Lecture bioinformatic databases - Tuesday - Theory of clinical trials
18 important questions on Lecture bioinformatic databases - Tuesday - Theory of clinical trials
How to deal with confounding by indication (when randomization is not possible)?
--> usually, adjustment in statistics cannot remove all confounding
What are the phases of a clinical trial?
- phase I = safety study (20-80 people)
- phase II = side effects and effectiveness (100-300)
- phase III = effectiveness and monitor side effects (1000-3000)
- phase IV = monitor long-term side effects
What is tested in the phase I of a clinical trial?
- Higher grades + faster learning
- Never study anything twice
- 100% sure, 100% understanding
What is tested in a phase III trial of a clinical trial?
What is tested in a phase IV trial?
What is a placebo?
Difference medication, prosthesis and surgical experiment
Prosthesis = new material versus old
Surgical = new technique vs classical
What is the index group and what the control group?
Control = placebo or standard treatment
Why is a control group necessary?
What is a cross-over experiment?
There is a wash out period in between the two treatments.
What is the problem with simple randomisation?
In case of interim analysis, unequal distribution of subjects
Treatment groups not balanced with respect to risk factors of prognostic factors
What is block randomisation and what is the advantage?
Advantage: Equal number of subjects in the treatment groups (even if study is terminated early). Is used in multicenter trials.
Randomization of blocks of subjects of even size (4, 6 8 etc.)
Half of the patients of each block are assigned to treatment A and other half to B
Termination of trial: always approximately equal number of subjects
What is stratiefed randomisation and what is the purpose?
Before study entrance subjects are subdivided in groups, strata, based on baseline characteristics.
Subjects in each stratum are randomly assigned to a treatment group.
Use stratified randomisation in combination with block randomisation.
Knowledge of treatment will influence the evaluation of the outcome, how can this be solved?
- single blind = participants are given no information about the identity of their treatment
- double blind = neither participants nor investigators are aware of the identity of the assigned treatment
What are open label studies?
What is internal validity and external validity?
External = generalizability of results, clinically important outcomes, benefit vs harm
What are two types of clinical trials?
- how does treatment work and what is the efficacy (werkzaamheid) of treatment
Pragmatic = effectiveness trial
- How should I treat patients with this disorder?
- Effectiveness (doeltreffendheid) of treatment
How can one present effect sizes?
Relative risk (hazard ratio)
Absolute risk difference
Relative risk reduction
Number needed to treat / number needed to harm
The question on the page originate from the summary of the following study material:
- A unique study and practice tool
- Never study anything twice again
- Get the grades you hope for
- 100% sure, 100% understanding