The authors thank Dr Wei for his interest and support provided to their research and its findings. We are cognisant that the experience of addressing mental illness, criminal justice contact, and homelessness is a worldwide issue.
Naturally, as with many cross-sectional cohort studies examining associations between independent characteristics and outcomes, causation is not able to be inferred. The authors agree that contact with the criminal justice system, mental disorders and/or substance use may both precede and contribute to homelessness. Although we do not infer a causal relationship, the high proportion of the cohort released from prison (∼30%) and the high rate of recidivism during the study show that the existing interventions are not preventive.
The authors look forward to prospective studies that may be able to disentangle the contribution and the effects of criminal recidivism, mental illness, substance use, and homelessness and seek to offer evidence-based solutions to provide stable accommodation.
Alcohol dependence is a risk factor for experiencing victimisation, but little is known about how peer and behavioural mechanisms may explain this relationship.
To test deviant peer association and/or heavy-episodic drinking frequency as mediators between alcohol dependence and risk for being victimised.
The Pathways to Desistance data were analysed. Generalised structural equation modelling was used to determine whether either or both of the hypothesised pathways significantly mediated the relationship between alcohol dependence and victimisation.
Alcohol dependence at recruitment (age 14–17) was associated with higher odds of being a victim of some form of violence in Wave 3 (age x–y). This relationship was significantly mediated by deviant peer association but not heavy-episodic drinking frequency between Waves 2 and 3.
These findings add knowledge about the mechanism of a link between early alcohol dependence and later violent victimisation among young offenders. They suggest that more focus on reducing delinquent peer association, or reducing its impact, is crucial to reducing further harms to these young people, in turn possibly affecting continuing substance use and reoffending risks. Peer mentoring programmes help to provide prosocial modelling and reduce deviant peer ties in some circumstances, and these findings suggest that the next step should be their evaluation specifically among justice-involved young people with alcohol dependence. Providing additional funding and/or opportunities for involvement in such mentoring programmes may help to reduce the public health and financial costs associated with alcohol dependence in the juvenile justice system.
Men who commit violence against an intimate partner differ in their motives. Classifying the proactivity of men's partner violence may reveal important differences that could be treatment targets.
To examine the differences between proactive and reactive partner violence based on coded descriptions of past violent events.
Community cohabiting couples reporting intimate partner violence were recruited via advertisements. Men and women were independently interviewed about past male-to-female violent events. The narratives of a male perpetrator and a female victim were coded using a Proactive–Reactive coding system, yielding three categories of violence: reactive, mixed proactive/reactive and proactive. The three categories were compared for differences in personality disorder features, attachment, psychophysiological reactivity during a conflict discussion task, and a self- and partner report of men's proactive and reactive aggressive tendencies.
The results revealed a 54% classification agreement between perpetrator and victim reports. No differences were found on personality or attachment measures between the groups regardless of gender of the reporter. Reactive violence was related to a tendency to self-report more reactive aggression and higher heart rate reactivity during a laboratory conflict discussion compared to the group reporting both proactive and reactive violent incidents.
This study suggests that a coding system for intimate partner violence can be applied to community volunteers, and it is a reliable report and valid. However, there are discrepancies when the coding is based on the perpetrator or victim reports.
There are not many longitudinal studies examining people experiencing homelessness and interacting with the criminal justice system over time.
To describe the type of criminal offences committed, court outcomes, identify probable predictors of reoffending, and estimate the criminal justice costs in a cohort of homeless hostel clinic attendees.
A retrospective cohort study of 1646 people attending a homeless clinic who had had contact with the criminal justice system (CJS) in New South Wales (NSW), Australia, using linked clinic, criminal offence, health and mortality data from 1 July 2008 to 30 June 2021. Initial comparisons were made with the 852 clinic attendees without CJS contact in the period. Multivariable logistic regression was used to identify predictors of recidivism.
There were 16,840 offending episodes, giving an offence rate of 87.8 per 100 person-years (95%CI: 86.5–89.1). The most common index offences were acts intended to cause injury (22%), illicit drug (17%) and theft-related (12%) offences. Most people (83%) were found guilty of the index offence and received a fine (37%) or community-based sentence (29%). Total court finalisation costs were AUD $11.3 million. Three-quarters of those convicted reoffended within 24 months. Offenders were more likely to be younger, have a diagnosis of personality disorder (AOR: 1.31; 95% CI: 1.04–1.67), a substance use disorder (AOR: 1.60; 95% CI 1.14–2.23) and/or to have a previous charge dismissed on mental health grounds (AOR: 1.79; 95% CI: 1.31–2.46). Within the offending cohort, reoffenders had almost twice the odds of having theft-related offences as their principal index offence (AOR: 1.85; 95% CI: 1.29–2.66).
This longitudinal study finding of not only a high rate of criminal justice contact, but also a high rate of recidivism among people who have been homeless, lends support to a need for strategies both to address the root causes of homelessness and to provide a comprehensive systems-based response to reduce recidivism, that includes secure housing as well as mental health and substance use treatment programmes for homeless offenders.
Although there is general consensus about the behavioural, clinical and sociodemographic variables that are risk factors for reoffending, optimal statistical modelling of these variables is less clear. Machine learning techniques offer an approach that may provide greater accuracy than traditional methods.
To compare the performance of advanced machine learning techniques (classification trees and random forests) to logistic regression in classifying correlates of rearrest among adult probationers and parolees in the United States.
Data were from the subgroup of people on probation or parole who had taken part in the National Survey on Drug Use and Health for the years 2015–2019. We compared the performance of logistic regression, classification trees and random forests, using receiver operating characteristic curves, to examine the correlates of arrest within the past 12 months.
We found that machine learning techniques, specifically random forests, possessed significantly greater accuracy than logistic regression in classifying correlates of arrest.
Our findings suggest the potential for enhanced risk classification. The next step would be to develop applications for criminal justice and clinical practice to inform better support and management strategies for former offenders in the community.
Attention deficit/hyperactivity disorder (ADHD) is known to be a risk factor for antisocial and delinquent behaviour, but there is still a lack of information on how features of ADHD relate to offending behaviour among adults not already defined by their offending.
Our aim was to add to knowledge about relationships between ADHD and antisocial behaviour among adults in the general population by answering the following questions: (A) Does the level of self-reported ADHD features relate to criminal and non-criminal antisocial behaviour? (B) To what extent are self-ratings of ADHD features independent of socio-demographic features previously identified as predictors of antisocial behaviour?
A sample of adults was originally recruited to study public response to the COVID-19 outbreak through an online panel to be representative of the Israeli population. Among other scales, the 2025 participants completed an ADHD self-report scale, an antisocial behaviour self-report scale and a socio-demographic questionnaire probing for age, gender, urbanity, place of birth, socioeconomic status (education and income), family status (being in a relationship and having children) and religiosity.
Higher mean totals for the inattention and hyperactivity ADHD scale scores were associated with higher mean antisocial behaviour scores. These relationships were only slightly affected by socio-demographic variables, including sex, age, education and income.
Our findings suggest that not only may features of ADHD, even below a diagnostic threshold, constitute a risk factor for antisocial behaviour, but also that the self-rated levels of these problems covary. These findings are important for informing the early detection of risk of antisocial behaviour in the general population and its prevention.
The prevalence of mental disorders and substance use among prisoners is high. Convicted prisoners of ‘good behaviour’ can be part of a peer support system in prisons.
To evaluate the feasibility of a peer support programme for prisoners with common mental disorders and substance use in prison.
The study used a mixed method research design, with a quasi-experimental approach (single group pre-post without control). It was conducted in two phases: Phase I. Thirty-five peers/convicted prisoners were recruited through advertisements on the prisoners' community radio station. Volunteers with good behaviour reports were given training over 5 days to recognise mental and substance use disorders and provide basic peer support in prison; their attitudes and knowledge were tested before and after the training. Phase II. Feasibility of the peer support programme was tested by (i) recording the number of cases identified and referred, (ii) pre- and post-evaluation of well-being, coping, and symptom severity of those supported and (iii) evaluating qualitatively the experience of the peer supporters and service users.
Thirty-five peer supporters identified 49 cases over 3 months. These cases showed significant improvement in well-being (Z −1.962; p < 0.050) and reduction in symptom severity (Z −1.913; 0.056). There was a significant improvement in the peers supporters' self-esteem from pre- to post-training (t −3.31; p < 0.002), improvement in their benevolence (t −4.37; p < 0.001) and a significant reduction in their negative attitudes to mental illness (Z −3.518; p < 0.001). A thematic model of peer support encompassed self-experienced benefits for the peer supporter, wider recognition of peer supporters in the prison, challenges to this kind of support, experience of training and visions for future work.
The peer support programme was experienced positively by the peer-supporters and supported. Common mental disorders, substance use and suicidality were recognised and appropriately referred. A full-scale evaluation of this promising programme is warranted.
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：