lecture 6 Flashcards
clinical trials
what factors must be considered when setting up and running clinical trial
consent, controls, randomisation, blinding
consent
given before research, full information given, enter freely
controls
control group used as a standard, don’t receive treatment
randomisation
treatments assigned to patients by chance, avoid bias
randomisation types
simple, block and stratified
simple randomisation
based on a single sequence of assignments, groups can still be uneven
block randomisation
researcher determines the block size, gives equal sample sizes, calculates all combinations
stratified randomisation
extension of block randomisation, controls and balances influenced of covariates, generates a separate block for each combination of covariates
blinding
prevents reuses from being influenced by researchers and participants, single/double blinding, allows accurate data collection
placebo
looks like the real thing but has no active ingredient
hypothesis
proposed assumption that may or may not be true
hypothesis testing
evaluation done to either confirm/dissaprove a hypothesis
null hypothesis
captures our current situation “new medicine is no better than the current treatment’
alternative hypothesis
captures what we want to achieve by doing the trial “new medicine is better than the current”
sample size must be
large enough to have a good chance of distinguishing real effects
