Global web ALERT button

Comparative Effectiveness Research Study Design

 Click to view the video in a new window.

Notes

Recorded at the Comparative Effectiveness Research with Population-Based Data Conference, Baker Institute at Rice University, 2012. Presented by Matt Maciejewski, PhD, Professor in the Department of Population Health Sciences. 

 

Learning Module Notes Modules

  1. Introduction (0:00-8:20)
  2. Mary Beth Landrum’s presentation: Use of Geographic Variation to Estimate Effectiveness With Nonrandomized Data (8:20-1:11:36)
  3. Principles around study design (1:13:30-1:16:54)
    • Objectives
    • Limitations to internal validity
    • Strengths of external validity in quasi-experiments 
  4. Experimental studies/RTCs enable causality statements (1:16:54-1:18:31)
    • RCT is considered the gold standard
    • Not the case in quasi-experiments – conditions for causality in the absence of randomization
  5. Value & Motivation of Quasi-Experimental Studies (1:18:31-1:26:11)
    • What makes an ideal Quasi-Experimental Study
    • Internal validity: Refers to the ability to make casual inferences
    • The controversy between experimental and quasi-experimental research
    • Overall point of lecture: What are the design choices that must be considered to ensure a well-designed quasi-experiment in CER?
  6. 4 Design Choices for Rigorous Quasi-Experiments: (1:26:11-1:27:10)
    • In absence of randomization: incident or prevalent users, control group, pre-period measurement, post-period measurement 
  7. Incident & Prevalent User Designs: (1:27:10-1:30:32)
    • 3 Cohort Choices: Incident (new) users, prevalent users, both
    • Tradeoffs of incident user design
    • Defining pre-period for incident cohort designs 
  8. Example: Incident Cohort Study (1:30:32-1:32:47)
  9. Considerations before using Prevalent User Cohort: (1:32:47-1:40:55)
    • Tradeoffs in prevalent user design
    • Defining pre-period for prevalent cohort designs
    • Choice of inclusion criteria can impact adherence estimates
    • Incident vs Prevalent Examples
  10. The Role of Counterfactuals for Internal Validity (1:40:55-1:53:12)
    • Importance of a control group
    • What is a counterfactual? Ways of estimating.
    • Counterfactuals in quasi-experiments
    • Assuming you have non-equivalent controls
    • Selection of control group
    • Take-aways of user cohorts
  11. Threats to Internal Validity (1:53:12-2:00:47)
    • Selection, ambiguous temporal precedence
    • Example
    • History & maturation; threats to internal validity 
  12. Measurement Issues for Internal Validity (2:00:47-2:04:25)
    • Pretests, posttests, non-equivalent dependent variables
    • External validity
    • CER/Translation research
  13. Bottom line for quasi-experimental study designs & Conclusions (2:04:25-2:05:11)
  14. Q&A (2:05:11-2:17:33)
  Learning Modules