Sexually Violent Predator Commitment – What We Know and What Works

Anthony D. Perillo, Indiana University of Pennsylvania
Cynthia Calkins, John Jay College of Criminal Justice
Elizabeth L. Jeglic, John Jay College of Criminal Justice

We examined statewide data of persons evaluated for sexually violent predator (SVP) commitment and compared risk-relevant data of three groups: those committed as SVPs (n _ 374), those not recommended for commitment (n _ 2,707), and those nearly committed (recommended for commitment but ultimately not committed; n _ 117). Consistent with legal language for SVP commitment, binary and multinomial regression analyses revealed that risk scores predicted SVP commitment recommendations in addition to some historical factors (e.g., psychiatric history, never being married, prior sex offenses). For those recommended for commitment, prior sexual offenses predicted ultimate commitment. Those nearly committed had significantly higher sexual recidivism rates than did others who were not committed; however, these recidivism rates were still low (11.5%). Findings suggest that evaluators’ SVP decisions incorporate risk data and follow empirically supported trends; however, the observed recidivism rates of a subset of people that SVP commitment appears to target suggests that the potential for reducing sexual recidivism effectively and efficiently through commitment appears to be quite limited.

sexually violent predator, sex offending, recidivism, civil commitment, risk assessment

“Sexually violent predator (SVP) statutes permit the postsentence civil commitment of persons convicted of a sex offense and considered a heightened risk for reoffending. In the United States, 20 states and the federal government have enacted SVP legislation since 1990. Though statutory language differs across states, SVP commitment generally requires a history of sexual offending, a mental abnormality with evidence of volitional impairment, and a link between the mental abnormality and an elevated risk to commit future sexual violence.”

“Much published research examining SVP-committed persons reports demographics and characteristics of those held in SVP commitment programs. Age is a key consideration as recidivism risk has been estimated to decline 4% per year as people with sex offending histories age. SVP-committed persons are, on average, approximately in their mid-forties, with some variability across jurisdictions. SVP-committed persons in Minnesota, for example, were found to be slightly younger at admission than were those in Arizona. Findings related to the victims of SVP-committed persons have been particularly variable: having adult, stranger victims; having child victims; and having both related and unrelated victims have all been associated with SVP commitment…SVP-committed persons were much more likely to have used a weapon during their index sex offense. A later study in New York found that factors suggestive of sexual deviance particularly predicted SVP commitment, including assessment of deviant sexual preferences or pedophilia/pedophilic disorder.”

“…5- and 10-year sexual recidivism rates among released from SVP commitment at 9.2% and 13.1%, respectively. Most recidivists were between age 30 and 49, but recidivism rates were relatively low even for these groups (11.5 to 15.9%). Ten-year recidivism rates for those released from SVP commitment were no higher than were 10-year recidivism rates for a random subset of persons convicted of sex offenses who were previously released from incarceration without being committed. Five-year recidivism rates for those released from SVP commitment were higher than were those for other released individuals (9.2% vs. 5.3%); however, the results suggest fairly low recidivism rates, with over 90% of SVP-committed persons who were later released desisting through the first 5 years of release.”

“An additional proxy for assessing SVP commitment may thus be to examine the recidivism rates of people nearly committed: those who were referred for commitment by legal decision makers (e.g., district attorney), then recommended for commitment by an expert evaluator (e.g., psychiatrist or psychologist), but were ultimately not committed. In fact, such focus may better reflect prospective recidivism rates of SVP-committed persons than actual recidivism rates of SVP-commtted persons after release from commitment. Examination of recidivism among those committed, although useful, would be limited by the fact that SVP-committed persons often spend a decade or more committed; thus, they would be expected to show notable age-related declines in recidivism during their commitment (regardless of any effects of treatment while committed).”

“Given questions about how effective SVP commitment can be in reducing sexual violence, research must provide policymakers with findings that offer insight into efficient and efficacious focus of scarce resources. Examination of SVP commitment decision making offers insight into the effectiveness of SVP commitment in multiple ways. One, such investigations can evaluate the extent that commitment decisions accurately target those at elevated risk by examining whether such decisions are consistent with factors empirically associated with recidivism risk (e.g., actuarial risk scores) rather than factors with no relationship to recidivism (e.g., race). Two, if SVP commitment decision making were empirically based, comparisons of recidivism of noncommitted individuals with the subset of nearly committed individuals would be expected to result in elevated recidivism rates among those nearly committed. Finally, review of nearly committed individuals offers a better proxy comparison to the potential recidivism rates of individuals considered for commitment than merely examining the broad scope of people released after incarceration for a sex crime.”

“Using a quasi-experimental design in a state-wide sample, the overall aim of the current study is to add to existing SVP literature by examining characteristics of those recommended for SVP commitment and recidivism rates of those recommended for commitment by expert evaluators but nonetheless not committed (i.e., nearly committed). First, we compared SVP-committed persons to those released from prison following their sentence (which includes all those released regardless of whether they were referred and/or recommended for commitment) on a series of offense and risk factors. Using those factors that distinguished SVP-committed persons from those not SVP-commtted, we used a series of logistic regressions to estimate the strength of the relationship between these factors and decisions to commit. Finally, we examined the sexual and nonsexual recidivism rates of nearly committed persons via Cox regressions to offer insight into the prospective recidivism rates of those being recommended for SVP commitment. Based on findings from SVP samples in other states, we anticipated that those who had been committed under SVP legislation would have the highest risk scores followed by those who were nearly committed and those that were not considered for SVP commitment. Further, we anticipated the group that was nearly committed would have higher rates of recidivism than the group not recommended for SVP commitment by an expert evaluator.”

“We found persons ultimately SVP-committed to be higher risk on actuarial risk instruments than those not committed. SVP-committed persons in our sample tended to have more victims, have a juvenile record, and have a psychiatric history relative to those not committed. Further, the SVP-committed persons in our sample had significantly higher actuarial risk scores.”

“We found that among the entirety of those not SVP-committed, those recommended for SVP commitment (nearly committed) had higher static risk scores and were more likely to have a psychiatric history than others who were not committed. These predictors are relevant to both general prongs of SVP commitment: elevated recidivism risk and mental abnormality. The key difference between those recommended for commitment who were or were not committed was a record of sexual offenses beyond the index offense, as those with prior sexual charges or convictions were over five times more likely to be committed.”

“Although sexual reconviction rates of those nearly committed were more than double (11.5%) those of the general sample (5.0%), it is notable also that these rates of recidivism are still relatively low. Nonetheless, we found those nearly committed were indeed more likely to recidivate, particularly for more serious offenses (sexual and violent offenses). Indeed, offenses for which those nearly committed had no higher likelihood of recidivism were generally those less directly relevant to sexual risk assessment.”

“Recidivism results again suggest clinicians are effectively identifying those that better fit the criteria and aims of SVP commitment, as those recommended for commitment but nonetheless released were over twice as likely to sexually recidivate than others who were released.”

“Although the current study includes no economic data and thus no specific cost-effectiveness data, the fact that 88.5% of persons targeted for SVP commitment (the nearly committed group) did not reoffend suggests SVP commitment could only prevent additional sexual violence for a small minority of those committed. Such findings are consistent with prior conclusions that, given the narrow scope of SVP commitment (in terms of those targeted and the types of offenses prevented) relative to the resources we devote to it, SVP commitment is arguably best summarized as a well intended practice that unfortunately offers little return on investment. Findings of this study suggest that clinicians are generally identifying those at higher risk for reoffending (i.e., those considered for SVP commitment but ultimately not committed); therefore, we can develop services to better supervise and monitor this higher risk group. In the long run, such a focus would likely prevent future sexual violence more effectively and more efficiently.”

“…some states have at some point developed outpatient SVP schemes (e.g., Texas), whereas other states have specific designations for people deemed at high risk for committing sex crimes. Programs like these focus resources on outpatient services to monitor and supervise higher risk persons without extensive resources for inpatient SVP commitment. Although in theory these programs appear to address many of the concerns highlighted in inpatient SVP commitment programs, we still know little about their effectiveness at preventing sexual violence. Data from New York State’s civil management initiative (i.e., Strict and Intensive Supervision and Treatment), which closely monitors those in the community at high risk to reoffend sexually, have been associated with small reductions in sexual re-arrest rates and at considerably less cost per client per year (i.e., approximately $30,000 per year as compared with $100,000 per year.”

As always, please join the discussion below if you have thoughts or comments to add!