Clinical Trials Flashcards
What is a clinical trial?
A clinical trial is any research study that prospectively assigns human participants to one or more health-related interventions to evaluate the effects on health outcomes. Otherwise known as interventional trials.
Describe the 4 trial phases of clinical trials
Trial Phase I (10-30 participants): test if drug is safe for use
Trial Phase II (30-100 participants): test if drug is efficacious
Trial Phase III (>100 participants): test for safety and efficacy on a larger group of patients
Trial Phase IV (thousands of patients): long-term safety and efficacy evaluation in a real world context
What is the hierarchy of evidence?
A pyramid where the type of studies at the bottom are least suited as direct evidence to make clinical decisions, while the type of studies at the top are most suited to make clinical decisions.
Bottom: in-vitro and animal research
Top: Randomised controlled double-blind trials, systematic reviews and meta-analyses
Difference between explanatory and pragmatic trials
- Explanatory trials measure efficacy - the benefit of a treatment under ideal conditions.
- Pragmatic trials measure effectiveness - the benefit of a treatment in routine clinical practice.
Validity: explanatory trials have high internal validity, pragmatic trials have high external validity
Sample size: explanatory trials have smaller sample size, pragmatic trials have larger sample size
Design: explanatory trials have sophisticated design, pragmatic trials have simple design
Environment: explanatory trials are conducted in a controlled environment, pragmatic trials are conducted in diverse settings.
Phases: explanatory trials are mostly phase I-III, pragmatic trials are mostly phase IV.
Superiority vs equivalence vs non-inferiority trials
Superiority trials: aim to determine whether either of two treatments is superior to the other [two-sided test].
Equivalence trials: aim to determine whether one treatment is similar/equivalent to another within a predefined acceptable margin [two-sided test].
Non-inferiority trials: aim to determine whether one treatment is not worse than another treatment by more than a predefined acceptable margin [one-sided test].
What are randomised controlled trials (RCT)?
- Randomized control trial (RCT) is a study in which people are allocated at random to receive clinical intervention(s). One of these interventions is the standard of comparison or control.
- Regarded as the ‘gold standard’ for evaluating clinical interventions.
Strengths of randomisation in RCT
- Ensures that all participants have the same chance of being assigned to each of the study groups [i.e. eliminates selection bias].
- Ensures that all baseline characteristics (both known and unknown factors), are, on average, distributed equally among the study groups. [i.e. eliminates confounding]
» Thus, any significant differences between study groups in the measured outcome can be attributed to the intervention and not to some other factors.
What are the two important features for successful randomization?
1). Complete unpredictability of intervention assignment
- Preferred approach: Simple randomization
» First, generate a list of random numbers (e.g. using Excel)
» Come up with a system for categorization (assign odd numbers to Group A and even numbers to Group B (for randomization to 2 groups).
- Blocked randomization is a commonly used technique to ensure that the number of participants is equally distributed among study groups.
» First, decide on the block size (e.g. block size of 4).
» Next, list down all possible permutations to randomize participants into 2 groups.
» Then, randomly select a list of permutations (e.g. by generating a list of random numbers using excel), and assign participants to groups based on the random numbers.
- Allocation concealment
- Assignments are tamperproof such that the person enrolling participants does not know in advance which intervention the next person will get
What are the different RCT designs?
- Parallel design:
» Most common and simple design
» Each participant is exposed only to one of the interventions. - Crossover design:
» Each participant is given all of the interventions in successive periods
» Order of interventions is determined at random
» Washout period in between to prevent carryover effect
» Minimizes between-subject variability
» Reduces sample size required
» Appropriate only for:
»> Interventions that have a short-term effect
»> Chronic conditions that are fairly stable over time - Factorial Design
» Used to evaluate 2 or more interventions not only separately but also in combination and against control.
» Allows evaluation of interaction between interventions.
Differentiate between allocation concealment and blinding
Allocation concealment
- Assignments are tamperproof such that the person enrolling participants does not know in advance which intervention the next person will get. Usually done before/along with randomization.
Blinding
- Aka masking
- Refers to withholding information about the assigned interventions from people involved in the trial who may potentially be influenced by this knowledge. Usually done after randomization.
- Minimizes problems of compliance and information bias (observation bias).
- Level of blinding:
» Open or open label – no blinding
» Single blind – participants blinded
» Double blind – participants and investigators (who are usually also the outcome assessors) blinded
» Triple blind – participants, investigators and data analysts blinded
Discuss problems posed by loss to follow-up
- Losses can:
» Affect internal validity
» Introduce bias (esp. for differential rates of loss)
» Decrease power
Describe and apply the different data analysis approaches in RCTs
1). Primary analysis
- Intention-to-treat (ITT) analysis
» “Once randomized, always analyze as assigned”.
» Compares outcomes between study groups with every participant analyzed according to his randomized group assignment, regardless of whether he adhered to the assigned intervention.
» Preserves comparability
- Secondary analysis
A. Per protocol (PP) analysis
»Analysis only includes participants who fully comply with the trial protocol.
B. As-treated analysis
» Participants analyzed according to treatment received instead of treatment assigned.
Ethical considerations of clinical trials
1) . Clinical equipoise
- The assumption that there is not one ‘better’ intervention present (for either the control or experimental group) during the design of a RCT.
- A true state of equipoise exists when one has no good basis for a choice between two or more care options.
- The assumption of a state of uncertainty has been identified as the central ethical principle for human experimentation.
2). Data and Safety Monitoring Board (DSMB):
- Independent group with expertise in various disciplines
- Charge to safeguard participants in the trial, so as to ensure integrity of trial and to monitor for unexpected benefit or harm.
- Reviews progress of trial and unblinded data on outcomes
- Can recommend modification or termination of a component of trial based on:
» Information from trial itself
» From other trials
» New basic science information
Regulatory considerations of clinical trials
- Health Science Authority (HSA):
» Clinical Trials Branch of HSA has regulatory oversight of clinical trials on Therapeutic Products and Medicinal Products
»> Health Products (Clinical Trials) Regulations 2016
»> Medicines (Clinical Trials) Regulations 2016
Registration considerations of clinical trials
- Different registers available for researchers from all around the world to register their study and to find active clinical trials.
- Singapore: HSA Clinical Trials Register
