2. Focus areas of the course
Introduction
Study design
Data collection methods and tools
Drug utilization studies
Pharmacovigilance and drug safety
05/26/25 2
3. Introduction to Epidemiology
Learning objectives
By the end of this unit the students will be able to:
–Define epidemiology and explain its objectives;
–Discuss historical development of epidemiology;
–Define and understand the natural history of diseases;
–Know the different measures of disease prognosis;
–Understand disease causation in epidemiology
05/26/25 3
4. Epidemiology: definition
The word has a Greek origin: epi meaning “on or upon”,
“demos” meaning “ the common people” and “logs”
meaning “study”
(Pickett, JP, ed. The American Heritage Dictionary of English Language. 4th
Edition,
Boston, MA: Houghton Mifflin; 2000)
05/26/25 4
5. Definition cont’d
Epidemiology can also be defined as the branch of
medical science which treats epidemics
(Simpson, JA, Weiner ESC, eds. Oxford English Dictionary. 2nd
edition. Oxford,
England: Clarendon press; 1989, London Epidemiological Society, 1850.)
05/26/25 5
6. Definition cont’d
The study of the distribution and determinants of
disease frequency in
specified populations, and the application of this study to
control health problems
Source:
1.MacMoahon, B.,Trichopoulos, D. Epidemiology: Principles and Methods. 2nd
ed.
Boston, MA: Little, Brawn and co; 1996
2. Last, JM. A dictionary of Epidemiology, 4th
ed.. New York, NY: Oxford University Press;
2001.
05/26/25 6
7. Definition cont’d
Five key words/phrases
Population;
Disease frequency;
Disease distribution;
Disease determinants and
Disease control
05/26/25 7
8. Population
Population
Population refers to a group of people with a common
characteristics
such as place of residence, gender, age, or use of certain
medical services
Epidemiologists are concerned with the disease
occurrence in groups of people rather than individuals
Determining the size of the population in which disease
occurs is important to know the true frequency of
disease
05/26/25 8
9. Disease frequency
Quantifying how often a disease occurs in a population
Counting is a key activity of epidemiologists which
include three steps:
Developing a definition of disease;
Instituting a mechanism for counting cases of disease
within specified population and
Determining the size of the population
05/26/25 9
10. Disease Distribution
Analysis of disease patterns according to person, place,
and time
Who is getting the disease?
Where is it occurring?
When is it occurring?
How is it changing overtime?
05/26/25 10
11. Disease Distribution
Variations in disease frequency by the three
characteristics provide useful information that helps:
Understand the health status of a population;
Formulate hypothesis about the determinants;
Plan, implement and evaluate health programs
05/26/25 11
12. Disease Determinants
Factors that bring about a change in a person’s health
Determinants include both causal and preventive
factors
Determinants also include individual, environmental,
and societal
05/26/25 12
14. The objectives of Epidemiology
1. Causation:
To identify the etiology or cause of a disease and relevant risk
factors; example, description of disease in terms of
The kind of people affected and its geographic distribution led to discover
the association of pellagra and maize diet
Knowing how the disease is transmitted would enable us to
intervene , reduce morbidity & mortality
05/26/25 14
15. Objectives cont’d
2. Study the natural history and prognosis of diseases
Some diseases are more severe than others;
some may be rapidly lethal and others may have
longer/shorter duration of survival
Defining the baseline natural history of a disease in
quantitative terms help to
develop new treatment/ prevention modalities
05/26/25 15
16. Objectives cont’d
3. Description of health status of the population
Determine the extent of disease found in the
community:
What are the actual and potential health problems in the
community?
Where are they?
Who is at risk?
Which problem is declining/increasing over time?
How these patterns relate to the level of and distribution of
services available?
05/26/25 16
17. Objectives cont’d
4. Evaluation of intervention
Example: Effectiveness of residual DDT spraying is
measured by reduction in the incidence of malaria
05/26/25 17
18. Epidemiological purpose and sequence:
Summary
1. Identifying disease/health problem
2. Linking with cause/risk factors
3. Establishing causal relationship
4. Designing an intervention for controlling health problem
5. Evaluating effectiveness of intervention
(MAXCY)
05/26/25 18
20. Hippocrates (460-377BC)
The First Epidemiologist
Attempted to explain disease occurrence from rational instead
of supernatural viewpoint
Wrote three books: Epidemic I, Epidemic III, and On Airs,
Waters and Places
Recognized that different diseases occurred in different
places.
Example: malaria & yellow fever most commonly occurred in
swampy areas.
05/26/25 20
21. Hippocrates (460-377BC)…
The essentials of epidemiology noted by Hippocrates
included observation on how diseases
affected population and how disease spread particularly,
issues related to
time, seasons, place, & environmental conditions
His broader contribution to epidemiology is
“Epidemiological Observation”
05/26/25 21
22. John Graunt & Vital Statistics Development
In 1602 in London a systematic recording of deaths was
commenced,
“ Bills of Mortality”- weekly count of deaths
Graunt systematically recorded age, sex, who died, of
what, and where they died and when;
how many persons per year died of what kind of event.
1662: Published Natural & Political observations on the
Bills of Mortality
05/26/25 22
23. John Graunt & Vital Statistics Development
He drew many inferences about the patterns of fertility,
morbidity and mortality
by tabulating the Bills of Mortality,
He was the first to calculate &develop life tables and life
expectancy
Considered as Epidemiologist, statistician and
demographer
05/26/25 23
24. James Lind- Mid 1700s
Ship’s surgeon
1747: Conducted one of earliest experimental studies on
the treatment of scurvy, a common disease & cause of death
that time
He dismissed the popular ideas that scurvy was a
hereditary or infectious disease
The principal/main predisposing cause was moist air and
that its “occasional” cause was diet
05/26/25 24
25. Lind’s Experiment
Took 12 patients with scurvy on board
Patients are similar situation (putrid gums, the spots and lassitude, with
weaknesses of their knees, had one diet common to all)
Design
2 were with a quart of cyder a day, 2 took 25 gutts of elixir vitriol 3x a day;
2 took 2 spoonful of vinegar 3x/day; 2 of the worst patient under a course
of sea water; 2 had each two oranges & one lemon/day; 2 took nutmeg
3x/day.
Observation: sudden and visible good effects were perceived
from the use of oranges and lemons.
05/26/25 25
26. William Farr (1807-1883)
1839: Farr was appointed as Registrar General in England
and built on the ideas of Graunt
Farr advanced the use of vital statistics.
Devised a system which is the antecedent of modern
international classification of diseases.
Farr was considered as the father of modern vital statistics
and disease surveillance
05/26/25 26
27. London Cholera Epidemic (1849-1854)
Finding the cause of Cholera was important issue
William Farr: adhered to “Miasmic theory”
05/26/25 27
28. Farr’s observational study
Elevation (ft) Deaths/10,000 inhabitants
< 20 120
20-40 65
40-60 34
60-80 27
80-100 22
100-120 17
340-360 8
Table 1. Deaths from cholera in 10,000 inhabitants by
elevation of residence above sea level, London (1848-
1849)
The lower the
elevation the higher
the mortality
05/26/25 28
29. John Snow
Anesthesiologist during the mid-1800s
Interested in the cause and spread of cholera
1849: he published his views on the causes and
transmission of cholera.
Snow conducted his classic study in 1854
Determine where persons with cholera lived and worked-
used the information to map the distribution of cases (spot map)
05/26/25 29
31. Natural History of Diseases
What is health and disease?
Disease, illness and sickness
What does it mean by natural history of disease?
05/26/25 31
32. Natural History of Diseases
Health is a difficulty concept to define
According to Advanced Learner’s Dictionary:
Health is the state of being well and free from illness
There are two opposing models in the concept of Health:
a negative and positive models.
A negative model : illness-wellness model
Health is the absence of the constraints of illness i.e. you are
healthy if you are not ill.
Disease A + Treatment A = Health
05/26/25 32
33. Definition of Health and Disease cont’d
Positive model of health – WHO’s definition
Health is a state of complete physical, mental, and social
wellbeing and not merely the absence of disease or infirmity.
It also includes the ability to lead a socially and economically
productive life.
05/26/25 33
34. Definition of Health and Disease cont’d
Disease is relatively simple to define and conceptualize
The opposite of ease.
Disease: physiological/psychological dysfunction
Illness: a subjective state of a person who feels aware of not
being well
Sickness: a state of social dysfunction.
05/26/25 34
35. Natural History of Diseases (NHD)
Exposure Onset of
symptoms
Pathological
changes
Usual time of diagnosis
Immunity &
Resistance
NHD: refers to the progress of a disease process in an individual overtime, in the
absence of an intervention
Incubation/latency period
05/26/25 35
36. Timing of prevention efforts in the NHD
primordial
& 1o
secondary tertiary
05/26/25 36
37. Timing of prevention efforts in the NHD..
A P
S
M D T
A: Biologic onset
P: Pathological evidence could be found if sought
S: Sign and symptoms develop
M:Medical care sought
D: Diagnosis
T: Treatment
05/26/25 37
38. Disease prognosis
Prognosis is a quantitative expression of the likelihood of
a specific outcome (survival)”
General issues:-
1. At what point to begin counting survival?
2. How is diagnosis made?
05/26/25 38
39. Identifying the endpoints of disease
Death
Cure
Remission (A decrease in, or disappearance of, signs and
symptoms of disease)
Recurrences
A return of disease
05/26/25 39
41. Ways of expressing prognosis
1. Case Fatality Rate (CFR)
Given that a person has a disease, what is the likelihood that
he/she will die of the disease?
Number of people who died of a disease divided by the number
of people who have the disease.
Suitable for diseases that are short –term/acute.
In chronic diseases in which death occur many years after
diagnosis and the possibility of death from other causes
becomes more likely a CFR becomes less useful.
05/26/25 41
42. Ways of expressing prognosis…
2. Five Year Survival
A percent of patients who are alive 5-years after
treatment begins/ after diagnosis
Used as an index of success in cancer treatment.
Not appropriate measure if we want to measure less than five
years survival experience
Measure of success may be misleading if treatment start
either after the sign and symptoms appear or before that
through screening programs- Lead time bias
05/26/25 42
43. Five year survival
Example: Treatment outcome of breast cancer patient. Scenario I:
BO D&T→ 4-years survival Death
______↓_____________↓__________________↓_______
1987 1991 1995
Scenario II:
BO S&D&T → 6-years survival ← Death
______↓_____________↓__________________↓_______
1987 1989 1995
05/26/25 43
44. Ways of expressing prognosis…
Five year survival cont’d
Death occurred in 2015 in both scenarios.
Same patient could be classified as treatment success in
scenario II and failure in scenario I
05/26/25 44
45. Ways of expressing prognosis…
To use the actual observed survival overtime using a life table
approach.
Table 1: A hypothetical study of treatment results in patients
who were treated from 2010 to 2014 and followed to 2015.
Yr of
treatment
No of
patients
No alive on anniversary of treatment
2010
2011
2012
2013
2014
84
62
93
60
76
2011 2012 2013 2014 2015
44 21
31
13
14
50
10
10
20
29
8
6
13
16
43
05/26/25 45
46. Ways of expressing prognosis:
Observed survival..
Table 2: Analysis of survival in patients treated from 2010 to 2014 and followed
to 2015 (no lost to follow up)
Yr of
treatment
No. of
patients
No. alive at the end of Year
2010
2011
2012
2013
2014
Total
84
62
93
60
76
375
1st 2nd 3rd 4th 5th
44
31
50
29
43
197
21
14
20
16
71
13
10
13
36
10
6
16
8
8
05/26/25 46
47. Ways of expressing prognosis:
Observed survival…
The probability of surviving the first year is 197/375 = 0.52
P2 =71/197-43 = 0.46
P3= 36/71-16 = 0.65
P4 = 16/36-13 =0.70
P5 = 8/16-6 = 0.80
PT = p1xp2xp3xp4xp5 = 0.088 or 8.8%
05/26/25 47
48. Survival curve: percent surviving vs. years
of follow up
Fig 1: Survival curve for hypothetical example of patients treated from
2010-2014 and followed to 2015
05/26/25 48
49. Ways of expressing prognosis:
Median survival
The length of time that half of the study observation
Why should we use median rather than mean?
Less affected by extremes unlike mean
We need only to observe the deaths of half of the group to
calculate the median.
05/26/25 49
50. Ways of expressing prognosis:
relative survival
The ratio of the observed survival to expected survival
rate.
RSR = Observed survival in people with disease
Expected survival if disease were absent
Let us consider 5-year survival rate for a group of 30-years old
men with colon cancer
What would you expect their survival to be if they did not
have colon cancer? What if we consider a group of 80-years
old men with the same disease? We wouldn’t expect
anything near 100% survival in a population of this age, even
without colon cancer.
05/26/25 50
51. Cause of Diseases
Disease could be classified as
Acute or chronic based on its course
Infectious or non-infectious based on its cause
A cause of disease is a factor (characteristic, behavior, event etc..)
that influences the occurrences of disease.
If disease does not develop without the factor being present then
we term the causative factor “ sufficient”
Example: Exposure to MTB is necessary for TB to develop but it is
not sufficient, because not everyone infected develops the
disease.
05/26/25 51
53. Epidemiologic Triangle/Triad
Traditional model of disease causation;
Three components:
an external agent,
a susceptible host, and
an environment that brings the host and agent together;
05/26/25 53
54. Epidemiologic Triangle/Triad
Agent factors originally referred to infectious
microorganisms
These agents must be present for the disease to occur
(necessary but may not be sufficient)
In non-infectious diseases agent includes chemical and physical
causes of diseases
• Example: Nutritive elements (deficiencies like in Xerophthalmia,
kwashiorkor; : obesity); Chemical agents like poisons with co; physical
agents e.g. radiation
05/26/25 54
55. Epidemiologic Triangle/Triad
Host factors
factors that influence an individual’s exposure
susceptibility or response to a causative agent Example:
age, sex, race, socio-economic status, behaviors (smoking, drug
abuse, life style, sexual practices, eating habits), nutritional and
immunological status etc
05/26/25 55
56. Component causes and causal pies
Agent – host – environment model does not work well
for some non-infectious diseases
Another approach for explaining disease causation is
causal pies model – concept of multi-factorial nature of
disease causation
05/26/25 56
57. Conceptual scheme for causes of hypothetical
disease
Sufficient cause I Sufficient Cause II Sufficient Cause III
05/26/25 57
58. Component causes
Factors represented by the pieces of pie are called component
causes
Disease may have more than one sufficient cause with each
sufficient cause being component of several factors – different
ways leading to the same end.
A single component cause is rarely a sufficient cause by itself
Blocking any single component of a sufficient cause can prevent a
disease at least through that pathway.
05/26/25 58
59. Establishing the cause of diseases
Causal inference :
Process of determining whether the observed association is
likely to be causal
Before the association is assessed for causality, chance, bias
and confounding have to be excluded
05/26/25 59
60. Establishing causation
Bias
Any systematic error in the design, conduct or analysis of
the study that results
mistaken estimate of an exposure’s effect on the risk of the
disease.
Common types of bias in epidemiologic studies:
Selection bias
Misclassification bias (due to inaccuracies in the methods of
data acquisition)
Information bias (includes bias in abstracting records, recall
bias, interviewer bias, social desirability bias, measurement
bias)
05/26/25 60
61. Establishing causation…
Confounding
In a study of the association between exposure to a cause
and the occurrence of disease,
confounding occur when another exposure exists in the study
population and is associated both with the disease and
exposure being studied
In a study of whether factor A is a cause of disease B,
we say that a third factor, factor X is confounder, if the
followings are true:
05/26/25 61
62. Establishing causation - confounding
1. Factor X is a known risk factor for disease B.
2. Factor X is associated with factor A but not the result of
factor A.
X
? B
A
The confounder X is associated both with Exposure A and the disease B
05/26/25 62
63. Establishing causation - confounding
Examples:
In the relationship between coffee and cancer of pancreas,
smoking is a confounder because:
It is a known risk factor for pancreatic cancer,
It is associated with coffee drinking but not the result of coffee
drinking
Confounding may be the explanation for the relationship
between coffee consumption and coronary heart disease,
because:
It is well known that cigarette smoking is the cause of
coronary heart disease
It is also known that coffee consumption is associated with
cigarette smoking.
05/26/25 63
64. Establishing causation - confounding
Control of confounding
In the design stage:
Randomization – applicable only to experimental studies
Restriction – can limit the study to people who have particular
characteristics
Matching
oGroup matching: selecting controls in such a manner that
the proportion of controls with a certain characteristics is
identical to the proportion of cases with the same
characteristics
oIndividual matching: for each case selected , control is
selected who is similar to the case in terms of the specific
variable of concern
05/26/25 64
65. Establishing causation - confounding
At the analysis stage :
Stratification which involves the measurement of
strength of associations in well defined and
homogenous categories (strata) of the confounding
variable.
Statistical modeling: multivariate model controls a
number of confounding variables simultaneously.
05/26/25 65
66. Guidelines for establishing causation
The best known criteria for assessing causation were
proposed in 1965 by Sir Austin Bradford Hill : Hill’s
nine criteria or more appropriately termed as
guidelines.
05/26/25 66
67. Criteria for Causal Association
Surgeon General’s Report (1964)
1. Consistency
2. Strength —Dose-response
3. Specificity
4. Temporality
5. Coherence
Hill’s Criteria (1965)
1. Strength
2. Consistency
3. Specificity
4. Temporality
5. Biological gradient
6. Plausibility
7. Coherence
8. Experiment
9. Analogy
Source: Hill AB. The Environment and Disease: Association or Causation? Proceedings of
the Royal Society of Medicine 1965; 58:295-300.
05/26/25 67
68. Guidelines for establishing causation
1. Temporal relation:
Does the cause precede the effect?
2. Plausibility:
Is the association consistent with other
knowledge? Mechanism of action; evidence from
experimental animal.
3. Consistency
Have similar results been shown in other studies?
05/26/25 68
69. Guidelines contd..
4. Strength
What is the strength of the association between cause and
effect? RR>2 can be considered as strong. Cigarette smoking
is approximately twofold increase in the risk of MI
compared to non-smokers.
5. Dos- response relationship:
Is increased exposure to possible cause associated with
increased effect? Occurs when changes in the level of
possible cause are associated with changes in the
prevalence and incidence of the effect. Example, prevalence
of hearing loss increases with noise level and exposure time.
05/26/25 69
70. Guidelines contd..
6. Reversibility:
Does the removal of a possible cause lead to
reduction of disease risk?
7. Study design:
Is the evidence based on strong study design?
Strength of evidence:
RCT : Strong evidence
Cohort : Moderate
Case-control : Moderate
Cross-sectional: Weak
Ecological: Weak
05/26/25 70
71. Guidelines contd..
8. Specificity of association:
Certain exposure should be associated with single
outcome
Weakest of all criteria
Considered as an additional support if others hold
true.
71
05/26/25
