FAIIR: Building Toward A Conversational AI Agent Assistant for Youth Mental Health Service Provision

TL;DR

FAIIR is an AI tool designed to assist crisis responders in youth mental health support by accurately identifying issues in conversations, reducing their workload, and improving response quality through advanced transformer models trained on extensive data.

Contribution

This work introduces FAIIR, a novel transformer-based system that enhances issue identification in youth mental health crisis conversations with high accuracy and human-in-the-loop refinement.

Findings

FAIIR achieved 94% AUCROC and 64% F1-score on retrospective data.

CRs' responses agreed with FAIIR predictions over 90% of the time.

The system maintained robustness with less than 2% performance drop during testing.

Abstract

The world's healthcare systems and mental health agencies face both a growing demand for youth mental health services, alongside a simultaneous challenge of limited resources. Here, we focus on frontline crisis support, where Crisis Responders (CRs) engage in conversations for youth mental health support and assign an issue tag to each conversation. In this study, we develop FAIIR (Frontline Assistant: Issue Identification and Recommendation), an advanced tool leveraging an ensemble of domain-adapted and fine-tuned transformer models trained on a large conversational dataset comprising 780,000 conversations. The primary aim is to reduce the cognitive burden on CRs, enhance the accuracy of issue identification, and streamline post-conversation administrative tasks. We evaluate FAIIR on both retrospective and prospective conversations, emphasizing human-in-the-loop design with active CR engagement for model refinement, consensus-building, and overall assessment. Our results indicate that FAIIR achieves an average AUCROC of 94%, a sample average F1-score of 64%, and a sample average recall score of 81% on the retrospective test set. We also demonstrate the robustness and generalizability of the FAIIR tool during the silent testing phase, with less than a 2% drop in all performance metrics. Notably, CRs' responses exhibited an overall agreement of 90.9% with FAIIR's predictions. Furthermore, expert agreement with FAIIR surpassed their agreement with the original labels. To conclude, our findings indicate that assisting with the identification of issues of relevance helps reduce the burden on CRs, ensuring that appropriate resources can be provided and that active rescues and mandatory reporting can take place in critical situations requiring immediate de-escalation.