RCT (Jackson)

Question 1

Why do RCT?

Answer 1

Why Do RCT?

RCT is a prospective study in which participants are allocated at random to receive:

• Study treatment or control (i.e. no study treatment), or
• One of two (or more) alternative treatments

RCT is gold standard study design for determining if intervention works (but only when it is possible to do a good one!). Trials are only ethical where uncertainty exists about research question. Types of RCT includes:

• Phase 1 is small dose ranging studies of toxicity, typically not randomised
• Phase 2 is small, short-term studies of surrogate efficacy, sometimes randomised
• Phase 3 is large, long-term studies of true efficacy and safety, always randomised

Question 2

What are the strengths and weaknesses of RCTs?

Answer 2

Strength (only one!)

• includes addresses confounding

Weaknesses include:

• Too expensive
• Usually too small (lots of random error)
• Usually not ‘real world’
• Poor compliance (not useful for long-term)

Question 3

Describe Confounding factors

Answer 3

Randomisation

Definition

Random allocation means all participants have the same chane of being assigned to each study groups.

Randomisation is any method that allows foreknowledge treatment that would be assigned, e.g. alternate participants identified, date invited to participate, alternate days/weeks/months, participants surname/initials/date of birth.

Simple Randomisation

Simple random allocation scheme is a process by which each participant has equal likelihood of being assigned to treatment versus referent groups.

However, by chance an unequal number of individuals may be assigned to each arm of the study and thus decrease the power to detect statistically significant differences between groups.

Stratified Randomisation

Stratified randomization is two-stage procedure. Patients who enter a clinical trial are:

• Firstly, divide/stratify participants into key groups according to clinical features before randomisation
• Secondly within each stratum, patients are then assigned to a treatment according to separate randomization schedules

This ensures balance between groups for values of ‘key’ variables. It is important if study is small size, where chance of imbalance is more likely (reduce confounding in small studies)

Block Randomisation

Block randomization ensures numbers of participants in study groups kept even by creating ‘blocks’ of sequences

• If there are two study groups A and B, then you could create blocks of four participants, then design a series of possible sequences (1) AABB; (2) BBAA; (3) ABAB; (4) BABA; (5) ABBA; (6) BAAB
• Then randomly choose a sequence for every four consecutive participants.
• In every sequence, two participants get A and two get B.

This method increases probability that each arm will contain an equal number of individuals by sequencing participant assignments by block.

Question 4

What are the 3 key messages about RCT?

Answer 4

1) Only thing they deal with well is confounding
2) The worst thing about them is that they’re really small
3) For you to use them, you need to translate the relative risk (30% reduction) into benefit (what is the risk in the patients without drugs)

• e.g. from 10/100 to 7/100 effect vs 10/100 7/1000
  
  • Both are 30% reduction, but more benefit if the trial had 1000