Developing a clinically useful actuarial tool for assessing violence risk

A Scoping Review of Patient Involvement in Violence Risk Assessment

OBJECTIVE

This scoping review aimed to summarize the published literature on patient involvement in violence risk assessment. Two research questions reviewed the extent of patient involvement and what evidence exists.

INCLUSION CRITERIA

English-language peer-reviewed published articles of any methodology related to violence risk assessment toward others were included. Articles were related to forensic and mental health practice and involve patients directly in the process.

METHODS

Five electronic databases were comprehensively searched, as well as the reference lists of included articles. Both authors reviewed articles for inclusion and extracted data from included articles.

RESULTS

Fifteen articles met the inclusion criteria. Articles reported on three approaches to patient engagement in structured violence risk assessment: how patients were involved or experienced the process, using rating scales, and using questions related to patient self-perceived risk. In relation to what evidence existed, four main themes emerged: patient views about risk and their involvement in risk assessment, comparing the predictive accuracy of patient self-rated tools with clinician-rated tools, predictive accuracy of a patient self-rated tool, and comparing risk ratings between patients and clinicians.

CONCLUSIONS

There is a dearth of research published about involving patients in their own risk assessment. Patients report both positive and negative experiences of the process. From cohort-type studies, results have shown that patient self-risk assessment can have a similar predictive ability to the clinician ratings related to adverse violence outcomes. Findings from studies can pave the way for future clinical research around the tools that have been developed thus far.

Risk assessment for aggressive behaviour in schizophrenia

BACKGROUND

Aggressive or violent behaviour is often associated with people with schizophrenia in common perceptions of the disease. Risk assessment methods have been used to identify and evaluate the behaviour of those individuals who are at the greatest risk of perpetrating aggression or violence or characterise the likelihood to commit acts. Although many different interventions have been developed to decrease aggressive or violent incidences in inpatient care, staff working in inpatient settings seek easy-to-use methods to decrease patient aggressive events. However, many of these are time-consuming, and they require intensive training for staff and patient monitoring. It has also been recognised in clinical practice that if staff monitor patients' behaviour in a structured manner, the monitoring itself may result in a reduction of aggressive/violent behaviour and incidents in psychiatric settings.

OBJECTIVES

To assess the effects of structured aggression or violence risk assessment methods for people with schizophrenia or schizophrenia-like illnesses.

SEARCH METHODS

We searched the Cochrane Schizophrenia Group's Study-Based Register of Trials, which is based on CENTRAL, MEDLINE, Embase, CINAHL, PsycINFO, PubMed, ISRCTN registry, ClinicalTrials.gov, and WHO ICTRP, on 10 February 2021. We also inspected references of all identified studies.

SELECTION CRITERIA

We included all randomised controlled trials (RCTs) comparing structured risk assessment methods added to standard professional care with standard professional care for the evaluation of aggressive or violent behaviour among people with schizophrenia.

DATA COLLECTION AND ANALYSIS

At least two review authors independently inspected citations, selected studies, extracted data, and appraised study quality. For binary outcomes, we calculated a standard estimation of the risk ratio (RR) and its 95% confidence interval (CI). For continuous outcomes, we calculated the mean difference (MD) and its 95% CI. We assessed risk of bias in the included studies and created a summary of findings table using the GRADE approach.

MAIN RESULTS

We included four studies in the review. The total number of participants was not identifiable, as some studies provided number of participants included, and some only patient days. The studies compared a package of structured assessment methods with a control group that included routine nursing care and drug therapy or unstructured psychiatric observations/treatment based on clinical judgement. In two studies, information about treatment in control care was not available. One study reported results for our primary outcome, clinically important change in aggressive/violent behaviour, measured by the rate of severe aggression events. There was likely a positive effect favouring structured risk assessment over standard professional care (RR 0.59, 95% CI 0.41 to 0.85; 1 RCT; 1852 participants; corrected for cluster design: RR 0.59, 95% CI 0.37 to 0.93; moderate-certainty evidence). One trial reported data for the use of coercive measures (seclusion room). Compared to standard professional care, structured risk assessment may have little or no effect on use of seclusion room as days (corrected for cluster design: RR 0.92, 95% CI 0.27 to 3.07; N = 20; low-certainty evidence) or use of seclusion room as secluded participants (RR 1.83, 95% CI 0.39 to 8.7; 1 RCT; N = 20; low-certainty evidence). However, seclusion room may be used less frequently in the standard professional care group compared to the structured risk assessment group (incidence) (corrected for cluster design: RR 1.63, 95% CI 0.49 to 5.47; 1 RCT; N = 20; substantial heterogeneity, Chi2 = 0.0; df = 0.0; P = 0.0; I2 = 100%; low-certainty evidence). There was no evidence of a clear effect on adverse events of escape (RR 0.2, 95% CI 0.01 to 4.11; 1 RCT; n = 200; very low-certainty evidence); fall down (RR 0.33, 95% CI 0.04 to 3.15; 1 RCT; n = 200; very low-certainty evidence); or choking (RR 0.2, 95% CI 0.01 to 4.11; 1 RCT; n = 200; very low-certainty evidence) when comparing structured risk assessment to standard professional care. There were no useable data for patient-related outcomes such as global state, acceptance of treatment, satisfaction with treatment, quality of life, service use, or costs.

AUTHORS' CONCLUSIONS

Based on the available evidence, it is not possible to conclude that structured aggression or violence risk assessment methods are effective for people with schizophrenia or schizophrenia-like illnesses. Future work should combine the use of interventions and structured risk assessment methods to prevent aggressive incidents in psychiatric inpatient settings.

Psychometric properties of an Arabic translation of the long (12 items) and short (7 items) forms of the Violent Ideations Scale (VIS) in a non-clinical sample of adolescents

BACKGROUND

Accurately measuring violent ideations would be of particular high relevance in Arab countries, which are witnessing an unprecedented increase in violence rates among adolescents because of the widespread social, economic and political unrest in the region. Therefore, the present study aimed to test the psychometric properties of an Arabic translation of the 12-item and the 7-item forms of the Violent Ideations Scale (VIS and VIS-SF) in a sample on non-clinical Arabic-speaking adolescents.

METHODS

Five hundred seventy-seven community adolescents (mean age of 15.90 ± 1.73 years, 56.5% females) answered an anonymous online survey comprising an Arabic translation of the Violent Ideations Scale (VIS) and a measure of physical aggression.

RESULTS

Confirmatory factor analyses (CFA) offered support for the single-factor structure of the Arabic VIS and the VIS-SF. Both the long and short forms of the scale yielded excellent internal consistency, with McDonald's ω coefficients of 0.96 and 0.94 and Cronbach's α coefficients of 0.96 and 0.94, respectively. Multi-group CFA established measurement invariance across gender groups. Finally, results revealed significant and positive correlations between the two forms of the VIS and physical aggression scores, thus supporting concurrent validity.

CONCLUSION

Both the VIS and VIS-SF have demonstrated good psychometric properties in their Arabic versions, and suitability for sound assessment of violent ideations. We therefore expect that these measures assist clinicians in risk assessment and management of violence, and help foster research in this area in Arab countries.

An analysis of Chinese nursing electronic medical records to predict violence in psychiatric inpatients using text mining and machine learning techniques

BACKGROUND

The prevalence of violence in acute psychiatric wards is a critical concern. According to a meta-analysis investigating violence in psychiatric inpatient units, researchers estimated that approximately 17% of inpatients commit one or more acts of violence during their stay. Inpatient violence negatively affects health-care providers and patients and may contribute to high staff turnover. Therefore, predicting which psychiatric inpatients will commit violence is of considerable clinical significance.

OBJECTIVE

The present study aimed to estimate the violence rate for psychiatric inpatients and establish a predictive model for violence in psychiatric inpatients.

METHODS

We collected the structured and unstructured data from Chinese nursing electronic medical records (EMRs) for the violence prediction. The data was obtained from the psychiatry department of a regional hospital in southern Taiwan, covering the period between January 2008 and December 2018. Several text mining and machine learning techniques were employed to analyze the data.

RESULTS

The results demonstrated that the rate of violence in psychiatric inpatients is 19.7%. The patients with violence in psychiatric wards were generally younger, had a more violent history, and were more likely to be unmarried. Furthermore, our study supported the feasibility of predicting aggressive incidents in psychiatric wards by using nursing EMRs and the proposed method can be incorporated into routine clinical practice to enable early prediction of inpatient violence.

CONCLUSIONS

Our findings may provide clinicians with a new basis for judgment of the risk of violence in psychiatric wards.

The effects of emotion and juvenile diagnoses on parole release decisions: An experimental approach

Parole boards are often required to make many release recommendations after reviewing a substantial amount of information in a short timeframe. While making release decisions, parole board members might be motivated to sacrifice accuracy and, instead, use heuristics, such as their emotions. Emotions might increase the reliance on risk or threat related information, such as an inmate's mental illness status. The current study applies the appraisal tendency framework of emotion to assess the impact of emotion on parole decisions involving juvenile mental health diagnoses. Using a 3(emotion: anger, compassion, and control) x 4(mental illness: conduct disorder, oppositional defiant disorder, depression, and control) factorial design, this study examined the extent to which mock parole board members' emotions impacted evaluations of inmates with juvenile mental illness diagnoses and subsequent release decisions. Results indicated that there was no effect of emotion on parole decisions. However, the inmates' mental illnesses did play a role in parole release decisions. Specifically, parole candidates with depression were released on parole at higher rates compared to conduct disorder, oppositional defiant disorder, and the control condition. Policy implications are discussed.

Predicting criminal and violent outcomes in psychiatry: a meta-analysis of diagnostic accuracy

Although reducing criminal outcomes in individuals with mental illness have long been a priority for governments worldwide, there is still a lack of objective and highly accurate tools that can predict these events at an individual level. Predictive machine learning models may provide a unique opportunity to identify those at the highest risk of criminal activity and facilitate personalized rehabilitation strategies. Therefore, this systematic review and meta-analysis aims to describe the diagnostic accuracy of studies using machine learning techniques to predict criminal and violent outcomes in psychiatry. We performed meta-analyses using the mada, meta, and dmetatools packages in R to predict criminal and violent outcomes in psychiatric patients (n = 2428) (Registration Number: CRD42019127169) by searching PubMed, Scopus, and Web of Science for articles published in any language up to April 2022. Twenty studies were included in the systematic review. Overall, studies used single-nucleotide polymorphisms, text analysis, psychometric scales, hospital records, and resting-state regional cerebral blood flow to build predictive models. Of the studies described in the systematic review, nine were included in the present meta-analysis. The area under the curve (AUC) for predicting violent and criminal outcomes in psychiatry was 0.816 (95% Confidence Interval (CI): 70.57-88.15), with a partial AUC of 0.773, and average sensitivity of 73.33% (95% CI: 64.09-79.63), and average specificity of 72.90% (95% CI: 63.98-79.66), respectively. Furthermore, the pooled accuracy across models was 71.45% (95% CI: 60.88-83.86), with a tau squared (τ²) of 0.0424 (95% CI: 0.0184-0.1553). Based on available evidence, we suggest that prospective models include evidence-based risk factors identified in prior actuarial models. Moreover, there is a need for a greater emphasis on identifying biological features and incorporating novel variables which have not been explored in prior literature. Furthermore, available models remain preliminary, and prospective validation with independent datasets, and across cultures, will be required prior to clinical implementation. Nonetheless, predictive machine learning models hold promise in providing clinicians and researchers with actionable tools to improve how we prevent, detect, or intervene in relevant crime and violent-related outcomes in psychiatry.

Psychopathy subdomains in violent offenders with and without a psychotic disorder

BACKGROUND

Violence in psychosis has been linked to antisocial behavior and psychopathy traits. Psychopathy comprises aspects of interpersonal, affective, lifestyle, and antisocial traits which may be differently involved in violent offending by persons with psychotic disorders. We explored psychopathy subdomains among violent offenders with and without a psychotic disorder.

METHODS

46 males, with a history of severe violence, with (n = 26; age 35.85 ± 10.34 years) or without (n = 20; age 39.10 ± 11.63 years) a diagnosis of a psychotic disorder, were assessed with the Psychopathy Checklist-Revised (PCL-R). PCL-R was split into subdomains following the four-facet model. Group differences in total and subdomain scores were analyzed with a general linear model with covariates.

RESULTS

Total PCL-R scores did not differ between the groups (p = 0.61, Cohen's d = 0.17). The violent offenders without psychotic disorders had higher facet 2 scores than the patient group with psychotic disorders (p = 0.029, Cohen's d = 0.77). Facet 1, 3, or 4 scores did not differ between the groups. Controlling for age did not alter the results.

CONCLUSION

Patients with a psychotic disorder and a history of severe violence have lower affective psychopathy scores than violent offenders without psychotic disorders. This observation may point toward distinct underlying mechanisms for violence and may provide a target for focused treatment and prevention.

The age of violence: Mapping brain age in psychosis and psychopathy

Young chronological age is one of the strongest predictors for antisocial behaviour in the general population and for violent offending in individuals with psychotic disorders. An individual's age can be predicted with high accuracy using neuroimaging and machine-learning. The deviation between predicted and chronological age, i.e., brain age gap (BAG) has been suggested to reflect brain health, likely relating partly to neurodevelopmental and aging-related processes and specific disease mechanisms. Higher BAG has been demonstrated in patients with psychotic disorders. However, little is known about the brain-age in violent offenders with psychosis and the possible associations with psychopathy traits. We estimated brain-age in 782 male individuals using T1-weighted MRI scans. Three machine learning models (random forest, extreme gradient boosting with and without hyper parameter tuning) were first trained and tested on healthy controls (HC, n = 586). The obtained BAGs were compared between HC and age matched violent offenders with psychosis (PSY-V, n = 38), violent offenders without psychosis (NPV, n = 20) and non-violent psychosis patients (PSY-NV, n = 138). We ran additional comparisons between BAG of PSY-V and PSY-NV and associations with Positive and Negative Syndrome Scale (PANSS) total score as a measure of psychosis symptoms. Psychopathy traits in the violence groups were assessed with Psychopathy Checklist-revised (PCL-R) and investigated for associations with BAG. We found significantly higher BAG in PSY-V compared with HC (4.9 years, Cohen'sd = 0.87) and in PSY-NV compared with HC (2.7 years, d = 0.41). Total PCL-R scores were negatively associated with BAG in the violence groups (d = 1.17, p < 0.05). Additionally, there was a positive association between psychosis symptoms and BAG in the psychosis groups (d = 1.12, p < 0.05). While the significant BAG differences related to psychosis and not violence suggest larger BAG for psychosis, the negative associations between BAG and psychopathy suggest a complex interplay with psychopathy traits. This proof-of-concept application of brain age prediction in severe mental disorders with a history of violence and psychopathy traits should be tested and replicated in larger samples.

Associations between amygdala nuclei volumes, psychosis, psychopathy, and violent offending

The amygdala is involved in fear perception and aggression regulation, and smaller volumes have been associated with psychotic and non-psychotic violence. We explored the relationship between amygdala nuclei volumes in violent offenders with and without psychosis, and the association to psychopathy traits. 3T MRI scans (n = 204, males, 18-66 years) were obtained from psychotic violent offenders (PSY-V, n = 29), non-psychotic violent offenders (NPV, n = 19), non-violent psychosis patients (PSY-NV, n = 67), and healthy controls (HC, n = 89). Total amygdala and 9 amygdala nuclei volumes were obtained with FreeSurfer. Psychopathy traits were measured with the Psychopathy Checklist-revised (PCL-R). Multivariate analyses explored diagnostic differences in amygdala nuclei volumes and associations to psychosis, violence, and psychopathy traits. PSY-V had a smaller basal nucleus, anterior amygdaloid area, and cortical amygdalar transition area (CATA), whereas PSY-NV had a smaller CATA than HC. Volumes in NPV did not differ from HC, and there were no associations between PCL-R total or factor scores and any of the nuclei or whole amygdala volumes. The lower volumes of amygdala nuclei involved in fear modulation, stress responses, and social interpretation may point towards some mechanisms of relevance to violence in psychosis, but the results warrant replication in larger subject samples.

White matter microstructure in schizophrenia patients with a history of violence

Schizophrenia (SCZ) is associated with increased risk of violence compared to the general population. Neuroimaging research suggests SCZ to be a disorder of disrupted connectivity, with diffusion tensor imaging (DTI) indicating white matter (WM) abnormalities. It has been hypothesized that SCZ patients with a history of violence (SCZ-V) have brain abnormalities distinguishing them from SCZ patients with no history of violence (SCZ-NV). Yet, a thorough investigation of the neurobiological underpinnings of state and trait measures of violence and aggression in SCZ derived from DTI indices is lacking. Using tract-based spatial statistics, we compared DTI-derived microstructural indices: fractional anisotropy (FA), mean, axial (AD) and radial diffusivity across the brain; (1) between SCZ-V (history of murder, attempted murder, or severe assault towards other people, n = 24), SCZ-NV (n = 52) and healthy controls (HC, n = 94), and (2) associations with current aggression scores among both SCZ groups. Then, hypothesis-driven region of interest analyses of the uncinate fasciculus and clinical characteristics including medication use were performed. SCZ-V and SCZ-NV showed decreased FA and AD in widespread regions compared to HC. There were no significant differences on any DTI-based measures between SCZ-V and SCZ-NV, and no significant associations between state or trait measures of aggression and any of the DTI metrics in the ROI analyses. The DTI-derived WM differences between SCZ and HC are in line with previous findings, but the results do not support the hypothesis of specific brain WM microstructural correlates of violence or aggression in SCZ.

Validation of the English Language Version of the Violent Ideations Scale

This study used a within-participant design to evaluate the concurrent validity and test-retest reliability of the Violent Ideations Scale in a general population, English-speaking opportunistic sample. Data from 116 adult participants (M age = 33.7, SD = 11.9, male = 30 [25.9%]) were used to compare scores on the Violent Ideations Scale and Aggression Questionnaire and responses to the Schedule of Imagined Violence. A subgroup of 27 participants (M age = 37.2, SD = 13.6, male = 8 [29.6%]) completed the Violent Ideations Scale on a second occasion, 2 weeks later. The Violent Ideations Scale was found to correlate significantly with the Aggression Questionnaire subscale and total scores, with the strongest correlations being with physical aggression and total scores. Participants were more likely to be categorized as having experienced a violent ideation based on responses to the Violent Ideation Scale, compared with the Schedule of Imagined Violence, most likely due to the Schedule of Imagined Violence underestimating the prevalence of violent ideation. A significant, strong correlation was found between total Violent Ideations Scale scores at Time 1 and Time 2. Overall, the Violent Ideations Scale was found to have concurrent validity when compared with the Aggression Questionnaire and good test-retest reliability, suggesting that it would be suitable for use with a nonclinical, English-speaking sample.

The Deficits of Individual Morphological Covariance Network Architecture in Schizophrenia Patients With and Without Violence

Background: Schizophrenia is associated with a significant increase in the risk of violence, which constitutes a public health concern and contributes to stigma associated with mental illness. Although previous studies revealed structural and functional abnormalities in individuals with violent schizophrenia (VSZ), the neural basis of psychotic violence remains controversial. Methods: In this study, high-resolution structural magnetic resonance imaging (MRI) data were acquired from 18 individuals with VSZ, 23 individuals with non-VSZ (NSZ), and 22 age- and sex-matched healthy controls (HCs). Whole-brain voxel-based morphology and individual morphological covariance networks were analysed to reveal differences in gray matter volume (GMV) and individual morphological covariance network topology. Relationships among abnormal GMV, network topology, and clinical assessments were examined using correlation analyses. Results: GMV in the hypothalamus gradually decreased from HCs and NSZ to VSZ and showed significant differences between all pairs of groups. Graph theory analyses revealed that morphological covariance networks of HCs, NSZ, and VSZ exhibited small worldness. Significant differences in network topology measures, including global efficiency, shortest path length, and nodal degree, were found. Furthermore, changes in GMV and network topology were closely related to clinical performance in the NSZ and VSZ groups. Conclusions: These findings revealed the important role of local structural abnormalities of the hypothalamus and global network topological impairments in the neuropathology of NSZ and VSZ, providing new insight into the neural basis of and markers for VSZ and NSZ to facilitate future accurate clinical diagnosis and targeted treatment.

Strategies for Assessing and Preventing Inpatient Violence in Forensic Hospitals: A Call for Specificity

Abstract. Violence within inpatient forensic hospitals is a significant and enduring problem that leads to harm to staff and patients and causes significant expenditures. This paper provides comprehensive recommendations for developing and implementing violence reduction strategies within forensic settings that are predicated on appropriate evaluation for violence risk. This paper posits that proper strategies must take into account subtypes of violence and classifying risk with systematic and continuous evaluations. Treatment interventions should be geared to patients most at-risk for violence. By recognizing the dynamic nature of violence, hospital administrators can work closely with institution staff to provide support for improving the environment of forensic hospitals. By employing empirically based treatment interventions on both acute and long-term units, forensic hospitals can provide a safer environment.

Hippocampal subfield and amygdala nuclei volumes in schizophrenia patients with a history of violence

Schizophrenia (SCZ) is associated with an increased risk of violence compared to the general population. Previous studies have indicated smaller hippocampal and amygdala volumes in violent than non-violent psychotic patients^. However, little is known about volumetric differences at the subdivision level of these structures. In the present study, hippocampal subfields and amygdala nuclei volumes were estimated with FreeSurfer from 3 T MRI of SCZ patients with (SCZ-V, n = 24) and without (SCZ-NV, n = 51) a history of severe violence and 90 healthy controls (HC). Volumetric differences between groups were explored with a general linear model covarying for confounders, in addition to follow-up analyses in patient groups controlling for clinical characteristics such as antipsychotic medication, duration of illness and illicit substance use. SCZ-V had smaller total hippocampal volume and smaller CA1, HATA, fimbria, and molecular layer of DG volumes compared to HC. Total amygdala volume together with basal nucleus, accessory basal nucleus, CTA, and paralaminar nucleus volumes were smaller in SCZ-V compared to HC. In SCZ-NV, compared to HC, the observed smaller volumes were limited to basal and paralaminar nucleus. There were no significant differences in hippocampal subfield and amygdala nuclei volumes between SCZ-V and SCZ-NV. Follow-up analyses showed that the results in patient groups were not affected by clinical characteristics. The results suggest that smaller hippocampal subfield and amygdala nuclei volumes may be relevant to violence risk in SCZ. However, the neurobiological signature of violence in SCZ should be further investigated in larger cohorts.

The Classification of Violence Risk

(Reprinted with permission from Behav. Sci. Law 24: 721-730, 2006).

How dangerousness evolves after court-ordered compulsory psychiatric admission: explorative prospective observational cohort study

BACKGROUND

Compulsory admission is commonly regarded as necessary and justified for patients whose psychiatric condition represents a severe danger to themselves and others. However, while studies on compulsory admissions have reported on various clinical and social outcomes, little research has focused specifically on dangerousness, which in many countries is the core reason for compulsory admission.AimsTo study changes in dangerousness over time in adult psychiatric patients admitted by compulsory court order, and to relate these changes to these patients' demographic and clinical characteristics.

METHOD

In this explorative prospective observational cohort study of adult psychiatric patients admitted by compulsory court order, demographic and clinical data were collected at baseline. At baseline and at 6 and 12 month follow-up, dangerousness was assessed using the Dangerousness Inventory, an instrument based on the eight types of dangerousness towards self or others specified in Dutch legislation on compulsory admissions. We used descriptive statistics and logistic regression to analyse the data.

RESULTS

We included 174 participants with a court-ordered compulsory admission. At baseline, the most common dangerousness criterion was inability to cope in society. Any type of severe or very severe dangerousness decreased from 86.2% at baseline to 36.2% at 6 months and to 28.7% at 12 months. Being homeless at baseline was the only variable which was significantly associated with persistently high levels of dangerousness.

CONCLUSIONS

Dangerousness decreased in about two-thirds of the patients after court-ordered compulsory admission. It persisted, however, in a substantial minority (approximately one-third).Declaration of interestNone.

Characteristics of a suicide attempt predict who makes another attempt after hospital discharge: A decision-tree investigation

The year following discharge from psychiatric hospitalization is a high-risk period for suicidal behavior, particularly among patients initially hospitalized after a suicide attempt. Demographic and clinical correlates have been identified; however, characteristics of the initial attempt may provide insight into risk for subsequent attempts as well. This investigation examined whether individual or a combination of suicide attempt characteristics predicted future attempts. Two hundred and eighteen psychiatric inpatients from the MacArthur Violence Risk Assessment Study with a recent suicide attempt were administered items from the Suicide Intent Scale and followed one year after discharge. Sixty-nine (31.65%) made a subsequent attempt. Data were analyzed by a stepwise logistic regression, followed by an iterative receiver operator curve (IROC) analysis, a recursive partitioning classification tree. The cross-validated IROC, but not logistic regression, predicted subsequent suicide attempts. Furthermore, the IROC found that participants who made definite plans and underwent extensive preparation were at highest risk for subsequent attempts. These findings suggest that suicide attempt characteristics preceding psychiatric hospitalization can help identify patients at elevated risk for another attempt post-discharge.

The use of model based recursive partitioning as an analytic tool in child welfare

Child welfare agencies are tasked with investigating allegations of child maltreatment and intervening when necessary. Researchers are turning to the field of predictive analytics to optimize data analysis and data-driven decision making. To demonstrate the utility of statistical algorithms that preceded the current predictive analytics, we used Model Based (MOB) recursive partitioning, a variant of regression analysis known as decision trees, on a dataset of cases and controls with a binary outcome of serious maltreatment (defined as hospitalization or death). We ran two models, one which split a robust set of variables significantly correlated with the outcome on the partitioning of a proxy variable for environmental poverty, and one which ran the same variable set partitioned on a variable representing confirmed prior maltreatment. Both models found that what most differentiated children was spending greater than 2% of the timeframe of interest in foster care, and that for some children, lack of Medicaid eligibility almost doubled or tripled the odds of serious maltreatment. We find that decision trees such as MOB can augment risk assessment tools and other data analyses, informing data-driven program and policy decision making. We caution that decision trees, as with any other predictive tool, must be evaluated for inherent biases that may be contained in the proxy variables and the results interpreted carefully. Predictive analytics, as a class, should be used to augment, but not replace, critical thinking in child welfare decision making.

Violent children: a framework for assessment

Violence by and towards young people has become a major public health issue. Increased lethality, more random violence and fewer safe places largely account for the high levels of fear experienced by both children and adults. In the field of child protection and domestic violence, where traditionally the child is referred as the victim not the perpetrator, child psychiatrists are well versed, practised and skilled in the assessment of children and families. However, young people are increasingly being referred to child and adolescent mental health teams for assessment because of violent acts that they have carried out. This is reflected in heavy case-loads of children with conduct disorder who have multi-morbidity and complex need. In England and Wales, health (including mental health), social care and education services are mandated to assist youth offending teams.

Evaluating risks

Mistakes are inevitable in any branch of medicine, but psychiatry is a particularly risky business (Holloway, 1997). When psychiatrists get it wrong there may serious consequences for their patients, the clinical team and the wider public. The Government introduced a series of initiatives in the 1990s: the Care Programme Approach (1990), the supervision register (Department of Health & Home Office, 1994) and supervised discharge (Secretary of State for Health, 1997). One of the main purposes of this legislation was to minimise the risk psychiatric patients pose to the community. Future service provision will be shaped by clinical governance and the National Service Framework for Mental Health (Secretary of State for Health, 1997), and evaluation and management of risk will become increasingly important.

Use and perceived utility of structured violence risk assessments in English medium secure forensic units

Aims and Method

We surveyed the usage and perceived utility of standardised risk measures in 29 forensic medium secure units (a 62% response rate).

Results

The most common instruments were Historical Clinical Risk–20 (HCR–20) and Psychopathy Checklist – revised (PCL–R); both were rated highly for utility. the Risk Matrix 2000 (RM2000), Sex Offender Risk Appraisal Guide (SORAG) and Static-99 were the most common sex offender assessments, but the Sexual Violence Risks–20 (SVR–20) was rated more positively for its use of dynamic factors and relevance to treatment.

Clinical Implications

Most medium secure units use structured risk assessments and staff view them positively. As HCR–20 and PCL–R/PCL–SV (Psychopathy Checklist – Screening Version) are so widely used they should be the first choices considered by other services.

Differences between British and Japanese perspectives on forensic mental health systems: A preliminary study

AIM

To clarify the differences in views on forensic mental health (FMH) systems between the United Kingdom and Japan.

METHODS

We conducted a series of semi-structured interviews with six leading forensic psychiatrists. Based on a discussion by the research team, we created an interview form. After we finished conducting all the interviews, we qualitatively analyzed their content.

RESULTS

In the United Kingdom the core domain of FMH was risk assessment and management; however, in Japan, the core domain of FMH was psychiatric testimony. In the United Kingdom, forensic psychiatrists were responsible for ensuring public safety, and psychopathy was identified as a disease but deemed as not suitable for medical treatment. On the other hand, in Japan, psychopathy was not considered a mental illness.

CONCLUSION

In conclusion, there are considerable differences between the United Kingdom and Japan with regard to the concepts of FMH. Some ideas taken from both cultures for better FMH practice were suggested.

Situating violent ideations within the landscape of mental health: Associations between violent ideations and dimensions of mental health

Violent ideations occur more frequently in individuals with mental health problems. They may be of interest in clinical contexts as possible indicators of dangerousness, as corollaries of mental health problems, as candidate treatment targets and as potentially playing a role in perpetuation or onset of symptoms. In spite of their relevance to mental health, some fundamental questions about their place within the broader landscape of mental health problems remain unanswered. To provide a basic characterisation of the relations between violent ideations and dimensions of mental health and provide a foundation for future research in this area we factor analysed a measure of violent ideations and an omnibus measure of mental health dimensions in a normative sample of 1306 youth (at age 17). Results supported a separate dimension of violent ideations with a small to moderate correlation with five other dimensions of mental health: internalising, prosociality, ADHD, indirect/proactive aggression, and physical/reactive aggression. Controlling for comorbidity among mental health dimensions, all but ADHD had unique relations with violent ideations. This suggests that violent ideations are potentially of broad relevance to mental health and related behaviours and there should be a greater research effort aimed at understanding their possible role in mental health.

Suicide attempts in bipolar I patients: impact of comorbid personality disorders

Objective

To evaluate the association between personality disorders (PDs) and suicide attempts (SAs) in euthymic patients with type I bipolar disorder (BD).

Methods

One-hundred twenty patients with type I BD, with and without history of SA, were evaluated during euthymia. The assessment included a clinical and sociodemographic questionnaire, the Hamilton Depression Rating Scale, the Young Mania Rating Scale, the Barratt Impulsiveness Scale, and Structured Clinical Interviews for DSM-IV Axis I and II Disorders. Logistic regression was employed to determine associations between history of SA and patient characteristics.

Results

History of SA was significantly associated with comorbid axis I disorder, rapid cycling, high impulsivity (attentional, motor, non-planning, and total), having any PD, and cluster B and C PDs. Only cluster B PDs, high attentional impulsivity, and lack of paid occupation remained significant after multivariate analysis.

Conclusions

Cluster B PDs were significantly associated with SA in patients with type I BD. High attentional impulsivity and lack of gainful employment were also associated with SA, which suggests that some cluster B clinical and social characteristics may exacerbate suicidal behavior in this population. This finding offers alternatives for new therapeutic interventions.

Improving risk management for violence in mental health services: a multimethods approach

Background

Mental health professionals increasingly carry out risk assessments to prevent future violence by their patients. However, there are problems with accuracy and these assessments do not always translate into successful risk management.

Objectives

Our aim was to improve the accuracy of assessment and identify risk factors that are causal to be targeted by clinicians to ensure good risk management. Our objectives were to investigate key risks at the population level, construct new static and dynamic instruments, test validity and construct new models of risk management using Bayesian networks.

Methods and results

We utilised existing data sets from two national and commissioned a survey to identify risk factors at the population level. We confirmed that certain mental health factors previously thought to convey risk were important in future assessments and excluded others from subsequent parts of the study. Using a first-episode psychosis cohort, we constructed a risk assessment instrument for men and women and showed important sex differences in pathways to violence. We included a 1-year follow-up of patients discharged from medium secure services and validated a previously developed risk assessment guide, the Medium Security Recidivism Assessment Guide (MSRAG). We found that it is essential to combine ratings from static instruments such as the MSRAG with dynamic risk factors. Static levels of risk have important modifying effects on dynamic risk factors for their effects on violence and we further demonstrated this using a sample of released prisoners to construct risk assessment instruments for violence, robbery, drugs and acquisitive convictions. We constructed a preliminary instrument including dynamic risk measures and validated this in a second large data set of released prisoners. Finally, we incorporated findings from the follow-up of psychiatric patients discharged from medium secure services and two samples of released prisoners to construct Bayesian models to guide clinicians in risk management.

Conclusions

Risk factors for violence identified at the population level, including paranoid delusions and anxiety disorder, should be integrated in risk assessments together with established high-risk psychiatric morbidity such as substance misuse and antisocial personality disorder. The incorporation of dynamic factors resulted in improved accuracy, especially when combined in assessments using actuarial measures to obtain levels of risk using static factors. It is important to continue developing dynamic risk and protective measures with the aim of identifying factors that are causally related to violence. Only causal factors should be targeted in violence prevention interventions. Bayesian networks show considerable promise in developing software for clinicians to identify targets for intervention in the field. The Bayesian models developed in this programme are at the prototypical stage and require further programmer development into applications for use on tablets. These should be further tested in the field and then compared with structured professional judgement in a randomised controlled trial in terms of their effectiveness in preventing future violence.

Funding

The National Institute for Health Research Programme Grants for Applied Research programme.

Development and Validation of a Brief Measure of Violent Thoughts: The Violent Ideations Scale (VIS)

Violent ideations (VIs) have potential significance across clinical, forensic, and research contexts. They feature in dominant theories of violence, are a candidate risk factor in violence prediction, and are a potential target for therapeutic intervention. Given this, there is a need for multi-item psychometrically supported measures of VIs. We report on the development and validation of the “Violent Ideations Scale” (VIS): a brief measure of VIs. In a normative sample of N = 1,276 older adolescents, we evaluated the dimensionality, sex invariance, concurrent validity, and discriminative power of the VIS. The VIS showed unidimensionality, minor measurement differences across males and females, correlated well with a preexisting measure of VIs and showed a strong relation to criminal violence. These features support the use of the VIS as a research tool and as a possible source of information regarding violence risk in clinical and forensic settings.

Comparison of patients who were violent, victimized and violent-victimized during the first year after discharge from emergency psychiatry

This prospective observational study included 345 (70%) of 489 patients discharged from an emergency psychiatric hospital during one year. Episodes of offending and victimization were recorded during first year after discharge. Forty-eight persons (14%) committed violent offenses only, 27 persons (8%) were violence victims only, and 42 persons (12%) were both offenders and victims. Significant differences in demographic and clinical variables were found between the three groups. The results pointed to two distinct groups of victims: one group with a robust offender-victim overlap and another group without offender-victim overlap. The latter group was difficult to distinguish from other discharged patients.

Innovations in Practice: Piloting a new child and adolescent risk assessment suite in the UK

BACKGROUND

A prototype risk assessment suite (FACE-CARAS) was developed for use within CAMHS and evaluated for acceptability and reliability.

METHOD

Clinicians underwent brief training in the system and invited 69 young people to an assessment using the FACE-CARAS. A second rater produced a separate set of blind ratings for most patients. Clinicians also provided qualitative feedback.

RESULTS

The component schedules of the FACE-CARAS could be reliably rated with 'near perfect' to 'moderate' agreement observed. Internal reliability consistency values, as indexed by Cronbach's alpha, were moderate to high in all cases.

CONCLUSIONS

The assessment schedules that make up the FACE-CARAS can be reliably rated by clinicians with minimal training.

Offense Characteristics of Incompetent to Stand Trial Defendants Charged With Violent Offenses

The current study compared offender and offense characteristics of pretrial defendants found incompetent to stand trial (IST) against those described as general offenders by victims in the 2008 Bureau of Justice Statistics (BJS) survey and evaluated factors that differentiated IST defendants who allegedly used weapons from those who did not during the course of a violent offense. IST defendants were older and used "weapons" more frequently than those reported in the BJS survey; however, other characteristics, including use of firearms, did not differ. No demographic, clinical, or legal factors differentiated pretrial defendants who used weapons from those who did not. Overall, pretrial defendants were frequently diagnosed with a comorbid substance use disorder, and were homeless, unemployed, and had an extensive history of psychiatric hospitalizations and prior arrests at the time of their alleged offenses. Such results indicate that models for comprehensive discharge planning may have utility in addressing the unique needs of this subgroup of mentally disordered offenders. The findings also raise questions about the federal and state prohibition of gun rights to all IST defendants.

Improving risk assessment in schizophrenia: epidemiological investigation of criminal history factors

BACKGROUND

Violence risk assessment in schizophrenia relies heavily on criminal history factors.

AIMS

To investigate which criminal history factors are most strongly associated with violent crime in schizophrenia.

METHOD

A total of 13 806 individuals (8891 men and 4915 women) with two or more hospital admissions for schizophrenia were followed up for violent convictions. Multivariate hazard ratios for 15 criminal history factors included in different risk assessment tools were calculated. The incremental predictive validity of these factors was estimated using tests of discrimination, calibration and reclassification.

RESULTS

Over a mean follow-up of 12.0 years, 17.3% of men (n = 1535) and 5.7% of women (n = 281) were convicted of a violent offence. Criminal history factors most strongly associated with subsequent violence for both men and women were a previous conviction for a violent offence; for assault, illegal threats and/or intimidation; and imprisonment. However, only a previous conviction for a violent offence was associated with incremental predictive validity in both genders following adjustment for young age and comorbid substance use disorder.

CONCLUSIONS

Clinical and actuarial approaches to assess violence risk can be improved if included risk factors are tested using multiple measures of performance.

Using Dual Beta Distributions to Create "Proper" ROC Curves Based on Rating Category Data

BACKGROUND

Receiver operating characteristic (ROC) analysis helps investigators quantify and describe how well a diagnostic system discriminates between 2 mutually exclusive conditions. The conventional binormal (CvB) curve-fitting model usually produces ROCs that are improper in that they do not have the ever-decreasing slope required by signal detection theory. When data sets evaluated under the CvB model have hooks, the resulting ROCs can contain misleading information about the diagnostic performance of the method at low and high false positive rates.

OBJECTIVE

To present and evaluate a dual beta (DB) ROC model that assumes diagnostic data arise from 2 β distributions. The DB model's parameter constraints assure that the resulting ROC curve has a positive, monotonically decreasing slope.

DESIGN/METHOD

Computer simulation study comparing results from CvB, DB, and weighted power function (WPF) models.

RESULTS

The DB model produces results that are as good as or better than those from the WPF model, and less biased and closer to the true values than curves obtained using the CvB model.

CONCLUSIONS

The DB ROC model expresses the relationship between the false positive rate and true positive rate in closed form and allows for quick ROC area calculations using spreadsheet functions. Because it posits simple relationships among the decision axis, operating points, and model parameters, the DB model offers investigators a flexible, easy-to-grasp ROC form that is simpler to implement than other proper ROC models.

Criminogenic factors, psychotic symptoms, and incident arrests among people with serious mental illnesses under intensive outpatient treatment

Although research robustly indicates that general or "criminogenic" factors predict various measures of recidivism, there is controversy about the extent to which these factors, versus untreated symptoms, lead to justice involvement for people with mental illnesses. Based on a sample of 183 people in intensive outpatient treatment followed for an average period of 34.5 months, the present study tested whether criminogenic factors (i.e., factor-analytically derived proxies of some of the "Central Eight"; Andrews & Bonta, 2010) and psychotic symptoms were independently associated with arrest. The study also compared the predictive utility of these domains. In the fully adjusted model, the antisocial subscale and male sex were associated with increased arrest rates, whereas psychosis and age were associated with decreased arrest rates. Criminogenic factors and psychotic symptoms had comparable predictive utility. We conclude that criminogenic factors-chiefly arrest history-and psychotic symptoms predict arrest rates. Both sets of variables appear useful for assessing risk of arrest among people with mental illnesses who are not under current correctional supervision.

Forensic mental health assessment in France: recommendations for quality improvement

The quality of forensic mental health assessment has been a growing concern in various countries on both sides of the Atlantic, but the legal systems are not always comparable and some aspects of forensic assessment are specific to a given country. This paper describes the legal context of forensic psychological assessment in France (i.e. pre-trial investigation phase entrusted to a judge, with mental health assessment performed by preselected professionals called "experts" in French), its advantages and its pitfalls. Forensic psychiatric or psychological assessment is often an essential and decisive element in criminal cases, but since a judiciary scandal which was made public in 2005 (the Outreau case) there has been increasing criticism from the public and the legal profession regarding the reliability of clinical conclusions. Several academic studies and a parliamentary report have highlighted various faulty aspects in both the judiciary process and the mental health assessments. The heterogeneity of expert practices in France appears to be mainly related to a lack of consensus on several core notions such as mental health diagnosis or assessment methods, poor working conditions, lack of specialized training, and insufficient familiarity with the Code of Ethics. In this article we describe and analyze the French practice of forensic psychologists and psychiatrists in criminal cases and propose steps that could be taken to improve its quality, such as setting up specialized training courses, enforcing the Code of Ethics for psychologists, and calling for consensus on diagnostic and assessment methods.

Male inmate profiles and their biological correlates

OBJECTIVE

Borderline and antisocial personality disorders (PDs) share common clinical features (impulsivity, aggressiveness, substance use disorders [SUDs], and suicidal behaviours) that are greatly overrepresented in prison populations. These disorders have been associated biologically with testosterone and cortisol levels. However, the associations are ambiguous and the subject of controversy, perhaps because these heterogeneous disorders have been addressed as unitary constructs. A consideration of profiles of people, rather than of exclusive diagnoses, might yield clearer relationships.

METHODS

In our study, multiple correspondence analysis and cluster analysis were employed to identify subgroups among 545 newly convicted inmates. The groups were then compared in terms of clinical features and biological markers, including levels of cortisol, testosterone, estradiol, progesterone, and sulfoconjugated dehydroepiandrosterone (DHEA-S).

RESULTS

Four clusters with differing psychiatric, criminal, and biological profiles emerged. Clinically, one group had intermediate scores for each of the tested clinical features. Another group comprised people with little comorbidity. Two others displayed severe impulsivity, PD, and SUD. Biologically, cortisol levels were lowest in the last 2 groups and highest in the group with less comorbidity. In keeping with previous findings reported in the literature, testosterone was higher in a younger population with severe psychiatric symptoms. However, some apparently comparable behavioural outcomes were found to be related to distinct biological profiles. No differences were observed for estradiol, progesterone, or DHEA-S levels.

CONCLUSIONS

The results not only confirm the importance of biological markers in the study of personality features but also demonstrate the need to consider the role of comorbidities and steroid coregulation.

Systematic meta-analysis of outcomes associated with psychosis and co-morbid substance use

OBJECTIVE

To compare the symptoms and social function of patients with psychosis and current substance use to those with psychosis and no history of substance use.

METHOD

The databases EMBASE, MEDLINE and PsycINFO were searched for peer-reviewed publications in English that reported the characteristics of patients with psychotic illness who were current substance users and those who had never used substances. The searches yielded 22 articles that met the inclusion criteria. Meta-analysis was used to compare four key outcome variables: positive symptoms, negative symptoms, depression and social function - and three secondary outcomes: violence, self-harm and hospital admissions.

RESULTS

Current substance-using patients were significantly younger than non-substance-using patients and were more likely to be male, but did not differ in age at onset of psychosis or in their level of education. Current substance users had higher ratings of positive symptoms and were more likely to have a history of violence. Older studies reported a stronger association between current substance use and positive symptoms than more recently published studies. Current substance users did not differ from non-users on measurements of negative symptoms, depressive symptoms, social function, self-harm, or the number of hospital admissions.

CONCLUSION

Current substance users with psychosis may have more severe positive symptoms than patients who have never used substances, but this result should be interpreted with caution because of demographic differences between substance users and non-substance users.

Psychiatry on trial: the Norway 2011 massacre

On July 22, 2011, Anders Breivik, a Norwegian citizen, detonated a fertilizer bomb near government buildings in Oslo, killing eight people, and then proceeded to a nearby island where the Labor Party was holding a youth camp. There, he killed 69 people before being arrested. Just before these events, he posted a "compendium" on the Web explaining his actions and encouraging others to do likewise. Much of the ensuing media coverage and trial focused on whether he was sane and whether he had a psychiatric diagnosis. One team of court-appointed psychiatrists found him to be psychotic with a diagnosis of paranoid schizophrenia and legally insane. A second team found him neither psychotic nor schizophrenic and, thus, legally sane. Their contrary opinions were not reconciled by observing his behavior in court. We discuss why experienced psychiatrists reached such fundamentally opposing diagnostic conclusions about a "home-grown" terrorist holding extreme political views.

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.

Course and predictors of physical aggressive behaviour after discharge from a psychiatric inpatient unit: 1 year follow-up

The present study analyzes course and predictors of physically aggressive behaviour over a 1-year follow up in a sample of patients discharged from a psychiatric inpatient unit. One hundred and eighty-six patients discharged from a locked short-term Psychiatric Inpatient Unit at the Bologna University Hospital. After discharge, two data collection contacts at 1 month and at 1 year were scheduled. In particular, psychiatrists, nurses, and other professionals were interviewed by the research staff using the Overt Aggression Scale. About 20 % of discharged patients showed physical aggressiveness in subsequent follow-up contacts. Risk factors for physical violence in the short-time period were social problems and a longer time from the first psychiatric contact. Living in residential facilities and physical aggressiveness during hospitalization were correlated to violence in the long-time period. Risk factors for physically violent behaviour differed in the short-term and long-term follow-ups; different causes of violent behaviour could be hypothesized.

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.

Dorsal/ventral parcellation of the amygdala: relevance to impulsivity and aggression

Investigations into the specific association of amygdala volume, a critical aspect of the fronto-limbic emotional circuitry, and aggression have produced results broadly consistent with the 'larger is more powerful' doctrine. However, recent reports suggest that the ventral and dorsal aspects of the amygdala play functionally specific roles, respectively, in the activation and control of behavior. Therefore, parceling amygdala volume into dorsal and ventral components might prove productive in testing hypotheses regarding volumetric association to aggression, and impulsivity, a related aspect of self-control. We sought to test this hypothesis in a group of 41 psychiatric patients who received standard magnetic resonance imaging and a psychometric protocol including aggression and impulsivity measures. Whole amygdala volumes were not associated with aggression or impulsivity, but significant correlations were found when dorsal/ventral amygdalae were analyzed separately. Specifically, left and right ventral amygdala volume was positively associated with motor impulsivity, and left dorsal amygdala was negatively associated with aggression. Results are discussed in terms of an activation and control model of brain-behavior relations. Potential relevance to the continuum of amygdala hyper- to hypo-activation and aggression is discussed.

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.

The development of a population-level clinical screening tool for self-harm repetition and suicide: the ReACT Self-Harm Rule

BACKGROUND

Self-harm is a common reason for Emergency Department (ED) attendance. We aimed to develop a clinical tool to help identify patients at higher risk of repeat self-harm, or suicide, within 6 months of an ED self-harm presentation.

METHOD

The tool, the ReACT Self-Harm Rule, was derived using multicentre data from a prospective cohort study. Binary recursive partitioning was applied to data from two centres, and data from a separate centre were used to test the tool. There were 29 571 self-harm presentations to five hospital EDs between January 2003 and June 2007, involving 18 680 adults aged ⩾16 years. We estimated sensitivity, specificity and positive and negative predictive values to measure the performance of the tool.

RESULTS

A self-harm presentation was classified as higher risk if at least one of the following factors was present: recent self-harm (in the past year), living alone or homelessness, cutting as a method of harm and treatment for a current psychiatric disorder. The rule performed with 95% sensitivity [95% confidence interval (CI) 94-95] and 21% specificity (95% CI 21-22), and had a positive predictive value of 30% (95% CI 30-31) and a negative predictive value of 91% (95% CI 90-92) in the derivation centres; it identified 83/92 of all subsequent suicides.

CONCLUSIONS

The ReACT Self-Harm Rule might be used as a screening tool to inform the process of assessing self-harm presentations to ED. The four risk factors could also be used as an adjunct to in-depth psychosocial assessment to help guide risk formulation. The use of multicentre data helped to maximize the generalizability of the tool, but we need to further verify its external validity in other localities.

Predicting community violence from patients discharged from acute mental health units in England

PURPOSE

To investigate the validity of risk factors and established risk measures in predicting community violence in an acute mental health sample up to 20 weeks post-discharge.

METHOD

Prospective cohort follow-up study conducted between January 2006 and August 2007. Baseline assessments were conducted while participants were inpatients. The measures were rated following interview with the participants, record review and speaking to someone who knows the person well (e.g. friend, relative, carer). Baseline measures were then compared with frequency and severity of violence in the community post-discharge at 20 weeks.

RESULTS

In the 20-week period post-discharge, 29 (25.4%) of the 114 participants were violent. All the risk measures and measures of impulsiveness and anger were predictive of violence where p < 0.05. The HCR-20 total, psychopathy and clinical factors were strongly correlated with the frequency of violence where p < 0.05.

CONCLUSIONS

The risk factors and risk measures that have been found to be predictive in forensic samples are also predictive in acute mental health samples, although the effects are not as large. Future research needs to be conducted with a larger sample to include investigation of differences in risk factors based on gender and social support. Services and clinicians need to consider how to integrate findings into useful frameworks to support decisions and contribute to managing risk. This should assist in identifying interventions aimed at preventing community violence.