When do drugs trigger criminal behavior? a machine learning analysis of offenders and non-offenders with schizophrenia and comorbid substance use disorder.

Introduction: Comorbid substance use disorder (SUD) is linked to a higher risk of violence in patients with schizophrenia spectrum disorder (SSD). The objective of this study is to explore the most distinguishing factors between offending and non-offending patients diagnosed with SSD and comorbid SUD using supervised machine learning. Methods: A total of 269 offender patients and 184 non-offender patients, all diagnosed with SSD and SUD, were assessed using supervised machine learning algorithms. Results: Failures during opening, referring to rule violations during a permitted temporary leave from an inpatient ward or during the opening of an otherwise closed ward, was found to be the most influential distinguishing factor, closely followed by non-compliance with medication (in the psychiatric history). Following in succession were social isolation in the past, no antipsychotics prescribed (in the psychiatric history), and no outpatient psychiatric treatments before the current hospitalization. Discussion: This research identifies critical factors distinguishing offending patients from non-offending patients with SSD and SUD. Among various risk factors considered in prior research, this study emphasizes treatment-related differences between the groups, indicating the potential for improvement regarding access and maintenance of treatment in this particular population. Further research is warranted to explore the relationship between social isolation and delinquency in this patient population.

Predictive Factors Associated with Declining Psycho-Oncological Support in Patients with Cancer.

(1) Background: International cancer treatment guidelines recommend low-threshold psycho-oncological support based on nurses' routine distress screening (e.g., via the distress thermometer and problem list). This study aims to explore factors which are associated with declining psycho-oncological support in order to increase nurses' efficiency in screening patients for psycho-oncological support needs. (2) Methods: Using machine learning, routinely recorded clinical data from 4064 patients was analyzed for predictors of patients declining psycho-oncological support. Cross validation and nested resampling were used to guard against model overfitting. (3) Results: The developed model detects patients who decline psycho-oncological support with a sensitivity of 89% (area under the cure of 79%, accuracy of 68.5%). Overall, older patients, patients with a lower score on the distress thermometer, fewer comorbidities, few physical problems, and those who do not feel sad, afraid, or worried refused psycho-oncological support. (4) Conclusions: Thus, current screening procedures seem worthy to be part of daily nursing routines in oncology, but nurses may need more time and training to rule out misconceptions of patients on psycho-oncological support.

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.

Supplementary Motor Area Activity Differs in Parkinson's Disease with and without Freezing of Gait.

The study aimed to investigate the neural changes that differentiate Parkinson's disease patients with freezing of gait and age-matched controls, using ambulatory electroencephalography event-related features. Compared to controls, definite freezers exhibited significantly less alpha desynchronization at the motor cortex about 300 ms before and after the start of overground walking and decreased low-beta desynchronization about 300 ms before and about 300 and 700 ms after walking onset. The late slope of motor potentials also differed in the sensory and motor areas between groups of controls, definite, and probable freezers. This difference was found both in preparation and during the execution of normal walking. The average frontal peak of motor potential was also found to be largely reduced in the definite freezers compared with the probable freezers and controls. These findings provide valuable insights into the underlying structures that are affected in patients with freezing of gait, which could be used to tailor drug development and personalize drug care for disease subtypes. In addition, the study's findings can help in the evaluation and validation of nonpharmacological therapies for patients with Parkinson's disease.

Proof of concept: Predicting distress in cancer patients using back propagation neural network (BPNN).

Background: Research findings suggest that a significant proportion of individuals diagnosed with cancer, ranging from 25% to 60%, experience distress and require access to psycho-oncological services. Until now, only contemporary approaches, such as logistic regression, have been used to determine predictors of distress in oncological patients. To improve individual prediction accuracy, novel approaches are required. We aimed to establish a prediction model for distress in cancer patients based on a back propagation neural network (BPNN). Methods: Retrospective data was gathered from a cohort of 3063 oncological patients who received diagnoses and treatment spanning the years 2011-2019. The distress thermometer (DT) has been used as screening instrument. Potential predictors of distress were identified using logistic regression. Subsequently, a prediction model for distress was developed using BPNN. Results: Logistic regression identified 13 significant independent variables as predictors of distress, including emotional, physical and practical problems. Through repetitive data simulation processes, it was determined that a 3-layer BPNN with 8 neurons in the hidden layer demonstrates the highest level of accuracy as a prediction model. This model exhibits a sensitivity of 79.0%, specificity of 71.8%, positive predictive value of 78.9%, negative predictive value of 71.9%, and an overall coincidence rate of 75.9%. Conclusion: The final BPNN model serves as a compelling proof of concept for leveraging artificial intelligence in predicting distress and its associated risk factors in cancer patients. The final model exhibits a remarkable level of discrimination and feasibility, underscoring its potential for identifying patients vulnerable to distress.

A short screening tool identifying systemic barriers to distress screening in cancer care.

INTRODUCTION: International guidelines on cancer treatment recommend screening for early detection and treatment of distress. However, screening rates are insufficient. In the present study, a survey was developed to assess perceived systemic barriers to distress screening. METHODS: A three-step approach was used for the study. Based on qualitative content analysis of interviews and an expert panel, an initial survey with 53 questions on barriers to screening was designed. It was completed by 98 nurses in a large comprehensive cancer center in Switzerland. From this, a short version of the survey with 24 questions was derived using exploratory principal component analysis. This survey was completed by 150 nurses in four cancer centers in Switzerland. A confirmatory factor analysis was then performed on the shortened version, yielding a final set of 14 questions. RESULTS: The initial set of 53 questions was reduced to a set of 14 validated questions retaining 53% of the original variance. These 14 questions allow for an assessment within 2-3 min that identifies relevant barriers to distress screening from the perspective of those responsible for implementation of distress screening. Across several hospitals in Switzerland, the timing of the first distress screening, lack of capacity, patient and staff overload, and refusal of distressed patients to be referred to support services emerged as major problems. CONCLUSION: The validated 14 questions on barriers to screening cancer patients for distress enable clinicians and hospital administrators to quickly identify relevant issues and take action to improve screening programs.

Steriod-associated psychiatric burden in cancer patients.

This study explores the role of steroid administration in identifying distressed or even mentally disordered cancer patients (so-called case finding). Charts of 12 298 cancer patients (4499 treated with prednisone equivalents) were analysed descriptively. A subset of 10 945 was further explored via latent class analysis (LCA). LCA avoids confounding by indication because it subgroups patients without prior preconceptions based on homogeneous expression of traits (i.e. the variables examined). LCA identified four subgroups: two subgroups with high dosages of prednisone equivalent (≥80 mg/day on average over all treatment days) and two with low dosages. The two subgroups with high average dosages had an increased likelihood of psychotropic drug administration, but only one was more likely to require 1:1 observation. In one subgroup, low dosages of prednisone equivlents correlated with a slightly increased probability for a psychiatric assessment and psychotropic drug administration. The subgroup least likely to receive steroid treatment was also the least likely to receive a psychiatric assessment and psychotropic drug administration. Descriptive statistics on age, sex, cumulative inpatient treatment, type of cancer, stage of cancer at first diagnosis, mental disorders, severe mental disorders and psychotropic drug administration (antidepressants, antipsychotics, benzodiazepines, anticonvulsants/mood stabilizers, opioids) are provided for patients receiving no, less and more than 80 mg of prednisone equivalent.

Early Impact of the COVID-19 Pandemic on Psycho-Oncological Support: A Latent Class Analysis.

INTRODUCTION: Research suggests a global shortfall of psycho-oncological assessment and care during the COVID-19 pandemic in addition to delayed diagnosis of cancer. The present study is the first to explore the effect of the pandemic on the provision of psycho-oncological care, stage of cancer at first diagnosis, and duration of hospitalizations. METHOD: Retrospective latent class analysis of 4,639 electronic patient files with all types, treatment types, and stages of cancer, 370 of which were treated during the pandemic prior to availability of vaccinations. DISCUSSION: Latent class analysis identified four subgroups based on differences in screening for distress, provision of psycho-oncological support (consultation with a psychiatrist or clinical psychologist), administration of psychotropic medication, use of 1:1 observation, stage of cancer at first diagnosis, and duration of hospitalizations. Yet, the pandemic had no effect on subgrouping. Thus, the COVID-19 pandemic had no effect on the provision of psycho-oncological support. CONCLUSION: Results are contrary to prior research. The efficiency and quality of procedures implemented to provide psycho-oncological support during and prior to the pandemic are critically reflected.

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.

Differential associations of emotional and physical domains of the MacNew Heart with changes in 6-min walking test.

AIMS: Cardiac rehabilitation (CR), a key component of secondary prevention in cardiac patients, contributes fundamentally to improved cardiovascular health outcomes. Health-related quality of life (HRQOL) represents a widely employed outcome measure in CR, yet, its predictive properties on exercise capacity change during CR are poorly understood. Aim of this study was to examine the association between baseline HRQOL and its subdomains on improvement of exercise capacity during CR. METHODS: Study participants were 13,717 inpatients of six Swiss CR clinics from 2012 to 2018. We measured HRQOL at admission to CR with the MacNew Heart (MNH) questionnaire and exercise capacity at admission and discharge using the six minutes walking test (6MWT). Following factorial analyses, we performed univariate and multivariate analyses to test the predictive properties of baseline global HRQOL and its domains for improvement in exercise capacity, adjusting for demographic and clinical characteristics. RESULTS: Mean improvement in 6MWT was 114 m (SD = 90), achieved after 17.4 days (SD = 5.5). Lower emotional HRQOL (b = 7.85, p = < .001, 95% CI [- 5.67, 10.03]) and higher physical HRQOL (b = - 5.23, p < .001, 95% CI [- 6.56, - 3.90]) were associated with less improvement in the 6MWT. Global MNH and social HRQOL showed no association with exercise capacity improvement. CONCLUSION: Patients entering CR with low emotional and high physical HRQOL are at risk for a lower gain in exercise capacity during CR. Global MNH alone does not provide a reliable assessment of HRQOL; thus a focus on specific domains of HRQOL is needed.

Clinically Significant Distress and Physical Problems Detected on a Distress Thermometer are Associated With Survival Among Lung Cancer Patients.

BACKGROUND: The distress thermometer is a well-established screening tool to detect clinically significant distress in cancer patients. It is often administered in combination with the problem list, differentiating further between various (e.g., physical and emotional) sources of distress. OBJECTIVE: The present study aimed to extend previous research on the association between distress and overall survival. A further exploratory analysis aimed to evaluate the predictive value of the problem list for overall survival. METHODS: Patients (n = 323) with newly diagnosed lung cancer were recruited from a large cancer center. Patients were split into 2 groups, those with (distress thermometer score ≥5) and those without significant distress. The overall survival time was illustrated by a Kaplan-Meier curve and compared with a log-rank test. Univariable Cox proportional hazard models were built to control the association of distress with overall survival for age, gender, disease stage, comorbidity, and their interaction terms. Multiple linear regression was used to investigate the association of the items from the problem list with survival time. RESULTS: Patients with significant distress had a shorter survival time than patients without significant distress (25 vs. 43 months). Regression analysis revealed more problems with both "bathing and dressing" and "eating," as well as absence of "diarrhea" and increased "nervousness," to negatively associated with the overall survival time. CONCLUSIONS: Our results show that estimation of the survival function using cancer-related distress is possible. However, when using Cox regression, distress shows no significant value for survival as a predictor. Moreover, our study did not reveal an interaction effect among disease stage, comorbidity, and distress. Overall, results suggest that physical and emotional problems that arise from lung cancer may be useful to identify patients at risk of poor prognosis (on the basis of Kaplan-Meier estimator).

High Risk, High Dose?-Pharmacotherapeutic Prescription Patterns of Offender and Non-Offender Patients with Schizophrenia Spectrum Disorder.

Compared to acute or community settings, forensic psychiatric settings, in general, have been reported to make greater use of antipsychotic polypharmacy and/or high dose pharmacotherapy, including overdosing. However, there is a scarcity of research specifically on offender patients with schizophrenia spectrum disorders (SSD), although they make up a large proportion of forensic psychiatric patients. Our study, therefore, aimed at evaluating prescription patterns in offender patients compared to non-offender patients with SSD. After initial statistical analysis with null-hypothesis significance testing, we evaluated the interplay of the significant variables and ranked them in accordance with their predictive power through application of supervised machine learning algorithms. While offender patients received higher doses of antipsychotics, non-offender patients were more likely to receive polypharmacologic treatment as well as additional antidepressants and benzodiazepines. To the authors' knowledge, this is the first study to evaluate a homogenous group of offender patients with SSD in comparison to non-offender controls regarding patterns of antipsychotic and other psychopharmacologic prescription patterns.

Reading Wishes from the Lips: Cancer Patients' Need for Psycho-Oncological Support during Inpatient and Outpatient Treatment.

BACKGROUND: Psycho-oncological support (PO) is an effective measure to reduce distress and improve the quality of life in patients with cancer. Currently, there are only a few studies investigating the (expressed) wish for PO. The aim of this study was to evaluate the number of patients who request PO and to identify predictors for the wish for PO. METHODS: Data from 3063 cancer patients who had been diagnosed and treated at a Comprehensive Cancer Center between 2011 and 2019 were analyzed retrospectively. Potential predictors for the wish for PO were identified using logistic regression. As a novelty, a Back Propagation Neural Network (BPNN) was applied to establish a prediction model for the wish for PO. RESULTS: In total, 1752 patients (57.19%) had a distress score above the cut-off and 14.59% expressed the wish for PO. Patients' requests for pastoral care (OR = 13.1) and social services support (OR = 5.4) were the strongest predictors of the wish for PO. Patients of the female sex or who had a current psychiatric diagnosis, opioid treatment and malignant neoplasms of the skin and the hematopoietic system also predicted the wish for PO, while malignant neoplasms of digestive organs and older age negatively predicted the wish for PO. These nine significant predictors were used as input variables for the BPNN model. BPNN computations indicated that a three-layer network with eight neurons in the hidden layer is the most precise prediction model. DISCUSSION: Our results suggest that the identification of predictors for the wish for PO might foster PO referrals and help cancer patients reduce barriers to expressing their wish for PO. Furthermore, the final BPNN prediction model demonstrates a high level of discrimination and might be easily implemented in the hospital information system.

General practitioner-centred paediatric primary care reduces risk of hospitalisation for mental disorders in children and adolescents with ADHD: findings from a retrospective cohort study.

BACKGROUND: General practitioners (GPs) play an essential role in the sustainable management of attention-deficit/hyperactivity disorder (ADHD). To our knowledge, the healthcare programme described here is the first integrated care programme for paediatric ambulatory care embedded in GP-centred-healthcare in Germany. OBJECTIVES: To compare the health-service-utilisation of patients with ADHD enrolled in a GP-centred-paediatric-primary-care-programme with usual care in terms of disease-related hospitalisation, pharmacotherapy and psychotherapy. METHODS: In 2018, we conducted a retrospective cohort study of 3- to 18-year-old patients with ADHD in Baden-Wuerttemberg, southern Germany. The intervention group (IG) comprised patients enrolled in a GP-centred-paediatric-primary-healthcare-programme and consulted a participating GP for ADHD at least once. GP-centred-paediatric-primary-care provides high continuity of care, facilitated access to specialist care, extended routine examinations and enhanced transition to adult healthcare. Patients in the control group (CG) received usual care, meaning they consulted a non-participating GP for ADHD at least once. Main outcomes were disease-related hospitalisation, pharmacotherapy and psychotherapy. Multivariable logistic regression was performed to compare groups. RESULTS: A total of 2317 patients were included in IG and 4177 patients in CG. Mean age was 8.9 ± 4.4. The risk of mental-disorder-related hospitalisations was lower in IG than CG (odds ratio (OR): 0.666, 95% confidence interval (CI): 0.509-0.871). The prescription rate for stimulants was lower in IG (OR: 0.817; 95% CI: 0.732-0.912). There was no statistically significant difference in the participation rate of patients in cognitive behavioural therapy between groups (OR: 0.752; 95% CI: 0.523-1.080). CONCLUSION: Children and adolescents with ADHD enrolled in GP-centred-paediatric-primary-care are at lower risk of mental-disorder-related hospitalisation and less likely to receive stimulants.

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.

Severe mental illness in cancer is associated with disparities in psycho-oncological support.

Patients with both cancer and a severe mental illness (SMI) have a higher risk of advanced stage cancer at diagnosis and poorer survival in comparison to individuals with cancer alone. The present study explores if similar disparities exist in terms of psycho-oncological support. Latent class analysis (LCA) was used to group 10,945 patients with any type of cancer, of which 72 (0.7%) had been diagnosed with a SMI (ICD10-codes F20-F22, F24, F25, F28-F31, F32.3, F33.3), and 1056 (9.6%) with another mental disorder. Subgrouping was based on presence of SMI, other mental illnesses, stage of cancer at its first detection, screening for distress and receipt of information on psycho-oncology, consultation with a psychotherapist and/or psychiatrist, prescription of different psychotropic medication, and use of a patient care attendant. Five subgroups were identified. Patients with SMI were most likely to suffer from further mental comorbidities, to be prescribed antipsychotics, antidepressants, or mood stabilizers, and be in need of a patient care attendant. In comparison to patients without SMI, the larger one of 2 subgroups of patients with SMI had a low probability to be screened for distress and informed about psycho-oncological support services. A smaller subgroup of patients with SMI was probable to be diagnosed with an advanced stage of cancer. In subgroups without patients with mental disorders, screening for distress and offering psycho-oncological support seemed to be economized unless benzodiazepines or opioids were prescribed. Contrary to published evidence, distress screening and offering psycho-oncological support is neglected in patients with SMI unless an advanced stage of cancer is being diagnosed.

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.

Distinct psycho-oncological support inclinations and needs in patients with cancer: A large sample latent class analysis approach.

BACKGROUND: In patients with cancer, the routine recording of distress symptoms has been widely established in recent years. Psycho-oncological support has proven to reduce distress and increase quality of life. Despite high levels of distress as well as physical and emotional challenges in patients with cancer, a significant proportion forgoes psycho-oncological services. METHODS: A cross-sectional retrospective evaluation was carried out. Latent class analysis was used to examine the relationship between distress, physical and emotional challenges, and desire for psycho-oncological services in 2191 patients with cancer. RESULTS: Latent class analysis revealed four homogeneous subgroups: a) patients with high distress, high physical and low emotional challenges and no desire for psycho-oncology, b) patients with high distress, low physical and high emotional challenges and no desire for psycho-oncology, c) patients with high distress, high physical and emotional challenges and a desire for psycho-oncology, d) patients with low distress, low physical and emotional challenges and no desire for psycho-oncology. CONCLUSION: The identification of these subgroups of patients with cancer is useful for health care providers in order to focus their efforts in patients with cancer. It might contribute to a more tailored treatment offer for specific subgroups whose needs have so far been insufficiently met.

The Reconstruction of Causal Networks in Physiology.

We systematically compare strengths and weaknesses of two methods that can be used to quantify causal links between time series: Granger-causality and Bivariate Phase Rectified Signal Averaging (BPRSA). While a statistical test method for Granger-causality has already been established, we show that BPRSA causality can also be probed with existing statistical tests. Our results indicate that more data or stronger interactions are required for the BPRSA method than for the Granger-causality method to detect an existing link. Furthermore, the Granger-causality method can distinguish direct causal links from indirect links as well as links that arise from a common source, while BPRSA cannot. However, in contrast to Granger-causality, BPRSA is suited for the analysis of non-stationary data. We demonstrate the practicability of the Granger-causality method by applying it to polysomnography data from sleep laboratories. An algorithm is presented, which addresses the stationarity condition of Granger-causality by splitting non-stationary data into shorter segments until they pass a stationarity test. We reconstruct causal networks of heart rate, breathing rate, and EEG amplitude from young healthy subjects, elderly healthy subjects, and subjects with obstructive sleep apnea, a condition that leads to disruption of normal respiration during sleep. These networks exhibit differences not only between different sleep stages, but also between young and elderly healthy subjects on the one hand and subjects with sleep apnea on the other hand. Among these differences are 1) weaker interactions in all groups between heart rate, breathing rate and EEG amplitude during deep sleep, compared to light and REM sleep, 2) a stronger causal link from heart rate to breathing rate but disturbances in respiratory sinus arrhythmia (breathing to heart rate coupling) in subjects with sleep apnea, 3) a stronger causal link from EEG amplitude to breathing rate during REM sleep in subjects with sleep apnea. The Granger-causality method, although initially developed for econometric purposes, can provide a quantitative, testable measure for causality in physiological networks.

Stress, Schizophrenia, and Violence: A Machine Learning Approach.

This study employs machine learning algorithms to examine the causes for engaging in violent offending in individuals with schizophrenia spectrum disorders. Data were collected from 370 inpatients at the Centre for Inpatient Forensic Therapy, Zurich University Hospital of Psychiatry, Switzerland. Based on findings of the general strain theory and using logistic regression and machine learning algorithms, it was analyzed whether accumulation and type of stressors in the inpatients' history influenced the severity of an offense. A higher number of stressors led to more violent offenses, and five types of stressors were identified as being highly influential regarding violent offenses. Our findings suggest that an accumulation of stressful experiences in the course of life and certain types of stressors might be particularly important in the development of violent offending in individuals suffering from schizophrenia spectrum disorders. A better understanding of risk factors that lead to violent offenses should be helpful for the development of preventive and therapeutic strategies for patients at risk and could thus potentially reduce the prevalence of violent offenses.