RCT Flashcards
Advantage of RCT
Reduces confounders by distributing equally between groups via randomisation
Increases internal validity
Purpose of allocation concealment
Prevent selection bias
Purpose of blinding (where possible, single or double)
Prevent observer bias and recall bias, view of outcome may be affected
Purpose of intention to treat analysis
Prevent selection bias due to loss to follow up
Why use post hoc adjustment
Adjust for differences in study groups not minimised by randomisation
Relative risk
Risk in treatment group/risk in control group
Can be said as: Treatment group has x% risk of outcome compared to control
Absolute risk reduction (ARR)
Risk in control - risk in treatment
Can be said as: Treatment has x% less risk of outcome vs control
Number needed to treat
100/ARR
x number need to be treated to prevent 1 outcome
Risk vs odds
Risk is chance out of total whereas odds is ratio of positive outcome vs negative
Risk = positive outcome/total treated Odds= positive outcome/negative outcome
Type 1 error
Finding a difference when there is no difference, false positive
Known as α
Type 2 error
Finding no difference when there truly is a difference in the population, false negative
Known as β
p value definition
The probability of the observed result or more extreme results, if the null hypothesis is true
Essentially shows probability of type 1 error
what is power of an rct
Ability to detect a difference if one exists
1-β(type 2 error)
What determines power
Effect size
Sample size
Distributions (S.D)
