Using predictive analytics to tailor care for patients with newly diagnosed type 2 diabetes

In a population of adults newly diagnosed with type 2 diabetes (T2D) from 2010-2015 and followed for 5 years this study will use machine learning-based predictive analytic methods to develop and validate predictive models using EHR-derived patient data available at the time of T2D diagnosis to identify patients at increased risk for suboptimal 1-year and 5-year glycemic trends. Secondary predictive models will be developed to identify individuals at increased risk for other adverse T2D-outcomes, including microalbuminuria, microvascular disease, and macrovascular disease. The study will also develop and validate predictive models that include data from the first year following T2D diagnosis to identify patients at increased risk for suboptimal 5-year glycemic trends.

0