Randomised control trial

Question 1

What is the hierarchy of evidence from highest to lowest?

Answer 1

Sys Reviews Metanalysis
RTCs (Randomised Control Trials)
Cohort Studies
Cases - Control
Cross Sectional Studies
Case series, Case reports
Ideas, Opinions, Editorials, Anecdotal

Question 2

What is the definition of a randomised control trial (RTC)?

Answer 2

A study in which a number of similar people are randomly assigned to 2 (or more) groups to test a specific drug, treatment or other intervention.
Experimental may be laboratory studies clinical or field trials.

Question 3

What are the 2 types of groups in a RTC?

Answer 3

Treatment group (intervention)
Control group

All individuals in both groups have the disease under treatment
Subjects are assigned at random to get treatment or not.
Randomisation means that differences between subjects are divided between study groups.

Question 4

How can bias be reduced?

Answer 4

Subject randomisation minimises chance of bias
Blinding can be several layered.

Question 5

What are the key 6 steps in a RTC?

Answer 5

• Enrolment
• Assessed for eligibility
• Randomized
• Allocation
• Follow up
• Analysis

Question 6

What are 6 strengths of RTCs?

Answer 6

• Direct comparison of treatments
• Can make casual inference
• Randomisation minimises bias
• Randomisation minimises confounding
• Statistical test of significance is easily
  interpretable
• If sufficient power, avoids Type 1 and Type 2
  errors

Question 7

What are 5 weaknesses of RTCs?

Answer 7

• If large sample size required (e.g. small
  effect, rare outcomes) can be very expensive
  and logistically difficult
• If outcomes occur after long delay, increases
  cost and loss to follow up
• Results may not be achieved in “real life”
  outside study
• Ethical implications - RTC not possible
  without clinical equipoise
• Informed consent may be impossible

Question 8

What does Equipoise mean?

Answer 8

Genuine clinical uncertainty between 2 options

Question 9

What is a type 1 error?

Answer 9

Concluding that there is an effect when there isn’t one

Question 10

What is a type 2 error?

Answer 10

Concluding that there isn’t an effect when there is