Commentary: Inventing Diagnosis for Civil Commitment of Rapists

SVP Civil Commitment Cases and Forensic Diagnosis

Just as SVP commitment laws have generated considerable legal controversy, so have certain practices of forensic psychiatrists and psychologists who testify as experts in the trials of SVP commitment cases. The forensic assessments of sexual offenders facing commitment under SVP commitment laws are almost exclusively performed by psychologists. There are at least three possible reasons for the near absence of forensic psychiatrists from this arena. First, most of these cases have a fee retainer potential that is less than most forensic psychiatrists can expect in other types of cases. Second, one of the primary tasks in these cases—sexual offender risk assessment—is an area of specialized expertise that has been pioneered both in research and practice primarily by psychologists. Finally, the American Psychiatric Association (APA) has taken a strong position in opposition to SVP commitment laws. In 1998, its Task Force Report on Sexually Dangerous Offenders declared: [S]exual predator commitment laws represent a serious assault on the integrity of psychiatry, particularly with regard to defining mental illness and the clinical conditions for compulsory treatment. Moreover, by bending civil commitment to serve essentially non-medical purposes, sexual predator commitment statutes threaten to undermine the legitimacy of the medical model of commitment… . [P]sychiatry must vigorously oppose these statutes, to preserve the moral authority of the profession and to ensure continuing societal confidence in the medical model of civil commitment [Ref. 10, p 173].

In addition, the APA filed amicus curiae briefs arguing against the constitutionality of SVP commitment in the two cases in which the U.S. Supreme Court upheld such laws. One psychiatrist even resigned his position as chief psychiatrist of a state hospital rather than acquiesce to the civil commitment of a sexual offender.¹¹ Thus, it is not surprising that very few psychiatrists are involved in forensic evaluations of persons facing SVP commitment.

Given the imprimatur that the U.S. Supreme Court placed on SVP commitment laws,⁹ the debates about SVP commitment among forensic psychiatrists and forensic psychologists are about appropriate practice in performing forensic evaluations. Generally, these laws require the state to prove that the person facing commitment, referred to as the respondent, has a mental disorder that predisposes him to engage in sexual violence and that he is likely to do so. The U.S. Supreme Court's decision in Kansas v. Crane ⁹ added the requirement that the respondent have a volitional impairment, which the Court defined as “serious difficulty in controlling behavior” (Ref. 9, p 413). Most of the forensic assessment controversy has revolved around the methodology of the risk assessment necessary to determine if the respondent is likely to commit future acts of sexual violence. The dispute in SVP commitment trials and in the published literature has concentrated most heavily on the appropriateness of the use of actuarial risk assessment instruments.⁷ By comparison, the subjects of mental disorder and volitional impairment have received short shrift. This article focuses on forensic assessment related to mental disorders and, specifically, on a particular diagnosis often relied on to make the determination that the respondent has a mental disorder that qualifies him for SVP commitment. This article will refer to psychiatrists and psychologists who perform forensic evaluations in SVP commitment cases as “forensic evaluators.”

How Rapists Are Diagnosed in SVP Commitment Cases

Because civil commitment of sexual offenders is an exercise of the state's police power to prevent crime, it can only avoid being unconstitutional preventive detention if it is based on a judicial finding that the committed individual has a diagnosed mental disorder.⁹ In Kansas v. Crane, the U.S. Supreme Court referred to the need for such a diagnosis in its summary of the constitutional prerequisites for civil commitment of a sexual offender: [T]here must be proof of serious difficulty in controlling behavior. And this, when viewed in light of such features of the case as the nature of the psychiatric diagnosis, and the severity of the mental abnormality itself, must be sufficient to distinguish the dangerous sexual offender whose serious mental illness, abnormality, or disorder subjects him to civil commitment from the dangerous but typical recidivist convicted in an ordinary criminal case (emphasis added) [Ref. 9, p 413].

The most prevalent diagnosis among men subject to SVP commitment laws is pedophilia.^12–14 These are men who have been convicted of crimes prohibiting sexual contact with children under the age of 14. The second most prevalent diagnosis among men subject to SVP commitment is paraphilia not otherwise specified (PNOS).^12–14 This diagnosis is often applied to men convicted of sexual assaults of women (i.e., rapists).¹³ For example, a typical study showed that, in Washington state, this diagnosis applied to 42.6 percent of the men held pursuant to that state's SVP commitment law.¹⁴

The application of PNOS as a diagnosis of rapists was popularized by Dennis Doren, a state hospital psychologist from Wisconsin, whose 2002 book, Evaluating Sex Offenders: A Manual for Civil Commitments and Beyond,¹⁵ has become the bible for many forensic evaluators in SVP commitment cases, especially those who testify primarily for the prosecution. Doren contends that the diagnosis of PNOS, along with his self-coined specifier, nonconsent, is appropriate “[i]f the offender has repetitively and knowingly enacted sexual contact with nonconsenting persons over a period of at least six months (specifically for sexual arousal to the nonconsensual interaction), and the behavior has caused him significant impairment in social, occupational, or other areas of functioning” (Ref. 15, p 67). Doren sets forth a list of 13 polythetic diagnostic criteria (9 inclusionary and 4 exclusionary) for forensic evaluators to use in assessing whether the rapist's pattern of behavior has been specifically for sexual arousal to the nonconsensual interaction.¹⁵ Doren's unprecedented recommendation of the application of PNOS to diagnose rapists, and the apparent adoption of that recommendation by many forensic evaluators in SVP commitment cases,¹⁴ raise important questions of professional conduct. Before a discussion of these topics, a review of DSM history preceding this diagnostic practice is in order.

Are Nonsadistic Rapists Legitimately Diagnosable? History of the DSM Answer

Why is it necessary for Doren and forensic evaluators following his lead to rely on the miscellaneous category of PNOS to diagnose rapists? The only adult diagnosis in DSM-IV-TR that specifically refers to rape in its text description is sexual sadism, a paraphilia that requires that “the suffering (including humiliation) of the victim [be] sexually exciting to the person” (Ref. 16, p 574). The prevalence of this diagnosis among rapists is quite rare, estimated at 5 to 10 percent.¹⁷ Even among men civilly committed as sexually violent predators, the prevalence of the diagnosis of sexual sadism is generally low, with one study finding it at six percent in 14 states, and not all of these offenders were rapists.¹³

No edition of the DSM has ever included a specific diagnosis for which the criteria included nonsadistic rape. Nor does the current International Classification of Diseases (ICD-10) include such a diagnosis.¹⁸ However, two researchers have posited the existence of a psychopathology involving rape behavior. Money,¹⁹ who asserted the existence of 136 paraphilias (including fellatio and transgenderism), applied the terms “biastophilic rapism or raptophilia” to rapists whose sexual arousal requires that victims, “typically a stranger, should be unsuspecting of what is about to happen, and should be maximally terror-stricken and resistant.”²⁰ However, this definition describes rapists who are already diagnosable with sexual sadism, according to the DSM-IV-TR classification.¹⁶ Abel and colleagues acknowledged that DSM does not include a paraphilia for nonsadistic rapists, but argued that it should do so “because many individuals report having recurrent, repetitive, and compulsive urges and fantasies to commit rapes” (Ref. 21, p 18). However, they conceded, “[T]he scientific evidence must be balanced with society's acceptance of such a categorization” (Ref. 21, p 19).

From 1983 to 1986, attempts were made to insert a proposed diagnosis called paraphilic coercive disorder (PCD) into DSM-III-R,²² with the following diagnostic criteria:

The proposed diagnosis of PCD was made along with proposals for two other diagnoses: masochistic personality disorder and premenstrual dysphoric disorder.²⁴ The three proposed diagnoses generated what the New York Times reported as “vigorous opposition” (Ref. 25, p C1). The American Psychological Association, the American Orthopsychiatric Association, the National Association of Social Workers, and the National Organization for Women mounted strong opposition to the proposed diagnoses.²⁴ Even the U.S. Department of Justice, which rarely takes public policy positions on matters related to mental health, argued that the proposed diagnosis of PCD would be used by criminal defendants to avoid legal responsibility in criminal prosecutions for rape.²⁴ After opposition to the three proposed diagnoses was expressed at a meeting of the Work Group to Revise DSM-III, the group's chairman, Robert Spitzer, admitted, “We didn't anticipate the strong objections of many, both in psychiatry and psychology, to this proposal. The discussion was very heated” (Ref. 25, p C1).

The opposition to the proposal for PCD should have been expected given that, by the mid 1980s, it was widely accepted that rape is a violent assault motivated by the rapist's desire for power and dominance rather than by sexual arousal, a concept advanced in popular culture by Susan Brownmiller's 1975 best-selling book, Against Our Will: Men, Women, and Rape,²⁶ and in the research literature by a body of empirical studies of the time. For example, a 1977 study that ranked the rapist's motivation in accounts from 132 rapists and 92 victims found, “[T]he offenses could be categorized as power rape … or anger rape… . There were no rapes in which sex was the dominant issue; sexuality was always in the service of other, nonsexual needs” (Ref. 27, p 1239). Similarly, in 1984 a study that examined social and demographic variables in a sample of 985 incarcerated males to compare rapists to child molesters, nonsexual violent offenders, and property offenders found that, “in general, rapists were most often similar to both serious property and violent offenders. Rapists differed most often from other sex offenders” (Ref. 28, p 157). The proposal for PCD also contradicted the then-recent reform of the criminal law, which reconceptualized rape from being the result of the rapist's uncontrolled sexual desire to being a crime of intentional assault.²⁹ This reform included new evidentiary rules that prevented the victim's dress and demeanor from being portrayed as provocative or insufficiently resistive.²⁹

Within professional psychiatry, there was considerable opposition to the proposed diagnoses. Psychiatrist Judith Herman said of the proposed diagnosis of PCD, “The diagnosis is not sufficiently based on behavioral criteria. …[T]he diagnosis of paraphilic coercive disorder does not belong in DSM-III” (Ref. 30, p 25). The then President of the American Academy of Psychiatry and the Law wrote, “Indeed it is the consensus that such categories may well be embarrassing to psychiatry as a whole” (Ref. 31, p 1). At a meeting of the two APA committees primarily responsible for revising DSM-III, forensic psychiatrist Loren Roth expressed reservations about PCD, which “included concerns about its reliability, the lack of information about its prevalence and epidemiology among rapists, problems with differential diagnosis…and possible forensic implications” (Ref. 32, p 1). The APA received hundreds of letters and petitions containing thousands of signatures from mental health professionals and others who objected to the validity of the proposed diagnoses. The controversy over the proposed diagnoses was covered by news media internationally.³³

Even one of the initial proponents of the proposed diagnosis of PCD eventually expressed doubts about its validity and acceptability. Days before the APA Board vote on the proposed diagnosis, forensic psychiatrist Park Dietz noted that, since research showed that both rapists and nonoffenders are aroused by images of sexual coercion, the evidence was “insufficient to draw strong conclusions from these two types of data alone as to the strength of the presumed relationship between the arousal pattern…[of rapists] and the commission of sexual assaults” (Ref. 34, p 1). He added, “The proposed diagnosis is not generally accepted in the psychiatric community. … I should add that most forensic psychiatrists oppose the diagnosis…” (Ref. 34, pp 1–2).

Based on the strong opposition and the concerns raised as to the validity of the proposed diagnosis of PCD, the APA Board of Trustees rejected it by a vote of 10 to 4 on June 28, 1986.^24,35 The minutes of this board meeting refer to PCD and the other simultaneously considered diagnoses as having been “extensively discussed” with a focus on “the scientific foundation for the three categories proposed for inclusion in DSM-III-R” (Ref. 35, p 34). The minutes also state, “It was suggested by a number of members that further research was needed to validate the status of these categories as diagnostic classifications” (Ref. 35, p 34). Describing the board's decision, a member of both committees that had originally recommended the diagnosis, explained that it “was found wanting credibility and lacking acceptance among psychiatrists…” (Ref. 36, p 1). The newspaper of the American Medical Association explained that the APA Board rejected the proposed diagnosis “because of the preliminary nature of the data and the difficulty physicians have in differentiating PCD from other disorders” (Ref. 37, p 41).

The history and text of DSM following the APA's rejection of PCD does not suggest any intent that the diagnostic concept that it embodied be used even within the PNOS miscellaneous category, as the following evidence demonstrates. First, after rejecting PCD, the APA Board approved DSM-III-R, which defines PNOS as describing paraphilias “that are less commonly encountered” (Ref. 22, p 280)—a definition that essentially remains in DSM-IV-TR as “less frequently encountered” (Ref. 16, p 567). It is unlikely that the rape behavior postulated by the proponents of PCD would be “less commonly encountered” than the more extreme behavior diagnosed as sexual sadism.^13,17

Second, whereas the APA Board rejected PCD outright, it placed the two other, simultaneously proposed (and equally controversial) diagnoses in the DSM-III-R²² appendix,³⁵ thereby allowing their application under an NOS category.¹⁶ Had the APA Board intended that PCD also be diagnosed via PNOS, it would have placed PCD wording in the PNOS category or, at least, in the DSM appendix. In fact, documentary evidence that is contemporaneous with the APA's rejection of PCD confirms the intent of the APA Board not to include the failed diagnosis in the PNOS category. After the board's rejection of PCD, Fred Berlin, one of the original proponents of PCD, criticized the board-approved draft of DSM-III-R as “a conscious effort to leave out the fact that some men rape as a consequence of being turned on by the coercive rather than the sadistic elements of rape” (Ref. 38, p 4). Referring again to the board's conscious effort not to include the PCD concept in the PNOS category, he added, “Again…it troubles me a great deal that no mention of those individuals who are preoccupied with thoughts and urges that center around issues of coercion are not at least mentioned” (Ref. 38, p 5).

Third, following the controversy over PCD, the APA never reconsidered it,³⁹ but, starting in DSM-IV⁴⁰ and continuing in DSM-IV-TR,¹⁶ new V-codes were inserted to cover perpetrators of sexual coercion and rape. V-codes, as defined in DSM-IV and DSM-IV-TR, are assigned when the person has a problem that is not by itself considered to be a mental disorder, but that may nevertheless be what DSM-IV-TR refers to as “a focus of clinical attention” (Ref. 16, p 731). One such classification is sexual abuse of an adult, which is coded either as V61.12, if the focus of clinical attention is on a perpetrator who has abused a partner, or V62.83, if the perpetrator has abused a nonpartner.¹⁶ The perpetrators of sexual abuse to which this V-code applies are those who engage in “sexual coercion, rape” (Ref. 16, p 738), the very persons to whom Doren would apply PNOS. The fact that DSM-IV-TR includes this rape-related designation in the V-codes is further evidence that the drafters of DSM-IV-TR meant to exclude nonsadistic rapists, not only from the paraphilias category, but, also, from the entire classification of mental disorders.

In the 16 years following the major controversy and ultimate rejection of the proposed diagnosis of PCD, there remained a broad consensus in clinical and forensic subspecialties of psychiatry and psychology that nonsadistic rapists are not appropriately diagnosable. However, Doren challenged that consensus with his 2002 book, recommending the diagnosis of PNOS-nonconsent for rapists who act “specifically for sexual arousal to the nonconsensual interaction” (Ref. 15, p 67), a definition essentially identical to the rejected diagnosis of PCD. A 2005 article that reviewed the extent to which Doren's recommendation had been accepted by published experts in the assessment and treatment of paraphilias concluded: In summary, the weight of opinion among experts in the treatment of paraphilias is that the omission of non-sadistic rape from the paraphilias category in the DSM was, and continues to be, a deliberate decision of the American Psychiatric Association. None of these experts has published a statement supporting the use of the diagnosis of paraphilia-NOS for rapists. Thus, Doren (2002) appears to stand alone among published experts in defending this use of the diagnosis [Ref. 12, p 45].

However, more recently, three published discussions of the diagnosis of PNOS have also applied it to rapists to various degrees. The proposed diagnostic criteria of DeClue⁴¹ encompass anyone who has had “over a period of at least six months, recurrent, intense, sexually arousing fantasies, sexual urges, or behaviors involving sexual activity with a nonconsenting person, and…the person has acted on these urges, or the sexual urges or fantasies cause distress or interpersonal difficulty” (Ref. 41, p 512). Because DeClue allows that the behavioral requirement of this proposed diagnosis be satisfied simply by repeated nonconsensual sexual behavior, any rapist who reoffended over at least six months would qualify. Thus, DeClue's use of PNOS is broader than Doren's, since Doren requires that the rapist act “specifically for sexual arousal to the nonconsensual interaction” (Ref. 15, p 67).

Frances et al.,⁴² assert that “PNOS, nonconsenting partners, is an inherently weak construct, given the lack of a set of defined criteria” (Ref. 42, p 381). These authors do not acknowledge Doren's 13 polythetic diagnostic criteria for assessing whether the rapist's pattern of behavior has been “specifically for sexual arousal to the nonconsensual interaction” (Ref. 15, pp 67–77), but they do suggest, “[S]uch diagnosis would require considerable evidence documenting that the rapes reflected paraphilic urges and fantasies linking the coercion to arousal” (Ref. 42, p 380). They add, “One acceptable standard for using it may be to demonstrate clear substantiation of urges and fantasies, either inferred by the acts perpetrated on the victim, or by the interview information…” (Ref. 42, p 380).

Similarly, First and Halon state, “There are certainly some dangerous sexual offenders in our society whose sexual offenses are clearly driven by a paraphilic sexual arousal pattern involving fantasies and urges to commit rape, and it may be appropriate to apply a diagnosis of paraphilia NOS to such individuals” (Ref. 43, p 452). They urge that this diagnosis be applied to rapists “with extreme caution” and only based on evidence “to establish the presence of a deviant sexual arousal pattern in which the offender is aroused specifically by the nonconsensual nature of the sexual act” (Ref. 43, p 452). Like Doren and Frances et al., First and Halon offer suggested diagnostic criteria from which diagnosticians may infer whether the rapist was aroused by the nonconsensuality of the rape and allow an inference of paraphilic arousal to be made from “the pattern of rape behavior” (Ref. 43, p 452). And, like Doren and Frances et al., First and Halon offer no empirical evidence to support their assertions that men rape because they are “aroused specifically by the nonconsensual nature of the sexual act” (Ref. 43, p 452).

In light of the clear DSM history of deliberate rejection of a proposed paraphilia for rapists whose rape behavior is motivated by sexual arousal to the nonconsensuality of the rape (PCD), the prevalent use of PNOS as a proxy for that rejected diagnosis and the advocacy of such use by Doren, DeClue, Frances et al., and First and Halon raise important concerns regarding professional conduct in forensic psychology and forensic psychiatry. The balance of this article will analyze these concerns.

Are DSM's NOS Categories Intended for Forensic Assessment?

According to DSM-IV-TR, a diagnosis may be based on an NOS category when the clinical presentation “conforms to the general guidelines for a mental disorder in the diagnostic class, but the symptomatic picture does not meet the criteria for any of the specific disorders” (Ref. 16, p 4). The NOS category is also available when the clinical presentation “conforms to a symptom pattern that has not been included in the DSM-IV classification but that causes clinically significant distress or impairment” (Ref. 16, p 4). Obviously, these provisions give the diagnostician virtual carte blanche to omit DSM-required criteria for specific diagnoses when the general criteria for the diagnostic class are met or simply to invent new diagnoses based on the subjective determination that “the symptom pattern…causes clinically significant distress or impairment” (Ref. 16, p 4). No wonder the NOS category is often referred to as a “wastebasket diagnosis” (Ref. 44, p 310).

While such license may have its place in clinical settings where the major consequence of its use is the ability of the clinician to bill third-party payers, in the forensic context, the unbridled use of NOS categories may have critical consequences for case outcome. In the SVP commitment context, the ability of a forensic evaluator to assign a paraphilia diagnosis to a rapist-respondent may literally determine whether the respondent is civilly committed. For example, in one study of a sample of 450 men evaluated for SVP commitment in Florida, a diagnosis of PNOS raised the odds of commitment by an astounding 10,580 percent,⁴⁵ an outcome that may be explained by the fact that a forensic evaluator need only resort to a miscellaneous diagnosis when some diagnosis is needed to justify a recommendation of civil commitment. However, the DSM allowance of an NOS diagnosis in clinical settings does not mean that standards of professional conduct allow such use for forensic assessments in which the stakes are typically much higher.

The text of DSM-IV-TR itself suggests that appropriate use of the manual for forensic assessments relies on specified diagnoses, not the ad hoc invention and application of a diagnosis by a forensic evaluator employing an NOS category. In this regard, DSM-IV-TR states: The use of DSM-IV in forensic settings should be informed by an awareness of the risks and limitations… . When used appropriately, diagnoses…can assist decision makers in their determinations. For example, when the presence of a mental disorder is the predicate for a subsequent legal determination (e.g., involuntary civil commitment), the use of an established system of diagnosis enhances the value and reliability of the determination. By providing a compendium based on a review of the pertinent clinical and research literature, DSM-IV may facilitate the legal decision makers' understanding of the relevant characteristics of mental disorders. The literature related to diagnoses also serves as a check on ungrounded speculation about mental disorders and about the functioning of a particular individual. Finally, diagnostic information regarding longitudinal course may improve decision making when the legal issue concerns an individual's mental functioning at a past or future point in time (emphasis added) [Ref. 16, p xxxiii].

Thus, DSM-IV-TR emphasizes that appropriate use of the manual in forensic settings relies on “an established system of diagnosis,” which it then describes as being “based on a review of the pertinent clinical and research literature” and a description of the disorder's “longitudinal course” (Ref. 16, p xxxiii). Clearly, this delineation of the appropriate use of DSM in forensic contexts limits such use to DSM's specified diagnoses, which provide text detailing “clinical and research literature” and “longitudinal course,” not to NOS diagnoses, which neither include such text nor are part of the “established system of diagnosis,” but are, instead, ad hoc inventions of individual diagnosticians. Whereas the DSM basis in research, field testing, and professional consensus makes for the “established system” of specified diagnoses, the reliance on NOS categories by forensic evaluators amounts to the “ungrounded speculation about mental disorders” (Ref. 16, p xxxiii) in the forensic context that DSM warns against.

Ethical and Legal Constraints on the Use of NOS Diagnoses in Forensic Assessment

In most jurisdictions, an expert opinion from a forensic psychiatrist or forensic psychologist is admissible in court only if she or he testifies that the opinion is made “with reasonable medical certainty” or “with reasonable psychological/scientific certainty,” respectively.⁴⁶ Black's Law Dictionary defines this standard as one “based on the general consensus of recognized medical thought” (Ref. 47, p 1273). Similarly, most states require that expert testimony be grounded in “scientific validity…and reliability” (Ref. 48, pp 594–595) or “general acceptance in the particular field to which it belongs” (Ref. 49, p 1014). NOS diagnoses are the antithesis of “reasonable certainty,” because they are ad hoc applications based on the idiosyncratic opinion of an individual diagnostician, not the consensus of recognized thought that would be reflected in a specified diagnosis. In fact, DSM-IV-TR defines NOS diagnoses as one of the “ways in which a clinician may indicate diagnostic uncertainty” (Ref. 16, pp 4–5). Thus, it is difficult to imagine how a forensic evaluator could testify “with reasonable certainty” that a diagnosis of PNOS, used as an equivalent for PCD, is scientifically valid, reliable, and accepted in the field, when, in fact, it is based on a construct that was deliberately rejected as unsupported by research and the behavioral sciences when it was considered for inclusion in the manual of diagnosis that is based on professional consensus.

Forensic psychologists using the diagnosis of PNOS as a proxy for the rejected diagnosis of PCD may be violating at least two standards of the American Psychological Association's “Ethical Principles of Psychologists and Code of Conduct.”⁵⁰ Sec. 9.01(a) of that Code, entitled “Bases for Assessments” states, “Psychologists base the opinions contained in their recommendations, reports, and diagnostic or evaluative statements, including forensic testimony, on information and techniques sufficient to substantiate their findings” (Ref. 50, p 13). That standard then cross-references Section 2.04, entitled “Bases for Scientific and Professional Judgments,” which requires that “[p]sychologists' work [be] based upon established scientific and professional knowledge of the discipline” (Ref. 50, p 5). These cross-referenced standards are contrary to the practice of using the diagnosis of PNOS as a substitute for PCD, because the practice would be a “diagnostic or evaluative statement” unsupported by “established scientific and professional knowledge of the discipline” (Ref. 50, pp 5, 13). This lack of support from the discipline is evidenced not only by the APA's deliberate, informed rejection of the proposed diagnosis of PCD, but also by the fact that the rejection was based on a determination of a lack of both scientific validity and professional consensus to support the proposed diagnosis.

Similarly, the American Psychology-Law Society's “Specialty Guidelines for Forensic Psychologists” require forensic psychologists to use knowledge “consistent with accepted clinical and scientific standards in selecting data collection methods and procedures for an evaluation” (Ref. 51, p 661). Their statements and testimony must be “communicated in ways that will promote understanding and avoid deception,” and, when testifying, they must present findings “in a fair manner” that precludes “distortion or misrepresentation” (Ref. 51, pp 663–664). A proposed update of these guidelines would also admonish forensic psychologists to be “sensitive to the problems posed by using a clinical diagnosis in forensic contexts and consider and qualify their opinions and testimony appropriately” (Ref. 52, p 14). These ethics mandates are inconsistent with the forensic practice of using the miscellaneous PNOS category as a proxy for an equivalent diagnosis that was deliberately rejected for inclusion in DSM. A psychologist engaging in this practice despite these mandates would exacerbate the ethics breach by failing to disclose to all parties that the NOS diagnosis is being used despite DSM history and text to the contrary and an absence of professional consensus or research supporting the validity or interrater reliability of the diagnosis.

Psychologists who violate the foregoing ethics may be simultaneously violating comparable legal provisions that regulate their licensure. For example, two states that have SVP commitment laws, Wisconsin and Florida, prohibit licensed psychologists from “[r]eporting distorted, erroneous, or misleading psychological information” (Ref. 53, p 11) or “[m]aking misleading, deceptive, untrue, or fraudulent representations in the practice of any profession licensed under this chapter,”⁵⁴ respectively. Licensing laws in most states also reference the American Psychological Association's Code of Ethics in their statutes and/or administrative code, making violation of standards of ethics a potential basis for discipline by the licensing board. Finally, misdiagnosis of prisoners by psychologists may constitute negligence under state law⁵⁵ or, in certain circumstances, a civil rights violation under federal law.⁵⁶

Forensic psychiatrists are guided by the Ethics Guidelines for the Practice of Forensic Psychiatry,⁵⁷ which admonish them to “adhere to the principle of honesty and…strive for objectivity” (Ref. 57, p 3). Appelbaum has referred to this ethics principle as “truth-telling” (Ref. 58, p 2). In application, this principle means “Testimony…should accurately reflect the scientific data on the subject at hand and the consensus of the field. When the testifying expert goes beyond the data or controverts generally accepted professional understandings, that deviation should be made clear” (Ref. 58, p 2). Given that the proposed diagnosis of PCD was considered, debated, and rejected by the APA, a proxy PNOS diagnosis does not “reflect the consensus of the field.” Forensic evaluators who choose to do so anyway should at least disclose that their use of the miscellaneous diagnosis is contrary to the history and text of DSM, and that it is unsupported by research to establish its validity. This lack of research grounding for a purported rape paraphilia was recognized by the APA Board of Trustees when, in 1998, it approved publication of a report that stated: Whether or not any rapist has a paraphilia represents a controversial issue in the research literature. DSM-IV has not classified paraphilic rapism as a mental disorder. Some researchers believe that a small group of rapists have diagnostic features similar to those with other paraphilias. The ability to make the diagnosis with a sufficient degree of validity and reliability remains problematic. In addition, other research has shown that many rapes are not the product of primary sexual interests, but rather represent an exercise in power and control [Ref. 10, pp 169–170].

Research Support for Validity or Reliability of a Rape Paraphilia Diagnosis Is Still Lacking

Twenty-two years after the APA's rejection of PCD and a decade after its reconfirmation of the lack of empirical support for a rape-related paraphilia, there still is little research to support such a diagnosis and much to call it into question. Given the infinite range of human sexual response, there may indeed be some men who rape, to use Doren's phrase, “specifically for sexual arousal to the nonconsensual interaction” (Ref. 15, p 67). However, numerous questions remain to be answered before such behavior may be considered a valid diagnostic taxon. Doren attempts to justify his claim as to the legitimacy of PNOS to diagnose rapists by simply asserting, “Finding some cases throughout the physiological assessment literature strongly supports the existence of this paraphilic condition” (Ref. 15, p 65). The fact that “some cases” may be found of incarcerated rapists who show sexual arousal to video or audio depictions of rape scenes does not mean that such arousal is pathologic or distinguishable from the arousal that nonrapists may exhibit under the same circumstances. Nor does it mean that men showing such arousal to artificial stimuli will experience or act on such arousal in the community. In fact, as Park Dietz suggested to the APA Board in 1986,³⁴ the assumed positive correlation between such arousal and rape behavior may be illusory.

Doren also attempts to justify his use of PNOS to describe a hypothesized rape paraphilia by citing the description of this hypothesis in the research literature.⁵⁹ The mere description or assertion of the existence of a rape paraphilia by researchers no more makes it a valid diagnostic taxon than did the invention of a mental disorder called drapetomania, the pre-Civil War diagnosis applied to African-American slaves seeking freedom.⁶⁰ The question of whether acting on sexual arousal to nonconsensual sexual interactions is pathological (and not just antisocial and morally reprehensible) is, ultimately, more of a value judgment than a scientific one. Is it logical to label rape behavior that is hypothetically driven by the rapist's sexual arousal as pathological when that label is not applied to armed robbery driven by a heroin addict's craving? If a mental disorder diagnosis can be applied to any criminal act that satisfies the criminal's cravings, then much of the criminal code could be engrafted into the DSM. Thus, it should not be surprising that leading rape researchers have not concluded that rape behavior is either pathological or caused by mental disorder.⁶¹ In fact, current research continues to support the research of the 1980s, when PCD was rejected by the APA Board, that, psychologically, rape is sexualized violence in which the perpetrator desires to dominate or control the injured party rather than to experience ultimate erotic fulfillment.⁶²

The legitimacy of a diagnostic classification must also depend on its discriminative validity (i.e., its ability to discriminate disorder from nondisorder).⁶³ A PCD-equivalent PNOS diagnosis raises the question of whether it is even possible to measure accurately a rapist's arousal to nonconsensuality. Putting aside the debate about the validity of penile plethysmography (PPG) for this purpose,⁶⁴ even Doren admits, “In general, PPG physiological findings for rapists are not consistent for all groups of such offenders and under all conditions” (Ref. 15, p 64). A 2005 review of the entire body of PPG research comparing rapists to nonrapists concluded: Rapists are characterized by high mating effort and antisociality. Are individual differences in propensity to rape also related to a paraphilic (or other) sexual disorder? There is strong phallometric evidence that many rapists are sexually different from men who do not commit rape. The meaning of the difference is yet unclear (emphasis added) [Ref. 61, p 127].

This review, by leading researchers on rape, showed that, in PPG studies, while rapists generally showed more arousal in response to rape stories than nonrapists, this difference was often explained as a function of the stories depicting particularly brutal and violent scenes more suggestive of the diagnosis of sexual sadism than a PCD type of arousal. When rapists were compared with nonrapists using stimuli that were coercive but not sadistic, and therefore more consistent with a PCD-supportive hypothesis, the differences were often nonexistent.^65,66 In any event, since PPG testing is rarely part of forensic evaluations for SVP commitment cases,⁶⁷ the discriminability of purported PCD-type rapists from other rapists or even nonrapists using PPG is largely academic.

While the PPG evidence supporting a PCD rape theory is as inconclusive today as it was in 1986 when Park Dietz brought that inconclusiveness to the attention of the APA Board, the evidence of diagnostic reliability of the diagnosis of PNOS is even worse. Levenson⁶⁸ analyzed the cases of 295 sex offenders who were assessed by two independent forensic evaluators to determine whether the men should be referred for SVP commitment. She found that the diagnosis of PNOS had an interrater reliability κ coefficient of only .36, which is regarded as poor reliability.⁶⁹ When reanalyzing the data two years later, Packard and Levenson⁷⁰ used statistical methods more likely to find higher rates of reliability, but the proportions of agreement for PNOS were still only .68. In an analysis of the Levenson and Packard study, Wollert⁷¹ found that their reliability data were inflated. Wollert's own empirical study of interrater reliability with the Doren version of PNOS-nonconsent yielded evidence of very poor reliability.⁷¹

Conclusion

Thus, the APA's decision to reject the proposed diagnosis of PCD is as justified today by the paucity of evidence as to the validity and reliability of such a diagnosis as it was in 1986. Yet that decision is minimized by Doren,¹⁵ Frances et al.,⁴² and First and Halon.⁴³ For example, Frances et al., erroneously claim, “The discussion about paraphilic coercive disorder was not widely promulgated to the general clinical community” (Ref. 42, p 380), an assertion contradicted by extensive APA records and news media accounts,^{25,30–33,35–37} by the hundreds of letters to the APA in opposition to PCD,³³ and the year-long furor within the APA that was rivaled only by the 1973 debate and rejection of another DSM paraphilia: homosexuality.⁷² If the APA's explicit 1986 rejection of PCD may be overridden by a forensic evaluator's ad hoc application of an equivalent diagnosis couched as PNOS, could homosexuality again be diagnosed, now by using PNOS, despite the APA's 1973 rejection of homosexuality as a valid diagnosis? Since the scope of PNOS is limited only by the imagination of the diagnostician and his or her willingness to assert that the diagnosed condition causes “clinically significant distress or impairment” (Ref. 16, p 4), such an abuse is not inconceivable, especially given the prevalence of so-called reparative/conversion therapies⁷³ to attempt alteration of homosexual orientation. This illustrates why heeding the history and intent behind DSM is so important to ensure the legitimacy of diagnosis and the professional integrity of those who apply it.

Diagnosis should never be a pretext for social control,⁷⁴ be it to enforce sexual morality or preventive detention. Imagine the bewilderment of a man who has been civilly committed to a prison-like facility, potentially for the rest of his life, based on a diagnosis that the court described as one not recognized by psychiatry but advanced by a psychologist who “perceived a gap in the American Psychiatric Association's Diagnostic and Statistical Manual” (Ref. 1, p 2). Standards of professional conduct demand that forensic evaluators resist rogue, politically-driven diagnosis. Forensic evaluators must report and testify about diagnosis with reasonable certainty, not invented pretextuality.