Hospital readmissions depend on numerous factors. Automated risk calculation using electronic health record(EHR) data could allow targeting care to prevent them. EHRs usually are incomplete with respect to data relevantto readmissions prediction. Lack of standard data representations in EHRs restricts generalizability of predictivemodels. We propose developing such models by first generating derived variables that characterize clinicalphenotype. This reduces the number of variables, reduces noise, introduces clinical knowledge into model building,and abstracts away the underlying data representation, thus facilitating use of standard data mining algorithms. Wecombined this preprocessing step with a random forest algorithm to compute risk for readmission within 30 daysfor patients in ten disease categories. Results were promising for encounters that our algorithm assigned very highor very low risk. Assigning patients to either of these two risk groups could be of value to patient care teams aiming
【 预 览 】
附件列表
Files
Size
Format
View
Leveraging Derived Data Elements in Data Analytic Models forUnderstanding and Predicting Hospital Readmissions