This page was translated automatically from Turkish. Read the original

This page is an AI-assisted Turkish summary of the study presented at the 26th World Congress of Psychiatry. The study belongs to the presenting authors; please refer to the original source when citing it.

SO137

From innovation to implementation: patient and staff perspectives, readiness, and needs for artificial intelligence in mental health

FROM INNOVATION TO IMPLEMENTATION: PATIENT AND STAFF PERSPECTIVES, READINESS, AND NEEDS FOR AI IN MENTAL HEALTH

Both patients and staff are interested in the use of artificial intelligence in mental health and willing to receive training, but staff report medicolegal liability, transparency, and job security as their primary concerns, while patients prioritize data privacy and quality of care.

The study aimed to examine patient and staff perceptions, readiness, and training needs regarding the integration of artificial intelligence (AI) applications into mental health services. A mixed-methods design was used at a large mental health hospital in Canada; 5 staff focus groups, 6 patient focus groups, and 4 individual patient interviews were conducted with 21 staff members and 20 patients, and surveys were also completed via REDCap. Qualitative data were analyzed using thematic analysis in NVivo 14, and quantitative data were analyzed using descriptive statistics in SPSS (v13). The focus groups yielded 52 codes, which were organized into five themes and associated subthemes. According to the survey results, 90.5% of staff and 42.1% of patients reported being very or somewhat familiar with AI in mental health; 61.9% of staff and 79% of patients believed that AI could support or improve clinical or administrative tasks. Women accounted for 62% of the staff sample, and men accounted for 74% of the patient sample; 63% of patients were from the forensic inpatient program and 37% from the regional inpatient program. Across the themes, staff identified medicolegal policies, transparency and trust, and job security as priorities, while patients prioritized personal data and privacy strategies and quality of care. The authors recommend focusing on implementation strategies beyond model development and performance, embedding transparency and explainability, establishing robust privacy and data governance policies, designing tools with an 'AI first, human second' rather than an 'AI first, then human' approach, and co-designing with stakeholders.

All WCP 2026 summaries