A classification tree approach to the development of actuarial violence risk assessment tools

Identifying young adults at risk of Medicaid enrollment lapses after inpatient mental health treatment

OBJECTIVE

This study sought to describe Medicaid disenrollment rates and risk factors among young adults after discharge from inpatient psychiatric treatment.

METHODS

The sample included 1,176 Medicaid-enrolled young adults ages 18 to 26 discharged from inpatient psychiatric care in a mid-Atlantic state. Medicaid disenrollment in the 365 days postdischarge and disenrollment predictors from the 180-day predischarge period (antecedent period) were identified from administrative records. Classification and regression tree and probit regression analysis were used.

RESULTS

Thirty-two percent were disenrolled from Medicaid within a year of discharge. Both analytical approaches converged on four main risk factors: being in the Medicaid enrollment category for persons with a nondisabled low-income parent or for a child in a low-income household, being age 18 or 20 at discharge, having a Medicaid enrollment gap in the antecedent period, and having no primary care utilization in the antecedent period. For the 48% of the sample continuously enrolled in the antecedent period who were in the enrollment categories for disabled adults or foster care children, the disenrollment rate was 13%.

CONCLUSIONS

A substantial minority of Medicaid-enrolled young adults discharged from inpatient care were disenrolled from Medicaid within a year. About half the sample had a low disenrollment risk, but the other half was at substantial risk. Risk factors largely reflected legal status changes that occur among these transition-age youths. Identifying inpatients at high risk of disenrollment and ensuring continuous coverage should improve access to needed postdischarge supports. Regular primary care visits may also help reduce unintended Medicaid disenrollment in this population.

Psychopathic traits modulate brain responses to drug cues in incarcerated offenders

Recent neuroscientific evidence indicates that psychopathy is associated with abnormal function and structure in limbic and paralimbic areas. Psychopathy and substance use disorders are highly comorbid, but clinical experience suggests that psychopaths abuse drugs for different reasons than non-psychopaths, and that psychopaths do not typically experience withdrawal and craving upon becoming incarcerated. These neurobiological abnormalities may be related to psychopaths' different motivations for—and symptoms of—drug use. This study examined the modulatory effect of psychopathic traits on the neurobiological craving response to pictorial drug stimuli. Drug-related pictures and neutral pictures were presented and rated by participants while hemodynamic activity was monitored using functional magnetic resonance imaging. These data were collected at two correctional facilities in New Mexico using the Mind Research Network mobile magnetic resonance imaging system. The sample comprised 137 incarcerated adult males and females (93 females) with histories of substance dependence. The outcome of interest was the relation between psychopathy scores (using the Hare Psychopathy Checklist-Revised) and hemodynamic activity associated with viewing drug-related pictures vs. neutral pictures. There was a negative association between psychopathy scores and hemodynamic activity for viewing drug-related cues in the anterior cingulate, posterior cingulate, hippocampus, amygdala, caudate, globus pallidus, and parts of the prefrontal cortex. Psychopathic traits modulate the neurobiological craving response and suggest that individual differences are important for understanding and treating substance abuse.

Protective mechanisms and prevention of violence and aggression in veterans

Although a subset of Iraq and Afghanistan Veterans show aggression toward others after they return home from military service, little is known about protective mechanisms that could be bolstered to prevent violence. A national longitudinal survey was conducted between 2009 and 2011 using a random sample of veterans who served in Operation Iraqi Freedom or Operation Enduring Freedom. One thousand and ninety veterans, from 50 states representing all military branches, completed 2 waves of data collection, 1 year apart (retention rate = 79%). The final sample resembled the U.S. military post 9/11 in terms of age, sex, ethnicity, geography, and service branch. Protective mechanisms in socioeconomic (money to cover basic needs, stable employment), psychosocial (resilience, perceiving control over one's life, social support), and physical (healthy sleep, no physical pain) domains were examined. We found these protective mechanisms predicted decreased aggression and violence at follow-up, particularly among higher risk veterans. Multivariable analyses confirmed that protective mechanisms lowered violence through their interaction with risk factors. This study identifies protective mechanisms related to decreased community violence in veterans and indicates that rehabilitation aimed at improving socioeconomic, psychosocial, and physical well-being has potential promise to reduce aggression and violence among veterans after returning home from military service.

Improving Clinical Prediction of Bipolar Spectrum Disorders in Youth

This report evaluates whether classification tree algorithms (CTA) may improve the identification of individuals at risk for bipolar spectrum disorders (BPSD). Analyses used the Longitudinal Assessment of Manic Symptoms (LAMS) cohort (629 youth, 148 with BPSD and 481 without BPSD). Parent ratings of mania symptoms, stressful life events, parenting stress, and parental history of mania were included as risk factors. Comparable overall accuracy was observed for CTA (75.4%) relative to logistic regression (77.6%). However, CTA showed increased sensitivity (0.28 vs. 0.18) at the expense of slightly decreased specificity and positive predictive power. The advantage of CTA algorithms for clinical decision making is demonstrated by the combinations of predictors most useful for altering the probability of BPSD. The 24% sample probability of BPSD was substantially decreased in youth with low screening and baseline parent ratings of mania, negative parental history of mania, and low levels of stressful life events (2%). High screening plus high baseline parent-rated mania nearly doubled the BPSD probability (46%). Future work will benefit from examining additional, powerful predictors, such as alternative data sources (e.g., clinician ratings, neurocognitive test data); these may increase the clinical utility of CTA models further.

Self-report and longitudinal predictors of violence in Iraq and Afghanistan war era veterans

This study, using a longitudinal design, attempted to identify whether self-reported problems with violence were empirically associated with future violent behavior among Iraq and Afghanistan war veterans and whether and how collateral informant interviews enhanced the risk assessment process. Data were gathered from N = 300 participants (n = 150 dyads of Iraq and Afghanistan war veterans and family/friends). The veterans completed baseline and follow-up interviews 3 years later on average, and family/friends provided collateral data on dependent measures at follow-up. Analyses showed that aggression toward others at follow-up was associated with younger age, posttraumatic stress disorder, combat exposure, and a history of having witnessed parental violence growing up. Self-reported problems controlling violence at baseline had robust statistical power in predicting aggression toward others at follow-up. Collateral report enhanced detection of dependent variables: 20% of cases positive for violence toward others would have been missed relying only on self-report. The results identify a subset of Iraq and Afghanistan war veterans at higher risk for problematic postdeployment adjustment and indicate that the veterans' self-report of violence was useful in predicting future aggression. Underreporting of violence was not evidenced by most veterans but could be improved upon by obtaining collateral information.

Quantifying the relative risk of sex offenders: risk ratios for static-99R

Given the widespread use of empirical actuarial risk tools in corrections and forensic mental health, it is important that evaluators and decision makers understand how scores relate to recidivism risk. In the current study, we found strong evidence for a relative risk interpretation of Static-99R scores using 8 samples from Canada, United Kingdom, and Western Europe (N = 4,037 sex offenders). Each increase in Static-99R score was associated with a stable and consistent increase in relative risk (as measured by an odds ratio or hazard ratio of approximately 1.4). Hazard ratios from Cox regression were used to calculate risk ratios that can be reported for Static-99R. We recommend that evaluators consider risk ratios as a useful, nonarbitrary metric for quantifying and communicating risk information. To avoid misinterpretation, however, risk ratios should be presented with recidivism base rates.

Recidivism after treatment in a forensic youth-psychiatric setting: the effect of treatment characteristics

The aim of this study was to examine the effect of treatment characteristics on recidivism in a forensic youth-psychiatric outpatient clinic. The treatment offered comprised functional family therapy (FFT), individual cognitive behavioural therapy (CBT), or CBT in combination with parent training. Some of the youth additionally participated in aggression replacement training (ART). FFT and ART were implemented as a trial version, meaning that most therapists had not received formal training yet. Treatment characteristics related to recidivism were length of treatment, type of treatment, number of sessions, and the therapist. The longer the period of treatment and the greater the number of sessions, the higher the recidivism, even after controlling for risk of recidivism based on static risk factors. Juveniles who participated in ART reoffended more often than juveniles who had not participated in such training. Given the fact that FFT and ART were not well-implemented trial versions, it can be concluded that poorly implemented treatment leads to poor outcomes.

Managing aggression and violence: the clinician's role in contemporary mental health care

OBJECTIVE

From time to time misconceptions about violence risk assessment raise debate about the role mental health professionals play in managing aggression, with associated concerns about the utility of violence risk assessment. This paper will address some of the misconceptions about risk assessment in those with serious mental illness.

METHODS

The authors have expertise as clinicians and researchers in the field and based on their accumulated knowledge and discussion they have reviewed the literature to form their opinions.

RESULTS

This paper reflects the authors' views.

CONCLUSION

There is a modest yet statistical and clinically significant association between certain types of mental illness and violence. Debate about the appropriateness of clinician involvement in violence risk assessment is sometimes based on a misunderstanding about the central issues and the degree to which this problem can be effectively managed. The central purpose of risk assessment is the prevention rather than the prediction of violence. Violence risk assessment is a process of identifying patients who are at greater risk of violence in order to facilitate the timing and prioritisation of preventative interventions. Clinicians should base these risk assessments on empirical knowledge and consideration of case-specific factors to inform appropriate management interventions to reduce the identified risk.

Youth Actuarial Risk Assessment Tool (Y-ARAT): The development of an actuarial risk assessment instrument for predicting general offense recidivism on the basis of police records

Estimating the risk for recidivism is important for many areas of the criminal justice system. In the present study, the Youth Actuarial Risk Assessment Tool (Y-ARAT) was developed for juvenile offenders based solely on police records, with the aim to estimate the risk of general recidivism among large groups of juvenile offenders by police officers without clinical expertise. On the basis of the Y-ARAT, juvenile offenders are classified into five risk groups based on (combinations of) 10 variables including different types of incidents in which the juvenile was a suspect, total number of incidents in which the juvenile was a suspect, total number of other incidents, total number of incidents in which co-occupants at the youth's address were suspects, gender, and age at first incident. The Y-ARAT was developed on a sample of 2,501 juvenile offenders and validated on another sample of 2,499 juvenile offenders, showing moderate predictive accuracy (area under the receiver-operating-characteristic curve = .73), with little variation between the construction and validation sample. The predictive accuracy of the Y-ARAT was considered sufficient to justify its use as a screening instrument for the police.

Differences in the prevalence and impact of risk factors for general recidivism between different types of juveniles who have committed sexual offenses (JSOs) and juveniles who have committed nonsexual offenses (NSOs)

To date, there is surprisingly little research on differences in the prevalence and impact of risk factors for general recidivism between juveniles who have committed sexual offenses (JSO) and juveniles who have committed nonsexual offenses (NSO). Therefore, we examined differences in the prevalence and impact of dynamic risk factors for general delinquency between youth with nonsexual offenses (NSO, n = 504), youth with misdemeanor sexual offenders (MSO, n = 136), youth with felony sexual offenders (FSO, n = 116) and youth with offenses against much younger children (CSO, n = 373). The sample consisted of boys with a mean age of 15.3 years (SD = 1.5). The prevalence of dynamic risk factors for general delinquency was significantly lower in JSOs than in NSOs. More serious sexual offenses were associated with a lower prevalence of dynamic risk factors. In contrast, the impact of most dynamic risk factors on general recidivism proved to be significantly larger among JSOs compared to NSOs. The relative importance of the dynamic risk factors varied for each type of JSO, resulting in differences in the dynamic risk profiles of the various types of JSOs.

A comparison of scoring models for computerised mental health screening for federal prison inmates

BACKGROUND

There are high rates of mental disorder in correctional environments, so effective mental health screening is needed. Implementation of the computerised mental health screen of the Correctional Service of Canada has led to improved identification of offenders with mental health needs but with high rates of false positives.

AIMS

The goal of this study is to evaluate the use of an iterative classification tree (ICT) approach to mental health screening compared with a simple binary approach using cut-off scores on screening tools.

METHODS

A total of 504 consecutive admissions to federal prison completed the screen and were also interviewed by a mental health professional. Relationships between screening results and more extended assessment and clinical team discussion were tested.

RESULTS

The ICT was more parsimonious in identifying probable 'cases' than standard binary screening. ICT was also highly accurate at detecting mental health needs (AUC=0.87, 95% CI 0.84-0.90). The model identified 118 (23.4%) offenders as likely to need further assessment or treatment, 87% of whom were confirmed cases at clinical interview. Of the 244 (48.4%) offenders who were screened out, only 9% were clinically assessed as requiring further assessment or treatment. Standard binary screening was characterised by more false positives and a comparable false negative rate.

CONCLUSIONS

The use of ICTs to interpret screening data on the mental health of prisoners needs further evaluation in independent samples in Canada and elsewhere. This first evaluation of the application of such an approach offers the prospect of more effective and efficient use of the scarce resource of mental health services in prisons. Although not required, the use of computers can increase the ease of implementing an ICT model.

Valuing structured professional judgment: predictive validity, decision-making, and the clinical-actuarial conflict

Structured professional judgment (SPJ) has received considerable attention as an alternative to unstructured clinical judgment and actuarial assessment, and as a means of resolving their ongoing conflict. However, predictive validity studies have typically relied on receiver operating characteristic (ROC) analysis, the same technique commonly used to validate actuarial assessment tools. This paper presents SPJ as distinct from both unstructured clinical judgment and actuarial assessment. A key distinguishing feature of SPJ is the contribution of modifiable factors, either dynamic or protective, to summary risk ratings. With modifiable factors, the summary rating scheme serves as a prognostic model rather than a classification procedure. However, prognostic models require more extensive and thorough predictive validity testing than can be provided by ROC analysis. It is proposed that validation should include calibration and reclassification techniques, as well as additional measures of discrimination. Several techniques and measures are described and illustrated. The paper concludes by tracing the limitations of ROC analysis to its philosophical foundation and its origin as a statistical theory of decision-making. This foundation inhibits the performance of crucial tasks, such as determining the sufficiency of a risk assessment and examining the evidentiary value of statistical findings. The paper closes by noting a current effort to establish a viable and complementary relationship between SPJ and decision-making theory.

Financial well-being and postdeployment adjustment among Iraq and Afghanistan war veterans

Research has yet to examine the relationship between financial well-being and community reintegration of veterans. To address this, we analyzed data from n = 1,388 Iraq and Afghanistan War Era Veterans who completed a national survey on postdeployment adjustment. The results indicated that probable major depressive disorder, posttraumatic stress disorder, and traumatic brain injury were associated with financial difficulties. However, regardless of diagnosis, veterans who reported having money to cover basic needs were significantly less likely to have postdeployment adjustment problems such as criminal arrest, homelessness, substance abuse, suicidal behavior, and aggression. Statistical analyses also indicated that poor money management (e.g., incurring significant debt or writing bad checks) was related to maladjustment, even among veterans at higher income levels. Given these findings, efforts aimed at enhancing financial literacy and promoting meaningful employment may have promise to enhance outcomes and improve quality of life among returning veterans.

Protective factors and risk modification of violence in Iraq and Afghanistan War veterans

OBJECTIVE

After returning home, a subset of Iraq and Afghanistan War veterans report engaging in aggression toward others. This study is the first to identify variables empirically related to decreased risk of community violence among veterans.

METHOD

The authors conducted a national survey from July 2009 to April 2010 in which participants were randomly drawn from over 1 million US military service members who served after September 11, 2001. Data were collected from a total of 1,388 Iraq and Afghanistan War era and theater veterans. The final sample included veterans from all 50 states and all military branches.

RESULTS

One-third of survey respondents self-identified committing an act of aggression toward others during the past year, mostly involving minor aggressive behavior. Younger age, criminal arrest record, combat exposure, probable posttraumatic stress disorder, and alcohol misuse were positively related to violence toward others. Controlling for these covariates, multivariate analyses showed that stable living situation and the perception of having control over one's life were associated with reduced odds of severe violence (R2 = 0.24, χ27 = 145.03, P < .0001). Greater resilience, perceiving positive social support, and having money to cover basic needs were linked to reduced odds of other physical aggression (R2 = 0.20, χ28 = 188.27, P < .0001).

CONCLUSIONS

The study identifies aggression as a problem for a subset of Iraq and Afghanistan War veterans who endorsed few protective factors. Analyses revealed that protective factors added incremental value to statistical modeling of violence, even when controlling for robust risk factors. The data indicate that, in addition to clinical interventions directed at treating mental health and substance abuse problems, psychosocial rehabilitation approaches aimed at improving domains of basic functioning and psychological well-being may also be effective in modifying risk and reducing violence among veterans.

Mental disorder and violence: is there a relationship beyond substance use?

PURPOSE

A general consensus exists that severe mental illness (SMI) increases violence risk. However, a recent report claimed that SMI "alone was not statistically related to future violence in bivariate or multivariate analyses." We reanalyze the data used to make this claim with a focus on causal relationships between SMI and violence, rather than the statistical prediction of violence.

METHODS

Data are from the National Epidemiologic Survey on Alcohol and Related Conditions (NESARC), a two-wave study (N = 34,653: Wave 1: 2001-2003; Wave 2: 2004-2005). Indicators of mental disorder in the year prior to Wave 1 were used to examine violence between Waves 1 and 2.

RESULTS

Those with SMI, irrespective of substance abuse status, were significantly more likely to be violent than those with no mental or substance use disorders. This finding held in both bivariate and multivariable models. Those with comorbid mental and substance use disorders had the highest risk of violence. Historical and current conditions were also associated with violence, including childhood abuse and neglect, household antisocial behavior, binge drinking and stressful life events.

CONCLUSIONS

These results, in contrast to a recently published report, show that the NESARC data are consistent with the consensus view on mental disorder and violence: there is a statistically significant, yet modest relationship between SMI (within 12 months) and violence, and a stronger relationship between SMI with substance use disorder and violence. These results also highlight the importance of premorbid conditions, and other contemporaneous clinical factors, in violent behavior.

Predictive Validity of the Washington State Juvenile Court Pre-Screen Assessment in the Netherlands

This study examined the predictive validity of the Washington State Juvenile Pre-Screen Assessment (WSJCA pre-screen) in the Netherlands. Previous research conducted in the United States showed the predictive validity of the WSJCA pre-screen to be modest, as is the case with the predictive validity of most other risk assessment instruments for juveniles. Therefore, it was also examined whether the predictive validity of the WSJCA pre-screen can be improved by modifying the scoring procedure. The sample consisted of 520 youths who had been referred to the juvenile probation service by court. The present study showed the predictive validity of the WSJCA pre-screen in the Netherlands to be modest too, with an area under the receiver operating characteristic curve (AUC) of .625. Modifying the scoring procedure by means of chi-squared automatic interaction detector analyses significantly improved the predictive validity to an AUC of .702. The modified scoring procedure is time-saving because only variables that uniquely contribute to the prediction of recidivism are included, which at the same time leads to a more accurate prediction of recidivism.

Integrating Risk Context Into Risk Assessments

The context in which offenders are released is an important component of conducting risk assessments. A sample of 257 supervised male parolees were followed in the community ( M = 870 days) after an initial risk assessment. Drawing on community-based information, the purpose of this study was to evaluate the recently developed Risk Context Scale. Four domains from the Risk Context Scale include Resource Engagement, Social Network Presence, Integration of Care, and Social Stability. Using mediation analysis, an initial static risk probability was altered up to 26% by accounting for risk context. Implications of the present results include a broader explanation of recidivism, offering intervention strategies, and further individualizing risk assessments.

Derivation and validation of a short emergency department screening tool for perpetrators of intimate partner violence: the PErpetrator RaPid Scale (PERPS)

BACKGROUND

There is no short screening tool for perpetrators of intimate partner violence (IPV), although one is needed.

OBJECTIVE

To retrospectively derive and prospectively validate a brief screening tool for perpetrators of IPV: the PErpetration RaPid Scale (PERPS).

METHODS

In the derivation phase of the study, we developed the PERPS based on historical data. The PERPS consists of three Yes/No questions about physical abuse of a partner. In the validation phase, we prospectively screened subjects during randomized 4-h shifts in a busy emergency department (ED). Subjects were asked to complete the newly derived three-question PERPS as well as the Physical Abuse of Partner Scale (PAPS), a 25-question Likert scale that is the gold standard for detection of physical abuse of a partner. The main outcome measures were sensitivity, specificity, predictive values, accuracy, and Cronbach alpha of the PERPS for internal consistency.

RESULTS

The PERP Scale derivation was based on a 207-subject historical database, and resulted in a three-question PERPS. Validation was completed on a new set of 214 patients presenting to the ED during 52 randomized 4-h shifts. The prevalence of IPV perpetration using the PERPS was 47/207 (22.7%; 95% confidence interval [CI] 16-27). For the PAPS, prevalence was 56/207 (27%; 95% CI 20-32). Compared with the PAPS, the sensitivity of the PERPS was 66%, specificity 93%, negative predictive value 87%, positive predictive value 78%, with an accuracy of 85%. Cronbach alpha of the PERPS was 0.68. Age, gender, and race were not predictive of positive results on either scale.

CONCLUSION

We successfully derived and validated a three-question perpetrator of IPV scale that can be used in a busy ED or office setting.

Disparities in the moral intuitions of criminal offenders: The role of psychopathy

The present study examined whether and in what ways psychopathy is associated with abnormal moral intuitions among criminal offenders. Using Haidt et al.'s Moral Foundations Questionnaire, 222 adult male offenders assessed for clinical psychopathy reported their degree of support for five moral domains: Harm Prevention, Fairness, Respect for Authority, Ingroup Loyalty, and Purity/Sanctity. As predicted, psychopathy total score explained a substantial proportion of the variance in reduced support for Harm Prevention and Fairness, but not the other domains. These results confirm that psychopathy entails a discrete set of moral abnormalities and suggest that these abnormalities could potentially help to explain the characteristic antisocial behavior of individuals with psychopathy.

The impact of psychopathy on violence among the household population of Great Britain

BACKGROUND

The public health impact of psychopathy on violence has not previously been measured in the general population.

METHODS

Psychopathy was measured using the Psychopathy Checklist:Screening Version (PCL:SV) in adults 16-74 years (n = 638) in households in England, Scotland and Wales in a two-phase survey which included self-reported measures of violence.

RESULTS

Participants with PCL:SV scores 11 or above demonstrated a prevalence of 2.1% (95% CI 1.2, 3.8) but accounted for 18.7% of violent incidents, a population-attributable risk of 16.6%, and demonstrated an exponential rise in reported violent incidents. Psychopathic traits correlated with victim injury, multiple victim subtypes and locations.

CONCLUSIONS

Psychopathy makes a substantial impact on violence among the general population despite a low prevalence. Explanations of this association include interpersonal difficulties due to psychopathic traits, fearlessness, thrill seeking, and antisocial lifestyle, but not impulsivity independent of antisocial lifestyle.

An approach to the patient in crisis: assessments of the risk of suicide and violence

Suicide is one of the leading causes of death in the United States and is defined as intentional self-harm with the intent of causing death. Various mental disorders may be a cause for increased violence. This article outlines the elements of the risk assessment (for harm to self and/or others) in patients in crisis and addresses which contributing factors may be modifiable. This article also proposes a practical framework for the management of risk regarding suicide and violence.

Clinical decisions in psychiatry should not be based on risk assessment

OBJECTIVE

Risk assessments that place patients in high or low risk categories have been widely adopted by mental health services in an attempt to reduce the harms associated with psychiatric disorders. This paper examines the effects of categorization based on the results of a risk assessment.

METHODS

The violence prediction instrument derived from the MacArthur Study of Mental Disorder and Violence was used to illustrate the nature and effects of risk assessment and the consequent categorization of patients.

RESULTS

The majority of patients categorized as being at high risk will not commit any harmful acts.

CONCLUSIONS

Patients who are classified as high risk share the cost of efforts to reduce harm in the form of additional treatment and restrictions, although the majority will not go on to commit a harmful act. Clinical decisions made on the basis of risk assessment also divert resources away from patients classified as low risk, even though a significant proportion do go on to a commit harmful act. We argue that psychiatric professionals should discuss the risks of treatment and of non-treatment with patients (or with their substitute decision-makers) and should maintain a duty to warn about the consequences of not having treatment. However, assessment of risk of harm should not form the basis for clinical decision making. We should aim to provide optimal care according to the treatment needs of each patient, regardless of the perceived risk of adverse events.

A time series modeling approach in risk appraisal of violent and sexual recidivism

For over half a century, various clinical and actuarial methods have been employed to assess the likelihood of violent recidivism. Yet there is a need for new methods that can improve the accuracy of recidivism predictions. This study proposes a new time series modeling approach that generates high levels of predictive accuracy over short and long periods of time. The proposed approach outperformed two widely used actuarial instruments (i.e., the Violence Risk Appraisal Guide and the Sex Offender Risk Appraisal Guide). Furthermore, analysis of temporal risk variations based on specific time series models can add valuable information into risk assessment and management of violent offenders.

Improving risk assessment of violence among military veterans: an evidence-based approach for clinical decision-making

Increased media attention to post-deployment violence highlights the need to develop effective models to guide risk assessment among military Veterans. Ideally, a method would help identify which Veterans are most at risk for violence so that it can be determined what could be done to prevent violent behavior. This article suggests how empirical approaches to risk assessment used successfully in civilian populations can be applied to Veterans. A review was conducted of the scientific literature on Veteran populations regarding factors related to interpersonal violence generally and to domestic violence specifically. A checklist was then generated of empirically-supported risk factors for clinicians to consider in practice. To conceptualize how these known risk factors relate to a Veteran's violence potential, risk assessment scholarship was utilized to develop an evidence-based method to guide mental health professionals. The goals of this approach are to integrate science into practice, overcome logistical barriers, and permit more effective assessment, monitoring, and management of violence risk for clinicians working with Veterans, both in Department of Veteran Affairs settings and in the broader community. Research is needed to test the predictive validity of risk assessment models. Ultimately, the use of a systematic, empirical framework could lead to improved clinical decision-making in the area of risk assessment and potentially help prevent violence among Veterans.

Psychopathy: assessment and forensic implications

Psychopathy is commonly viewed as a personality disorder defined by a cluster of interpersonal, affective, lifestyle, and antisocial traits and behaviours, including grandiosity, egocentricity, deceptiveness, shallow emotions, lack of empathy or remorse, irresponsibility, impulsivity, and a tendency to violate social norms. In our article, we outline standard methods for the assessment of psychopathy, its association with antisocial personality disorder (ASPD), and its implications for clinical and forensic issues, including crime and violence, risk assessment, and treatment options.

Cognitive behaviour therapy for violent men with antisocial personality disorder in the community: an exploratory randomized controlled trial

BACKGROUND

Little information exists on treatment effectiveness in antisocial personality disorder (ASPD). We investigated the feasibility and effectiveness of carrying out a randomized controlled trial of cognitive behaviour therapy (CBT) in men with ASPD who were aggressive.

METHOD

This was an exploratory two-centre, randomized controlled trial in a community setting. Fifty-two adult men with a diagnosis of ASPD, with acts of aggression in the 6 months prior to the study, were randomized to either treatment as usual (TAU) plus CBT, or usual treatment alone. Change over 12 months of follow-up was assessed in the occurrence of any act of aggression and also in terms of alcohol misuse, mental state, beliefs and social functioning.

RESULTS

The follow-up rate was 79%. At 12 months, both groups reported a decrease in the occurrence of any acts of verbal or physical aggression. Trends in the data, in favour of CBT, were noted for problematic drinking, social functioning and beliefs about others.

CONCLUSIONS

CBT did not improve outcomes more than usual treatment for men with ASPD who are aggressive and living in the community in this exploratory study. However, the data suggest that a larger study is required to fully assess the effectiveness of CBT in reducing aggression, alcohol misuse and improving social functioning and view of others. It is feasible to carry out a rigorous randomized controlled trial in this group.

Examining the incidence of and time to recidivism within the risk contingency framework: a 20-year follow up study

This study examined the extent to which the contingencies risk framework (consisting of dispositional, historical, contextual, and clinical domains) predicted detected recidivism (i.e., arrest or conviction). Secondary data were extracted on 413 prisoners who underwent a psychiatric evaluation, were assessed on the risk domains, and followed up over 20 years. There were 273 (66.1%) cases of recidivism for serious offenses (e.g., assaults). Criminal career outcomes examined included: years to and the incidence of recidivism. Statistics showed that chance classification of the incidence of recidivism was 21% more accurate for the recursive partitioning than the bilinear model. These results are consistent with the contingencies risk framework, support its use over linear models, and highlight its predictive utility.

Prevalence and correlates of psychopathic traits in the household population of Great Britain

There are no previous surveys of psychopathy and psychopathic traits in representative general population samples using standardized instruments. This study aimed to measure prevalence and correlates of psychopathic traits, based on a two-phase survey using the Psychopathy Checklist: Screening Version (PCL: SV) in 638 individuals, 16-74 years, in households in England, Wales and Scotland. The weighted prevalence of psychopathy was 0.6% (95% CI: 0.2-1.6) at a cut score of 13, similar to the noncriminal/nonpsychiatric sample described in the manual of the PCL: SV. Psychopathy scores correlated with: younger age, male gender; suicide attempts, violent behavior, imprisonment and homelessness; drug dependence; histrionic, borderline and adult antisocial personality disorders; panic and obsessive-compulsive disorders. This survey demonstrated that, as measured by the PCL: SV, psychopathy is rare, affecting less than 1% of the household population, although it is prevalent among prisoners, homeless persons, and psychiatric admissions. There is a half-normal distribution of psychopathic traits in the general population, with the majority having no traits, a significant proportion with non-zero values, and a severe subgroup of persons with multiple associated social and behavioral problems. This distribution has implications for research into the etiology of psychopathy and its implications for society.

Can Violent (Re)offense be Predicted? A Review of the Role of the Clinician and Use of Actuarial Tests in Light of New Data

This article is an evaluation of established actuarial probation-parole tests in light of new data on violent and homicidal behavior. Probation–parole tests originally were developed by observing risks related with recidivism or return to court after release (i.e., the “danger” of releasing an individual) by following offenders in court records for up to 10 years. Commonly used probation-parole tests together comprise 82 distinct items related to characteristics of the offender: home, school, peers. Job, family, individual-medical, community, and court contacts. The risks for violence and homicide found by Zagar, et al. were compared with prior meta-analyses of risks with the criterion of violent delinquency. Bootstrapped logistic regressions in Zagar and colleagues' new data yielded highly accurate predictions of violence, showing that improved methods and sampling can lead to still higher accuracy than had been achieved by established probation-parole tests. A general discussion of the usefulness of actuarial tests and answers to challenges of their validity for decision making are provided.

Issues regarding the clinical use of the Classification of Violence Risk (COVR) assessment instrument

The MacArthur Study of Mental Disorder and Violence produced an instrument for classifying hospitalized psychiatric patients according to their risk of behaving violently following discharge. The instrument, Classification of Violence Risk (COVR) has been computerized and is now commercially available to clinicians. A validation study performed by the original researchers showed that when the instrument was applied to a new sample of patients, it demonstrated a considerable reduction in positive predictive power. Potential factors affecting the instrument's accuracy in applied settings are reviewed. It is concluded that, until additional research clarifies uncertainty about the instrument, clinicians would do well to be very cautious in utilizing COVR results to make judgments as to violence risk, particularly when the test results suggest a high risk of future violence.

Utility of R0 as a predictor of disease invasion in structured populations

Early theoretical work on disease invasion typically assumed large and well-mixed host populations. Many human and wildlife systems, however, have small groups with limited movement among groups. In these situations, the basic reproductive number, R0, is likely to be a poor predictor of a disease pandemic because it typically does not account for group structure and movement of individuals among groups. We extend recent work by combining the movement of hosts, transmission within groups, recovery from infection and the recruitment of new susceptibles into a stochastic model of disease in a host metapopulation. We focus on how recruitment of susceptibles affects disease invasion and how population structure can affect the frequency of superspreading events (SSEs). We show that the frequency of SSEs may decrease with the reduced movement and the group sizes due to the limited number of susceptible individuals available. Classification tree analysis of the model results illustrates the hierarchical nature of disease invasion in host metapopulations. First, the pathogen must effectively transmit within a group (R0>1), and then the pathogen must persist within a group long enough to allow for movement among the groups. Therefore, the factors affecting disease persistence--such as infectious period, group size and recruitment of new susceptibles--are as important as the local transmission rates in predicting the spread of pathogens across a metapopulation.

Are violence risk assessment tools clinically useful?

Despite increasing concerns regarding the prevalence of violent behaviour in mainstream mental health settings, the impressive body of forensic research on violence risk assessment has thus far had only limited impact on front-line general mental health practice. The common objection raised by clinicians that risk assessment tools lack utility for clinical practice may contribute to this. The present paper argues that this objection, although understandable, is misplaced. Usage of appropriate, validated risk assessment tools can augment standard clinical approaches in a number of ways. Some of their advantages derive simply from having a well-structured approach, others from consideration of specific kinds of risk factors: 'static' and 'dynamic'. The inappropriate use of tools without a firm evidence base, however, is unlikely to enhance clinical practice significantly.

Identifying subgroups at high risk of dropping out of domestic batterer treatment: the buffering effects of a high school education

This study identifies the subgroups of domestic batterers who are at a low or high risk of failing to complete domestic batterer cognitive behavioral treatment. The sample is composed of 355 domestic batterers ordered to complete treatment, with 31.8% not completing treatment. Three subgroups of batterers were identified as having at least a 60% chance of treatment failure: (a) unemployed generalized aggressors, (b) high school dropouts ordered into substance abuse treatment, and (c) unemployed offenders ordered into substance abuse treatment. Furthermore, a high school education, even when offenders are unemployed or living in poverty, buffers the negative effects of a substance abuse problem among court-mandated batterers required to participate in both domestic violence treatment and substance abuse treatment. Two thirds of substance-abusing high school graduates completed both domestic violence and substance abuse treatment, compared to only one third of the substance-abusing high school dropouts. Implications are discussed.

Predicting sexual aggression: the role of pornography in the context of general and specific risk factors

The main focus of the present study was to examine the unique contribution (if any) of pornography consumption to men's sexually aggressive behavior. Even after controlling for the contributions of risk factors associated with general antisocial behavior and those used in Confluence Model research as specific predictors of sexual aggression, we found that high pornography consumption added significantly to the prediction of sexual aggression. Further analyses revealed that the predictive utility of pornography was due to its discriminative ability only among men classified (based on their other risk characteristics) at relatively high risk for sexual aggression. Other analyses indicated that the specific risk factors accounted for more variance in sexual aggression than the general risk factors and mediated the association between the general risk factors and sexual aggression. We illustrate the potential application of the findings for risk assessment using a classification tree.

Treatment engagement and violence risk in mental disorders

BACKGROUND

Research has uncovered many characteristics related to violence committed by people with mental illness. However, relatively few studies have focused on understanding the connection between violence and dynamic, malleable variables such as a patient's level of treatment engagement.

AIMS

To explore the link between community violence and patients'beliefs about psychiatric treatment benefit.

METHOD

A sample of 1011 adults receiving out-patient treatment for a psychiatric disorder in the public mental health systems of five US states were interviewed.

RESULTS

Bivariate analyses revealed community violence was inversely related to treatment adherence, perceived treatment need and perceived treatment effectiveness. Multivariate analyses showed these three variables were associated with reduced odds of violent and other aggressive acts.

CONCLUSIONS

The results suggest clinical consideration of patients' perceptions of treatment benefit can help enhance violence risk assessment in psychiatric practice settings.

Psychopathy: a clinical and forensic overview

There is a substantial amount of empirical evidence that psychopathy, as measured by the PCL-R and its derivatives, is a predictor of recidivism and violence in prison, forensic psychiatric, and civil psychiatric populations. The PCL-R is one of the most generalizable of the risk factors identified thus far, and for this reason it is included in various actuarial and structured clinical risk assessment procedures. Although psychopathy is not the only risk factor for recidivism and violence, it is too important to ignore, particularly with respect to violence. Treatment and management are difficult, time-consuming, and expensive, but new initiatives based on current theory and research on psychopathy and the most effective correctional philosophies may help to reduce the harm done by psychopaths.

Regression Tree Approach to Studying Factors Influencing Acoustic Voice Analysis

Multiple factors influence voice quality measurements (VQM) obtained during an acoustic voice assessment including: gender, intrasubject variability, microphone, environmental noise (type and level), data acquisition (DA) system, and analysis software. This study used regression trees to investigate the order and relative importance of these factors on VQM including interaction effects of the factors and how the outcome differs when the acoustic environment is controlled for noise. Twenty normophonic participants provided 20 voice samples each, which were recorded synchronously on five DA systems combined with six different microphones. The samples were mixed with five noise types at eight signal-to-noise ratio (SNR) levels. The resulting 80,000 audio samples were analyzed for fundamental frequency (F(0)), jitter and shimmer using three software analysis systems: MDVP, PRAAT, and TF32 (CSpeech). Fifteen regression trees and their Variable Importance Measures were utilized to analyze the data. The analyses confirmed that all of the factors listed above were influential. The results suggest that gender, intrasubject variability, and microphone were significant influences on F(0). Software systems and gender were highly influential on measurements of jitter and shimmer. Environmental noise was shown to be the prominent factor that affects VQM when SNR levels are below 30 dB.

Psychopathy: theory, measurement, and treatment

The purpose of this article is to link empirical literature to the theoretical background of the concept of psychopathy and the impact that this has had on the development of treatment and intervention procedures for psychopathic offenders. This article begins with a discussion of the different theories of psychopathy, which leads into considerations of different developmental pathways of psychopathy in the individual. The discussion will then lead on to the psychometrics and measurement tools used to assess psychopathy in the individual. The measurement section will primarily be focused on Hare's Psychopathy Checklist-Revised (PCL-R), as this is the most frequently used and validated measure of psychopathy. The relationship between psychopathy and different types of crime is also discussed. The final section of the article considers the treatment and interventions that are available to psychopathic offenders and the implications this has for future research.

Contacts with public services, with special reference to mental health care, preceding a serious crime: a retrospective study of 268 subjects of forensic psychiatric investigations

Between July 1995 and June 1996, 533 subjects underwent forensic psychiatric investigation I Sweden. Odd case numbers (n = 268) were included in the study. Of these 268 people, 50% had been in contact with psychiatric services during the six-month period preceding the crime leading to forensic psychiatric assessment. Contacts with psychiatric services during the six-month period preceding the crime were significantly more common in three categories of individuals than others. These were: women, individuals who were diagnosed as having a psychotic disorder during the forensic psychiatric investigation, and individuals found to suffer from a serious mental disorder as defined in the legislation. Subjects who were found not to suffer from a serious mental disorder and were thus not eligible for special sanctions on medicolegal grounds had significantly less psychiatric contacts before the crime, as did subjects under 20 years of age. However, these two groups still had considerably more psychiatric contacts than the general population. The present results suggest that the patient category studied requires special monitoring and case management in general psychiatry in Sweden. To this end, we call for closer studies of high-risk individuals, particularly of previously violent offenders and potentially violent offenders, and closer studies of their psychiatric contacts. This will provide a basis for the development of adequate programs and guidelines for effective care and treatment within the psychiatric sector.

The need for the development and evaluation of preventive psychosocial forensic interventions in mainstream adult community mental health services

OBJECTIVE

This paper provides a selective review of forensic research with the aim of making recommendations for the development and evaluation of psychosocial interventions for the mainstream community mental health setting to address the needs of patients with a history of offending.

CONCLUSION

Mainstream community mental health services can be guided by existing findings in the design of psychosocial intervention and prevention programmes. There is growing need to develop and evaluate such interventions.