Slide 1
Intention-to-treat keeps people in their assigned groups
An intention-to-treat analysis compares participants in the groups to which they were randomized. It keeps that assignment even when treatment use changes later.
More to Read
This estimates the effect of assignment to a strategy under the trial’s conditions. It is a different question from asking what would happen if every participant followed the treatment exactly as planned.
Slide 2
Excluding people who change treatment can bias the comparison
People who stop or change treatment may differ systematically from those who continue. Excluding them can distort the randomized comparison.
More to Read
A departure might relate to side effects, worsening illness or practical obstacles. Those reasons can also relate to outcomes, so removing departures is not necessarily a neutral cleanup of the data.
Slide 3
The label alone does not show who was analyzed
Reports use intention-to-treat terminology inconsistently. Readers need to inspect who was actually included and how the groups were analyzed.
More to Read
A survey of published randomized trials found variation in how the term was applied. A methods label is therefore weaker evidence than a transparent account of exclusions and analysis decisions.
Slide 4
Keeping an assignment cannot recover a missing outcome
Keeping a person’s assigned group does not reveal an outcome that was never measured. Missing data require explicit methods and assumptions.
More to Read
Analyses can explore how conclusions change under different plausible assumptions about missing outcomes. Such sensitivity checks help expose dependence on information that the study could not directly observe.
Slide 5
Assignment and actual treatment use answer different questions
Assignment and adherence effects can both be useful. They should not be confused, because they describe different consequences of an intervention strategy.
More to Read
A simple comparison restricted to adherent participants does not automatically recover a causal adherence effect. Departures after randomization create an analysis problem that requires more than deleting inconvenient records.
Sources
- Assessing risk of bias in a randomized trial
Cochrane Handbook, chapter 8
- What is meant by intention to treat analysis?
Hollis and Campbell, BMJ