Review of an RCT Flashcards
What can be evaluated with RCTs?
Drugs
Surgery (ethical consideration for control group)
Health technology
Health promotion programs
Public policy
What are the general aspects of an RCT?
Exposure/treatment determined by researcher
Control / placebo / comparison group
Random allocation of subjects to treatment groups
Blinding of treatment
Eligibility criteria to filter participants (improved internal validity at the expense of external validity)
What is randomisation?
System of assigning study subjects to treatment groups that are independent of the individual participants’ characteristics.
Why do we randomize?
To eliminate selection bias. This is because it removes systematic differences between treatment and control groups.
What stages is blinding conducted?
In the stage of selection of partcipants that get intervention and those that are a control group.
The follow up measurement should be completely random. No knowledge of who was circumcised should take place when measuring HIV.
What are the criteria for a confounder to be associated with a disease?
Confounder needs to be associated with exposure of interest
or
Confounder should be a risk factor for the disease
or
Confounder should be causally linked to the disease.
What are stopping rules?
Essential for most clinical trials; they are rules made objectively prior to recruitment to stop a trial.
What are possible reasons a trial may be stopped?
Success
Harm
Adverse events
New evidence that changes nature of the trial
What factors affect internal validity?
Chance (random error) - P value 0.0031
Confounding
Bias
What is done in RCTs to reduce selection and information bias?
Randomisation reduces selection bias by removing bias in allocating study groups
Information bias is reduced by blinding
