Evidence Based Medicine

Question 1

Name the 5 types of randomised control trial designs

Answer 1

• Parallel group RCTs
• Factorial design
• Crossover
• Cluster randomisation
• Stepped-wedge

Question 2

Describe parallel group RCTs

Answer 2

Participants randomly separated into intervention and control groups

Question 3

Describe factorial design RCTs

Answer 3

Each participant is randomly assigned to a group that receives a particular combination of interventions or non-interventions e.g., Group A: steroid/placebo, Group B: steroid/Acyclovir, Group C: placebo/placebo,…

Question 4

Describe crossover RCTs

Answer 4

Over time, each participant receives all interventions in a random order and with a wash-out period between each treatment

Question 5

Describe cluster RCTs

Answer 5

Pre-existing groups of participants are randomly selected to receive an intervention or placebo e.g., schools, villages, GPs

Question 6

Describe stepped-wedge RCTs

Answer 6

• Observations are collected at a baseline period when no clusters are receiving treatment
• At regular intervals, each cluster is randomised to receive the intervention and all participants are once again measured

Question 7

When is a result declared statistically significant?

Answer 7

When p<0.05

<5% probability of the result occurring by chance

Question 8

How is absolute risk (AR) calculated?

What are the units?

Answer 8

Number of events in the group divided by the number of people in that group

Units are ‘per 1000 patient years’

Question 9

How is absolute risk reduction (ARR) calculated?

Answer 9

The absolute risk of the control group (ARC) minus the absolute risk of the treated group (ART)

ARR = ARC - ART

Question 10

How is relative risk (RR) calculated?

Answer 10

Absolute risk of the treatment group divided by absolute risk of the control group

RR = ART/ARC

Question 11

How is relative risk reduction (RRR) calculated?

Answer 11

Absolute risk reduction (ARR) divided by absolute risk of the control group (ARC)

RRR = (ARC-ART)/ARC

Question 12

How is number needed to treat (NNT) calculated?

Answer 12

1 divided by absolute risk reduction

NNT = 1/(ARC-ART)

Question 13

What is the main drawback of an observational study?

Answer 13

It is impossible to control all confounding variables

i.e., variables other than that being tested

Question 14

What is the difference between a prospective and retrospective cohort study?

Answer 14

Prospective: cohort followed in time from present

Retrospective: cohort formed in the past and followed up to the present

Question 15

What are the advantages of a retrospective cohort study compared to a prospective?

Answer 15

Less time consuming

Cheaper - people don’t need to be employed or followed up regularly

Question 16

What is a case-control study?

Answer 16

• A group of subjects with a condition are identified
• A suitable control group is identified
• Frequency of exposure to a risk factor is compared between the two groups