Mental Health World Conference 2026

Abstract

Abstract: Short description of the presentation topic about 250 to 600 words.

Objective - To identify the latent classes of change trajectories of medication deviation among patients with mental disorders and to explore their influencing factors, thereby providing a theoretical basis for the development of targeted clinical interventions.

MethodsA convenience sampling method was employed to enroll 285 patients with mental disorders who were hospitalized in a Class-A tertiary general hospital in Chongqing from June 2025 to June 2026. Baseline data were collected using a general information questionnaire, the Frailty Screening Scale, and the Medication Literacy Questionnaire. Medication deviation was assessed using the Medication Deviation Assessment Tool at four follow-up time points: 1 week (T1), 1 month (T2), 3 months (T3), and 6 months (T4) postdischarge. The latent class growth model (LCGM) was utilized to identify distinct trajectories of medication deviation, and multinomial logistic regression was performed to determine the influencing factors associated with these trajectory classes.

A total of 253 patients were ultimately included in the analysis. The incidence rates of medication deviation at T1, T2, T3, and T4 were 29.6, 56.2, 68.6, and 76.8, respectively. Three distinct latent trajectory classes were identified: the rapid deterioration group (34.8), the stable maintenance group (26.0), and the persistent deviation group (39.2). Multinomial logistic regression analysis revealed that medication literacy, frailty status, adverse drug reactions, educational level, duration of illness, number of medications taken, and caregiver support were significant influencing factors for the trajectory class membership (all P  0.05).

ConclusionMedication deviation among patients with mental disorders exhibits a general upward trend over time, with significant inter-individual heterogeneity in developmental trajectories. Healthcare providers can implement tailored, stage-specific interventions