Experimental Clinical studies Flashcards
Draw a table suggesting the likelihood of problems with each of the main three types of studies. Include Bias, Confounding and Random Error
Give four examples of suitable treatments to test using a RCT, and 2 that would be more difficult.
Name 5 key useful features an RCT could have to ensure accuracy.
Controlled (placebo controlled)
Randomized
Double blind
Large
Analysed by `intention to treat’ method
Name 3 common reasons a patient’s condition can improve in the absence of a specific treatment.
Natural tendency to recover –> biological healing and repair (especially with good nursing care)
Probability – individuals at the extreme of a distribution tend to come towards the mean - `regression to the mean’
The placebo effect – when people believe that they are receiving a treatment (even if actually neutral) their recovery is improved
Why is it not advisable to use a control group of historical case?
Because outcomes often change (tend to improve) with time….
What are three different options of how to treat your control group?
CONTROL groups –
- Receiving no active treatment
- Receiving USUAL CARE (i.e. existing pattern of treatment, but without the new intervention being tested)
- Either of these categories can also be receiving PLACEBO for the active intervention under test
Outline the importance of randomisation within the allocation process. Specifically which type of bias does it directly inhibit?
The best way of ensuring that the characteristics of patients in the INTERVENTION and CONTROL groups are similar
Specifically avoids allocation bias
Indirectly avoids…
Selection bias and confounding (problem in CCS and CS) are excluded if the study is well designed
Randomization also facilitates the process of `blinding’ which reduces the risk of information bias
What is the difference between Selection and Allocation Bias?
Selection bias occurs if the selected sample is not representative of the patient population.
If there is a systematic difference between participants in how they are assigned to treatment groups, then it is referred to as Allocation Bias.
This is minimised by the use of random allocation.
NB: The randomisation of patients to treatment groups will not affect selection bias.
Why is blinding important to the assessment process in an RCT?
- There are always vested interests in the outcome of trial!
- Objective outcomes (e.g. death) are pretty robust, however problems with subjective, patient assessed outcomes (`I feel so much better on this new treatment doctor…’) can lead to – ASSESSMENT BIAS
- This can also happen with observer assessed outcomes (e.g. reported `health’, blood pressure) – ASSESSMENT BIAS
Explain the difference between a Single, Double and Triple Blind study
Patient OR outcome assessor blind = single blind
Patient AND outcome assessor blind = double blind
Patient + outcome assessor + statistician = triple blind
What are two options to minimise assessment bias if the study is not Triple Blind?
- Statistician can carry out analysis blind to randomisation code
- Trial analyses are pre-specified in great detail, then less need for blinding
What is a crossover study? And what are it’s advantages?
Crossover Study
A longitudinal study in which subjects receive a sequence of different treatments (or exposures).
Nearly all crossover are designed to have “balance”, whereby all subjects receive the same number of treatments and participate for the same number of periods.
In most crossover trials each subject receives all treatments, in a random order.
Advantages
- Very efficient trial design because every participant acts as both intervention and control
- For a given number of participants, get a more precise estimate of effect than with parallel group
- Only works for RAPIDLY ACHIEVED and REVERSIBLE outcomes (e.g blood pressure)
- NOT for irreversible events (e.g development of cancer, diabetes, death)
What is a factorial study?
Involves 2 interventions (e.g. blood pressure lowering A and lipid lowering treatments B)
4 equal sized randomized groups
- None (control group)
- A
- B
- A + B (dual intervention)
Efficient design, because all groups involved in examining both the effect of A and the effect of B
What is a type 2 error?
A type II error is a statistical term used within the context of hypothesis testing that describes the error that occurs when one accepts a null hypothesis that is actually false. A type II error produces a false negative, also known as an error of omission.
What are the advantages of a larger sample size?
- Higher statistical power (likelihood of detecting a true intervention effect when present)
- More precise estimate of effect size
- Ability to look at the impact of the intervention in a range of different subgroups (different age-groups, genders, disease characteristics)
N.B. Does NOT make trial more representative of the population at large (that depends on the source of the trial population)
