ANALYSIS AND INTERPRETATION OF CENSORED COST DATA USING REAL-WORLD EVIDENCE- A STEP-BY-STEP APPROACH
Author(s)
Abdalla Aly, PhD, Pharmerit International, Bethesda, USA; Ebere Onukwugha, PhD, MS, University of Maryland School of Pharmacy, Baltimore, USA; Ruchit Shah, PhD, Pharmerit International, Bethesda, USA; Shuo Yang, PhD, Bristol-Myers Squibb, Lawrenceville, USA
Presentation Documents
PURPOSE: To discuss censoring of costs and how to adjust for and interpret censored cost data.
DESCRIPTION: Healthcare cost data available in retrospective claims databases are typically censored because patients are lost to follow up for reasons such as health plan disenrollment or data availability. The objectives of this workshop are to: 1) provide an overview of the different types of censoring; 2) explain when and why is it important to adjust for censoring in the analysis; and 3) explain 2 methods to deal with right censoring and their impact on research findings if not accounted for. We plan to use the Surveillance, Epidemiology, and End Results (SEER)-Medicare linked database to estimate the economic burden of urothelial carcinoma in the presence of right censoring as a case study to demonstrate the 2 methods of adjustment. The audience will be guided in a stepwise manner through the following methods of adjustment: 1) Lin’s descriptive method and 2) inverse probability weighted generalized linear models. The workshop will include a detailed discussion of the 2 methods with respect to their application in other published studies and interpretation. The discussion leaders will seek audience feedback on the methods including personal experiences with adjusted costs. This workshop is appropriate for clinicians, health care decision makers, and economists who are interested in conducting health economics research using real-world evidence data.
Conference/Value in Health Info
2017-05, ISPOR 2017, Boston, MA, USA
Code
W24
Topic
Organizational Practices, Patient-Centered Research