Violent and non-violent offending in patients with schizophrenia: Exploring influences and differences via machine learning

Sonnweber, Martina; Lau, Steffen; Kirchebner, Johannes

doi:10.1016/j.comppsych.2021.152238

Cited by 21 publications

(45 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Through comparing with each other, the nnet had better performance, and its AUC of 0.6673 (0.5599-0.7748) was significantly better than chance. In terms of the ability to recognize male schizophrenia patients with violence, our model performance showed similar precision as was obtained in the previous studies ( 14 , 22 ). Moreover, the nnet algorithm can calculate the probability of an individual committing violence.…”

Section: Discussionsupporting

confidence: 84%

“…For instance, Wang et al utilized seven classification algorithms to predict violence status in schizophrenia individuals and found random forests showed better performance, its accuracy and AUC achieving 62% and 0.63, respectively ( 14 ). Another study determined gradient boosting as the best algorithm among seven algorithms, with its accuracy and AUC being 0.678 and 0.764 in predicting violent offending of forensic offender patients with schizophrenia, respectively ( 22 ). In this study, we conducted eight ML algorithms to differentiate violent and non-violent behaviors of male patients with schizophrenia.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

The Prediction and Influential Factors of Violence in Male Schizophrenia Patients With Machine Learning Algorithms

Zhang

Liu

et al. 2022

Front. Psychiatry

View full text Add to dashboard Cite

BackgroundEarly to identify male schizophrenia patients with violence is important for the performance of targeted measures and closer monitoring, but it is difficult to use conventional risk factors. This study is aimed to employ machine learning (ML) algorithms combined with routine data to predict violent behavior among male schizophrenia patients. Moreover, the identified best model might be utilized to calculate the probability of an individual committing violence.MethodWe enrolled a total of 397 male schizophrenia patients and randomly stratified them into the training set and the testing set, in a 7:3 ratio. We used eight ML algorithms to develop the predictive models. The main variables as input features selected by the least absolute shrinkage and selection operator (LASSO) and logistic regression (LR) were integrated into prediction models for violence among male schizophrenia patients. In the training set, 10 × 10-fold cross-validation was conducted to adjust the parameters. In the testing set, we evaluated and compared the predictive performance of eight ML algorithms in terms of area under the curve (AUC) for the receiver operating characteristic curve.ResultOur results showed the prevalence of violence among male schizophrenia patients was 36.8%. The LASSO and LR identified main risk factors for violent behavior in patients with schizophrenia integrated into the predictive models, including lower education level [0.556 (0.378–0.816)], having cigarette smoking [2.121 (1.191–3.779)], higher positive syndrome [1.016 (1.002–1.031)] and higher social disability screening schedule (SDSS) [1.081 (1.026–1.139)]. The Neural Net (nnet) with an AUC of 0.6673 (0.5599–0.7748) had better prediction ability than that of other algorithms.ConclusionML algorithms are useful in early identifying male schizophrenia patients with violence and helping clinicians take preventive measures.

show abstract

Section: Discussionsupporting

confidence: 84%

Section: Discussionmentioning

confidence: 99%

The Prediction and Influential Factors of Violence in Male Schizophrenia Patients With Machine Learning Algorithms

Zhang

Liu

et al. 2022

Front. Psychiatry

View full text Add to dashboard Cite

show abstract

“…Over the past decades, evidence of the association between schizophrenia and violence has accumulated (e.g., [40]), thereby identifying a multitude of relevant risk factors. Risk factors of violence mainly concern individual diagnoses such as schizophrenia, genetic influences between mental illness and violence, a history of victimization, comorbid substance abuse, alcohol use, antisocial behavior, poor adherence to treatment, and overall symptom severity, etc.…”

Section: Introductionmentioning

confidence: 99%

Effects of Integrated Violence Intervention on Alexithymia, Cognitive, and Neurocognitive Features of Violence in Schizophrenia: A Randomized Controlled Trial

Hsu

Ouyang

2021

Brain Sciences

View full text Add to dashboard Cite

Patients with schizophrenia and repetitive violence express core impairments that encompass multiple domains. To date, there have been few interventions integrating neurocognition, social cognition, alexithymia, and emotion regulation together as an approach to manage repetitive violence. The aim of this open-label randomized controlled trial was to examine more comprehensively the effectiveness of a novel Integrated Cognitive Based Violence Intervention Program on management of repetitive violence in patients with schizophrenia (vSZ). Sixty recruited patients were aged ≥20 years, diagnosed with schizophrenia for >2 years, had repetitive violent behavior within one year, and were psychiatrically hospitalized. The vSZ patients were randomly allocated to two groups and received either the intervention or treatment as usual. The intervention module, consisting of all defined 11 cognitive and social cognitive domains as well as emotion regulation, which were grouped into four modules. The intervention placed emphasis on the patients’ behavioral problems or intrinsic conflicts in relation to repetitive violence. The results indicate a statistically significant trend toward reducing impulsivity, anger with resentment, physical aggression, suspicion, and hostility (p < 0.05). The intervention significantly alleviated the intensity of cognitive failure, improved the management of alexithymic features and attribution styles and errors, and fostered adequate decision-making styles and emotion regulation capacity (p < 0.05). The intervention, when applied in conjunction with psychiatric standard care, could exert synergistic effects on alexithymia and cognitive, clinical, and neurocognitive features of repetitive violence in schizophrenia. This intervention provided patients a more active role to manage their violent behavior with the involvement of alexithymia.

show abstract

“…Although previous studies have reported that environmental factors, such as low socio-economic status and childhood trauma, may lead to violence in SZ ( 3 – 5 ), increasing evidence indicates that neurobiological factors may also play a key role in the increased risk of violence in individuals with SZ ( 6 , 7 ). The origins of violent behavior in people with SZ are not yet sufficiently understood ( 8 ). Moreover, the management of aggression in SZ patients is a challenging clinical dilemma given that violence or aggressive behavior is heterogeneous in origin ( 8 – 10 ).…”

Section: Introductionmentioning

confidence: 99%

“…The origins of violent behavior in people with SZ are not yet sufficiently understood ( 8 ). Moreover, the management of aggression in SZ patients is a challenging clinical dilemma given that violence or aggressive behavior is heterogeneous in origin ( 8 – 10 ). Therefore, delineating the underlying neurobiological basis of violence in SZ may facilitate its management and effective therapy.…”

Section: Introductionmentioning

confidence: 99%

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

Shen

Liu³

et al. 2021

Front. Psychiatry

View full text Add to dashboard Cite

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.

show abstract

Violent and non-violent offending in patients with schizophrenia: Exploring influences and differences via machine learning

Cited by 21 publications

References 40 publications

The Prediction and Influential Factors of Violence in Male Schizophrenia Patients With Machine Learning Algorithms

The Prediction and Influential Factors of Violence in Male Schizophrenia Patients With Machine Learning Algorithms

Effects of Integrated Violence Intervention on Alexithymia, Cognitive, and Neurocognitive Features of Violence in Schizophrenia: A Randomized Controlled Trial

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

Contact Info

Product

Resources

About