Developing a predictive model for factors related to risk of aggression in psychiatric inpatients using physiological and clinical data

Unpredictable, aggressive behavior in psychiatric inpatients remains a challenge in mental health, emphasizing the negative impact on both patients and staff. With estimates suggesting a significant percentage of patients exhibiting aggression during psychiatric stays, the study employs big data analysis on an actual clinical data set extracted from patients’ medical records to develop a predictive model for clinical aggression risk factors. Retrospective analysis covers variables such as heart rate, blood pressure, age, incidents, medication, hospitalization, suicide risk, education, type of incident, and gender. Statistical analyses, including t-test, stepwise regression, and logistic regression, reveal a significant correlation between a history of aggression and a lower resting heart rate. The final model identifies predictors such as systolic and diastolic blood pressure, medication refusal, and gender. The study highlights the potential of big data in enhancing medical insights and recommends future exploration of streaming and temporal data for more precise disease prevention.

Name thesis:degree_name

Master of Health Sciences (MHSc)

Discipline thesis:degree_discipline

Health Informatics

Grantor

University of Ontario Institute of Technology

Year dc:date.issued

2025

Author dc:creator

• Farsi, Leila

Advisor dc:contributor.advisor

• McGregor, Carolyn

Language dc:language.iso

en

Handle dc:identifier.uri

https://hdl.handle.net/10155/1929

OAI identifier oai:identifier

oai:ontariotechu.scholaris.ca:10155/1929