Making sense of violence risk predictions using clinical notes

TL;DR

This paper investigates methods to interpret violence risk prediction models using clinical notes, emphasizing understanding data and model behavior to enhance generalizability and trustworthiness in psychiatric settings.

Contribution

It introduces two interpretability approaches—topic models with random forests and evaluation metric analysis—to improve understanding of violence risk classifiers from clinical notes.

Findings

Enhanced understanding of classifier data and behavior

Framework for assessing model generalizability

Insights into improving violence risk prediction models

Abstract

Violence risk assessment in psychiatric institutions enables interventions to avoid violence incidents. Clinical notes written by practitioners and available in electronic health records (EHR) are valuable resources that are seldom used to their full potential. Previous studies have attempted to assess violence risk in psychiatric patients using such notes, with acceptable performance. However, they do not explain why classification works and how it can be improved. We explore two methods to better understand the quality of a classifier in the context of clinical note analysis: random forests using topic models, and choice of evaluation metric. These methods allow us to understand both our data and our methodology more profoundly, setting up the groundwork to work on improved models that build upon this understanding. This is particularly important when it comes to the generalizability of evaluated classifiers to new data, a trustworthiness problem that is of great interest due to the increased availability of new data in electronic format.