DSM5 in Distress

The DSM's impact on mental health practice and research

My Review Of 28 Sexually Violent Predator Cases

Paraphilia NOS is misdiagnosed and misused.

During the past two years, I have been asked by public defenders to review 28 SVP cases, mostly in California, but two each in Iowa and Washington, and one each in illinois and New Hampshire. All had previously been evaluated by two or three psychologists testifying for the prosecution and also by two or three psychologists testifying for the defense. There was the predictable split of opinion. Prosecution psychologists usually made a diagnosis of 'Paraphilia Not Otherwise Specified' and defense psychologists did not (but five prosecution psychologists did change their minds, dropped the diagnosis, and then either did not testify at all or flipped and supported the defense).

Of course, there are inherent and serious limitations in this kind of unsystematic review. The sample is very small, was not selected randomly, and may differ in important respects from the entire population of offenders diagnosed with 'Paraphilia NOS'. Conceivably, public defenders call on me only for their easiest cases, where they believe they have the best chance of winning. Equally conceivably, I could be called in as a last resort for their most hopeless cases. The defenders say I am seeing routine and representative cases who just happened to be next in line. But I have no way of confirming this and small samples of convenience can always be misleading. An even greater limitation is that I offer only my own unverified and possibly biased observations—others would make different judgments.

Find a Therapist

Search for a mental health professional near you.

Despite these caveats, I think the review exposes serious and systematic problems in how the SVP statutes are currently being implemented, and suggests the obvious next necessary steps toward improving future diagnostic practice. My method was to read the most recent reports of all the evaluators on both sides of the case, supplemented as needed by their earlier reports and by police, parole prison, and hospital records. I did no personal interviews. My job was to assess the accuracy of diagnostic judgments, the quality of the supportive documentation, and of the thoughtfulness of the diagnostic rationale.

Most striking, the diagnosis of Paraphilia NOS was justified in only 2 of the 28 cases. In both instances, the offender's sex life pivoted centrally around the committing of the offense and his ability to experience sexual arousal was predicated on performing the criminal act. Both these offenders spent much time in planning their crimes and stalking victims, maintained extensive pornography collections pertinent to their particular interests, had no other sexual life, and were preoccupied with performing their specific and stereotyped sexual act.

The other 26 cases clearly did not meet DSM IV criteria for Paraphilia NOS. Instead, their sexual offenses had been opportunistic crimes forming part of a pattern of generalized criminal behavior, very often facilitated by substance intoxication.

With remarkable consistency, the state evaluators who incorrectly over-diagnosed Paraphilia NOS provided no convincing evidence (and quite often, no evidence or rationale at all) to support their diagnoses. They typically wrote extremely long and detailed reports that were almost uniformly lacking in any meaningful discussion of diagnostic rationale. A 70-page report might contain one short paragraph on diagnosis, asserting that the offender meets the DSM IV diagnosis of Paraphilia NOS based only on the fact that he had committed more than one sex crime. The evaluators routinely confused ordinary criminal recidivism with psychiatric illness and incorrectly diagnose Paraphilia NOS based on a simple repetition of the criminal acts, as if these by themselves could possibly indicate the presence of mental disorder.
This ignores the compelling fact that rape has been explicitly rejected as grounds for diagnosing mental disorder by DSM III, DSM IIIR, DSM IV, and DSM V and that hebephilia has never been a DSM diagnosis. Rape is almost always just a crime and only very rarely is caused by mental disorder. The state evaluators consistently failed to present evidence to confirm their idiosyncratic DSM IV diagnoses and never ruled out the much more frequent correlates of sexual crimes: predator opportunism; substance intoxication; a longstanding pattern of self serving and unempathic criminal behavior; or anger or disrespect toward woman.

As I read the Supreme Court's Hendrik's decision, the Court allowed the constitutionality of SVP psychiatric commitment only if offenders are mentally disordered. It remains unconstitutional to psychiatrically detain, for preventive public safety purposes, offenders who are really no more than simple criminals.

Evaluators working with the prosecution have failed in their professional responsibility to make the crucial distinction between recidivist criminality and mental disorder. They betray a fundamental misunderstanding of DSM IV and seem unaware of their obligation to justify diagnostic decisions with solid evidence and well reasoned rationale. Their diagnostic testimony was always far short of expert- and too often was simply incompetent.

In contrast, the reports done by the evaluators who testified for the defense usually (but not always) contained a thoughtful review of the SVP literature and a detailed consideration of whether the offender met DSM IV criteria for a mental disorder. They tend to write much shorter but much more relevant reports. The diagnostic sections are done carefully and in sufficient detail to make the distinction between simple crime and mental disorder.

Fortunately, the California Department of Mental Heath has recently recognized that there are serious problems in the quality and validity of reports done by evaluators working with the prosecution. It is trying hard to improve diagnostic accuracy through retraining programs meant to instruct evaluators in proper diagnostic habits. The slides used for the DMH retraining specifically instruct evaluators that DSM-IV-TR-listed diagnoses must be established by "substantial evidence" as articulated in the case of People v. Dodd (2005) 133Cal.App.4th1564. Such instruction is desperately needed precisely because most state appointed evaluators have displayed no inclination or ability to offer any supportive evidence whatever for the key element that would legitimize a Paraphilia NOS diagnosis, i.e. a specific pattern of sexual arousal triggered by the coercion.

Another slide sets out a relevant section of the Specialty Guidelines for Forensic Psychology, adopted and published by the American Psychological Association's Council of Representatives, August 3, 2011. This document directs psychologists to maintain integrity by testing plausible rival hypotheses, i.e., the concept of differential diagnosis.

Again, state evaluators have shown a remarkable indifference to their clear obligation to conduct and document a thorough differential diagnosis before forwarding diagnostic conclusions that will bear such important conseqiences in a forensic setting.

The great news is that the California DMH has identified the problem of sloppy psychologist diagnostic habits in SVP cases and has prepared an excellent retraining program to bring them up to the standards of expert testimony. It is far too early to tell if this will work and some of the recent reports do show some reference to, and impact of, the retraining. But the bad news is that other recent reports remain completely inadequate; clearly some psychologists remain completely ignorant of proper procedure or have decoded simply to ignore it.

So far 10 of my cases have come to a decision. Six offenders were released; four continued on commitment. Most disturbing was the randomness of the decision making. On the merits alone, two of the losing cases seemed the very most likely to be obvious winners. My experience indicates that the SVP laws are being implemented in a highly arbitrary and idiosyncratic fashion with judges and juries easily confused by misleading expert testimony.

Some conclusions:

1) A very small proportion of criminal rapists committed under SVP statutes do qualify for a Paraphilia diagnosis.

2) But the overwhelming majority of committed rapists do not qualify for a diagnosis of 'Paraphilia NOS.'. This term is currently being misused and wildly over-diagnosed by evaluators who have a fundamental misunderstanding of DSM IV.

3) Since most diagnoses of 'Paraphilia NOS' are inaccurate and unreliable, they should be viewed with great skepticism and accepted only when there is convincing affirmative evidence that the criminal sexual behavior is the offender's preferred or obligatory means of sexual excitement. The differential diagnosis should require the ruling out of all the much more common causes: opportunistic sexual exploitation that is part of a criminal lifestyle, substance disinhibition, and anger or revenge.

4) The good news is that SVP cases can be won. This proves that the SVP system can work to protect individual civil rights, constitutional authority, and the integrity of involuntary psychiatric commitment.

5) The bad news is that much of the diagnostic work done by SVP evaluators is simply wrong and misleading to juries and judges- resulting in SVP decisions that are arbitrary and questionably constitutional.

6) The remedial actions to improve SVP diagnosis instituted by the California Department of Mental Health should be copied by other states. Correcting the current diagnostic misunderstandings is crucial both for the legal and the mental health systems, and absolutely necessary if they are to have an appropriate interaction.


Allen Frances, M.D., was chair of the DSM-IV Task Force and is currently professor emeritus at Duke.


Subscribe to DSM5 in Distress

Current Issue

Love & Lust

Who says marriage is where desire goes to die?