Clinical Trials Flashcards
1
Q
Randomization Purpose
A
- Each patient has an equal chance to be allocated into any treatment group
- Control for known and unknown cofounders: related to prognosis
- Check if randomization worked to see if baseline characteristics are similar (Table 1)
2
Q
Simple Randomization
A
- Random number generator
- Flipping a coin
3
Q
Block Randomization
A
- Divide subjects into subgroups (“blocks”)
- Subjects within blocks are then randomly assigned to treatment groups
- Ensures equal number from each block into treatment groups
4
Q
Stratified Randomization
A
- Goal: assign all the other variables equally except for treatment
- Create predefined strata based on the prognostic factors (age, severity, etc)
- Final groups more balanced for cofounders
5
Q
Allocation Concealment
A
- No a priori knowledge of group assignment
- The allocation sequence is concealed from those assigning participants to intervention groups until moment of assignment
- Prevents selection bias
- Prevents researchers from influencing which participants get which intervention
6
Q
Blinding
A
- Hiding who gets what
- Minimize influence of pre-existing biases or expectations
- Blind patients AND investigators if their responses have a possible subjective component
- Blinding is more important as subjectivity increases
- Prevents ascertainment/information bias
- Not all studies are blinded or need to be blinded
7
Q
Blinding Schemes
A
- Open label: everybody knows who is getting what
- Single-blind: either the patient or the investigator doesn’t know
- Double-blind: both the patient and the investigator don’t know
- Triple-blind: the patient, investigators, AND data analysts don’t know
8
Q
Description of Blinding
A
- Mechanism should be included
- Similarity of treatment characteristics
- Allocation schedule
- How successful was the blinding (were people able to guess which group they were)
9
Q
Intervention Information
A
- Treatment
- Dosing: is it similar to customary
- Timing of efficacy assessment: given the amount of time needed to work
- Formulation: comparable to commercial product
10
Q
Selection of Control Groups
A
- Placebo control
- Active controls (standard of care)
- Historical (external) controls
11
Q
Placebo Control
A
- Commonly thought to provide a comparison to “doing nothing”
- Actually allows comparison with the act of drug administration
- Inert resembles the active drug
- Placebo therapeutic and adverse effects
12
Q
Active Treatment Controls
A
- Use when its inappropriate to withhold therapy
- Permits comparison to established therapies
- Is the established therapy the most appropriate and being used correctly (fair fight)
13
Q
Double-Dummy
A
- When treatments differ by route and dosing schedule
- Used to compare single v.s. combination therapy
- Groups: Dummy A + Drug B, Drug A + Drug B, Dummy A + Dummy B, Drug A + Dummy B
14
Q
Active Treatment Control Clinical Trial Types
A
- Superiority: is drug A better than B
- Equivalence: is A just as good as B
- Non-inferiority: is A no worse than B
- The definitions of better, worse, etc. need to be defined in advance
15
Q
Common RCT Study Designs
A
- Parallel study design
- Cross-over design: fewer subjects needed, includes washout period
- Factorial randomized trials: multiple dose levels/regimens
- Cluster randomization design: group of subjects is randomized
- Stepped Wedge Cluster: random and sequential crossover of clusters from control to intervention until all clusters are exposed
16
Q
Crossover Studies
A
- Study patients should return to pre-treatment status between periods
- Washout might be needed to dissipate treatment effects
- Good news: half as many patients needed (serve as own control)
- Bad news: twice as much time needed
17
Q
Result Interpretation
A
- Were there important differences between eligible, recruited, and enrolled patients
- Were the study groups similar at baseline (Table 1)
- Identify confounders
- Are the differences important
- Caution: hard to find differences in baseline when groups are small
18
Q
Results: Loss
A
- Are ALL patients accounted for including dropouts and dismissed patients
- Difference in dropout rates/reasoning between the groups
- 20% loss considered a threat to study validity
19
Q
Intention-to-treat
A
- Analyzes individual patient outcomes according to group they were assigned
- Regardless of whether they finished the trial or even received any drug
20
Q
Per-protocol
A
- Analyzes the outcomes of individual patients who finished the trial
- According to treatment they received regardless of what group they were assigned to
21
Q
Reporting Results
A
- Outcomes reported the same as those in the objective
- Miss any important outcomes
- Components of composite also reported separately
- How large was the treatment effect and if it was important
- Results reported in actual or relative change
- Survival analysis shows the development of the outcome over time
22
Q
Assessment of Adherence
A
- Pill counts
- Patient diaries
- Serum or urin drug concentrations
- Refill frequencies
- Medication containers that record opening (MEMS)
23
Q
Descriptors of Risk
A
- AR: Absolute Risk
- RR: Relative Risk
- RRR: Relative Risk Reduction
- ARR: Absolute Risk Reduction (attributable risk or risk difference)
- NNT: Number Needed To Treat
- NNH: Number Needed to Harm
24
Q
AR
A
- Probability of an event
- Could represent incidence or prevalence
- Incidence (exposed): Successful Treatment/Total Treatment
25
Q
RR
A
- Risk in one group compared with the risk in another
- AR (treatment)/AR (control)
- _X greater relative risk of a positive outcome compared to control
26
Q
AAR
A
- Difference in absolute risks
- AR (treatment) - AR (control)
- Receiving treatment prevented _% of treated patients from experiencing unsuccessful outcome
27
Q
NNT
A
- Reciprocal of ARR
- 1/ARR
- About _ patients in this study had to receive treatment instead of control to avoid one adverse outcome
OR
-About one in _ patients will achieve successful outcome by using the treatment instead of the control
28
Q
NNH
A
- How many people need to be treated in order for one person to have a particular adverse effect
- EX: 1 of 138 had a particular AE, then NNH is 138
