- Intention to Treat Analysis (ITT):
- all participants that are randomized must be included in the final analysis and analyzed according to the treatment group to which they were originally assigned, regardless of the treatment received, withdrawals, loss to f/u or cross-overs
- However…in many instances in randomized trials the term intention-to-treat was inappropriately described and participants improperly excluded.
- Generally preferred as they are unbiased, and also because they address a more pragmatic and clinically relevant question.
Keep participants in the intervention groups to which they were randomized, regardless of the intervention they actually received.
Measure outcome data on all participants.
Include all randomized participants in the analysis.
There is no clear consensus on whether all criteria should be applied (Hollis 1999). While the first is widely agreed, the second is often impossible and the third is contentious, since to include participants whose outcomes are unknown (mainly through loss to follow-up) involves imputing (‘filling-in’) the missing data.
- Modified Intention to Treat Analysis (mITT):
- No clear definition and can vary from trial to trial
- Primary factor that characterizes the mITT: Post-randomization exclusion
- Lies between true ITT and per-protocol analyses
- E.g. patient receives at least 1 dose of medication and 1 efficacy measurement…as long as that happens they were included in that study
- Also consider the number that was censored
- More likely to report post-randomization exclusion and more likely to be associated with industry sponsorship
- As-Treated: subjects analyzed according to whether they got treatment or not (regardless of allocation) → May be used for safety events
- Sensitivity analysis: a repeat of the primary analysis or meta-analysis, substituting alternative decisions or ranges of values for decisions that were arbitrary or unclear
- Asks the question: “Are the findings robust to the decisions made in the process of obtaining them?”
For dichotomous outcomes, should odds ratios, risk ratios or risk differences be used?
And for continuous outcomes, where several scales have assessed the same dimension, should results be analysed as a standardized mean difference across all scales or as mean differences individually for each scale?
- Different from subgroup analysis:
I. sensitivity analyses do not attempt to estimate the effect of the intervention in the group of studies removed from the analysis, whereas in subgroup analyses, estimates are produced for each subgroup.
II. in sensitivity analyses, informal comparisons are made between different ways of estimating the same thing, whereas in subgroup analyses, formal statistical comparisons are made across the subgroups.
Intention to Treat and Per Protocol
|Intention to Treat||Per Protocol|
|All subjects who were randomized are analyzed at the end of the trial according to the treatment to which they were originally assigned
This does not fix problems related to loss to f/u
Superiority studies: Conservative
Non-inferiority studies: Less conservative
|All subjects are included only if they received the intended intervention in accordance with the protocol
= ideal patients
Superiority studies: less conservative
Non-inferiority studies: conservative
- Modified vs. Intention to Treat Analysis:
- Cochrane Handbook – Sensitivity Analyses: