Assessing risk of sex offenders with major mental illness: integrating research into best practices
| DOI | https://doi.org/10.1108/JACPR-02-2015-0162 |
| Date | 12 October 2015 |
| Pages | 258-274 |
| Published date | 12 October 2015 |
| Author | Sharon M. Kelley,David Thornton |
| Subject Matter | Health & social care,Criminology & forensic psychology,Aggression, conflict & peace |
Assessing risk of sex offenders with major
mental illness: integrating research into
best practices
Sharon M. Kelley and David Thornton
Dr Sharon M. Kelley is a
Forensic Evaluator at
Evaluation Unit, Sand Ridge
Secure Treatment Center,
Madison,
Wisconsin, USA.
Dr David Thornton is Research
Director at Research Unit, Sand
Ridge Secure Treatment
Center, Madison,
Wisconsin, USA.
Abstract
Purpose –Sex Offenders with a Major Mental Illness (SOMMI) are doubly stigmatized, as these individuals
are members of two highly marginalized social groups (Guidry and Saleh, 2004). Within each of these groups
SOMMI only represent a small minority. For professionals seeking to base their practice in empirical research
this has led to a significant problem since the literature related specifically to this group is both limited and
hard to locate. Additionally, intensity of psychological risk factors varies as a function of psychiatric
decompensation for some SOMMI making it hard to apply certain procedures that work with ordinary sexual
offenders. The purpose of this paper is to provide a review of the relevant literature and recommendations
for clinical practice that are responsive to the particular challenges posed by this unusual group of
sexual offenders.
Design/methodology/approach –The current paper provides a review of literature on risk factors for
sexual recidivism and validity of current risk tools as it pertains to SOMMI. Recommendations for risk
assessment with SOMMI are provided.
Findings –The static actuarial tools appear to be useful with SOMMI. However, risk assessments measuring
dynamic risk factors have poorer predictive validity. Additional factors that will need to be considered involve a
possible higher recidivism rate for SOMMI and a variable relationship between major mental illness and sex
offending with it sometimes predisposing, sometimes exacerbating existing risk factors, and sometimes
mitigating risk.
Originality/value –There is a paucity of research and guidance in assessment and risk management
of SOMMI. The current paper is the first to thoroughly explore the efficacy of current sex offender risk
assessment tools with SOMMI and provide structured guidance for making decisions about risk and risk
management needs for this challenging population.
Keywords Risk assessment, Risk management, Prediction, Sex offenders, Criminogenic need,
Major mental illness
Paper type Literature review
Background of the problem
Sex Offenders with a Major Mental Illness (SOMMI) are doubly stigmatized, as these individuals
are members of two highly marginalized social groups (Guidry and Saleh, 2004). Within each of
these groups SOMMI represent only a small minority. Surveys of individuals in psychiatric
settings indicate that only a minority is known to have committed sex offenses with numbers
ranging from 1.6 to 20 percent (Baker and White, 2002; Fisher et al., 2006; Wallace et al., 2004).
Similarly surveys of sexual offenders indicate that the prevalence of major mental illness is low,
ranging from 1.4 to 16 percent, with higher rates seen among criminal defendants who were
referred for forensic evaluations at court clinics (Becker et al., 2003; Cochrane et al., 2001;
Langstrom et al., 2004; Packard and Rosner, 1985; Raymond et al., 1999). Sexual offenders do
Received 3 February 2015
Revised 7 June 2015
Accepted 13 June 2015
The views expressed are those of
the authors and not necessarily
those of the Wisconsin
Department of Health Services.
PAG E 25 8
j
JOURNAL OF AGGRESSION, CONFLICTAND PEACE RESEARCH
j
VOL. 7 NO. 4 2015, pp. 258-274, © Emerald Group Publishing Limited, ISSN 1759-6599 DOI 10.1108/JACPR-02-2015-0162
have mental disorders; however, these more commonly include substance use disorders,
paraphilias, anxiety, and depression (Kafka and Prentky, 1994; McElroy et al., 1999) rather than
major mental illness (psychosis or bipolar disorder). Probably as a consequence of this dual
stigma and dual minority status, SOMMI patients are often underserved by the mental health
system, which lacks expertise in the management of sexual deviance and are equally poorly
served by traditional sex offender-specific treatment programs, which do not consider the unique
psychiatric issues within this population (Drake and Pathe, 2004).
A central difficulty for those seeking to provide evidence-based risk management services for
SOMMI is that their double minority status means that it is not clear how well research into
persons with either major mental illness or persons with a history of sexual offending applies
to them. Given the cautions by Grisso (2000) against the generalization of test results outside the
population(s) sampled to create norms, this leaves clinicians in the untenable situation of having to
choose between using the existing assessment tools despite validity concerns or using
unstructured clinical judgment, which has been shown to have weak predictive accuracy
(Mann et al., 2010).
The present paper seeks to assist professionals involved in carrying out risk management
assessments for SOMMI by speaking to three inter-related questions:
1. How does major mental illness relate to sexual offending?
2. How predictive are commonly used risk assessment tools when used with SOMMI?
3. Is it possible to make individualized assessments of how major mental illness relates to
sexual offending?
Unfortunately, research speaking directlyto these questions is limited, so in endeavoring to provide
a helpful answer wesupplement it by drawing on wider bodies of literature.
How does major mental illness relate to sexual offending?
Before turning specifically to studies of SOMMI and sexual offending, it may be helpful to put this
in the context of the larger body of studies that have examined the relationship between
major mental illness and violent or general offending. Two recent, large meta-analyses of the
relationship between psychosis and violence have been reported (Douglas et al., 2009; Bonta
et al., 2014). Results indicate that individuals with psychosis are more likely to be violent than the
general population or persons with internalizing mental disorders but less likely to be violent than
those with externalizing disorders (e.g. Attention Deficit/Hyperactivity Disorder, Oppositional
Defiant Disorder, Conduct Disorder, Antisocial Personality Disorder, and Substance Abuse
Disorders). As a consequence they are not consistently more violent than the average mentally
disordered offender.
Thus, psychosis may be seen as a risk factor for violence but as a less important risk factor than
those embodied in externalizing disorders. This conclusion is consistent with other studies which
found that treatment non-compliance, suicidal ideation, preoccupation with violence, impulsivity,
hostility, and residing in neighborhoods with high-poverty rates appear to increase the risk of
violence for individuals with, and individuals without, a major mental illness (Monahan et al., 2001;
Silver et al., 1999; Swanson et al., 2006). Somewhat similarly, Bonta et al.’s (2014) study also
demonstrated that within the mentally disordered offender population in general, and specifically
among those determined to be Not Guilty by Reason of Insanity, well-known general
criminogenic factors such as procriminal attitudes, antisocial personality traits, and alcohol/
substance abuse were more predictive than psychosis.
Consistent with interpreting the literature as indicating that major mental illness is a risk factor for
violence is the finding that, among those with major mental illness, treatment specific to these
disorders should lower risk. Van Dorn et al. (2013) found a protective effect related to mental
health services aimed to maintain psychiatric stabilization among 4,056 individuals diagnosed
with schizophrenia or bipolar disorder who had been released from a hospital. Specifically, having
access to psychotropic medication for at least 90 days significantly reduced the likelihood
of felony offenses.
VOL. 7 NO. 4 2015
j
JOURNAL OF AGGRESSION, CONFLICTAND PEACERESEARCH
j
PAG E 25 9
To continue reading
Request your trial