Block 3 Week 3 - RCT

Question 1

What are the lessons we have learnt from history about how we should conduct trials?

Answer 1

1) Collect data- don’t rely on anecdotal
2) The need for Controls
3) Avoid bias

Question 2

What may the differences in outcomes in Tx groups be due to?

How must the controled and the trial group differ?

Why must the sample be sufficiently large

Answer 2

Treatment effect, Bias or Chance

2) Must be alike in every possible way except the Tx
3) To avoid chance imbalance

Question 3

What is a controlled clinical trial?

Answer 3

Prospective study comparing the effects & value of an intervention against a control in human subjects

Question 4

What is an uncontrolled trial?

Answer 4

Trial involving no control group

Question 5

What is an Randomised Clinical Trial?

What are the advantages

Answer 5

Contolled clinical trial where the therapies are allocated by a chance mechanism- neither patient or physician know which therapy they will revieve in advance

Advanatage:

a) Limit selection bias
b) Balance prognostic factors
c) Validity of satistical tests

Question 6

How are clinical trials classified by design in order of weakest strenght - most strength

Answer 6

a) Uncontrolled trials
b) Historical controls (the comparison group is running concurrently to the treatment group- comparison between two different times)
c) Current non-randomised controlled
d) Randomised controlled

Overall- Improved design reduced bias

Question 7

What is the concept of statistical inference?

What are its aims

Answer 7

Analysed data results in a selected sample are used to infer how all the population would behave

Population- people going into the study → Protocol Treatment Patients- are people fit to go through the process → Observed Result → Extrapolate

Estimate treatment effect (bias beware)

Aware of: Variabilty of estimate due to samplaing from population (How may population variabilty affect the study)

Goal: Control bias & reduced variability

Question 8

Randomised allocation means……

Answer 8

a) Gives an equal chance of receiving each Tx
b) In long run leads to groups that are likely to be similar in characteristics by chance
c) Reduced selection bias if patients enter trial before randomisation
d) is used in other expirimental settings

Question 9

What is the placebo effect?

Answer 9

Even if the therapy is irrelavent to the patient’s condition, the patient’s atttidue, to their illness & the illness may be improved by a feeling that something is being done about it

Question 10

If you are comparing a group with Tx and a group without Tx what could the results be down to?

Answer 10

True tx effect

OR

Placebo effect

Question 11

What are the two types of blind trial?

What is the aim of this?

Answer 11

Single Blind Trial: Either the patient, clinician or assessor does not the Tx allocation

Double Blind Trial: Two or more of the patient, clinician or assessor do not know the Tx allocation

Aim: Reduce placebo effect that could bias comparisons between Tx

Question 12

What is the Outcome evaluation?

Answer 12

a) If there is a difference between groups
b) Could the difference have arisen by chance- is it statistically significant
c) How big is the difference- is it clinicall relavent
d) If the difference is attributable to the Tx

Question 13

Non-compliance:

Why may participants not complete the trials

Why doesn’t everyone comply with their allocated Tx?

Answer 13

A) Due to clinical condition necessitating their removal

OR

Want to withdraw

B) Misunderstood the instructions

Don’t like taking the Tx

Don’t feel the Tx is working

Question 14

Draw a diagram showing from randomisation how a clinical trial is conducted

Answer 14

-

Question 15

What are the 2 different interpretiations of the new treatment?

Answer 15

1) Is the Physiological Action better (drug effects)

OR

2) Is the new tx better than the standard tx in Clinical Practice (tx package)

Question 16

What is an Explanatory trial- ‘As treated analysis’?

What does it do & what are the pro/cons?

Answer 16

Analyses only those who completed the follow-ups & complied with the trial

Compares the physiological effects of Tx

b) Looses the effect of randomisation: non-compliers are likely to be systemically different from compliers- selection bias/ confounding

Question 17

What is an Pragmatic trial- ‘Intention to treat analysis’?

Answer 17

-

Question 18

Explanatory Trials vs Pragmatic Trials

Which should difinitive clincal trials normally be analysed on?

Answer 18

Explanatory analyses: tend to give larger sizes of effect

Pragmatic analyses: tend to give smaller sized of effect & reflect effect in clincal practice

b) Pragmatic

Question 19

What is the ethical dilemma of clincal trials?

Answer 19

Clinicians should provide best Tx to patients

BUT

Scientific enquiry require Tx to be chosen randomly

Question 20

What is clinical equipoise?

Answer 20

Reasonable uncertainty about which Tx (including non-Tx) is better

Thus randomisation does not deny any patient the best Tx

Question 21

Infromed consent in trials

Explanation & information to be given

Answer 21

Explanation:

a) Patient is invited to trial
b) What the alternative Tx are including SE
c) Tx allocation will be random
d) Choice to withdraw

Information given:

Verbally and in writing with ‘cooling off’ time

By knowledgeable informant

Question 22

What is the difference between standard randomisation and Zelen method?

Answer 22

Zelen: Randomisation gives experimental & non-experimental group

Non experimental: No consent needed & standard Tx given

Experimental group: Consent to reciving the experimental treatment ONLY. If decline they go onto standard Tx