Hospital Readmissions

The Problem

Existing studies of hospital readmissions typically focus on specific diagnoses, age groups, discharge dispositions, payers, or hospitals, and often use small samples. It is not clear how predictive models generated from such studies generalize across diseases, hospitals, or time frames. The goal of this project is to construct a generic model of readmission risk which can be applied to the majority of inpatient admissions, and validate the model’s ability to extrapolate across hospital sites and time frames. To better convey the nature of the model, a computer application will be built for demonstration purposes.


  1. Refine readmissions risk statistical models based on latest research findings
  2. Embellish the Readmission App code to be better aligned internally
  3. Update and examine the fit of the readmission prediction model with the latest input factors
  4. Conduct a comparative analysis of the readmission prediction model against another hospital site
  5. Disseminate the readmission research findings to involved hospitals and other appropriate organizations


Use logistics regression to predict 30-day readmission rates for patients. Construct an optimization model to establish patient risk classes and recommended discharge interventions for each patient class based on objective-based intervention characteristics, such as efficacy and cost. Build a web-based app for showcasing a Discharge Intervention Decision Support System that can convey effective and appropriate use of prediction and intervention recommendation models.

Potential Impact

To help mitigate partnering hospitals’ readmission risks by implementing patient-appropriate care transition strategies in order to improve the hospital’s quality of care and to have them avoid costly penalties.

Purdue University, 610 Purdue Mall, West Lafayette, IN 47907, (765) 494-4600

© 2015 Purdue University | An equal access/equal opportunity university | Copyright Complaints | Maintained by Regenstrief Center for Healthcare Engineering

Trouble with this page? Disability-related accessibility issue? Please contact Regenstrief Center for Healthcare Engineering at