Loading
Analysing every patient in the group they were randomised to, whether or not they actually took the treatment. It preserves the fairness of randomisation and reflects what happens when a treatment is prescribed in real life; per-protocol analysis, by contrast, counts only those who complied.
ITT is the primary analysis for superiority trials because excluding non-compliers or early dropouts reintroduces bias. Per-protocol analyses matter for non-inferiority trials, where non-compliance can mask a real difference. 'Modified ITT' variants exclude patients who never received a dose or lacked a baseline measurement and should be pre-specified. 'Response-evaluable' populations in single-arm trials inflate response rates by dropping early progressors, and 'safety population' counts patients as treated. Trial and drug pages on this site report ITT results unless stated.
Shares Non-inferiority margin and equivalence trials, Non-inferiority trial, Randomised trial.
Shares Blinding (double-blind, open-label, placebo-controlled), Control arm and comparator (investigator's choice), Why trials fail: underpowered, wrong endpoint, control arm drift, subgroup fishing, crossover.
Shares Control arm and comparator (investigator's choice), Non-inferiority margin and equivalence trials, Non-inferiority trial, Why trials fail: underpowered, wrong endpoint, control arm drift, subgroup fishing, crossover.
Shares Pre-specified vs post-hoc analysis, Non-inferiority trial, Why trials fail: underpowered, wrong endpoint, control arm drift, subgroup fishing, crossover.
Shares Blinding (double-blind, open-label, placebo-controlled), Control arm and comparator (investigator's choice), Randomised trial.
Shares Pre-specified vs post-hoc analysis, Estimands and intercurrent events (ICH E9(R1)), Why trials fail: underpowered, wrong endpoint, control arm drift, subgroup fishing, crossover.
Shares Blinding (double-blind, open-label, placebo-controlled), Randomised trial.
Shares Control arm and comparator (investigator's choice), Estimands and intercurrent events (ICH E9(R1)), Why trials fail: underpowered, wrong endpoint, control arm drift, subgroup fishing, crossover.