Aksum University

1
College of Health Science
Department of Epidemiology and Biostatistics

Basic Epidemiology
TEKLIT ANGESOM (MPH IN EPIDEMIOLOGY)

DECEMBER, 2017
AKSUM, ETHIOPIA

Teklit Angesom 09/18/2020

Introduction
What is epidemiology? 2

 Epidemiology is a fundamental science of public health

 Has made major contributions to improving population health

 Essential to process of identifying and mapping emerging disease

 There is often a frustrating delay between acquiring

epidemiological evidence and applying the evidence to health
policy

Teklit Angesom 09/18/2020

Introduction…
Historical context 3

 Hippocrates observation(>2000 years) that the environmental

factors influence the occurrence of disease (agent, host and env’t)

 In the 19th century, the distribution of disease in specific human

population groups was measured to a large extent

 The finding by John Snow that the risk of cholera in London was
related to drinking of water supplied by a particular company

 Comparing rates of disease in subgroups of human population

become common practice in the late 19th and early 20th century

Teklit Angesom 09/18/2020

Introduction…
4

 The approach was initially applied to the control of communicable

disease

 But later (late 20th) it was applied to chronic non communicable

disease like heart disease and cancer

 Epidemiology in its modern form is relatively new discipline and

uses quantitative methods to study diseases, to inform control and
prevention of diseases

Example : Richard Doll and Andrew Hill study tobacco use and
lung cancer in 1950s
Teklit Angesom 09/18/2020
Introduction…
5
 Experimental study (carcinogenicity of tobacco tars/nicotine)

 Cohort study (long term observation of smokers and non smokers)

Example : the case of cholera Vs water supply and lung cancer Vs

smoking (case study)

 Definition, scope, and uses of Epidemiology

Teklit Angesom 09/18/2020

Introduction…
6
Definition: it is the study of frequency, distribution and determinants
of diseases and other health related states or events in a specified
populations, and the application of this study to the prevention and
control of health problems of the specified population

 Greek words epi, meaning “on or upon,” demos, meaning

“people,” and logos, meaning “the study of.”

 It is not only concerned with death, illness and disability, but also
with more positive health states/with means to improve health

Teklit Angesom 09/18/2020

Introduction…
Terms included in the definition 7
 Study: surveillance, observation, hypothesis testing, analytical research

 Frequency: quantification of magnitude/amount of disease

 Distribution: the analysis by places (where), persons (who), Times

(when)

 Determinants: factors that influence heath status (biological, chemical,

physical, social, cultural, economic, behavioral, genetic etc.)… what,
how, why

Teklit Angesom 09/18/2020

Introduction…
8
 Health related conditions: diseases, behaviors, accidents,
injuries

 Specified population: with identified characteristics

(individual Vs community)

 Application to prevention and control: aim of public heath

Achievements and types of epidemiology

Teklit Angesom 09/18/2020

Introduction
9
Types of epidemiology
 Social
 Occupational
 Environmental
 Nutritional
 Reproductive
 Molecular/Genetic
 Clinical

Teklit Angesom 09/18/2020

Introduction…
10

Epidemiology in Relation to Other Disciplines

 Biomedical, clinical and other related disciplines sometimes

claim that epidemiology belongs to their particular research area

 Physicians Vs epidemiologists

Teklit Angesom 09/18/2020

Introduction …
11
Scope of epidemiology
 Community diagnosis

 Measuring the frequency and mortality of diseases/events

(mortality, births, morbidity incidence and prevalence, hospital
stays, drug use, injuries, health behaviors etc);

 Solving epidemics/outbreaks;

 Look for causes of disease and risk factors;

 Evaluating new diagnostic tests;

 Evaluating new treatments;

Teklit Angesom 09/18/2020

Introduction…
12
 Surveillance for new diseases and changes in old ones (specific
diseases);

 Searching health services (plan effective health services), their

availability and their problem;

 Costing out alternative diagnostics, treatments or health service

provisions

 Monitor the change of health in a community over a period of

time (immunization program, health education, nutritional supply)

Teklit Angesom 09/18/2020

Basic Epidemiologic Assumptions
13

1. Human disease does not occur at random: there are patterns of

occurrence in which some behavioral and environmental factors
(exposures) increase the risk of acquiring/developing a particular
disease among group of individuals

2. Human disease has causal and preventive factors: that can be

identified through systematic investigation of populations or group
of individuals within a population in different places or at different
times

Teklit Angesom 09/18/2020

Use/application of epidemiology
14

Purpose of Epidemiology
 How big is the problem (magnitude)?
 Prevalence, incidence, mortality

 What, who, when and where of any health problem?

 Person characteristic of affected population
 Place characteristics (locality)
 Time characteristics

 What factors are associated with certain disease

 Specific factors related to causation (determinants)

Teklit Angesom 09/18/2020

Introduction…
15

 To evaluate interventions
 Which drug is best for patients with X disease
 To evaluate any program
 In general Epidemiology helps to answer the following questions

 Define (Identification and magnitude)

1. What are the main health problems in the area?
2. How many cases or health events are there?

 Describe (Distribution)
3. When do the cases occur?
4. Where do they occur?
5. Who is affected?
Teklit Angesom 09/18/2020
Introduction…
16
 Analyze
6. Why and how does the problem occur

 Apply and evaluate intervention measures

7. What kind of measures were taken to deal with the problem
8. What results were achieved
9. What else could be done?

Teklit Angesom 09/18/2020

17

Epidemiology of communicable diseases

Teklit Angesom 09/18/2020

Cont’d
18

Communicable diseases continue to account

for a major proportion of disease burden

Occur in epidemic forms

The problem is exacerbated by:

 Poor socio-economic status
 Poor personal and environmental hygiene
 Inadequate health service coverage, etc.

Teklit Angesom 09/18/2020

Natural History of Diseases
19

Usual time
of diagnosis
Pathologic Onset of
Exposure
Changes Symptoms

Stage of Stage of Stage of Stage of Recovery,

Susceptibility Subclinical Clinical Disability or Death
Disease Disease

Teklit Angesom 09/18/2020

Cont’d
20

Begins with exposure with the causative

agent capable of causing disease.

Without intervention, the process ends with

recovery, disability or death.

The course can be halted at any time in the

progression by intervention, host factors,
other influences.

Teklit Angesom 09/18/2020

21

The natural history of infectiousness

includes:
Prepatent period: the time interval from infection to
becoming infectious (shedding of the agent).

Infectious period: the time during which an infected host

could infect another host or vector.

Incubation period: the time from infection to

symptomatic disease.

Teklit Angesom 09/18/2020

Components of Infectious disease process
22

Infectious diseases result from the

interaction between the infectious agent,
host/reservoir and environment.

Host

Agent Environment

Teklit Angesom 09/18/2020

Cont’d
23
Agent: An infectious micro-
organism depends on: Environment: encompasses all
 Pathogenicity,
 infectivity,
extrinsic of the human host
 infective dose,
 Physical: describes the
 immunogenicity, geography and climate,
 Virulence tropical/temperate, urban/rural;

Host: Related to human factors.  Biological: made up of plants,

 Influences individual’s exposure, animals, and other life forms;
susceptibility or response to a
causative agent, and it depends on:
 Age, - gender,
 Socioeconomic: includes
 race, - habits,
factors like housing, sanitation,
 sexual activities, - immunization,
population density, crowding,
 contraception, - diet,
education, occupation, public
 nutrition, -etc. health resources.

Teklit Angesom 09/18/2020

Causal Concepts of Disease
24

 Not all associations between exposure and disease are

causal.

A cause: A factor that preceded the disease and

without which the disease would not occur.

 If disease does not develop without the factor, then

the causative factor is “necessary”.

 No specific factor is sufficient to produce a

disease.

Teklit Angesom 09/18/2020

Cont’d
25

A “sufficient cause” : defined as a set of

conditions and events that inevitably produce
disease;

The occurrence of all of the conditions or

events is necessary.

For example,
Tobacco smoking is a cause of lung cancer, but by itself
it is not a sufficient cause.

Teklit Angesom 09/18/2020

Cont’d
26

If a single factor alone become sufficient to

develop the disease, then we term the
causative factor as both “Necessary” and
“sufficient”.

 Example:
 Tubercle bacilli is a necessary factor for TB

 Rabies virus is sufficient for developing

clinical rabies

Teklit Angesom 09/18/2020

Model of disease causation theories
27

1. Epidemiologic triangle and triad (balance

beam)
-Traditional model of infectious disease causation

Agent Host
Agent

Environment
environment Host
Epidemiologic triangle Epidemiologic beam

Teklit Angesom 09/18/2020

Cont’d
28

2. Multi-causality of Diseases

Teklit Angesom 09/18/2020

An example of three sufficient causes of disease
Cont…
U
U U
A B
A E B E
I II III
 Assume that these three causes are operating in the diagram

 Without U, there is no disease. U is considered as necessary

cause, but all disease is not due to U alone.

 E causes disease through two mechanisms, II and III, all

diseases arising from II and III are due to E.

 No component cause acts alone, the factors interact with their

complementary factors to produce disease
Teklit Angesom
29
Reading assignment
30

 Contagion theory
 Supernatural theory
 Personal behavior theory
 Miasma theory
 The Germ Theory
 The Life Style Theory
 The Environmental Theory
 The Multi Causal Theory

Teklit Angesom 09/18/2020

Chain model of infectious diseases
31

Chain of infection/transmission cycle

Causative
agent
Reservoir
Susceptible
host
Portal of
exit
Portal of
entry Mode of
transmission

Teklit Angesom 09/18/2020

Cont’d
32

Teklit Angesom 09/18/2020

Levels of Disease Prevention
33

I. Primary prevention
The objectives here are to
-promote health,
-prevent exposure,
-and prevent disease

Teklit Angesom 09/18/2020

Cont’d
34

A. Health promotion (Primordial):

 This consists of general non-specific
interventions that enhance health and the body’s
ability to resist disease – including:

 The improvement of socioeconomic status

through the provision of adequately.
 paid jobs,
 education,
 affordable and adequate housing and clothing, etc.

Teklit Angesom 09/18/2020

Cont’d
35

B. Prevention of exposure:

 There are many examples of interventions aimed at this

stage,

 Relatively to specific compared to primordial prevention

 the provision of safe and adequate water, of proper excreta disposal,

 Provision of vector control;

 Provision of a safe environment at home

Teklit Angesom 09/18/2020

Cont’d
36

C. Prevention of disease:

 An example of intervention, which acts at this stage, is

passive immunization.

 Some times it may be difficult to differentiate interventions

in what form of prevention they involved

Teklit Angesom 09/18/2020

Cont’d
37

II. Secondary prevention

 Interventions that act after the biological onset of disease,
but before permanent damage sets in.

 The objective here is to stop or slow the progression of

disease so as to prevent or limit permanent damage.

 Strategy at this stage is through early detection and

treatment of disease.

Teklit Angesom 09/18/2020

Cont’d
38

III. Tertiary prevention

 Intervention that acts after permanent damage has set
in, and the objective of tertiary prevention is to limit the
impact of that damage.

 The impact can be physical, psychological, social (social

stigma or avoidance by others), and financial.

 Strategy at this stage in general is rehabilitative.

Teklit Angesom 09/18/2020

Levels of Disease Occurrence
39

Diseases occur in a community

 by difference in level of disease at a point in time

 Excess or predictable levels of what is expected

1. Expected level of occurrence of disease

Endemic: the usual presence of disease from low to
moderate level

Hypo/Hyper-endemic: a persistently lower or high

level of disease

Sporadic: Normally does not occur, but occasional cases

occur at irregular intervals

Teklit Angesom 09/18/2020

Cont’d
40

Excess of expected levels

Epidemic: An excess occurrence of disease over
expected level at certain time.

Outbreak: Synonymous with epidemic, but

characterized by a sharp rise and fall in incidence,
(usually to occurrence in a limited area.

Pandemic: An epidemic that affects several countries or

continents. (eg HIV/AIDS)

Teklit Angesom 09/18/2020

Infection and Disease Outcome
41

Exposure to an infectious agent does not

necessarily lead to infection, and

An infection does not necessarily lead to

disease

Infection may remain asymptomatic or sub-

clinical, or may lead to overt clinical disease

Teklit Angesom 09/18/2020

Cont’d
42

Outcomes at each stage of infection

Exposure Infection Disease Disease outcome

Infectiousness Pathogenicity Virulence

Teklit Angesom 09/18/2020

Cont’d
43

1. Infectiousness: the proportion of an exposed

susceptible host who become infected (measured by
infection rate)

2. Pathogenicity: the proportion of infected people who

develop clinical disease, and measured by the clinical-to
sub-clinical ratio

3. Virulence: the proportion of persons with clinical disease

who become severely ill or die, and measured by Case-
fatality-rate and hospitalization rate

Teklit Angesom 09/18/2020

Basic measurements in epidemiology

44

1. Measures of disease frequency

Frequency measures compare one part of the
distribution to another part of the distribution, or to
the entire distribution

Common frequency measures are count, ratios,

proportions, and rates

Teklit Angesom 09/18/2020

Cont’d
45
 Ratio: The numerator and denominator need not be
related
: The values of x and y may be completely
independent, or x may be included in y
: May be same or different variable

 Proportion: a ratio in which x is included in y

 Rate: is often a proportion, with time dimension: it

measures the occurrence of an event in a population
over time (how quickly the event occurs/speed)
Teklit Angesom 09/18/2020
Algorism for distinguishing ratio, proportion and rate

46

Teklit Angesom 09/18/2020

Example
47
 # beds per doctor
 120 beds/10 doctors
 in 2006
 # students per facilitator
 # inhabitants per latrine
 Sex ratio:
2
-Male / Female -----= 0.02 / year
-Female / Male 10
Odds ratio 0
Rate ratio
 Sex :
-Male / whole population
-Female / whole population

Teklit Angesom 09/18/2020

2. Measures of disease occurrence

48

 Prevalence: point and period prevalence

 Incidence: incidence rate/incidence density and
cumulative incidence/incidence proportion
 Attack rate and secondary attack rate
 Measures of prognosis
 Measures of quality of life

Teklit Angesom 09/18/2020

1. Prevalence
49

: measures the population’s disease status

-The proportion of persons in a population who have a
particular disease at a specified point in time or over a
specified period of time

 Expressed as a %tage
 HIV/AIDS in X city in 1999:
 Population 210,000
 Cases 3,200

 Prevalence 1.5%
Teklit Angesom 09/18/2020
1. Point prevalence
50

It is proportion of a population that is affected by

disease at a given point in time
The amount of disease in a population is constantly changing

Thus may not be useful for assessment of diseases with short

generation period
2. Period prevalence
Prevalence in a given period of time

If we want to know how much disease is present over a

longer period of time, period prevalence would be preferred.
Teklit Angesom 09/18/2020
Example 1
51
 A total of 100 people were at risk of disease x, which has no life
time immunity

t1 t2
 What is the prevalence of disease X during time t 1?
 What is the prevalence of disease X during time between t 1 and t2

Teklit Angesom 09/18/2020

Example 2
52
 The following figure represents ten episodes of an
illness in a population of 20 over a period of 16
months. Each horizontal line represents the portion
of time one person spends being ill. The line begins
on the date of onset and ends on the date of death or
recovery. Calculate:

A. Point prevalence on October 1, 1990

B. Period prevalence October 1, 1990 to September
30, 1991

Teklit Angesom 09/18/2020

Cont’d
53

Teklit Angesom 09/18/2020

2. Incidence
54

The number of new events/disease in a defined

population within a specified period of time.

Number of NEW cases of disease during a specific

period

Population at risk during that period

Teklit Angesom 09/18/2020

1. Cumulative Incidence
55

 CI assumes that the entire population is at risk and

is followed up for specified time of period

x
i sk
x
R
CI = 3/12 per yr
x = 0.25 per yr

Month 1 Month 12
Teklit Angesom 09/18/2020
Cont’d…….CI
56
 Denominator is the size of the population at the start of
the time period

 It is a measure of the probability or risk of disease, i.e.,

what proportion of the population will develop illness
during the specified time period

 The probability/risk that an event will occur

 What proportion of the popn develop the event, risk, attack rate
Number of new cases X10n
Population at risk

Teklit Angesom 09/18/2020

2. Incidence density (rate)
57
It indicates how quickly people become ill measured in
people per year
Expressed as number of new cases per person-time at
risk
Person-time can be person-days, person-months,
person years,

Incidence Rate = Number of new cases X10n

Population-time at risk

Teklit Angesom 09/18/2020

Cont’d….IR
58
 Typically, each person is observed from a set beginning
point to an established end point (onset of disease,
death, migration out of the study, or end of the study)

 The numerator is still the number of new cases, but the

denominator is a little different, the sum of the time
each person is observed free of the outcome
It is a good measure in a dynamic cohort

Dynamic cohort is a cohort of people leaving and

entering a study at different time of period
Teklit Angesom 09/18/2020
Cont’d….IR
59

Dynamic cohort

Birth In-migrants

Death Out-migrants

Teklit Angesom 09/18/2020

90 91 92 93 94 95 96 97 98 99 00 Time at risk

A 6.0
B x 6.0
C 10.0
D 8.5
E x 5.0

Total years at risk 35.5

time followed ID = 2 / 35.5 person- years
x disease onset
= 0.056 person-year
Example
61

1. 1000 HIV negative persons were followed for one

year and 50 were found HIV positive.

What is the incidence (cumulative incidence) of HIV

infection?

What is the incidence density (person-time rate) of

HIV infection?

Teklit Angesom 09/18/2020

Estimating Incidence Density
62
Assume disease is acquired on the mid-point of the
interval between the last disease-free visit and the
first visit when disease diagnosed, ID

950 persons not infected = 950 person-years

50 persons at risk for ½ year = 50 x ½ = 25 person-years

50 new cases/975 person-years = .05 case per person-year, or

5.1 cases per 100 person-years.

Teklit Angesom 09/18/2020

Example
63

A. Two surveys were done of the same community 12 months apart. Of

5,000 people surveyed the first time, 25 had antibodies to
Histoplasmosis. Twelve months later, 35 had antibodies, including
the original 25.
1. Prevalence at the second survey:
= antibody positive at second survey = 35
= population = 5,000
= 35/5,000 ×1,000 = 7 per 1,000
2. Incidence during the 12-month period:
= number of new positives during the 12-month period = 35 −25 = 10
= population at risk = 5,000 −25 = 4,975
= 10/4,975 ×1,000 = 2 per 1,000
Teklit Angesom 09/18/2020
attack rate
64

 An is a variant of an incidence rate, applied to a

narrowly defined population observed for a limited
time, such as during an epidemic.

 Number of new cases among the population during the period x100
population at risk at the beginning of the period

Teklit Angesom 09/18/2020

secondary attack rate
65
 A is a measure of the frequency of new cases of a
disease among the contacts of known cases.

Number of cases among contacts of 10 cases during the period x100

total number of contacts

Teklit Angesom 09/18/2020

Example
66

 Seven cases of hepatitis A occurred among 70 children

attending a child care center. Each infected child came from
a different family. The total number of persons in the 7
affected families was 32. One incubation period later, 5
family members of the 7 infected children also developed
hepatitis A.

: Calculate the AR in the child care center and the 20AR

among family contacts of those cases.

Teklit Angesom 09/18/2020

Cont’d….AR
67

Solution A.
 cases of hepatitis A among children in child care center = 7
 number of children enrolled in the child care center = 70
=(7/700)x100=10%
Solution B.
 cases of hepatitis A among family contacts of children with
hepatitis=5
 number of persons at risk in the families (total number of
family members—children already infected) = 32 −7 = 25
=(5/25)x100=20%

Teklit Angesom 09/18/2020

Exercise
68

Teklit Angesom 09/18/2020

Cont’d…exercise
69

Compute
 Period prevalence between Sep.1 and Nov.30

 Point prevalence in Sep.1

 Incidence in month Sep.1 to Oct.1

 Incidence in month Oct.1 to Nov.1

Teklit Angesom 09/18/2020

Comparing Incidence and Prevalence
70

Incidence Prevalence

New cases or events All cases at

over period of time point/period of time

Useful to study factors Useful for measuring

causing disease, size of problem and
disease “risk” planning

Teklit Angesom 09/18/2020

Relationship between Incidence and Prevalence
71

Incidence

 Deaths,
 Cure,
 Lost to
Prevalence follow up

Teklit Angesom 09/18/2020

Factors influencing prevalence
72

Increased by Decreased by
 Longer duration of the disease Shorter duration of the disease

 Prolongation of life of patients High case fatality

without cure
Decrease in new cases (decrease in
 Increase in new cases incidence)

 In-migration of cases In-migration of health people

 Out-migration of healthy people Out-migration of cases

 In-migration of susceptible people Out-migration of susceptible people

 Improved diagnostic facilities (better

Improved cure rate of cases)
reporting)
Teklit Angesom 09/18/2020
Cont’d
73

3. Measures of disease mortality

 Crude rates
 Specific rates
 Adjusted rates

Teklit Angesom 09/18/2020

Cont’d…mortality measures
74
Crude death rate(CDR) =
Total no. of deaths reported during a given time interval x 1000
Estimated mid interval population
Cause specific death rate (CSMR)
No. of deaths from a specific cause during a given time x 100,000
Total no. of mid interval population

 The CDR is the mortality rate from all causes of death for the population
 CSMR The mortality rate from a specified cause for a population

 The denominator for both is the size of the population at the midpoint of
the time period

Teklit Angesom 09/18/2020

Cont’d…mortality measures
75

Proportionate mortality ratio=

No. of deaths from a specific cause during a given time x 100
Total no. of deaths from all causes in the same time
Age- specific mortality rate=
Deaths in a specific age group during a given time X 1000
Estimated mid interval population of specific age group

Sex- specific mortality rate=

No. of deaths in a specific sex during a given time X 1000
Estimated mid interval population of same sex

Teklit Angesom 09/18/2020

Cont’d…mortality measures
76
Infant Mortality Rate
One of the most commonly used measures for comparing health services
among nations.
Deaths among children under 1 year of ageX1000
Number of live births

Perinatal Mortality Rate:

Number of stillbirths 28 weeks or more and infant deaths under 7 days
Number of live and still births 28 weeks or more in the same year

Neonatal Mortality Rate:

Number of deaths among children under 28 days of age in a year
Number of live births in the same year
Teklit Angesom 09/18/2020
Cont’d…mortality measures
77

Post neonatal mortality rate=

Infant death aged from 28 days -1 yearX1000
total live birth
Under- five mortality rate=
No. of deaths of under five years of age X1000
total live birth
Child Mortality Rate:
Number of deaths in children aged 1-4 years in a year
Number of children aged 1-4 in the same year

Teklit Angesom 09/18/2020

Cont’d…mortality measures
78

Maternal Mortality Ratio=

No. of pregnancy associated deaths of mothers
in a given time X 100,000
No. of live births in the same time

* Actually a ratio used to measure mortality associated with

pregnancy

Teklit Angesom 09/18/2020

Cont’d…mortality measures
79

Case Fatality Rate (CFR)=

No. of deaths from a specific disease during a given time x 100
No. of cases of that disease during the same time

Teklit Angesom 09/18/2020

Standardization of rates and ratios
80
Virtually every large An overall measure that does
population is heterogeneous in not take explicit account of
regard to the composition of the
 socio-demographic (e.g., age, population is called crude
gender, education, religion),
 geographic,
 Crude is an average of the
 genetic,
values for the individual
 occupational,
subgroups, weighted by their
 dietary,
relative sizes
 medical history,
 and innumerable other
 The larger the subgroup, the
personal attributes and
environmental factors related more influence it will have
to health on the crude measure
Teklit Angesom 09/18/2020
Cont’d…standardization
81

 Comparison of rates across populations or time

periods having different age composition

Table 1 in 1970, 5,022 out of the 562,887 white women

in Miami died, and 285 of the 106,917 white Alaskan
women died

CDR= 8.92 and 2.67 per 1,000 respectively

What do you conclude from this?

Teklit Angesom 09/18/2020
Cont’d…standardization
82

Teklit Angesom 09/18/2020

Cont’d…standardization
83

 Although the crude rates suggest that the force of mortality is

stronger in Miami than in Alaska, the table shows that for
any given age the two populations have very similar mortality
rates

 Then what accounts for the difference in the crude death

rates?

 Look at the age distributions in both states

 Which state have much greater proportion of women in older

age groups…..higher mortality
Teklit Angesom 09/18/2020
Cont’d…standardization
84
 Two populations may have the same overall size and
identical age-specific death rates, but different total
numbers of deaths and different overall death rates,
due to differences in their age distributions

 Standardization seeks to provide numbers and

comparisons that minimize the influence of age

Teklit Angesom 09/18/2020

Cont’d…standardization
85

1. Direct standardization method

 The stratum-specific rates of study populations are

applied to the age distribution of a standard
population

 When the same standard is used and if two study

populations have the same age-specific rates, then
their directly standardized rates will be identical,
independent of the age distributions in the study
populations
Teklit Angesom 09/18/2020
Cont’d…standardization
86
Where
• r= rate for each stratum
of the study population

• N= number of persons
for each stratum of the
standard population

• ∑= means summation
over each strata

Teklit Angesom 09/18/2020

Cont’d…standardization
87
DSR Miami= DSR Alaska=
=(1.19x23,961)+0.71x15,420 =(1.59x23,961)+(0.9x15,420)
+…(39.11x10,685) +…(39x10,685)=
91,028 91,028
=6.92 deaths per 1000 =6.71 deaths per 1000

Teklit Angesom 09/18/2020

Cont’d…standardization
88

2. Indirect standardization
 When stratum-specific numbers are small, stratum-
specific rate estimates are too susceptible to being
heavily influenced by random variability

 It takes stratum-specific rates from a standard

population of sufficient size

 Direct standardization, the study population provides

the rates and the standard population provides the
weights….here vise versa
Teklit Angesom 09/18/2020
Cont’d…standardization
89
 SMR= Observed death
Expected death
 Observed death= ∑ dk
 Expected death= ∑(stratum specific rates from standard

population x stratum specific weights from study

population)= Rknk

Where:
Dk=number of deaths in the kth stratum of study population
Rk=death rates in the kth stratum of standard population
nk=size of the kth stratum of the study population
Teklit Angesom 09/18/2020
Cont’d…standardization
90

The choice of method may be affected by

 Direct standardization requires the age-specific rates in all
the populations under study

 Indirect standardization only requires the total number of

deaths (or cases) and the age structure of the study
population, and thus indirect standardization may be the
only feasible method if age-specific rates are not available

 Indirect standardization is preferable when there are small

numbers in particular age groups

Teklit Angesom 09/18/2020

Individual Assignment
91

1. Choice of Standard Population in

standardization

2. Epidemiologic Disability measures and

disease prognosis measures

Teklit Angesom 09/18/2020