Assessment, Diagnosis, and Risk Management of Sexual Offenders with Intellectual Disabilities

Abstract

This chapter discusses characteristics associated with intellectual disabilities and those who have sexually offended and modifications needed in the assessment process including adapting interviewing and testing techniques and procedures selected for lower cognitive levels of functioning. In addition, issues related to comorbid psychiatric and psychological conditions and, more broadly, diagnostic challenges are addressed. The applications of actuarial risk assessment, as well as other approaches to risk assessment, are considered relative to offenders with intellectual disabilities. Finally, evidence-based interventions are considered, including both those designed to assist the offender-client in reducing his own risk and interventions to assist persons in the offenders’ “risk management circle.”

Persons with intellectual disabilities represent a unique and important subgroup of sexual offenders. There is evidence that such individuals commit sexual offenses at rates higher than sex offenders without intellectual disadvantage or delay. Available research indicates that there are both important similarities and differences between the sex offender with intellectual disability and other neurotypical sex offenders. As a result, there can be complex issues associated with assessment, diagnosis, and risk management with persons with intellectual disabilities who have sexually offended. The purpose of this chapter is to discuss characteristics associated with intellectual disabilities and those who have sexually offended, address modifications needed in the assessment process including adapting interviewing and testing techniques and procedures selected for lower cognitive levels of functioning. In addition, issues related to comorbid psychiatric and psychological conditions and, more broadly, diagnostic challenges are addressed. The applications of actuarial risk assessment, as well as other approaches to risk assessment, are considered relative to offenders with intellectual disabilities. Finally, evidence-based interventions are considered, including both those designed to assist the offender-client in reducing his own risk and interventions to assist persons in the offenders’ “risk management circle.”

A primary area of concern for evaluators and treatment providers working with sexual offenders who have intellectual or other developmental disabilities is that of risk of reoffense. Evaluators are also concerned with other forms of aggression, mental health diagnosis, and treatment planning. Evaluating sexual offenders typically entails evaluating the degree of risk of what behavior occurring under what circumstances or contexts. Evaluators attempt to discern what the individual’s needs are and how those relate to his potential for reoffense. Finally, evaluators also look into how to implement treatment and supervision in order to gain the maximum therapeutic benefit for both the individual being treated and the people in the individuals’ proximity.

Unfortunately, no singular characteristic or trait is so strongly correlated with reoffense that it alone can be relied upon to ascertain an individual person’s risk of sexual reoffense (Hanson & Morton-Bourgon, 2007). This makes comprehensive, holistic assessments necessary (Beech, Fisher, & Thornton, 2003; Blasingame, 2005). This is true for neurotypical as well as intellectually disabled sexual offenders.

Definitions and Characteristics Associated with Sexual Offenders with Intellectual Disabilities

Researchers unfortunately do not always use the same criteria or definition of intellectual disability or cognitive impairment. Some studies include individuals with full-scale IQs up to 80, while other studies do not (Crocker, Cote, Toupin, & St-Onge, 2007; Lindsay, Hastings, Griffiths, & Hayes, 2007). It is commonly understood that two individuals with the same FSIQ score will have different strengths and weaknesses. Clinicians often take liberties regarding which clients they describe as developmentally or learning disabled. This affects how they manage treatment planning for various individuals, based on the effects of chronic mental illnesses, general learning disabilities, borderline intellectual functioning, and illiteracy. However, with these diverse levels of cognitive abilities in different studies, it makes it difficult to compare various studies and findings.

For the purposes of this chapter, persons whose full-scale intellectual quotient (FSIQ) is 70 or below are considered to be intellectually disabled and those whose FSIQ is 71–84 are characterized by “borderline” intellectual functioning (APA, 2000). Borderline intellectual functioning implies the person is on the border between normal cognitive functioning and mental retardation. Many persons with borderline intellectual functioning appear to function in the normal range in some skill areas or may present as if they understand an evaluator’s vocabulary or meanings of words that they actually don’t understand. Upon further questioning, however, individuals with borderline intellectual functioning often demonstrate a lack of understanding and a tendency to acquiesce as if they understand when they do not. This may undermine assessment accuracy if not attended to or if measurement tools used do not consider this. Evaluators often recognize that many individuals whose FSIQ is between 71 and 84 would more readily benefit from assessment and treatment approaches similar to those commonly used for persons with intellectual disabilities.

By definition, intellectual disabilities (previously known as mental retardation in the USA and often referred to as learning disabled in the UK) have their onset before the age of 18 (APA, 2000). They have ongoing, lifelong effects on the individual. Having cognitive functioning two standard deviations below the mean of the population as well as functional impairment in two or more adaptive skills domains is an element of the formal APA diagnostic criteria.

Intellectual disabilities are themselves heterogeneous conditions. Two primary pathways are commonly identified as influential in development or appearance of intellectual disability. One is related to low functioning cultural-familial upbringings; the other is based on genetic and/or organic contributions (APA, 2000; Dykens, Hodapp, & Finucane, 2000; Holland, 2004). The impairing contributions from the genetic and organic etiologies undermine development of cognitive and social competencies, among others. Behavioral phenotypes have varying degrees of expression. There is no question that genes have influences on behavior in multiple, nonspecific ways (Dykens et al., 2000). There are hundreds of genetic etiologies for intellectual disability (Dykens et al., 2000) but with a similar outcome of impaired cognitive and social functioning and adaptive behavior deficits. Authorities agree that intellectual disability alone does not dismiss culpability for criminal or sexual conduct; it nonetheless affects court disposition, treatment planning, and risk management (Baroff, Gunn, & Hayes, 2004; Blasingame, 2005; Melton, Petrila, Poythress, & Slobogin, 1997).

Prevalence Issues

Intellectual disability occurs in a small percentage of the overall population. The diagnosis of intellectual disability requires the individual have an FSIQ two or more standard deviations below the population norm. The broader category of intellectual disability represents about 2.5 % of the general population (Kaufman & Lichtenberger, 2006). Those with mild mental retardation/intellectual disability comprise approximately 85 % of those individuals who have any level of intellectual disability (APA, 2000). Those with FSIQs of 71–84, those with borderline intellectual functioning, represent 14 % of the population.

A question is often raised regarding the co-occurrence of crime and intellectual disability and the prevalence of intellectually disabled persons in the criminal justice system (Crocker et al., 2007; Lindsay & Taylor, 2005). There is a large contingent of intellectually disabled persons among the criminal justice population, but percentages vary from setting to setting and state to state (Petersilia, 2000). Several studies report a strong link between low IQ and later delinquency (e.g., Babinski, Hartsough, & Lambert, 1999; White, Moffitt, & Silva, 1989). A study of Ireland’s prison inmate population found that 28 % of the inmates had intellectual disabilities/mental handicaps (Murphy, Harrold, Carey, & Mulrooney, 2000). The frequency in a Canadian study was approximately 20 % (Crocker et al., 2007). An international review found that the range was from 2 % to 40 % depending on varying methodological strategies and definitions (Jones, 2007). The fact that different studies use different methods of measurement or degrees of intellectual functioning makes it very difficult, if not impossible, to compare groups or studies.

An older study reported that 25 % of sex offenders who have intellectual disabilities also had histories of other violent behavior (Lund, 1990). Lindsay (2002) reported that of 62 offenders with intellectual disabilities, 62 % had prior convictions or documented evidence of prior sex offenses. Other studies reported that community-based outpatient samples have a low frequency of serious crimes (Hayes, 1991). Again, however, the setting in which the study occurs has a significant relationship with such prior histories.

Although it is clear that there is an overrepresentation of persons with intellectual disabilities within the criminal justice system, it is not clear that people with intellectual disabilities as a group commit more crimes (Holland, 2004; Lindsay & Taylor, 2005). Holland describes filter points or decision points that affect who is criminally charged or otherwise held accountable. These include whether a criminal behavior is detected or identified, whether that behavior is reported to the authorities, whether law enforcement action follows the report or if it is dropped, whether the alleged offender is arrested, and whether the individual is charged, taken to court, and if he is found guilty. Studies investigating the percentage of offenders who have intellectual disabilities are challenged by differences in criteria for intellectual disability (as discussed above), undetected intellectual disabilities among the general criminal population, tolerance of victims due to the subject’s apparent disability, a victim’s credibility as a witness due to having her/his own disability issues, and whether the law enforcement community believes the alleged offender understood that what he was doing was actually a crime (Holland, 2004). These factors all influence how prevalence data is tracked and tallied.

Ward, Trigler, and Pfeiffer (2001) estimated that approximately 5 % of persons with intellectual and/or other developmental disability engage in some form of sexually inappropriate behavior. People who have developmental disabilities often do not face adjudication for their sexual misconduct; only 15 % of a community-based sample had ever experienced incarceration for their misconduct (Ward et al., 2001). This suggests that as many as 80 % of offenders with intellectual disabilities are never incarcerated for a particular sex offense. Thirty-seven percent of community-based programs in their survey do not serve those with sexual behavior problems. Per Ward et al., 81 % of the respondents indicated that services in their regions were inadequate to serve the people with developmental disabilities who have sexual behavior problems. This lack of community resources may have an impact on court decisions regarding disposition of cases.

Differentiating subtypes of sexual offenders who have intellectual disabilities is another complex issue. Offenders with intellectual disabilities known to have victimized a child may well have other forms of sexual misconduct and other types of victims (Blasingame, Abel, Jordan, & Wiegel, 2011; Heil, Ahlmeyer, & Simons, 2003; McGrath, Livingston, & Falk, 2007b). McGrath and colleagues found that 54 % of their sample of 153 adult males had a history of more than one type of sexual offense. Of the subjects in the McGrath et al. study, 27 % had assaulted adults, 27 % had male victims of child sexual abuse, 17 % had female child sexual abuse victims, nearly 12 % had committed incest, and 16 % were identified as noncontact sexual offenders.

Description of Sexual Offenders Who Have Intellectual Disabilities

There are a number of similarities between intellectually disabled and intellectually typical sexual offenders (Courtney, Rose, & Mason, 2006; Crocker et al., 2007; Haaven & Coleman, 2000; Haaven & Schlank, 2001; Kalal, Nezu, Nezu, & McGuffin, 1999; Leonard, Shanahan, & Hillery, 2005; Lindsay, Elliot, & Astell, 2004; Quinsey, 2004). Common characteristics among offenders include having poor social support, attitudes supportive of sexual abuse, antisocial lifestyles, poor self-regulation and self-management, poor cooperation with supervision and treatment, and increased anger and stress prior to reoffending (Lindsay et al., 2004). Negative problem-solving strategies and poor skills are correlated with sexual deviancy among persons with impaired as well as normal intellectual functioning (Nezu, Nezu, Dudek, Peacock, & Stoll, 2005). The correlations of several of these characteristics with risk for sexual reoffense are also known (Hanson, 1997; Hanson & Bussiere, 1996; Hanson & Morton-Bourgon, 2007), but no singular characteristic is predictive of reoffense (Table 1).

Table 1 Characteristics of sexual offenders with and without intellectual disabilities

There are many additional needs and challenges for persons who have intellectual disabilities. Low cognitive functioning impairs the person’s ability to manage information, formulate concepts, create internal cognitive scripts, and absorb information. Low cognitive functioning, as measured by intelligence or achievement tests, indicates inadequate vocabularies; slower performance on timed skills tasks; limited working memory, i.e., how much information the person can manage at a given moment in time; and a slower speed for processing the information at hand. This also causes deficiencies in recalling information, i.e., short-term and/or long-term memory skills are often impaired. These challenges make it difficult to take in information in the short term as well as impair the person’s ability to transfer short-term memory to long-term memory. Later retrieval of information from memory is compromised. Many children with learning disabilities attempt to please the authority figures that may be talking to them or asking them questions. These cognitive impairments make these individuals susceptible to leading questions. Others perhaps admit behaviors they actually did not commit out of the belief that they do not always understand things yet they assume the authority figure knows and they trustingly go along with the leading questions, as they want to please authority figures. This is a problematic area for law enforcement and mental health evaluators.

Other challenges come in the form of socialization constrictions. Individuals who have intellectual disabilities often have poor social skills or inappropriate boundaries. As such, their parents can be hypervigilant in their supervision of the children. This functions to protect the child from the consequences of his actions, which would typically be part of the learning process. Without social feedback about one’s behavior or not having the ability to learn from the consequences of one’s actions, an individual may not have sufficient social learning experiences. This can impair the process of internalizing social boundaries. In addition, children who are so closely supervised often do not have the opportunity to experiment with social-sexual behaviors such as flirting, holding hands, kissing, or making out with a girlfriend or boyfriend. While close supervision is helpful in many ways, it can impair the opportunity for these children to learn through normal experimentation. This can contribute to the child getting the sense that such behaviors, that are otherwise natural and normal, are not acceptable to the parent and the child may learn to sneak about or take advantage of clandestine opportunities to experiment with these behaviors. Given that many people with intellectual disabilities are socialized with younger family members or children, their biological urges for sexual experimentation may be acted upon with the younger mental age-mates rather than their chronological age-mates (Blasingame, 2005).

Another significant challenge for people with intellectual disabilities is their deficits in communication skills. The poor vocabulary, social understanding, and information processing skills impair these individuals’ abilities to formulate ideas and to express those ideas. This impairs their ability to identify internal emotions and conceptualize subjective experiences and needs. The inability to communicate one’s needs along with poor coping skills often leads to maladaptive behaviors in attempt to meet one’s normal needs. These and a variety of medical, mental health, and behavioral problems are well documented (Alloy, Jacobsen, & Acocella, 1999; APA, 2000; Blasingame, 2005; Dykens et al., 2000; Fletcher, Loschen, Stavrakaki, & First, 2007; Haaven & Schlank, 2001; Seghorn & Ball, 2000; Sherak, 2000).

Parental and/or caretaker limitations, perceptions, and frustrations may further complicate family life, sexuality education, and socialization opportunities for some individuals who have intellectual disabilities (Blasingame, 2005; Lund, 1992). These several factors complicate the decision-making process regarding distinctions made between sexually offensive behavior, sexual offending behavior, and sexual deviance (Blasingame, 2005).

People who have intellectual or other developmental disabilities are reported to have two to four times greater risk of developing any comorbid psychiatric disorder (APA, 2000; Fletcher et al., 2007). Mood disorders, attention deficit hyperactivity disorder, conduct disorder, and movement disorders are common comorbid conditions albeit potentially with atypical clinical presentations (APA, 2000; Fletcher et al., 2007; Hurley, 2006). One of the challenges in making mental health diagnoses is that individuals who experience intellectual disabilities may well lack the ability to verbalize complaints regarding psychological symptoms, internal feeling states, emotions, social distress, or identify symptoms (Fletcher et al., 2007). Interviews with family members or other knowledgeable informants are helpful in discerning the presence of such symptoms.

People with intellectual disabilities and comorbid mental illnesses are at higher level of risk for reoffending criminally, particularly if they experience a major mental disorder or substance abuse history (Smith & O’Brien, 2004). Day (1997) reported 25–33 % of intellectually disabled sex offenders are dually diagnosed. Likewise, Lindsay et al. (2002) reported that 32 % of their sample of 62 offenders had a diagnosis of a major mental disorder. In another study, Smith, Quinn, and Lindsay (2000) reported that from a sample of 153 individuals with intellectual disabilities who had sexually acted out, 22 % were diagnosed with a significant mental illness, 12 % with a mood disorder, and 10 % with schizophrenia. These frequencies are in greater proportions than in the general population.

Personality disorders and maladaptive personality traits are also relatively common among individuals who have intellectual disabilities (Lindsay et al., 2007). The frequency of personality disorder diagnosis again varies by setting and methodology for making the diagnosis. Some studies have identified a large percentage of individuals with intellectual disability as having traits of personality disorders (e.g., maladaptive personality traits). Goldberg, Gitta, and Puddephatt (1995) found traits in 57 % of their sample of institutionalized individuals and 91 % of those in a community setting. By contrast, Flynn, Matthews, and Hollins (2002) found 92 % of their sample of 36 subjects who had severe behavior problems were diagnosed with a personality disorder. Setting and context of evaluation appear to have a significant influence on whether personality disorders are diagnosed. Unfortunately, the wide variations in prevalence reported in the literature make it difficult to generalize the findings.

In a study of 164 cases that have intellectual disabilities from three forensic settings, Lindsay et al. (2007) found a preponderance of antisocial personality disorder but very low percentages of other personality disorders. Thirty-nine percent of these forensic subjects were diagnosed with a personality disorder; no correlation was found between FSIQ level and personality disorder diagnosis. Additionally, Lindsay et al. found through confirmatory factor analysis that there were two primary factor loadings. The first was referred to as acting out; the second was avoidant/rumination/inhibited. These factors crossed over several personality disorder diagnostic categories and when combined made up 37 % of the variance between diagnostic groupings. It should be noted that this study did not provide separate information regarding subtypes of forensic subjects, i.e., they did not separate sex offender data from violent offenders.

A more severe form of personality disorder is that of psychopathy. The construct of psychopathy is valid among criminal offenders who have intellectual disability (Morrissey, Mooney, Hogue, Lindsay, & Taylor, 2007). However, psychopathy may present differently in offenders who are intellectually disabled, due to their communication and adaptive functioning challenges as discussed above. These individuals may present behaviors that are poor matches for the descriptors used in the PCL-R coding manual. Prevalence studies report that approximately 10 % of criminal offenders have elevated traits of psychopathy (Morrissey et al., 2007). The prevalence rates vary from setting to setting, similar to variations in diagnoses of other mental health disorders discussed above. Morrissey et al. further suggest that while some of the characteristics of psychopathy overlap with characteristics of intellectual disability, such as impulsivity or lack of empathic awareness, there remains strong evidence of a percentage of offenders who have intellectual disabilities who are also psychopathic.

Characteristics of psychopathy that are particularly relevant to offenders who have intellectual disabilities are strongly associated with Factor 1 on the PCL-R (Hare, 1991). These are glibness/superficial charm, inflated self-esteem, pathological lying, conning and manipulation, lack of remorse or guilt, shallow affect, lack of empathy or callousness, and failure to take responsibility. Morrissey et al. (2007) reported that higher scores on the PCL-R Factor 1 scale correlates with problems during treatment such as moves to higher levels of supervision and termination from treatment programs.

The sexual knowledge base of persons who have intellectual disabilities and sexual behavior problems is heterogeneous. For example, Lunsky, Frijters, Griffiths, Watson, and Williston (2007) found that among intellectually disabled persons who were evaluated in association with sexual misconduct, there was a significant variation in the amount of sexual knowledge held by the offenders. Those who committed offenses that are more serious had much higher than expected sexual knowledge, while those who committed offenses such as inappropriate touching or public masturbation were less informed about sexuality. Such variation calls for individualized assessment, diagnosis, and treatment planning.

Although it has been postulated that persons with intellectual disabilities who commit sexual offenses did so with motivations other than sexual deviance, the counterfeit deviance theory (Hingsburger, Griffiths, & Quinsey, 1991) has met challenges in recent years. The counterfeit deviance theory postulated that while clearly inappropriate, some of these individuals’ sexual misconduct is due to characteristics associated with their disabling conditions. Hingsburger et al. (1991) proposed 11 counterfeit deviance hypotheses. Their theoretical constructs are as follows:

• The Structural Hypothesis: The system of care these individuals are required to live in may have failed to address the needs for sexual expression. The iatrogenic effect of such restricted environments could cause these individuals to seek opportunities in settings or with individuals that are inappropriate.

• The Modeling Hypothesis: Some sexual behaviors are misguided repetitions or reenactments of “caring behaviors” previously done by family members or staff persons. These might include clients being naked in front of staff for hygienic/bathing purposes. The individual may later fail to discriminate with whom such behavior is appropriate.

• The Behavioral Hypothesis: Some sexual behaviors may be mechanisms to gain attention from family or care home staff persons. Some clients learn that one way to get attention is through negative behaviors, including sexually oriented behaviors. There is significant positive reinforcement (attention) for committing such behaviors.

• The Partner Selection Hypothesis: Because many people with developmental disabilities are not afforded age-appropriate opportunities to develop fulfilling relationships, they may well seek out opportunities to relate to staff or an available child in attempt to develop intimate relationships.

• The Inappropriate Courtship Hypothesis: Lacking the interpersonal skills needed to move through the stages of relationship development and unable to discern the nuances of private versus public behaviors, some persons with developmental disabilities may become too aggressive in their pursuits of personal friendships and relationships.

• The Sexual Knowledge Hypothesis: Given their problems with learning through subtle social learning experiences, some people with developmental disabilities have sexual knowledge deficits. When they are afforded sex education, it is often in the context of biology and body parts rather than in the context of social relationships, appropriate consent, and self-control. Some cases involve individuals being given too much information to process, exciting excessive curiosity.

• The Perpetual Arousal Hypothesis: Some persons with developmental disabilities appear to be perpetually aroused due to their inability to fulfill their sexual needs in a normal way, or they may not have the knowledge or skills required for achieving orgasm.

• The Learning History Hypothesis: People with developmental disabilities brought up in overly protective homes or nonnormative environments may not have normal learning opportunities. This would also include being abused or those who lack socialization opportunities.

• The Moral Vacuum Hypothesis: People with developmental handicaps may not comprehend the effects of their behavior on others. As such they may not realize their behavior can inflict pain or discomfort on others.

• The Medical Hypothesis: Some individuals may not realize that particular symptoms are the result of a medical condition that needs attention. For example, scratching one’s genitals, albeit inappropriate in social settings, may be indicative of an infection rather than an attempt at masturbation.

• The Medication Side Effect Hypothesis: Many people with developmental disabilities take psychotropic medications. Some people experience side effects such as inhibited sexual desire or diminished ability to achieve orgasm. If these side effects are not effectively explained to the individual or caregiver, sexual dysfunction or behavioral issues might result.

Hingsburger et al. (1991) recommended interventions such as system modification, policy modification, better sex education, staff education and training, and provision of counseling to the individual to address the problems. From this perspective, the client is not considered criminally culpable for his or her actions that are otherwise sexually inappropriate. Rather, the care delivery system is often considered the source of the problem and/or challenged to make corrective adjustments.

Hingsburger et al. (1991) also defined what they refer to as a hypothesis of “real deviance” for individuals with developmental disabilities.

• Benign Paraphilia: Unusual sexual behaviors typically done in private and not dangerous to oneself or others. Hingsburger et al. suggest that stealing underwear for masturbation is an example of a benign paraphilia. If this were done by a noninstitutionalized or unsupervised person, it would not be known about, e.g., it would be benign. It would have no victim and not be offensive to others.

• Offensive Paraphilia: Offensive paraphilia is a sexual arousal in circumstances or behaving in ways that are harmful to others or are offensive, such as an adult sexually interacting with a child.

• Hypersexuality: Ruminating about sexual themes or sexual acts may leave the individual feeling controlled by sexual obsessions and may interfere with daily life. Hingsburger et al. suggested this type of rumination may be an indication of a physical hypersexuality and may need management with hormone therapy.

Recent research, however, found that different offenders with intellectual disabilities have different levels of sexual knowledge, as discussed above. The research provides mixed results as to whether the sexual knowledge hypothesis is a sufficient descriptor for offenders with intellectual disability. Talbot and Langdon (2006) found that intellectually disabled sexual offenders had higher levels of sexual knowledge than their non-offender control group. Michie, Lindsay, Martin, and Grieve (2006) made similar findings.

Another aspect of concern is in that of cognitive distortions. Lindsay, Whitefield, and Carson (2007) found that sexual offenders with intellectual disabilities held to a greater number of cognitive distortions supportive of sexual offending than did the control group of other persons with intellectual disabilities. In another study, Lindsay et al. (2006) found that those intellectually disabled offenders who abused children held a higher number of child-oriented cognitive distortions than the offenders who abused adults and vice versa. These findings strongly support that cognitive distortions are not only relevant to assessment and treatment of offenders with intellectual disabilities, they also suggest that some offenders with intellectual disabilities have the ability to differentiate their targets of sexual interest.

Recent theoretical developments have identified multiple routes or pathways that people take on their way to becoming sexual offenders (Ward & Hudson, 2000). Different goals and motivations drive the individual to follow different pathways to goal attainment. These pathways involve diverse behavioral scripts and patterns of cognitive distortions (Lindsay, Steptoe, & Beech, 2008). Some offenders may want to offend and make efforts to do so, while others are aware of their potential to offend but want to avoid offending. The former pathway is known as the approach pathway; the latter is known as the avoidance pathway. Offenders are also different in their degree of overt effort and planning invested in their acting out. For the approach-oriented offender, these strategies may include making direct, explicit efforts to achieve sexual conquests or thoughtlessly acting on habituated behavioral scripts that are consistent with sexual offending (Lindsay et al., 2008; Ward & Hudson, 2000). The avoidance-oriented offenders may make efforts to avoid offending that are simply inadequate and do not sufficiently interrupt the underlying propensity to sexually act out.

Keeling, Rose, and Beech (2006) investigated the application of the multiple pathways theoretical constructs with offenders who have intellectual disabilities. Their findings supported the use of the self-regulation model with this group of offenders. Keeling, Rose, and Beech found that more than 90 % of their sample of 16 subjects was classified in two of the four pathways. Thirty percent were classified as approach explicit and 62 % were classified as approach automatic. These findings suggest that there is little difference between mainstream and intellectually disabled offenders in regard to their pathways to offending. Lindsay et al. (2008) also found that over 90 % of the intellectually disabled sex offenders in their sample were classified in the approach pathways. These findings also suggest that those intellectually disabled offenders had intact self-regulation and control and may well have engaged in conscious planning. These findings further suggest that a large percentage of intellectually disabled sex offenders are not as naïve or simply impulsive as previously thought.

Blanchard et al. (1999) found a significant correlation between intellectual deficiencies and maternal age in relation to the prevalence of male-oriented pedophilia. School grade failure and/or assignment to a special education class was found to be a significant educational background variable among sexual offenders (Cantor et al., 2006). Offenders with intellectual disability are less discriminating in their victim selection and offenses than are their neurotypical counterparts (Rice, Harris, Lang, & Chaplin, 2008). These characteristics suggest there is likely a neurodevelopmental disruption that contributes to the onset of maladaptive sexual behaviors (Cantor et al., 2006; Rice et al., 2008).

Blanchard et al. (1999) found that mental retardation or lower intellectual functioning was correlated with a diagnosis of pedophilia. In a large study of adult male sexual offenders, Blanchard et al. report that the presence of lowered intellectual capacities decreased the likelihood of exclusive sexual interest in girls. They also found that maternal age at the birth of the child increased the likelihood of exclusive sexual interest in boys. When both of these characteristics were present, there was a greater likelihood of sexual interest in boys; when only one was present, that likelihood was lessened (Blanchard et al., 1999). Based on penile plethysmography findings, the Blanchard et al. data indicate that the victim selection choices made by child sexual abusers who had intellectual disabilities were not due simply to situational availability; their decisions were made because of relative sexual interest in children.

Recidivism studies in recent years have improved our abilities to identify several characteristics to target in treatment. Lindsay et al. (2004) studied 52 adult male offenders with intellectual disabilities. They included suspicion of reoffending in their investigation to attempt to capture those unreported reoffenses. Variables identified with reoffending and suspicions of reoffending were separately considered. Those variables associated with reoffense were antisocial attitudes, low self-esteem, lack of assertiveness, poor relationship with mother, allowances made by staff, staff complacency, poor response to treatment, and offenses involving violence (Lindsay et al., 2004).

In terms of variables related to suspicion of reoffending, Lindsay et al. (2004) reported somewhat different findings, with some overlap. They reported antisocial attitudes, attitudes tolerant of sexual crimes, denial of a crime, sexual abuse in childhood, low self-esteem, lack of assertiveness, low treatment motivation, erratic attendance, unexplained breaks from routine, deterioration in family attitudes, allowances made by staff, staff complacency, unplanned discharge, and poor response to treatment. While suspicion of reoffending may include persons who have indeed not reoffended, these characteristics can aid in identifying those offenders who are prone to persist in making poor decisions and putting themselves in situations where they will be scrutinized.

Base rates for sexual reoffense vary depending on the setting the subjects are in, i.e., institutional versus community-based treatment settings. Many studies comingle general criminal data on violent recidivism, including sexual recidivism, making the findings less useful when attempting to ascertain risk with a sexual offender from another setting (Phenix & Sreenivasan, 2009). Individuals in high security settings such as prisons and forensic hospitals have higher frequencies of prior sex crimes than those in community settings; these differences may well affect how outcome studies can be generalized. Lindsay (2004) reported that several older recidivism studies reported 30–70 % of general criminal recidivism among offenders who were intellectually disabled.

However, when looking specifically at sexual offenders with sexual recidivism, the data is more hopeful. Lindsay et al. (2002) reported on a sample of 48 sex offenders who had a 4 % reoffense rate in the first year of follow-up, 12 % reoffense at 2 years, and 13 % reoffense at 3 years. In their 11-year follow-up study, McGrath et al. (2007b) detected 11 recidivists, or approximately 11 %, who committed 20 new crimes. Eleven of those 20 new crimes involved noncontact crimes, and six of the victims were staff members. Finally, Tough (2001) reported a recidivism rate of 16 % in a sample of 76 treated sexual offenders when including informal file documentation as well as formal arrest and conviction records with up to a 19-year follow-up.

Criminal Investigation and Disposition of Sex Offenders with Intellectual Disabilities

Legal systems are involved in a significant number of cases with intellectually disabled sexual offenders. There are several steps along the way before an offender is found guilty or innocent, sentenced, and/or the case disposed of (Holland, 2004). Some studies have reported high frequencies of false confessions by intellectually impaired suspects, who believed they would be “allowed to go home” if they agreed with the police or otherwise acquiesced to the pressures of the interrogation by authority figures (Petersilia, 2000). The Miranda warning that is read to criminal suspects is estimated to require a seventh grade reading comprehension level, far above the comprehension level of an adult with a mild intellectual disability (Baroff et al., 2004; Petersilia, 2000). As much as some individuals with intellectual or other developmental disabilities try to present themselves as normal, it may be difficult for untrained law enforcement professionals to identify that the person is indeed intellectually impaired (Petersilia, 2000). Without such awareness, investigating officers may not make the necessary accommodations.

Once arrested, it is common for inmates with intellectual disabilities to have their cognitive impairments be undetected (Scheyett, Vaughn, Taylor, & Parish, 2009). Without identifying such impairments, these individuals cannot receive appropriate referrals to support agencies or advocates. Further, protective supervision in custody cannot be provided if correctional staff is unaware of the true level of functioning of the inmate. Inmate rights may well be compromised without such awareness by correctional staff (Petersilia, 2000; Scheyett et al., 2009). Given that the average FSIQ within the prison population tends to be lower than the mainstream population, those with somewhat lower intellectual functioning may not stand out to correctional staff (Hayes, 2007). Even though their needs may be similar to their non-impaired counterparts, the need to address those needs differently is clear (Crocker et al., 2007).

Many individuals who engage with the legal system are not formally prosecuted. Some are released as they are deemed not guilty or there is insufficient evidence to move the case forward at the prosecutor’s office. Others are charged but later found not guilty. Some are charged but diverted to community-based programming, such as mandated residential treatment through developmental disability service programs. Some are formally prosecuted and ordered to be supervised by probation or may be sent to prison or mental health facilities. It may not be an easier route for the intellectually disabled person to be diverted to a developmental center or psychiatric hospital, as those placements often turn out to be longer sentences than if they had done regular prison time (Hayes, 2007).

Many individuals who have intellectual or other developmental disabilities, such as autism spectrum disorders, have difficulty dealing with the investigation processes. Some are poor historians due to time frame distortions or have difficulty differentiating what was their own idea versus what someone suggested that they do. Others misinterpret questions and their own answers. Some are incautiously frank in their answers, are overly compliant with authority figures, and use words they do not fully understand (Allen et al., 2008).

Before a trial can take place, competency to stand trial must be determined. Competence to stand trial requires the defendant be able to aid his attorney in his own defense, understand the crime and consequences for the charges he is facing, and understand the roles of the participants in the court processes and the purpose of the trial (Baroff et al., 2004; Blasingame, 2005). If the person is found not competent to stand trial due to issues associated with intellectual disability, he may be incarcerated in a hospital or residential setting for training. The great majority of those with mild intellectual disability are found competent to stand trial (Baroff et al., 2004). In some cases, the charges may be dropped although the individual is still court ordered into a nonjudicial alternative such as care home placement or be placed under guardianship. In some cases, these individuals are trained to be competent and are then returned to court for continuation of the prosecution process.

Another issue facing the court is whether the defendant who has intellectual disabilities should be declared not guilty by reason of insanity (NGRI; Baroff et al., 2004). Those who are acquitted on the basis of NGRI are determined to have a mental disease or defect that decreases the person’s capacity to control their behavior (Melton et al., 1997). Those who are acquitted as NGRI are often ordered into hospital placements for longer sentences than they would have faced if they would have been sent to prison, so being declared NGRI may not be particularly advantageous (Blasingame, 2005; Salekin & Rogers, 2001). Individuals incarcerated in a psychiatric setting due to NGRI status are to receive treatment for the condition that diminished their capacity and be prepared to return to the community (Salekin & Rogers, 2001). In order to return to the community, a judicial hearing must take place and the hospitalization order be altered.

Courts have several legal options for sentencing an offender who has intellectual disability. One of those options is a diversion plan established by the regional service agency responsible for assisting individuals who have developmental disabilities. These types of plans often involve mandated placement in board and care facilities in the community, participation in approved treatment programming, specialized day programs or sheltered work settings, and being evaluated for psychotropic medications. Sometimes these plans also include incarceration in a state hospital or developmental center where the individual can be contained and treated for long periods of time.

Other sentencing options the courts have are to grant probation or to sentence the individual to prison. Individuals with intellectual disabilities are at increased risk of victimization when incarcerated in prison or institutions for persons with developmental disabilities (Haaven & Schlank, 2001). Many states now have civil commitment programs for sexual offenders who have completed their prison sentences. Those civil commitment programs have both mental health and correctional components (Haaven & Schlank, 2001). They nonetheless need to make adjustments to routines and protocols to accommodate the physical and psychosocial needs of those inmates who have intellectual disabilities. Inmate safety becomes a significant concern due to the risk of continued sexual acting out by the offender or his being victimized by other inmates or patients. Some civil commitment programs or institutions have been reported to attempt to include the intellectually disabled offenders in the general sex offender population and simplify the treatment curricula (Haaven & Schlank, 2001). This is inadequate in many ways, as the inmates with intellectual disabilities have a variety of learning problems that are very different from mainstream offenders, as discussed above. The author is aware of State institutions being subject to lawsuits for failure to make appropriate accommodations and adaptations in the programming for those patients who have intellectual disabilities.

Some court orders for probation also include the case management plan noted above in the context of a diversion plan. The intent of such an order is rehabilitation based on the assumption the individual can benefit from such a plan while also maintaining community safety. Collaborative management and treatment of these individuals appears to improve public safety through reduced recidivism (Hayes, 2004).

Community placement after incarceration or hospitalization, or as a diversion from custodial sentencing, requires thoughtful reentry planning, including addressing employment, housing, substance abuse treatment, mental health treatment, and of course sexual offender treatment. Ongoing support services are needed during institution to community transitions, including close supervision from parole or aftercare clinicians (Haaven & Schlank, 2001). Locating housing and appropriate services is both critical and difficult, as most communities do not have sufficient resources (Ward et al., 2001).

General Assessment Issues for Individuals with Intellectual Disabilities

Interviewing and testing individuals who have intellectual or other developmental disabilities has several inherent challenges. Evaluators need to be thoughtful about matching their vocabulary with these individuals to ensure accurate communication. The usual interview techniques of asking questions and expecting a relatively prompt response may not generate helpful information. Evaluators need to monitor their voice for suggestive tones, avoid leading questions, and slow the pace of the interview process. Providing multiple-choice options and the use of plain language are strongly recommended (Blasingame, 2005).

Acquiescence is a significant concern when evaluating individuals who have intellectual disabilities (Finlay & Lyons, 2002). Many of these individuals have a yea-saying response pattern, regardless of what is being asked. This may be associated with fears of disapproval from the evaluator, confusion about the questions being posed, feared consequences of disclosing certain information, or simply wanting to be compliant. Acquiescence should not be confused with dissimulation, lying, or socially desirable responding (Blasingame, 2005; Finlay & Lyons 2002).

For diagnosis, treatment planning, and case management purposes, measures of adaptive functioning should commonly be used in conjunction with intelligence tests to ascertain behavioral functioning levels, i.e., in what skill areas can the individual function independently and/or in what skill areas might the individual need supports and assistance (Fletcher et al., 2007). The Adaptive Behavior Assessment System, Second Edition (ABAS-II; Harrison & Oakland, 2003) and the Supports Intensity Scale (Thompson et al., 2004) are examples of standardized tools for this purpose. Such assessments are needed to discern environmental supports and contextual factors associated with risk management and intervention.

Intelligence tests are about information processing and accomplished learning. Intellectual functioning levels reveal information about one’s ability to learn, what he has learned, and the person’s adaptation to the environment (Kaufman & Lichtenberger, 2006). When assessing persons with intellectual disabilities, evaluators need to be cautious when interpreting results from instruments that have high demands of the executive functioning system, abstract thought, and information processing. Ultimately, to qualify for a diagnosis of mild intellectual disability, one has to obtain an intellectual quotient (IQ) score between 55 and 70, manifesting subaverage intelligence (American Psychiatric Association (APA), 2000; Fletcher et al., 2007). As mentioned earlier, this represents about 85 % of persons with intellectual disability. Individuals in this mild level of intellectual disability are said to be able to achieve about a sixth grade level of academic skills, significantly impacting treatment strategies and risk management efforts. Persons diagnosed with moderate intellectual disability, having IQ scores between 40 and 55, are said to be able to achieve about a second grade level of academic functioning (APA, 2000). To be diagnosed with intellectual disability, there needs to be evidence the person met or would have met the criteria prior to the age of 18 (APA, 2000).

A baseline risk assessment based on an actuarial procedure is very useful in treatment planning (Boer, Tough, & Haaven, 2004; Quinsey, 2004). However, the actuarial risk estimates (a) do not include all risk factors known in the literature and (b) do not discern the conditions under which that baseline risk may be increased or decreased dependent on a given offender in a given situation. The latter variables, also known as dynamic risk factors (Hanson, Harris, Scott, & Helmus, 2007), may be of keen interest to those performing risk assessments on individuals with intellectual disabilities particularly due to the issue of inaccurate or incomplete “official records” (Beech et al., 2003; Boer et al., 2004; Keeling et al., 2006; Tough, 2001). It is important to assess each individual in the primary areas associated with changes in risk, specifically self-management, socio-affective functioning, degree of deviant sexual interests, and pro-offending attitudes (Thornton, 2002). Therefore, holistic assessments for persons with intellectual disabilities should include not only the actuarially based risk estimate (discussed below) as a beginning baseline; they also should address the following additional areas:

• Psychiatric or mental health assessment

• Psychosexual assessment

• Psychosocial assessment

• Contextual assessment

Persons with intellectual disabilities are sometimes marginal to poor historians and reporters. Parents and/or other caretakers should participate in collateral clinical and/or psychosocial interviews when possible (Harrison & Oakland, 2003). Additional knowledgeable informants include care home staff members, previous treatment or care providers, teachers, physicians, probation officers, or service coordinators/case managers. Discussing the case history and current functioning with these informants may also help ascertain the reasonable veracity of any client testing or interview information.

Assessing Psychiatric or Mental Health Conditions Among Persons with Intellectual Disabilities

Individuals with intellectual disabilities have an increased risk for comorbid mental disorders, as discussed above (APA, 2000; Fletcher et al., 2007). The presence of these additional mental disorders may impact the individual’s capacity to self-regulate his or her behavior, meaningfully participate in an assessment or treatment process, and may or may not diminish his or her capacities to such an extent as to undermine culpability for their actions (Melton et al., 1997). These of course may also be relevant concerning the needs, risks, and responsivity factors when treatment planning (Andrews & Bonta, 2003).

Due to the diversity of issues and idiosyncratic nature of challenges faced by these individuals, mental health assessments are also complex but not overwhelmingly difficult. Comprehensive review of case records, client interviews, informant interviews, and, in some cases, psychological testing are called for (Blasingame, 2005; Fletcher et al., 2007; Hurley, 2006; Mikkelsen, 2004).

The types of file information needed for evaluative processes are fairly broad in range, including early social and behavioral histories, academic and school testing, school behavioral concerns, child protective services and/or police reports, court reports, civil commitment reports, leisure activities and recreational interests, index (present concern) incident reports, medication history, and treatment history (Blasingame, 2005). If the client has resided in institutional or residential care, summary information about in-home and out-of-home placement functioning is also helpful in determining any behavioral problems of a pervasive nature. Indeed, many intellectually disabled offenders’ victims may be other care home or institution residents. The evaluator needs to have access to this full range of background information in order to perform a comprehensive evaluation (Blasingame, 2005).

Mental health conditions among persons with intellectual disabilities may not easily yield themselves to the traditional diagnostic interview process (Fletcher et al., 2007; Hurley, 2006). In some diagnostic situations, it may be necessary to discern the presence of behavioral phenomena and consider these as “behavioral equivalents” in place of diagnostic criteria (Hurley, 2006). As such, maladaptive behavioral symptoms may be considered as substitutes for some criteria when reviewing diagnostic checklists, e.g., when completing a diagnosis based on the Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM 5; APA 2013), or the Diagnostic Manual-Intellectual Disability (DM-ID; Fletcher et al., 2007).

Mental disorders are neither static variables nor are they always associated with increased risk for sexual or violent reoffense. Nonetheless, it is important to recognize that having a mental disorder may undermine, at different points in time, the individual’s capacity to self-regulate and self-manage his behaviors, subsequently contributing to behavioral dysregulation and/or impulsivity.

Psychosexual Assessment of Persons with Intellectual Disability

Significant research points to the fact that the presence of deviant sexual arousal involving children or other sexual deviations is indicative of increased risk for sexual reoffense (Hanson & Bussiere, 1996). Hanson and Morton-Bourgon (2004) further identified indicators of current risk for reoffense to include, among others, sexual preoccupation, emotional identification with children, having any deviant sexual interests, and general self-regulation problems.

These findings strongly suggest that measurement for these factors is important in the assessment and risk management processes. Psychosexual variations are measured by a variety of means. These include the following:

• Penile plethysmography

• Attentional or viewing behavior measurements

• Self-report questionnaires

• Clinical interview

These methods each have its own strengths and weaknesses such that these should not be used independently or outside the clinical domain (Keeling, Beech, & Rose, 2007; Kalmus & Beech, 2005). It should not be expected that all sexual abusers would exhibit measurable preferences during phallometric or similar assessment (Reyes et al., 2006). Sexual arousal or sexual attraction assessment procedures should not be used independent of other sources of information gathering and clearly should not be used in the courtroom as evidence of guilt or innocence (Association for the Treatment of Sexual Abusers (ATSA), 1997).

Many child sexual abusers have what appear to be “normal” patterns of sexual interest under laboratory conditions (Reyes et al., 2006). Nonfamilial child sexual abusers with multiple victims are identified as more deviant during assessments compared to nonfamilial offenders with one victim or incest-only abusers. In fact, incest-only abusers are often nonresponsive to child stimuli in phallometric tests (Marshall, Anderson, & Fernandez, 1999). Marshall et al. note that nonfamilial child sexual abusers have the most consistent phallometric measurement results, i.e., they exhibit the most consistent sexual attraction for children. Nonetheless, there is significant variance in outcomes of phallometric assessments even within the subtypes of nonfamilial child sexual abusers, rapists, or exhibitionists (Kalmus & Beech, 2005; Reyes et al., 2006).

Not all individuals who sexually abuse children have deviant fantasies about children prior to the offense behavior (Marshall et al., 1999). Nonetheless, it is important to rule in or out the presence of sexual deviance, current sexual interest or arousal involving children, or other forms of sexual preoccupation if one is to assess current risk for reoffense and develop realistic intervention plans and programming to manage any current risk for reoffense.

A confounding issue regarding penile plethysmography (PPG) or the viewing time measures (discussed below) in assessing sexual abusers has to do with undisclosed offenses in the histories of the offender. Heil et al. (2003) summarize a number of studies as well as their own data which demonstrates that a majority of sexual offenders have fluid interests and victim types, i.e., offenses that cross over the lines of age and gender differing from the victim of record. Official records often do not have complete information about other, non-adjudicated victims such as other adults, children, or animals (Blasingame, 2005; Heil et al., 2003). Therefore, it should be no surprise that the tools attempting to assess sexual preferences and interests will often be perceived as falling short given that sexual abusers’ interest patterns are not as discriminating or as stable as had previously been assumed.

Penile plethysmography is used with intellectually disabled males and is effective in ascertaining sexual arousal patterns (Haaven & Coleman, 2000; Haaven, Little, & Petre-Miller, 1990; Haaven & Schlank, 2001; Hingsburger et al., 1999; Reyes et al., 2006; Seghorn & Ball, 2000). Programs using it adjust the administration protocol to allow more adaptation time between stimuli and making sure the client understands the instructions. There is no research regarding the lower limits of intellectual functioning for use of PPG with persons with intellectual disabilities. Some programs perform pre- and posttreatment phallometric assessments to assess change of arousal patterns over time.

In a PPG study involving adult male sex offenders with developmental disabilities, Reyes et al. (2006) found three response patterns. While their sample size was only ten, they found one subset of sexual abusers who showed distinct differentiated arousal to deviance, another subset who showed undifferentiated deviant arousal, and the third subset of sexual abusers who showed no measured deviant arousal. Differentiated deviant arousal was defined as measured arousal to the presence of a specific gender or age, and this was measured higher than neutral stimuli or other categories. Undifferentiated deviant arousal was defined as measured arousal to deviant and nondeviant stimuli at a higher level than arousal to neutral stimuli. The nondeviant subset of offenders did not demonstrate measured arousal to deviant stimuli and was at a comparable level with neutral stimuli (Reyes et al., 2006). The subjects in this study all had child victims; the data indicates that not all intellectually disabled sex offenders with child sex abuse victims will exhibit measurable sexual interest/arousal in the testing situation despite their known history of abusive behaviors.

Another method for attempting to measure the sexual interests of abusers is that of viewing behavior measures. These have also been called stimulus viewing time, viewing time, and visual reaction time. Abel, Huffman, Warberg, and Holland (1998) reported high reliability and validity comparing the Abel Assessment for Sexual Interest visual reaction time (VRT) assessment with penile plethysmography. Others have also found the Abel Assessment for Sexual Interest (AASI) outcome comparable to penile plethysmography outcome data (Johnson & Listiak, 1999; Letourneau, 2002).

One viewing time instrument that has been used in the evaluation of individuals with intellectual disabilities or other developmental disabilities is the Abel-Blasingame Assessment System for individuals with intellectual disabilities (ABID; Abel & Blasingame, 2005). The ABID was adapted from the Abel Assessment for Sexual Interest and is designed specifically for the evaluation of individuals with intellectual disabilities or borderline intellectual functioning. The ABID has two components: a self-report questionnaire administered by the clinician as a semi-structured interview and an objective measure of sexual interest, using visual reaction time. The design of the ABID questionnaire intends that the evaluator will read the questions aloud to the test subject and record the responses on the computer. The questionnaire requires only a second grade level of reading comprehension for the questions.

The ABID questionnaire inquires about involvement in 16 problematic sexual behaviors or potential paraphilic sexual behaviors. In addition, the questionnaire component of the ABID is of value to evaluators as it includes endorsement of sexual fantasy vignettes, measurement of cognitive distortions, social desirability, alcohol and drug history, social-sexual history, and additional self-report information. The ABID contains a number of concrete visual aids, scheduled breaks, and instructions to facilitate transitioning between content topics. The stimuli used in the objective measure of sexual interest are the same as used with the AASI; however, the instructions have been simplified. The VRT portion of the ABID assesses sexual interest in several age categories, including preschool, grade school, adolescent, and adult in both males and females. These components combine as a system of assessment rather than the seemingly more popular focus on the viewing component alone.

The self-report data collected on the questionnaire portion of the Abel-Blasingame Assessment System for individuals with intellectual disabilities (Abel & Blasingame, 2005) indicates a significant amount of sexual behavior crossover between age of victims, gender of victims, and a variety of other potential paraphilic behaviors. The preliminary data, based on a sample of 495 adult males, on the utility of the ABID questionnaire is very promising as it solicits significant information from subjects in sixteen areas of sexual history. ABID data evidences significant increases in the number and types of sexual misconduct admitted in the assessment processes of the ABID; increased reporting of paraphilic behaviors, types of victims, and types of sexual misconduct improve the clinicians’ ability to discern clinical and protective supervision needs of these offenders. Preliminary data analyses indicated that there was an average of about a 50 % increase in disclosure of the 16 problematic sexual behaviors as a result of the administering the ABID, with the more significant disclosures found among domains involving behaviors done in secret. Preliminary analyses of the fantasy vignette card sort found that endorsement of the female child sexual fantasies correlated (r = .34) with the number of self-reported child victims. Endorsement of the male child sexual fantasies also correlated (r = .33) with the number of self-reported child victims (Blasingame et al., 2011).

Viewing time measurements, as well as plethysmography, are confronted with the issue of crossover interests and behaviors among sexual offenders. The sexual interests of sexual offenders who have intellectual disabilities appear to be fluid and are often undifferentiated. Other times their interests are fixed and clearly defined. Additionally, offenders often use measures to try to “beat the test” including socially desirable responding and purposeful means of preventing the evaluator from discerning the person’s deviant sexual interests or preferences. Using the visual reaction time portion alone of the ABID should not/will not necessarily discriminate between perpetrators referred for evaluation of child sexual abuse as opposed to those who are identified as having adult victims due to the fluidity of interest measured and reported on the ABID (Blasingame et al., 2011).

Other card sorts and similar questionnaires seek to discern the presence of sexual interests, fantasies, and/or preferences by presenting the individual with information and soliciting responses associated with the degree of sexual desirability or interest the person may have in the stimulus.

The Multiphasic Sex Inventory (MSI; Nichols & Molander, 1984) has 560 items on 20 scales but requires a seventh grade reading ability (Keeling et al., 2006). Scales include child molest, rape, fetish, voyeurism, social-sexual desirability, cognitive distortion, immaturity, and others. The MSI is able to detect faking by subjects and level of denial (Kalmus & Beech, 2005).

The Questionnaire of Attitudes Consistent with Sexual Offending (QACSO; Broxholme & Lindsay, 2003; Lindsay et al., 2006) measures cognitive distortions supportive of sexual offending among intellectually disabled males. The QACSO surveys the offender’s attitudes regarding rape, voyeurism, exhibitionism, dating abuse, stalking, homosexual assault, and sex with children (Lindsay & Taylor, 2005). It is reported to be able to discriminate intellectually disabled offenders from non-offenders as well as non-offenders without intellectual disability (Keeling et al., 2006; Lindsay et al., 2007).

While measurements of cognitive distortions, fantasies, and attitudes have the risk of transparency, these tools have demonstrated adequate effectiveness and should be incorporated within the evaluation process. A combination of these tools is recommended to assist in ascertaining the psychosexual histories and interests of individuals with intellectual or other developmental disabilities who have sexual behavior problems. Discerning these potential risk areas is critical in the later development of risk management strategies.

Psychosocial and Contextual Variables Associated with Assessment and Risk Management

Self-report information gained during clinical interviews with intellectually disabled offenders can be helpful. Many individuals with intellectual disabilities referred for treatment for sexual behavior problems are sufficiently motivated to make disclosures regarding their sexual histories and interests in an effort to engage with an evaluator. While this information may be held with some degree of question, it may be possible to ascertain the veracity of the information through review with knowledgeable informants or by reviewing the person’s file information to discern the degree of credibility to attribute to the self-report information. This self-report information may be helpful in determining the presence of additional factors not included in the actuarial or other procedures administered. Given the low level of accuracy of assessments based on clinical judgment (Hanson & Bussiere, 1996), reliance on self-report information alone, such as that from a psychosocial interview, is counter indicated.

Assessment of persons with intellectual disabilities who have sexual behavior problems may require consideration of a number of idiographic factors associated with the offender himself and the environment he lives in (Boer, McVilly, & Lambrick, 2007). Within this paradigm, staff members and other care providers become part of the risk management equation (Boer et al., 2004).

Boer et al. (2004) outlined a number of contextual, dynamic risk management variables that need to be assessed in the course of ascertaining the degree of risk an individual presents at a given time. These include the following:

• Staff member attitudes toward intellectually disabled sex offenders and the degree of effort they are willing to put forth to understand them

• Communications among supervising staff persons including care providers, therapists, and case managers

• Client-specific knowledge retained by supervisory staff persons

• Consistency of and between supervisory staff members

• Consistency of the environment and environmental changes

• The presence of new staff members and changes in support system

• The degree of monitoring of the offender by staff members

• The degree or opportunity for victim access

• Offender compliance with supervision and attitude toward supervision and treatment

• Offender knowledge of his own problem thinking, crime history, risk factors, and relapse prevention plan

• Offender sexual knowledge and self-regulation of sexuality and degree of sexual preoccupation

• Offender capacity to manage impulses, cope with change, and manage emotions (Boer et al., 2004)

These dynamic factors can be discerned by observing and interviewing care providing staff members, family members, and the offender-client himself in pursuit of this information.

The Treatment Intervention and Progress Scale for Sexual Abusers with Intellectual Disability (TIPS-ID; McGrath, Livingston, & Falk, 2007a) is another structured approach to gathering dynamic variable information that is well documented. The TIPS-ID has 25 factors that are examiner-scored on a four-point scale of zero to three. Items rated involve domains such as sexual knowledge, attitudes, and behaviors; criminality; mental health and substance abuse; social influences; cooperation with treatment and supervision; and risk management application. The TIPS-ID serves as a structured approach when evaluating dynamic, changeable characteristics within the individuals’ psychosocial and contextual environment.

Actuarial Risk Assessment of Sex Offenders with Intellectual Disabilities

Risk assessment of sexual offenders with intellectual disabilities is approached by multiple strategies: actuarial assessment, structured clinical assessment, or by use of a combination of these (Blasingame, 2005; Boer et al., 2004). Quinsey (2004) has noted that while intellectual disability is an associated risk factor for general antisocial behavior and pedophilia, it does not correlate in and of itself with general sexual reoffense.

Actuarial assessment focuses on a limited number of clear and distinct factors or offender characteristics identified from the research literature. Completing actuarial ratings typically relies on official records, e.g., from official criminal records. Actuarial instruments yield a specific score so the offender can be compared to other offenders with similar histories and a comparison of reoffense rates can be done. While the accuracy of actuarial risk assessments regarding groups of persons with a similar score is very high, the application of group-based actuarial ratings to individuals has been challenged (Hart, Michie, & Cooke, 2007). This is likely exacerbated when discussing individuals with intellectual or other developmental disabilities. The use of actuarial estimates is nonetheless recommended in the formation of baseline risk assessments, particularly if used in the larger context of the evaluative process, i.e., not as stand-alone assessment procedures (Blasingame, 2005; Boer et al., 2004; Hart et al., 2007; Tough, 2001). Structured clinical assessments are more idiosyncratic and contextual; these were discussed earlier in this chapter.

Actuarial procedures rely on a limited number of variables associated with reoffense that are delineated in advance of the assessment (Boer et al., 2004). There are limitations to these tools; they do not encompass every risk factor that is identified in the research literature (Quinsey, 2004). Nonetheless, they are well researched and offer validation data to support their use. Actuarial tools offer a baseline risk rating that can aid in determining an individual’s needed level of case management, supervision, and/or treatment intensity (Boer et al., 2004; Hart et al., 2007).

Actuarial tools validated on samples of males with intellectual disabilities who had committed sexual offenses include the following:

• The Violence Risk Assessment Guide (VRAG; Quinsey, Harris, Rice, & Cormier, 1998; Rice et al., 2008)

• The Static-99 Structured Risk Assessment (Hanson & Thornton, 1999; Tough, 2001)

• The Rapid Risk Assessment of Sex Offender Recidivism (RRASOR; Hanson, 1997; Tough, 2001)

The Violence Risk Assessment Guide (Quinsey et al., 1998) estimates long-term risk for violent and/or sexual recidivism. It has been cross-validated on a variety of offender types, including forensic psychiatric and correctional facility populations and offenders with intellectual disabilities. The VRAG includes 12 domains, including psychopathy measured by use of the Psychopathy Checklist-Revised (PCL-R; Hare, 1991). Due to the use of the PCL-R, the VRAG is somewhat more complicated to use with persons with intellectual disabilities as the PCL-R relies significantly on information in the person’s file. Since a significant number of intellectually disabled persons referred for sexual behavior problems have poor documentation and/or no official charges in their file information, it may not be possible to complete the PCL-R (Quinsey, 2004). Nonetheless, in the original development samples for the VRAG, it was as accurate with persons with intellectual disabilities as it was with intellectually typical individuals (Quinsey et al., 1998).

PCL-R trait scores are reported to correlate with IQ scores among sex offenders who have normal intellectual functioning (Beggs & Grace, 2008). PCL-R scores were found to correlate (r = .18; p < .01) with prior sexual offenses and with reoffending (r = .25; p < .01) although the IQ scores did not. Beggs and Grace report that higher PCL-R trait scores are associated with increased risk of reoffending and that there is an interaction between PCL-R score and IQ. This particular study did not use the traditional cut-point of 25 or 30 for PCL-R scores; the authors considered scores of 12–15 as high. Beggs and Grace found the “high” PCL-R /lower IQ group to have the highest reoffense rate. Given the low level of psychopathy reported in the study, it may be more accurate to refer to these phenomena as antisociality rather than psychopathy. These findings suggest that evaluators can strengthen their assessments by including the PCL-R in the assessment regimen and that there should be raised concern if there are more modest PCL-R scores in combination with lower intellectual functioning.

Another advance in the development of risk assessment procedures is the Static-99 (Hanson & Thornton, 1999). The research included two large meta-analyses of sexual offender recidivism studies. On the Static-99, points are assigned based on the presence of several factors. These are prior offense convictions, age over/under 25, male gendered victims, the presence or absence of a relationship with the victim, prior nonsexual crimes, offense of immediate concern relating to nonsexual violence, having stranger victims, length and presence of marital/relational status, and the total number of prior sentencing dates. Individual cases are compared to the frequency of recidivism known among groups of individuals with similar ratings.

The Static-99 was slightly more accurate than the RRASOR (discussed below; Hanson, 1997) in classifying risk categories among the general sexual offender population (Hanson & Thornton, 1999). The Static-99 requires access to criminal justice documentation such as police or court records. For a variety of reasons, many persons with developmental disabilities that are referred for evaluation and treatment of sexual behavior problems have no formal criminal record (Beech et al., 2003). There does not appear to be an interaction of Static-99 scores and IQ (Beggs & Grace, 2008).

Harris, Phenix, Hanson, and Thornton (2003) note that the original data samples for the Static-99 included developmentally delayed offenders. They indicate that research to date supports the utility of the Static-99 with the developmentally delayed population, and where formal legal documentation does not exist, the use of documentation from informal hearings and sanctions such as placement in treatment facilities and residential moves are counted as both a charge and a conviction for a sexual offense. When such documents are available, the Static-99 is a useful tool in classifying levels of risk for reoffense among intellectually disabled offenders.

The Rapid Risk Assessment of Sex Offender Recidivism (RRASOR; Hanson, 1997) is another tool useful for sexual offenders with intellectual disabilities. Tough (2001) found the RRASOR to provide a good estimate of overall risk for recidivism among intellectually disabled sex offenders.

The RRASOR consists of only four items: prior history of sexual convictions, age of the offender at the time of the RRASOR assessment, victim(s’) gender, and the offenders’ relationship to the victim. The coding rules for the RRASOR can be modified, as discussed above regarding the Static-99, to overcome the fact that many offenses committed by intellectually disabled persons are not reported to law enforcement or that the legal system may dismiss the charges due to the individual having a developmental disability (Harris et al., 2003; Keeling et al., 2006; Tough, 2001). In her study, Tough used institutional and counseling center records in addition to official legal system records in scoring the RRASOR. Tough suggests the RRASOR to be the more appropriate tool for use with the intellectually disabled population of sexual offenders as it has fewer items than the Static-99. She suggested that her finding that the RRASOR more accurately classified risk for sexual reoffense than the Static-99 may be related in part to the fact that the additional six items on the Static-99 may be absent or poorly documented in intellectually disabled individuals’ charts, given the documentation problems noted above. By using institutional and/or clinical records that include information regarding what would otherwise have been a matter brought to the attention of the criminal justice system, except that the alleged perpetrator was an individual with intellectual or other developmental disability, Tough found that the subjects’ risk estimate scores were indeed increased as was the overall accuracy of the RRASOR.

Evaluators must recognize the weight of their opinion and how these opinions influence restriction of civil liberties (Blasingame, 2005; Hart et al., 2007). Further, it is understood that given a number of documentation challenges and complications, an individuals’ risk estimate may be an underestimate (Keeling et al., 2006). Nonetheless, the current data on these actuarial tools does support their use in assessment and risk management with those sexual offenders who have intellectual disabilities (Phenix & Sreenivasan, 2009).

Actuarial tools offer a baseline risk rating that can aid in determining an individual’s needed level of supervision and/or treatment intensity (Boer et al., 2004). However, evaluators should be cautious in making decisions from outcomes based on documents that have not been subject to legal scrutiny and due process in the legal arena. Additionally, given the limitations of actuarial tools, it is critical to integrate all sources of information in the evaluation and treatment planning process. Incorporating the individuals’ psychosocial and sexual histories, the PCL-R, sexual interest measures, and measures of cognitive distortions, along with the actuarial risk measures creates the most helpful evaluation. The TIPS-ID (McGrath, 2005) provides a structured format for summarizing information about 25 risk factors that should be addressed. Bringing all these various pieces of information together in a systematic fashion allows evaluators to accomplish more holistic assessments with more specific discernment of an individual’s risk, needs, and strategies to engage the individual in treatment.

Current Treatment Strategies

Therapeutic treatment of individuals with intellectual disabilities has been described from a variety of approaches. These include behavior management, problem-solving skills training, psychoeducational activities, and cognitive-behavioral therapies (Blasingame, 2005; Lindsay & Taylor, 2005). Many programs, including the Developmentally Disabled Sexual Offender Rehabilitative Treatment model (DD-SORT; Blasingame, 2005), integrate multiple components using these varied strategies, making it difficult to ascertain which individual elements of treatment have significant, if any, effects.

At least one meta-analysis of studies on the effectiveness of psychotherapy with individuals with intellectual disabilities found a moderate level of effectiveness (Prout & Nowak-Drabik, 2003). This exploratory study suggested that individual therapies might be more effective than group therapies. The study also suggested that behaviorally oriented treatments showed more promise for bringing about change.

Cognitive-behavioral approaches have been the most popular in recent years and have relatively good support in the literature (Blasingame, 2005; Lindsay & Taylor, 2005). While randomized control studies are yet to be located, cognitive-behavioral treatments have been reported to reduce recidivism among intellectually disabled sex offenders (Lindsay & Taylor, 2005; Rose, Jenkins, O’Conner, Jones, & Felce, 2002). Small sample sizes challenge broad-based comparisons, but these findings are promising. These findings are also consistent with various studies suggesting that cognitive-behavioral treatment helps reduce recidivism in the mainstream population.

McGrath et al. (2007b) reported a skills training and cognitive-behavioral group therapy program was the primary treatment approach for intellectually disabled sex offenders in their statewide program. As discussed earlier, this study found an approximate 11 % reoffense rate among over a hundred subjects. The authors included in their reoffense data incidents that may not have been prosecuted, but under state law, the behaviors could have been considered criminal had prosecution been pursued. The findings from this study suggest that a multifaceted management strategy, including cognitive-behavioral treatment, can be an effective tool in reducing recidivism among intellectually disabled sexual offenders.

Keeling et al. (2006) found that the self-regulation model of relapse prevention can reasonably be applied to intellectually disabled sex offenders. While the great majority of these offenders were classified in the approach pathways, these findings aid in defining the types of risk and needs issues that are to be addressed, particularly self-regulation deficits and abuse-oriented goals. Suggested treatment, teaching, and training targets include correcting cognitive distortions and pro-offending attitudes, developing victim empathy or awareness, controlling deviant sexual interests, and developing motivation to change (Blasingame, 2005, 2006a, 2006b; Haaven & Schlank, 2001; Keeling et al., 2006).

Effective and adaptive solutions to the problems of day-to-day life need to be developed (Nezu, Fiore, & Nezu, 2006). Teaching an individual with intellectual disabilities to consciously monitor his own reaction to a situation and purposely change his reaction to a problem is part of this approach. Internal thoughts, fantasies, and habituated scripts each contribute to problem interpretation and efforts at solving those problems. However, some individuals avoid their problems and therefore fail to implement any problem-solving efforts. Reducing sexual aggression can be aided by discovering alternative ways to cope and resolve issues. Problem-solving interventions need to be multimodal and address cognitive, affective, and behavioral skills development (Nezu et al., 2005). Addressing problems directly, developing a positive attitude about problem solving, and reducing impulsive approaches to problem solving are examples of this multimodal schema.

Collaborative Management of Sex Offenders with Intellectual Disabilities

There are a large number of potential factors influencing an individual client’s functioning at any given time. These affect risk management strategies for offenders who have intellectual disabilities. Until the offender can reduce and manage his own risk, treatment providers, family members, and supervising care providers pursue risk management to prevent further sexual misconduct (Blasingame, 2005).

Andrews and Bonta (2003) articulately point out that engaging the offender-client in a manner that increases responsivity to treatment is associated with reduced recidivism. The methods discussed above address ways of discovering the offenders’ comparative risk level for sexual recidivism and the criminogenic needs which have accumulated in the person’s life.

Risk management efforts will need to address at least seven areas. These are environmental contingencies, coordinated case management, supervisory staff competencies, psychiatric care, cognitive-behavioral treatment, law enforcement supervision, and victim advocacy (Table 2).

Table 2 The risk management circle for individuals with intellectual disabilities

Environmental Contingencies: Risk management efforts in this domain include restricting access to children or potential opportunities to engage in inappropriate sexual conduct; restriction of access to alcohol or other mind-altering substances that might contribute to impulsivity and undermine self-regulation; monitoring types of peers and associates; creating opportunities to have pro-social and age-appropriate social interactions; and providing line-of-sight supervision when the offender is in the proximity of potential victims (Blasingame, 2005). Providing housing and supported employment, access to medical and psychological care, and transportation services are important aspects of the overall life management assistance for these offenders, provided in effort to reduce risk of harm to others in their communities.

Coordinated Case Management: Risk management efforts in this domain include concerted collaborative communication and shared responsibilities between the multiple professionals and supervisory staff persons who form the offenders’ risk management circle. Regular communications between such persons provides for greater continuity across venues, such as case managers, residential facilities, day programs, therapeutic services, and psychiatric services (Blasingame, 2005).

Staff Competencies: Risk management efforts in this domain include having well-trained and adequately motivated staff persons involved in the day-to-day life of the offender. Many intellectually disabled sexual offenders who are known to the service delivery system will be placed in board and care facilities and attend sheltered workshops or day programs. The staff members in these settings play a key role in risk management. Staff members who supervise the daily activities of the offender need to be competently trained across a number of areas (Blasingame, 2005; Mussack, 2006).

Psychiatric Care: Risk management efforts in this domain include evaluation and prescription of appropriate psychotropic medications for those individuals who have dual diagnoses. The frequency of comorbid mental disorders is high and should be attended to in effort to assist the individual in self-regulation of mental health issues, relapse prevention efforts, and the pursuit of a better life for the individual.

Cognitive-Behavioral Treatment: Risk management efforts in this domain include provision of relapse prevention training, sexual education training, relationship skills, adaptive functioning skills, self-regulation skills training, pragmatic problem-solving skills, and other sex offender-specific training based on his needs, risk level, and strategies for garnering engagement. The use of behavioral reinforcement principles and developmentally appropriate adaptations to the therapy delivery is critical (Blasingame, 2005).

Law Enforcement Supervision: Risk management efforts are enhanced for many offenders who have intellectual disabilities when there is significant collaboration with the law enforcement supervision agents, such as probation officers or parole agents. Not all offenders with intellectual disabilities have probation or parole status. When this is the case, it is imperative that all the members of the risk management circle understand the legal conditions imposed on the individual and that there is open communication between the professionals and care providers.

Victim advocacy: Risk management efforts in this domain require that the offender and all the professionals and paraprofessionals involved are working toward the goal of no more victims. While a better life for the offender is desirable, that is secondary to prevention of further harm to other children or other vulnerable persons.

Conclusion and Recommendations for Future Research

Sexual offenders who have intellectual disabilities present evaluators and treatment providers with a number of unique challenges. While the needs of these individuals who have intellectual impairments are very similar to neurotypical offenders, how their needs are assessed and met requires thoughtful intervention. This chapter has reviewed a number of these areas and identified strategies and potential tools to aid evaluators, treatment providers, and other members of the risk management circle.

Research in this specialized area of sexual offender treatment and management has flourished in the recent decade. However, there continue to be a number of areas that need further investigation and clarification. These include developing additional assessment procedures to evaluate mental health, personality characteristics, applied treatment strategies, risk assessment and management procedures, and continued cross-validation of the few tools that have been specifically developed for offenders who have intellectual disabilities. Many of the extant studies involving sex offenders who have intellectual disabilities have small samples from select settings, making it difficult to generalize the findings. Using different definitions or criteria also complicates use of the data. Researchers are encouraged to be more consistent in inclusion/exclusion criteria and collaborate across settings to help remedy these issues.

Ultimately, these sexual offenders will require a risk management circle of professionals, family members, and others to communicate effectively and work in support of the individuals’ success. The evidence is that treatment and intervention with offenders who have intellectual disabilities can be effective in reducing recidivism. Our shared mission is to have no more victims and better lives for these individuals.