Epidemiology

a. Long Questions-
i. Define Epidemiology. What are the uses of
epidemiology? Describe various epidemiological
methods.

Definition: study of distribution & determinants of

health related events in specified populations&the
application of this study to the control of health
problems
3 components
Frequency of disease: prevalence rate, incidence
rate
Distribution of the disease:w.r.t time place and
person
Determinants of disease : identifying the underlying
cause or risk factors of disease
Uses:

1. To study rise & fall of the disease

•Study of disease profiles & time trends in human
population
2. Community diagnosis
• Identification & quantification of health problems
in a community in terms of mortality & morbidity
rates & ratios
3. Planning & evaluation of health services
• Epidemiological information aboutthe distribution
of health problems over time & place provides the
fundamental basis for planning & developing the
health services & for assessing the impact of these
services
 4. Evaluation of individual risks
• Epidemiologists calculate Relative Risk &
Attributable Risk for a factor related to be a
cause of the disease
5. Syndrome identification
• Medical syndromes are identified by observing
frequently associated findings in individual
patients
6. Completing the natural history of disease
• Epidemiologists by studying disease patterns
in the community in relation to agent, host &
environmental factors is in a better position to
fill up the gaps in the natural history of disease
7. Searching for causes & risk factors
• Epidemiology, by relating disease to
interpopulation differences & other attributes,
tries to identify the causes of disease

Epidemiological methods
I) Observational studies

1)Descriptive studies (hypothesis formulation)

2) Analytical studies (hypothesis testing)

1. Ecoloqical/ Correlational study (unit of study is
population)
2. Crossectional/ prevalence survey/ snapshot of
population survey (unit of study is individual)
3. Case control (unit of study is individual!
[Link] series study
[Link] case control study
[Link] (unit of study is individual)
*Prospective cohart
*Retrospective cohart
*Mixed chart/ combined prospective retrospective
cohart

II)Experimental studies (hypothesis confirmation

1)RCT (unit of study is patients/ cases)
2)Field trials (unit of study is healthy people)
3)Community trials (unit of study is community)
4)Clinical trial (unit of study is patients/ cases)

Migration study- used to study environmental &

genetic factors in a disease in population

KAP (knowledge Attitude Practice) study was first

used in Indid to study family planning

Sequence of studies in epidemiology

Descriptive study -> analytical study -> Experimental
study

Metaanalvsis (Strongest) > RCT Cohart (retrospective

prospective) > case control study > crossectional
study > ecological study (weakest)
 ii. Define and describe descriptive epidemiology.

DEFINITION: The study of distribution of health

related events in specified populations by time, place
and person.
PROCEDURES:

[Link] the population to be studied

[Link] the Disease under study
[Link] the Disease by
a. Time
b. Place
c. Person
4. Measurement of Disease
[Link] with known indices
[Link] of an a etiological hypothesis

1 . Defining the population

The "defined population" can be the whole

population in a geographic area, or more often a
representative sample taken from it. The defined
population can also be a specially selected group
such as age and sex groups, occupational groups,
hospital patients, school children, small communities
as well as wider groupings - in fact, wherever a
group of people can be fairly accurately counted.
 CRITERIA:
*Large
*Stable
*Not different from other communities
*Community participation
*Accessibility of health services

2 . Defining the disease under study

*Once the population to be studied is defined or

specified, one must then define the disease or
condition being investigated. Here the needs of the
clinician and epidemiologist may diverge. The
clinician may not need a precise definition of
disease (e.g.. migraine) for immediate patient care.

* the epidemiologist looks out for an "operational

definition", i.e., a definition by which the disease or
condition can be identified and measured in the
defined population with a degree of accuracy. For
example, tonsillitis might be defined clinically as an
inflammation of the tonsils caused by infection.
usually with streptococcus pyogenes. This definition,
like many other clinical definitions , serves to
convey particular information, but cannot be used
to measure disease in the community. On the other
hand, an "operational definition" spells out clearly
the criteria by which the disease can be measured.
Such criteria in the case of tonsillitis would include
the presence of enlarged, red tonsils with white
exudate, which on throat swab culture grow
predominantly S. pyogenes.
3 . Describing the disease
The primary objective of descriptive epidemiology is
to describe the occurrence and distribution of
disease by
*time,
*place and
*person,
and identifying those characteristics associated with
presence or absence of disease in individuals.

Characteristics frequently examined in descriptive studies

4. Measurement of disease

This information should be available in terms of

mortality, morbidity, disability and so on, and
should preferably be available for different
subgroups of the population.
5. Comparing with known indices

By making comparisons between different populations,

and subgroups of the same population, it is possible to
arrive at clues to disease aetiology. We can also
identify or define groups which are at increased risk
for certain diseases.

6. Formulation of a hypothesis
*By studying the distribution of disease, and
utilizing the techniques of descriptive epidemiology,
it is often possible to formulate hypotheses relating
to disease aetiology.
*An epidemiological hypothesis should specify the
following :
a. the population - the characteristics of the persons
to whom the hypothesis applies
b. the specific cause being considered
c. the expected outcome - the disease
d. the dose-response relationship - the amount of the
cause needed to lead to a stated incidence of the
effect
e. the time-response relationship - the time period
that will elapse between exposure to the cause and
observation of the effect.

* A hypothesis should be formulated in a manner that

it can be tested taking into consideration the above
elements. In practice, the components of a hypothesis
are often less well- defined.
 iii. Discuss distribution and determinants of a
disease process in a community
giving suitable examples

Distribution is based on
*Time
*Place
*person

Time distribution

The pattern of disease may be described by the time

of its occurrence. There are three kinds of time
trends or fluctuations in disease occurrence.
I. Short-term fluctuations
II. Periodic fluctuations, and
III. Long-term or secular trends

I. Short-term fluctuations:
1. Common source epidemics:
i. Single exposure or point source epidemics
*explosive
*sharp rise &sharp fall in no. of cases
*clustering of cases in narrow interval of time
*all case develop within one incubation period of
the disease
*uniform curve with no secondary waves
Example:food poisoning, Bhopal gas tragedy
 ii. Continuous or repeated or multiple exposure
epidemics
*sharp rise in number of cases
*fall in number of cases is interrupted by secondary
waves/peaks
Example: STD-prostitutes, contaminated well in a
village
2. Propagated epidemics
* gradual rise & gradual fall over a long period of
time
* results from person to person transmission
* transmission continues till no susceptibles are left
or they are no longer exposed to infected
individuals
* speed of spread, depends on herd immunity and
secondary attack rate
Example -HIV,TB,Polio

II. Periodic fluctuations

1. Seasonal trend
*it is a seasonal varied a disease occurrence may
be related to environmental conditions
Example-URI infection -WINTERS,
GIT infection-summers
2. Cyclical trend
Rubella - every 6-9years
Influenza - every 7-10 years
III. Long-term or secular trends
*trends changes over a long period of time or
decades
*CHD,Cancer,diabetes-increased in last few decades
*polio,diphtheria,typhoid-decreased in last few
decades

Place distribution

1. Local variation
*spot maps(ex:cholera)
*shaded maps
2. Rural-urban variation
*rural - zoonotic diseases, soil transmitted
HELMINTHS
*urban -accidents ,CVD
[Link] variation
*northern region-goitre
*southern region malaria
*central region-lathyrism
4. International variation
*japan -cancer stomach
*india -cancer in oral cavity
 Person distribution
1. Age
Childhood- measles
• Middle age- cancer
Old age - atherosclerosis
2. Sex
Females- hyperthyroldism
Males- lung cancer
3. Marital status
Marled females- cancer cervix
Mortality rates are lower In married males
or females
4. Occupation
Coal mines- pneumoconiosis
Desk job- CHD
5. Social class
- Upper class- non communicable diseases-
DM, HIN
Lower class- communicable diseases- TB,
AIDS
Determinants

• They are causative agents of the disease

• Most important component of the definition of
epidemiology
•
• Finding causative agent helps in prevention and
control of disease
• E.g. Swineflu
-> First detected in Mexico 2009. Health authorities
thought it is due to swine (pig) , applying
epidemiology helps to find them there is no role of
pig in spreading the disease and the exact causative
agent is H1N1 virus and the route of transmission is
respiratory route.
- DOC is Oseltamivir
- Two types of vaccines: live and killed vaccines
aré used.
- It is k/as H1N1 type A Influenza pm 09
iv. Describe and discuss Analytical Epidemiology.

*Analytical studies are the second major type of

epidemiological studies. In contrast to descriptive
studies that look at entire populations, in analytical
studies, the subject of interest is the individual
within the population.
* Analytical studies comprise two distinct types of
observational studies :
a. case control study
b. cohort study.
From each of these study designs, one can
determine :
a. whether or not a statistical association exists
between a disease and a suspected factor; and
b. if one exists, the strength of the Association.
v. Describe and discuss the elements of Cohort study.

Cohort study is another type of analytical study

which is usually undertaken to obtain additional
evidence to support the existence of an association
between suspected cause and disease

ELEMENTS OF A COHORT STUDY

The elements of a cohort study are :
1. Selection of study
2. subjects Obtaining data on exposure
3. Selection of comparison groups
4. Follow-up, and
5. Analysis.

1. Selection of study subjects

1. General population
2. Special groups
[Link] groups- doctors, teachers, lawyers
[Link] groups- industrial workers,
radiologists
2. Obtaining data on exposure
[Link] interviews
ü.Review of records
[Link] examination/ tests
[Link] surveys
3. Selection of comparison groups i.
[Link] comparison
*according to degree of exposure
*smoking
[Link] comparison
• Chart of radiologists with Chart of
ophthalmologists

4. Follow up
• Periodic medical examination
• Periodic home visit

5. Analysis

The data are analyzed in terms of :

(a) Incidence rates of outcome among exposed and
non- exposed,
(b) Estimation of risk.
vi. Discuss the principle and the steps of Case
Control study.

Case control studies, often called "retrospective

studies" are a common first approach to test causal
hypothesis.

The case control method has three distinct

features :

a. both exposure and outcome (disease) have

occurred before the start of the study

b. the study proceeds backwards from effect to

cause; and

c. it uses a control or comparison group to support

or refute an inference.

There are four basic steps in conducting a case

control study :

1. Selection of cases and controls

2. Matching
3. Measurement of exposure, and
4. Analysis and interpretation.
1.1. Selection of cases
• Definition of case
•Diagnostic criteria
• Eligibility criteria
• Sources of cases
•Hospitals
• General population
1.2 Selection of controls
•Controls must be as similar to the cases as possible,
except for the absence of disease
*If study group is small choose up to 4 controls per
case(1:4) & in larger studies 1: 1 of cases & controls
should be taken
.
Sources of controls-
• Hospital (often source of selection bias)
• Relatives
• Sibling controls are unsultable In genetic studies
• Neighborhood
•General population

2. Matching
• To ensure comparability between cases & controls
• Types-
[Link] matching- age, occupation, social class
[Link] in pairs
3. Measurement of exposure
• Obtained by- interviews, by questionnaire, or by
studying past records(hospital record, employment
record)
4. Analysis

The final step is analysis, to find out

(a)Exposure rates among cases and controls to
suspected factor
(b) Estimation of disease risk associated with
exposure (Odds ratio)

I. Exposure rates
Cases =a/a + c
Controls= b/b + d
II. Estimation of risk (strength of assoclation)
• Cannot provide with Incidences, so relative risk cannot be
calculated, so an estimate of relative risk that is known as
odds ratio or cross product ratio is calculated (less accurate
than relative risk as it is an estimate)
Odds ratio(Cross product ratio)
OR = axd/ bxc

Interpretation- Exposed showed a risk of having disease

times that of non exposed
Indicator of increased risk of disease in predisposed
population
It is just on estimate of relative risk not the calculation of
relative risk similor to relative risk
vii. Define the terms Epidemic, Endemic and
Pandemic giving suitable examples.

Epidemics- unusual occurrence of a disease in a

community, clearly in excess of expected occurrence
*Outbreak - small localized epidemic
*The area declared free of epidemic when no new
case is reported from twice the incubation period of
disease since the last case
*Verification of diagnosis is the first step in
investigation o an epidemic
Example :yellow fever , small pox

Endemic
• constant presence or usual or expected frequency
of a disease within a given geographic area

[Link]-endemic -
• Constant presence of disease àt high incidence/
prevalence
• affects all age group
b. Holo-endemic -
• High level of infection beginning early in life
• affecting mostly children

• for the disease to be endemic in a

steadystate = RO X S =1,
RO=basic reproduction number of an infection
(number of secondary case caused by a single case
in a population with no immunity & intervention)
S = proportion of susceptibles in population
Example: malaria
• Endemic curve

*it's not a straight line as number of cases for the

endemic disease in a population will not be fixed
throughout a year but it will show seasonal
variations

*Endemic curve & epidemic curve are different as

baseline of endemic curve never touches zero

Pandemic -
*an epidemic usually affecting large proportion of
population over a wide geographic area such
as entire nation or a continent or world (country to
country spread)
*pandemic are caused by influenza A
Example: covid19
i. Uses of epidemiology

1. To study rise & fall of the disease

Study of disease profiles & time trends In human
population
2. Community diagnosis
*Identification & quantification of health problems
in a community in terms of mortality & morbidity
rates & ratios
3. Planning & evaluation of health services
*Epidemiological Information about the
distribution of health problems over time & place
provides the fundamental bass for planning &
developing the health services & for assessing the
impact of these services
4. Evaluation of Individual risks
• Epidemiologists calculate relative risk &
attributable risk for a factor related to be a cause
of the disease
5. Syndrome Identification
• Medical syndromes are identified by observing
frequently assocated findings in individual
patients
6. Completing the natural history of disease
• Epidemiologists by studying disease patterns in
the community in relation to agent, host &
environmental factors is in a better position to
fill up the gaps in the natural history of disease
7. Searching for causes & risk factors
• Epidemiology, by relating disease to
interpopulation differences & other attributes,
tries to identify the causes of disease
Concept of Bias

* Any systematic error in an epidemiological

study, occurring during data collection,
compilation, analysis & interpretation
* Arise from human errors of assessment of the
outcome due to human element
1. Subject bias
*Error introduced by study subjects.
*participants who may subjectively feel better if
knew they were receiving a new treatment
*Attention bias / Hawthorne effect
-Study subject may systematically alter their
behaviour when they know they are being watched
-Seen in cohart study
*Memory / Recall bias
- Cases are more likely to remember exposure
than controls
-It is a systematic distortion of retrospective
study that can be eliminated by a prospective
design
2. Observer/ Investigator bias
*Investigator may be influenced if he knows
beforehand that what therapy is given to which
patient
*Example- blas due to wrong interpretation of
laboratory test
*Selection bias / Susceptibility bias
*Groups are more susceptible to desired outcome
if biased during selection Occurs during
recruitment
Berksonian bias
*It is a type of selection bias
*It is due to different rates of admission to
hospital due to different diseases
3. Analyzer bias- Introduced by analyzer
4. Interviewer bias- Interviewer devotes more
time with cases as compared to controls
5. Lead time bias/ zero time shift bias- Bias of
over estimation of survival time due to
backward shift of starting point
6. Neyman/ Prevalance-Incidence blas- Bias due
to missing of fetal cases. Mild/ silent cases &
short duration cases from the study
7. Bias In evaluation- Investigator may
subconsciously give a favorable report of the
outcome of the trial

Blinding, Randomization & Matching help to

reduce blas

Concept of randomization

*Statistical procedure by which the

participants are allocated into groups called
study & control groups to receive or not to
receive the intervention
*Randomization is done while dividing patients
into experimental group & reference group
* Randomization is the heart of trial
* Best done by using a table of random numbers

Purpose of randomization
-To equalize the effects of extraneous variables,
thus quarding against bias
-Participants have equal &known chance of
falling into either of the 2 groups
-To eliminate selection bias
- To ensure comparability of 2 groups
-To ensure that study groups are comparable on
baseline characteristics
- To have similar prognostic factor among 2
groups
-Facilitates blinding of treatment
-Increase internal validity ofstudy

It is the heart of RCT

Randomization is better than matching as it

removes both confounding & bias

Stratified randomization
*ideal to ensure similarity between experimental
& control groups
a) Relative Risk (risk ratio)
RR = Incidence of disease among exposed
Incidence of disease among non exposed

• RR measures the strength of association between

suspected cause & effect
RR = 1 indicates no association
• Ex- smoking & HIV/ AIDS
• Interpretation- Chance of disease development is same
among exposed as compared to non exposed
RR > 1 indicates positive association between exposure &
disease
• Ex- smoking & lung cancer
• Interpretation-_RR of ___indicates that incidence of
disease is ___times higher in the exposed group as
compared with unexposed
RR < 1 negative association between exposure & disease
• Ex- vitamin A intake & epithelial cancers
• Interpretation--Chances of disease development is less
among exposed as compared to non exposed
Ex. RR of 0.25 indicates 75 % reduction in incidence of
disease in exposed as compared to unexposed

b) Attributable Risk
AR = incidence of disease among exposed - incidence of
disease among non exposed × 100
incidence of disease among exposed

*Interpretation- _so much of disease is attributed to

exposure
*Good measure of extent of public health problem caused
by the exposure
*Most appropriate method to know about contribution of
risk factors to disease
*Assess etiological role or factor in disease
*It is the risk difference between exposed & non exposed
*Also known as absolute risk or excess risk or risk
difference
Framingham Heart Study
* cohort study
* 1948 by US puble health services at Framingham
to study relationship of risk factors (serum
cholesterol, blood pressure, weight & smoking) &
development of cardiovascular disease

*age group - 30-62 vears

*Sample size- 5127
*Multiple exposure were studied by using multivariate
methods
*Follow up: study population was examined every 2
years for 20 years
*Findings of study

- Increasing risk of CHD with Increasing age & more

in males
- Hypertensive have a greater risk of CHD
- Elevated blood cholesterol assoclated with ciD
- Tobacco smoking & habitual use of alcohol
Increases risk of CHD
- Increase body weight predisposes CHD
- DM Increases risk of CHD
- Increased physical activity decreases CHD
development
Odds ratio(Cross product ratio)
OR = exposed &developed cases X neither exposed nor
developed cases
Exposed but not developed X not exposed but
developed disease

* Interpretation- Exposed showed a risk of having

disease___times that of non exposed
* Indicator of increased risk of disease in
predisposed population
* It is just on estimate of relative risk not the
calculation of relative risk

• OR > 1- Positive association

*so many times odds that cases were exposed to a
risk factor is more to the odds that the controls were
exposed
* example- OCPs & thromboembolism
• OR = 1- No association
* Odds that cases were exposed to a risk factor is
same as the odds that the controls were exposed
* Example- smoking & HIV/ AIDS
• OR < 1- Negative association
* So many times odds that cases were exposed to risk
factor is less than the odds that the control were
exposed
* Example- regular physical exercise CHD
 Randomized Control Trial (RCT)

*Gold standard for clinical research is RCT

*For new programmes & theraples RCT Is the best
method of evaluation
*Baseline characteristles of Intervention are simllar
in both arms
*Baseline characteristics are comparable
*Investigator bias Is minimized by double blinding
*Sample size required depends on the hypothesis&
type of study
*The dropouts from the trial are not excluded
from the study
*Intention to retreat analysis Is done in RCT

Confounding
*Any factor viz associated with both exposure &
outcome & has an independent effect in causation
of outcome is a confounder
*Unequally distributed between study & control
group
*Associated with both exposure & outcome
*Independent effect in causation of outcome(so
itself a risk factor
*Source of bias is interpretation
*Methods used to control confounding
-Randomization- most ideal method
-Matching- mostly useful in case control studies
-Restriction
-Stratification
-Statistical modeling
-Multivariate analysis
Example
A study revealed that lesser Incidence of carcinoma
colon in pure vegctarians than non vegetarians by
which it was concluded that beta carotene is
protective against cancer. This may not be true
because the vegetarian subjects may be consuming
high fibre diet which is protective against cancer

Criteria of casual association

Hills (surgeon general's) criteria of causal assoclation
• Temporal association
*Cause preceded effect or effect follows cause
*Suspected cause preceding the observed effect is an
example for temporality
*Considers both order of appearance & length of
Interval between exposure & disease
*Most Important/ essential criterion of causal
association
*Best established by a cohart study (especially
concurrent chart study)
• Strength of association
*Relative risk (cohart study)- Cohart study is
associated with antecedent causation
*Odds ratio (case control study)
• Specificity of association
*Implies that Disease is caused by risk factor
*Most difficult criterion to establish
* Weakest criterion
• Consistency of association- results are replicable in
different settings
• Biological plausibility- Credibility of association by
anatomical/physiological justification
• Coherence of association- association is supported
with relevant facts & studies
• Dose response relationship- increase in dose cause
Increase in incidence of effect
• Reversibility- removal of cause reduces risk of
disease
- Example- current smokers are at higher risk of
developing lung cancer as compared to ex smoken

• RCT are best studles for establishing causation

Ecological study Is the weakest study to test the
assocation between risk factor & diseass
• Indirect association - example- assoclation of high
altitude areas with goitre