Machine learning to predict dementia for American Indian and Alaska native peoples - a retrospective cohort study

Aug 5, 2026

Upload by Miles

Dementia risk is rising in American Indian and Alaska Native communities, yet machine learning models using electronic health record data had not been developed or validated for this population. The study builds a two-year dementia risk prediction model using seven years of Indian Health Service and related EHR data, splitting into a five-year baseline and two-year prediction window. Four algorithms are trained and compared with performance assessed using AUC.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages13

Reading Time33 min

AccessFree to download

Background

Methods

Data and modeling

Algorithms and evaluation

Findings

Interpretation

Keywords

UC Irvine UC Irvine Previously Published Works Title Machine learning to predict dementia for American Indian and Alaska native peoples: a retrospective cohort study. Permalink [https://escholarship.org/uc/item/7991h7gs](https://escholarship.org/uc/item/7991h7gs) Authors Ports, Kayleen Dai, Jiahui Conniff, Kyle et al. Publication Date 2025-03-01 DOI 10.1016/j.lana.2025.101013 Peer reviewed [eScholarship.org](eScholarship.org) Powered by the California Digital Library University of California Articles Machine learning to predict dementia for American Indian and Alaska native peoples: a retrospective cohort study Kayleen Ports,a Jiahui Dai,a Kyle Conniff,b Maria M. Corrada,a,c Spero M. Manson,d Joan O’Connell,d and Luohua Jianga,∗ aDepartment of Epidemiology & Biostatistics, Joe C. Wen School of Population & Public Health, Susan and Henry Samueli College of Health Sciences, University of California, Irvine, 856 Health Sciences Quad, Irvine, CA 92697-7550, USA bDepartment of Statistics, Donald Bren School of Information and Computer Sciences, University of California, Irvine, Bren Hall 2019, Irvine, CA 92697-1250, USA cDepartment of Neurology, School of Medicine, University of California, Irvine, 1513 Hewitt Hall, 843 Health Sciences Rd, Irvine, CA 92697, USA dCenters for American Indian and Alaska Native Health, Colorado School of Public Health, University of Colorado Anschutz Medical Campus, 13055 East 17th Place, Aurora, CO 80045, USA Summary Background Dementia is an increasing concern among American Indian and Alaska Native (AI/AN) communities, yet machine learning models utilizing electronic health record (EHR) data have not been developed or validated for this population. This study aimed to develop a two-year dementia risk prediction model for AI/AN individuals actively using Indian Health Service (IHS) and Tribal health services. Methods Seven years of data were obtained from the IHS National Data Warehouse and related EHR databases and divided into a ﬁve-year baseline period (FY2007–2011) and a two-year dementia prediction period (FY2012–2013) . Four algorithms were assessed: logistic regression, Least Absolute Shrinkage and Selection Operator (LASSO), random forest, and eXtreme Gradient Boosting (XGBoost) . Dementia Risk Score (DRS)-based and extended models were developed for each algorithm, with performance evaluated by the area under the receiver operating characteristic curve (AUC) . The Lancet Regional Health -Americas 2025;43: 101013 Published Online xxx [https://doi.org/10](https://doi.org/10) . 1016/j.lana.2025 . 101013 Findings The study cohort included 17,398 AI/AN adults aged ≥ 65 years who were dementia-free at baseline, of whom 59.8% were female. Over the two-year follow-up, 611 individuals (3.5%) were diagnosed with incident dementia. Extended models for logistic regression, LASSO, and XGBoost performed comparably: AUCs (95% CI) of 0.83 (0.79, 0.86), 0.83 (0.79, 0.86), and 0.82 (0.79, 0.86) . These top-performing models shared 12 of the 15 highest-ranked predictors, with novel predictors including service utilization. Interpretation Machine learning algorithms utilizing EHR data can effectively predict two-year dementia risk among AI/AN older adults. These models could aid IHS and Tribal health clinicians in identifying high-risk individuals, facilitating timely interventions and improved care coordination. Funding NIH. Copyright © 2025 The Author(s) . Published by Elsevier Ltd. This is an open access article under the CC BY license ([http://creativecommons.org/licenses/by/4.0/](http://creativecommons.org/licenses/by/4.0/)). Keywords: American Indian and Alaska Native Peoples; Alzheimer’s disease and related dementia (ADRD); All-cause dementia; LASSO; Machine learning; Risk prediction; XGBoost Introduction The population of older adults aged 65 and older in the United States is rapidly growing.1 In line with the general population, the p

FAQs

What population was studied and how many participants were included?

The cohort included 17,398 American Indian and Alaska Native adults aged 65 and older who were dementia-free at baseline.

Which machine learning algorithms were evaluated for dementia risk prediction?

The study assessed logistic regression, LASSO, random forest, and XGBoost, then built DRS-based and extended models for each approach.

How was model performance evaluated?

Performance was evaluated using the area under the receiver operating characteristic curve (AUC), with reported 95% confidence intervals for top models.

Aug 5, 2026

Upload by Miles