Researchers have developed a new algorithm that uses patient data to make predictions about an individual’s future health conditions, according to a paper published in the Annals of Applied Statistics, Popular Sciencereports.
How It Works
Massachusetts Institute of Technology and Columbia University researchers took data from a multiyear clinical drug trial involving tens of thousands of patients age 40 and older. Patient information includes gender, ethnicity, medical histories and prescriptions (Boyle, Popular Science, 6/5).
They then used the Hierarchical Association Rule Model to predict future medical conditions based on combined data on various conditions from the larger patient pool.
Tyler McCormick — assistant professor of statistics and sociology at the University of Washington — said the statistical model “provides physicians with insights on what might be coming next for a patient, based on experiences of other patients.” For example, patients who already have dyspepsia and epigastric pain could be more likely to develop heartburn, according to McCormick.
Access to Tool
McCormick said he intends to make the tool available to both doctors and patients. He said, “We hope that this model will provide a more patient-centered approach to medical care and to improve patient experiences” (Hall, FierceHealthIT, 6/6).