Cohort studies Flashcards
What is a cohort?
Defined group of people with known exposure
What two ways can you compare the cohort?
Subdividing the level of exposure and comparing (IRR)
Comparing with external reference pop (SMR)
What can be difficulties with calculating SMR?
Could have limited data in reference pop, eg no incidence data or morbidity rates.
Could have a healthy worker effect (selection bias)
What is the precision measured by?
Width of CI, based on ef
What is the sequence for a cohort study?
Recruit disease free individual. Follow individuals over time until end if study period. Analyse and interpret results.
What happens in a retrospective study?
Recruit historically disease free individuals using existing records.
What are person years?
Sum of the total number of years of everyone followed up in the study, even if they come in and out.
Why might you not collect data straight away?
To check you have recruited disease free individuals and not just people who havent been diagnosed yet.
What are the advantages of a cohort study over case-based?
Can study a range if different outcomes.
Better to study a rare exposure as can select people with it.
Better at establishing exposure precedes outcome.
What do you start with in a cohort study?
Disease free individuals
Disadvantages of cohort studies
Large and resource intensive.
Long time (historical less so).
Definitions of outcome and exposure can require expensive and intensive investigation.
Survivor bias - risk of high number of losses in follow up as people may leave the study etc.
Results can take a long time.
Not good for rare outcomes.
Difficulty with confounding.
