The violence risk assessment of the community patients with severe mental illness and Logistic regression analysis on associated factors

Abstract

Objective To assess the violence risk levels of the community patients with severe mental illness and Analyze the associated factors. Methods A working group composed of psychiatrists, community police, FIMITIC staff, community physicians, neighborhood or village committee members was established. 501 community patients with severe mental illness were evaluated with Violence Risk Assessment Scale by working group.While the data of the general demographic, medical history and violence history was investigated. Logistic regression analysis was used to explore the factors associated with violence. Results In 501 patients, risk levels of grade 0: 124 cases (24.8%), grade 1: 165 cases (32.9%), grade 2: 118 cases (23.5%), grade 3: 59 cases (11.8%), grade 4: 28 cases (5.6%) and grade 5: 7 cases (1.4%). Logistic regression analysis showed four significant risk factors, which were schizophrenia (OR 4.139), no taking medicine(OR 6.883), unstable condition (OR 234.534) the number of previous violence(OR 3.457) (P<0.05). Logistic regression equation: Logit (severe violence) =-8.511+ 1.420(schizophrenia) + 1.929(no taking medicine) + 5.458(unstable condition) + 1.241(the number of previous violence). Conclusion The results of violence risk assessment and risk factors have directive significance for management of community patients with severe mental illness. Key words: Severe mental illness; Violence; Risk assessment; Logistic regression analysis