The researchers extracted 823,627 patient EHRs from the Maine Health Information Exchange network to train their machine learning algorithm. The algorithm assigns individual patients to one of five risk categories, based on their likelihood of being diagnosed with incident hypertension within one year.
Researchers found the model recognized Type 2 diabetes, lipid disorders, mental illnesses and socioeconomic factors as associated with incident hypertension. The “very high risk population” was comprised mainly of patients over the age of 50 with multiple chronic conditions.
To assess the prediction model, the researchers extracted an additional 680,810 patient EHRs.
“With statewide EHR datasets, our study prospectively validated an accurate 1-year risk prediction model for incident essential hypertension,” the study authors concluded. They noted the model has been deployed in Maine to provide “implications in interventions for hypertension and related diseases and hopefully [enhance] hypertension care.”
More articles on data analytics & precision medicine:
Intermountain’s precision genomics arm begins tests to identify hereditary cancer genes
IBM, U of Melbourne researchers use data to predict epilepsy patients’ likelihood of having a seizure
Atrius Health taps Linguamatics Health to process unstructured patient data
