Identifying Direct Coercion in a High Risk Subgroup of Offender Patients With Schizophrenia via Machine Learning Algorithms

Delusions and Delinquencies: A Comparison of Violent and Non-Violent Offenders With Schizophrenia Spectrum Disorders

The relationship between schizophrenia spectrum disorders (SSD) and violent offending has long been the subject of research. The present study attempts to identify the content of delusions, an understudied factor in this regard, that differentiates between violent and non-violent offenses. Limitations, clinical relevance, and future directions are discussed. Employing a retrospective study design, machine learning algorithms and a comprehensive set of variables were applied to a sample of 366 offenders with a schizophrenia spectrum disorder in a Swiss forensic psychiatry department. Taking into account the different contents and affects associated with delusions, eight variables were identified as having an impact on discriminating between violent and non-violent offenses with an AUC of 0.68, a sensitivity of 30.8%, and a specificity of 91.9%, suggesting that the variables found are useful for discriminating between violent and non-violent offenses. Delusions of grandiosity, delusional police and/or army pursuit, delusional perceived physical and/or mental injury, and delusions of control or passivity were more predictive of non-violent offenses, while delusions with aggressive content or delusions associated with the emotions of anger, distress, or agitation were more frequently associated with violent offenses. Our findings extend and confirm current research on the content of delusions in patients with SSD. In particular, we found that the symptoms of threat/control override (TCO) do not directly lead to violent behavior but are mediated by other variables such as anger. Notably, delusions traditionally seen as symptoms of TCO, appear to have a protective value against violent behavior. These findings will hopefully help to reduce the stigma commonly and erroneously associated with mental illness, while supporting the development of effective therapeutic approaches.

Exploring Characteristics of Homicide Offenders With Schizophrenia Spectrum Disorders Via Machine Learning

The link between schizophrenia and homicide has long been the subject of research with significant impact on mental health policy, clinical practice, and public perception of people with psychiatric disorders. The present study investigates factors contributing to completed homicides committed by offenders diagnosed with schizophrenia referred to a Swiss forensic institution, using machine learning algorithms. Data were collected from 370 inpatients at the Centre for Inpatient Forensic Therapy at the Zurich University Hospital of Psychiatry. A total of 519 variables were explored to differentiate homicidal and other (violent and non-violent) offenders. The dataset was split employing variable filtering, model building, and selection embedded in a nested resampling approach. Ten factors regarding criminal and psychiatric history and clinical factors were identified to be influential in differentiating between homicidal and other offenders. Findings expand the research on influential factors for completed homicide in patients with schizophrenia. Limitations, clinical relevance, and future directions are discussed.

Machine Learning for prediction of violent behaviors in schizophrenia spectrum disorders: a systematic review

Background

Schizophrenia spectrum disorders (SSD) can be associated with an increased risk of violent behavior (VB), which can harm patients, others, and properties. Prediction of VB could help reduce the SSD burden on patients and healthcare systems. Some recent studies have used machine learning (ML) algorithms to identify SSD patients at risk of VB. In this article, we aimed to review studies that used ML to predict VB in SSD patients and discuss the most successful ML methods and predictors of VB.

Methods

We performed a systematic search in PubMed, Web of Sciences, Embase, and PsycINFO on September 30, 2023, to identify studies on the application of ML in predicting VB in SSD patients.

Results

We included 18 studies with data from 11,733 patients diagnosed with SSD. Different ML models demonstrated mixed performance with an area under the receiver operating characteristic curve of 0.56-0.95 and an accuracy of 50.27-90.67% in predicting violence among SSD patients. Our comparative analysis demonstrated a superior performance for the gradient boosting model, compared to other ML models in predicting VB among SSD patients. Various sociodemographic, clinical, metabolic, and neuroimaging features were associated with VB, with age and olanzapine equivalent dose at the time of discharge being the most frequently identified factors.

Conclusion

ML models demonstrated varied VB prediction performance in SSD patients, with gradient boosting outperforming. Further research is warranted for clinical applications of ML methods in this field.

Differentiating Between Sexual Offending and Violent Non-sexual Offending in Men With Schizophrenia Spectrum Disorders Using Machine Learning

Forensic psychiatric populations commonly contain a subset of persons with schizophrenia spectrum disorders (SSD) who have committed sex offenses. A comprehensive delineation of the features that distinguish persons with SSD who have committed sex offenses from persons with SSD who have committed violent non-sex offenses could be relevant to the development of differentiated risk assessment, risk management and treatment approaches. This analysis included the patient records of 296 men with SSD convicted of at least one sex and/or violent offense who were admitted to the Centre for Inpatient Forensic Therapy at the University Hospital of Psychiatry Zurich between 1982 and 2016. Using supervised machine learning, data on 461 variables retrospectively collected from the records were compared with respect to their relative importance in differentiating between men who had committed sex offenses and men who had committed violent non-sex offenses. The final machine learning model was able to differentiate between the two types of offenders with a balanced accuracy of 71.5% (95% CI = [60.7, 82.1]) and an AUC of .80 (95% CI = [.67, .93]). The main distinguishing features included sexual behaviours and interests, psychopathological symptoms and characteristics of the index offense. Results suggest that when assessing and treating persons with SSD who have committed sex offenses, it appears to be relevant to not only address the core symptoms of the disorder, but to also take into account general risk factors for sexual recidivism, such as atypical sexual interests and sexual preoccupation.

Self-Harm Among Forensic Psychiatric Inpatients With Schizophrenia Spectrum Disorders: An Explorative Analysis

The burden of self-injury among offenders undergoing inpatient treatment in forensic psychiatry is substantial. This exploratory study aims to add to the previously sparse literature on the correlates of self-injury in inpatient forensic patients with schizophrenia spectrum disorders (SSD). Employing a sample of 356 inpatients with SSD treated in a Swiss forensic psychiatry hospital, patient data on 512 potential predictor variables were retrospectively collected via file analysis. The dataset was examined using supervised machine learning to distinguish between patients who had engaged in self-injurious behavior during forensic hospitalization and those who had not. Based on a combination of ten variables, including psychiatric history, criminal history, psychopathology, and pharmacotherapy, the final machine learning model was able to discriminate between self-injury and no self-injury with a balanced accuracy of 68% and a predictive power of AUC = 71%. Results suggest that forensic psychiatric patients with SSD who self-injured were younger both at the time of onset and at the time of first entry into the federal criminal record. They exhibited more severe psychopathological symptoms at the time of admission, including higher levels of depression and anxiety and greater difficulty with abstract reasoning. Of all the predictors identified, symptoms of depression and anxiety may be the most promising treatment targets for the prevention of self-injury in inpatient forensic patients with SSD due to their modifiability and should be further substantiated in future studies.

Prediction of Chinese clients' satisfaction with psychotherapy by machine learning

Background

Effective psychotherapy should satisfy the client, but that satisfaction depends on many factors. We do not fully understand the factors that affect client satisfaction with psychotherapy and how these factors synergistically affect a client's psychotherapy experience.

Aims

This study aims to use machine learning to predict Chinese clients' satisfaction with psychotherapy and analyze potential outcome contributors.

Methods

In this cross-sectional investigation, a self-compiled online questionnaire was delivered through the WeChat app. The information of 791 participants who had received psychotherapy was used in the study. A series of features, for example, the participants' demographic features and psychotherapy-related features, were chosen to distinguish between participants satisfied and dissatisfied with the psychotherapy they received. With our dataset, we trained seven supervised machine-learning-based algorithms to implement prediction models.

Results

Among the 791 participants, 619 (78.3%) reported being satisfied with the psychotherapy sessions that they received. The occupation of the clients, the location of psychotherapy, and the form of access to psychotherapy are the three most recognizable features that determined whether clients are satisfied with psychotherapy. The machine-learning model based on the CatBoost achieved the highest prediction performance in classifying satisfied and psychotherapy clients with an F1 score of 0.758.

Conclusion

This study clarified the factors related to clients' satisfaction with psychotherapy, and the machine-learning-based classifier accurately distinguished clients who were satisfied or unsatisfied with psychotherapy. These results will help provide better psychotherapy strategies for specific clients, so they may achieve better therapeutic outcomes.

Model Building in Forensic Psychiatry: A Machine Learning Approach to Screening Offender Patients with SSD

Today’s extensive availability of medical data enables the development of predictive models, but this requires suitable statistical methods, such as machine learning (ML). Especially in forensic psychiatry, a complex and cost-intensive field with risk assessments and predictions of treatment outcomes as central tasks, there is a need for such predictive tools, for example, to anticipate complex treatment courses and to be able to offer appropriate therapy on an individualized basis. This study aimed to develop a first basic model for the anticipation of adverse treatment courses based on prior compulsory admission and/or conviction as simple and easily objectifiable parameters in offender patients with a schizophrenia spectrum disorder (SSD). With a balanced accuracy of 67% and an AUC of 0.72, gradient boosting proved to be the optimal ML algorithm. Antisocial behavior, physical violence against staff, rule breaking, hyperactivity, delusions of grandeur, fewer feelings of guilt, the need for compulsory isolation, cannabis abuse/dependence, a higher dose of antipsychotics (measured by the olanzapine half-life) and an unfavorable legal prognosis emerged as the ten most influential variables out of a dataset with 209 parameters. Our findings could demonstrate an example of the use of ML in the development of an easy-to-use predictive model based on few objectifiable factors.

Thinking Aloud or Screaming Inside: Exploratory Study of Sentiment Around Work

Background

Millions of workers experience work-related ill health every year. The loss of working days often accounts for poor well-being because of discomfort and stress caused by the workplace. The ongoing pandemic and postpandemic shift in socioeconomic and work culture can continue to contribute to adverse work-related sentiments. Critically investigating state-of-the-art technologies, this study identifies the research gaps in recognizing workers’ need for well-being support, and we aspire to understand how such evidence can be collected to transform the workforce and workplace.

Objective

Building on recent advances in sentiment analysis, this study aims to closely examine the potential of social media as a tool to assess workers’ emotions toward the workplace.

Methods

This study collected a large Twitter data set comprising both pandemic and prepandemic tweets facilitated through a human-in-the-loop approach in combination with unsupervised learning and meta-heuristic optimization algorithms. The raw data preprocessed through natural language processing techniques were assessed using a generative statistical model and a lexicon-assisted rule-based model, mapping lexical features to emotion intensities. This study also assigned human annotations and performed work-related sentiment analysis.

Results

A mixed methods approach, including topic modeling using latent Dirichlet allocation, identified the top topics from the corpus to understand how Twitter users engage with discussions on work-related sentiments. The sorted aspects were portrayed through overlapped clusters and low intertopic distances. However, further analysis comprising the Valence Aware Dictionary for Sentiment Reasoner suggested a smaller number of negative polarities among diverse subjects. By contrast, the human-annotated data set created for this study contained more negative sentiments. In this study, sentimental juxtaposition revealed through the labeled data set was supported by the n-gram analysis as well.

Conclusions

The developed data set demonstrates that work-related sentiments are projected onto social media, which offers an opportunity to better support workers. The infrastructure of the workplace, the nature of the work, the culture within the industry and the particular organization, employers, colleagues, person-specific habits, and upbringing all play a part in the health and well-being of any working adult who contributes to the productivity of the organization. Therefore, understanding the origin and influence of the complex underlying factors both qualitatively and quantitatively can inform the next generation of workplaces to drive positive change by relying on empirically grounded evidence. Therefore, this study outlines a comprehensive approach to capture deeper insights into work-related health.

Effectiveness of Artificial Intelligence Methods in Personalized Aggression Risk Prediction within Inpatient Psychiatric Treatment Settings—A Systematic Review

Aggression risk assessments are vital to prevent injuries and morbidities amongst patients and staff in psychiatric settings. More recent studies have harnessed artificial intelligence (AI) methods such as machine learning algorithms to determine factors associated with aggression in psychiatric treatment settings. In this review, using Cooper’s five-stage review framework, we aimed to evaluate the: (1) predictive accuracy, and (2) clinical variables associated with AI-based aggression risk prediction amongst psychiatric inpatients. Databases including PubMed, Cochrane, Scopus, PsycINFO, CINAHL were searched for relevant articles until April 2022. The eight included studies were independently evaluated using critical appraisal tools for systematic review developed by Joanna Briggs Institute. Most of the studies (87.5%) examined health records in predicting aggression and reported acceptable to excellent accuracy with specific machine learning algorithms employed (area under curve range 0.75–0.87). No particular machine learning algorithm outperformed the others consistently across studies (area under curve range 0.61–0.87). Relevant factors identified with aggression related to demographic and social profile, past aggression, forensic history, other psychiatric history, psychopathology, challenging behaviors and management domains. The limited extant studies have highlighted a potential role for the use of AI methods to clarify factors associated with aggression in psychiatric inpatient treatment settings.

Exploring substance use as rule-violating behaviour during inpatient treatment of offender patients with schizophrenia

BACKGROUND

Rule-violating behaviour in the form of substance misuse has been studied primarily within the context of prison settings, but not in forensic psychiatric settings.

AIMS

Our aim was to explore factors that are associated with substance misuse during hospitalisation in patients among those patients in a Swiss forensic psychiatric inpatient unit who were suffering from a disorder along the schizophrenia spectrum.

METHODS

From a database of demographic, clinical and offending data on all residents at any time between 1982 and 2016 in the forensic psychiatric hospital in Zurich, 364 cases fulfilled diagnostic criteria for schizophrenia or a schizophrenia-like illness and formed our sample. Any confirmed use of alcohol or illicit substances during admission (yes/no) was the dependent variable. Its relationship to all 507 other variables was explored by machine learning. To counteract overfitting, data were divided into training and validation set. The best model from the training set was tested on the validation set.

RESULTS

Substance use as a secure hospital inpatient was unusual (15, 14%). Prior substance use disorder accounted for so much of the variance (AUC 0.92) that it was noted but excluded from further models. In the resulting model of best fit, variables related to rule breaking, younger age overall and at onset of schizophrenia and nature of offending behaviour, substance misuse as a minor and having records of complications in prior psychiatric treatment were associated with substance misuse during hospitalisation, as was length of inpatient treatment. In the initial model the AUC was 0.92. Even after removal of substance use disorder from the final model, performance indicators were meaningful with a balanced accuracy of 67.95, an AUC of 0.735, a sensitivity of 81.48% and a specificity of 57.58%.

CONCLUSIONS

Substance misuse in secure forensic psychiatric hospitals is unusual but worthy of clinical and research consideration because of its association with other rule violations and longer hospitalisation. More knowledge is needed about effective interventions and rehabilitation for this group.

Towards identifying cancer patients at risk to miss out on psycho-oncological treatment via machine learning

Objective

In routine oncological treatment settings, psychological distress, including mental disorders, is overlooked in 30% to 50% of patients. High workload and a constant need to optimise time and costs require a quick and easy method to identify patients likely to miss out on psychological support.

Methods

Using machine learning, factors associated with no consultation with a clinical psychologist or psychiatrist were identified between 2011 and 2019 in 7,318 oncological patients in a large cancer treatment centre. Parameters were hierarchically ordered based on statistical relevance. Nested resampling and cross validation were performed to avoid overfitting.

Results

Patients were least likely to receive psycho‐oncological (i.e., psychiatric/psychotherapeutic) treatment when they were not formally screened for distress, had inpatient treatment for less than 28 days, had no psychiatric diagnosis, were aged 65 or older, had skin cancer or were not being discussed in a tumour board. The final validated model was optimised to maximise sensitivity at 85.9% and achieved an area under the curve (AUC) of 0.75, a balanced accuracy of 68.5% and specificity of 51.2%.

Conclusion

Beyond conventional screening tools, results might contribute to identify patients at risk to be neglected in terms of referral to psycho‐oncology within routine oncological care.

Integrating machine learning with electronic health record data to facilitate detection of prolactin level and pharmacovigilance signals in olanzapine-treated patients

BACKGROUND AND AIM

Available evidence suggests elevated serum prolactin (PRL) levels in olanzapine (OLZ)-treated patients with schizophrenia. However, machine learning (ML)-based comprehensive evaluations of the influence of pathophysiological and pharmacological factors on PRL levels in OLZ-treated patients are rare. We aimed to forecast the PRL level in OLZ-treated patients and mine pharmacovigilance information on PRL-related adverse events by integrating ML and electronic health record (EHR) data.

METHODS

Data were extracted from an EHR system to construct an ML dataset in 672×384 matrix format after preprocessing, which was subsequently randomly divided into a derivation cohort for model development and a validation cohort for model validation (8:2). The eXtreme gradient boosting (XGBoost) algorithm was used to build the ML models, the importance of the features and predictive behaviors of which were illustrated by SHapley Additive exPlanations (SHAP)-based analyses. The sequential forward feature selection approach was used to generate the optimal feature subset. The co-administered drugs that might have influenced PRL levels during OLZ treatment as identified by SHAP analyses were then compared with evidence from disproportionality analyses by using OpenVigil FDA.

RESULTS

The 15 features that made the greatest contributions, as ranked by the mean (|SHAP value|), were identified as the optimal feature subset. The features were gender_male, co-administration of risperidone, age, co-administration of aripiprazole, concentration of aripiprazole, concentration of OLZ, progesterone, co-administration of sulpiride, creatine kinase, serum sodium, serum phosphorus, testosterone, platelet distribution width, α-L-fucosidase, and lipoprotein (a). The XGBoost model after feature selection delivered good performance on the validation cohort with a mean absolute error of 0.046, mean squared error of 0.0036, root-mean-squared error of 0.060, and mean relative error of 11%. Risperidone and aripiprazole exhibited the strongest associations with hyperprolactinemia and decreased blood PRL according to the disproportionality analyses, and both were identified as co-administered drugs that influenced PRL levels during OLZ treatment by SHAP analyses.

CONCLUSIONS

Multiple pathophysiological and pharmacological confounders influence PRL levels associated with effective treatment and PRL-related side-effects in OLZ-treated patients. Our study highlights the feasibility of integration of ML and EHR data to facilitate the detection of PRL levels and pharmacovigilance signals in OLZ-treated patients.

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

BACKGROUND

Early 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.

METHOD

We 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.

RESULT

Our 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.

CONCLUSION

ML algorithms are useful in early identifying male schizophrenia patients with violence and helping clinicians take preventive measures.

Maintaining social capital in offenders with schizophrenia spectrum disorder-An explorative analysis of influential factors

The importance of "social capital" in offender rehabilitation has been well established: Stable family and community relationships offer practical assistance in the resettlement process after being released from custody and can serve as motivation for building a new sense of self off the criminal past, thus reducing the risk of re-offending. This also applies to offenders with severe mental disorders. The aim of this study was to identify factors that promote or hinder the establishment or maintenance of social relationships upon release from a court-ordered inpatient treatment using a modern statistical method-machine learning (ML)-on a dataset of 369 offenders with schizophrenia spectrum disorder (SSD). With an AUC of 0.73, support vector machines (SVM) outperformed all the other ML algorithms. The following factors were identified as most important for the outcome in respect of a successful re-integration into society: Social integration and living situation prior to the hospitalization, a low risk of re-offending at time of discharge from the institution, insight in the wrongfulness of the offense as well as into the underlying psychiatric illness and need for treatment, addressing future perspectives in psychotherapy, the improvement of antisocial behavior during treatment as well as a detention period of less than 1 year emerged as the most predictive out of over 500 variables in distinguishing patients who had a social network after discharge from those who did not. Surprisingly, neither severity and type of offense nor severity of the psychiatric illness proved to affect whether the patient had social contacts upon discharge or not. The fact that the majority of determinants which promote the maintenance of social contacts can be influenced by therapeutic interventions emphasizes the importance of the rehabilitative approach in forensic-psychiatric therapy.

Potential markers for predicting delayed encephalopathy in patients with acute carbon monoxide poisoning

BACKGROUND

Acute carbon monoxide poisoning (ACOP) commonly results in delayed neuropsychiatric sequelae (DNS). Currently, there are no reliable predictors. The aim of this article is to establish a practical model for predicting the development of delayed encephalopathy clinically.

METHODS

Retrospective analysis of clinical data were performed at a single institution for the past 6 years. 107 patients with ACOP were recruited, of who 67 developed DNS and 40 did not. Clinical characteristics of the patients were analyzed between the two groups. The risk factors associated with DNS development were screened to identify the potential markers for predicting DNS. A predictive model was then built, and the receiver operating characteristic (ROC) curve analysis was used to assess its predictive ability.

RESULTS

There were significant differences in 13 clinical features between the two groups. Four potential markers were identified. They were age, source of CO, Glasgow Coma Scale score and the initiation of HBOT. The potential predictive model showed an area under the curve (AUC) of 0.93 in the training set and 0.97 in the testing set.

CONCLUSIONS

Our model could calculate the probability of DNS after acute CO poisoning.

Schizophrenia and substance use disorder: Characteristics of coexisting issues in a forensic setting

BACKGROUND AND AIMS

Recent research has identified higher prevalence of offending behavior in patients with comorbid schizophrenia spectrum disorder (SSD) and substance use disorder (SUD) compared to patients with SSD only and to the general population. However, findings on the subgroup of patients with SUD, SSD and offending behavior in forensic psychiatric care are scarce and inconsistent. The present study used machine learning to uncover more detailed characteristics of offender patients in forensic psychiatric care with comorbid SSD and SUD.

METHODS

Using machine learning algorithms, 370 offender patients (91.6 % male, mean age of M = 34.1, SD = 10.2) and 558 variables were explored in order to build three models to differentiate between no substance use disorder, cannabis use disorder and any other substance use disorder. To counteract the risk of overfitting, the dataset was split, employing variable filtering, machine learning model building and selection embedded in a nested resampling approach on one subset. The best model was then selected and validated on the second data subset.

RESULTS

Distinguishing between SUD vs. no drug use disorder yielded models with an AUC of 70 and 78. Variables assignable to demographics, social disintegration, antisocial behavior and illness were identified as most influential for the distinction. The model comparing cannabis use disorder with other substance use disorders provided no significant differences.

CONCLUSIONS

From a clinical perspective, offender patients suffering from schizophrenia spectrum and comorbid substance use disorder seem particularly challenging to treat, but initial differences in psychopathology will dissipate over inpatient treatment. Our data suggest that offender patients may benefit from appropriate treatment that focuses on illicit drug abuse to reduce criminal behavior and improve social integration.

Uncovering Barriers to Screening for Distress in Patients With Cancer via Machine Learning

BACKGROUND

Psychologic distress and manifest mental disorders are overlooked in 30-50% of patients with cancer. Accordingly, international cancer treatment guidelines recommend routine screening for distress in order to provide psychologic support to those in need. Yet, institutional and patient-related factors continue to hinder implementation.

OBJECTIVE

This study aims to investigate factors, which are associated with no screening for distress in patients with cancer.

METHODS

Using machine learning, factors associated with lack of distress screening were explored in 6491 patients with cancer between 2011 and 2019 at a large cancer treatment center. Parameters were hierarchically ordered based on statistical relevance. Nested resampling and cross validation were performed to avoid overfitting and to comply with assumptions for machine learning approaches.

RESULTS

Patients unlikely to be screened were not discussed at a tumor board, had inpatient treatment of less than 28 days, did not consult with a psychiatrist or clinical psychologist, had no (primary) nervous system cancer, no head and neck cancer, and did have breast or skin cancer. The final validated model was optimized to maximize sensitivity at 83.9%, and achieved a balanced accuracy of 68.9, area under the curve of 0.80, and specificity of 53.9%.

CONCLUSION

Findings of this study may be relevant to stakeholders at both a clinical and institutional level in order to optimize distress screening rates.

Predicting involuntary hospitalization in psychiatry: A machine learning investigation

Background

Coercion in psychiatry is a controversial issue. Identifying its predictors and their interaction using traditional statistical methods is difficult, given the large number of variables involved. The purpose of this study was to use machine-learning (ML) models to identify socio-demographic, clinical and procedural characteristics that predict the use of compulsory admission on a large sample of psychiatric patients.

Methods

We retrospectively analyzed the routinely collected data of all psychiatric admissions that occurred between 2013 and 2017 in the canton of Vaud, Switzerland (N = 25,584). The main predictors of involuntary hospitalization were identified using two ML algorithms: Classification and Regression Tree (CART) and Random Forests (RFs). Their predictive power was compared with that obtained through traditional logistic regression. Sensitivity analyses were also performed and missing data were imputed through multiple imputation using chain equations.

Results

The three models achieved similar predictive balanced accuracy, ranging between 68 and 72%. CART showed the lowest predictive power (68%) but the most parsimonious model, allowing to estimate the probability of being involuntarily admitted with only three checks: aggressive behaviors, who referred the patient to hospital and primary diagnosis. The results of CART and RFs on the imputed data were almost identical to those obtained on the original data, confirming the robustness of our models.

Conclusions

Identifying predictors of coercion is essential to efficiently target the development of professional training, preventive strategies and alternative interventions. ML methodologies could offer new effective tools to achieve this goal, providing accurate but simple models that could be used in clinical practice.

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

OBJECTIVES

The link between schizophrenia and violent offending has long been the subject of research with significant impact on mental health policy, clinical practice and public perception of the dangerousness of people with psychiatric disorders. The present study attempts to identify factors that differentiate between violent and non-violent offenders based on a unique sample of 370 forensic offender patients with schizophrenia spectrum disorder by employing machine learning algorithms and an extensive set of variables.

METHODS

Using machine learning algorithms, 519 variables were explored in order to differentiate violent and non-violent offenders. To minimize the risk of overfitting, the dataset was split, employing variable filtering, machine learning model building and selection embedded in a nested resampling approach on one subset. The best model was then selected, and the most important variables applied on the second data subset.

RESULTS

Ten factors regarding criminal and psychiatric history as well as clinical, developmental, and social factors were identified to be most influential in differentiating between violent and non-violent offenders and are discussed in light of prior research on this topic. With an AUC of 0.76, a sensitivity of 72% and a specificity of 62%, a correct classification into violent and non-violent offences could be determined in almost three quarters of cases.

CONCLUSIONS

Our findings expand current research on the factors influencing violent offending in patients with SSD, which is crucial for the development of preventive and therapeutic strategies that could potentially reduce the prevalence of violence in this population. Limitations, clinical relevance and future directions are discussed.

Exploring Similarities and Differences of Non-European Migrants among Forensic Patients with Schizophrenia

Migrants diagnosed with schizophrenia are overrepresented in forensic-psychiatric clinics. A comprehensive characterization of this offender subgroup remains to be conducted. The present exploratory study aims at closing this research gap. In a sample of 370 inpatients with schizophrenia spectrum disorders who were detained in a Swiss forensic-psychiatric clinic, 653 different variables were analyzed to identify possible differences between native Europeans and non-European migrants. The exploratory data analysis was conducted by means of supervised machine learning. In order to minimize the multiple testing problem, the detected group differences were cross-validated by applying six different machine learning algorithms on the data set. Subsequently, the variables identified as most influential were used for machine learning algorithm building and evaluation. The combination of two childhood-related factors and three therapy-related factors allowed to differentiate native Europeans and non-European migrants with an accuracy of 74.5% and a predictive power of AUC = 0.75 (area under the curve). The AUC could not be enhanced by any of the investigated criminal history factors or psychiatric history factors. Overall, it was found that the migrant subgroup was quite similar to the rest of offender patients with schizophrenia, which may help to reduce the stigmatization of migrants in forensic-psychiatric clinics. Some of the predictor variables identified may serve as starting points for studies aimed at developing crime prevention approaches in the community setting and risk management strategies tailored to subgroups of offenders with schizophrenia.

Aims to Reduce Coercive Measures in Forensic Inpatient Treatment: A 9-Year Observational Study

Protecting the human rights is particularly important within the forensic context because patients in forensic psychiatry are not admitted voluntarily and so the treatment itself is of a coercive nature. Coercive measures (i.e., actions against the will of the patient such as forced medication, seclusion or restraint) form an additional incision of personal rights. Although the use of coercion within forensic psychiatric institutions remains controversial, little empirical research has been conducted on the use of coercive measures within forensic settings. The study presented here can contribute to close this research gap by informing about rates of coercive measures within the present institution. National and international organizations on the prevention of torture or inhuman or degrading treatment have emphasized the need to keep the incidents of coercive measures to a minimum. Criticisms by such organizations on high rates of seclusion, restraint, and compulsory medication have led to organizational changes within the present institution which is Switzerland's largest forensic clinic with an average of 124 patients per year. After a first visit of such a committee, e.g., the detailed documentation of coercive measures became obligatory and part of special reports. Changes in the use of coercive measures are presented here. Data on coercive measures was analyzed for years 2010 to 2018. With respect to the most invasive coercive measurement, restraint, a minimum of four patients in 2017 and a maximum of 14 patients in 2010 have been subject to this form of coercive measurement. A minimum of sixteen patients in 2012 and a maximum of 40 patients in 2010 were secluded. Though total number and duration show a trend towards a reduction in severity of coercive measures on average, a few patients are not responsive to deescalating interventions. Preventive mechanisms, documentation standards, and efforts to ensure humane and adequate treatment are discussed under ethical considerations of coercive measures within court mandated treatment.