Methodological Issues in Outcome Research

1
Methodological Issues in Outcome Research

• Effective Healthcare
• AHRQ 2008 Conference
• September 7-10, 2008

2
Purpose

• To discuss state of the art analytical approaches
  to adjust for confounding in observational
  research within the Effective Health Care Program
• Measured confounders Propensity score
• Unmeasured confounders
• Sensitivity analysis
• Instrumental variables

3
Panelists

• John D. Seeger, PharmD., DrPH.
• Senior scientist, Ingenix/i3 Drug Safety
• Transparency in the Use of Propensity Score
  Methods
• Sebastian Schneeweiss, M.D., ScD
• Department of Medicine, Harvard Medical School
• Principal Investigator of Brigham and Womens
  Hospital DEcIDE Center
• Sensitivity Analysis for Residual Confounding
• Alan M. Brookhart, Ph.D.
• Brigham and Womens Hospital, Harvard medical
  School
• Instrumental variable analysis for comparative
  effectiveness research

4
Ground Rules

• Each panelist will present for 20-25 minutes
• Questions and comments should be made at the end
  of session.
• If you have a question, please come to the
  microphone if possible.
• Please identify the panelist to whom the question
  is addressed.