9. Randomised Controlled Trials

Question 1

Define relative risk.

Answer 1

The likelihood of an effect with one exposure compared to the chance of the effect with another exposure.

Question 2

Define a clinical trial.

Answer 2

Any form of planned experiment which involved patients and is designed to elucidate the most appropriate method of treatment of future patients with a given medical condition.

Question 3

What is the purpose of a clinical trial?

Answer 3

To provide reliable evidence of treatment efficacy and safety.

Question 4

Define efficacy.

Answer 4

The ability of a health care intervention to improve the health of a defined group under specific conditions.

Question 5

Define safety.

Answer 5

The ability of a health care intervention not to harm a defined group under specific conditions.

Question 6

What are three criteria of clinical trials?

Answer 6

Reproducible - in experimental conditions
Controlled - comparison of interventions
Fair - unbiased without confounding

Question 7

What are the stages in drug development?

Answer 7

Pre-clinical phase - laboratory studies on cell cultures
Phase I - volunteer studies on a few healthy volunteers
Phase II - treatment studies on several patients to work out effects and dosages
Phase III - clinical trials on lots of patients to compare to standard treatment
Phase IV - post-marketing surveillance to look for potential new uses and monitoring for adverse reactions.

Question 8

What are problems with non-randomised clinical trials?

Answer 8

Allocation bias - by patient, clinician or investigator.

Confound - known and unknown.

Question 9

How does comparison with historical controls work in clinical trials?

Answer 9

The group of patients who have had the standard treatment is compared with a group of patients receiving the new treatment.

Question 10

What are problems with historical controls in clinical trials?

Answer 10

Selection can be less well defined and rigorous, treated differently from the new treatment group, less information about potential bias or confounders and unable to control for confounders.

Question 11

What is the first step of a randomised control trial?

Answer 11

Definition of factors: the disease of interest, the treatments to be compared, the outcomes to be measured, possible bias and confounders, patients eligible for the trial and patients to be excluded from the trial.

Question 12

What is the second step of a randomised control trial?

Answer 12

The trial is conducted: a source of eligible patients are identified and invited to join in the trial, patients willing are consented, participants are fairly allocated to the treatments, patients are followed up identically. Losses to follow up are minimised and compliance with treatment is maximised.

Question 13

What is the third and final step in randomised control trials?

Answer 13

Comparison of the outcomes: is there an observed difference in outcome between treatment groups? Could the observed difference be down to chance? How big is the observed difference? Is the observed difference because of the treatments?

Question 14

Why are outcomes pre-defined?

Answer 14

It prevents data dredging and repeated analysis, it gives protocol for data collection, there’s an agreed criteria for measurement and assessment of outcomes.

Question 15

What are primary and secondary outcomes?

Answer 15

Studies should preferably only have one primary outcome and its used in the sample size calculation.
Secondary outcomes are other outcomes of interest like occurrence of side effects.

Question 16

What are the three types of outcome?

Answer 16

Patho-physiological, clinically defined and patient focused.

Question 17

What are six criteria for an ideal outcome?

Answer 17

Appropriate and relevant, valid and attributable, sensitive and specific, reliable and robust, simple and sustainable and cheap and timely.

Question 18

What is non-random allocation?

Answer 18

Allocation of participants to treatment by a person, historical basis etc. that leads to potential for allocation bias and confounding factors to unwittingly cause unidentified differences between the treatment groups being compared.

Question 19

What does random allocation minimise?

Answer 19

Minimal allocation bias and minimal confounding.

Question 20

How may an open label trial lead to bias?

Answer 20

Non treatment effect - clinician may alter their treatment, care or interest in the patient.
Measurement bias - investigation may alter their approach when making measurements and assessing outcomes.

Question 21

How does randomisation minimise allocation bias?

Answer 21

It ensures that randomisation gives each participant an equal chance of being allocated to each of the treatments.

Question 22

What is a single blind?

Answer 22

One of the patient, clinician or assessor don’t know the allocated treatment.

Question 23

What is a double blind?

Answer 23

Two of patient, clinician or assessor don’t know the treatment allocation.

Question 24

What does triple blind mean?

Answer 24

All three of patient, clinician and assessor don’t know the treatment allocation. Usually referred to as double blind.

Question 25

When can blinding be difficult?

Answer 25

In surgical procedures, psychotherapy vs anti depressant, alternative medicine, lifestyle interventions, prevention programmes.

Question 26

What is the placebo effect?

Answer 26

The effect the thought of having a new treatment will have on patients, their expectations change and a proportion of the effect of a treatment seen is down to the patient’s expectations.

Question 27

What is a placebo?

Answer 27

An inert substance made to appear identical in every way to the active formulation with which it is to be compared.

Question 28

What is the aim of the placebo?

Answer 28

To cancel out any placebo effect that may exist in the active treatment.

Question 29

What are some ethical issues with placebo?

Answer 29

It should only be used if there is no standard treatment available, it is a form of deception so participants but be informed that they may receive a placebo.

Question 30

What are the two causes of losses to follow up?

Answer 30

Their clinical condition means they have to be removed from the trial, appropriate, or the may choose to withdraw from the trial, unfortunate.

Question 31

How can losses to follow up be minimised?

Answer 31

Make the follow up practical and minimise inconvenience, be honest about the commitment required form participants from the start, avoid coercion or inducements, maintain contact with the participants.

Question 32

Why might participants have noncompliance with the allocated treatment?

Answer 32

They may have understood the instructions, they may not like taking their treatment, they may think their treatment is not working, they may prefer another treatment, they can’t be bothered.

Question 33

How can noncompliance with treatments be minimised?

Answer 33

Simplify instructions, ask about compliance, ask about effects and side effect and monitor compliance.

Question 34

Which two ways can the new treatment be compared to the existing treatment in clinical trials?

Answer 34

Look at the physiological action of the new treatment vs old treatment, as treated analysis. Or look at the new treatment vs current in clinical practice, takes compliance into account, intention to treat analysis.

Question 35

How do as treated and intention to treat analyses differ?

Answer 35

As treated analysis only looks at those who completed the follow up and complied with their treatments. Intention to treat looks at all data so compares the likely effect in clinical practice.

Question 36

In what way is intention to treat analysis better than as treated analysis?

Answer 36

It keeps randomisation, whereas as treated gets rid of non compliance data which gets rid of some randomisation. So there is lower selection and confounding with intention to treat.

Question 37

What are the collective and individual ethics associated with clinical trials?

Answer 37

Collective - all patients should have treatments properly tested for efficacy and safety.
Individual - principles of beneficence, non-maleficence, autonomy and justice.

Question 38

What five things must a clinical trial be to be approved by the research ethics committee?

Answer 38

Clinical equipoise, scientifically robust, ethical recruit, valid consent and voluntariness.

Question 39

What is clinical equipoise?

Answer 39

Where there’s reasonably uncertainty or genuine ignorance about better treatment or intervention.

Question 40

What are problems with measuring clinical equipoise?

Answer 40

Is the uncertainty from the individual clinician or the whole scientific community? What counts as reasonable uncertainty? Is better for the individual or patients generally?

Question 41

What is meant by scientifically robust for clinical trials?

Answer 41

It addresses a relevant, important and valid issue/ question, has an appropriate study design and protocol, has the potential to reach sound conclusions, has acceptable risks of possible harm compared to possible benefits, has provision of monitoring safety and has arrangements for appropriate reporting and publication.

Question 42

What issues could there be with ethical recruitment?

Answer 42

Inappropriate inclusion of participants form communities that are unlikely to benefit, participants with high risk of harm vs benefit, participants who will likely be excluded later.
In appropriate exclusion of people who differ from an ideal homogenous group or people who are difficult to get valid consent from.

Question 43

Who does valid consent come from?

Answer 43

A knowledgable informant who’s been given appropriate information. They must be a competent decision maker and a legitimate authoriser.

Question 44

How could voluntariness be compromised in clinical trials?

Answer 44

Coercion or perceived coercion or manipulation. All invalidate consent, the person should not be under any influence when making the decision.

Question 45

What do the research ethics committee focus on?

Answer 45

Scientific design and conduct of studies, recruitment of participants, care and protection of participants, protection of participants’ confidentiality, informed consent process and community considerations.

Question 46

Define absolute risk.

Answer 46

The likelihood of an effect compared to no effect.