Give 6 types of study design
Cohort study, case-control, ecological, cross-sectional, RCT, meta-analysis
describe the design of a cohort study
An observational study
Longitudinal study in similar groups but with different risk factors/treatments
Follow up over time (constantly checking in)
(in contrast to case-control, the outcome has not occurred yet)
Describe case-control studies
An observational study looking at the cause of a disease. Compares similar participants w/ disease and those without (controls). Looks RETROSPECTIVELY for exposure/cause (potential RF data is collected - e.g. questionnaires)
The outcome has already occurred at the time of investigation
important idea: looks backwards in time
describe ecological studies
Observational study of a disease or outcome and exposure of interest are measured in a number of populations/groups (e.g. between electoral wards or different hospitals)
describe a cross-sectional study
Observational study collecting data from a population at a specific point in time (no follow-up) - a SNAPSHOT STUDY
Example: the census (conducted every 10 years) –> other examples include prevalence at a specific time (e.g. number of smokers in a GP practice)
describe an RCT
Experimental study
Similar participants are randomly assigned to an intervention or control group to study effect of intervention
What are the advantages and disadvantages of an RCT?
Advantages:
- Low risk of bias and confounding
- Comparative
- randomised
- blind - reduces possibility of information bias
Disadvantages:
- High drop out rate
- Ethical issues
- Time consuming and expensive
- prior knowledge required
What are the advantages and disadvantages of an ecological study?
Advantages: can lookout geographical correlations, good for generating a hypothesis
Disadvantages:
- subject to bias + confounding
What are the advantages and disadvantages of a cohort study?
Advantages:
- Can follow up rare exposure
- Allows to identify risk factors
- Data on confounders collected prospectively
Disadvantages:
- Large sample size required
- Impractical for rare diseases
- Expensive
- People drop out
- Confounding/bias
What are the advantages and disadvantages of a case control study?
Advantages:
- Quick
- Good for rare outcomes
Disadvantages:
- Difficult finding appropriately matched controls
- Prone to selection and information bias (e.g. recall bias)
- dropout
- Reverse causality (as the outcome interest has already occurred)
n.b. all the observational studies (case control, ecological, cohort, cross-sectional) are subject to bias and confounding
What are the advantages and disadvantages of a cross-sectional study?
Advantages:
- Large sample size
- Provides data on prevalence of risk factors and disease
- Quick to carry out
- Repeated studies show changes over time
Disadvantages:
- Risk of reverse causality – which came first?
- Less likely to include those who recover quickly or short recovery
- Not useful for rare outcomes
- subject to bias + confounding
- Temporality: impossible to be certain whether an outcome developed before or after the exposure concerned
What is a meta-analysis?
An evidence synthesis where all the data of similar studies are pooled together and a forest plot is created
What is an ethical dilemma of RCTs?
is it ethical to withhold potentially life-saving treatment form the control group?
Give an example of a cross-sectional study
Census