Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: May 13, 2026

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack

Published on: May 15, 2020

Combining Machine Learning Models and Screening to Enhance Suicide Risk Identification for American Indian Patients:

Novalene Alsenay Goklish¹, Emily E Haroz¹, Rohan R Dayal¹

¹Department of International Health, Center for Indigenous Health, Johns Hopkins Bloomberg School of Public Health, 415 N. Washington St, Baltimore, MD, 21205, United States, 1 410-955-0011.

Journal of Medical Internet Research

|May 11, 2026

Summary

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

A Guide to Constructing Indigenous Statistical Spaces for Prevention Science Research.

Prevention science : the official journal of the Society for Prevention Research·2026

Same author

Addressing Health Disparities through Community Engagement in Artificial Intelligence-Driven Prevention Science.

Prevention science : the official journal of the Society for Prevention Research·2026

Same author

Self-harm and suicide prevention in humanitarian and fragile contexts: A systematic scoping review.

Global mental health (Cambridge, England)·2025

Same author

Impact of long-term systolic blood pressure on cognitive disorder risk: Across different age groups and antihypertensive medication timing.

Journal of Alzheimer's disease : JAD·2025

Same author

Exploring the Perspectives of Pediatric Health Care Providers, Youth Patients, and Caregivers on Machine Learning Suicide Risk Classification: Mixed Methods Study.

Journal of medical Internet research·2025

Same author

AI-Y: An AI Checklist for Population Ethics Across the Global Context.

Current epidemiology reports·2025

This study found that parallel testing using a machine learning (ML) suicide risk model with the Ask Suicide-Screening Questions (ASQ) offers the highest sensitivity for identifying at-risk American Indian patients. This approach acts as a crucial safety net in emergency departments.

Area of Science:

Public Health
Health Informatics
Mental Health Research

Background:

American Indian and Alaska Native communities face disproportionately high suicide rates.
Machine learning (ML) models using electronic health records show promise for suicide risk identification.
Optimal integration of ML models with current screening practices requires further investigation.

Purpose of the Study:

To compare parallel and serial testing strategies combining an ML suicide risk model and the Ask Suicide-Screening Questions (ASQ).
To evaluate these combined strategies against using the ASQ alone for suicide risk identification.
To assess the effectiveness in a real-world clinical setting.

Main Methods:

Retrospective analysis of electronic health record data from adult emergency department visits.

Keywords:

American Indian and Alaska Native electronic health records machine learning predictive modeling screening suicide prevention suicide risk identification

More Related Videos

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

Published on: January 11, 2020

Related Experiment Videos

Last Updated: May 13, 2026

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack

Implementation of a Real-Time Psychosis Risk Detection and Alerting System Based on Electronic Health Records using CogStack

Published on: May 15, 2020

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment

Published on: January 11, 2020

Inclusion of visits at an Indian Health Service facility between October 2019 and October 2021.

Comparison of ML model thresholds (95th and 75th percentiles) and ASQ results using sensitivity, specificity, and predictive values.

Main Results:

The ML medium-risk threshold alone showed higher sensitivity (0.782) and negative predictive value (0.999) compared to ASQ alone.
Serial testing with the ML high-risk threshold and ASQ maximized positive predictive ability (PPV: 0.050) but reduced sensitivity.
Parallel testing with the ML medium-risk threshold achieved the highest sensitivity (0.795) without missing cases.

Conclusions:

Parallel testing with ML models and ASQ serves as a valuable clinical safety net for identifying at-risk individuals missed by standard screening.
Serial testing offers high predictive accuracy but may be less feasible in practice.
Integrating ML for suicide prevention necessitates balancing statistical performance with real-world clinical workflows.