Machine learning models using electronic health records can predict dementia risk among American Indian and Alaska Native elders, as revealed by a recent study led by the University of California, Irvine. This research identifies new predictors for dementia and sets a framework for other healthcare systems, especially those with limited resources, to follow.
The study, published in the Lancet Regional Health – Americas and supported by the National Institutes of Health, marks the first time machine learning has been harnessed to predict dementia risk in this historically understudied population, as defined by the U.S. Census Bureau.
“Public health researchers play a significant role in helping clinicians and policymakers make informed decisions about population health,” said Luohua Jiang, professor of epidemiology and biostatistics at UC Irvine. “If future studies confirm these results, our findings could prove valuable to the Indian Health Service and Tribal health clinicians in identifying high-risk individuals, facilitating timely interventions and improving care coordination.”
Analyzing data from the Indian Health Service’s National Data Warehouse, researchers used seven years of records from 2007 to 2013, involving 17,400 American Indian/Alaska Native adults aged 65 or older who were initially free from dementia. They discovered 611 new dementia diagnoses during the two-year prediction period.
The team evaluated four machine-learning algorithms based on performance and identified several novel predictors for dementia, including health service utilization. The research team included Kayleen Ports, Jiahui Dai, Kyle Conniff, Maria M. Corrada, Spero M. Manson, and Joan O’Connell.
Funding for the study was provided by the National Institutes of Health AIM-AHEAD and the National Institute on Aging.
UC Irvine continues to advance health research through its Brilliant Future campaign, which aims to raise significant support for the university and its programs, including the Joe C. Wen School of Population & Public Health.
###
