Online Program Home
My Program

Abstract Details

Activity Number: 626 - Health Policy and Real World Evidence with Administrative Data and Electronic Health Records
Type: Contributed
Date/Time: Thursday, August 2, 2018 : 8:30 AM to 10:20 AM
Sponsor: Health Policy Statistics Section
Abstract #328899 Presentation
Title: Quantile Regression Application in Cost Analysis of Peripheral Artery Intervention in U.S. Healthcare Setting
Author(s): Haekyung Jeon-Slaughter* and Shirling Tsai and Bala Ramanan and Abigail Wheeler and Houman Khalili and Subhash Banerjee and Ishita Tejani
Companies: University of Texas Southwestern Medical Center and Dallas VA Medical Center and Dallas VA Medical Center and R Gare and Dallas VA Medical Center and Dallas VA Medical Center and UT Southwestern Medical Center
Keywords: Costs; Quantile regression; peripheral artery disease; Heathcare ; cost effectiveness; Claudication

Peripheral artery disease (PAD) is atherosclerosis disease caused by plagues in the artery walls thus reducing blood flow. Based on severity of disease, PAD patients can be categorized as claudication with leg pain while walking but no resting pain and critical limb ischemia (CLI) patients with resting pain or tissue loss. Endovascular revascularization may mitigate symptoms but at a high cost. Treatment costs were bimodal and heavily right skewed. Thus, the regression at mean fails to convey accurate cost estimation of PAD intervention. To address this, the study introduced quantile regression to examine PAD intervention cost differential between claudication and CLI patients. General Linear Model estimates of procedural and 1 year PAD-related costs between claudication and CLI were similar, while differential in cost quantile estimates between CLI and claudication increased linearly in quantiles with significantly higher 1 year PAD-related costs in CLI patients than claudicants at 75% and higher. Treating claudication was cost effective in 1-year PAD-related costs, while actual procedural costs for treating claudication were significantly higher than CLI at all quantile levels.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2018 program