L17 Anatomy of RCT

Question 1

What are the strengths and weaknesses of an RCT

Answer 1

Strength:
- Both groups having same confounders at baseline: important for drug studies where benefits/harms small and can be hidden by larger confounding effects

Weaknesses:
- Expensive so usually small: lots of random error

• Often not real world (ideal settings)
• participants not representative of population wanna treat (healthy volunteer)
• Poor compliance: not stay in the groups.

Question 2

What does PECOT stand for

Answer 2

Participants, Exposure (intervention)
Comparison (control)
Outcome:
Time (longitudinal (incidence) or cross sectional (prevalence)

Question 3

What is the difference between cohort study vs RCT

Answer 3

Cohort: Investigator observes people already exposed / who is not.

RCT : Investigator randomises participants to exposure & comparison groups

Question 4

What does double blind vs single blind mean for RCT. why is participant blinding important

Answer 4

Neither investigators nor participants know whether their are in EG or CG in double blind, in single blind only one doesn’t know but not both

Double blind is important because when outcomes are self reported, participant bias can effect results.

Question 5

What is the difference between simple , stratified and block randomisation

Answer 5

Single randomisation: flipping coin process designed to give all participants equal change of being allocated to EG or CG

2. Stratified randomisation: Before single randomisation, study group is separated into high/ low risk groups and then randomised
3. Block: ensures numbers of participants in study groups kept even (more small studies) by designing possible sequences of the E and C group put in different order but same number of participants for each group for each block eg. EECC, ECEC

Question 6

How do you determine if randomisation in RCT is successful and why

Answer 6

1. Generating single randomisation with random sequence to allocate participants
2. Concealing those involved (participants and investigators) from knowing which group a patient will be allocated to. eg. double blind
   - Randomisation done externally so investigator bias doesn’t try and change the group they allocate the participant to.

Question 7

How is Exposure group outcomes (EGO) and CGO reported.

What does intention to treat vs On-treatment analysis mean, and which is better

Answer 7

Number of positive outcome / number of people in the group

ITT includes everyone intended to treat in denominator (allocated right beginning)
vs
OTT only includes the actual number of people that were treated – excludes not treated/

ITT is better because less prone to confounding than OT/PP analysis