Linking Mental Disorder and Risk in Sexually Violent Person Assessments

Mental Condition Criterion by Jurisdiction

All 21 jurisdictions that have SVP laws require a mental condition for civil commitment. The terms include mental abnormality (Florida,² Iowa³, Kansas,⁴ Massachusetts¹², Missouri,⁵ New Hampshire,⁶ New Jersey,⁷ New York,¹⁹ South Carolina,⁸ Virginia,¹⁰ and Washington¹¹); mental disorder (Arizona,¹⁵ Illinois,¹⁶ Minnesota,¹³ North Dakota,²⁰ and Wisconsin¹⁷); diagnosed mental disorder (California¹); behavioral abnormality (Texas⁹); and mental illness (Nebraska¹⁸ and Tennessee³⁷). The federal government¹⁴ uses the terms mental illness, abnormality, or disorder. Irrespective of the specific term used, the statutory definitions tend to be more descriptive than symptom identifying; for example, “a congenital or acquired condition affecting the emotional or volitional capacity that predisposes the person to” commit sexually dangerous acts. This definition, or a derivative, is used by most jurisdictions (California,¹ Florida,² Illinois,¹⁶ Iowa,³ Kansas,⁴ Massachusetts,¹² Missouri,⁵ New Hampshire,⁶ New Jersey,⁷ New York,¹⁹ South Carolina,⁸ Texas,⁹ Virginia,¹⁰ Washington,¹¹ and Wisconsin¹⁷).

The terms listed are not the only qualifying mental conditions for an SVP commitment. The statutes in Arizona,¹⁵ Florida,² Kansas,⁴ Massachusetts,¹² Minnesota,¹³ Nebraska,¹⁸ New Hampshire,⁶ New Jersey,⁷ North Dakota,²⁰ South Carolina,⁸ Virginia,¹⁰ and Washington¹¹ also include personality disorder. Moreover, the statutory definitions of a qualifying disorder in other states (Minnesota,¹³ North Dakota,²⁰ and Arizona¹⁵) explicitly identify paraphilia or sexual disorder; Arizona includes conduct disorder.

With respect to case law for the 21 jurisdictions, most states' supreme and appellate courts have acknowledged that any mental condition, and not necessarily a sex-related disorder, can be used as the basis for an SVP commitment. This criterion is acceptable as long as the condition relates to the individual's “inability to control sexual impulses,” and thus supports the person's predisposition to commit sexually dangerous acts, including an antisocial personality disorder (Arizona,³⁸ Florida,³⁹ Iowa,⁴⁰ Kansas,⁴¹ Massachusetts,⁴² Minnesota,⁴³ Missouri,⁴⁴ Nebraska,⁴⁵ New Hampshire,⁴⁶ New Jersey,⁴⁷ New York,⁴⁸ North Dakota,⁴⁹ Virginia,⁵⁰ Washington,⁵¹ and Wisconsin⁵²).

DSM Codes and SVP Mental Conditions

Is a specific DSM-5 diagnostic disorder legally necessary for an SVP commitment? Mental health professionals who conduct SVP evaluations may be under the impression that the law requires listing specific diagnostic codes as found in the DSM-5,³⁶ and more specifically, that the diagnosis must be a paraphilia. As demonstrated above, the law may not hold clinicians to this standard. In the landmark case of Kansas v. Hendricks,³⁵ the U.S. Supreme Court wrote: Indeed, we have never required state legislatures to adopt any particular nomenclature in drafting civil commitment statutes. Rather, we have traditionally left to legislators the task of defining terms of a medical nature that have legal significance. As a consequence, the States have, over the years, developed numerous specialized terms to define mental health concepts. Often, those definitions do not fit precisely with the definitions employed by the medical community [Ref. 35, p 359].

Further, diagnostic debate has focused generally on two specific DSM codes and whether antisocial personality disorder (ASPD) can or should qualify, and whether the disorder has to be a paraphilia. Case law in most states has noted that ASPD can serve as a qualifying mental condition as long as a connection is formed between that individual's mental disorder and its current relationship to the risk for sexually dangerous behavior if released from confinement. Moreover, the DSM-IV-TR²⁸ definition for paraphilia not otherwise specified (and more recently, the DSM-5's³⁶ other specified and unspecified paraphilic disorder) have led some to levy this criticism: criminal offenses involving a sexual assault, such as rape, have been used inappropriately to justify a paraphilic mental disorder.^53,54

We believe that forensic evaluators are placing themselves in an unnecessary constraining box when trying to fit the individual's mental state into DSM codes so as to substantiate that an SVP qualifying mental condition exists. Doing so reflects a silo approach; particularly, when it is based heavily on historical offenses. Rather, it is more important that the forensic evaluator recognize and articulate DSM conditions through a discussion of the traits and features present in the individual that drive sexual violence risk, and not necessarily focus solely on one diagnostic code. Indeed, it is frequently the presence of multiple diagnostic conditions with specific traits and features that collectively interact with one another to enhance deviant impulses and lower sexual impulse control, thereby presenting the predisposing risk (e.g., a person with paraphilia and stimulant use disorder).

Conducting an SVP evaluation is an individualized enterprise. It should not be presumed that an individual's history of a previous paraphilic diagnosis is sufficient to support a current diagnosis of a paraphilic disorder. In addition, historical sexual offending behavior informs the diagnosis of a paraphilic disorder, but should not be the sole basis for a diagnosis. The DSM-5 acknowledges that pedophilic disorder “includes other elements that may change over time with or without treatment. … Therefore, the course of pedophilic disorder may fluctuate, increase, or decrease with age” (Ref. 36, p 699).

Relying heavily on historical criminal case facts to identify a qualifying mental condition for an SVP commitment may be not only misleading, but false. For example, there may be contextual factors (e.g., marital stress, poor interpersonal skills, use of alcohol or drugs) that led to the behavioral expression of pedophilia 20 years ago, but are no longer present. Thus, basing a diagnosis only on historical data and with no discussion of such factors as the individual's current psychosocial maturation, interest in age-appropriate activities, long-term pattern of sobriety, and adherence to treatment conditions raises the question of how this individual's current mental condition predisposes him to future sexual offenses. On the other hand, if an individual who committed child molestations 20 years ago and was diagnosed with pedophilia and a comorbid stimulant use disorder now presents with current evidence of preoccupation with children (e.g., has pictures of children in states of undress in his possession; shows interest in child-oriented activities and movies) and use of substances frequently while in a secure setting, an argument could be made that this individual's mental condition predisposes him to future sexual offenses.

The DSM-5 acknowledges the limitations of categorical characterization of mental disorders (i.e., the criteria set forming a specific diagnostic code). The introduction of the DSM-5³⁶ includes the following statement, “Indeed, the once plausible goal of identifying homogeneous populations for treatment and research resulted in narrow diagnostic categories that did not capture clinical reality, symptom heterogeneity within disorders, and significant sharing of symptoms across multiple disorders” (Ref. 36, p 12). Sexual deviancy is driven by multiple factors; it may be related to paraphilic disorders, learning processes, neurobiological disturbances, alcohol or drug abuse, or other pathological conditions. Moreover, many of the specific traits and features relevant for SVP evaluations could overlap across several DSM diagnoses. Consequently there can be criticism and confusion of diagnostic impressions instead of clarifications of the mental condition that puts the individual's at risk.

Reliance on Actuarials for Risk Assessment

Because of the initiation of SVP laws, actuarials such as the Static 99/Static 99R have assumed prominence in sexual violence risk assessments and testimony.^31,55,–,60 They frequently form the basis for forensic evaluators quantifying the level of risk an individual has for future sexual violence.^31,57,61 Caution should be taken when using, and perhaps relying too heavily on, the results from actuarial instruments at the expense of clinical judgment and the consideration of individual differences.^62,–,64 For example, at one time, the SVP commitment laws of the state of Virginia required the use of a specific actuarial risk assessment instrument as a screening tool. If the score was at or above an established number, the individual would undergo further assessment (including an in-person evaluation) to determine whether the criteria for an SVP civil commitment were met. However, if the score fell below the established number, no further attempt for an SVP civil commitment would be made, and the individual would be released from prison at the end of his sentence. The reliance on a specified actuarial instrument and score was challenged. Among the reasons cited were that it “limits the ability of qualified individuals to use their professional discretion to decide which offenders should be further evaluated, including some offenders who are likely sexually violent predators” (Ref. 65, p 29) Recently, the statute was amended wherein an actuarial is not required but may be included in the assessment.⁶⁶

The acknowledgment that people being considered for an SVP commitment should have an individual clinical examination, particularly as it relates to the person's risk of sexual harm, underscores the importance of clinical judgment. The over-reliance on risk assessment tools that are based on group data and do not allow for the consideration of individual differences is a naïve and faulty approach to identifying people at risk of violence, and it leads to the silo approach.

By way of brief explanation, actuarials yield a score based on the sum of the number of risk factors present and that number in turn is associated with a risk percentage (i.e., individuals studied in that sample with an X score had a Y rate of sexual recidivism, defined usually as arrests, over Z period of time). Actuarials use a small number of factors identified through statistical analyses as related to sexual recidivism; they are nontheoretical and are not a facet of any particular mental condition, DSM defined or otherwise. (It should be noted that an exception is the Static-2002R, which according to the test developers includes theoretical categories; for example, persistence of sexual offending, deviant sexual interests, and general criminality.⁶⁷) In addition, actuarials are often based on limited group data and therefore do not give a risk assessment unique to the individual assessed. Finally, actuarials require no mental health expertise to administer, score, or interpret. For example, in some jurisdictions, parole and probation officers perform an actuarial to determine the level of supervision for sex offenders while in the community.^68,69

Historically, this drive to quantify risk has been prompted by the statutory language of “likely,” which in some jurisdictions has been defined as “more likely than not,” “more probable than not,” or “greater than 50 percent.”⁷⁰ Although it may be argued that, for those few states where likely has been defined as greater than 50 percent, the current literature related to actuarial risk percentages associated with a score make it clear that these percentages cannot be applied to a specific individual.⁷¹ Some actuarial proponents suggest that the scores can be used to assign relative risk; that is, compared with the average offender, this score is either higher or lower than that reference group.^72,73

Forensic evaluators must acknowledge that we have no ability to predict with quantification whether an individual will act out sexually, let alone do so within a specified time frame. Risk assessment is not equivalent to risk prediction. Whether providing a score derived from an actuarial offers clarification or obfuscation may be debatable. However, it is clear that a risk score has no logical association with the mandate of civil commitment; viz, how does it relate to the person's risk of sexual offending due to a mental disorder?

The Silo Approach and the Missing Link

A silo process fosters a missing link type of SVP evaluation. By offering separate opinions regarding the diagnosis and the risk, without connecting the two, the forensic evaluator fails to address adequately the ultimate issue; viz, whether the individual does or does not meet the standard for an SVP commitment. Case law discussion supports the reasoning behind establishing the nexus between a mental condition and sexual offending behavior.^{44,49,74,–,76} Consequently, the forensic evaluator in an SVP evaluation should describe to the trier of fact those pertinent aspects of the evaluee's psychological functioning that are features of a mental condition (or DSM diagnoses) that contributed to the person's engaging in sexual offending previously, and whether (and how) those conditions remain current so as to predispose the individual to sexual reoffense if not contained in a forensic facility. Simply stated, if a forensic evaluator identifies the individual with a specific diagnosis, the evaluator must then explain in a narrative manner what traits, symptoms, and impairments the individual manifests currently that predispose him to act in a sexually dangerous manner. Historical offense facts may be useful in understanding past behavior, but should not be the only basis for identifying a current mental condition.

The second aspect of the silo approach pertains to the risk criterion, and in particular, the use of actuarial scores to identify risk. Actuarial scores are not a mental disorder and thereby have no connection to a current mental condition. Regardless of how high or low the score may be, it offers no information related to what there is clinically about the individual's mental condition that places him at risk.²⁷ The mere statistical correlation of demographic factors, criminal history, and relationship to previous victims may yield a risk score, but contributes no clinical data on why this particular offender's current mental condition places him at risk or not. In parallel, dynamic risk factors identified through structured professional judgment (SPJ) tools such as “lack of concern for others,” “impulsivity,” and “negative attitudes” can be viewed as meaningless with respect to the requirements of the SVP law if there is no explanation as to how they relate to a person's risk for sexual violence precipitated by a mental condition.

Moving Beyond the Silo Approach

The ways in which mental health experts render diagnoses can engender doubt as to the credibility of the diagnostic enterprise, and indeed the expert. The failure to articulate the link between the mental disorder and current predisposition to commit sexually violent criminal offenses can, at minimum, result in a level of frustration to the trier of fact and, at a more detrimental level, cast doubt as to the legitimacy of forensic psychiatry and psychology.