Topic 18: interventional studies Flashcards
(28 cards)
1
Q
Describe intervention studies
A
- Investigator controls exposure = type + amount + duration + who receives it
- Attributes cause + effect
2
Q
Why is intervention studies regarded as vigorous?
A
- Random = reduces confounding bias
- Concealment = reduces selection bias
- Blinding = reduces information bias
3
Q
What is the function of a trial?
A
- Test efficacy of intervention + assess safety
4
Q
Define prophylactic trials
A
- Evaluate efficacy of intervention to prevent disease
- E.g. vaccine/vitimin
5
Q
Define treatment trials
A
- Evaluate efficacy of curative drugs/intervention + drug to manage symptoms
6
Q
Define RCT trials
A
- Individuals
- Tighly controlled
- Highly selected
- Short/long period
7
Q
Define community/cluster trails
A
- Cities/regions/schools/hospitals
- Less controlled
- Long duration
- Primary prevention
8
Q
Describe phase 1 of clinical trial
A
- Researchers test new drug treatment = small group
- Evaluate safety
- Determine safe dosage range
- Identify side effects
- No control
9
Q
Describe phase 2 of clinical trials
A
- Larger group
- See if effective
- Further evaluate safety
- No control group
10
Q
Describe phase 3 of clinical trials
A
- Given to larger group = confirm effectiveness
- Monitor side effects
- Compare to common treatment/placebo
11
Q
Describe phase 4 of clinical trials
A
- Studies after drug marketed = gather information on effect in various populations
- Side effects of long-term use
12
Q
Describe randomized controlled trials
A
- Participants assigned study group
- Procedures controlled = all treated same
- Assigned treatment condition at random
13
Q
Describe randomization
A
- Random assignment ensures anything that could distort outcome = evenly distributed = control confounders
14
Q
Describe allocation concealment
A
- Person generating random assignment blind to condition person is entering
- Stops staff assigning better patient to intervention than control = bias treatment
- Before intervention starts
15
Q
Describe blinding
A
- Researchers collecting data = prevented from knowing details about patient
- Minimize information bias = both patient + researcher blind to participant’s assignment
- During intervention
16
Q
Describe placebo control in RCT
A
- Placebo identical to intervention
- No therapeutic ingredient but still placebo effect
- Placebo effect = psychological effect of getting treatment
- These trails = show effectiveness of intervention beyond placebo effects
17
Q
Describe double blinded trails
A
- Reduces influence of expectation by participant + staff
- Best for trial = effect of medicine compared to similar looking placebo
- Reduce bias
18
Q
Give the control groups of RCT
A
- New treatment vs old = old is control
- Different dosages = lowest is control
19
Q
Describe partial blinding
A
- If assignment is known to research staff deliverin treatment = keep staff assessing study outcome blind to assignment
20
Q
Describe unblinded trials
A
- When neither can be blinded = best if both hold positive expectations
21
Q
What problems arise with loss of follow-up + non-compliance + contamination?
A
- Major bias
- Decreased power
- Loss of credibility
22
Q
Why is there loss of follow-up?
A
- Side effects
- Moved
- Died
- Recovered
- Lost interest
23
Q
Why is there poor compliance?
A
- Side effects
- Iatrogenic reaction
- Recovered
- Lost interest
24
Q
How to maximize follow-up + compliance?
A
- 2 screening visits prior to enrolling
- Pre-randomization run-in period using placebo/active
- Maintain blinding
25
Give the advantages/disadvantages of cross-over design
ADVANTAGES:
- Each subject = own control
- Smaller number of pateints needed compared to parallel
DISADVANTAGES:
- Longer than parallel
- Difficulty for multiple dosage
- Drop outs
26
Describe quasi-experimental design
- Participants assigned using non-random
27
Give the advantages of RCT design
- Measures efficacy + safety
- High internal validity = can control selection + biases
28
Give the disadvanatges of RCT
- Time + energy intensive
- Expensive
- Not feasible for all interventions
- Not good for rare outcomes
- Validity depends on how well design + conducted + reported
