Scholarlink

Background <p>Suicide represents a major public health concern. Suicide risk assessment, considered one of the most complex clinical situations in psychiatry, is crucial for prevention. However, mental health clinicians report low self-efficacy and willingness to conduct these assessments. Current training processes are predominantly theoretical, limiting opportunities to practice asking challenging questions. Furthermore, there are global challenges in achieving international standardization of training while maintaining linguistic and cultural appropriateness.</p> Methods <p>The study evaluated the SEE THE PAIN system, a Generative AI (GenAI) simulator providing skills-based suicide risk assessment training across 20 languages. Twenty-seven psychiatrists attending the European Psychiatric Association Summer School participated. Participants conducted a risk assessment of a virtual character via the simulator and completed measures of self-efficacy and willingness to treat at pre- and post-intervention points.</p> Results <p>Analysis demonstrated significant improvements in self-efficacy (from M = 6.7/10 to M = 7.6/10, <i>p</i> = .001) and willingness to treat (from M = 6.5/10 to M = 7.59/10, <i>p</i> &lt; .001) following the intervention. User experience ratings were notably positive, with mean scores of 7.19/10 (SD = 1.73) for future utility and 7.81/10 (SD = 1.69) for feedback quality. Improvements in self-efficacy showed a significant negative correlation with years of clinical experience (<i>r</i> = -.54, <i>p</i> = .01), suggesting particular effectiveness for early-career practitioners.</p> Conclusions <p>GenAI based multilingual simulation training effectively addresses the complex challenges in suicide risk assessment training by enhancing both self-efficacy and willingness to conduct assessments. The system bridges the gap between theoretical knowledge and practical skills while providing standardized, culturally adapted training across language barriers. These findings suggest promising potential for improving global psychiatric training standards in this critical clinical skill.</p> Clinical trial number <p>Not applicable.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

The effectiveness of multilingual AI-based simulator for suicide risk assessment training in improving self-efficacy among young psychiatrists: a pilot study across twenty languages

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年1月6日

查看全文
  • Zohar Elyoseph,
  • Yossi Levi-Belz,
  • Inbar Levkovich,
  • Yuval Haber,
  • Carla Maria Gramaglia,
  • Jorge López Castroman,
  • Hanon Cecile,
  • Emilie Olie

摘要

Background

Suicide represents a major public health concern. Suicide risk assessment, considered one of the most complex clinical situations in psychiatry, is crucial for prevention. However, mental health clinicians report low self-efficacy and willingness to conduct these assessments. Current training processes are predominantly theoretical, limiting opportunities to practice asking challenging questions. Furthermore, there are global challenges in achieving international standardization of training while maintaining linguistic and cultural appropriateness.

Methods

The study evaluated the SEE THE PAIN system, a Generative AI (GenAI) simulator providing skills-based suicide risk assessment training across 20 languages. Twenty-seven psychiatrists attending the European Psychiatric Association Summer School participated. Participants conducted a risk assessment of a virtual character via the simulator and completed measures of self-efficacy and willingness to treat at pre- and post-intervention points.

Results

Analysis demonstrated significant improvements in self-efficacy (from M = 6.7/10 to M = 7.6/10, p = .001) and willingness to treat (from M = 6.5/10 to M = 7.59/10, p < .001) following the intervention. User experience ratings were notably positive, with mean scores of 7.19/10 (SD = 1.73) for future utility and 7.81/10 (SD = 1.69) for feedback quality. Improvements in self-efficacy showed a significant negative correlation with years of clinical experience (r = -.54, p = .01), suggesting particular effectiveness for early-career practitioners.

Conclusions

GenAI based multilingual simulation training effectively addresses the complex challenges in suicide risk assessment training by enhancing both self-efficacy and willingness to conduct assessments. The system bridges the gap between theoretical knowledge and practical skills while providing standardized, culturally adapted training across language barriers. These findings suggest promising potential for improving global psychiatric training standards in this critical clinical skill.

Clinical trial number

Not applicable.