Lecture #10 - Cohort Studies Flashcards
Group assignments in Cohort Study designs are based on:
GROUP MEMBERSHIP or EXPOSURE
True or False: Cohort studies are observational, descriptive studies
False: Observational, analytical
In a 2x2 table for a cohort study, the columns and rows are labled what, AND which is known at the start of the study?
Column: Disease
Row: Exposure
KNOW EXPOSURE AT START (beacuse thats how you allocated groups)
Pro’s of cohort studies include:
o Unable to “randomize”
Impossible – genetic traits
Unethical/Illegal – effect of cocaine use during pregnancy (self select then observe in natural setting)
o Limited resources
A cohort is defined as:
a group with something in common
Birth cohort refers to:
• Individuals assembled based on being born in a
geographic region in a given time period
Inception cohort refers to:
• Individuals assembled at a given point based on some common factor
• Where people live or where they work, or something they
have in common
• Useful for single‐group non‐comparisons for incidence rate
determination
Example: The Framington study (1948 study of town)
Exposure cohort refers to:
- Individuals assembled based on some common exposure
- Frequency connected to environmental or other events
Example: 9/11 Firefighters, Hiroshima, etc
Fixed Cohort refers to:
A cohort (derived from an irrevocable event) which can’t gain members but can have loss‐to‐follow‐ups
Closed Cohort refers to:
A Fixed cohort with no loss‐to‐follow‐ups
Open (or Dynamic) Cohort refers to:
A cohort with new additions and some loss‐to‐follow‐ups
During Hiroshima 1 million people were exposed to the bomb’s radiation. Forty years later, a study was done to determine how the radiation effected their bodies. The cohort studied consisted of 800,000 people. What type of cohort would this have been?
Fixed cohort - once specific event (can’t gain, but can lose to loss of follow up)
What type of study’s Main purpose is to follow (or assess) both exposed and unexposed groups for 1 or more outcomes of interest (disease or outcome/event of
interest)
Cohort Study
Incidence study / Follow‐up study / Longitudinal study are also known as
Cohort studies
2 Primary purposes of Cohort studies are:
o Descriptive (Measures of Frequency) Incidences / Incidence Rates
o Analytic (Measures of Association) Risk ratios & predictive risk‐factors associated with outcome
GO OVER (but referred to as analytical?)
In which of the following ways can cohort studies be conducted?
Prospective
Retrospective
Ambidirectional
ALL
State the type of cohort study:
Exposure group is selected on the basis of a past or
current exposure and both groups (exposure and non‐
exposure) followed into the future to assess for
outcome(s) of interest (which has yet to occur), and
then compared
Prospective
State the type of cohort study:
At the start of the study, both the exposure and the outcome of interest have occurred
Start (historically) at time of exposure and follow forward to the point of outcome occurrence (known), in the present
Exposure still has to occur before outcome of interest
Retrospective
State the type of cohort study:
Uses retrospective design to assess past differences but
adds all data collected on additional outcomes
prospectively from start of study (into future)
Ambidirectional
True or False:
Cohort studies are good for assessing multiple outcomes of multiple exposures
False:
Good for assessing multiple outcomes of one exposure
Hard to control for other exposures if more than one
plausible for causing (or being associated with) an outcome
What are some disadvantages of prospective cohort studies?
o Time, Expense & Lost‐to‐follow‐up’s
o Not efficient for rare diseases - use Case‐Control study
o Not suited for long Induction/Latency conditions
o Exposure (or its “amount”) may change over time
ex: exposure to cell phone towers changes over time
What are some disadvantages of restrospective cohort studies?
o Requires access to charts, databases, employment records (may not be complete or thorough enough for your study)
o “Information” may not factor in other exposures to harmful elements
o Patients may not be available for interview if contact necessary for missing or incomplete data
o Exposure (or its “amount”) may have changed over time
Explain Issues Affecting Outcome Occurrence in Groups:
o Level of exposure
o Induction Period
o Latency Period
Level of exposure - example: smokers may smoke a pack a day or a single cigarette. Stratify refers to layering out exposure levels.
Induction period - interval between exposure which causes disease and onset of disease process (may not be diagnosable at this stage)
Latency period: Interval between disease-process onset and clinical diagnosis of disease.
The most important thing to consider when selecting unexposed for a study is to:
Make them AS CLOSE AS POSSIBLE, yet not exposed.
Name of the 3 groups that unexposed can come from and definitions:
Internal (best, if feasible) - Patients from the same “cohort”, yet who are unexposed (most similar)
• If there are only levels of exposure, you may have to use thelowest exposure group as comparator (if there is no “no”exposure group internally‐available)
General Population - used as a second choice when the best‐possible comparison group(internal) is not realistically possible
Comparison Cohort - This is the least acceptable group (but still can be utilized)
• Simply attempt to match groups as close as possible on numerous
personal characteristics (can’t control for other potentially harmful
exposures in comparison cohort; also causing disease)
Definition of Healthy-worker effect:
If healthy, you work (even if exposed). If too ill to work (due
to exposure?) you may be unemployed (now part of non‐
working general population). So a cohort of “working” population would be difficult to find worst cases.
Definition of Selection Bias in Cohort Studies:
How exposure status is defined/determined (less of an issue
with exposure status)
Classification issue (Misclassification)
Cohort Study general info:
Analysis of Cohort Study Data:
o Primary Objective:
Compare outcome(s) of interest between the Exposed and Un‐exposed groups
o Descriptive Comparisons used in Cohort Studies:
Absolute & Relative differences in Incidence/Risk & Incidence
Rates (between groups)
o Analytic Comparisons used in Cohort Studies:
Relative Risk (RR) (between groups comparison)
Statistical comparison (of RR and predictive factors)
General questions to ask of cohort studies:
A few questions to ask of Cohort studies:
o Was duration appropriate/adequate for outcome(s) of
interest?
o How as exposure defined and measured?
o How as outcome of interest defined and measured?
o How was LTFU minimized, was it excessive and/or
equal in the groups?
o Were appropriate other‐factors (confounders)
evaluated?
