CAT stuff part 2: Return of the MEEEEOOOOOW Flashcards
What is the most appropriate study design the investigating diagnosis?
Cross sectional analytic study
What is a cross sectional analytic study?
An observational study that analyses data from a population at a specific time period
What are the most appropriate study designs for investigating aetiology?
- Cohort study
- Case-control study
What is a cohort study?
A longitudinal study that follows a population to observe outcomes, often in the context of one group having a particular exposure (e.g. smoking) and one group not
What is a case-control study?
A retrospective population study in which 2 groups with different outcomes are identified and compared on the basis of some supposed causal attribution
What is the most appropriate study design for prognosis?
Cohort study
What is the most appropriate study design when investigating treatment?
- Randomised control trial
- Systematic review of RCTs
What is the most appropriate study design for evaluation?
Systematic review with meta-analysis
What are some advantages of cohort studies?
- Best information about causation
- Able to examine a range of outcomes
- Good for rare exposure
What are some disadvantages of cohort studies?
- Long follow up so expensive and time consuming
- Bad for rare outcomes
- Bad for long latency periods
- Can have different follow up for exposed and non-exposed
- Confounders not recognised
What are some advantages of a case-control study?
- Don’t require long latency periods so quick and quite easy to carry out
- Good for long latency periods
- Good for rare outcomes
What are some disadvantages to case-control studies?
- Inferior to cohort studies
- Bad for rare exposure
- Controls may not represent where sample is from = often requires several controls per case
- Cases don’t recognise the full disease spectrum
- Particularly vulnerable to recall bias
- Confounders not recognised
What is procedure bias?
When subjects in different groups receiving different treatment other than the intervention
What is information bias?
When information about about participants is incorrect
What are 3 types of randomisation?
- Simple randomisation
- Stratified randomisation
- Minimisation
What is an example of simple randomisation?
Computer random number generator
What is stratified randomisation?
- Randomisation that ensures different groups have equal numbers of disease/equal levels of disease severity in each group
- Essentially randomisation that takes possible confounders into account and distributes them evenly among the groups
What is minimisation and when is it used?
- A computer algorithm forming groups to ensure comparability
- Used in smaller studies
What is the difference between precision and accuracy in stats?
- Precision: narrow CIs
- Accuracy: Result is close to real value
What is treatment fidelity?
How well a treatment is reproduced from a protocol or model
What is the purpose of randomisation?
To ensure any confounders are equally distributed among study groups, thus avoiding selection bias
What is internal validity?
How well a study was conducted, taking into account confounders and removing bias
What is external validity?
- Generalisability
- How well a study can be applied to different real world situations/patients/environments
What is association in relation to variables?
2 variables are said to be associated when one is found more commonly in the presence of the other
What are the 3 types of association and describe them?
- Spurious: association has arisen purely by chance and not a real association
- Indirect: association due to the presence of a third factor (confounding factor)
- Direct: a true association not affected by a confounding factor
What is used to assess whether or not an association is causal?
The Bradford-Hill Criteria