# Understanding the Information Needs and Practices of Human Supporters of an Online Mental Health Intervention to Inform Machine Learning Applications

**Authors:** Anja Thieme  
**arXiv:** [2111.06667](https://arxiv.org/abs/2111.06667)  
**Date:** 2021-11-12  
**Summary page:** https://www.summarizepaper.com/en/arxiv-id/2111.06667v1/

## Key points

- Role of human supporters or coaches in internet-delivered Cognitive Behavioral Therapy (iCBT) for depression and anxiety
- Advancements in AI and machine learning (ML) to enhance work practices of iCBT supporters
- Interviews with 15 iCBT supporters to understand their work practices and information needs
- Six themes summarizing strategies and challenges faced by iCBT supporters in delivering personalized feedback
- Concrete opportunities for ML methods to address challenges and support iCBT treatment
- Potential social, emotional, and pragmatic implications of introducing machine-generated data insights within client review practices
- Aim to inform the development of implementable ML applications in the context of iCBT for depression and anxiety, focusing on maximizing the effectiveness of human support.

## Summary

This research paper explores the role of human supporters or coaches in internet-delivered Cognitive Behavioral Therapy (iCBT) for depression and anxiety. It investigates how advancements in AI and machine learning (ML) can enhance the work practices of iCBT supporters by providing useful data insights. The study involves interviews with 15 iCBT supporters to understand their existing work practices and information needs. The analysis identifies six themes that summarize the strategies and challenges faced by iCBT supporters in delivering personalized feedback to mental health clients. Based on these findings, the paper presents concrete opportunities for ML methods to address these challenges and support iCBT treatment. The paper also discusses potential social, emotional, and pragmatic implications of introducing machine-generated data insights within supporter-led client review practices. Overall, this research aims to inform the development of implementable ML applications in the context of iCBT for depression and anxiety, with a focus on maximizing the effectiveness of human support in therapy interventions.