Violence in inpatients with schizophrenia: a prospective study

Area of Science:

• Psychiatry
• Clinical Psychology
• Forensic Psychiatry

Background:

• Accurate violence risk assessment in psychiatric inpatients is essential but challenging.
• Previous studies have yielded mixed results regarding the predictability of violence.
• Schizophrenia and schizoaffective disorder patients present unique challenges in risk assessment.

Purpose of the Study:

• To prospectively evaluate variables for predicting violence in inpatients diagnosed with schizophrenia or schizoaffective disorder.
• To identify key clinical and demographic factors associated with violent incidents.
• To develop a predictive model for violence in this patient population.

Main Methods:

• Prospective evaluation of 63 inpatients with schizophrenia or schizoaffective disorder.
• Violence incidents rated using the Overt Aggression Scale.
• Logistic regression analysis to identify predictors of violence, including sociodemographic data, clinical history, neurological signs, substance abuse, EEG abnormalities, and PANSS scores.

Main Results:

• No significant differences were found in sociodemographic variables, clinical history, neurological signs, substance abuse, or EEG abnormalities between violent and nonviolent groups.
• Violent patients exhibited significantly more positive symptoms (Positive and Negative Syndrome Scale - PANSS), higher general psychopathology scores, and poorer insight.
• A logistic regression model including insight, PANSS general psychopathology, and recent violence correctly classified 84.13% of patients.

Conclusions:

• Clinical variables, particularly insight into symptoms, are more predictive of violence than sociodemographic factors in psychiatric inpatients.
• Insight into psychotic symptoms emerged as a novel and significant predictor of violence in schizophrenia inpatients.
• These findings have practical implications for therapeutic interventions aimed at reducing violence risk.