Online Program

Return to main conference page

All Times EDT

Friday, June 5
Practice and Applications
Practice and Applications Posters, Part 2
Fri, Jun 5, 2:00 PM - 5:00 PM
TBD
 

WITHDRAWN: Basket Analysis Methods in Big Data: The Case of Diabetes (308492)

Tianyuan Guan, University of Cincinnati 
Hung-I Hsiao, University of Cincinnati 
Marepalli B Rao, University of Cincinnati 

Keywords: Diabetes, Basket Analysis, Trends in Hospital Admissions, Comorbidities

The focus is on hospital admissions due to diabetes. The source is the NIS (National Inpatient Sample) from 2012--2017. We will identity comorbidities that most commonly are associated with diabetes. We will use basket analysis methods to identify top association rules year by year. For example, in 2012, the subset of the NIS data connected to diabetes gives rise to 70,677 admissions and in 28,514 (40.3%) cases, the primary cause of admission was the long-term use of insulin. Data are missing from a few states in every year of the study. We will use a novel imputation method to fill in the missing values. Gender and racial differences will be examined critically. Hospital charged and length of stay will be documented year by year.