Cohort Study Designs Flashcards
What can give the strongest level of evidence in the observational studies?
Cohort
What is group-allocation based on?
1) Exposure status - yes or no 2) Group membership (start off with a group of ppl who have something in common – then would break them up) ex) COB class of 2020
What are cohort studies useful for?
Studying rare exposures
What do cohort studies generate?
Risk of dz/outcome –> RR
What are some reasons why choose to do a cohort study?
-Unable to force group allocation -limited resources - exposure of interest is RARE - more interested in incidence rates or risks (than effects of interventions)
What are prospective cohort studies?
Exposure group selected on basis of past or current exposure and both groups followed into future to assess for outcomes of interest (which haven’t occurred) and then compared
- at time of start of study the outcome hasn’t occurred yet
What are retrospective cohort studies?
At start of study both exposure and outcome of interest have already occurred, but groups still allocated based on past hx of exposure
- exposure occurs before outcome of interest
Ambidirectional cohort studies:
use retrospective design to assess past differences BUT also adds future data collected on additional outcomes prospectively from start of study
- looking for outcomes in the past, and known in the present, but also into the future
1) A cohort can be defined as:
- Birth cohort =
- Inception cohort =
- Exposure cohort =
1) A group with something in common
- assembled ppl born at same time in same region
- assembled based on a common factor (ex = nurses cohort) – group of ppl who simply have something in common
- assembled based on common exposure (ex = 9/11) they were present at the moment of an event
What is a 1) fixed cohort, 2) closed cohort, 3) open cohort?
1) can’t gain members but CAN have LTFUs- die
2) not adding or losing anyone (short time period) (a fixed cohort with no LTFUs)
3) ppl can come and go (ex = Nurse’s Health Study)
How do you select for an exposed study population?
Allocate subjects based on pre-defined criteria of exposure
What are the 3 ways to select for the unexposed study population?
Which is the most reliable and which is the least reliable?
1) Internal - best option – same cohort, yet unexposed
2) General population
3) Comparison cohort – attempt to match groups as closse as possible on numerous characteristics
Most reliable = internal (most similar - same cohort, they’re just unexposed)
Least = comparison cohort
What are general strengths and weaknesses for Cohort study designs?
Strengths:
- assessing multiple outcomes for one exposure
- when exposures are rare
- Risk and RR
- less expensive
Weaknesses: -
- cannot demonstrate causation
- hard to control for other exposures (primarily in retrospective)
- can’t control for other exposures or pot changes in amount of study exposure ( primarily retrospective)
What are strengths and weaknesses for Prospective Cohort study designs?
Strengths:
- increased amount of important info from patients
- more control over data collection
- F/u tracking pt’s is easier
- answer to temporality
- multiple outcomes from a single exposure
- can calculate incidence and RR
Weaknesses:
- time, expensive, LTFUs
- not efficient for rare dz’s ****** ( think case control for these)
- not suited for long induction/latency period
- exposure may change over time
What are strengths and weaknesses for retrospective Cohort study designs?
Strengths:
- good for long induction periods
- can study rare exposures ***
- good if data already exists
- less expensive and time consuming
Weaknesses:
- possible incomplete data or different data collection completely
- can be impacted by recall or selection biases
- possible other exposures during that time
- requires access to charts, databases, and employment records
- pt may not be available for interview
- exposure or its amount may have changed over time
