Epidemiology Study Designs Flashcards
We use samples to infer populations. What are they samples affected by?
Demographic shape
Economic composition
Behavioural and lifestyle factors
What’s the difference between incidence and prevalence?
Incidence = number of new cases Prevalence = number of people who currently have the disease
What is a case controlled study?
Comparing a sample of people with the disease with a control group.
You look back retrospectively and try and find a relationship between risk factor and disease.
Measuring exposure
What is a cohort study?
A a group of people who have the disease and watch them over a period of time.
Observational not interventional
A prospective study - planned in advance
What is a cross sectional study?
An observational study which looks at a population at a specific tie and place e.g. Prevalence of disease in an area.
What is a sample average?
A mean of a sample
What is a 95% confidence interval?
You are likely to contain the mean of the population values 95% of the time
OR the range in which you are 95% sure the mean of the population values lie
What two factors widen the confidence interval?
Greater variation
A smaller sample size
What is inference?
Drawing a conclusion about a population from a sample.
What is selection bias? Give an example.
When the population selection is not representative.
Participants drawn from both samples aren’t representative of the population
Participants of one of groups are not representative of that sampling frame.
Groups being compared are not of the same population.
Name three types of errrors.
Recall error
Observer/interviewer error
Measurement error
What is information bias?
An error what dealing with study information.
Recall, observer/interviewer, measurement
What is a confounding factor?
When the experimenter cannot eliminate other explanation for an observed relationship.
A DOUBT
What is the difference between direct standardisation and indirect standardisation?
Direct - age gender standardisation e.g. Age specific rates of mortality - preferred.
Indirect - total number of deaths - good for small numbers in a specific age range.