Predictive Care: An institutional ethnography of risk assessment in acute psychiatric care
(2021-2023)

Project Leads: Laura Sikstrom, Marta Maslej

With: Katrina Hui, Daniel Buchman, Juveria Zaheer, Zoe Findlay, Gillian Strudwick

Funding: SSHRC IDG & Google Award for Inclusion Research

Project Summary. Managing violence or aggression is an ongoing challenge in emergency psychiatry. Efforts to automate the assessment of risk involve training machine learning models on electronic health records to predict these behaviours; however, no studies to date have examined which patient groups may be over-represented in false positive predictions, despite evidence of social and clinical biases that may lead to higher perceptions of risk in patients defined by intersecting features (eg, race, gender). This project pilots a computational ethnography to study how the integration of machine learning into risk assessment might impact acute psychiatric care, with a focus on how electronic health records are compiled and used to predict a risk of violence or aggression.