POPH192 Lecture 21 - Randomised Controlled Trials

Question 1

what are the 3 essential elements of a randomised control trial?

Answer 1

1) randomised= participants randomly allocated to groups
2) controlled= always have a comparison (control) group
3) trail= testing effect of treatments/interventions

Question 2

does the intervention increase or decrease the likelihood of the outcome?

Answer 2

hard to distinguish whether the change is due to the intervention or sample group

Question 3

what is confounding?

Answer 3

where two groups differ by some factor which is related/associated with the outcome

Question 4

what is the point of random allocation?

Answer 4

• equal chance for each participant to be in either group
• if enough people are randomly allocated, should have the same proportion of confounder in each group (this applies to known and unknown confounders)

Question 5

what is the result of successful randomisation?

Answer 5

successful randomisation means confounding is an unlikely reason for differences in outcomes between groups

Question 6

when does random allocation occur?

Answer 6

occurs when people who have already been randomly selected are randomly allocated in treatment and control groups

Question 7

when does random selection occur?

Answer 7

recruit people form the source population into your sample

Question 8

what is cluster randomisation?

Answer 8

randomise groups of people (e.g households or schools)

-macrointerventions

Question 9

what is the key strength of a random control trial?

Answer 9

randomisation

Question 10

what are the 2 ways to protecting randomisation?

Answer 10

1) concealment of allocation

2) intention-to-treat analysis

Question 11

what is important in terms of allocation?

Answer 11

important that the allocation sequence is concealed and unpredictable (removing bias)

Question 12

what is analyse as randomised?

Answer 12

analyse people in the group they were originally assigned in by intention-to-treat
-can more accurately reflect ‘real-world’ effect of intervention

Question 13

what is per-protocol analysis?

Answer 13

• analyse as treated

- tends to be more appropriate for efficacy trials

Question 14

to protect randomisation the most effectively should a study use intention-to-treat or per-protocol?

Answer 14

intention-to-treat analysis

Question 15

what are 3 potential sources of bias for randomised control trials?

Answer 15

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

Question 16

what is blinding?

Answer 16

with-holding information of which arm a person of a trial is in

Question 17

what is the challenge to achieving blinding in practise?

Answer 17

• can be obvious which group the participators is in

- safety concerns

Question 18

what is non-adherence?

Answer 18

participants not doing what they are supposed to do

-can include doing what the other group is doing

Question 19

what are the advantages of a randomised control trial?

Answer 19

• bets way to evaluate an intervention
• can calculate incidence (so can directly calculate relative risk and risk differences)
• strongest design for demonstrating a casual association

Question 20

are randomised control trials intervention or observational study designs?

Answer 20

intervention study designs

Question 21

what is clinical equipoise?

Answer 21

genuine uncertainty about the benefit or harm of intervention

Question 22

what is unethical if given in a randomised control trial?

Answer 22

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

Question 23

what are the practical issues of a randomised control trial?

Answer 23

• very resource intensive
• exposures needs to be modifiable
• highly selective (can affect generalisability)