Lecture 21: Randomised Control Trials

Question 1

What type of studies are randomised control trials?

Answer 1

analytic and intervention

Question 2

What are our two groups that we compare in a randomised control trial?

Answer 2

treatment group and comparison group

Question 3

What does the “randomised” part of RCT involve?

Answer 3

participants are randomly allocated into groups

Question 4

What does the “contolled” part of RCT involve?

Answer 4

always have a comparison (control) group

Question 5

What does the “trial” part of RCT involve?

Answer 5

testing effect of treatment/interventions

Question 6

What does a trial involve?

Answer 6

we take a sample from the source population and assign participants to the exposed (treatment) or comparison (control) group

Question 7

What does randomisation involve and how does this differ from random selection?

Answer 7

Randomisation is the same as random allocation which is when people are randomly allocated into either the treatment or control group. There must be an equal chance for each participant to be in either group.
Random selection is the random selection of people from a source population into the sample

Question 8

What is the advantage of randomisation?

Answer 8

if enough people are randomly allocated, there should be the same proportion of a confounder in each group (for both known and unknown confounders).

Question 9

What does successful randomisation mean?

Answer 9

confounding is an unlikely reason for differences in outcome between groups

Question 10

What are the steps to carry out an RCT?

Answer 10

• Identify source population
• Randomly select sample who don’t have the outcome of interest
• Randomise the sample to either the intervention or control group
• Follow up over time to see who develops the outcome

Question 11

What two things can be done to protect the randomisation?

Answer 11

• concealment of allocation

- intention to treat analysis

Question 12

What is meant by “we can protect randomisation by concealment of allocation”?

Answer 12

It is important that the allocation sequence is concealed and unpredictable. This means that by looking at someone, we can’t know the allocation of that person and each allocation is independent of another

Question 13

What is meant by intention-to-treat analysis and why is it important?

Answer 13

If people are randomised into the treatment group, you analysis their results as if they are in this group, even if they did not take the treatment. This is important because it can reflect the “real-world” effect of intervention ie. not everyone is going to take the treatment in real life

Question 14

What are three potential sources of bias?

Answer 14

1. blinding
2. loss to follow up
3. non-adherence

Question 15

Why is blinding important?

Answer 15

We need to not let the researcher or the participant know which group they are in because other wise people may react differently depending on which group they are in

Question 16

What is non-adherance?

Answer 16

When participants do not do what they are supposed to which can include doing what the other group is doing

Question 17

What are the advantages/strengths of an RCT?

Answer 17

• it is the best way to evaluate an intervention
• we can calculate incidence and therefore relative risk and risk difference
• we can demonstrate a causal association

Question 18

How can loss to follow up lead to bias?

Answer 18

People who leave the study which can lead to bias and confounding as those who leave the study may be different from those who stayed

Question 19

What are the two ways to randomise and what do these involve?

Answer 19

• Cluster randomisation: when subgroups are randomised rather than individuals
• stratified/block stratification: where participants are randomised within blocks

Question 20

What are 6 limitations of an RCT?

Answer 20

• Many exposures we cannot randomise
• Need clinical equipoise
• Expensive
• Time consuming
• Participants may not represent the general population
• Not good for rare outcomes

Question 21

What is clinical equipoise?

Answer 21

genuine uncertainty about the benefit/harm of the intervention

Question 22

In terms of clinical equipoise, what three things its it unethical to do?

Answer 22

• give known harmful interventions to people
• give interventions known to be less effective than current treatments
• waste resource and risk people’s well-being if you already know the answer

Question 23

In an RCT, the exposure needs to be

Answer 23

modifiable

eg. age and vaccination status are not exposures because they can not be changed

Question 24

What is the problem with RCTs being highly selective?

Answer 24

Often a very select group of people are included which can affect generalisability