(0)

Already a DIA Member? Sign in. Not a member? Join.

Sign in

Forgot User ID? or Forgot Password?

Not a Member?

Create Account and Join

Bethesda North Marriott Hotel and Conference Center

Apr 23, 2018 8:30 AM - Apr 23, 2018 12:00 PM

5701 Marinelli Road, , North Bethesda, MD 20852 , USA

Course 2: Causal Inference: Weighting Methods and Case Studies

Overview

Short Course* Registration: 7:30AM-12:00PM

Inverse probability weighting (IPW) methods have been used to control for confounding in observational studies, as well as to account for switching or stopping therapy, missing information, and informative censoring. The methods offer powerful and flexible approaches useful in comparative safety and efficacy studies. This course will start with a general overview on causal inference methods using weighting and a case-study using IPW method, then we will introduce recently developed methodology of overlap weights, which places emphasis on clinical equipoise and has statistical advantages over IPW. Simulated data and R code will be provided to demonstrate practical issues of implementation: How to 1) fit the different models; 2) deal with tails of the propensity distribution; 3) check balance of the covariates in the weighted population; and 4) describe the target population to a clinical audience. The various weighting methods will be compared in a real world example.
*Short Courses are not included in the conference registration and require a separate fee.

Return to Biostatistics Industry and Regulator Forum.

Learning objectives

At the conclusion of this short course, participants should be able to:

  • Translate study questions into a causal inference framework using weighting to control for confounding
  • Use best practices of weighting methods to a causal inference problem
  • Weigh pros and cons of different weighting methods to a causal inference problem

Have an account?

Be informed and stay engaged.

Don't miss an opportunity - join our mailing list to stay up to date on DIA insights and events.