Predictive care: a protocol for a computational ethnographic approach to building fair models of inpatient violence in emergency psychiatry
Introduction Managing violence or aggression is an ongoing challenge in emergency psychiatry. Many patients identified as being at risk do not go on to become violent or aggressive. Efforts to automate the assessment of risk involve training machine learning (ML) models on data from electronic healt...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMJ Publishing Group
2023-04-01
|
| Series: | BMJ Open |
| Online Access: | https://bmjopen.bmj.com/content/13/4/e069255.full |
| _version_ | 1797838701733085184 |
|---|---|
| author | Gillian Strudwick Daniel Z Buchman Juveria Zaheer Katrina Hui Laura Sikstrom Marta M Maslej Zoe Findlay Sean L Hill |
| author_facet | Gillian Strudwick Daniel Z Buchman Juveria Zaheer Katrina Hui Laura Sikstrom Marta M Maslej Zoe Findlay Sean L Hill |
| author_sort | Gillian Strudwick |
| collection | DOAJ |
| description | Introduction Managing violence or aggression is an ongoing challenge in emergency psychiatry. Many patients identified as being at risk do not go on to become violent or aggressive. Efforts to automate the assessment of risk involve training machine learning (ML) models on data from electronic health records (EHRs) 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). Because risk assessment can impact psychiatric care (eg, via coercive measures, such as restraints), it is unclear which patients might be underserved or harmed by the application of ML.Methods and analysis We pilot a computational ethnography to study how the integration of ML into risk assessment might impact acute psychiatric care, with a focus on how EHR data is compiled and used to predict a risk of violence or aggression. Our objectives include: (1) evaluating an ML model trained on psychiatric EHRs to predict violent or aggressive incidents for intersectional bias; and (2) completing participant observation and qualitative interviews in an emergency psychiatric setting to explore how social, clinical and structural biases are encoded in the training data. Our overall aim is to study the impact of ML applications in acute psychiatry on marginalised and underserved patient groups.Ethics and dissemination The project was approved by the research ethics board at The Centre for Addiction and Mental Health (053/2021). Study findings will be presented in peer-reviewed journals, conferences and shared with service users and providers. |
| first_indexed | 2024-04-09T15:46:12Z |
| format | Article |
| id | doaj.art-571a226b489643d4af54bb16a27ccf7e |
| institution | Directory Open Access Journal |
| issn | 2044-6055 |
| language | English |
| last_indexed | 2024-04-09T15:46:12Z |
| publishDate | 2023-04-01 |
| publisher | BMJ Publishing Group |
| record_format | Article |
| series | BMJ Open |
| spelling | doaj.art-571a226b489643d4af54bb16a27ccf7e2023-04-27T01:00:06ZengBMJ Publishing GroupBMJ Open2044-60552023-04-0113410.1136/bmjopen-2022-069255Predictive care: a protocol for a computational ethnographic approach to building fair models of inpatient violence in emergency psychiatryGillian Strudwick0Daniel Z Buchman1Juveria Zaheer2Katrina Hui3Laura Sikstrom4Marta M Maslej5Zoe Findlay6Sean L Hill7Institute of Health Policy, Management and Evaluation, University of Toronto, Toronto, Ontario, CanadaCentre for Addiction and Mental Health, Toronto, Ontario, CanadaGerald Sheff and Shanitha Kachan Emergency Department, Centre for Addiction and Mental Health, Toronto, Ontario, CanadaCentre for Addiction and Mental Health, Toronto, Ontario, CanadaThe Krembil Centre for Neuroinformatics, The Campbell Family Mental Health Research Institute, Centre for Addiction and Mental Health, Toronto, Ontario, CanadaCentre for Addiction and Mental Health, Toronto, Ontario, CanadaDepartment of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, CanadaCentre for Addiction and Mental Health, Toronto, Ontario, CanadaIntroduction Managing violence or aggression is an ongoing challenge in emergency psychiatry. Many patients identified as being at risk do not go on to become violent or aggressive. Efforts to automate the assessment of risk involve training machine learning (ML) models on data from electronic health records (EHRs) 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). Because risk assessment can impact psychiatric care (eg, via coercive measures, such as restraints), it is unclear which patients might be underserved or harmed by the application of ML.Methods and analysis We pilot a computational ethnography to study how the integration of ML into risk assessment might impact acute psychiatric care, with a focus on how EHR data is compiled and used to predict a risk of violence or aggression. Our objectives include: (1) evaluating an ML model trained on psychiatric EHRs to predict violent or aggressive incidents for intersectional bias; and (2) completing participant observation and qualitative interviews in an emergency psychiatric setting to explore how social, clinical and structural biases are encoded in the training data. Our overall aim is to study the impact of ML applications in acute psychiatry on marginalised and underserved patient groups.Ethics and dissemination The project was approved by the research ethics board at The Centre for Addiction and Mental Health (053/2021). Study findings will be presented in peer-reviewed journals, conferences and shared with service users and providers.https://bmjopen.bmj.com/content/13/4/e069255.full |
| spellingShingle | Gillian Strudwick Daniel Z Buchman Juveria Zaheer Katrina Hui Laura Sikstrom Marta M Maslej Zoe Findlay Sean L Hill Predictive care: a protocol for a computational ethnographic approach to building fair models of inpatient violence in emergency psychiatry BMJ Open |
| title | Predictive care: a protocol for a computational ethnographic approach to building fair models of inpatient violence in emergency psychiatry |
| title_full | Predictive care: a protocol for a computational ethnographic approach to building fair models of inpatient violence in emergency psychiatry |
| title_fullStr | Predictive care: a protocol for a computational ethnographic approach to building fair models of inpatient violence in emergency psychiatry |
| title_full_unstemmed | Predictive care: a protocol for a computational ethnographic approach to building fair models of inpatient violence in emergency psychiatry |
| title_short | Predictive care: a protocol for a computational ethnographic approach to building fair models of inpatient violence in emergency psychiatry |
| title_sort | predictive care a protocol for a computational ethnographic approach to building fair models of inpatient violence in emergency psychiatry |
| url | https://bmjopen.bmj.com/content/13/4/e069255.full |
| work_keys_str_mv | AT gillianstrudwick predictivecareaprotocolforacomputationalethnographicapproachtobuildingfairmodelsofinpatientviolenceinemergencypsychiatry AT danielzbuchman predictivecareaprotocolforacomputationalethnographicapproachtobuildingfairmodelsofinpatientviolenceinemergencypsychiatry AT juveriazaheer predictivecareaprotocolforacomputationalethnographicapproachtobuildingfairmodelsofinpatientviolenceinemergencypsychiatry AT katrinahui predictivecareaprotocolforacomputationalethnographicapproachtobuildingfairmodelsofinpatientviolenceinemergencypsychiatry AT laurasikstrom predictivecareaprotocolforacomputationalethnographicapproachtobuildingfairmodelsofinpatientviolenceinemergencypsychiatry AT martammaslej predictivecareaprotocolforacomputationalethnographicapproachtobuildingfairmodelsofinpatientviolenceinemergencypsychiatry AT zoefindlay predictivecareaprotocolforacomputationalethnographicapproachtobuildingfairmodelsofinpatientviolenceinemergencypsychiatry AT seanlhill predictivecareaprotocolforacomputationalethnographicapproachtobuildingfairmodelsofinpatientviolenceinemergencypsychiatry |