Maine is piloting a project to use predictive modeling software to comb through nearly all of the state residents' electronic health records to determine which patients are most at risk of visiting the emergency room, being admitted to the hospital, suffering a stroke or heart attack or developing type 2 diabetes, according to an article in The Bangor Daily News.
The program, spearheaded in part by HealthInfoNet, the state's health information exchange (HIE), applies the predictive analytics on the records in its database. Patients identified as at risk then are contacted and offered support, education and resources. It has already proven to be effective in reducing patient readmissions. Unlike other predictive modeling software, which uses insurance claims data that can be 90 days old, the data used is at most 24 hours old. HealthInfoNet plans on marketing the tool to HIEs in other states. Article