EPIDEMIOLOGIC DATA SOURCES

AND MEASUREMENT
CHAPTER 2
 Learning Objectives
1. Describe sources of health data and provide a
situation for discussion and analysis by the class
2. Recognize and recall the different types of
epidemiological measurements
3. Analyse the different types of information that
are relevant to public health practice
4. Assess the types of information routinely
available on population morbidity and mortality
5. Apply epidemiologic data sources and
measurements and apply it to actual data collected
and gathered from different sources with
interpretation and analysis
6. Analysing types of data used to create community
rates and ratios for planning and evaluating health
progress and/ or health status of a population
7. Identify a situation for discussion and analysis by
the class.
MOST COMMONLY USED RATES FROM
POPULATON DATA
No. live births
Crude Birth Rate = ---------------------- X 1000
Mid period population

No. of Deaths
Crude Death rate = --------------------------- X 1000
Mid period population
 No. of deaths in a
particular age group
Age Specific Death Rate = -------------------- X 1000
Mid period
population
( same age group )
 No. of deaths due to a particular cause
Cause Specific Death Rate = ------------- x100,000
Mid period population
Infant Mortality Rate = IMR
No. of deaths <1 year old
IMR = ------------------------------------------ X 1000
No. of live births
 No. of deaths <28days old
Neonatal Mortality Rate = ----------------- X 1000
No. of live births

Post Neonatal mortality rate = IMR - NMR

 no. stillbirths + no. of deaths < 7days old
Perinatal mortality rate = --------------------- X
1000
no. stillbirths + no. live births
Maternal Mortality Rate = MMR
No. of pregnancy related deaths
MMR = --------------------------------------- X 100,000
No. of live births

• defined place and time period

• Same place and time period
 SOURCES OF HEALTH DATA
Denominator Data define the population at risk
Numerator Data define the events or conditions of
concern

Census statistics often are used in the denominator

Statistics gathered from health, disease, birth and

death registries and surveys are used in the
numerator data
 National Notifiable Disease
Surveillance System
- Physicians, Hospitals, Clinics, and Laboratories
are required to report all cases of many
infectious diseases and certain non infectious
diseases to local health authorities
 Epidemiologic Measurements
- The fundamental epidemiologic measure is
the frequency with which an event of interest
(ex. disease, injury or death) occurs in a
population to be studied.
 EPIDEMIOLOGIC MEASUREMENTS
FREQUENCY
- The frequency of a disease, injury or death can
be measured in different ways and it can be
related to different denominators
Incidence – is the frequency ( number ) of
occurrences of disease, injury or death in the
study population during the time period of the
study
Prevalence ( Prevalent Cases )
- Is the number of persons in a defined
population who have a specified disease or
condition at a point in time, usually the time a
survey is done
Difference Between Point Prevalence and Period
Prevalence

Point Prevalence is meant the prevalence at a point in

time

Period Prevalence
- refers to the number of persons who had the disease at
any time during the specified time interval.
- It is the sum of the point prevalence at the beginning
of the interval plus the incidence during the interval
RISK
- Epidemiologically, risk is defined as the
proportion of persons who are UNAFFECTED
at the beginning of the study period, but who
undergo the risk event during the study period
- Risk event maybe death, disease or injury and
the persons at risk for the event at the
beginning of the study period are called a
COHORT.
RATES
- Is the frequency ( number ) of events that
occur in a defined time period, divided by the
average number of people at risk during the
period being studied multiplied by a constant
multiplier
 Definitions of Specific Types of Rates
Incidence Rate
- Is calculated as the number of incident cases
over a defined study period divided by the
population at risk at the midpoint of that
study period
- Is usually expressed per 1,000, per 10,000 or
per 100,000
Prevalence Rate
- Is actually a proportion and not a rate
- Indicate a proportion (usually expressed as a
percentage ) of persons with a defined disease
or condition at the time they are studied
Use of Crude Rates vs. Specific Rates
- Rates applied to an entire population without
reference to any characteristics of the individuals
in it are crude death rates
- When a population is divided into a more
homogenous subgroups based on a particular
characteristics of interest ( ex. age, sex, race, risk
factors) and rates are calculated within these
groups, the rates are called SPECIFIC RATES ex.
age-specific, sex specific rates.
 Commonly Used Rates That Reflect
Maternal and Infant Health
- Infant Mortality Rate reflect the health of mothers
and infants
Infant Mortality Rate ( IMR )
- The IMR is often used as an overall index of the
health status of a nation
- Reason: Because the health of infants is unusually
sensitive to maternal health practices ( especially
the mother’s nutrition and use of tobacco,
alcohol, and illegal drugs ) environmental factors
and the quality of health services.
 - Most infant deaths occur in the first week of life
and are due to prematurity or intra- uterine
growth retardation and the resulting immaturity,
which often leads to respiratory failure
Neonatal and Post Neonatal Mortality Rates:
Neonatal deaths are those occurring in infants
younger than 28 days of age, whereas post neonatal
are those occurring in infants from the 28th day of
life to the first birthday
Perinatal Mortality Rate and Ratio:
- Perinatal means “around the time of birth”
- Stillbirths are included in the numerator to
capture the deaths around the time of birth
- Stillbirths are included in the denominator
because of the criteria for a valid rate
Perinatal Mortality Ratio – in which the
denominator does not include still births
Primary Used of the Perinatal Mortality Rate:
- To evaluate the care of pregnant women
before and during delivery
- Care of the mothers and their infants in the
immediate post partum period
Maternal Mortality Rate
- The equation is based on the number of pregnancy
related (puerperal) deaths.
- Most common cause of maternal deaths or puerperal
deaths are hemorrhage, infection, and toxemia of
pregnancy.
- Pregnancy also complicates the course of other
conditions, such as heart disease, diabetes and
tuberculosis
- A useful measure of the progress of a nation in
providing adequate nutrition and medical care for
pregnant women
- The equation is based on the number of
pregnancy related (puerperal) deaths. In cases
of accidental injury or homicide, the death of a
woman who is pregnant or has recently
delivered is not usually considered “pregnancy
related”
Definition of Terms:
Live Birth – the delivery of a product of
conception that shows any sign of life after
complete removal from the mother
Sign of life may consist of a breath or a cry, any
spontaneous movement, a pulse or a heart beat
or pulsation of the umbilical cord
 Categories of Foetal Death
Miscarriage – occurs when a dead foetus is
delivered within the first 20 weeks of gestation. This
is called early foetal death

Intermediate foetal death – is one in which a dead

foetus is delivered between 20 and 28 weeks of
gestation

Stillbirth (Late foetal death) – a foetus born dead at

28 weeks of gestation or greater
Over the course of 7 years 146 people were
referred to a hospital plastic surgery unit
because they had been bitten by a dog. Details
of the breed of dog causing the bite were
collected from 107 people. The main breeds
responsible were as follows:
Staffordshire bull terrier 15 cases
Jack Russell 13 cases
Medium-sized mongrel 10 cases
Alsatian 9 cases
Labrador 8 cases
Collie 6 cases

Does this mean that Staffordshire bull terriers are more likely
to bite people than collies?
If your answer is NO or perhaps NOT SURE, what other
information would you like before you could answer this
question properly?
Here are some figures from the hypothetical town of
Barking. In Barking, in 2005, 20 people were bitten by
Staffordshire bull terriers and 15 people were bitten by
collies. Barking has a dog registration scheme and
assuming that all dogs are registered, it is known that in
2005 the average population of Staffordshire bull terriers
in Barking was 200, and collies was 150. Two dog owners
are having a fierce debate over which breed is more
likely to bite people.
What do you think is the answer to the question of
which dog bites more? And Why?
A nursing home has 100 residents. On the first day
in January ten residents had a cold. Over the month
of January another 18 residents developed a cold.
Assuming that the number of residents did not
change over January, answer the following
questions:
What proportion of the residents had a cold on the
first day of January?
What proportion of the residents had a cold some
time during the month of January?
The table below shows the number of deaths, and
mid-year populations for England and Wales and three
surveillance sites combined in Tanzania in 2001.
Complete the table by calculating the crude death rates.

Total deaths Mid-year Death

pop. rate per
1000 pop.
England and
Wales 530, 373 52,084,000
Tanzania 3, 625 345, 935
What is the answer? Is this the answer you
expected?

Does this mean, in your view, that the risk of

death in these two populations is about the
same?

What reasons can you think of for the similarity

in the crude rates?
 Deaths per 1000 persons per year
Cause of Tot. Non All cig. >25/day
Death pop. smokers smokers

All causes 14.05 12.06 16.32 19.67 Lung CA

0.65 0.07 1.20 2.23
CHD 3.99 3.31 4.57 4.97
What are the risks of death in the non-smokers and
those smoking 25 or more cigarettes a day?

What is the risk of death from lung cancer in

non-smokers, and the risk of death from lung cancer
in those smoking 25 or more cigarettes a day?

Can you conclude from the figures in the table that

smoking increases the risk of death?
Taking the figures from the table, the relative risk of
all those smoking for death from lung cancer is?

How many times then that those smoking are more

likely to die from lung cancer than non-smokers?

The relative risk of death from lung cancer for heavy

smokers compared to non-smokers?
Now calculate the relative risk of death from
coronary heart disease for smokers compared to
non-smokers and heavy smokers compared to
non-smokers?