STUDY Designs Flashcards
What are the 4 Study designs
Case-control
Cohort
Descriptive
Experimental
purpose of analytical studies?
Designed to test hypotheses
Quantify risks
What is a case control study
Analytical and Observational study
Participants selected based on having the disease (cases) and not having the disease (control)
Attempting to identify differences between groups to identify possible causes for the disease
how are potential exposures found in case-control studies
If a certain exposure is more common in cases than controls → it may be the causing exposure (e.g. smoking habits)
What Measure of Association is used to quantify risk in Case-control studies
Odds Ratio
→ used as an estimate of relative risk in case-control studies
→ Can’t calculate rates in case-control (RR) as CC studies only provides the prevalence of exposures
What is the Odds Ratio and how is it calculated?
The odds ratio tells us how many times higher (or lower if below 1) is the risk of exposed group developing the disease compared to the non-exposed group
OR = (A * D) / (B * C)
What is Relative Risk and how is it calculated?
Is a ratio between the risk of getting disease in the EXPOSED GROUP compared to the risk of getting the disease in the NON-EXPOSED GROUP.
RR = Risk of disease in Exposed / Risk of disease in NON-Exposed
Advantages of a Case-control Study?
More efficient than a cohort study (time, money and effort wise)
Suited for diseases with long latent period (outcome occurs later in life e.g. 50s)
Good for rare diseases
Can examine multiple exposures
Analytical (tests hypothesis)
Cheap and quick (relatively)
Disadvatages of a Case-control Study?
Exposure assessed after development of outcome - (recall bias)
Prone to selection bias in control group
Usually limited to one disease or outcome per study
Inefficient for rare exposures
Confounding: 3rd factor which distorts relationship between exposure and disease
Time sequence → looking back for exposure after outcome has occurred
No direct measure of risk
Method for increasing POWER of a Case-control Study?
increasing the sample size by increasing the number of controls per CASE - 1:4 usually being the highest
What is a cross-sectional study?
Examines association between health conditions and other variables of interest as they exist in a defined population at one particular time.
- Descriptive, but can be analytical (to test hypothesis)
- Analytical cross-section studies require information about the correct time sequence (exposure precedes outcome)
- Collects data on exposure and outcome (both also assessed at the same time)
- Limitation of “prevalence pool” being affected by incidence, survival and mortality
- Inappropriate for assessing conditions of short duration
Sample selection methods for cross-sectional study
Non-probability Sampling
→ Asking for volunteers (or other methods)
Probability Sampling
Simple Random Sample → each individual has equal chance of selection
Systematic Sample → selecting at intervals (e.g. every 10th name)
Stratified Random Sample → random sample taken from each stratum (e.g. age or ethnic groups)
Cluster Sample → Random sample of groups instead of individuals
why is it important to follow-up participants and have high response rates?
non-respondents = lost information
can increase response rate by using phone calls, questionnaires, letters and media coverage
Advantages of cross-sectional study
Quick
Cheap
Representative samples
Service planning
Generate hypothesis (or test)
Disadvantages of cross-sectional study
Time sequence (does exposure precede outcome?)
Difficult to control for confounding
Inefficient for rare exposure
