Principles of RCTs (Epidemiology)

Question 1

Advantages of random allocation

Answer 1

• Ensures that the baseline characteristics of patients in control and intervention groups are similar
• Avoids allocation bias
• Simplifies interpretation of differences in outcomes between intervention and control. No need to account for bias/confounding

Facilitates blinding - reducing the risk of information bias

Question 2

Define per protocol analysis

Answer 2

Analysis carried out on the basis of the treatment actually taken

• Should only be used as a subsidiary analysis

-Can over estimate the effect of treatment

• Can introduce bias into the trial - how much depends on rate of attrition and characteristics of dropouts and crossovers

Question 3

Define Intention to Treat Analysis

Answer 3

Analysis carried out on the basis of the original random allocation.

• May underestimate the effect of intervention
• Provides an unbiased estimate of intervention effect - sticks to principles of randomisation and does not introduce allocation bias into the study

-More representative of real life

• Favoured method and should be the main analysis

Question 4

Pros and cons of RCT

Answer 4

Strengths
- Conducted well, RCT can exclude possibility of confounding a bias (more likely to be a true association)
- Wide range of study topics, e.g. surgical treatment, pharmacological
- Able to determine size of treatment effect
- Level 1 and 2 evidence
- Can have subgroup analysis enhancing usefulness in clinical practise

Weaknesses
- More difficult to assess long term interventions and lifestyle changes
- Lack of choice in traditional RCTs may lead to attrition
- May not be ethical to give intervention/exposure
- Expensive
- Logistically challenging

Question 5

Advantages of larger sample size

Answer 5

Higher statistical power

More precise estimate of effect size

Ability to look at impact of intervention in a range of subgroups (e.g. ages, gender)

Question 6

Ethics of RCTs

Answer 6

Presumed that:
Patients will not come to harm
Will not be excluded from the standard effective treatment currently available
Fully informed of risks and benefits

Therefore:
Interventions must be likely to benefit rather than cause harm
Control group needs to be focused on usual standard of care available
Proper informed consent procedures should be in place
Monitoring and reporting any adverse effects as and when they occur