The Forensic Risks of DSM-V and How to Avoid Them

The DSM-IV Experience

Every revision of the Diagnostic and Statistical Manual of Mental Disorders (DSM) runs the inherent risk of creating unforeseen forensic problems. This vulnerability arises first and foremost from the fact that none of the three primary purposes of the DSM addresses its frequent use in forensic settings. The clear and necessary goals in preparing any DSM are to promote clinical, research, and educational utility.² Although high sensitivity to forensic misuse should certainly be emphasized, this problem plays only a secondary role in the construction of criteria sets and the writing of text. It is a fact of life that forensic concerns can never be at the forefront of work group effort or interest.

The inherent risk of introducing forensic concerns is further heightened by the general lack of forensic expertise among work group members. They are selected precisely because of their special contributions to the research in their own, usually narrow, area of interest. They are rarely expert in the forensic implications of diagnoses and are usually quite naïve about, and not particularly interested in, the way the suggestions they make may someday be misused in forensic settings.

Work group members do not understand that the DSM is read very differently by lawyers and by psychiatrists and other mental health practitioners. Even when the DSM criteria sets and text are written with a consistency that is sufficient for clinical, research, and educational purposes, the wording does not always stand up well to the technical rigor of precise legal dissection. By training and inclination, lawyers parse every phrase for meanings never foreseen by those writing primarily for a psychiatric audience.

In preparing the DSM-IV, we were acutely aware of, and frightened by, the underlying forensic risks of our work and the lack of expertise on the DSM-IV Task Force to aid us in foreseeing the unforeseeable. We developed what we considered to be a carefully staged process designed to mitigate the risks and avoid unpleasant surprises. The process began with a firm resolution to be very conservative in allowing changes in the DSM-IV criteria sets.³ We insisted that there be a rigorous risk/benefit analysis for each change, supported by extensive empirical evidence.⁴ Part of the rationale was that the material in the current manual had already withstood the test of time and was less likely to cause trouble than anything new that we might add. We also understood the absolute necessity of recruiting a dedicated group of forensic advisors to review carefully each word in every criteria set to identify the potential for forensic misuse. We realized that only the most careful forensic review by the most knowledgeable experts could protect us from unintended consequences and that the more eyes on the document, the less likely a gremlin might sneak through.

Our reviewers did a great job, and for many years after the publication of the DSM-IV, we believed complacently that we had dodged the potential bullets. There were no complaints or reported problems of forensic misuse due to difficulties with the DSM-IV criteria or text. This optimism turned out to be dramatically premature. As detailed extensively, especially in this Journal,^5–7 a seemingly trivial “or” for “and” editorial change in the DSM-IV definition of paraphilia has been misconstrued in a way that allows the act of rape alone to support a definition of mental disorder. This designation has facilitated the indefinite psychiatric commitment of rapists after they have completed their prison terms. Plausible arguments can certainly be made on both sides in this public policy question. However, the outcome was definitely not the intention of the DSM-IV wording change; it represents a much undesired, accidental, inadvertent, and unintended consequence that emerged unbidden and almost casually out of what had seemed to be our very careful forensic review process.

The DSM-V: A Riskier Enterprise

Even though the DSM-IV project was unambitious, superconservative, and obsessively careful, it caused this one great forensic difficulty. What then are the safety prospects for the very differently conceived and executed DSM-V? The following ways in which the DSM-V is being prepared may create a storm of forensic problems.

• Those preparing the DSM-V have had the unrealizable ambition of promoting a paradigm shift in psychiatric diagnosis, and they have articulated their openness to change. Their plan is not good news for forensic practice. Anything new is more likely to have unintended forensic consequences than are old standbys that have never caused trouble.

• The construction of the DSM-V has been a secretive process closed to the usual iterative interchange with the field that has protected previous DSMs from many mistakes.⁸

• I do not believe that the DSM-V work group includes anyone skilled in the highly technical art of writing criteria. The few criteria sets that have surfaced display internal incoherence and some external inconsistency. The more general papers meant to describe the DSM-V process are difficult to decipher.^9,10 Confusing criteria sets are a prescription for forensic confusion.

• Early signs are that the work groups are seriously considering the inclusion of new diagnoses that are likely to become forensic or insurability disasters.

• The shroud of secrecy covering the development of the DSM-V does not allow us access to even minimal information about timelines, methods for revising work group drafts, and the possibility of a forensic review. I do not have the impression that there is a sound method at work for identifying and eliminating errors.

• An application to the National Institutes of Health to fund field trials has been rejected, which suggests that any field trials that are conducted will be poorly executed.

Any one of these problems would, by itself, occasion serious concern that the DSM-V may create an array of forensic trouble. Interacting with one another, however, the combination of high aspirations, poor technical skills, secrecy, and time pressure makes for a grim prognosis unless there is a sharp midterm correction.

Everyone with a stake in forensic psychiatry should try to find ways to save the DSM-V from itself. There is some reason for hope. The leadership of the DSM-V has responded to critical commentaries¹¹ and to a pointed letter to the Board of Trustees of the American Psychiatric Association (APA) outlining these and other concerns. These beneficial results include the appointment of an external oversight committee¹²; the postponement of field trials until after work groups options have been posted and reviewed; a delay in the publication deadline; and a quieting of statements about achieving a paradigm shift along with promises of increased caution. Unfortunately, however, the secrecy still is so pervasive that everyone outside the DSM-V work groups remains in the dark about future timelines and methods.

The goal of the forensic community should be to influence the DSM-V leadership by identifying the forensic problems in DSM-V suggestions and by finding alternate preferable solutions. There are three complementary ways to exert what could be a profound influence on the heretofore closed DSM-V process. The first is for all those APA components that are related to the forensic, disability, and insurance areas to act in concert within APA governance structures to insist that they be given a formal role in systematically reviewing and vetting all suggested changes in the DSM-V drafts. The Board of Trustees and DSM-V leadership will have no choice but to grant these requests, and the oversight committee may serve as a useful intermediary. Second, formal pressure should be applied by all organizations outside the APA with an interest in having a manual that will be safe and usable in the forensic system. It is only by historical accident that the APA has temporary control of the psychiatric classification system in the United States. If it fails to produce a quality document, another auspice can always be found. The APA will be sensitive to this risk and likely to take seriously any formal requests coming from outside groups to participate in improving the DSM-V for forensic use. The third venue of influence will come from individual forensic practitioners once the work group options are finally posted. It will be very valuable to have a large number of forensic experts read each option critically with a view as to how they may be misused in the legal system.

Another focus for the forensic community should be improving and updating the two forensic cautions that appear in the introduction to DSM-IV-TR (Ref. 13, pp xxxii–xxxiii). These discuss crucial conceptual concerns, with profound practical implications, that create special confusion between psychiatry and the law. The first caution is that the concept “mental disorder” is inherently fuzzy, has never been defined precisely by either the psychiatric or the legal professions, and may be used differently across disciplines and jurisdictions. This concept should be spelled out in more detail and illustrated with examples—the paraphilia confusion being a good one.

The second caution is that the presence of a mental disorder does not by itself indicate loss of responsibility. Assigning responsibility for a given behavior can be informed, but is not governed, by the presence of a mental disorder and is determined more by legal than psychiatric constructs. The clinical terminology in some of the DSM-IV-TR criteria sets and text creates confusion about responsibility when parsed in legal settings. For example, the DSM-IV-TR wording for substance dependence includes the phrases “compulsive drug-taking behavior” and unsuccessful efforts to “control substance use” (Ref. 13, p 197). It should be made clear that this wording does not imply the same thing as loss of control or responsibility in a legal sense. The caution could deal more explicitly with the different implications of terms in legal and clinical settings.

In cautioning about the possible misuses of the DSM system, we are also aware that the caution itself may be misused to suggest that psychiatric diagnosis has no role whatever in legal determinations, a claim sometimes made by lawyers. For this reason, we also indicated ways in which the DSM can be valuable in forensic settings by providing a standardized and more reliable method of diagnosing mental disorders; a compendium of the characteristics of mental disorders; a check on ungrounded speculation; and information on longitudinal course. This list should be expanded and refined.

Conclusions

The DSM-V Task Force should first open up its process to ensure that there is the most careful vetting of DSM-V by forensic experts. But, second, the task force should be aware that even the very careful forensic review before the DSM-IV failed to anticipate the large problems introduced by small changes in the paraphilia section. All changes should be made with great caution, since each carries the potential for unforeseen forensic risk.

The forensic psychiatry community should not assume passively that problems with the DSM-V will work themselves out. The DSM-V process has so far consistently lacked a self-correcting internal homeostatic mechanism and seems to respond only to persistent external pressure. A stitch in time now will save many later years of forensic nightmares.