Analysis of outcome measures (Statistics)

Question 1

Define Intention to Treat + pros/cons

Answer 1

Study participants analysed according to the group to which they were originally randomised regardless of whether they actually received the intervention (or even started it) or dropped out of the study

Pros: Reduces probability of overestimating effectiveness (Type 1 error)

Reduces selection bias

Preserves sample size and statistical power

More reflective of the real world where non adherence and other protocol deviations occur

Cons: Improves the estimate of the effect of treatment under optimal conditions

Question 2

Define Per Protocol

Answer 2

Only participants who adhered to the study protocol are analysed

Pros: Non adherence