Activity Number:
|
164
- SPEED: Causal Inference and Related Methodology
|
Type:
|
Contributed
|
Date/Time:
|
Monday, July 30, 2018 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Statistics in Epidemiology
|
Abstract #330778
|
Presentation
|
Title:
|
Assessing Therapeutic Equivalence of Brand and Generic Drugs Using Observational Data
|
Author(s):
|
Lamar Hunt* and Daniel Scharfstein and Irene Murimi and Jodi Segal and Ravi Varadhan and Ramin Mojtabai
|
Companies:
|
Johns Hopkins Bloomberg SPH & OptumLabs Visiting Fellows and Johns Hopkins University and Johns Hopkins Bloomberg SPH & OptumLabs Visiting Fellows and Johns Hopkins Bloomberg SPH & OptumLabs Visiting Fellows and Johns Hopkins University and Johns Hopkins Bloomberg SPH
|
Keywords:
|
Survival Analysis;
Generic Drugs;
Therapeutic Equivalence;
Regression Discontinuity;
G-Computation;
Insurance Claims Data
|
Abstract:
|
Although generic drugs are required to be bioequivalent to brand, generic producers are not required to establish therapeutic equivalence through clinical trials. We describe a method to assess therapeutic equivalence of brand and generic drugs using insurance claims data for the anti-depressant venlafaxine and time to drug failure as an outcome. Generic market entry is typically followed by a large shift among new users towards initiation on generic, resulting in little overlap in initiation times of brand and generic users. This creates temporal confounding if observed survival times are affected by changes over time in unmeasured variables. There is also time varying confounding. Our method addresses both of these concerns by applying Regression Discontinuity to counterfactual survival curves, with a discontinuity in the probability of initiation to generic at the date when generic becomes available. The survival curves themselves are estimated using G-Computation to account for the time-varying confounding.
|
Authors who are presenting talks have a * after their name.