Clinical Trials in CVD Flashcards
What is a clinical trial?
A longitudinal study that looks at an intervention and its effect on incidence - involves an intervention group and a control group
What are the key outcomes of a clinical trial?
relative measures of intervention effect, absolute measures of intervention effect and survival analysis
What are the three key features of clinical trials?
randomisation, blinding, intention to treat analysis
What is randomisation?
A tool used to deal with confounding - random allocation of subjects into each arm of the clinical trial
What is blinding?
A tool used to deal with information bias - means that the subject or the subject and the investigator are unaware of which arm of the clinical trial they are allocated to
What is intention to treat analysis?
A tool used to deal with selection bias - means that you assume that participants remain in the group which they were assigned regardless of whether they did or not - introduces an overlap- will lead to an underestimate of the the true effect of the intervention
What is a hazard?
A continually updated instantaneous rate to look at incidence in longitudinal studies e.g. week 1 10/1000 die, week 2 15/990 die etc.
What is a survival analysis?
Looks at outcomes and their specific time of occurrence - typically done by plotting hazard against time on a kaplan-meier curve
What is a hazard ratio?
Similar to relative risk but derived from statistical analysis and looks at whole period of follow up
What does a hazard ratio of 0.5 mean?
That at any given point in time, the probability of the outcome for the intervention group was half that of the control group
What is a risk/rate reduction?
The reduction of the outcome in either relative or absolute terms
How is risk/rate reduction in relative terms calculated?
rate of outcome in intervention arm divided by rate of outcome in control arm e.g. 7/100py divided by 10/100py = 0.7
How is risk/rate reduction in absolute terms calculated?
rate of outcome in control arm minus rate of outcome in intervention arm e.g. 10/100py - 7/100py = 3/100py
What is the number needed to treat?
The number of people who are needed to undergo the intervention in order to prevent the outcome in one person
How is the number needed to treat calculated?
1 divided by absolute risk/rate reduction e.g. 1 divided by 3/100py = 33.3
