Letter to the Editor

Michael B. First1 

(1)  Division of Clinical Phenomenology, Department of Psychiatry, New York State Psychiatric Institute and Columbia University, 1051 Riverside Drive, Unit 60, New York, NY 10032, USA

Michael B. First 
Email: mbf2@columbia.edu

Published online: 10 August 2010

During each of the revisions of the American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders, Workgroups comprised of experts in their respective fields are convened with the goal of considering changes to the diagnostic criteria. Based on my experiences working on the DSM-III-R and DSM-IV revisions, there is a natural tendency for Workgroups to focus on making changes aimed at broadening the diagnostic umbrella of their assigned categories with the goal of increasing diagnostic coverage, i.e., reducing what they consider to be false negatives, an effort which inevitably comes at the cost of increasing false positives (Pincus, Frances, Davis, First, & Widiger, 1992).

In the overall scheme of the DSM, false positives (i.e., erroneously giving a diagnostic label to an individual for whom it is not justified) are problematic because of the unnecessary stigma and inappropriate treatment that may result. However, because of their central role as a proxy for the “mental abnormality” requirement in sexually violent predator commitment statutes (First & Halon, 2008), a false positive diagnosis of a paraphilia has a uniquely negative outcome, namely inappropriate and potentially indefinite civil commitment to a secure forensic psychiatric facility. For this reason, the potential false positive implications of changes to the paraphilia criteria demand special consideration. Most of the proposed changes, as embodied in the draft DSM-5 proposals in the www.dsm5.org website, have a significant potential to lead to false positive diagnoses and should consequently be reexamined and reconsidered before the DSM-5 drafts are finalized in mid-2012. Proposed changes that have significant potential for false positives include (1) expanding the “official” roster of paraphilias to include sexual arousal to pubescent children (hebephilia) and sexual arousal to coercive sex (paraphilic coercive disorder); (2) continuing the inadvertent and erroneous DSM-IV-TR inclusion of behavior in the core definition of paraphilias, and (3) allowing a paraphilia diagnosis simply by exceeding an arbitrary number of sexual offenses. Because of space limitations, this Letter will focus only on considerations of the latter two problems; concerns about the inclusion of hebephilia and paraphilic coercive disorder have been raised elsewhere (Franklin, 2009; Green, 2010; Knight, 2010; Plaud, 2009; Quinsey, 2010; Tromovitch, 2009; Zander, 2009).

The definitions of the various paraphilias in DSM-IV-TR reflect the same overall diagnostic construct. The first component of the definition lays out the core psychopathology of a paraphilia, namely the fact that the person is intensely aroused by deviant sexual stimuli. The second part requires that the deviant pattern carries negative consequences for the individual or society: for those paraphilias which involve the participation of an unwilling victim (i.e., exhibitionism, voyeurism, frotteurism, pedophilia, sexual sadism), the diagnosis is made if the person has acted on his urges or else if the urges or fantasies cause marked distress or interpersonal difficulty; for the remaining paraphilias (e.g., fetishism, sexual masochism, and transvestic fetishism), the diagnosis is made if the urges, fantasies, or behaviors cause clinically significant distress or impairment in functioning.

As described in First and Frances (2008) and First and Halon (2008), however, a wording change introduced to the diagnostic criteria for paraphilias during the last stages of the DSM-IV production process inadvertently opened the door for false positive diagnoses of paraphilias, a fact that only came to light once the diagnostic criteria for paraphilias came under intense scrutiny in the context of civil commitment litigation of sexually violent predators. It is first important to understand that this decision to adjust the wording was made under the mistaken belief that the new wording and the old wording were diagnostically equivalent, i.e., that it would identify exactly the same individuals as having a paraphilia as did DSM-III-R. In reviewing the limited empirical data base on paraphilias, the DSM-IV Sexual Disorders Workgroup recommended making no changes in the diagnostic criteria for any of the paraphilias, retaining the DSM-III-R wording of criterion A (“over a period of at least 6 months, recurrent intense sexual urges and sexually arousing fantasies involving…”). The only proposed change to the Paraphilias section of DSM-IV was the addition of a “with gender dysphoria” specifier to the criteria for Transvestic Fetishism. However, as part of a system-wide effort in the final stages of the preparation of DSM-IV to incorporate the clinical significance criterion (CSC) (Frances, 1998) into the criteria sets for most DSM-IV disorders, criterion B for all of the paraphilias (i.e., “the person has acted on these sexual urges, or the sexual urges or fantasies cause marked distress or interpersonal difficulty”) was replaced by the standard CSC wording (i.e., “the fantasies, sexual urges, or behaviors cause clinically significant impairment in social, occupational, or other important areas of functioning”). Moreover, to make up for removal from criterion B of the behavioral indicator “acted on these urges,” the phrase “or behaviors” was amended to criterion A in acknowledgement of the fact that it is typically behaviors which prompt a person to receive treatment for his or her paraphilia. Thus, the final wording of criterion A in DSM-IV and DSM-IV-TR was as follows: “over a period of at least 6 months, recurrent, intense sexually arousing fantasies, sexual urges, or behaviors involving [paraphilic focus].” This seemingly minor change in criterion A went virtually unnoticed until enterprising forensic evaluators in sexually violent predator commitment cases started using this change in wording to justify making a diagnosis of a paraphilia in cases in which the only evidence available to support the diagnosis was the nature of the sexual offenses themselves because the person either did not cooperate with the forensic evaluation or else denied having a deviant sexual arousal pattern.

There are both conceptual and practical problems related to including behaviors as part of the core definitional component of a paraphilia. As noted by Blanchard (2010) in his literature review of pedophilia, paraphilias “are erotic preferences or orientations that inhere in the individual and that have some existence independent of specific observable actions” (p. 310). A paraphilia is thus fundamentally a disturbed internal mental process (i.e., a deviant focus of sexual arousal) which is conceptually distinguishable from its various clinical manifestations, such as the person having sexual fantasies centered around the paraphilic focus, seeking and using pornography thematically linked to the paraphilic focus, having sexual urges triggered by stimuli related to paraphilic focus, or engaging in sexual behavior revolving around the paraphilic focus. Including behaviors as part of the core definition of a paraphilia conflate the underlying phenomenology of a paraphilia with its clinical manifestations.

From a practical perspective, offering “behaviors” as one of the defining elements of a paraphilia is potentially a significant source of false positive diagnoses, given that, of all of the clinical manifestations of a paraphilia, the person’s behavior is the least specific and most fallible as an indicator of an underlying paraphilic focus of sexual arousal. Given a particular behavior, there are many potential mental states that could be driving that behavior. Take, for example, the behavior of exposing one’s genitals in public. In addition to being a manifestation of an exhibitionistic paraphilic sexual arousal pattern, this behavior could be a manifestation of disinhibition or poor impulse control related to substance intoxication, a manic episode, or personality change due to a dementing illness. Thus, the particular mental state underlying this inappropriate behavior must be elucidated before one can attribute the behavior to a particular motivation, such as acting on a paraphilic urge. This “disconnect” between sexual behavior and underlying mental state is further illustrated by studies suggesting that paraphilic sexual interest may be the underlying explanation in only a minority of cases of sexual offenses. For example, Marshall and Fernandez (2003) reviewed 10 studies of exhibitionists using penile plethysmography and found that 9 out of the 10 studies suggested that exhibitionists in clinical settings did not have a preference for exposing themselves. Similarly, a study by Seto and Lalumiere (2001) of over 1000 child molesters using phallometric testing as a validator demonstrated that less than one-third had an underlying pedophilic arousal pattern.

Patient self-report of recurrent sexual urges and sexual fantasies are explicit indicators of a person’s sexual arousal pattern that require little, if any, inferential suppositions on the part of the clinician. The relationship between a person’s behavior and his sexual arousal preferences, on the other hand, requires making an inference that the behavior is driven by a paraphilic preference rather than other reasons. Making behaviors definitionally equivalent to sexual urges and sexual fantasies in criterion A leads to false positive diagnoses of paraphilias by virtue of giving the clinician permission to skip this crucial step in the diagnostic assessment process, allowing for the diagnosis to potentially be applied to individuals whose sexual offenses are motivated by non-paraphilic reasons (e.g., opportunism in a person with antisocial personality disorder, intoxication-related disinhibition in a person with substance dependence).

The DSM-5 proposal to define paraphilias in terms of behaviors alone is motivated by Workgroup members’ concerns regarding potential false negatives that may result from the DSM-III-R requirement that there be intense sexual urges or sexually arousing fantasies involving the paraphilic focus. Particularly in forensic settings, individuals being evaluated for the presence of paraphilic interests have little objective motivation to be truthful regarding their sexual urges and fantasies and are thus often non-cooperative with evaluators. As noted by Blanchard (2010) with respect to pedophilia, “offenders are not necessarily rewarded for being truthful about pedophilic impulses [and]… might experience even more severe consequences of their actions if they acknowledge being pedophiles” (p. 306). Thus, the DSM-5 proposals for the majority of the paraphilias involving non-consenting persons (i.e., Exhibitionism, Frotteurism, Sexual Sadism, and Voyeurism) will continue to allow a paraphilia diagnosis to be made based entirely on behaviors in order to “lessen the dependence of diagnosis on patients’ self-reports regarding urges and fantasies,” perpetuating the already identified false positive problem and its potential legal ramifications in terms of civil commitment.

Furthermore, in recognition of the fact that allowing a diagnosis of a paraphilia to be based entirely on sexual acts alone has the potential to lead to false positives (Blanchard, 2010; Långström, 2010); for those paraphilias involving non-consenting persons, the Workgroup has proposed setting a “minimum number of separate victims for diagnosing the paraphilia in uncooperative patients” (DSM-5 Paraphilia SubWorkgroup, 2010). For Pedohebephilia, at least two different child victims are required if the children are prepubescent, and three or more if at least one is pubescent; for Sexual Sadism, two or more victims on separate occasions are required, and for Exhibitionism, Frotteurism, Paraphilic Coercive Disorder, and Voyeurism, three or more non-consenting persons on separate occasions are necessary. According to the DSM-5 web site, the use of different cut-offs for different disorders is “an attempt to obtain similar rates of false positives and false negative diagnoses for all of the paraphilias” (DSM-5 Paraphilia SubWorkgroup, 2010).

Basing the diagnosis of a paraphilia entirely on exceeding a proscribed count of the number of victims is problematic for a number of reasons. First of all, although providing a precisely defined cut-off might give the appearance that these cut-off thresholds have a firm empirical basis, this is not in fact the case. As was the case with DSM-IV, one of the central requirements of the DSM-5 revision process is for recommendations to be grounded in empirical evidence (Kupfer, Regier, & Kuhl, 2008). Although the “rationale” sections on the DSM-5 web site explain why the Workgroup proposed using a victim count (“to lessen the dependence of diagnosis on patients’ self-reports regarding urges and fantasies”), no empirical data were cited on the website to explain how or why these specific thresholds were set. An examination of the literature reviews conducted by the Workgroups to provide the empirical basis for the changes reveals that only the literature review for pedophilia offered any justification. In the Exhibitionsim, Frotteurism, and Voyeurism literature review, Långström (2010) stated that “to my knowledge, there are no published data that could directly advise on such behavioral determinants for the paraphilias reviewed here” (p. 322). The review justifying proposed changes for Sexual Sadism (Krueger, 2010) did not even address the cut-off issue.

The DSM-5 paraphilia literature review cites a single study by Blanchard, Klassen, Dickey, Kuban, and Blak (2001) as the justification for adopting a diagnostic threshold based on counting the number of victims, stating that “the results of Blanchard et al. (2001) show that absolute cut-off scores matter, at least up to three known offenses” (Blanchard, 2010, p. 309). Typically, when empirical data are used to justify a particular diagnostic cut-off, receiver operator characteristic (ROC) curves are calculated to determine the optimal balance of false positives and false negatives based on different diagnostic thresholds. For example, a study examining the optimal diagnostic cut-off of panic symptoms conducted as part of the MacArthur-funded DSM-IV secondary data reanalysis used ROC analysis to demonstrate that a diagnostic cut-off of four symptoms performed best using psychiatric hospitalization, ER visits, and suicide attempts as diagnostic validators (Leon, Klerman, Weissman, Fyer, & Johnson, 1992). An analogous ROC analysis to determine the optimal cut-off for the number of victims would have demonstrated that three or more victims yielded the best balance of false negatives vs. false positives based on some gold standard for a diagnosis of pedophilia, such as the individual admitting to a preferential sexual attraction to children.

The Blanchard et al. (2001) study, however, was not in any way designed to determine the diagnostic validity of a three victim diagnostic threshold. What the study actually showed was that the diagnostic sensitivity of phallometric testing for pedophilia (i.e., the percentage of men who are correctly diagnosed by the test as pedophilic) varied based on the number of victims, which were divided a priori into three groups: one victim, two victims, or three or more victims. For men with a history of offending children outside the family, diagnostic sensitivity was 30% for one victim, 42% for two victims, and 61% for three or more victims. For men with a history of offending entirely within the family, diagnostic sensitivity was 33% for one victim, 39% for two victims, and 29% for three or more victims. Blanchard et al. concluded in the discussion that “our analyses for offenders against unrelated children confirmed the expected result that men with greater numbers of victims had a greater likelihood of being diagnosed as pedophilic.” They go on to say that “we were not able to demonstrate the same relation among incest offenders; however, that negative result is probably not meaningful because there were few incest offenders with multiple victims” (Blanchard et al., 2001, p. 124). The only other conclusion reached by the study was that “the more adult women with whom a patient has had sexual contact, the less likely he is to be diagnosed as pedophilic” (p. 124). It thus appears that the three victim cut-off is entirely the product of expert consensus rather than being based on any empirical footing, leaving its actual impact on rates of false positive diagnoses unknown.

Moreover, using a fixed cut-off, as is proposed for the paraphilias, introduces a significant false negative problem into the picture, most evident in the diagnostic criteria for Pedohebephilia. Unlike the other non-consenting paraphilias, which typically involve victims picked opportunistically at random, pedophilia often involves multiple sexual offenses against one or two child victims who are known to the perpetrator (e.g., family members, children of neighbors). The proposed criteria for Pedohebephilia, which requires seeking sexual stimulation from two or more different children if both are prepubescent, or three or more different children if one or more are pubescent, would not diagnose such individuals as pedophiles.

This problem here stems from attempting to define a disorder in a way that is independent of the proper exercise of clinical judgment. Although a paraphilia diagnosis made in this way would most likely have improved reliability over the current approach, this increase in reliability would come at the expense of validity in terms of both false positives and false negatives. As noted in the introduction to the DSM, “the specific diagnostic criteria included in the DSM-IV are meant to serve as guidelines to be informed by clinical judgment and are not meant to be used in a cookbook fashion” (American Psychiatric Association, 2000, p. xxxii). As discussed above, a paraphilic arousal pattern is a mental phenomenon that may be manifested in a particular patient in a variety of ways. In those cases in which the forensic evaluator’s evidentiary base is by necessity limited to the person’s history of sexual offending behavior, a diagnosis of paraphilia is legitimate only if a paraphilic arousal pattern can be validly inferred from the overall pattern of behavior. Such an inference involves both ruling out other possible causes for the sexual behavior (e.g., other mental disorders such as bipolar disorder, opportunity coupled with disinhibition related to substance intoxication), and making the case that, given the nature of the pattern itself, the overwhelmingly likely motivating factor would be a paraphilic arousal pattern (e.g., involvement of a large number of victims over time coupled with the absence of sexual behavior involving non-paraphilic arousal stimuli).

In order to reduce the false positive problem inherent in allowing a paraphilia diagnosis to be made based entirely on external behavior, I recommend a return to the DSM-III-R framework which places the person’s deviant sexual arousal pattern at the core of the definition and clarifying that this arousal pattern may be manifested by sexual urges, sexual fantasies, or sexual behaviors. This approach serves to clearly differentiate between the core psychopathological construct of a paraphilia (i.e., a pattern of deviant sexual arousal) and its clinical manifestations (i.e., sexual urges, fantasies, and behaviors). The current proposed wording for pedohebephilia, which separates these two constructs in its criterion A, can be used as the model for this wording change, which should be applied uniformly to the definitions of all of the paraphilias. The following wording is suggested: “Over a period of at least 6 months, recurrent and intense sexual arousal from [paraphilic focus of sexual interest], as manifested by fantasies, urges, or behaviors.” Although this wording allows for the diagnosis to be made based on a consideration of the person’s behavioral pattern, there is an implicit requirement that the diagnostician make the clinical judgment that the behaviors are a manifestation of the deviant sexual arousal pattern that is at the core of the paraphilia and to provide adequate explication of this in the medical record. Furthermore, I recommend rejecting the imposition of an arbitrary threshold for the number of victims and instead continue with the DSM-IV-TR requirement that “the person has acted on these sexual urges, or the sexual urges or fantasies cause marked distress or interpersonal difficulty.”

A strategy for reducing diagnostic false positives that is used in the vast majority of diagnostic criteria sets in the DSM is to provide exclusionary criteria for the purpose of alerting the clinician to alternative diagnostic formulations that must be considered and ruled out (First, Frances, & Pincus, 2004). For example, the diagnostic criteria for specific phobia instruct the clinician to consider alternative diagnostic explanations for the avoidance behavior before making the diagnosis, such as a diagnosis of Obsessive–Compulsive Disorder in someone with fear of dirt related to obsessions about contamination, a diagnosis of posttraumatic stress disorder in someone who avoids stimuli associated with a severe stressor, a diagnosis of Separation Anxiety Disorder in someone who avoids going to school, etc. However, with the exception of the diagnostic criteria set for Fetishism that excludes the diagnosis if “the fetish objects are limited to articles of female clothing used in cross-dressing (as in Transvestic Fetishism) or devices designed for the purpose of tactile genital stimulation (e.g., a vibrator),” none of the paraphilia criteria sets include an exclusionary criteria.

Given the unique forensic implications of this diagnostic category, great care should be taken in the definition of the paraphilias to prevent false positive diagnoses. The current DSM-IV-TR criteria and the proposed DSM-5 criteria promote false positive diagnoses by allowing the diagnosis to be made without a consideration of the underlying motivation for sexual behavior which may be driven by motivations other than a paraphilic sexual arousal pattern. The proposals discussed here (i.e., placing sexual arousal at the core of the definition and adding exclusionary criteria) serve to counterbalance this risk while still accommodating the clinical reality that individuals, especially in forensic settings, are likely to falsely deny a history of deviant sexual urges and fantasies.

References American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disorders (4th ed., text revision). Washington, DC: Author.   Blanchard, R. (2010). The DSM diagnostic criteria for Pedophilia. Archives of Sexual Behavior, 39, 304–316.
  Blanchard, R., Klassen, P., Dickey, R., Kuban, M., & Blak, T. (2001). Sensitivity and specificity of the phallometric test for pedophilia in nonadmitting sex offenders. Psychological Assessment, 13, 118–126.
  DSM-5 Paraphilia SubWorkgroup. (2010). Rationale for Exhibitionism changes. http://www.dsm5.org/ProposedRevisions/Pages/proposedrevision.aspx?rid=62#.   First, M., & Frances, A. (2008). Issues for DSM-V. Unintended consequences of small changes: The case of paraphilias. American Journal of Psychiatry, 165, 1240–1241.
  First, M. B., Frances, A., & Pincus, H. A. (2004). DSM-IV-TR guidebook. Washington, DC: American Psychiatric Publishing.   First, M., & Halon, R. (2008). Use of DSM paraphilia diagnoses in sexually violent predator commitment cases. Journal of the American Academy of Psychiatry and the Law, 36, 443–454.
  Frances, A. (1998). Problems in defining clinical significance in epidemiological studies. Archives of General Psychiatry, 55, 119.
  Franklin, K. (2009). The public policy implications of "hebephilia". A response to Blanchard et al. [Letter to the Editor]. Archives of Sexual Behavior, 38, 319–320.   Green, R. (2010). Sexual preference for 14-year-olds as a mental disorder: You can’t be serious!! [Letter to the Editor]. Archives of Sexual Behavior, 39, 585–586.
  Knight, R. A. (2010). Is a diagnostic category for paraphilic coercive disorder defensible? Archives of Sexual Behavior, 39, 419–426.
  Krueger, R. B. (2010). The DSM diagnostic criteria for Sexual Sadism. Archives of Sexual Behavior, 39, 325–345.
  Kupfer, D. J., Regier, D. A., & Kuhl, E. A. (2008). On the road to DSM-V and ICD-11. European Archives of Psychiatry and Clinical Neuroscience, 258(Suppl. 5), 2–6.
  Långström, N. (2010). The DSM diagnostic criteria for Exhibitionism, Voyeurism, and Frotteurism. Archives of Sexual Behavior, 39, 317–324.
  Leon, A. C., Klerman, G. L., Weissman, M. M., Fyer, A. J., & Johnson, J. (1992). Evaluating the diagnostic criteria for panic disorder: Measures of social morbidity as criteria. Social Psychiatry and Psychiatric Epidemiology, 27, 180–184.
  Marshall, W. L., & Fernandez, Y. (2003). Phallometric testing with sexual offenders: Theory, research, and practice. Brandon, VT: Safer Society Press.   Pincus, H., Frances, A., Davis, W., First, M., & Widiger, T. (1992). DSM-IV and new diagnostic categories: Holding the line on proliferation. American Journal of Psychiatry, 149, 112–117.
  Plaud, J. J. (2009). Are there "Hebephiles" among us? A response to Blanchard et al. (2008) [Letter to the Editor]. Archives and Sexual Behavior, 38, 326–327.   Quinsey, V. L. (2010). Coercive Paraphilic Disorder. Archives of Sexual Behavior, 39, 405–410.
  Seto, M. C., & Lalumiere, M. L. (2001). A brief screening scale to identify pedophilic interests among child molesters. Sexual Abuse: A Journal of Research and Treatment, 13, 15–25.
  Tromovitch, P. (2009). Manufacturing mental disorder by pathologizing erotic age orientation: A comment on Blanchard et al. (2008) [Letter to the Editor]. Archives of Sexual Behavior, 38, 328.   Zander, T. K. (2009). Adult sexual attraction to early-stage adolescents: Phallometry doesn't equal pathology [Letter to the Editor]. Archives of Sexual Behavior, 38, 329–330.

Download: 

DSM-5 Proposals for Paraphilias - pdf