orlansky -- whose client is slated for execution despite a lower-than-70 IQ -- told Reuters, "If the Texas court system is willing to ignore the DSM-IV, I don't know why they wouldn't be just as willing to ignore the DSM-5."

By all means take a moment to familiarize yourself with the changes in the new diagnostic manual that are relevant to your work. Just don't be conned into taking this whole diagnostic enterprise too seriously. After all, that's what the American Psychiatric Association is counting on to keep itself financially solvent.

I welcome comments, especially if you know of other changes of potential forensic relevance that are not listed here, or if you have a different take on the changes I highlighted.

And, if you are planning to attend the American Psychological Association convention in Honolulu, I invite you to my full-day CE training on psychiatric diagnosis in legal settings on July 31.