Superiority trial
A trial that seeks to demonstrate that the investigational product is better than a comparator, in terms of efficacy and/or safety.
Equivalence trial
A trial that seeks to demonstrate that the investigational product is the same as a comparator, in terms of efficacy and/or safety.
Non-inferiority trial
A trial that seeks to demonstrate that the investigational product is no worse than a comparator, in terms of efficacy and/or safety.
Dose-response trial
A trial that seeks to examine the dose-response relationship of an investigational product.
Parallel group design
A trial design in which subjects are randomized to one of two or more arms and each arm is allocated a different treatment.
Crossover design
A trial design in which each subject is his own control; subjects receive two treatments consecutively, typically with a washout period in between.
Factorial design
A trial design in which two or more treatments are tested simultaneously; each subject gets a combination of treatments and/or doses.
Group sequential design
A trial design in which groups of subjects are treated and analyzed sequentially to facilitate the conduct of interim analysis.
Multicenter trial
A trial that is conducted at multiple sites, usually to increase recruitment potential and/or generalizability of findings.
Primary variable
The primary variable provides direct evidence related to the primary objective. Typically an efficacy variable, depending on the intent and purpose of the trial.
Secondary variable
Secondary variables support the evaluation of the primary objective or measure effects specified in the secondary objectives.
Composite variable
Composite variables combine multiple variables to support a particular objective.
Global assessment variable
Global assessment variables are more subjective; for example, scaled ratings, or the investigator’s overall impression of how the subject is doing.
Multiple primary variables
It may sometimes be desirable to use more than one primary variable to account for the effects observed. While not ideal due to the increased risk of study failure due to the lack of both endpoints being achieved, it may be warranted in certain disease states.
Surrogate variable
Surrogate variables allow prediction of clinical benefit without measuring actual clinical efficacy.
Categorized variable
Categorized variables represent continuous variables as discrete units.
Open-label trial
A trial in which treatment identities are known to everyone involved.
Single-blind trial
A trial in which the investigator and/or staff may be aware of treatment identities but subjects are not (or vice versa).
Double-blind trial
A trial in which treatment identities are unknown to everyone involved (this is the optimal approach).
Double-dummy trial
A trial in which subjects receive the active form of one treatment and the placebo form of the other. Typically used when the two treatments cannot be made identical.
Unrestricted randomization
A randomization method in which all subjects have the same chance to receive each treatment. Like flipping a coin.
Block randomization
A randomization method in which subjects are randomized in blocks, each containing a predefined combination of treatments.
Central randomization
A randomization method used in multicenter trials to obtain the desired overall distribution across treatment arms.
Stratified randomization
A randomization method in which subjects with shared characteristics are divided into strata; each stratum is then randomized separately.
