Effectiveness of Long-Acting Injectable Antipsychotics in Managing Violence among Patients with Co-Occurring Substance-Induced Psychotic Disorder and Serious Mental Illness with High Risk to Violence (SMI-V)

Abstract

Objective: To evaluate the effectiveness and safety of long-acting injectable antipsychotics (LAIs) compared to oral antipsychotics in controlling violent behavior, reducing relapse, and improving global functioning among patients with substance-induced psychotic disorder (SIPD) who are at high risk to violence (SMI-V).

Methods: This quasi-experimental study employed a non-randomized two-group pre-posttest design with a 6-month follow-up. Participants were recruited using purposive sampling from patients aged 15 – 65 years diagnosed with substance-induced psychotic disorder (SIPD) who met the criteria for serious mental illness with high risk of violence (SMI-V) at Trakanphuetphon Hospital between October 2025 and May 2026. Of the 56 patients initially enrolled, 50 completed the study, all of whom were male. Participants were divided into an LAI group (n = 25) and a control group (non-LAIs/oral medication, n = 25). Assessment tools included violent incidents, psychiatric relapse requiring emergency department visits or hospitalization (ER/IPD), and the global assessment of functioning (GAF) scale. Data were analyzed using paired t-tests, independent t-test, ANCOVA, and multivariable logistic regression.

Results: After 6 months, the LAI group demonstrated a significant reduction in mean violent incidents (from 3.36 to 0.08, p < 0.001). At the 6-month follow-up, the mean violent incidents in the LAI group (0.08) were significantly lower than in the control group (0.44) (p = 0.014). The LAI group also achieved significantly higher GAF scores than the control group (71.28 vs. 46.68, p < 0.001). Furthermore, the LAI group was 19.24 times more likely to successfully control violent behavior (95% CI: 1.65 – 224.35, p = 0.018) and 17.35 times less likely to be readmitted to the hospital or visit the emergency department (95% CI: 1.84 – 162.92, p = 0.012). Regarding safety, no significant difference in the incidence of serious adverse events was observed between the two groups (p = 0.312).

Conclusion: Patients with SIPD who were at high risk of violence (SMI-V) and received LAIs demonstrated a greater reduction in violent behaviors, lower rates of psychiatric relapse requiring rehospitalization, and greater improvement in overall functioning (GAF) compared with those receiving oral antipsychotics after 6 months of follow-up, with no significant difference in safety outcomes between treatment groups. In real-world settings characterized by widespread substance use, ongoing substance consumption among patients, and limitations in specialized mental health personnel, LAIs may represent a valuable therapeutic option for this high-risk population. These findings support consideration of expanding access to long-acting injectable antipsychotics to reduce harm and improve the safety of patients, families, communities, and society, while providing preliminary real-world evidence to inform mental health policy development and future violence prevention strategies.