This document discusses mortality measurements and standardization of crude death rates. It defines key mortality indicators such as crude death rate, age-specific death rate, infant mortality rate, and under-five mortality rate. It also explains how to directly standardize crude death rates to account for differences in population age structures and allow better comparison between locations. Direct standardization involves applying the age-specific death rates of each area to a standard population distribution to calculate expected deaths and a standardized crude death rate.

1. Mortality &
Standardization of Death Rates
Dr. Min Ko Ko
M.B.,B.S, M.Med.Sc. (Public Health)
Ph.D. in Demography (Mahidol University)
Associate Professor
Health Behaviour & Communication Department
University of Public Health

2. Ground Rules for today discussion?
1-
2-
3-
4-
5-
6-
7-

3. • Outline of Presentation
– Cover 2 topics:
1) Mortality measurements
2) Standardization of crude death rates

4. Warming up

5. Mortality : Meanings
• The process whereby deaths occur in a
population.
• The scientific study of demography began
with the study of mortality
– John Graunt- Bills of Mortality

6. Source of Statistics on Mortality
Source Characteristics
Death certificates Demographic and social characteristics of
the deceased
Details about the cause of death
Vital statistics Detail data are from death certificates
Cross-national data
(eg: Census, Survey)
Published comparative data:
Eg. ‘Demographic Yearbook’,
‘WHO’s World Health Statistics Annual’

7. Most commonly used, easily understood & calculated
Crude death rate (CDR)
Age-specific death rate (ASDR)
Mortality in early life including:
Infant mortality rate (IMR)
Under-five mortality rate (U5MR)
Cause Specific Death Rate
Maternal Mortality Ratio (MMR)
Case Fatality Rate (CFR)
Direct Measurements of Mortality

8. Example:
Deaths = 534,000
Mid-year pop. = 52,426,000
CDR = ?
Crude death rate (CDR)
CDR = 534,000/52,426,000 x 1,000 = 10.2 deaths/1000 MYP
CDR is defined as the number of people dying in a given year
divided by the number of people in the population in the middle of
that year. Conventionally it is expressed per 1000 persons.
CDR =
D
MYP
x K
D = total number of deaths in a given year
MYP = total population of the middle of the year
K = a constant usually taken as 1,000
Q: Where Denominator & Numerator come from?

9. Crude Death Rate (CDR)
Advantages:
• Simplest, Easy to understand.
• Easy to get data as denominator used MYP
• Can measure the natural growth rate.
• Can apply in DTM.
Disadvantages:
It is not a reliable indicator of comparative mortality levels.
• Can be affected by age & sex of population.
• Can’t make international comparison!

10. Q: Possible or not!
• CDR of developed country is higher than CDR of developing
country
Brain storming Q:!
Answer: Possible!
CDR of developed country may have higher than developing
country.
It does not mean that people in developing country are
healthier. It is because of age composition of the two population.
Other things being equal, the older a population, the greater
the annual number of death, and the higher the CDR

11. International Comparison on mortality data
CDR
Yes No
Why: affected by age & sex of population!
Age Specific Cause Specific
Age Specific Death Rate (ASDR)
Mortality in early life including:
Infant mortality rate (IMR)
Under-five mortality rate (U5MR)
--------
--------
--------
older a population ++> the greater annual
number of death > higher the CDR

12. Age Pattern of Mortality
• J Shaped curve (Mortality Curve)
– Mortality is high in the first year of life, drops rapidly to a minimum in
the age group 10-14, rises slowly to about age 50 after which it
increases at an ever accelerating rate.
• Different age groups in a pop: are exposed to different risk of
dying b/c of occupation, age, other characteristics.
– Death rate are highest for the very young and very old.
– Older person are more likely to die than younger person.
– Frontline army men are more likely to die than clerical worker.

13. 0.0100
0.1000
1.0000
10.0000
100.0000
1000.0000
Brunei
Darussalam
Cambodia Indonesia Lao People's
Democratic Republic
Malaysia Myanmar Philippines Singapore
Thailand Timor-Leste Viet Nam
ASDR(per1000)
Line Diagram of Age Specific Death Rates
of South East Asia Region in 2010
Source: United Nations, Department of Economic and Social Affairs, Population Division (2013).
World Population Prospects: The 2012 Revision, DVD Edition

14. Age-specific death rate (ASDR) :
ASDR =
Dx
MYPx
x 1,000
Dx = total number of deaths at specific age x in a given year
MYPx = total population of the middle of the year
K = a constant usually taken as 1,000
Age-specific death rate: measures the incidence of death at
each age
Numerator come from VS, while Denominator come from
census, survey or pop: estimates.
Q: Where Denominator & Numerator come from?

15. Please calculate!

16. 20 wks 28 wks 40 wks
At birth
7 days 28 days 1 yr
Late Fetal
Death
(Still birth)
Neonatal
death
Infant death
Perinatal Mortality
Post Neonatal
death
Fetal Death
Early
Neonatal
death
Late
Neonatal
death
Rate vs Ratio
Mortality in Early Life

17. 40 wks
At Birth
4 year 5 year
Child Mortality Rate(1-4 yrs)
Infant Mortality Rate
Under 5 Mortality Rate
1 year
Rate vs Ratio

18. Perinatal mortality rate (PMR):
Stillbirth rate (SBR):
Further Measures of Mortality in Early Life
PMR =
Annual deaths between 28 weeks gestation
and 7 days after birth
Total live births and stillbirths in a calendar
year
X 1,000
SBR =
Annual deaths from 28 weeks gestation
Total live births and stillbirths in a calendar
year
X 1,000

19. Neonatal mortality rate (NMR):
Further Measures of Mortality in Early Life
PNMR=
Deaths between 29 days and one year after birth
Total live births in a calendar year
X 1,000
Post-neonatal mortality rate (PNMR):
NMR=
Deaths in the first 28 days after birth
Total live births in a calendar year
X 1,000

20. IMR = Deaths under 1 year of age x 1,000
Total live births in a calendar year
IMR : is a key indicator of demographic development and health
conditions in different countries
Example:
Infant deaths = 63,813
Births = 1,737,000
IMR = ?
Infant mortality rate (IMR) :
IMR = 63,813/1,737,000 x 1,000
= 36.7 infant deaths per 1,000 live births

21. Advantages:
 It is a key indicator of demographic development & Health
conditions in different countries. (MDG, SDG indicator)
Disadvantages:
 Infant deaths are not evenly distributed through the first year of life
b/c a high proportion of infant mortality occur in the first month of
life.
 Seasonal fluctuation in birth
 Many babies are born and die in the same calendar year and are
omitted in a count of pop <1 yr at both the beginning and the end of
the year.
 Greater under-enumeration of infants b/c many parents do not think
of infants as persons when asked.
Infant mortality rate (IMR) (cont: )

22. Under-five mortality rate (U5MR) :
U5MR = Deaths under 5 year of age x 1,000
Total live births in a calendar year
U5MR : is a key indicator of demographic development and health
conditions in different countries
Example:
Under 5 deaths = 96, 356
Births = 1,737,000
U5MR = ?
Under-five mortality rate (U5MR) :
U5MR = 96,356/1,737,000 x 1,000
= 55.5 under 5 deaths per 1,000 live births

23. Cause-specific death rate (CSDR): Identify and address
specific health problems
The denominator of CSDR is not strictly the population at risk,
because the risk of dying from many causes varies by age and or
sex.
Cause-specific death rates, therefore, are mostly calculated
by sex and age
Cause-specific Death Rates
CSDR=
Deaths in a calendar year from a particular cause
Total mid-year population
X 100,000

24. Maternal Mortality Ratio (MMR)
- MDG, SDG indicator
MMR =
No: of deaths from puerperal causes in a given
population in a given year
No: of LBs registered in the same population in the
same area in the same year
x
1,000

25. CFR =
No of deaths assigned to a specific disease
Number of cases of the disease
X 1000
Case Fatality Rate

26.  Gender [male vs female]
 Marital status [married vs single]
 Ethnic group
 Religion [Buddhist vs Chritians vs Hindu vs Islam vs others]
 Occupation [high vs low] [Gov: staff vs Non-Gov staff]
 Income [high vs low]
 Residence [Rural vs Urban]
 Educational status [Edu: vs Non-edu:]
 Family structure
 [single mother vs both parents]
 [Nuclear family vs Extended family]
 Minority vs Majority
 Migrant vs non-migrant
Differentials in mortality
The risk of death varies according to people’s lifestyles
and life chances.
Brain storming Q:!

27. Break x 10 minutes

28. Standardization Of Crude Death Rates

29. Which area has higher mortality than the other one?
Mid-year
population
Deaths Death rate
(per 1,000)
Area A 5,000 185 = 185/5,000 x1,000
= 37
Area B 5,000 165 = 165/5,000 x1,000
= 33

30. Which area has higher mortality than the other one?
Mid-year
population
Deaths Death rate
(per 1,000)
Area A
0-44 1000 25 = 25/1,000 x1,000 = 25
45+ 4000 160 = 160/4,000 x1,000 = 40
Area B
0-44 4000 120 = 120/4,000 x1,000 = 30
45+ 1000 45 = 45/1,000 x1,000 = 45

31. Which area has higher mortality than the other one?
Mid-year
population
Deaths Death rate
(per 1,000)
Area A 5,000 185 = 185/5,000 x1,000 = 37
0-44 1000 25 = 25/1,000 x1,000 = 25
45+ 4000 160 = 160/4,000 x1,000 = 40
Area B 5,000 165 = 165/5,000 x1,000 = 33
0-44 4000 120 = 120/4,000 x1,000 = 30
45+ 1000 45 = 45/1,000 x1,000 = 45

32. Standardization of Crude Death Rates
3 possible causes of differences in CDRs:
1)Defective of data
2)Differences in age-specific rates of mortality
3)Differences in the age composition of the populations
To compare two, or more than two rates:
•Standardization
 Advantages: removing confounding effect of age structure
variations
 Policy & planning for facilities/resources allocation

33. Standardization of Crude Death Rates
There are 2 types of standardization.
1) Direct standardization
 Each area/country has ASDRs
 It employs a “standard age structure”
 (ie: If Myanmar has Population age structure of Thailand,
>>>>>>)
2) Indirect standardization
 Age structure and total deaths are available but ASDRs are not
available for each area.
 It employs a “Standard ASDR”
 (ie: If Myanmar has ASDR of Thailand, >>>>>>)

34. Standardization of Crude Death Rates
Direct standardization : steps of calculation
1. Select standard population
–Select a population age distribution (age structure) to be a “standard
population”
2. Calculate expected deaths and CDR
–Multiply ASDRs of each area by number of standard population in each
age group and sum of the deaths
–Divided sum of the deaths by total number of standard population
Differences between standard CDR and expected CDR are due
only to mortality levels of each area

35. Standardization of Crude Death Rates

37. Standardization of Crude Death Rates
Indirect standardization : steps of calculation
1. Select standard population
–Use age-specific death rates of standard population.
2. Calculate expected deaths and CDR
–Multiply population in each age group by ASDRs of standard population
and sum of the deaths.
3. Calculate standardized mortality ratio (SMR)
–Divided observed deaths by expected deaths
4. Calculate standardized CDR
–Multiply CDR of standard population by SMR

38. Standardization of Crude Death Rates

40. Calculation
1) Direct Standardization
2) Indirect Standardization
– Lets’ try together

41. What you learned today?
 ___________________________
 ___________________________
 ___________________________
 ___________________________
 ___________________________
 ___________________________
 ___________________________
 ___________________________
Conclusion

42. • DAVID LUCAS AND PAUL MEYER. 1994. Beginning of
Population Studies, National Center for Development Studies,
Australian National University
• PALMORE, J. A. & ROBERT W. GARDNER. 1983. Measuring Mortality,
Fertility, and Natural Increase: A Self- Teaching Guide to
Elementary Measures. Honolulu: The East-West Center.
Chapter I (p.1-7)
• PATAMA VAPATTANAWONG. August 28, 2014. PRRH 555: Analytical
Techniques in Demography. Session 4: Mortality Measures.
Mahidol University
• ROWLAND, D.. 2003. Demographic Methods and Concepts.
Oxford : Oxford University Press. Chapter I (p.13-44)
More Readings & References