Epidemiology by joud hassoun

We have three types of variables :

1. Number
2. Numerical / ordered(age, weight ,height)
3. Categorical / Nominal (yes or no)

epidemiology

Observation Experimental

(when? , (Why ? ,
where? , how?)
who?) Descriptive: Analytical :
o Cas report o Cohort
o Serial cases o Case
study(control)
o Cross-sectional
 Systematic
review

RCT
Increase the error

cohort

Decrease the error

Case control

Cross-sectional

Case report / serial-case

Expert opinion
Contingency table : Cross-sectional :
Is a 2 x 2 table (snapshot) a + b + c + d = population

Case D+ Control D-
Exposed E+ a b a+b
Unexposed E- c d c+d
a+c b+d

E+, D+

re
on

as su
E+, D-
lati

ise xpo
pu

e
ed e
po

sample

th ure
E-, D+

of eas
M E-, D-
 Cohort : (prospective ) = incidence study
a/ ( a + b)
RR (relative risk) =
c/ ( c + d)

Risk of exposed : a/(a + b)

Risk of unexposed : c/(c + d)

Risk attribute or excess risk : risk of exposed – risk of unexposed

OR (odds ratio)= ad / bc D+
E+
D-
sample
D+
Target
population E-
D-

We are looking for the relation between

the exposer and the disease
Examples :

Given : a cohort study of 156 persons with coronary artery disease smoker following 5 years showing death :
a= 27 b= 48 c= 14 d= 67
Calculate the risk if the smokers and non smokers and give the relative risk and attribute risk (excess of risk).

Answer :
dead Alive
Smoker 27 48
Non smoker 14 67

Risk of smokers= 27/27+48 = 0,35

Risk of non smokers = 14/14+67 = 0,1728
RR = 0,35/ 0,1728 = 2,1
Attribute risk = 0,35 – 0,1728 = ~ 0,19

This means these diseased patient that continued smoking even thought they are diseased in the 5
years interval have a 19 deaths with a risk base of 2,1

End of this exercise

Definition of ethically safe :
Subject can be matched and can establish timing and direction of all eligibility, criteria and outcome easier

Exposer may be linked to a hidden confounder

Bias confounder

Blinding the randomization is not present

Case control (retrospective study) (already diseased) OR= ad/bc

E+

D+
E-

E+ D-

E-

Already have the disease (we are looking for the reason of the disease)
True or false :

We can calculate the frequency of the disease and the incidence .

FALSE, WE CANNOT
Evaluation of the expository on a posterior of the diseased and non diseased
TRUE.

Hill Criteria :
Relation cause-effect:
1. Specificity
2. Biologic plausibility
3. Coherence

Descriptive study determine the frequency and the distribution of the disease using time, place, person. It comes in form
of rate
morbidity
We have three types of bias:
mortality 1. Selection
2. Information
natality 3. Confusion
Know the exact need for the population and constitute the health factor for the population
Exercise:

1. Illustrate a diagram of the retrospective study, explain the nature of the risk between exposure and disease.
2. Calculate the excess of risk + explain.

Answer:

3. . E+

po p
D+
E-

ula
tion
D-
E+

E-

2. This is a trap there is no excess of risk or any kind of risk in this study because people in this study
already have the disease , they’re already sick.
Exercise :

1. Make a 2 x 2 table using results of the table shown below .

2. Calculate the prevalence of this cross-sectional study .

Participants smoker Hypertension

1 Yes Yes
2 Yes No
3 Yes Yes
4 yes Yes
5 No No
6 No Yes
7 No No
8 No No
9 No Yes
10 No No
Answer:

1.
Hypertension D+ Not hypertension D- Total
Smoking E+ 3 1 4
Not smoking E- 2 4 6

total 5 5

2. Prevalence??
Since it’s a cross sectional study we calculate the prevalence which is a+c/a+b+c+d
3+2/3+2+1+4= 5/10= 0,5

End of the exercise

Note :
determinant can be (+) or (–) like poverty or obesity
Risk factor is always (–) like smoking