Confounding

• Confounding:

  • The potential weakness of all non-randomised studies
• Confounding:

  • Confounding may increase/decrease or totally manufacture any observed effect
  • How to control for confounding:
  • Prevent at design stage:
  • restrict, randomly select, match
  • Deal with at analysis stage:
  • stratification, multivariate modelling
  • but only for things you know about