Analyzing and Predicting Hospital Readmission Rates

For the 2014 Data Geek Challenge, I decided to build an analysis to look at one of the most costly issues resulting from a surgical procedure: complications requiring re-admission to the hospital.  In my dataset, I track a set of patients that were successfully discharged from Smalltown Hospital after a routine medical procedure and determine whether they were re-admitted to the hospital within 30 days.  In order to share the results of this analysis with others, I have created a SAP Lumira Cloud Infographic.

The Problem
With the introduction of the Affordable Care Act, more emphasis is being placed on the ability to understand and reduce re-admissions because of the financial incentives offered to hospitals with low or improving re-admission rates.  Smalltown Hospital has identified one particular surgical procedure as a high source of overall re-admissions.  Smalltown contacted me to request an analysis of their patient data to determine if there are any general risk factors that can be used to identify at-risk patients before they are discharged.

Analysis Results
After examining overall patient distributions across risk factors and noting there is sufficient data to analyze hypertension, depression, smoking, BMI, and gender as risk factors, I proceeded to examine re-admission rates by factor, noting that the re-admission rate is much higher for patients with hypertension and much lower for patients reporting healthy behaviors, as shown in the pie charts below.

Patients with both smoking and depression indicators have a re-admission rate almost 7 times higher than those with neither indicator.

To download full PDF and Continue Reading…

 About Hillary Bliss
Hillary Bliss is a Senior ETL Consultant at Decision First Technologies, and specializes in data  warehouse design, ETL development, statistical analysis, and predictive modeling. She works  with clients and vendors to integrate business analysis and predictive modeling solutions into  the organizational data warehouse and business intelligence environments based on their  specific operational and strategic business needs. She has a master’s degree in statistics and an  MBA from Georgia Tech.