Key definitions Flashcards
Clinically-meaningful
Represents a meaningful advance in healthcare/ patient outcomes
Reliable
Results can be trusted, reproducible if trial is repeated
Internal validity
Observed differences can be correctly attributed to the intervention
External validity
Trial results can be generalized to the real-world population
Superiority trials
Test whether new drug is better than comparator e.g. placebo or current best treatment
Equivalence trials
Test whether a new drug is the same as an existing treatment
Clinical equivalence trials
Test whether clinical outcomes for a new drug are the same as for an existing treatment
Bioequivalence trials
Test whether pharmacokinetic parameters e.g. blood concentration or receptor occupancy for a new drug are the same as for an existing treatment
Non-inferiority trials
Test whether a new treatment is no worse than an existing treatment
Bias
Systematic distortion of the results of a clinical study away from the truth
Selection bias
Systematic differences in way subjects are enrolled or treatments allocated
Bias in study management
Treatment groups not handled equally (e.g. participants, samples, investigational medicinal product)
Observer ascertainment bias
Participants and/or investigators are influenced by knowledge of treatment assignment
Introduced by exclusions after randomisation
Missing data due to participant dropouts or lack of individual measurement
Publication bias
Positive trial outcomes more likely to be published than negative outcomes
Intention-to-treat analysis
Analysis includes all randomised participants, irrespective of what happened to them during the study
Per-protocol analysis
Analysis includes only participants who completed the protocol fully
Confounder
A parameter that is related to both the study treatment and the outcome, but is not caused by the treatment. Confounders can interfere with the ability of a trial to produce a true result and should be controlled for in study design by randomisation
Randomisation
A process of allocating participants to study groups that gives everyone an equal chance of being allocated to any group
Stratification
Randomisation within subgroups of key confounders
Minimisation
Dynamic method of randomisation that assigns participants to treatment groups in an order that minimises overall imbalance between groups for a number of selected confounding factors
Type 1 error
Rejecting the null hypothesis when it is true, false positive result
Designated as alpha, usually set at 0.05 (5%)
Type 2 error
Failing to reject the null hypothesis when it is false, false negative result
Designated as beta, usually set at 20%
Sample size
The number of people to be enrolled in a trial or study
Power
The ability of a study to find a significant result if one exists
Calculated as 100-beta (if beta is 20%, power is 80%)
Sample
A set of observations from the population
Population
The complete set of possible observations
Multiplicity
Performing many statistical tests on one clinical trial - increases the risk of a type 1 error/false positive result
Risk of type 1 error
[1-(1-alpha)n], where n is the number of tests performed and alpha is the type 1 error rate
Repeated measurements
Measurements at multiple timepoints
Interim analysis
Analysis conducted during the trial for ethical or economic reasons - trial may be terminated early if significant differences in efficacy or safety are identified
Bonferroni correction
Reducing the alpha of statistical tests to reduce the risk of a false positive result - divide 0.05 by the number of tests to set the significance level for the subtests
Covariate
Baseline characteristic (e.g. age, sex) that accounts for some of the variance in an outcome measured in a clinical trial (e.g. age is associated with increased risk of mortality). May modify the effect of a new treatment (e.g. drug may work in younger but not older patients)