RCT lecture Flashcards
What is the aim of an RCT?
To find out how effective a treatment is?
What is observational evidence?
Observing individuals being treated to see if the treatment works
What is pathophysiological theory?
Use a theory of the underlying physiology to determine which drugs should be used to treat them. However this can sometimes be wrong and must be tested by RCT
What is empirical evidence?
It is the actual results of the trial and can often prove theory wrong
What are the four main explainations of an observation?
Bias
Confounders
Chance
Hypothesis correct: effective
What is internal validity?
Looking at whether bias, confounders or chance have played a role in the observation
What is external validity?
This is also known as generalisability and is whether this is applicable to the population.
What are some of the factors that have to be considered in outside of external validity for should I treat?
Benefits worth harms
Resource costs
Patient preferences
Ethics
What is bias?
A difference in measurement before and after
What is confounding?
Some other factor has changed before and after e.g. the patient is taking some other medication
What is regression to the mean?
If we observe an extreme value then part of the reason for it being extreme is chance
Therefore the next value is more likely to be closer to the mean
This can give the impression of things getting better however this is mainly due to chance
This means that unusually high levels of something are more likely to be better next time you see them due to chance
What is the definition of an RCT
It is an intervention study where participants are randomised into two or more groups where the outcome is assessed.
What is the advantage of randomisation?
Means it reduces the effect of confounders as the confounders are allocated by chance
What are the main issues of RCTs?
Choice of outcome measure Ethics Choice of control Contamination (crossover) Bias in assessment of outcome Losses to follow up Sample size
What is the ideal outcome to measure?
measure the outcome that is relevant to the patient
What is equipoise?
The belief that one treatment is not better than the other that makes it ethical to randomise people into different groups
When choosing a control when would you use no treatment?
When there is no usual care and we don’t believe new treatment is effective. Cant disadvantage patients so must give next best usual treatment if there is one
What is contamination?
Participants randomised to the control group recieving the intervention
What is the advantage of intention to treat analyses?
Reduces risk of allocation bias
However it may underestimate the effect of the analysis
What is nominal data?
This is catagorical data e.g. you either die or you don’t, blood type, gender
What is ordinal data?
This is ordered catagorical data e.g. disease stage
What is discrete data?
This is quantitive data that are intergers e.g. progression- free and overall survival
What is continous data?
Quantitative data that can be any value e.g. height, mass
What type of intervention can contamination be a problem?
When education or information is an intervention as it can be easily shared with the placebo group.
How can you reduce contamination in trials involving education?
Using cluster randomisation to randomise groups of people rather than individuals. This reduces contamination between individuals
What are the different degrees of blinding?
Unblinded/open
Single blind - participant doesn’t know if treatment of control
Double blind - researcher doesn’t know
Triple blind - Analyst reviewing results does not know
What is allocation concealment?
Researchers are blinded to whether a participant is allocated to the intervention or the control group. This prevents them from influencing the process and hence reduces selection bias
When do losses to followup effect the results?
when the loss is related to which treatment group they were allocated to and to their outcome
What sample size will studies with small expected effects require?
A larger sample size, imprecisely measured outcomes also require a larger sample size
What is eligibility criteria? What can this effect?
This is the criteria that the researchers set for who can be included in the study. If they have used a very specific eligibility criteria then you will get a very specific group of people and the results will not be very generalisable to all patients.
What is randomisation method?
refers to how the randomisation was actually done e.g. using random number generator etc.
