Professional Documents
Culture Documents
Objectives:
1. Review Basic Design of Cohort Study
- Prospective
- Retrospective
- Nested Case-Control
2. Review Measure of Effect (Relative
Risk)
COHORT STUDIES
Cohort Study
Key Point:
Presence or absence of risk
factor is determined before
outcome occurs.
COHORT STUDIES
Fixed Cohort
X = outcome
x
(+) Relative
x
risk
Exposure
= (2/3)/(1/3)
(-) x
= 2.0
COHORT STUDIES
Fixed cohort
Disease = Hepatitis A Risk = a/(a+b)
(+) (-) =0.3
S a+b
30 a 70 Risk = c/(c+d)
a (+) b 100
= 0.05
l
a (-) 3 c 57 d c+d
60
d Rel. risk=
a
a+b
c
= 0.3/0.05 =6
c+d
COHORT STUDIES
Rel. risk=
a
a+b
Disease = Hep A c
= 0.3/0.05 =6
c+d
(+) (-)
S
Odds Ratio:
a (+) 30 70 b a+b (a/c)/(b/d)=(a/b)/(c/d)
a
l = 100
a 3 57 d c+d (30/3)/(70/57)= 8.14
(-) c
d = 60
COHORT STUDIES
Dynamic Cohort
Rel. Risk
X
X = 2/3/2/3 =1
(+)
or
Exposure X
X 2/5py/2/10py
(-)
Years
= 2.0
COHORT STUDIES
Questions:
Findings due to confounding?
Could subclinical disease have
affected the risk factor (activity)?
COHORT STUDIES
Take-Home Message:
The best measure of effect is the “relative
risk.” For a fixed cohort, this will be the
ratio of the cumulative incidences. For a
dynamic cohort, this will be the ratio of the
incidence rates.
The odds ratio can be used for fixed cohorts
comparing cumulative incidences. It will
be close to the relative risk for rare
diseases.
COHORT STUDIES
Variations on a theme: