Intro To Epidemiological Study Designs

Question 1

Descriptive Epidemiology

Answer 1

Sampling so that information can infer back to the population

Rarely will ever achieve a perfect sample

Question 2

Analytical Epidemiology

Answer 2

Compares like for like samples e.g. between sample from population A and population B

Again - rarely gain exactly “like for like” samples

Question 3

Epidemiological study designs

Answer 3

Ecological studies
Cross - sectional surveys

Case control studies
Cohort studies

Question 4

Populations concept of ecological study

Answer 4

Can be a series of studies

Identify groups to study (splitting up your population into equal groups

Define the characteristics to be studied (i.e. exposure and outcome of this exposure)

Decide whether analysis is to involve -
Counting categorical data (using a nominal or ordinal scale) or
Measuring continuous date (so using intervals or a ratio scale)

Issues for ecological study: - definition of characteristics is an issue, measurement variation, chance (random error) and confounding factors (falsely inferring individual-level association from group level)

Question 5

Population concept of cross-sectional surgery

Answer 5

Need to consider
The theoretical population - i.e. who do you want to generalise to?
The study population - i.e. what population can you get access to?
The sampling frame - i.e. how can you get access to them>
The sample - i.e. Who is in your study

Issues for a cross sectional study:
Sampling bias
Responder/Participant bias
Chance (random error)

Question 6

Population concept of a case-control study

Answer 6

Always retrospective

Identify a group of cases (with disease)
Identify a suitable group of non-cases (non diseased controls)

Assess the past exposure status’s of everyone

Compare the level or exposure in cases and controls and see if there is a pattern for cases experiencing similar things e.g. group of people with lung cancer in a certain area - large proportion of this group worked in mines

Issues: - Selection bias (c
Intro should reflect study population and should be comparable to cases
Information bias (especially for exposure) (look out for differential misclassification and non-differential misclassification)
Confounding factors
Chance (random error)

Question 7

Population concept of a cohort study

Answer 7

This can be historical or retrospective

Find exposed people and unexposed people and compare and contrast these 2 different groups for potential outcomes

Ascertain outcomes for everyone and compare incidence rates for each exposure group

Analysis can use Odds ratio or rate ratio

Issues: -
Loss to follow-up – differential loss (people could move away) – survivor bias (those that survive can be checked up on, those that dont cant)
Information bias, especially for outcome
– differential misclassification – non-differential misclassification
Confounding variables
Chance (Random error)

Question 8

Epidemiological study design - analysis

Answer 8

Descriptive epidemiology study designs:
ecological study: unit of analysis is groups
cross-sectional survey: unit of analysis is individuals

Analytical epidemiology study designs:
case-control study: analysis only odds ratio
cohort study: analysis can be rate or odds ratio