Predictive value of insight for suicide,violence, hospitalization,and social adjustment for outpatients with schizophrenia: a prospective study

This prospective study explores the predictive value of insight for suicide, violence, hospitalization, and social adjustment for a group of schizophrenic outpatients over 1 year. The Schedule of Assessment of Insight (SAI) and its expanded version (SAI-E) were used to provide a baseline insight score for 74 schizophrenic outpatients considered to be in remission or to have minimal psychopathology. Subjects were reassessed at the end of the period to determine the predictive value of initial insight rating by comparison. Insight into treatment was associated with less hospitalization and better social adjustment, but insight into mental illness or psychotic experience was not. It was also found that insight could neither predict suicidal behavior or violent tendencies. Therefore, some dimension of insight may predict hospitalization and social adjustment in schizophrenics, although its predictive efficacy for other clinical outcomes and behaviors may be limited.     

Characteristic hostility in schizophrenic outpatients

In this study of 133 schizophrenic outpatients, we assessed characteristic hostility and correlates of hostility over a 6-month period. Results showed that 13 percent of the study group were characteristically violent, 18 percent were characteristically threatening, and another 21 percent were irritable and argumentative. About half (48%) were without hostility. A multiple regression model identified six variables--housing instability, hallucinations or delusions, schizoaffective diagnosis, lack of depression, alcohol use, and bizarre behavior--that together accounted for over 50 percent of the variance in observed characteristic hostility. Hostility also predicted rehospitalization and total inpatient days during 1-year followup. Implications of these findings for assessment and future research are discussed.     

The role of violence in decisions about hospitalization from the psychiatric emergency room.

OBJECTIVE: The authors evaluated the relationship between violent behavior and decision making about hospitalization from the psychiatric emergency room. METHOD: The medical charts of 321 patients evaluated in an urban psychiatric emergency room during a 4-week period were reviewed retrospectively. Violent behavior was defined as physical attacks on persons or fear-inducing behavior before or during the evaluation in the emergency room; and its value in predicting hospitalization decisions was assessed with logistic regression analyses that also included 12 demographic, clinical, and contextual variables. RESULTS: A model predicting hospitalization decisions was developed and cross-validated. Although violent patients were more likely to be hospitalized than nonviolent patients, clinical variables such as diagnosis and overall severity of psychiatric impairment were more important than violent behavior in predicting hospitalization decisions. CONCLUSIONS: Despite legal pressures to focus on overt behaviors such as violence as a basis for liability prevention and civil commitment, clinicians in this study did not allocate inpatient resources to preventively detain persons unlikely to benefit from treatment. Rather, they hospitalized the most severely disturbed patients, with diagnoses such as schizophrenic and manic disorders for which a widely accepted therapeutic armamentarium exists. The results are consistent with clinical recommendations that in the evaluation of the violent patient, attention needs to be given to the underlying disorder, since violent behavior itself can result from diverse causes only some of which require inpatient psychiatric treatment.     

Violence in inpatients with schizophrenia: a prospective study.

Accurate evaluations of the dangers posed by psychiatric inpatients are necessary, although a number of studies have questioned the accuracy of violence prediction. In this prospective study, we evaluated several variables in the prediction of violence in 63 inpatients with a DSM-IV diagnosis of schizophrenia or schizoaffective disorder. Nurses rated violent incidents with the Overt Aggression Scale. During hospitalization, sociodemographic variables, clinical history, neurological soft signs, community alcohol or drug abuse, and electroencephalographic abnormalities did not differ between violent and nonviolent groups. Violent patients had significantly more positive symptoms as measured by the Positive and Negative Syndrome Scale (PANSS), higher scores on the PANSS general psychopathology scale, and less insight in the different constructs assessed. A logistic regression was performed to discriminate between violent and nonviolent patients. Three variables entered the model: insight into symptoms, PANSS general psychopathology score, and violence in the previous week. The actuarial model correctly classified 84.13 percent of the sample; this result is significantly better than chance for the base rate of violence in this study. At hospital admission, clinical rather than sociodemographic variables were more predictive of violence. This finding has practical importance because clinical symptoms are amenable to therapeutic approaches. This study is the first to demonstrate that insight into psychotic symptoms is a predictor of violence in inpatients with schizophrenia.     

Risk factors and correlates of violence among acutely ill adult psychiatric inpatients

OBJECTIVE: The purpose of the study was to identify risk factors and correlates of violence committed by patients in an acute adult psychiatric inpatient unit in a district general hospital of the United Kingdom's National Health Service. METHODS: Incidents of violence committed by inpatients over a one-year period in 1997-1998 were retrospectively analyzed. The clinical characteristics of 49 violent patients were compared with those of all patients admitted to the unit during the study period (N=474) and with a random sample of nonviolent patients (N=140). Logistic regression analysis was used to identify clinical variables that predicted violent behavior. RESULTS: Violence was not positively associated with schizophrenia or negatively associated with depression. Frequent medication change, high use of sedative drugs, past violent behavior, an ICD-10 diagnosis of dissocial personality disorder or emotionally unstable personality disorder (DSM-IV antisocial personality disorder or borderline personality disorder), and long hospitalization were the most powerful predictors of violence. Together these variables had a sensitivity of 76 percent, a specificity of 97 percent, and a positive predictive value of 90 percent in predicting which patients became violent. Compulsory (involuntary) admission, comorbid diagnoses, past self-harm, and nonalcohol drug abuse were also associated with violent behavior. CONCLUSIONS: Clinicians' judgment about an inpatient's potential for violence may be augmented by knowledge of the risk factors identified in this study. Medication variables could be especially useful predictors, particularly when information about other risk factors is not available. Factors other than mental illness per se may be crucial determinants of violence in acute inpatient settings.     

Relationships between insight and medication adherence in outpatients with schizophrenia and bipolar disorder: prospective study

The aim of the present study was to explore the associations between insight and medication adherence at index interview and at 1-year follow-up interview in bipolar and schizophrenic outpatients. The Schedule for Assessment of Insight (SAI) and its expanded version (SAI-E) were used to provide a baseline insight score for 65 bipolar subjects and 74 schizophrenic subjects considered to be in remission or to have minimal psychopathology. Medication adherence of subjects was assessed at index interview and at 1-year follow-up interview, and the association between insight and medication adherence was analyzed cross-sectionally and prospectively. The results of the analysis reveal that in bipolar subjects, the index SAI scores for insight into treatment, mental-health status and psychotic experiences, and total SAI-E were positively correlated with medication adherence at both index and 1-year follow-up interviews. However, in schizophrenic subjects, insight into treatment and total SAI-E correlated positively with medication adherence at index interview but not at 1-year follow-up interview. Medication adherence at index interview could predict medication adherence 1 year later in both bipolar and schizophrenic subjects. These results indicate that the predictive value of insight for medication adherence differs between bipolar and schizophrenic patients, and building insight is an important step for establishing medication adherence in bipolar patients.     

Personality and substance use disorders as predictors of criminality

The study objective was to evaluate the relationship between a personality disorder (PD) diagnosis and criminal behavior among drug- and alcohol-dependent patients both retrospectively and prospectively. We examined 1-year pretreatment and 1-year post-treatment crime rates among 370 drug- and/or alcohol-dependent patients. Hierarchical logistic regression was used to examine the predictive value of DSM-III-R PD diagnoses after controlling for demographic features and type and severity of substance dependence. Patients with a diagnosis of antisocial PD (ASPD) were more likely to report having committed a variety of crimes, including violent crimes, during the pretreatment period. Individuals with more PD diagnoses or a diagnosis of borderline PD (BPD) or schizoid PD also reported a greater number of pretreatment violent crimes. In addition, the number of PD diagnoses was correlated with the number of crimes against property. During the pretreatment period, significant interactions were also found between PD measures and substance use in relation to both property crimes and violation of parole or probation. During the posttreatment period, a diagnosis of BPD predicted the commission of violent crimes. In contrast, a cluster A PD predicted a lower frequency of crimes against property. ASPD did not predict criminality during the 1-year follow-up period. In conclusion, a PD diagnosis, particularly ASPD, was associated with a variety of criminal behaviors during the 1-year period preceding substance abuse treatment. Following treatment, PD diagnosis had limited value in the prediction of criminal behavior. Possible explanations for these findings are discussed.     

Characteristics of voluntary participants versus nonparticipants in a psychoeducation program for euthymic patients with bipolar disorder

We aimed to assess the participation rate and predictive factors of participation in psychoeducation programs for euthymic outpatients with bipolar disorder. Ninety-five consecutive euthymic outpatients with bipolar disorder treated with lithium were recruited in a university department of psychiatry. The participants and nonparticipants in a program of psychoeducation were compared for sociodemographic, clinical, and psychological characteristics. According to univariate statistics, a younger age, a higher education level, a shorter duration of illness, a better initial knowledge about lithium, and a less external locus of control were predictive of participation in the program. A binary logistic regression model showed that an external locus of control was an independent predictor of participation. Among bipolar patients, the older, the less educated, those who have less knowledge about their treatment, and those with a more external locus of control were less likely to participate in hospital-based psychoeducation programs.     

Premorbid adjustment and violent behavior in schizophrenic patients

We examined the influence of premorbid adjustment on violent behavior in schizophrenic patients. There is some evidence that poor premorbid adjustment predicts violent behavior, then we decided to examine this hypothesis further. METHOD: We recruited 72 schizophrenic patients without concomitant substance abuse 6 months prior to the assessment. Diagnoses were based on the SCID-I. Premorbid adjustment was evaluated with the Premorbid Adjustment Scale and violent behaviors with the Overt Aggression Scale. RESULTS: Violent schizophrenic patients showed an overall worse premorbid adjustment during childhood. In addition, the area of "peer relationships" was significantly diminished in several life period sections such as childhood, early and late adolescence in violent patients. DISCUSSION: Our data indicate that difficulties in social relationships during early stages of life may increase the risk of future violent behavior among schizophrenic patients.     

Individual psychotherapy with schizophrenic outpatients in the public mental health system.

Forty-two therapists conducting individual psychotherapy with schizophrenic outpatients in a public mental health system responded to a questionnaire that focused on the types of interventions used and the issues and problems encountered in therapy. The therapists spent 59 percent of their time in supportive, problem-solving work and only 32 percent in traditional psychotherapeutic interventions, such as providing insight. The most common issues in therapy were relationship problems, family concerns, depression, losses, and the role of medications in the client's life. The most significant impediments to therapeutic work were the lack of community resources, the client's noncompliance with medications and lack of motivation, and dual diagnoses. Suggestions are offered for improving psychotherapy with schizophrenic outpatients in the public mental health system.     

Clinical characteristics, cognitive functioning, and criminal histories of outpatients with schizophrenia

OBJECTIVE: The authors examined the relationships between clinical characteristics, cognitive functioning, and history of violent behavior and substance use among outpatients with schizophrenia. METHODS: Ninety-six patients with a diagnosis of schizophrenia or schizoaffective disorder completed a clinical and neuropsychiatric battery that included tests of general intelligence, executive and frontal lobe function, visual-motor processing, and motor function. Violent behavior was defined on the basis of arrest records. Self-reported violent behavior and substance use were recorded. The study participants were separated into three groups: history of violent arrest (N=34), nonviolent arrest (N=23), and no arrest (N=39). The three groups were compared for differences in demographic characteristics, clinical symptoms, and scores on neuropsychological tests. RESULTS: Fifty-seven (59 percent) of the 96 participants had a history of arrest. Persons who were arrested for nonviolent crimes had a significantly lower mean+/-SD number of arrests (3.39+/-3.7) than those arrested for violent crimes (9.24+/-8.9). No significant differences in neuropsychological test scores or clinical ratings were found between the three groups. The prevalence of substance use disorders was 65 percent, 57 percent, and 36 percent among patients with a history of violent, nonviolent, and no arrest, respectively. Only 47 percent of participants with a criminal history accurately reported this history, and 11 percent of participants with a history of drug-related arrests acknowledged previous substance use. CONCLUSIONS: Performance on neuropsychiatric tests does not distinguish stable outpatients with schizophrenia who have a history of violent behavior from those who do not have such a history. Two established predictors of violence, a history of arrests and substance abuse, are unreliable when assessed by self-report.     

Associations with violent and homicidal behaviour among men with schizophrenia

Objective We aimed to assess the risk factors associated with homicidal behaviour in male patients diagnosed with schizophrenia. Methods In a period of 1 year, male schizophrenia cases between 18–65 years of age (n?=?210) were included. The clinical evaluation included the Positive and Negative Syndrome Scale (PANSS) and Overt Aggression Scale (OAS). The patients were divided into three groups in terms of violent behaviour history: (1) homicide group (n?=?30), (2) a violent act resulting in serious injury (n?=?71), (3) control group (patients without a history of a violent act) (n?=?109). Results Lower level of education, rural residence, being unemployed and living alone were found to be significantly more common in patients who had committed a violent act compared to the schizophrenia patients in the control group. In order to explore the predictive value of several factors associated with violent behaviour, a logistic regression model was used, and variables (shorter duration of education, living alone, and lack of insight) significantly predicted the presence of violent behaviour (murder and/or injury) (χ²=31.78, df?=?12, p?=?0.001). Conclusions In order to be able to determine causality of homicidal acts in schizophrenia patients, our significant findings between homicidal violence, non-homicidal violence and the control group would merit further attention and exploration in further studies.     

Hopelessness and its impact on rehabilitation outcome in schizophrenia -an exploratory study

The primary focus in contemporary psychiatry on symptoms and their neurobiological basis, although fundamentally important, is nevertheless incomplete. The long-term course and outcome of schizophrenia are determined not only by the disorder, but also by the interaction between the person and the disorder. Not only psychopathological symptoms but also cognitive variables such as negative self-concepts, low expectations and external loci of control can influence the patient's coping strategies and may lead to hopelessness and chronicity. Hopelessness here refers to a cognitive-affective state in which the patient perceives the disorder and its consequences to be beyond his control, feels helpless and has given up expecting to influence its course positively, thereby abandoning responsibility and active coping strategies. In a prospective study, we examined these relationships by using logistic regression in data from 46 schizophrenic outpatients who were participating in a vocational rehabilitation program. Negative self-concepts, external loci of control, and depression correlated to a higher extent with depressive-resigned coping strategies than did schizophrenic symptoms. Thus, poor rehabilitation outcome may be predicted to a high degree by the presence of external loci of control, pessimistic outcome expectancies, negative symptoms, and depressive-resigned coping strategies. After having eliminated the influence of negative symptoms, external control beliefs still had significant predictive value for the outcome. Rehabilitation outcome in schizophrenic patients can be only partially predicted by negative symptoms; the other predictive factor is whether the patient has already given up or not.     

The predictive value of benzodiazepine tolerance in persistently aggressive schizophrenia.

The aim of this study was to determine whether benzodiazepine tolerance might provide a predictive marker for persistent aggression in schizophrenia. Seventy-seven male schizophrenic patients newly admitted to our psychiatric intensive care unit due to violent behavior during a 4-month period were examined. As a result, a high dose of benzodiazepine required for sedation or a short duration until regaining consciousness after the initial sedation, was related to severer aggression on waking up after sedation. Despite the small number of subjects examined, a conservative claim can be made that the level of the effect of benzodiazepine required for sedation seems to predict persistent severe aggression in schizophrenia.     

A comparison of statistical models in predicting violence in psychotic illness

BACKGROUND: The application of statistical modeling techniques, including classification and regression trees, in the prediction of violence has increasingly received attention. METHODS: The predictive performance of logistic regression and classification tree methods in predicting violence was explored in a sample of patients with psychotic illness. RESULTS: Of 2 logistic regression models, the forward stepwise method produced a simpler model than the full model, but the latter performed better. The performance of the classification tree appeared to be high before cross-validation, but reduced when cross-validated. The standard logistic model was the most robust model. A simplified tree with extra weight given to violent cases was a reasonable competitor and was simple to apply. CONCLUSION: Although classification trees can be suitable for routine clinical practice, because of the simplicity of their decision-making processes, their robustness and therefore clinical utility was problematic in this sample. Further research is required to compare such models in large prospective epidemiologic studies of other psychiatric populations.     

Sensitivity and specificity of the Brøset Violence Checklist as predictor of violence in forensic psychiatry

Abstract

Background: Violence and aggressive behavior within psychiatric facilities are serious work environment problems, which have negative consequences for both patients and staff. It is therefore of great importance to reduce both the number and the severity of these violent incidents to improve quality of care. Aims: To evaluate the specificity and sensitivity of the Brøset Violence Checklist (BVC) as a predictor of violent incidents for Danish forensic psychiatry patients. Method: A total of 156 patients were assessed three times daily with the BVC for 24 months. All aggressive or violent incidents were recorded using the Staff Observation Aggression Scale-Revised (SOAS-R). SOAS-R scores of 9 or more defined violent incidents. Data were analyzed using standard logistic regression models as well as models incorporating a random person effect. We used receiver operating curve (ROC) analysis to evaluate different BVC thresholds. Results: Of a total of 139,579 BVC registrations we found 1999 scores above 0 and 419 violent incidents. The BVC score was a strong predictor of violence. For the standard cut-off point of 3, specificity was 0.997 and sensitivity was 0.656. For the general risk of violence seen in this study, the risk of violence given a BVC score > 3 (positive predictive value) was 37.2%, and the risk of violence given a BVC score < 3 (negative predictive value) was 0.1%. Conclusion: The BVC showed satisfactory specificity and sensitivity as a predictor of the short-term risk of violence against staff and others by patients in a forensic setting.     

Associated factors of rehospitalization among schizophrenic patients

The purpose of the present study was to identify the associated factors of rehospitalization in schizophrenic patients. A case-control study was conducted. The cases consisted of rehospitalized patients (n = 67) and controls selected from the outpatients who were matched by age, gender, and the period after the last discharge (n = 62). In the multiple logistic regression analysis, no clinic visits in the second month prior to entry, the number of clinic visits in the previous month, and junior high school graduation as education level were significantly (P < 0.01) associated with rehospitalization after controlling their present function as assessed by the Global Assessment of Functioning. Close monitoring of clinic visits and outreach service appear to be important in preventing rehospitalization of schizophrenic patients. These identified modifiable factors suggest further needs for development and implementation of integrated mental health services in the community.     

机器学习基于不平衡数据预测急性新发缺血性卒中患者院内死亡模型研究

目的 探索利用机器学习基于不平衡数据预测急性新发缺血性卒中患者的院内死亡风险,并比较机器学习模型和传统logistic模型的预测性能.方法 以中国卒中联盟多中心登记数据库中急性新发缺血性卒中患者为研究对象,分别基于机器学习[XGBoost模型、CatBoost模型、随机森林模型、支持向量机(support vector...     

Relationship between preadmission threats and later violent behavior by acute psychiatric inpatients

The medical charts of 253 patients admitted to an acute psychiatric inpatient unit were reviewed for evidence of threats of violence within the two weeks before admission and violent behavior during the first three days of hospitalization. Fifty-eight percent of the patients who had made threats before admission required seclusion for dangerous behavior in the hospital, and 32 percent of the patients who had made threats physically assaulted someone in the hospital. The association between preadmission threats and subsequent violence was especially strong among schizophrenic patients. Compared with schizophrenic patients who did not threaten others, those who had made preadmission threats were more likely to be physically and verbally assaultive and were more likely to require seclusion. Manic patients who had made threats were more likely than those who had not made threats to be verbally assaultive while in the hospital.     

A Machine Learning Approach to First Pass Reperfusion in Mechanical Thrombectomy: Prediction and Feature Analysis

IntroductionNovel machine learning (ML) methods are being investigated across medicine for their predictive capabilities while boasting increased adaptability and generalizability. In our study, we compare logistic regression with machine learning for feature importance analysis and prediction in first-pass reperfusion.MethodsWe retrospectively identified cases of ischemic stroke treated with mechanical thrombectomy (MT) at our institution from 2012–2018. Significant variables used in predictive modeling were demographic characteristics, medical history, admission NIHSS, and stroke characteristics. Outcome was binarized TICI on first pass (0-2a vs 2b-3). Shapley feature importance plots were used to identify variables that strongly affected outcomes.ResultsAccuracy for the Random Forest and SVM models were 67.1% compared to 65.8% for the logistic regression model. Brier score was lower for the Random Forest model (0.329 vs 0.342) indicating better predictive capability. Other supervised learning models performed worse than the logistic regression model, with accuracy of 56.2% for Naïve Bayes and 61.6% for XGBoost. Shapley plots for the Random Forest model showed use of aspiration, hyperlipidemia, hypertension, use of stent retriever, and time between symptom onset and catheterization as the top five predictors of first pass reperfusion.ConclusionUse of machine learning models, such as Random Forest, for the study of MT outcomes, is more accurate than logistic regression for our dataset, and identifies new factors that contribute to achieving first pass reperfusion. The benefits of machine learning, such as improved predictive capabilities, integration of new data, and generalizability, establish ML as the preferred model for studying outcomes in stroke.