Finding Cost-Effective Solutions to Health Care Problems

• Activity Number: 251
• Type: Contributed
• Date/Time: Monday, August 4, 2014 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistical Learning and Data Mining
• Author(s): Christian Lemieux*+ and Billie Anderson
• Companies: and Bryant University
• Keywords: BMI ; healthcare ; cost-effective ; statistical analysis ; data mining ; prevention

Abstract

This poster will present a statistical analysis utilizing data from the 2013 SAS Data Mining Shootout Competition. The competition examined 6 different programs in order to reduce health problems and in turn reduce healthcare costs by determining the most cost effective healthcare program to implement. This analysis will focus on implementing the most cost effective healthcare program for smoking cessation and reducing the body mass index (BMI) of individuals with excessive BMI; those individuals who are considered to be obese. The data used for this analysis is from the state of New Hampshire and covers the time period from 2006-2011. The data includes demographic, socio-economic, and healthcare cost data. The data will be used to analyze which preventative programs will be most cost effective over the next 9 years (2012-2020). A predictive model will be developed that will use demographic and socio-economic data to predict demographics and forecast the socio-economic data (household income, percent poverty level, and education measures). The healthcare cost data will be used build a model that will predict healthcare costs for the indicated diseases by age, gender, and county.