|
Activity Number:
|
572
- Addressing Complications in Causal Inference
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Thursday, August 6, 2020 : 3:00 PM to 4:50 PM
|
|
Sponsor:
|
Section on Statistics in Epidemiology
|
|
Abstract #314094
|
|
|
Title:
|
Using Instrumental Variables to Account for Unmeasured Confounding Under a Cox Model Extended to Non-Proportional Hazards
|
|
Author(s):
|
James O'Malley*
|
|
Companies:
|
Dartmouth
|
|
Keywords:
|
|
|
Abstract:
|
In this talk, I describe and justify an approach to estimating the Cox regression model for time-to-event data when the treatment variable is exposed to an unmeasured confounder and the effect of treatment varies over follow-up. The new procedure is shown to perform better than incumbent procedures under multiple assumptions regarding the distribution of the survival times and the unmeasured confounder. The methodology is applied to a unique data set constructed from the Vascular Quality Initiative (VQI) Registry and United States Medicare data to estimate the effect of carotid endarterectomy versus carotid artery stenting on the time to death of patients suffering from carotid artery disease. The results are benchmarked against the randomized trial evidence that exists.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2020 program
|
