|
Activity Number:
|
338
- BIOP Student Paper Awards
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2018 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Biopharmaceutical Section
|
|
Abstract #328873
|
Presentation
|
|
Title:
|
The Statistical Performance of Matching-Adjusted Indirect Comparisons
|
|
Author(s):
|
David Cheng* and Rajeev Ayyagari and Timothy Juday and Angelina Villasis Keever and James Signorovitch
|
|
Companies:
|
Harvard University and Analysis Group and Allergan and Janssen Research and Development and Analysis Group
|
|
Keywords:
|
average treatment effect on treated;
covariate balance;
indirect comparison;
generalizability;
propensity score
|
|
Abstract:
|
Indirect comparisons of outcomes across separate studies inform decision-making in the absence of direct randomized comparisons. Differences in baseline characteristics between study populations may introduce confounding bias. Matching-adjusted indirect comparison (MAIC) has been proposed (Signorovitch et al., 2010) to adjust for observed baseline differences when the individual patient-level data (IPD) are available for only one study and aggregate data (AGD) are available for the other. The approach weights outcomes from the IPD using estimates of trial selection odds that balance covariates between the IPD and AGD. Previously the statistical properties have not been assessed. We formulate identification assumptions for causal estimands that justify MAIC estimators. We then examine its asymptotic properties and propose strategies for estimating standard errors without the full IPD in both studies. The finite-sample bias of MAIC and alternative approaches, and the performance of confidence intervals based on proposed standard error estimators are evaluated through simulations. The method is illustrated through an example comparing two antiretroviral treatment regimens for HIV-1.
|
Authors who are presenting talks have a * after their name.
