Intention to Treat Analysis
Definition: "ITT analysis includes all randomized patients in the groups to which they were randomly assigned, regardless of their adherence with the entry criteria, regardless of the treatment they actually received, and regardless of subsequent withdrawal from treatment or deviation from the protocol” Fisher et al. (1990)
In other words, it ignores anything that happens after randomization, including noncompliance, protocol deviations, withdrawal, etc. ITT analysis. This results in an estimate of treatment effect that is generally conservative. It works on the assumption that subjects may be noncompliant or may drop out from the study due to their response of treatment.
“Once randomized, always analyzed”
- maintains prognostic balance generated from the original random treatment allocation
- reduces bias- avoids overoptimistic estimates of the efficacy of an intervention resulting from the removal of non-compliers by accepting that noncompliance and protocol deviations are likely to occur in actual clinical practice
- more applicable to real life
- CONSORT guidelines say you should
- interpretation might become difficult if a large proportion of participants cross over to opposite treatment arms
- may not analyze the treatment/intervention effectively if large amount of missing data
- more susceptible to type II error (false negative)
Modified ITT (mITT)
Some subjects are excluded in a justified way (ie patients who never started treatment)- this however is arbitrary and subjective. Some argue these are essentially “extended” per protocol analyses.