Scribd
Upload
34 views19 pages

Confounding and Effect Modification

The document discusses confounding, mediation, and effect modification in epidemiology. It provides examples and criteria for identifying confounding factors. It also explains the differences between confounding, mediation, and effect modification.

Uploaded by

Mohammed Saad
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
34 views19 pages

Confounding and Effect Modification

The document discusses confounding, mediation, and effect modification in epidemiology. It provides examples and criteria for identifying confounding factors. It also explains the differences between confounding, mediation, and effect modification.

Uploaded by

Mohammed Saad
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 19

Ministry of Higher Education and Scientific Research

Hawler Medical University


College of Medicine
Community Medicine Depatment

Confounding, Mediation and Effect Modification


A Seminar Submitted to
The Council of College of Medicine/ Community Medicine Department in
Partial Fulfillment of the Requirements for the Degree of Doctorate in Community Health

By:

Aram Salahuddin
Qassim Abbod Kareem
Mohammed Saad Abdullah
Supervised by:
Assist. Prof. Dr. Hatham Albana
Jan.9 2024
Outline
Confounding
-Introduction to confounding
-Criteria for confounding
-Controlling confounding at design and analysis stage
Mediation
Effect modification
-Definition and examples
-Confounding vs effect modification
Example

Suppose the incidence of CHD in a cohort study for smokers is twice that
for non-smokers (RR=2).

Question: does RR=2 reflect the effect of smoking?

Alternative explanation:

 Bias

 Confounder

 chance
Example

Suppose that smokers are much older than non-smokers.


We know that age is a risk factor for heart disease
Question: does RR=2 the effect
 Smoking?
 Older age?
 Both?
- Age is a confounder in the study of association between smoking and
CHD
Confounder

The distortion of a measure of the effect of an


exposure on an outcome due to the
association of the exposure with other factors
that influence the occurrence of the outcome.
Direct Acyclic graph (DAG)

Two pathway which could explain why smokers develop more CHD than
non- smokers.
- Direct effect of smoking ( top arrow).
- Backdoor pathway through age.
- Definition of confounding= existence of backdoor pathway.
How to Identify Confounding

1. Subject matter knowledge. Factors identified in existing literature or


plausible biological pathways can inform your decisions.

2. Three criteria for confounding:

(a) It is associated with the study exposure

(b) it is associated with the study outcome in the absence of study exposure; and

(c) it is not a consequence of exposure, i.e. it is not in the causal path between
the exposure and the disease.
Ie= 240/1000= 0.24
Io=120/1000= 0.12
RR= 2
Does age associated with exposure?

Old age and smokers= 600/1000= 60%


Old age and non-smokers= 200/1000=20%
Is age associated with CHD?

CHD and young non smoker= 80/800= 10%


CHD and old non-smoker= 40/200= 20%
How to Identify Confounding
3-Stratification. Stratify data by the extraneous variable to examine if the estimates
within strata of the extraneous variable are similar in magnitude and appreciably
different from the crude (pooled) estimate.

RR=1.5 RR=1.5
RR adjusted=1.5 (Mantel-Haenszel weights or modeling)
RR crude= 2 not equal to RR adjusted=1.5
Mediation

 Confounder should not be in the pathway between the exposure and


outcome
 If the other variable is in the pathway between the two, it is called a
mediator
 X Z Y
 A variable can act partially as a confounder and partially as a mediator
Effect Modification

Example

 Effectiveness of new antibiotic drug on pneumonia

 OR=1.5

 OR men= 1.1

 OR women = 1.9

Effect modification exists when the strength of association varies over


different levels of a third variable.
References
• Aschengrau, Ann_ Seage, George R - Essentials of epidemiology in
public health-Jones & Bartlett Learning (2020)
• Kenneth J. Rothman - Epidemiology_ An Introduction-Oxford
University Press (2012)
• Moyses Szklo, F. Javier Nieto - Epidemiology_ Beyond the Basics-Jones
& Bartlett Learning (2018)

You might also like