lecture 10: cohort studies

Question 1

cohort study

Answer 1

observational study used to observe naturally occuring events in naturally exposed/unexposed groups; useful for rare exposures

Question 2

two different ways to organize groups

Answer 2

1. group allocation by exposure status

2. group membership via something in common

Question 3

3 ways (fashions) that cohort studies can be designed

Answer 3

1. prospective
2. retrospective
3. ambidirectional

Question 4

prospective fashion

Answer 4

exposure group is selected based on past or current exposure and all groups are followed into FUTURE to assess outcomes

Question 5

pros/cons of prospective

Answer 5

PROS:
follow up/tracking could be easier if planned
better at answering temporality
may look at multiple outcomes for one exposure
calculate incidence/incidence rates
more study-important information available

CONS: time, expense, loss-to-follow up
not efficient for rare diseases
not suited for long latency/induction periods
exposure may change over time

Question 6

retrospective fashion

Answer 6

at start, BOTH exposure and outcome of interest have already occurred, but groups are based on EXPOSURE.

Question 7

pros/cons of retrospective

Answer 7

PROS:
best for long latency/induction periods
able to study rare exposures
useful if data already exists

CONS:
requires access to records
“info” may not factor in confounding or other exposures
patients may not be available to interview if data is missing
exposure may change over time

Question 8

cohort

Answer 8

a group with something in common

Question 9

3 types/examples of cohorts

Answer 9

1. birth = born in same region in same period of time
2. inception = assembled via common factor
3. exposure = assembled via common exposure or one-time event

Question 10

cohort sizes

Answer 10

1. fixed = fixed at start and may decline
2. closed= fixed at start and doesn’t change
3. dynamic/open= new additions and losses

Question 11

selecting exposed group

Answer 11

based on a pre-defined criteria of “exposure”

Question 12

selecting unexposed group/sources

Answer 12

need to make groups as close as possible (numbered best to worst)

1. internal = within same “cohort” group, yet unexposed
2. general pop= used when entire cohort is exposed OR exposures are taken from gen pop
3. comparison cohort= attempt to match groups as close as possible based on personal characteristics (can’t control)

Question 13

strengths of cohort studies

Answer 13

good for assessing multiple outcomes for ONE exposure
good when exposures are rare
calculate risk/risk ratios
less expensive
RETRO: good for long induction/latency periods
PROS: able to represent temporality

Question 14

weaknesses of cohort studies

Answer 14

can’t demonstrate causation
RETRO: hard to control for other exposures, impacted by confounders/bias, limited by available data
PROS: not good for long latency/induction periods

Question 15

biases encountered in cohort studies

Answer 15

healthy-worker effect

selection bias