WHO methods and data sources
for life tables 1990-2015
Department of Information, Evidence and Research
WHO, Geneva
May 2016
Global Health Estimates Technical Paper WHO/HIS/IER/GHE/2016.2
Acknowledgments
This Technical Report was written by Colin Mathers and Jessica Ho with inputs and assistance from Dan
Hogan, Wahyu Retno Mahanani, Doris Ma Fat and Gretchen Stevens. WHO life tables were primarily
prepared by Jessica Ho and Colin Mathers of the Mortality and Health Analysis Unit in the WHO
Department of Information, Evidence and Research (in the Health Systems and Innovation Cluster of
WHO, Geneva). Data and methods for the 2016 update of life tables were developed with advice and
assistance from a WHO Lifetables Working Group, established by the WHO Reference Group on Global
Health Statistics. We also drew on advice and inputs from Interagency Group on Child Mortality
Estimation (UN IGME), the UN Population Division and UNAIDS. We would particularly like to note the
assistance and inputs provided by Jeffrey Eaton, Patrick Gerland, Stephane Helleringer, Mary Mahy,
Bruno Masquelier, Francois Pelletier, John Stover, John Wilmoth and Danzhen You.
Estimates and analysis are available at:
http://www.who.int/gho/mortality_burden_disease/en/index.html
http://www.who.int/gho
For further information about the estimates and methods, please contact [email protected]
In this series
1. WHO methods and data sources for life tables 1990-2011 (Global Health Estimates Technical Paper
WHO/HIS/HSI/GHE/2013.1)
2. WHO-CHERG methods and data sources for child causes of death 2000-2011 (Global Health Estimates
Technical Paper WHO/HIS/HSI/GHE/2013.2)
3. WHO methods and data sources for global causes of death 2000-2011 (Global Health Estimates
Technical Paper WHO/HIS/HSI/GHE/2013.3)
4. WHO methods and data sources for global burden of disease estimates 2000-2011 (Global Health
Estimates Technical Paper WHO/HIS/HSI/GHE/2013.4)
5. WHO methods for life expectancy and healthy life expectancy (Global Health Estimates Technical
Paper WHO/HIS/HSI/GHE/2014.5)
6. CHERG-WHO methods and data sources for child causes of death 2000-2012 (Global Health Estimates
Technical Paper WHO/HIS/HSI/GHE/2014.6)
7. WHO methods and data sources for country-level causes of death 2000-2012 (Global Health Estimates
Technical Paper WHO/HIS/HSI/GHE/2014.7)
8. MCEE-WHO methods and data sources for child causes of death 2000-2015 (Global Health Estimates
Technical Paper WHO/HIS/HSI/GHE/2016.1)
Contents
Acknowledgments...........................................................................................................................................
Abbreviations ..................................................................................................................................................
1
Introduction .......................................................................................................................................... 1
2
WPP2015 life tables............................................................................................................................... 1
3 General approach for preparation of annual life tables ........................................................................... 2
3.1 Interpolation of mx for countries in the WPP and VR categories ....................................................... 4
3.2 Interpolation of mx for high HIV and “Other HIV” countries.............................................................. 6
4 Adjustments using death registration data ............................................................................................ 20
4.1 Updated assessments of VR completeness ...................................................................................... 40
4.2 Revision of imputed annual death rates ........................................................................................... 42
5 Neonatal, infant and under five mortality ............................................................................................ 133
6 WHO estimates of life expectancy and healthy life expectancy ........................................................... 133
References ................................................................................................................................................ 155
Annex A: Data sources and methods for WHO Life Tables....................................................................... 177
Annex B: Data sources and methods for mortality shocks ......................................................................... 34
Annex C: Estimated completeness of death registration data ................................................................... 41
Annex D: Estimated completeness of death registration data for most recent year ................................. 48
Annex E: Comparison of 45q15 estimates, WPP2015 and GHE2016 ......................................................... 49
ABBREVIATIONS
ARR
Annual rate of reduction
ART
Anti-retroviral therapy
CD
Coale Demeny
DHS
USAID-supported Demographic and Health Surveys
EPP
UNAIDS Epidemic Projection Package
GHE2015
WHO Global Health Estimates 2015
GHE2013
WHO Global Health Estimates 2013
ICD
International Classification of Diseases
IFHS
Iraq Family Health Survey 2006-2007
IHME Institute for Health Metrics and Evaluation
IMR
Infant mortality rate
MICS
UNICEF Multiple Indicator Cluster Surveys
mx
age-specific death rates calculated from information on deaths among persons in the age group
commencing at age x during a given time period and the total person-years for the population in
the same age group during the same time period. For WHO and WPP2015 abridged life tables,
all age groups from 5 onwards are 5-year age groups, m0 refers to infants aged 0 (first 12
months of life) and m1 refers to children aged 1 to 4 (ie. between exact ages 1 and 5).
NMR
Neonatal mortality rate
PCHIP Piecewise cubic Hermite interpolating polynomials
PMTCT Prevention of Mother to Child Transmission of HIV
PRIO
Peace Research Institute Oslo
nqx
probability of dying between exact ages x and x+n.
U5MR Under-5 mortality rate
UCDP Uppsala Conflict Data Program
UN-IGME
Inter-agency Group for Child Mortality Estimation
UNPD UN Population Division
VR
Vital registration
WHO
World Health Organization
WHS
2002 to 2003 WHO World Health Survey program
WPP2015
World Population Prospects 2015
1
Introduction
The World Health Organization (WHO) began producing annual life tables for all Member States in 1999.
These life tables are a basic input to all WHO estimates of global, regional and country-level patterns
and trends in all-cause and cause-specific mortality. After the publication of life tables for years to 2009
in the 2011 edition of World Health Statistics, WHO has shifted to a two year cycle for the updating of
life tables for all Member States, and has moved towards alignment of this revision cycle with that of the
World Population Prospects produced biennially by the UN Population Division.
Since 1998, WHO has been producing annual abridged life tables for Member States as part of its
mandate to monitor and report on global progress in improving health. During the MDG era, WHO has
been estimating time series of life tables from 1990 onwards. To support its reporting on progress
towards the 2030 Agenda for Sustainable Development, WHO has released updated annual life tables
for Member States for the period 1990-2015. These are available in the WHO Global Health Observatory
(1) and in World Health Statistics 2016 (2). These updated life tables also provide the all-cause mortality
estimates for the WHO Global Health Estimates 2015 (GHE2015) to be released in 2016.
In recent years, WHO has liaised more closely with the United Nations Population Division (UNPD) on life
tables for countries, in order to maximize the consistency of UN and WHO life tables, and to minimize
differences in the use and interpretation of available data on mortality levels. For almost all WHO
Member States, this update draws on the World Population Prospects 2015 (WPP2015) life tables
prepared by the UN Population Division (UNPD) (3), as well as on infant and under-5 mortality rates
(U5MR) have been developed and agreed upon by the Inter-agency Group for Child Mortality Estimation
(UN-IGME) which is made up of WHO, UNICEF, UNPD, World Bank and academic groups (4). death
registration data reported to WHO by Member States (5), and UNAIDS/WHO estimates of HIV mortality
for countries with high HIV prevalence (6).
The WHO Reference Group on Global Health Statistics convened a WHO Life Table Working Group in
December 2014 to advise WHO on data, methods and revision strategies for WHO life tables. This group
included academic experts as well as representatives of WHO, UNAIDS and UNPD and provided WHO
with advice on the GHE2015 revisions through 2015 to date.
Consultations with Member States were carried out for estimates of neonatal, infant and child mortality
in 2015, and for HIV mortality data, assumptions and estimates by UNAIDS in 2015 and 2016. Estimates
for non-HIV mortality for ages 5 and over are closely aligned with the WPP2015 life tables, with
adjustments for annual variations reported in death registration data and for mortality shocks (conflict
and natural disasters). As a result, WHO did not carry out a separate consultation for the life table
mortality rates, but will make these available to Member States for information, and for comments and
new data to contribute to future revisions.
2. WPP2015 life tables
WPP2015 was released in mid-2015 (2) and includes abridged life tables for five-year periods 19501955,……,2095-2100, for age groups 0, 1, 5, 10,….., 100+ by sex . Life tables are available for 183 of the
194 WHO Member States and for 3 non-Member territories with substantial populations (Occupied
Palestinian Territory, Puerto Rico and China: Province of Taiwan). The WPP2015 excluded the following
World Health Organization
Page 1
11 Member States (all with population <90,000 in 2015): Andorra, Cook Islands, Dominica, Marshall
Islands, Monaco, Nauru, Niue, Palau, Saint Kitts and Nevis, San Marino, and Tuvalu. The most recent
Global Health Estimates for causes of death (7) excluded 22 Member States with a population of less
than 550,000 in 2015. For the current update, life tables have been prepared for the 183 WHO Member
States included in WPP2015, as well as for the 3 largest non-Member territories. The latter will not be
released, but included only for the calculation of regional and global life tables and all-cause mortality.
For a group of 21 “high HIV countries”, the WPP life tables were developed using Spectrum to model the
HIV epidemic using Spectrum inputs and assumptions for HIV provided by UNAIDS in mid-2014. For
these countries, UN Population Division has provided estimates of non-AIDS death rates in the form of
model life tables indexed by e0 time series and specification of the model life table variant used for each
country (mostly Coale Demeny North). This allows calculation of implied age-sex-specific HIV death rates
for these countries.
3
General approach for preparation of annual life tables
For this update, the objective was to publish WHO annual period life tables for years 1990-2015. For
internal use in other analyses, we also aimed to prepare annual life tables for 1985-1989.The starting
point for the preparation of WHO annual life tables was to interpolate annual values for the age-specific
mortality rates mx from the WPP2015 5-year period average mx for each age-sex-country time series.
The WPP2015 uses the convention that a specified year (eg. 2015) refers to 1 July. Use of a hyphen (-)
between years, for example, 1995-2000, signifies the full period involved, from 1 July of the first year to
1 July of the second year. WHO references to calendar years (eg 2015) refer to the period 1 January to
31 December and statistics either refer to averages or totals for the calendar year period. In practice,
the annual average mortality rate for a calendar year is assumed to be essentially the same as the
mortality rate at 1 July. Similarly, we assume that the 1 July population for year T is a proxy for the
average population of calendar year T.
So if PT is the 1 July population for calendar year T, then the average population for the quinquennial
period 2010-2015 is the person-years for the period 1 July 2010 to 1 July 2015 divided by 5:
Average population = (0.5*P2010 + P2012 + P2010 + P2013 +P2014 + 0.5*P2015)/5
For purposes of interpolation, we assume that the quinquennial average death rate falls in the centre of
the quinquennial period eg. 1 January 2003 = 2002.5 This is usually an adequate approximation to the
extent that trends are reasonably linear across the quinquennial interval.
For interpolation of mx values for annual periods 1985, 1986, …..2015 we used piecewise cubic Herite
interpolating polynomials (usually referred to as PCHIP). This has the desirable property that he
piecewise cubics join smoothly, so that both the interpolated function and its first derivative are
continuous. In addition, the interpolant is shape-preserving in the sense that it cannot overshoot locally;
sections in which period mx is increasing, decreasing or constant with time remain so after
interpolation, and local extremes (maxima, maxima) also remain so (8). PCHIP interpolation was
World Health Organization
Page 2
implemented using a procedure called pchipolate.ado available for Stata from the Statistical Software
Components (SSC) Archive, often termed the "Boston College" archive (9).
When mx is not monotonically changing over time, and the mx for one period represents a local
maximum or minimum, the interpolated annual mx will result in a period average mx that is lower than
the local maximum input mx, or higher than the local minimum. This is illustrated in the following plot.
0.4
period average mx
0.35
3rd iteration
0.3
first iteration
0.25
0.2
0.15
0.1
0.05
0
1985
1990
1995
2000
2005
2010
2015
2020
2025
To ensure that the period average mx from the interpolated annual mx matches the inputs, the inputs
were adjusted by the ratio of the original input to the new period average and the imputation repeated.
Tests showed that three iterations were adequate to achieve reasonable convergence with average
relative deviations in period mx below 0.001.
We defined four groups of Member States for which data inputs and interpolation methods differed.
The four groups are:
High HIV countries
21 countries for which WPP2015 used Spectrum to explicitly model HIV
mortality. The UN Population Division has provided unpublished estimates of
non-HIV mortality for these countries.
“Other” HIV countries
An additional 22 countries with significant HIV epidemics for which WHO has
in the past explicitly modelled HIV and non-HIV mortality trends in order to
prepare life tables. These countries were not modelled using Spectrum for
WPP2015.
VR countries
85 countries for which the WHO Mortality Database held mortality data from
vital registration (VR) systems for 75% or more of years since 1990.
WPP countries
58 countries where interpolated mortality rates from WPP quinquennial life
tables were used directly to construct annual life tables
A full list of countries in each category is provided in Annex Table A.
World Health Organization
Page 3
3.1 Interpolation of mx for countries in the WPP and VR categories
Conflict and natural disasters (mortality shocks) may cause substantial increases in death rates for
specific country-years. These may or may not be reflected in available death registration or
survey/census data. WHO makes estimates of these deaths by country-year as part of its overall cause of
death analyses and hence the all-cause mortality and life tables need to be consistent. Methods used for
updating mortality shock estimates are summarized in Annex B.
The WPP 2015 includes the impact of large mortality shocks in some cases (eg. Rwanda 1994 genocide)
but not for others (eg. Haiti 2010 earthquake). This may be obvious from 45q15 plots for large isolated
mortality shocks, but much less clear for extended conflicts such as those in Afghanistan or Iraq. The
assumed impact of mortality shocks included in WPP estimates may or may not be consistent with the
WHO estimates of size of mortality shocks.
Annual estimates of conflict and natural disaster deaths by country, age, sex and year for the period
1985-2015 were prepared as described in Annex B. For countries in the VR and WPP groups, 45q15 plots
were reviewed to identify country-periods for which WPP2015 had included an impact of mortality
shocks. Separate plots for males and females were made for 45q15 calculated directly from the WPP
2015 estimates of mx and for “shock-free” mx from which the WHO estimates of shock mortality had
been subtracted. The following plots show examples where the WPP 2015 estimates of mx included
shock mortality (Bosnia and Herzegovina, left) and where they did not (Myanmar, right). For cases in
the first category, non-shock mx were interpolated from the mx for neighbouring periods.
For six countries with extended conflicts, covering many of the 5-year periods in the range 1985-2015, it
was assumed that the adult mortality data used to prepare WPP2015 life tables had included conflict
mortality. These six countries were Afghanistan, Iraq, Sri Lanka, Somalia, Sudan and Yemen. For these
countries, the WHO estimates of shock mortality for the 5-year periods were subtracted from the
WPP2015 mortality rates and the 45q15 time series computed. The non-shock 45q15 were smoothed
World Health Organization
Page 4
using Loess regression with bandwidth 0.8 and the smoothed 45q15 were used together with the UN
model life table system (10,11) to compute non-shock mortality rates by age and sex.
Adjustments were also made for specific country-periods listed in Table 1, where WPP2015 had included
some impact for a mortality shock. Table 2 lists country-periods with significant shock mortality
according to WHO estimates, for which WPP2015 did not include a shock adjustment.
Table 1. Shocks identified as included in WPP 2015 mortality rates
WHO estimated
Country
Period
Type of shock
Deaths/10,000
Deaths (‘000)
Algeria
1990-2005
Conflict
8
32
Azerbaijan
1990-1995
Conflict
24
9
Bosnia and Herzegovina
1990-2000
Conflict
182
38
Croatia
1990-1995
Conflict
8
2
Georgia
1990-1995
Conflict
20
5
Indonesia
2000-2005
2004 Asian tsunami
14
166
Iran
1985-1990
Conflict and 1990 earthquake
13
40
2000-2005
Bam earthquake 2003
9
29
Lebanon
1985-1990
Conflict
52
7
Libya
2010-2015
Conflict
19
6
Mexico
1985-1990
1985 Mexico City earthquake
23
95
Micronesia
2000-2005
Tropical storm Chataan 2002
9
0.1
Nicaragua
1985-1990
Conflict
16
3
1995-2000
1998 Hurricane Mitch
14
3
Palestine
1985-2015
Conflict
12
12
Pakistan
2005-2010
Conflict + 2005 Kashmir earthquake
13
106
5
47
2010-2015
Panama
1985-1990
Disaster
15
2
Peru
1990-1995
Conflict
7
8
Samoa
2005-2010
Disaster
17
0.2
Syria
2010-2015
Conflict
274
269
Tajikistan
1990-1995
Conflict
44
12
Turkey
1995-2000
1999 Izmit earthquake
14
43
Venezuela
1995-2000
1999 floods and mudslide
26
30
Rwanda
World Health Organization
Page 5
Table 2. Shocks identified as NOT included in WPP 2015 mortality rates
WHO estimated
Country
Period
Type of shock
Deaths/10,000
Deaths ('000)
Armenia
1985-1990
1988 earthquake
105
37
Bangladesh
1990-1995
1991 Bangladesh cyclone
25
142
Columbia
1985-1990
1985 volcanic eruption
16
23
Comoros
1995-2000
Conflict
6
147
Croatia
1995-2000
Conflict
4
1
El Salvador
1985-1990
Conflict
135
34
Georgia
2005-2010
Conflict
6
1
Honduras
1995-2000
1998 Hurricane Mitch
49
15
Kuwait
1990-1995
Conflict
236
22
Myanmar
2005-2010
Cyclone Nargis 2008
56
141
Nepal
2000-2005
Conflict
16
20
WPP2015 estimates of adult mortality for Lebanon were largely based on trends in U5MR and CD West
model life tables. Data on adult mortality from reported household deaths in some recent censuses and
surveys in the regions suggested that adult mortality is often higher than implied by CD west for a given
U5MR, and adult mortality rates for Lebanon were adjusted upwards based on the region-specific
relationship between 45q15, u5mr and income per capita. Additionally, Lebanon, Turkey and Jordan are
three countries with very substantial de facto resident populations of Syrian refugees from 2013
onwards. For years 2013-2015, mortality rates for these three countries were adjusted using a
population weighted average of WPP2015 estimates of the country mx and Syrian mx, using UNHCR
estimates of refugee populations (12).
After preparation of a complete set of “shock-free” period mx for the VR and WPP countries, annual mx
were interpolated and WHO annual estimates of shock mortality added to obtain total mortality rates by
5-year age group, sex and year from 1985-2015.
3.2 Interpolation of mx for high HIV and “Other HIV” countries
For countries with substantial proportion of younger adult deaths (15-60 years) due to HIV, the all-cause
mortality envelopes, trends and age patterns must be consistent with the HIV mortality estimates,
otherwise the “non-HIV envelopes” will have strange and implausible age and/or time trends. This will
then affect most other cause of death estimates.
For development of previous WHO life tables, 43 countries were classified as “high HIV” and explicit
efforts made to ensure consistency of all-cause and HIV mortality estimates. For the WPP 2015, UN
Population Division used Spectrum (13)with input assumptions consistent with those of UNAIDS in mid2014 to model all-cause mortality for 21 countries.
World Health Organization
Page 6
Following discussions at the WHO life table working group meeting in New York, October 2015, Avenir
Consulting prepared updated Spectrum models for 1985-2015 that took into account the WPP 2015
revisions to demographic data and all-cause mortality, as well as 2015 UNAIDS files with a range of 2016
updates to the Spectrum/AIM software including new patterns of adult mortality on ART and age at ART
initiation among pediatric patients and the re-fitting of all the EPP curves.
Among the most important Spectum/AIM updates were:
1. Improvements to the EPP model that fits smooth prevalence trends to surveillance and survey
data. The handling of entrants and exists from the population 15-49 was improved to reduce the
differences between the EPP model, which is a single age/sex group, and AIM which divides the
population by sex and single age.
2. The mother-to-child transmission rates were updated by reviewing new studies from the last
three years. This resulting in changes to the probability of transmission from mother-to-child for
some PMTCT regimens.
3. New patterns of mortality for adults on ART were developed by the IeDEA Consortium using
new data from 2011 to 2014. These updated patterns show somewhat higher mortality than the
patterns used last year mainly as a result of the inclusion of more countries in the IeDEA data
set.
4. The pediatric model now has a pattern describing the age distribution of children newly starting
ART. The pattern was derived from data from IeDEA treatment sites. Previously new ART
patients were distributed by age according to need. The new patterns are regional and show
shifts in the distribution by age over time as infant diagnosis has expanded.
For preparation of the WHO life tables, it was also necessary to address issues relating to large mortality
shocks in some countries, and the need to exclude these before interpolating from 5-year period to
annual mortality rates. The following table summarizes WHO analyses of mortality shocks likely
included in the mx estimates for the 43 high HIV and other HIV countries.
Table 3. Shocks identified as included in WPP 2015 mortality rates for HIV countries
WHO estimated
Country
Period
Type of shock
Angola
1985-1990
Conflict
36
19
High HIV
1990-1995
Conflict
65
39
High HIV
1995-2000
Conflict
36
12
High HIV
2000-2005
Conflict
28
10
High HIV
1985-1990
Conflict
16
4
Other HIV
1990-1995
Conflict
14
5
Other HIV
1995-2000
Conflict
6
2
Other HIV
2000-2005
Conflict
6
3
Other HIV
2005-2010
Conflict
12
6
Other HIV
Congo
1995-2000
Conflict
191
28
High HIV
DR Congo
1995-2000
Conflict
33
75
Other HIV
Rwanda
1990-1995
Conflict
1567
512
High HIV
Burundi
Chad
World Health Organization
Deaths/10,000
Deaths ('000)
Page 7
Table 4. Shocks identified as NOT included in WPP 2015 mortality rates for HIV countries
WHO estimated
Country
Period
Type of shock
Angola
1985-1990
Conflict
36
19
High HIV
Cameroon
1985-1990
1986 Lake Nyos Gas Disaster
0.3
2
High HIV
Djibouti
1990-1995
Conflict
17
1
Other HIV
Central African Republic
2010-2015
Conflict
25
6
High HIV
Eritrea
1995-2000
Conflict
224
37
Other HIV
2000-2005
Conflict
66
13
Other HIV
1985-1990
Conflict
23
50
High HIV
1990-1995
Conflict
15
39
High HIV
2005-2010
2010 Earthquake
14
188
Other HIV
Ethiopia
Haiti
Deaths/10,000
Deaths ('000)
Table 5. UN Model life table systems (UNMLT) used for non-HIV mortality estimates in HIV countries
High HIV countries
UNMLT
Other HIV countries
UNMLT
Angola
CD North
Burkina Faso
CD North
Burundi
CD North
Côte d'Ivoire
CD North
Botswana
CD West
Ghana
CD North
Central African Republic
CD North
Guinea
CD North
Cameroon
CD North
Haiti
CD North
Congo
CD North
Liberia
CD North
Ethiopia
CD North
Mali
CD North
Gabon
CD North
Nigeria
CD North
Equatorial Guinea
CD North
Chad
CD North
Kenya
CD North
Togo
CD North
Lesotho
CD West
Thailand
Mozambique
CD North
Malawi
CD South
Namibia
CD West
Rwanda
CD North
Swaziland
CD West
United Republic of Tanzania
CD North
Uganda
CD North
South Africa
UN Far_East_Asian
Zambia
CD North
Zimbabwe
CD North
World Health Organization
UN Far_East_A sian
Page 8
For high HIV countries, provisional non-HIV mx were calculated from the model life table assumptions
and e0 series provided by UN Population Division. We added estimates of HIV death rates based on the
revised Spectrum models to the non-HIV mx to recomputed total mortality mx. This led to consequential
changes in trends and/or levels of all-cause 45q15 for a number of countries. To reduce these
differences and to smooth trends for non-HIV mortality, revised model life tables for non-HIV mortality
were prepared for 8 countries: Central African Republic, Ethiopia, Gabon, Kenya, Lesotho, Malawi,
Mozambique, and Rwanda. The model life table assumptions for the high HIV countries are shown in
Table 5. Some adjustments to individual period mx were also required for 8 countries to take into
account mortality shocks. In the case of South Africa, all-cause death registration data adjusted for
completeness was also used to assess levels of all-cause mortality, resulting in HIV mortality estimates
somewhat lower than UNAIDS and WPP2015 estimates.
For “other HIV countries”, we subtracted the revised Spectrum modelled HIV mortality rates from the
WPP2015 all-cause mortality rates and examined the consistency and plausibility of the resulting nonHIV mortality time trends, age trends and sex ratios. Provisional WHO mx were calculated by smoothing
the implied non-HIV mortality trends and adding back the UNAIDS HIV mortality estimates. For problems
with isolated periods in 8 countries, the mx for the period were revised by interpolation. For the
following 11 countries, adjusted 45q15 time series were used to revise the model life tables for non-HIV
mortality: Burkina Faso (females only), Chad, Côte d'Ivoire, Ghana, Guinea, Haiti, Liberia, Mali, Nigeria,
Togo and Thailand (males only). The model life table assumptions for the “Other HIV” countries are
shown in Table 5.
Scatterplots of resulting 45q15 (non-HIV and
total) identified implausibly low mortality levels
600
for Tanzania in most recent years. The most
recent publicly available empirical mortality
HIV_high
500
"HIV_med"
Lesotho
pattern available for Tanzania (based on
reported household deaths in the 2012 census)
400
Swaziland
was also consistent with a higher 45q15 than the
300
WPP2015 estimate. This derived from an
estimated rapid decline of child mortality in
Mozambique
200
recent
years, used to predict adult non-HIV
Zambia
mortality from the model life table for WPP2015.
100
The most recent publicly available empirical
mortality pattern available for Tanzania (based
0
0
100
200
300
400
500
600
700
on reported household deaths in the 2012
HIV deaths per 100,000 (GHE2013)
census) was also consistent with a higher 45q15
than the WPP2015 estimate. Trends in adult 45q15 for Tanzania were revised to reduce the acceleration
in rate of decline, drawing also on IHME analyses of non-HIV mortality for Tanzania (14).
HIV deaths per 100,000 (GHE2015)
Figure 3. Comparison of HIV mortality rates for 2010 for
GHE2015 versus GHE2013
World Health Organization
Page 9
4 Adjustments using death registration data
The WPP2015 life tables draw extensively on available death registration data to assess age-specific
mortality rates mx. For 21 countries with high quality and complete death registration data, they make
use of life tables prepared for the Human Mortality Database (15, 16), which corrects for age
misstatement at older age groups.
WHO holds time series of death registration data for around 100 countries (5). These potentially
provide alternate data for preparing annual life tables, or additional data that would assist in imputation
from period life tables. For 85 countries with at least 75% of the years in range 1990-2015 available, we
evaluated the completeness of the all-cause deaths data and used completeness-adjusted death rates to
inform the imputation of annual death rates for life tables.
4.1 Updated assessments of VR completeness
Until now, WHO has used population data reported by Member States for the population covered by
death registration as the denominator for calculation of all-cause mortality rates, and has then
computed total numbers of deaths by applying these rates to WPP population estimates for the de-facto
resident population. This can result in completeness estimates varying from 100% for high income
countries, and to estimated total deaths lower than registered deaths. Additionally, country-reported
population data are not available for a substantial proportion of country-years in some regions.
For this revision of completeness estimates, we have switched to use of WPP2015 population estimates
as denominators. Implied completeness of death registration data has then been assessed against WPP
2015 by comparing reported registered deaths against the total deaths implied by the WPP life tables
for 5-year periods.
There are five countries where registered deaths reported to WHO do not include a province or territory
not under government control. These are:
Cyprus:
all data refer to government controlled areas
Georgia:
excluding Abkhazia and South Osetia
Moldova:
excluding Transnistria and Bender
Russia:
1993-2003 data exclude Chechnya
Serbia:
excluding Kosovo-Metohija province
Singapore does not report deaths for non-citizen residents, who represent approximately 30% of the defacto population. A number of European countries similarly exclude non-citizen residents from data
reported to WHO those these typically amount to at most a few percent of the de-facto resident
population. For these two groups of countries, the new method potentially
results in lower
completeness against UN estimates of resident population.
Overall completeness levels for ages 15+ were estimated by sex and compared with completeness
estimates from IHME derived using an ensemble of death distribution methods (17,18) and with
previous WHO completeness estimates based on WHO application of the Generalized Brass GrowthBalance and Bennett-Horiuchi methods for the 1990s and early 2000s. For the 85 countries with VR time
World Health Organization
Page 10
series, adult completeness was assessed by comparing total registered deaths for persons aged 15 years
and over in each five year period with 5-year period deaths for ages 15+ calculated from the WPP2015
life table mx together with WPP2015 population estimates for the 5-year periods. For 5-year periods
covering only 4 years of death registration data, total registered deaths for the missing year was
interpolated or extrapolated. For 5-year periods covering 2 or 3 years of death registration data,
completeness was assessed against the total deaths calculated from the corresponding annual lifetables
interpolated from WPP2015. In most cases, where a beginning or end period contained only 1 year of
death registration data, this was excluded from the analysis.
Adjustments to estimated completeness levels were made for some countries as follows:
Guyana:
Prior to 1995, IHME estimates of completeness were used
Israel:
inclusion of East Jerusalem from 1980 onwards would result in apparent completeness of 1.061.07 against WPP2015. Completeness was truncated at 1.0
Malta:
completeness assessed against WPP2015 exceeded 1.0 for all except most recent years.
Completeness was truncated at 1.0.
Mauritius:
completeness assessed against WPP2015 exceeded 1.0 for all except most recent years. The
deaths reported to WHO include all of Mauritius except for the island of Rodrigues.
Completeness for Mauritius was thus revised to 0.974.
Russia:
Deaths in Chechnya are missing from the data reported to WHO for years 1993-2003
(corresponding to approximately 1% incompleteness). Assessed completeness was slightly less
than 0.99 for most of this period, and was revised to 1.0 for males in 1990-1995 period, when it
slightly exceeded 1.0.
Suriname:
Prior to 1995-2000, previous WHO estimates of completeness were used
For a number of other countries mainly in Eastern Europe and Latin America and the Caribbean,
completeness estimates fluctuated both above and below 1.0. Apparent completeness above 100%
may reflect issues with the numerator for registered deaths, which can vary in some country-years in
term of (a) "year of occurrence vs year of registration", (b) "provisional vs. final", (c), de-jure vs. de-facto
or (d) inclusion of nationals only (as in Japan and several EU countries). It may also reflect mismatches
with denominators resulting from issues around the estimation of the "de-facto" population and the
inclusion of migrants and refugees in the countries (irrespective of their legal status). For countryperiods where estimated completeness exceeded 1.05, it was capped at 1.05.
After these adjustments, annual completeness estimates were smoothly interpolated from the 5-year
period completeness estimates. Rising and falling projections for latest partial 5-year period were
adjusted to avoid out-of-sample trends. In a number of cases where completeness rates for the last 5year period rose above 100%, completeness was capped at 100%. Identification of out of sample trends
was also informed by examination of IHME estimates of completeness trends (18). Annex C compares
the resulting final completeness time series for countries with previous estimates by WHO and IHME.
Annex Table D lists estimated completeness for the most recent year of death registration data for each
Member State meeting inclusion criteria.
World Health Organization
Page 11
4.2 Revision of imputed annual death rates
The WPP2015 life tables based on death registration data included adjustments for age mis-statement
and under-reporting at older ages based on analyses carried out by the Human Mortality Database
(15,16), analyses for consistency with population age structures and use of the Kannisto-Thatcher
method for assessing oldest age mortality rates (19). For this reason, we did not simply apply the
completeness estimated for ages 15+ to all VR deaths for those ages, but carried out a second
assessment of completeness against the WPP2015 life tables, for each sex separately for age groups 59,10-14, 15-69, 70-79, 80-84, 85+.
Smoothly varying annual completeness estimates for years 1985-2015 were imputed using PCHIP
interpolation (8,9). The age-sex specific completeness estimates were assumed constant outside the
lower and upper period mid-points for the available VR years for each country.
Age-specific
completeness estimates for age groups 70-74 and 75-79 were then imputed to ensure a smooth
transition between earlier and older age groups. Finally, the imputed age-sex specific completeness
estimates were adjusted to match the sex-specific completeness for ages 15+. Completeness for age
groups 5-9 and 10-14 was capped at 100%.
For 7 countries with less than 1 million population in 2000-2015, VR death rates were smoothed using a
3-year moving average. Annual mx for ages 5+ were calculated for each VR country-year as:
mx = VR deaths /(WPP population estimate)/(annual age-sex-specific completeness estimate)
While most VR countries had data up to 2012 or 2013, only a few had reported data to WHO for 2014,
and none for 2015 at the time of analysis (early 2016). VR-based mx estimates for age groups from 5-9
onwards were projected forward using Poisson regression to estimate the trend in latest 10 years of VR
data. Importance weights declining by a factor of 0.85 for each earlier year from last were used to give
greater weight to more recent trend. Estimated annual rates of change (ARR) for 5-year age groups were
smoothed across age groups using a 3-age group moving average. ARRs not statistically significant were
capped within the range of statistically significant ARRs across age groups. The mx values were projected
forward to 2015 using an ARR that changed smoothly from the VR-based ARR estimate to the WPP2015based ARR over a six year period. For most countries, this means that the 2015 estimated mx are quite
strongly influenced by the trends in the recent VR data, where for the handful of countries where the
latest year was earlier than 2009, the VR trend converged to the WPP trend over the project period.
Annual mx for all age groups from 5-9 onwards were replaced by the VR-based estimates and
projections for VR countries for which completeness analyses were carried out with the exception of the
following five countries: Singapore, Cyprus, Kuwait, Malaysia and Sri Lanka These five countries had
partial VR coverage and/or major fluctuations in implied completeness estimates (see Annex C).
World Health Organization
Page 12
5 Neonatal, Infant and Under-five mortality
Mortality rates for infants and age group 1-4 years for the WHO life tables were derived from the UNIGME estimates of infant mortality rates (IMR) and under 5 mortality rates (U5MR) by sex, for Member
States for years 1990-2015 (20). The United Nations Inter-agency Group for Child Mortality
Estimation (UN IGME), which includes UNICEF, WHO, the World Bank and United Nations Population
Division, was established in 2004 to advance the work on monitoring progress towards the achievement
of Millennium Development Goals regarding child mortality.
UN-IGME annually assesses and adjusts all available surveys, censuses and vital registration data, to
estimate country-specific trends in neonatal (NMR), infant (IMR) and under 5 (U5MR) mortality rates per
1,000 live births. All data sources and estimates are documented on the website
www.childmortality.org. For countries with complete recording of child deaths in death registration
systems, these are used as the source of data for the estimation of trends in neonatal, infant and child
mortality. For countries with incomplete death registration, all other available census and survey data
sources, which meet quality criteria, are used.
Due to fewer data available by sex than data for both sexes, UN IGME uses available data by sex to
estimates time trend in the sex ratio (male/female) of U5MR. Leontine Alkema and Fengqing Chao of the
National University of Singapore have developed new Bayesian methods for the UN IGME estimation of
sex ratios, with a focus on the estimation and identification of countries with outlying levels or trends
(21).
6
WHO estimates of life expectancy and healthy life expectancy
6.1 Life expectancy
Final estimates of age-sex-specific mortality rates for years 1990-2015 were used to compute abridged
life tables for 183 WHO Member States with population of 90,000 or greater in 2015. Life expectancies
at birth were reported in World Health Statistics 2016: and full life tables are available in the WHO
Global Health Observatory (www.who.int/gho). Annex E presents country plots showing the resulting
WHO annual estimates of 45q15 by sex for all-cause mortality and for non-HIV mortality excluding
disasters and conflict deaths. Five year period estimates of 45q15 from the WPP2015 life tables are
shown for comparison.
WHO applies standard methods to the analysis of Member State data to ensure comparability of
estimates across countries. This will inevitably result in differences for some Member States with official
estimates for quantities such as life expectancy, where a variety of different projection methods and
other methods are used. These WHO estimates of mortality and life expectancies should not be
regarded as the nationally endorsed statistics of Member States, which may have been derived using
alternative methodologies and assumptions.
INSERT SUMMARY OF DATA AVAILABILITY HERE
There remain substantial data gaps and deficiencies in data on levels of child and adult mortality,
particularly in those regions with the highest mortality levels. Quantifiable uncertainty ranges for adult
mortality are more complex to derive, and there is considerable research underway to develop
World Health Organization
Page 13
improved methods for measuring adult mortality in surveys, and in assessing the systematic biases in
such data. Table 6 summarizes the availability of data on levels of all-cause mortality for WHO Member
States and the methods used to assess mortality and life expectancy.
Table 6 Data availability for all-cause mortality
Number of WHO
a
Member States
Percentage of global
b
deaths 2015
59
28
Observed death rates
38
25
Adjusted death rates
Other population-representative data on aged
specific mortality
18 (3)
25
Estimated death rates and
model life table systems
Data on child (under 5 years) and
d
adult (15–59 years) mortality only
30 (18)
12
Estimated death rates and
model life table systems
37 (22)
10
Model life table systems
1
<1
Projected from data for
years before 2005
Available recent data (since 2005)
Complete death-registration data
c
Incomplete death-registration data
Data on child mortality only
d
No recent data
Methods
a
Only includes 183 Member States with population above 90 000 in 2015.
Percentage of global deaths that occur in the countries included in each category – not the percentage registered or included in datasets.
c
Completeness of 90% or greater for de facto resident population; as assessed by WHO and the United Nations, Population Division, 2016.
d
Numbers in parenthesis show the number of high HIV prevalence countries for which multistate epidemiological modelling for HIV mortality
was also carried out.
b
A qualititative guide to the uncertainty in adult mortality and life expectancy estimates is provided by
the listing of methods and data input types in Annex Table A. The most reliable estimates are those
based on death registration data assessed as complete, followed by those based on incomplete or
sample death registration data with adjustments for levels of completeness. For countries without
useable death registration data, uncertainties are substantially higher, and two categories can be
distinguished (a) those countries where there is independent evidence on adult mortality rates from
surveys or censuses and (b) those where estimates of adult mortality levels are derived from model life
tables with estimated infant and child mortality rates as inputs. Those countries with significant levels
of mortality due to conflict and natural disasters (say, greater than 1 death per 10,000 population per
annum) will usually have additional uncertainty associated with the difficulties in estimating conflict and
disaster death rates.
6.2 Healthy life expectancy
WHO has previously published estimates of healthy life expectancy (HLE or HALE) for years 2000 and
2012, drawing on the previous WHO life table series and estimates of years lost to disability (YLD) for
disease and injury causes from the Global Burden of Disease 2010 (GBD2010) study (22-24).
The same methods have been used to prepare estimates of healthy life expectancy for WHO Member
States for the year 2015 (2), using the updated WHO life tables and projections of YLD for year 2015
based on YLD estimates for 2010 and 2013 from the Global Burden of Disease 2013 (GBD2013)
study(25), with similar adjustments to disability weights and prevalences for certain causes as previously
(22).
World Health Organization
Page 14
References
(1)
The Global Health Observatory (GHO) is WHO’s portal providing access to data and analyses for
monitoring the global health situation. See: http://www.who.int/gho/en/
(http://www.who.int/gho/mortality_burden_disease/life_tables/en/index.html)
(2)
World Health Statistics 2016. Geneva: World Health Organization; 2016.
(http://apps.who.int/iris/bitstream/10665/206498/1/9789241565264_eng.pdf)
(3)
World Population Prospects: The 2015 Revision. DVD Edition. New York (NY): UnitedNations, Department
of Economic and Social Affairs, Population Division; 2015 (http://esa.un.org/unpd/wpp/)
(4)
Levels & Trends in Child Mortality. Report 2015. Estimates Developed by the UN Interagency Group for
Child Mortality Estimation. New York (NY), Geneva and Washington (DC): United Nations Children’s Fund,
World Health Organization, World Bank and United Nations; 2015
(http://www.unicef.org/publications/files/Child_Mortality_Report_2015_Web_9_Sept_15.pdf).
(5)
World Health Organization. Mortality Database. Available at:
http://www.who.int/healthinfo/mortality_data/en/index.html
(6)
UNAIDS (2015). HIV estimates with uncertainty bounds 1990-2014. Available at
2015http://www.unaids.org/en/resources/documents/2015/HIV_estimates_with_uncertainty_bounds_1
990-2014 (accessed 19 May 2016)
(7)
Global Health Estimates 2013: deaths by cause, age and sex; estimates for 2000–2012. Geneva: World
Health Organization; 2014 (http://www.who.int/healthinfo/global_burden_disease/en/).
(8)
For more information, see http://blogs.mathworks.com/cleve/2012/07/16/splines-andpchips/#ee54c20b-0ecc-4ac9-8986-8a0774a1763f
(9)
Boston College Department of Economics. PCHIPOLATE: Stata module for piecewise cubic Hermite
interpolation. Statistical Software Components no. S457561. Available at
http://fmwww.bc.edu/repec/bocode/p/pchipolate.ado
(10)
Coale AJ, P Demeny and B Vaughan. 1983. Regional Model Life Tables and Stable Populations. New York:
Academic Press.
(11)
UN Population Division. 2010. World Population Prospects 2012: Extended Model Life Tables. New York:
United Nations, Department of Economic and Social Affairs. http://esa.un.org/wpp/Model-LifeTables/data/MLT_UN2010-130_1y.xls
(12)
UNHCR (2015). Syria Regional Refugee Response: Inter-agency Information Sharing Portal. Available at
http://data.unhcr.org/syrianrefugees/regional.php
(13)
Stover J, Andreev K, Slaymaker E, et al. Updates to the Spectrum model to estimate key HIV indicators for
adults and children. AIDS (London, England). 2014;28(4):S427-S434. doi:10.1097/QAD.0000000000000483.
(14)
GBD 2013 Mortality and Causes of Death Collaborators. Global, regional, and national age–sex specific allcause and cause-specific mortality for 240 causes of death, 1990–2013: a systematic analysis for the
Global Burden of Disease Study 2013. Lancet 2014; 385: 117–71.
(15)
Human Mortality Database. University of California, Berkeley (USA), and Max Planck Institute for
Demographic Research (Germany). Available at www.mortality.org or www.humanmortality.de
(16)
Wilmoth JR, Andreev KF, Jdaov DA, Glei DA. Methods Protocol for the Human Mortality Database. Version
5. University of California, Berkeley and Max Planck Institute for Demographic Research; 2007.
World Health Organization
Page 15
(17)
Murray CJL, Rajaratnam JK, Marcus J, Laakso T, Lopez AD (2010) What Can We Conclude from
Death Registration? Improved Methods for Evaluating Completeness. PLoS Med 7(4): e1000262.
doi:10.1371/journal.pmed.1000262
(18)
Phillips DE, Rafael Lozano R, Mohsen Naghavi M, Charles Atkinson C, Diego Gonzalez-Medina D, Lene
Mikkelsen L, Christopher JL Murray CJL, Alan D Lopez AD. A composite metric for assessing data on
mortality and causes of death: the vital statistics performance index. Population Health Metrics 2014,
12:14. DOI: 10.1186/1478-7954-12-14
(19)
UNICEF, WHO, The World Bank and UN Population Division. Levels and Trends of Child Mortality - Report
2015, Estimates developed by the UN Inter-agency Group for Child Mortality Estimation. UNICEF, New
York, 2015
(20)
UNICEF, WHO, The World Bank and UN Population Division. Levels and Trends of Child Mortality - Report
2015, Estimates developed by the UN Inter-agency Group for Child Mortality Estimation. UNICEF, New
York, 2015
(21)
Alkema L, Chao F, Sawyer CC (2013). Sex Differences in U5MR: Estimation and identification of countries
with outlying levels or trends. Paper presented at the XXVII IUSSP International Population Conference,
Busan, Republic of Korea.
(22)
World Health Organization (2014). WHO methods for life expectancy and healthy life expectancy (Global
Health Estimates Technical Paper WHO/HIS/HSI/GHE/2014.5). Available at www.who.int/evidence/bod
(23)
Vos T, Flaxman AD, Naghavi M, Lozano R, Michaud C, Ezzati M et al (2012a). Years lived with disability
(YLDs) for 1160 sequelae of 289 diseases and injuries, 1990–2010: a systematic analysis for the Global
Burden of Disease Study 2010. Lancet.380:2163–2196.
(24)
Vos T, Flaxman AD, Naghavi M, Lozano R, Michaud C, Ezzati M et al (2012b). Years lived with disability
(YLDs) for 1160 sequelae of 289 diseases and injuries, 1990–2010: a systematic analysis for the Global
Burden of Disease Study 2010 [Supplementary appendix].
(http://download.thelancet.com/mmcs/journals/lancet/PIIS0140673612617292/mmc1.pdf?id=a02f57d18
11fcb77:-1b44796c:142333b8265:-259e1383841102443, accessed 7 November 2013)
(25)
Global Burden of Disease Study 2013 Collaborators: Vos T et al. Global, regional, and national incidence,
prevalence, and years lived with disability for 301 acute and chronic diseases and injuries in 188 countries,
1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. The Lancet. 2015 June 8.
doi: 10.1016/S0140-6736(15)60692-4.
World Health Organization
Page 16
Annex Table A: Data sources and methods for WHO Life Tables
Member State
WPP2015 methods for estimation of age-sex-specific mortality
ratesa
Life table method
GHE2015
method
WHO MBD VR
years available
Afghanistan
Estimated using the West model of the Coale-Demeny Model Life Tables
and three parameters: (1-2) direct and indirect estimates of infant and child
mortality, and (3) adjusted estimates of adult mortality (45q15). Adjusted
estimates of adult mortality were derived from: (a) recent household
deaths data from the 1979 census; (b) implied relationship between child
mortality and adult mortality based on the UN South Asian and West model
of the Coale-Demeny Model Life Tables; and (c) levels of adult mortality
based on sample registration data from neighbouring countries for recent
years. Estimates of adult mortality derived from (i) recent household deaths
data from the 2010 Afghanistan Mortality Survey (AMS), (ii) parental
orphanhood from the 2010 AMS (excluding the Southern region), and (iii)
siblings deaths from the 2010 AMS (excluding the Southern region) adjusted
for age misreporting and recall biases were also considered.
CD West relational
model for non-HIV
mortality
wpp
-
Angola
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the North model of the Coale-Demeny
Model Life Tables. The demographic impact of AIDS has been factored into
the mortality estimates.
CD North model life
tables for non-HIV
mortality
High HIV
-
Albania
Based on life tables for 1987-2013 derived from registered deaths by age
and sex and observed trends in infant and child mortality.
Death registration data
vr
1980, 1984-2009
United Arab
Emirates
Based on life tables derived from official estimates of registered deaths and
enumerated census population by age and sex from 1988 through 2010,
adjusted for infant and child mortality. Mortality rates for older ages were
adjusted.
Death registration data
wpp
2003, 2005-2010
Argentina
Based on registered deaths from 1950 through 2013, and the underlying
population from censuses, and revised projections by the National Statistics
Office (INDEC). The number of deaths was adjusted using the growthbalance method.
Death registration data
vr
1980-2013
Armenia
Based on: (a) a life table derived from reported deaths by age and sex in
2011 and the 2011 census population, adjusted for underreporting of infant
and child deaths, and (b) official estimates of life expectancy available from
2006 through 2013.
Death registration data
vr
1981-2012
Antigua and
Barbuda
Based on official estimates of life expectancy from 2000 to 2010.
Death registration data
wpp
1983-2009, 20122013
Australia
Based on official estimates of life expectancy available through 2009. The
age pattern of mortality is based on life tables through 2009 from the
Human Mortality Database.
Death registration data
vr
1980-2012
Austria
Based on official estimates of life expectancy at birth through 2013.
Death registration data
vr
1980-2014
Azerbaijan
Based on deaths registered through 2012 classified by age and sex and the
underlying population by age and sex. Death rates were adjusted for
underregistration.
Death registration data
vr
1981-2011
Burundi
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the North model of the Coale-Demeny
model life tables, and taking into account the number of deaths due to civil
strife. The demographic impact of AIDS has been factored into the mortality
estimates.
CD North model life
tables for non-HIV
mortality
High HIV
-
Belgium
Based on official estimates of life expectancy available through 2012.
Death registration data
vr
1980-2012
Benin
Estimated using the North model of the Coale-Demeny Model Life Tables
and implied relationships between life expectancy at birth and estimates of
infant and child mortality, and between life expectancy at birth and
estimates of adult mortality (45q15). The adjusted estimates of 45q15 were
derived from: (a) parental orphanhood from the 1981/83 multiround survey
and 2002 census, and (b) siblings deaths from the 1996, 2002 and 2006
CD North relational
model for non-HIV
mortality
Other HIV
-
World Health Organization
Page 17
DHS.
Burkina Faso
Estimated using the South model of the Coale-Demeny Model Life Tables
and three parameters: (1-2) direct and indirect estimates of infant and child
mortality, and (3) adjusted estimates of adult mortality (45q15). Data from
West African rural demographic surveillance sites and urban vital
registration were also considered. Adjusted estimates of adult mortality
were derived from: (a) recent household deaths data (unadjusted and
adjusted for underregistration using the growth-balance and syntheticextinct generation methods) from the 1960/61 survey, 1976, 1985, 1996
and 2006 censuses, the 1991 National Demographic Survey, and 2008
Global Fund survey; (b) parental orphanhood from the 1993, 2003 and
2010/11 DHS, 2006 MICS3 and 2006 census; (c) siblings deaths from the
1998/99, 2003 and 2010/11 DHS; (d) intercensal survivorship from
successive census age distributions (smoothed and unsmoothed) for
periods 1976-1985, 1985-1996 and 1996-2006.
CD South relational
model for non-HIV
mortality
Other HIV
-
Bangladesh
Based on life tables derived from age and sex-specific mortality rates from:
(a) the Sample Vital Registration System from 1981 up to 2010 adjusted for
infant and child mortality, (b) the 1974 Retrospective Survey of Fertility and
Mortality, and (c) the 1962/65 Population Growth Estimation Experiment.
Estimates are consistent with those from the 2001 and 2010 Bangladesh
Maternal Mortality Surveys (based on sibling histories and household
deaths in the preceding 36 months), and data gathered from Matlab Health
and Demographic Surveillance System up to 2012. For the period 19701975, mortality was adjusted to take into account the excess mortality
associated with the 1971 civil war and independence from Pakistan, and the
1974 flood and famine.
Death registration data
wpp
1980-1982, 19841986
Bulgaria
Based on official life tables through 2013.
Death registration data
vr
1980-2012
Bahrain
Based on life tables derived from official estimates of registered deaths and
enumerated census population by age and sex from 1980 to 2012, adjusted
for infant and child mortality. Mortality rates for older ages were adjusted.
For the period 1950-1980, life tables were derived from estimates of infant
and child mortality by assuming that the age pattern of mortality conforms
to the South model of the Coale-Demeny Model Life Tables in 1950-1955,
and converges over time toward the estimated 1980-1985 life table.
Death registration data
and CD South model life
tables
wpp
1980-2013
Bahamas
Derived from child and adult mortality estimates through 2013 by assuming
that the age pattern of mortality conforms to the West model of the CoaleDemeny Model Life Tables.
Death registration data
and CD West relational
model, adjusted for
UNAIDS estimates of HIV
mortality
Other HIV
1980-2012
Bosnia and
Herzegovina
Based on official estimates of life expectancy for 1988/89 and WHO
estimates for years 2000 to 2012. The estimates of war-related deaths in
the period 1992-1995 were also considered.
Death registration data
vr
1985-1991, 19982011
Belarus
Based on official life tables available through 2013.
Death registration data
vr
1981-2009, 20112012
Belize
Estimated using the West model of the Coale-Demeny Model Life Tables and
two parameters: (a) estimates of child mortality; and (b) adjusted estimates
of adult mortality from registered deaths and underlying population
through 2009. From 1950 to 1995, estimated using adjusted registered
deaths by age and sex and underlying population by age and sex.
Death registration data
and CD West relational
model, adjusted for
UNAIDS estimates of HIV
mortality
Other HIV
1980-2013
Bolivia
(Plurinational
State of)
Based on life tables derived from: (a) deaths by age and sex, adjusted using
the growth-balance method, and underlying population from the 1992,
2001 and 2012 censuses; (b) data on maternal orphanhood from the 1988
National Population and Housing Survey (ENPV); (c) official estimates of life
expectancy for 2010 and 2011; and (d) estimates of infant and child
mortality from the 2000 MICS and the 1989, 1994, 1998, 2003 and 2008
DHS.
Survey, census and
death registration data
vr
2000-2003
Brazil
Based on life tables derived from: (a) registered deaths by age and sex from
1979 through 2012, and the underlying census population by age and sex,
and (b) estimates of infant and child mortality. The number of deaths was
Death registration data
vr
1980-2013
World Health Organization
Page 18
adjusted using the growth-balance method.
Barbados
Derived from estimates of child mortality and adult mortality from vital
registration data through 2007 by assuming that the age pattern of
mortality conforms to the West model of the Coale-Demeny Model Life
Tables.
Death registration data
and CD West model life
tables
vr
1980-2012
Brunei
Darussalam
Derived from child and adult mortality estimates through 2011 by assuming
that the age pattern of mortality conforms to the West model of the CoaleDemeny Model Life Tables. Life tables are estimated using the Flexible twodimensional model life table and Lee-Carter method.
CD West relational
model for non-HIV
mortality
vr
1982-2013
Bhutan
Based on a life table derived from adjusted deaths in the past 12 months by
age and sex, and the population by age and sex from the 2005 census,
adjusted for infant and child mortality. For 1950-2000, life tables were
derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the West model of the Coale-Demeny
Model Life Tables in 1950-1955 and converges over time toward the
estimated 2000-2005 life table. Life tables based on adjusted annual deaths
from the 1994 National Health Survey were also considered.
Survey, census data and
CD West Model life
tables
wpp
-
Botswana
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the West model of the Coale-Demeny
Model Life Tables. The demographic impact of AIDS has been factored into
the mortality estimates.
CD West model life
tables for non-HIV
mortality
High HIV
1995
Central African
Republic
Estimated using the North model of the Coale-Demeny Model Life Tables
and implied relationships between life expectancy at birth and estimates of
infant, child, and adult (45q15) mortality. The adjusted estimates of adult
mortality were derived from (a) intercensal survivorship from successive
census age distributions (smoothed and unsmoothed) for the period of
1988-2000; (b) household deaths data (unadjusted and adjusted for
underregistration using the growth-balance and synthetic-extinct
generation methods) from the 1988 census; (c) parental orphanhood data
from the 1994/95 DHS and the 1988 census; and (d) siblings deaths from
the 1994/95 DHS. The demographic impact of AIDS has been factored into
the mortality estimates.
CD North relational
model for non-HIV
mortality
High HIV
1988
Canada
Based on official estimates of life expectancy available through 2011. The
age pattern of mortality is based on life tables through 2011 from the
Human Mortality Database.
Death registration data
vr
1980-2011
Switzerland
Based on official life tables from through 2012.
Death registration data
vr
1980-2012
Chile
Based on life tables derived from registered deaths, and population by age
and sex from 1950 to 2013 adjusted for infant and child mortality. The
number of deaths was adjusted using the growth-balance method,
Death registration data
vr
1980-2013
China
Based on life tables from: (a) the 1981, 1990, 2000 and 2010 censuses
(adjusted for underestimation of child mortality and overestimation of oldage mortality); (b) surveys on causes of death in 1973/75 and 2004/05; (c)
Disease Surveillance Points (DSP) system from 1991 to 2013; and (d) 1987,
1995 and 2005 population survey (1 per cent), and the annul survey on
population change (1 per thousand).
Survey, census and
sample death
registration data
wpp
1995-2000
Côte d'Ivoire
Estimated using the North model of the Coale-Demeny Model Life Tables
and three parameters: (1-2) direct and indirect estimates of infant and child
mortality, and (3) adjusted estimates of adult mortality (45q15). Adult
mortality estimates are derived from (a) recent household deaths data from
the 1978/79 follow-up survey, the 1998 census and the 2005 EIS, (b) parental
orphanhood from the 1978/79 follow-up survey, the 1988 and
1998 censuses, the 1994 DHS, the 2000 MICS2 and the 2006 MICS3 surveys,
(c) siblings deaths from the 1994 DHS and the 2005 EIS; (d) implied
relationship between child mortality and adult mortality based on the South
model of the Coale-Demeny Model Life Tables in 1950-1955 and assumed
to converge over time toward the North model of the Coale-Demeny Model
Life Tables by the 1970s.
CD North relational
model for non-HIV
mortality
Other HIV
1998
Cameroon
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the North model of the Coale-Demeny
Model Life Tables. The demographic impact of AIDS has been factored into
the mortality estimates.
CD North model life
tables for non-HIV
mortality
High HIV
1987
World Health Organization
Page 19
Democratic
Republic of the
Congo
Derived from (a) estimates of infant and child mortality by assuming that
the age pattern of mortality conforms to the North model of the CoaleDemeny Model Life Tables, and (b) data on survival of siblings from the
2007 and 2013/14 DHS.
CD North model life
tables
Other HIV
-
Congo
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the North model of the Coale-Demeny
Model Life Tables. The demographic impact of AIDS has been factored into
the mortality estimates.
CD West model life
tables for non-HIV
mortality
High HIV
-
Colombia
Based on life tables derived from registered deaths, and population by age
and sex from 1950 to 2013, adjusted for infant and child mortality. The
number of deaths was adjusted using the growth-balance method.
Death registration data
vr
1982-2012
Comoros
Derived from estimates of infant and child mortality, and the West model of
the Coale-Demeny Model Life Tables.
CD West model life
tables
wpp
-
Cabo Verde
Derived from estimates of child mortality, by assuming that the age pattern
of mortality conforms to the West model of the Coale-Demeny Model Life
Tables. Official estimates of life expectancy at birth by sex for 1990 and
2000 were also considered.
CD West model life
tables
wpp
-
Costa Rica
Based on life tables derived from registered deaths, adjusted using the
growth-balance method, and population by age and sex from 1950 to 2013
adjusted for infant and child mortality.
Death registration data
vr
1980-2013
Cuba
Based on: (a) deaths registered through 2012 classified by age and sex and
the underlying population by age and sex, and (b) estimates of infant and
child mortality. The number of deaths was adjusted using the growthbalance method.
Death registration data
vr
1980-2013
Cyprus
Based on: (a) official life tables; (b) deaths registered through 2013
classified by age and sex and on the underlying population by age and sex;
and (c) estimates from other areas were also considered.
Death registration data
wpp
1980-2012
Czech Republic
Based on official estimates of life expectancy available through 2013. The
age pattern of mortality is based on official life tables for 1950-2013.
Death registration data
vr
1982-2013
Germany
Based on official estimates of life expectancy available through 2013.
Death registration data
vr
1980-2013
Djibouti
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the North model of the Coale-Demeny
Model Life Tables.
CD North model life
tables
Other HIV
1991
Denmark
Based on official life tables available through 2012.
Death registration data
vr
1980-2012
Dominican
Republic
Based on: (a) registered deaths by age and sex through 2011, and
underlying mid-year population; (b) estimates of infant and child mortality
from 2000, 2006, and 2014 (preliminary) Multiple Indicator Cluster Survey
(MICS); (c) estimates of infant and child mortality from the 1986, 1991,
1996, 2002, 2007 and 2013 DHS. The number of deaths was adjusted using
the growth-balance method.
Survey, census and
death registration data
vr
1980-1992, 19942012
Algeria
Based on official estimates of life expectancy derived from the number of
deaths registered through 2013. Estimates were adjusted for underreporting of deaths and deaths of non-nationals. From 1950 to 2000, the
age patterns of mortality are derived from estimates of infant and child
mortality by assuming that the age pattern of mortality conforms to the
mortality patterns resulting from the blending from the South model of the
Coale-Demeny Model Life Tables to the West model of the Coale-Demeny
Model Life Tables from 1950 to 2000.
Death registration data
wpp
1980-1982, 19851986, 1998, 2000
Ecuador
Based on: (a) registered deaths by age and sex from 1954 to 2011, with
underlying mid-year population; (b) estimates from the 1989, 1994, 1999
and 2004 ENDEMAIN, and the 1987 ENDESA; and (c) estimates from the
1950, 1962, 1974, 1982, 1990, 2001 and 2010 censuses. The number of
deaths was adjusted using the growth-balance method.
Death registration data
vr
1980-2013
Egypt
Based on official estimates of life expectancy available through 2013. The
age pattern of mortality is based on official life tables for various years from
1960 to 2012 adjusted for infant and child mortality, and adult mortality.
Death registration data
vr
1980-1981, 19832013
Eritrea
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the North model of the Coale-Demeny
Model Life Tables.
CD North model life
tables
Other HIV
-
World Health Organization
Page 20
Spain
Based on: (a) official estimates of life expectancy available through 2013;
(b) registered deaths by age and sex through 2012 and underlying
population by age and sex; (c) estimates from the Human Mortality
Database; and (d) estimates from Eurostat were also considered.
Death registration data
vr
1980-2013
Estonia
Based on official life tables available through 2013.
Death registration data
vr
1981-2012
Ethiopia
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the North model of the Coale-Demeny
Model Life Tables. The demographic impact of AIDS has been factored into
the mortality estimates.
CD North model life
tables for non-HIV
mortality
High HIV
1984
Finland
Based on official life tables available through 2013.
Death registration data
vr
1980-2013
Fiji
Based on: (a) official 1976, 1986, 1996, 2001 and 2007 estimates, and (b)
deaths by age and sex registered from 1950 through 2007.
Death registration data
wpp
1980-1987, 19922009, 2011-2012
France
Based on official life tables through 2012.
vr
1980-2012
Micronesia
(Federated
States of)
Based on estimates of infant and child mortality by assuming that the age
pattern of mortality conforms to the West model of the Coale-Demeny
Model Life Tables.
CD West model life
tables
wpp
-
Gabon
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the North model of the Coale-Demeny
Model Life Tables. The demographic impact of AIDS has been factored into
the mortality estimates.
CD North model life
tables for non-HIV
mortality
High HIV
-
United Kingdom
Based on: (a) official life tables for 2010-2012, and (b) life tables derived
from official estimates of registered deaths and population from 1950 to
2011.
Death registration data
vr
1980-2013
Georgia
Based on official estimates of life expectancy available through 2012,
adjusted for underregistration.
Death registration data
vr
1981-2014
Ghana
Estimated using the South model of the Coale-Demeny Model Life Tables
and three parameters: (1-2) direct and indirect estimates of infant and child
mortality, and (3) adjusted estimates of adult mortality (45q15). Adult
mortality estimates were derived from: (a) recent female household deaths
data from the 2007 Ghana Maternal Health Survey; (b) parental
orphanhood from the 1988, 1993, 1998, 2003 and 2008 DHS as well as 2006
MICS3 survey; (c) siblings deaths from the 2007 Ghana Maternal Health
Survey; and (d) implied relationship between child mortality and adult
mortality based on the North model of the Coale-Demeny Model Life
Tables.
CD North relational
model for non-HIV
mortality
Other HIV
-
Guinea
Estimated using the South model of the Coale-Demeny Model Life Tables
and three parameters: (1-2) direct and indirect estimates of infant and child
mortality, and (3) adjusted estimates of adult mortality (45q15) derived from
(a) recent household deaths data from the 1954-1955 Demographic Survey,
and the 1983 and 1996 censuses; (b) parental orphanhood from the
1999 and 2005 DHS ; (c) siblings deaths from the 1999, 2005 and 2012 DHS ;
(d) implied relationship between child mortality and adult mortality based
on the North model of the Coale-Demeny Model Life Tables in 1950-1955
and assumed to converge over time toward the South model of the CoaleDemeny Model Life Tables by the 1990s. Data from West African rural
demographic surveillance sites and urban vital registration were also
considered, including from the 1957 Urban Survey (Konkoure).
CD South relational
model for non-HIV
mortality
Other HIV
-
Gambia
Estimated using the South model of the Coale-Demeny Model Life Tables and
three parameters: (1-2) direct and indirect estimates of infant and child
mortality, and (3) adjusted estimates of adult mortality (45q15). Adult
mortality was derived from the relationship to child mortality implied by the
North model of the Coale-Demeny Model Life Tables. Adult mortality
estimates derived from recent household deaths data from the 1973
census, and from parental orphanhood from the 1973, 1983 and 2003
censuses, 2001 Baseline Survey in Lower, Central and Upper River Divisions,
2000 and 2005/06 MICS surveys and rural demographic surveillance sites
were also considered. The results of the 2013 DHS were considered.
CD South relational
model for non-HIV
mortality
Other HIV
-
World Health Organization
Page 21
Guinea-Bissau
Estimated using the South model of the Coale-Demeny Model Life Tables
and three parameters: (1-2) direct and indirect estimates of infant and child
mortality, and (3) adjusted estimates of adult mortality (45q15). Adult
mortality estimates were derived from (a) parental orphanhood from the
2000 and 2006 MICS surveys, and (b) implied relationship between child
mortality and adult mortality based on the North model of the CoaleDemeny Model Life Tables in 1950-1955 and assumed to converge over
time toward the South model of the Coale-Demeny Model Life Tables by
the 1990s. Data from West African rural demographic surveillance sites
(e.g., Bandim) and urban vital registration were also considered.
CD South relational
model for non-HIV
mortality
Other HIV
-
Equatorial
Guinea
Estimated using the North model of the Coale-Demeny Model Life Tables
and implied relationships between life expectancy at birth and estimates of
infant and child mortality and between life expectancy at birth and
estimates of adult mortality (45q15). The adjusted estimates of 45q15 were
derived from (a) parental orphanhood data from the 2000 MICS; and (b)
intercensal survivorship from successive census age distributions (smoothed
and unsmoothed) for the period of 1983-1994. The demographic impact of
AIDS has been factored into the mortality estimates.
CD North relational
model for non-HIV
mortality
High HIV
-
Greece
Based on official life tables available through 2012.
Death registration data
vr
1980-2012
Grenada
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the West model of the Coale-Demeny
Model Life Tables.
CD West model life
tables
wpp
1985-2013
Guatemala
Based on: (a) registered deaths by age and sex and the underlying mid-year
population by age and sex through 2013; (b) age-sex-specific death rates
from the 1995, 1998/99, 2002 and 2008/09 Encuestas Nacionales de Salud
Materno Infantil (ENSMI); (c) age-sex-specific death rates from the 1987 and
1989 Encuestas Nacionales Socio-demográficas (ENSD); and (d) death rates
by age and sex from the 1950, 1964, 1973, 1981, 1994 and 2002 censuses.
The number of deaths was adjusted using the growth-balance method.
Death registration data
vr
1980-1981, 19832013
Guyana
Derived from child and adult mortality estimates through 2010 by assuming
that the age pattern of mortality conforms to the West model of the CoaleDemeny Model Life Tables. Adult mortality estimates based on the Global
Burden of Disease Study 2010 were considered.
CD West relational
model for non-HIV
mortality
vr
1984, 1988-2011
Honduras
Based on: (a) registered deaths by age and sex and the underlying mid-year
population by age and sex from 1950 through 1983 and from 2000 through
2011; (b) ages-sex-specific death rates from the 2005/06 and 2011/12
ENDESA (DHS); (c) ages-sex-specific death rates from the 1991/92, 1996 and
2001 ENESF; (d) ages-sex-specific death rates from the 1987 EFHS, the 1984
MCH/PF, the 1971/72 and 1983 EDENH, the 1981 National Contraceptive
Prevalence Survey (EPAH); and (e) estimates from the 1974, 1988 and 2001
censuses. The number of deaths was adjusted using the growth-balance
method.
Survey, census and
death registration data
wpp
1980-1983, 19871990, 2008-2013
Croatia
Based on deaths registered through 2013 by age and sex and the underlying
population by age and sex.
Death registration data
vr
1982-2013
Haiti
Based on: (a) estimates from the 1987, 1994/95, 2000 and 2005/06 DHS; (b)
estimates from the 1982 and 2003 censuses, (c) registered deaths by age and
sex adjusted for incompleteness using the growth-balance method and the
1971 census population by age and sex; and (d) estimates from the
1977 Enquête Haitienne sur la Fécondité (EHF).
Survey, census and
death registration data
Other HIV
1980, 1997, 1999,
2001-2004
Hungary
Based on official estimates of life expectancy available through 2013. The
age pattern of mortality is based on official life tables for through 2013.
Death registration data
vr
1980-2013
Indonesia
Derived from estimates of infant, child, adult and old-age mortality. Adult
and old age mortality estimates are based on: (a) the 2002/03, 2007 and
2012 DHS, (b) the 1990, 2000 and 2010 censuses, and (c) the 2007/08
Indonesia Family Life Survey (IFLS), and (d) the SUSENAS surveys (National
Socio-economic Surveys).
CD North relational
model for non-HIV
mortality
wpp
-
India
Based on life tables derived from age and sex-specific mortality rates from
the Sample Registration System from 1968-1969 up to 2013 adjusted for
infant and child mortality, and for adult death underregistration by using
the growth-balance and synthetic-extinct generation methods.
Death registration data
wpp
1988-2008
Ireland
Based on official estimates of life expectancy through 2009. Death rates
estimated from 2010 to 2012 were also considered.
Death registration data
vr
1980-2012
World Health Organization
Page 22
Iran (Islamic
Republic of)
Based on life tables derived from age and sex-specific mortality rates from
(a) registered 2000-2012 annual deaths adjusted for infant and child
mortality, and for adult death underregistration using the growth-balance
and synthetic-extinct generation methods; (b) the 1956-1966 intercensal
survival, 1973/76 Population Growth Survey, 1976, 1986 and 1991
censuses, and 2000 Demographic and Health Survey; and (c) estimates of
infant and child mortality by assuming that the age pattern of mortality
conforms to the West model of the Coale-Demeny Model Life Tables. For
1980-1988, excess mortality due to the war was factored in the overall
mortality levels based on the PRIO Battle Deaths Dataset version 3.0,
released in October 2009.
Death registration data
wpp
1983-1984, 1986,
1991, 1995-1999,
2001, 2005-2008
Iraq
Estimated using the West model of the Coale-Demeny Model Life Tables
and three parameters: (1-2) direct and indirect estimates of infant and child
mortality, and (3) adjusted estimates of adult mortality (45q15). Adult
mortality estimates are derived from: (a) recent household deaths data
from the 1973/74 Demographic Sample Survey and Sample Registration,
and 1999 Child and Maternal Mortality Survey (female only); (b) parental
orphanhood from the 1997 census, 2004 Iraq Living Conditions Survey and
2006 MICS3; (c) siblings deaths from the 1990 Iraq Immunization,
Diarrhoeal Disease, Maternal and Childhood Mortality Survey (female only),
and the 2006/07 Iraq Family Health Survey; (d) intercensal survivorship
from successive census age distributions (smoothed and unsmoothed) for
periods 1957-1965, 1965-1977, 1977-1987 and 1987-1997; (e) implied
relationship between child mortality and adult mortality based on the West
model of the Coale-Demeny Model Life Tables. For 1980-1988, excess
mortality due to the war was factored in the overall mortality levels based
on the PRIO Battle Deaths Dataset version 3.0, released in October 2009. For
2000-2005 and 2005-2010, excess mortality due to the war was
factored in the overall mortality levels; there is a high level of uncertainty in
the current estimates. The estimated numbers of war related deaths, as
provided by the Iraqi Ministry of Health, have also been taken into account.
CD West model life
tables
wpp
1987-1989, 2008
Iceland
Based on official life tables available through 2013.
Death registration data
vr
1980-2012
Israel
Based on life tables derived from official estimates of registered deaths and
enumerated census population by age and sex from 1948 to 2013. Mortality
rates for older ages were adjusted.
Death registration data
vr
1980-2014
Italy
Based on: (a) life tables for 2010, 2011 and 2012 from the National
Statistical Office (Istat) and Eurostat; (b) life tables through 2005-2009 from
the Human Mortality Database.
Death registration data
vr
1980-2012
Jamaica
Based on: (a) registered deaths by age and sex through 2005, adjusted for
underreporting of infant and child deaths; (b) official estimates for 1991,
2002, 2003 and 2006; and (c) estimates from the 2001 and 2011 censuses.
Census and death
registration data
Other HIV
1980-1991, 19962011
Jordan
Estimated using the West model of the Coale-Demeny Model Life Tables
and three parameters: (1-2) direct and indirect estimates of infant and child
mortality, and (3) estimates of adult mortality (45q15) implied by the
relationship between child mortality and adult mortality based on the South
model of the Coale-Demeny Model Life Tables in 1950-1955 and assumed to
converge over time toward the West model of the Coale-Demeny Model
Life Tables by the 1980s. Life tables based on the 1961 and 1979 censuses,
1972 National Fertility Survey and 1976 WFS, indirect estimates of adult
mortality based on widowhood data from the 1961 and 1979 censuses and
1976 WFS, as well as parental orphanhood from the 1976 WFS and 1981
Demographic Survey were also taken into account.
CD West model life
tables
wpp
2003-2004, 20082011
Japan
Based on life tables derived from official estimates through 2012.
Death registration data
vr
1980-2013
Kazakhstan
Based on official estimates of life expectancy available through 2008
adjusted for underreporting of infant and child mortality. The age pattern of
mortality is derived from a life table based on 2010-2013 data.
Death registration data
vr
1981-2010, 2012
World Health Organization
Page 23
Kenya
Estimated using the North model of the Coale-Demeny Model Life Tables
and implied relationships between life expectancy at birth and estimates of
infant and child mortality and between life expectancy at birth and
estimates of adult mortality (45q15). The adjusted estimates of 45q15 were
derived from: (a) household deaths data (unadjusted and adjusted for
underregistration using the growth-balance and synthetic-extinct generation
methods) from the 1969, 1979, 1989, 1999 and 2009 censuses; (b) parental
orphanhood from the 1983, 1989, 1993, 1998, 2003, and
2008/09 Kenya DHS, the 1977 World Fertility Survey, and all censuses
aforementioned; (c) siblings deaths from the 1989, 1993, 1998 2003, and
2008/09 Kenya DHS; and (d) intercensal survivorship from successive census
age distributions (smoothed and unsmoothed) for the period of 1969-2009.
The demographic impact of AIDS has been factored into the mortality
estimates.
CD North relational
model for non-HIV
mortality
High HIV
-
Kyrgyzstan
Based on official estimates of life expectancy available through 2013
adjusted for underreporting of infant and child mortality.
Death registration data
vr
1981-2013
Cambodia
Based on life tables derived from age and sex-specific mortality rates from:
(a) recent household deaths data from the 2004 Inter-Censal Population
Survey and 2008 census; (b) siblings deaths from the 2000, 2005 and 2010
DHS. Also, 1950-1955 life tables derived from estimates of child mortality
were used, by assuming that the age pattern of mortality conforms to the
West model of the Coale-Demeny Model Life Tables, as well as recent
household deaths data from the 1959 rural survey, and the 1962-1998
population reconstructions.
Survey, census data and
CD West Model life
tables
wpp
-
Kiribati
Based on: (a) estimates in the 2005 and 2010 censuses; (b) estimates from
deaths by age and sex from 1995 to 2001; (c) child mortality estimates by
assuming that the age pattern of mortality conforms to the West model of
the Coale-Demeny Model Life Tables; and (d) estimates from the Secretariat
of the Pacific Community were also considered.
CD West model life
tables
wpp
1991-2001
Republic of
Korea
Based on official estimates of life expectancy through 2012.
Death registration data
vr
1980-2013
Kuwait
Based on life tables derived from official estimates of registered deaths and
enumerated census population by age and sex from 1964 to 2010, adjusted
for infant and child mortality. Mortality rates for older ages were adjusted.
For 1950-1965, life tables were derived from estimates of infant and child
mortality by assuming that the age pattern of mortality conforms to the
South model of the Coale-Demeny Model Life Tables in 1950-1955 and
converges over time toward the estimated 1965-1970 life table.
Death registration data
wpp
1980-1989, 19912013
Lao People's
Democratic
Republic
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the West model of the Coale-Demeny
Model Life Tables.
CD West model life
tables
wpp
1995
Lebanon
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the West model of the Coale-Demeny
Model Life Tables. For the period 2010-2015, life expectancy at birth was
adjusted to account for different mortality patterns of large Syrian refugee
population.
CD West model life
tables
wpp
-
Liberia
Estimated using the South model of the Coale-Demeny Model Life Tables
and three parameters: (1-2) direct and indirect estimates of infant and child
mortality, and (3) adjusted estimates of adult mortality (45q15) derived from
(a) recent household deaths data from the 1969-1970 and 1970-1971
Population Growth Surveys ; (b) parental orphanhood from the 2007 DHS ;
(c) siblings deaths from the 2007 DHS ; (d) implied relationship between
child mortality and adult mortality based on the West model of the CoaleDemeny Model Life Tables in 1950-1955 and assumed to converge over
time toward the South model of the Coale-Demeny Model Life Tables by
the 1990s. Data from West African rural demographic surveillance sites and
urban vital registration were also considered.
CD South relational
model for non-HIV
mortality
Other HIV
-
Libya
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms the East model of the Coale-Demeny
Model Life Tables and converges over time toward the West model of the
Coale-Demeny Model Life Tables from 1950 to 2010.
CD West model life
tables
wpp
-
World Health Organization
Page 24
Saint Lucia
Based on: (a) official estimates of life expectancy available through 2005;
(b) registered deaths by age and sex through 2005 and underlying
population by age and sex; and (c) estimates from the 1991 and 2001
censuses.
Death registration data
vr
1980-2006, 20082012
Sri Lanka
Based on: life tables derived from official estimates of registered deaths and
population by age and sex from 1950 to 2010, adjusted for infant and child
mortality, and for adult death underregistration for males before 1980 by
using tabulations of paternal orphanhood (before marriage) by age of
respondent from the 1987 Sri Lanka DHS.
Death registration data
wpp
1980-2007
Lesotho
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the West model of the Coale-Demeny
Model Life Tables. The demographic impact of AIDS has been factored into
the mortality estimates.
CD West model life
tables for non-HIV
mortality
High HIV
-
Lithuania
Based on official estimates of life expectancy available through 2011. The
age pattern of mortality is based on life tables through 2011 from the
Human Mortality Database.
Death registration data
vr
1980-2013
Luxembourg
Based on official estimates of life expectancy available through 2012. The
age pattern of mortality is based on official life tables through 2012.
Death registration data
vr
1980-2013
Latvia
Based on official life tables available through 2012.
Death registration data
vr
1980-2012
Morocco
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality initially conforms to the mortality patterns
resulting from the blending from the East model of the Coale-Demeny
Model Life Tables to the West model of the Coale-Demeny Model Life
Tables from 1950 to 2015.
CD West model life
tables
wpp
1991-1998, 20082012
Republic of
Moldova
Based on official estimates of life expectancy available through 2012
adjusted for underreporting of infant and child mortality. The age pattern of
mortality is derived from a life table based on data for 2012.
Death registration data
vr
1981-2013
Madagascar
Based on: (a) estimates from the 1966 Demographic Survey and the 1973
and 1993 censuses; (b) estimates derived from registered age-sex-specific
deaths and underlying age-sex-specific population; (c) estimates derived
from implied relationships between child mortality and adult mortality from
the 1992, 1997, 2003/04 and 2008/09 DHS based on the North model of the
Coale-Demeny Model Life Tables; and (d) 1966 estimates from OECD were
also considered.
CD North relational
model for non-HIV
mortality
wpp
-
Maldives
Based on life tables derived from official estimates of registered deaths and
enumerated census population by age and sex from 1975 to 2012, adjusted
for infant and child mortality and for adult death underregistration for males
in 1980-1985 using the growth-balance and synthetic-extinct generation
methods. For 1950-1975, life tables were derived from
estimates of infant and child mortality by assuming that the age pattern of
mortality conforms to the South-Asian model of the United Nations Model
Life Tables in 1950-1955 and converges over time toward the estimated
1975-1980 life table.
Death registration data
vr
1984-2011
Mexico
Based on: (a) registered deaths by age and sex through 2013 and underlying
population by age and sex, (b) estimates from the 1992, 2006 and 2009
Encuesta Nacional de la Dinámica Demográfica (ENADID), (c) estimates from
the1978 and 1979 ENPUMA, and the 1976 WFS, (d) estimates from the
1970, 1990, 2000 and 2010 censuses. The number of deaths was adjusted
using the growth-balance method.
Death registration data
vr
1980-2013
The former
Yugoslav
Republic of
Macedonia
Based on official estimates of life expectancy available through 2012. The
age pattern of mortality is based on an official life table for 2010-2012.
Death registration data
vr
1982-2010
World Health Organization
Page 25
Mali
Estimated using the South model of the Coale-Demeny Model Life Tables
and three parameters: (1-2) direct and indirect estimates of infant and child
mortality, and (3) adjusted estimates of adult mortality (45q15) derived from
(a) recent household deaths data (unadjusted and adjusted for
underregistration using the growth-balance and synthetic-extinct
generation methods) from the 1957-1958 Demographic Survey (Central
Delta) and 1960-61 Demographic Survey, the 1976, 1987, 1998 and 2009
censuses; (b) parental orphanhood from the 1995-1996, 2001 and 2006
DHS ; (c) siblings deaths from the 1995-1996, 2001, 2006 DHS and 20122013 DHS; (d) intercensal survivorship from successive census age
distributions (smoothed and unsmoothed) for periods 1976-1987, 19871998, and 1998-2009 ; (e) implied relationship between child mortality and
adult mortality based on the North model of the Coale-Demeny Model Life
Tables in 1950-1955, and assumed to converge over time toward the South
model of the Coale-Demeny Model Life Tables by the 1980s.
CD South relational
model for non-HIV
mortality
Other HIV
1987
Malta
Based on official life tables through 2012.
Death registration data
vr
1980-2014
Myanmar
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the West model of the Coale-Demeny
Model Life Tables. Official estimates of life expectancy at birth by sex from
the 1991 Myanmar Population Change and Fertility Survey and the recent
household deaths data from the 2014 census were also considered.
CD West model life
tables
wpp
-
Montenegro
Based on official estimates of life expectancy available through 2010. The
age pattern of mortality is based on life tables for 1990, 2000 and 2006.
Death registration data
vr
1985-2010
Mongolia
Based on official estimates of life expectancy available through 2013
adjusted for underreporting of infant and child mortality.
Death registration data
vr
1991-2010
Mozambique
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the North model of the Coale-Demeny
Model Life Tables. The demographic impact of AIDS has been factored into
the mortality estimates.
CD North model life
tables for non-HIV
mortality
High HIV
-
Mauritania
Estimated using the South model of the Coale-Demeny Model Life Tables
and implied relationships between life expectancy at birth and estimates of
infant and child mortality, and estimates of adult mortality (45q15). Adult
mortality estimates were derived from: (a) parental orphanhood from the
2007 MICS3, 2000/01 DHS, 1964/65 Demographic Survey and 1981/82
Fertility Survey of Mauritania (WFS), (b) siblings deaths from the 2000/01
DHS, (c) household deaths data from the 1957 Fouta Toro survey, 1977 and
1988 censuses, and (d) intercensal survivorship from successive census age
distributions (smoothed and unsmoothed) for periods 1965-1977, 19771988 and 1988-2000.
CD South relational
model for non-HIV
mortality
wpp
1988
Mauritius
Based on official life tables through 2013.
Death registration data
vr
1980-2014
Malawi
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the South model of the Coale-Demeny
Model Life Tables. Estimates from the 1987, 1998 and 2008 censuses and
official estimates from the National Statistical Office of Malawi were also
considered. The demographic impact of AIDS has been factored into the
mortality estimates.
CD South model life
tables for non-HIV
mortality
High HIV
1987, 1998
Malaysia
Based on: (a) official estimates of life expectancy available through 2008,
and (b) 2011 and 2012 official estimates of deaths by age and sex and the
underlying population by age and sex.
Death registration data
wpp
1990-2009
Namibia
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the West model of the Coale-Demeny
Model Life Tables. Official estimates from Statistics Namibia were also
considered. The demographic impact of AIDS has been factored into the
mortality estimates.
CD West model life
tables for non-HIV
mortality
High HIV
-
World Health Organization
Page 26
Niger
Estimated using the South model of the Coale-Demeny Model Life Tables
and three parameters: (1-2) direct and indirect estimates of infant and child
mortality, and (3) adjusted estimates of adult mortality (45q15). Adult
mortality estimates were derived from (a) recent household deaths data
(unadjusted and adjusted for underregistration using the growth-balance
and synthetic-extinct generation methods) from the 1959/60 Demographic
Survey, 1977, 1988 and 2001 censuses; (b) parental orphanhood from the
1988 and 2001 censuses, 1992 and 1998 DHS, 2006 DHS-MISC3; (c) siblings
deaths from the 1992 DHS, 2006 DHS-MICS3 and 2012 DHS-MICS4; (d)
intercensal survivorship from successive census age distributions (smoothed
and unsmoothed) for periods 1977-1988, and 1988-2001; (e) implied
relationship between child mortality and adult mortality based on the North
model of the Coale-Demeny Model Life Tables in 1950-1955, and assumed
to converge over time toward the South model of the Coale-Demeny Model
Life Tables by the 1990s. Data from West African rural demographic
surveillance sites and urban vital registration were also considered.
CD South relational
model for non-HIV
mortality
wpp
-
Nigeria
Estimated using the South model of the Coale-Demeny Model Life Tables
and three parameters: (1-2) direct and indirect estimates of infant and child
mortality, and (3) adjusted estimates of adult mortality (45q15). Adult
mortality estimates were derived from: (a) recent household deaths data
from the 1965-1966 Nigerian rural demographic inquiry, the 2008 and 2013
DHS, and the 2010/11 GHS; (b) parental orphanhood from the 1986, 1999,
2003, 2008 and 2013 DHS, the 2007 MICS3 and 2010/11 GHS; (c) siblings
deaths from the 2008 DHS; (d) implied relationship between child mortality
and adult mortality based on the North model of the Coale-Demeny Model
Life Tables. Data from West African rural demographic surveillance sites
including for Malumfashi in 1962-1966 and 1974-1977 and urban vital
registration were also considered.
CD South relational
model for non-HIV
mortality
Other HIV
-
Nicaragua
Based on: (a) registered births and infant and child deaths from 1968
through 2011; (b) estimates from the 1998, 2001, 2006/07, and the
2011/12 (preliminary) ENDESA (DHS); (c) estimates from the 1992/93 Family
Health Survey, the 1993 and 2001 National Household Survey on Living
Standards Measurement (LSMS); the 1985/86 National Socio-Demographic
Survey, the 1978 National Retrospective Demographic Survey; and (d)
estimates from the 1953, 1963, 1971, 1995 and 2005 censuses. The number
of deaths was adjusted using the growth-balance method.
Death registration data
vr
1987-1994, 19962013
Netherlands
Based on official estimates of life expectancy derived from registered
deaths through 2013. The age pattern of mortality is based on official life
tables for 1950 to 2013.
Death registration data
vr
1980-2013
Norway
Based on official life tables available through 2013.
Death registration data
vr
1980-2013
Nepal
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the West model of the Coale-Demeny
Model Life Tables.
CD West model life
tables
wpp
1981, 1991
New Zealand
Based on official estimates of life expectancy available through 2009. The
age pattern of mortality is based on life tables through 2008 from the
Human Mortality Database.
Death registration data
vr
1980-2011
Oman
Based on life tables derived from official estimates of registered deaths for
2009-2011 and 2010 enumerated census population by age and sex,
adjusted for infant and child mortality. For 1950-2007, life tables were
derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the South model of the Coale-Demeny
Model Life Tables in 1950-1955 and converges over time toward the
estimated 2009-2011 life table.
Death registration data
wpp
2009-2010
Pakistan
Based on life tables derived from age and sex-specific mortality rates from:
(a) the 1962-1965 Population Growth Estimation Experiment, 1968-1971
Population Growth Survey I, 1976-1979 Population Growth Survey II; (b) the
1984-2007 annual Pakistan Demographic Surveys adjusted for infant and
child mortality, and for adult death underregistration for males in 19501970 using the growth-balance and synthetic-extinct generation methods,
as well as cross-validation with other countries experiencing similar
mortality levels; and (c) estimates of infant and child mortality by assuming
that the age pattern of mortality conforms to the South-Asian model of the
United Nations Model Life Tables. Estimates are consistent with those
based on parental survival and widowhood data from the 1984 PDS.
Mortality rates for older ages were adjusted.
Survey, census and
sample death
registration data
wpp
1984-1993
World Health Organization
Page 27
Panama
Based on: (a) registered deaths by age and sex and underlying population
by age and sex from 1952 through 2013; (b) estimates from the 1950, 1970,
1980, 1990, 2000, and 2010 censuses; and (c) official life tables for 1960,
1970, 1979, 1989 and 1999. The number of deaths was adjusted using the
growth-balance method.
Death registration data
vr
1980-2013
Peru
Based on: (a) registered deaths by age and sex through 2012 and the
underlying population by age and sex; (b) official estimates in 1961, 1965,
1980, 1990, 1995, 2000, and 2005; (c) estimates of infant and child
mortality from 2004-2014 continuous Encuestas Demográficas y de Salud
Familiar (ENDES/DHS), and the 1986, 1991/92, 1996 and 2000 ENDES; (d)
estimates of infant and child mortality from the 1977/78 World Fertility
Survey, the 1974/76 National Demographic Survey; and (e) estimates from
the 1961, 1972, 1981, 1993, and 2007 censuses. The number of deaths was
adjusted using the growth-balance method.
Survey, census and
death registration data
vr
1980-1992, 19942013
Philippines
Based on: (a) child mortality estimates from the 1998, 2003, 2006 and
2013 DHS, and 2006 Family Planning Survey, (b) estimates od infant and
child mortality, (c) official estimates from a life table of 2006, and (d) the
West model of the Coale-Demeny Model Life Tables and the Lee-Carter
method.
Death registration data
vr
1980-2005, 20072009
Papua New
Guinea
Based on: (a) infant and child mortality estimates, (b) parental survivorship
(orphanhood) data by age from the 2000 census, (c) child mortality data
from the 1996 and 2006 PNG DHS, by assuming that the age pattern of
mortality conforms to the Far Eastern model of the United Nations Model
Life Tables.
UN Far Eastern relational
model for non-HIV
mortality
wpp
1987-1998
Poland
Based on official estimates of life expectancy available through 2013. The
age pattern of mortality is based on official life tables through 2013.
Death registration data
vr
1980-2013
Puerto Rico
Based on: (a) registered deaths by age and sex through 2014 and underlying
population by age and sex, and (b) official estimates of life expectancy
available through 2010.
Death registration data
vr
1980-2013
Democratic
People's
Republic of
Korea
Based on the number of deaths in household during the 12-month period
preceding the 1993 and 2008 censuses classified by age and sex.
Census data
wpp
1993
Portugal
Based on: (a) official estimates of life expectancy available through 2012;
(b) registered deaths by age and sex through 2011 and underlying
population by age and sex; and (c) estimates from the Human Mortality
Database and Eurostat were also considered.
Death registration data
vr
1980-2013
Paraguay
Based on: (a) registered deaths by age and sex through 2006 and underlying
population by age and sex; (b) estimates from the 2004 and 2008 ENDSSR
and the1995/96 ENDSR; (c) estimates from the 2003 WHS, the 1998
National Maternal and Child Health Survey, the 1990 DHS, the 1987 RHS,
the 1979 WFS, and the 1977 National Demographic Survey; and (d)
estimates from the 1950, 1962, 1972, 1992, and 2002 censuses, and
preliminary results from the 2012 census. The number of deaths was
adjusted using the growth-balance method.
Survey, census and
death registration data
wpp
1980-1987, 1990,
1992, 1994-2013
Occupied
Palestinian
Territory
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the West model of the Coale-Demeny
Model Life Tables.
CD West model life
tables
wpp
2011
Qatar
Based on life tables derived from official estimates of registered deaths and
enumerated census population by age and sex from 1981 to 2011, adjusted
for infant and child mortality. Mortality rates for older ages were adjusted.
For 1950-1980, life tables were derived from estimates of infant and child
mortality by assuming that the age pattern of mortality conforms to the
South model of the Coale-Demeny Model Life Tables in 1950-1955 and
converges over time toward the estimated 1980-1985 life table.
Death registration data
wpp
1981-1983, 19852012
Romania
Based on official life tables through 2012.
Death registration data
vr
1980-2012
Russian
Federation
Based on official estimates of life expectancy available through 2012. The
age pattern of mortality is based on life tables through 2012 from the
Human Mortality Database. Both estimates incorporate an adjustment to
infant mortality.
Death registration data
vr
1980-2011
World Health Organization
Page 28
Rwanda
Based on the estimated level of infant mortality and taking into account the
unusual numbers of deaths caused by the 1993-1994 civil war. The
demographic impact of AIDS has been factored into the mortality estimates.
CD North model life
tables for non-HIV
mortality
High HIV
-
Saudi Arabia
Based on official estimates of life expectancy at birth for 2010-2013. For
1995-2010, based on life-tables, calculated from adjusted deaths in the past
12 months by age and sex, and the population by age and sex from the
1999 Demographic Survey, 2004 census and 2007 Demographic Survey
adjusted for infant and child mortality, and old-age mortality. For 19501995, life tables were derived from estimates of infant and child mortality
by assuming that the age pattern of mortality conforms to the South model
of the Coale-Demeny Model Life Tables in 1950-1955 and converges over
time toward the West model of the Coale-Demeny Model Life Tables and
the estimated 1999-2007 life tables. Life tables based on annual deaths
from the 2000 Demographic Survey, and on 2005 and 2009 registered
deaths were also considered.
Death registration data
wpp
2009, 2012
Sudan
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the North model of the Coale-Demeny
Model Life Tables.
CD North model life
tables
wpp
-
Senegal
Estimated using the South model of the Coale-Demeny Model Life Tables
and three parameters: (1-2) direct and indirect estimates of infant and child
mortality, and (3) adjusted estimates of adult mortality (45q15). Adult
mortality estimates were derived from (a) recent household deaths data
(unadjusted and adjusted for underregistration using the growth-balance
and synthetic-extinct generation methods) from the 1978/79 Multiround
Survey, 1988,2002 and 2013 censuses; (b) parental orphanhood from these
sources and the 1986, 1992/93, 2005 DHS and 2010/11 DHS-MICS, 1988
census, and 2000 MICS; (c) siblings deaths from the 1992/93, and 2005 DHS
and 2010/11 DHS-MICS; (d) intercensal survivorship from successive census
age distributions (smoothed and unsmoothed) for periods 1976-1988 and
1988-2002; (e) implied relationship between child mortality and adult
mortality based on the North model of the Coale-Demeny Model Life Tables
in 1950-1955, assumed to converge over time toward the South model of
the Coale-Demeny Model Life Tables by the 1990s; and (f) central deaths
rate by age from the 2013 census.
CD South relational
model for non-HIV
mortality
wpp
-
Singapore
Based on: (a) official estimates of life tables through 2010, and (b) 20112013 official estimates of deaths and population by age and sex.
Death registration data
wpp
1980-2014
Solomon Islands
Based on: (a) data on children ever born and surviving from the 1986 and
1999 censuses; (b) official estimates based on census analysis and WHOGBD estimates for 2006; and (c) 1980-1984 life table based on indirect
methods assuming that the pattern of mortality conforms to the West
model of the Coale-Demeny Model Life Tables..
CD West relational
model for non-HIV
mortality
wpp
-
Sierra Leone
Estimated using the South model of the Coale-Demeny Model Life Tables
and three parameters: (1-2) direct and indirect estimates of infant and child
mortality, and (3) adjusted estimates of adult mortality (45q15). Adult
mortality estimates were derived from: (a) recent household deaths data
from the 1992 Demographic and social monitoring survey and the 2004
census; (b) parental orphanhood from the 1973 pilot census. 1974, 1985
and 2004 censuses, 2000 MICS2, 2005 MICS3, 2007 CWIQ and 2008 DHS
surveys; (c) female sibling deaths from the 2005 MICS3, and sibling deaths
from the 2008 and 2013 DHS; (d) intercensal survivorship from successive
census age distributions (smoothed and unsmoothed) for periods 19631974, 1974-1985, 1985-2004; (e) implied relationship between child
mortality and adult mortality based on the South model of the CoaleDemeny Model Life Tables for males, and the North model for females for
the 1950-1970 period. Data from West African rural demographic
surveillance sites (including from the 1973/75 Ad-hoc survey in Greater
Freetown, the Western area and Makeni in the Northern Province) and
urban vital registration were also considered.
CD South relational
model for non-HIV
mortality
Other HIV
-
El Salvador
Based on: (a) registered deaths from 1975 through 2008, and underlying
population by age and sex; (b) estimates from the 1950, 1963, 1971, 1992
and 2007 censuses; (c) estimates from the 1973 to 2008 Encuesta Nacional
de Salud Familiar (FESAL), the 1992 EHS, and the 1985 DHS. The number of
deaths was adjusted using the growth-balance method.
Death registration data
vr
1980-2012
Somalia
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the North model of the Coale-Demeny
Model Life Tables. Estimates from GBD-WHO were also considered.
Additional deaths due to the famine of 1992 and the war have been
CD North model life
tables
wpp
-
World Health Organization
Page 29
factored into the mortality estimates.
Serbia
Based on official estimates of life expectancy available through 2011. The age
pattern of mortality is based on official life tables for 1997, and for 2005 to
2012.
Death registration data
vr
1985-2013
South Sudan
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the North model of the Coale-Demeny
Model Life Tables.
CD North model life
tables
Other HIV
-
Sao Tome and
Principe
Based on: (a) official estimates and life table derived from the 2001 census;
and (b) death rates calculated from registered deaths by age and sex
through 1979 and underlying population by age and sex. Estimates
from WHO-GBD and estimates derived from child and adult mortality using
North Model of the Coale-Demeny Model Life Table were also considered.
CD North relational
model for non-HIV
mortality
wpp
1984-1985, 1987
Suriname
Based on: (a) registered deaths by age and sex through 2013 and underlying
population by age and sex, and (b) official estimates for 1963, 1980, 2004
and 2006.
Death registration data
vr
1980-2012
Slovakia
Based on official life tables through 2013.
Death registration data
vr
1982-2014
Slovenia
Based on official estimates of life expectancy available through 2012. The
age pattern of mortality is based on blended life tables (from the East
model of the Coale-Demeny Model Life Tables assumed to convert over
time toward the empirical data in 1980) between 1950 and 1980, and
official life tables from 1980 to 2012.
Death registration data
vr
1982-2010
Sweden
Based on official life tables available through 2013.
Death registration data
vr
1980-2014
Swaziland
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the West model of the Coale-Demeny
Model Life Tables. The demographic impact of AIDS has been factored into
the mortality estimates.
CD West model life
tables for non-HIV
mortality
High HIV
-
Seychelles
Based on: (a) official estimates available through 2014, and (b) registered
deaths by age and sex through 2014 and underlying population by age and
sex.
Death registration data
wpp
1980-2014
Syrian Arab
Republic
For 2005-2010, based on a life-table calculated from 2005-2007 registered
deaths by age and sex, and post-censal population estimates by age and sex
derived from the 2004 census and 2010 official estimates adjusted for
infant and child mortality, and old-age mortality. For 1950-2005, due to the
lack of adult mortality information and life tables for this period, life tables
were derived from estimates of infant and child mortality by assuming that
the age pattern of mortality conforms for males to the South model of the
Coale-Demeny Model Life Tables in 1950-1955 and converges over time
toward the West model of the Coale-Demeny Model Life Tables and the
estimated 2005-2007 life table. For females a similar approach was used
assuming that the age pattern of mortality conformed since 1950 to the
West model. For each sex, the underlying mortality pattern and implied
adult mortality, are consistent with the life table from the 1976-1979 Syrian
Follow-up Demographic Survey. For the 2010-2015 period, excess mortality
due to the conflict was taken into account.
Death registration data
wpp
1983-1984, 1998,
2000-2001, 2004,
2008-2010
Chad
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the North model of the Coale-Demeny
Model Life Tables.
CD North model life
tables
Other HIV
-
Togo
Estimated using the South model of the Coale-Demeny Model Life Tables
and three parameters: (1-2) direct and indirect estimates of infant and child
mortality, and (3) adjusted estimates of adult mortality (45q15), Adult
mortality estimates were derived from (a) recent household deaths data
from the 1960 survey, 1970 and 1981 censuses; (b) parental orphanhood
from the 1998 DHS, 2000 MICS2 and 2006 MICS3; (c) siblings deaths from
the 1998 DHS; (d) implied relationship between child mortality and adult
mortality based on the North model of the Coale-Demeny Model Life Tables
in 1950-1955 and assumed to converge over time toward the South model of
the Coale-Demeny Model Life Tables by the 1990s.
CD South relational
model for non-HIV
mortality
Other HIV
-
Thailand
Based on life tables derived from official estimates of registered deaths and
enumerated census population by age and sex from 1948 to 2011, adjusted
Death registration data
Other HIV
1980-2009
World Health Organization
Page 30
for infant and child mortality and for underregistration of adult deaths.
Tajikistan
Based on registered deaths and population by age and sex through 2008,
adjusted for underregistration of deaths.
Death registration data
vr
1981-1982, 19852005
Turkmenistan
Based on official estimates of life expectancy available through 2006,
adjusted for underregistration of deaths.
Death registration data
vr
1981-1982, 19852013
Timor-Leste
Based on child mortality and adult mortality estimates from the 2009/10
Timor-Leste DHS. Life tables are estimated using the Flexible twodimensional model life table and Lee-Carter method. For 1950-2005,
derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the West model of the Coale-Demeny
Model Life Tables. Official estimates of life expectancy at birth for the year
2002 were also taken into account.
CD West model life
tables
wpp
-
Tonga
Based on: (a) the registered deaths by age and sex from 1957 to 1966 and
from 1982 to 2006 and underlying population by age and sex; and (b)
estimates from the 1996 and 2006 censuses by assuming that the age
pattern of mortality conforms to the Far Eastern model of the United
Nations Model Life Tables. Estimates from the Secretariat of the Pacific
Community were also considered.
UN Far Eastern model
life tables
wpp
1992-2003
Trinidad and
Tobago
Based on: (a) registered death by age and sex through 2005 and underlying
population by age and sex, and (b) official estimates through 2000.
Death registration data
vr
1980-2009
Tunisia
Based on official estimates of life expectancy from 1995 to 2012 from INS
Tunisia. The age pattern of mortality is based on national life table from
various years adjusted for under-five mortality.
Death registration data
wpp
1980, 1987-1989,
1991-2000, 2009,
2013
Turkey
Based on: (a) adjusted estimates from registered deaths by age and sex from
1952 to 2006 and for 2009 with underlying population by age and sex; (b)
official estimates for 1989, 2006, 2008 and 2011; and (c) estimates from
1990 to 2010 from the Turkish Institute of Statistics.
Death registration data
wpp
1999-2002, 20042013
China: Province
of Taiwan only
Based on official estimates of life expectancy derived from registered
deaths through 2009.
Death registration data
wpp
-
United Republic
of Tanzania
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the North model of the Coale-Demeny
Model Life Tables. The demographic impact of AIDS has been factored into
the mortality estimates. Estimates of adult mortality were also considered.
These were based on: (a) parental orphanhood from the 1978, 1988 and
2002 censuses, and the 1992, 1996, 1999, 2004/05 and 2010 DHS; (b)
siblings deaths from the above DHS; (c) intercensal survivorship from
successive census age distributions (smoothed and unsmoothed) for the
period of 1988-2012.
CD North model life
tables for non-HIV
mortality
High HIV
1988
Uganda
Derived from estimates of infant and child mortality, and adult mortality by
assuming that the age pattern of mortality conforms to the North model of
the Coale-Demeny Model Life Tables. Adult mortality (45q15) estimates were
based on: (a) parental orphanhood from the 1969, 1991, and 2002 censuses,
and the 1988/89, 1995, 2001, and 2006 DHS; (b) siblings deaths from the
above DHS; (c) intercensal survivorship from successive census age
distributions (smoothed and unsmoothed) for the period of 1991-2002. The
demographic impact of AIDS has been factored into the mortality estimates.
CD North relational
model for non-HIV
mortality
High HIV
-
Ukraine
Based on official estimates of life expectancy available through 2013. The
age pattern of mortality is based on life tables through 2013 from the
Human Mortality Database. Both estimates incorporate an adjustment to
infant mortality.
Death registration data
vr
1981-2012
Uruguay
Based on: (a) registered deaths by age and sex through 2013 and
underlying population by age and sex; (b) official estimates from 1964 to
2008; and (c) estimates from the 1963, 1975, 1985, 1996, 2004, and 2011
censuses. The number of deaths was adjusted using the growth-balance
method.
Death registration data
vr
1980-2010, 20122013
United States of
America
Based on official estimates of life expectancy available through 2011. The
age pattern of mortality is based on life tables through 2011 from the
Human Mortality Database.
Death registration data
vr
1980-2013
World Health Organization
Page 31
Uzbekistan
Based on official estimates of life expectancy available through 2008,
adjusted for underregistration of deaths.
Death registration data
vr
1981-2005
Saint Vincent
and the
Grenadines
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the West model of the Coale-Demeny
Model Life Tables. Registered deaths by age and sex through 2009 with
underlying population by age and sex were considered.
CD West model life
tables
wpp
1980-2013
Venezuela
(Bolivarian
Republic of)
Based on: (a) registered deaths by age and sex from 1950 through 2009 and
underlying population by age and sex; (b) estimates from the 1950, 1961,
1971, 1981, 1990, 2001 and 2011 censuses; (c) official estimates for 1974,
1975, 1985, 2000-2002 and 2007; and (d) estimates from the 1977 World
Fertility Survey and the 1998 Population and Family Survey. The number of
deaths was adjusted using the growth-balance method.
Death registration data
vr
1980-2012
Viet Nam
Based on life tables derived from age and sex-specific mortality rates from:
(a) recent household deaths data from the 1979, 1989, 1990 and 2009
censuses (unadjusted and adjusted for underregistration using the growthbalance and synthetic-extinct generation methods), and from the 2007
Population Change and Family Planning survey; (b) annual deaths for 2009
from the Viet Nam national sample mortality surveillance programme
adjusted for infant and child mortality, and for adult death completeness
according to capture-recapture survey; (c) direct and indirect estimates
based on parental orphanhood and siblings survival from the 1991 Vietnam
Life History Survey and 1995/98 Vietnam Longitudinal Survey; and (d) 19791989 intercensal survival estimates adjusted for outflows of refugees and
differential completeness of census enumeration. For 1950-1970 life tables
were derived from estimates of infant and child mortality by assuming that
the age pattern of mortality conforms to the average experienced of the
North and West models of the Coale-Demeny Model Life Tables in 19501955 and converged over time toward 1980s life tables. For 1965-1975,
excess mortality due to the war was factored in the overall mortality levels
based on direct and indirect adult mortality estimates derived from parental
orphanhood and siblings survival from the 1991 VHS and 1995/98
VLS, and from the PRIO Battle Deaths Dataset.
Death registration data
wpp
-
Vanuatu
Based on: (a) infant and child mortality estimates; (b) parental survivorship
(orphanhood) data by age of respondent from the 1999 census; and (c) the
assumption that the age pattern of mortality conforms to the Far Eastern
model of the United Nations Model Life Tables.
UN Far Eastern model
life tables
wpp
-
Samoa
Based on: (a) registered deaths by age and sex from 1980 through with the
underlying population by age and sex, and (b) estimates from the 1999 and
2009 Samoa DHS. The age pattern of mortality was assumed to conform to
the Far Eastern model of the United Nations Model Life Tables. Estimates
from the 2001, 2006 and 2011 censuses were also considered.
UN Far Eastern model
life tables
wpp
1980, 1992-1993
Yemen
Estimated using the West model of the Coale-Demeny Model Life Tables
and three parameters: (1-2) direct and indirect estimates of infant and child
mortality, and (3) estimates of adult mortality (45q15). Adult mortality
estimates were implied by the relationship between child mortality and
adult mortality based on the South model of the Coale-Demeny Model Life
Tables and assumed to converge over time toward the West model of the
Coale-Demeny Model Life Tables by the 1980s. Indirect estimates of adult
mortality based on widowhood data from the1979 WFS, as well as parental
orphanhood from this survey and the 2004 census were also considered.
Official estimates of life expectancy at birth from the Central Statistical
Organization of Yemen were also taken into account.
CD West relational
model for non-HIV
mortality
wpp
-
South Africa
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the Far Eastern model of the United
Nations Model Life Tables. Official estimates from Statistics South Africa
and the Actuarial Society of South Africa were also considered. The
demographic impact of AIDS has been factored into the mortality estimates.
Death registration data,
UN Far Eastern model
life tables for non-HIV
mortality
High HIV
1980-1982, 19842013
Zambia
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the North model of the Coale-Demeny
Model Life Tables. The demographic impact of AIDS has been factored into
the mortality estimates.
CD North model life
tables for non-HIV
mortality
High HIV
-
Zimbabwe
Derived from estimates of infant and child mortality by assuming that the
age pattern of mortality conforms to the North model of the Coale-Demeny
Model Life Tables. The demographic impact of AIDS has been factored into
the mortality estimates.
CD North model life
tables for non-HIV
mortality
High HIV
1982, 1986, 19901996, 1998, 2002
World Health Organization
Page 32
World Health Organization
Page 33
Annex B: Data sources and methods for mortality shocks
Natural disasters
Estimated deaths for major natural disasters were obtained from the EM-DAT/CRED International
Disaster Database (1). EM-DAT includes epidemics and some man-made disasters that are classified as
transport injuries etc, these are excluded from mortality estimates for natural disasters. Since 2000,
three major natural disasters that were associated with more than 100 000 deaths have dominated the
picture: the Asia tsunami in 2004; the Myanmar cyclone in 2008; and the Haiti earthquake in 2010
(Figure 1). The number of disasters has been declining in the last decade and the number of people
affected reached its lowest levels since 2000 in 2012 and 2013. Out of over one million disaster-related
deaths during 2000–2014, 61% occurred in Asia where 60% of the global population live, about one in
five people reported killed lived in the Americas. Africa (6%) and Oceania (less than 1% of deaths) had
much smaller proportions).
Figure 1 Number of people reported killed in natural and technological disasters, 2000–2014
Age-sex distributions were based on a number of studies of earthquake deaths (2, 3) and tsunami
deaths (4, 5).
Conflict deaths
Country-specific estimates of war and conflict deaths have been updated for the entire period 19902015 using revised methods together with information on conflict intensity, time trends, and mortality
World Health Organization
Page 34
obtained from a number of war mortality databases (described below). These estimates relate to deaths
for which the underlying cause (following ICD conventions) was an injury due to war, civil insurrection or
organized conflict, whether or not that injury occurred during the time of war or after cessation of
hostilities. The estimates include injury deaths resulting from all organized conflicts, including organized
terrorist groups, whether or not a national government was involved. They do not include deaths from
other causes (such as starvation, infectious disease epidemics, lack of medical intervention for chronic
diseases), which may be counterfactually attributable to war or civil conflict.
Methods used previously by WHO for estimation of direct conflict deaths were developed in the early
2000s and applied adjustment factors for under-reporting to estimates of battlefield or conflict deaths
from a variety of published and unpublished conflict mortality databases (5-9). Murray et al. (10)
summarized the issues with estimation of war deaths, and emphasized the very considerable
uncertainty in the original Global Burden of Disease estimates (11) and subsequent WHO estimates for
conflict deaths. WHO published estimates for the years 2000 through 2008 used adjustment factors
based on conflict intensity developed from an analysis of likely levels of under-reporting (12-15). These
adjustment factors ranged from around 3 to higher than 4 in sub-Saharan Africa.
Obermeyer, Murray and Gakidou (16) more recently analyzed data on deaths due to conflict from postconflict sibling histories collected in the 2002 to 2003 WHO World Health Survey (WHS) program. They
used data from 13 countries with more than 5 reported sibling deaths from war injuries in at least one
10-year period to estimate total war deaths for these countries for the period 1955-2002. The authors
then compared their estimates of war deaths to the number of war deaths estimated in the UCDP Battle
Deaths database (17) to derive an average adjustment factor of 2.96. Garfield and Blore (18) noted that
a very small number of war deaths for Georgia resulted in an outlier ratio of 12.0 which heavily
influenced the overall ratio of 2.96. They reanalyzed the WHS-derived war deaths dataset excluding
Georgia, to obtain an overall revised adjustment factor of 2.21.
The revised WHO country-specific estimates of war and conflict deaths for the period 1990-2015 make
use of estimates of direct deaths from three datasets: Battle-Related Deaths (version 5), Non-State
Conflict Dataset (UCDP version 2.4), and One-sided Violence Dataset (UCDP version 1.4) from 1989 to
2011 (19-21). Using these three datasets, instead of focusing solely on battle-related deaths, reduces
the likelihood that overall direct conflict deaths are underestimated. However, it is likely that a degree
of undercounting still occurs in the count-based datasets, and a revised adjustment factor of 1.91 has
been applied to the annual battle death main estimates for state-state conflicts. No adjustments were
applied to estimated conflict deaths (main estimates) for non-state conflict deaths, and one-sided
violence.
The adjustment factor 1.91 is the average of the factor of 2.21 obtained by Garfield and Blore (18) and
of a factor of 1.66 derived from comparing total deaths in the UCDP battle deaths dataset with those
estimated by the Peace Research Institute Oslo (PRIO) for the years 1989-2008 (22). As shown in the
graph below, the PRIO estimates are systematically higher than those of the UCDP.
World Health Organization
Page 35
The UCDP dataset is compiled primarily by counting the annual total of combat-related fatalities
(national and global) from reports of fatalities in individual violent incidents (battles, clashes, etc.) in
each state-based conflict. UCDP uses a variety of sources, including news reports, reports from human
rights organizations and nongovernmental organizations, etc. Since it is highly unlikely that all reports of
battle deaths will be recorded—particularly in conflicts where outside observers are banned from war
zones—this methodology will almost certainly underestimate the actual number of battle deaths. By
contrast, the PRIO dataset relies heavily on summary estimates—i.e., expert assessments of overall
fatalities. There is no reason to assume that summary estimates will systematically undercount battle
deaths as does UCDP’s incident-based estimation method.
Note that the application of a single adjustment factor for all state-state conflicts may result in deaths
for specific conflicts being over- or under-estimated. For the following countries, the multiplier was
adjusted downwards for low intensity years: Mexico (drug gangs), DR Congo, Columbia, Eritrea/Ethiopia
(1990-2000). For these conflict, estimated deaths from other sources suggest that UCDP figures provide
reasonable estimates without additional adjustment. For several conflicts where more specific sources
of information are available, these have been used to revise estimated deaths:
Iraq (81, 82).
Iraq
The conflict death toll in Iraq following the US-led invasion in March 2003 has been the
subject of much discussion with estimates for violent deaths to end June 2006 ranging
from 47,668 (Iraq Body Count) to 601,027 in a 2006 household survey (). The Iraq
Family Health Survey (IFHS), conducted in 2006-2007 by relevant Iraq Government
Ministries in collaboration with WHO, provided new evidence on mortality in Iraq for
the three years post-invasion (24). Latest counts of reported deaths in Iraq by the Iraq
Body Count (25) were compared with conflict deaths for the period 2003-2006
estimated from the Iraq Family Health Survey 2006 (24). This nationally representative
World Health Organization
Page 36
survey of 9,345 households included questions on deaths of adult siblings of
respondents, and deaths in the household. Sibling deaths were used to estimate adult
mortality rates using the Gakidou-King method (26). Calendar year adjustment factors
for under-reporting in the Iraq Body Count data ranged from 3.3 (2003) and 3.4 (2004)
to 2.3 (2006) and 2.2 (2007). An average adjustment factor of 2.17 was applied to Iraq
Body Count data for more recent years to derive a time series of estimated total
conflict deaths in Iraq.
Occupied Palestinian Territories. Estimates of Israeli and Palestinian deaths were derived from statistics
published by the Office for the Coordination of Humanitarian Affairs (OCHA) Occupied Palestinian Territory (OPT) (27) and the The Israeli Center for Human Rights
in the Occupied Territories (28).
Syria
For Syria, excess mortality in 2011 and 2012 due to the conflict was taken into account
based on UN estimates of overall conflict deaths by month and age distribution of deaths
(29, 30), as well as estimates by various human rights organizations (31, 32).
US, UK and the coalition of the willing. Military deaths in Afghanistan and Iraq were compiled from
various official sources and summary tables available on websites.
Deaths due to landmines and unexploded ordinance were estimated separately by country (33). Deaths
from terrorist events were separately estimated for many countries without ongoing general conflict
using data from the Global Terrorism Database (90) and Terrorism deaths. Terrorism deaths from this
database were not added to conflict deaths for Iraq, Pakistan, Afghanistan and a number of African
countries to avoid potential double counting.
Legal execution deaths are included in this cause category for GHE2015. Estimated execution deaths
were added for the main countries using capital punishment regularly (China, Iran, Iraq, DPR Korea,
Saudi Arabia, USA and Yemen), from UN Human Rights Reports, with additional information from
Amnesty International reports, Human Rights Watch reports and Wikipedia.
Age-sex distributions for conflict deaths were revised based on available distributions of conflict deaths
by age and sex for specific conflicts (10, 16, 24, 25, 27, 35, 36) and on age-patterns for certain countryperiods with high conflict deaths included in the WPP2015 life tables (37).
The following tables summarizes and compares various time series of conflict deaths estimates.
Table 5.2. Estimated total global injury deaths (thousands) due to conflict: comparison of
various time series and WHO estimates.
Year
GBD 1990
(a)
WHO
2000-2008
World Health Organization
IHME-GBD
2012 (j)
WHO 2013
(i)
PRIO
2015
IHME-GBD
2013 (k)
Current
revision 2016
Page 37
1990
502
-
63
138
94
72
131
2000
656
310 (b)
53
122
90
64
128
95
31
69
19
84
35
57
29
48
64
2013
82
31
157
2014
101
2000
230 (c)
2000
187 (d)
2004
182 (e)
2005
238 (f)
2008
182 (g)
2010
834
26
18
95
42
77
85
194
(a) Estimates and projections by Murray and Lopez (11)
(b) World Health Report 2001 (87) and World report on violence and health (38).
(c) World Health Report 2002 (12)
(d) Revision for Disease Control Priorities Study (13)
(e) Global burden of disease: 2004 update (14)
(f) World Health Statistics 2007 (39)
(g) WHO estimates of causes of death for year 2008 (15)
(h) Sum of main estimates of conflict deaths for state-state, state-nonstate and one-sided conflicts (19-21)
(i) Revised WHO estimates for years 1990-2011 (40).
(j) IHME Global Burden of Disease Study 2010 (41).
(k) IHME Global Burden of Disease Study 2013 (42).
The revised WHO estimates for total conflict deaths (in the final column) are considerably lower than
the previous WHO estimates for years 2000-2008 which used the earlier higher adjustment factor for
under-reporting, which in turn are lower than the previous estimates and projections in the original
Global Burden of Disease (GBD) study (11). The recently estimates for conflict deaths published by IHME
in the GBD 2013 study, shown in the rightmost column, are considerable lower than the revised WHO
estimates. The IHME estimates are also lower than the main estimate from the UCDC databases for the
same year. The IHME methods were based on a regression analysis of available all-cause mortality data
for country-years in which battle deaths were reported in various databases. Lozano et al (41) cite (43)
for more detailed documentation of their methods. The latter publication does not appear to exist.
World Health Organization
Page 38
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
CRED. EM-DAT: The CRED International Disaster Database. Belgium, Université Catholique de Louvain,
2013. Available at http://www.emdat.be/disaster-list (accessed 27 September 2013).
He H, Oguchi T, Zhou R, Zhang J, Qiao S. Damage and seismic intensity of the 1996 Lijiang earthquake,
Vhina: a GIS analysis. Technical report. Tokyo, Center for Spatial Information Science, University of Tokyo,
2001. Available at: http://www.csis.u-tokyo.ac.jp/english/dp/dp.html (accessed 18 January 2008).
Naghii MR. Public health impact and medical consequences of earthquakes. Pan American Journal of
Public Health, 2005, 18:216–221.
Nishikiori N, Abe T, Costa DG, Dharmaratne SD, Kunii O, Moji K. Who died as a result of the tsunami? Risk
factors of mortality among internally displaced persons in Sri Lanka: a retrospective cohort analysis. BMC
Public Health, 2006, 6:73.
Doocy S, Rofi A, Moodie C, Spring E, Bradley S, Burnham G et al. Tsunami mortality in Aceh Province,
Indonesia. Bulletin of the World Health Organization, 2007, 85:273–278.
Heidelberg Institute on International Conflict Research. Conflict barometer. Department of Political
Science, University of Heidelberg, 2012. Available at: http://www.hiik.de/en/konfliktbarometer/.
Project Ploughshares. Armed conflicts report. Waterloo, Canada, Project Ploughshares, 2005. Available at:
http://www.ploughshares.ca/.
Marshall MG, Gurr TR. Peace and conflict 2005: a global survey of armed conflicts, self-determination
movements, and democracy. University of Maryland, Center for International Development and Conflict
Management, 2005.
International Peace Research Institute. UCDP/PRIO Armed Conflict Dataset. Oslo, PRIO, 2009. Available at:
http://www.prio.no/CSCW/Datasets/Armed-Conflict/ (accessed 2 November 2009).
Murray CJ, King G, Lopez AD, Tomijima N, Krug EG. Armed conflict as a public health problem. British
Medical Journal, 2002, 324(7333):346-349.
Murray CJL, Lopez AD. The Global Burden of Disease: a comprehensive assessment of mortality and
disability from diseases, injuries and risk factors in 1990 and projected to 2020. Cambridge, Harvard School
of Public Health, 1996.
World Health Organization. World health report 2002. Reducing risks, promoting healthy life. Geneva,
World Health Organization, 2002.
Lopez, A.D., Mathers, C.D., Ezzati, M., Murray, C.J.L., & Jamison, D.T. Global burden of disease and risk
factors. New York, Oxford University Press, 2006.
World Health Organization. The global burden of disease: 2004 update. Geneva, World Health
Organization, 2008.
World Health Organization. Causes of death 2008: data sources and methods.
http://www.who.int/healthinfo/global_burden_disease/cod_2008_sources_methods.pdf .
Obermeyer Z, Murray CJL, Gakidou E. Fifty years of violent war deaths from Vietnam to Bosnia: analysis of
data from the world health survey programme. British Medical Journal, 2008, 336:1482-6.
Lacina B, Gleditsch NP. Monitoring trends in global combat: a new dataset of battle deaths. Eur J Popul,
2005, 21:145-166.
Garfield, R, Blore J. Direct Conflict Deaths. Unpublished report prepared on behalf of the Collective
Violence Expert Group for the Global Burden of Disease Study, 2009.
Uppsala Conflict Data Program. UCDP Battle-Related Deaths Dataset v.5-2015, 1989-2014. Oslo, Uppsala
University, 2015. Available at: http://www.pcr.uu.se/research/ucdp/datasets/ucdp_battlerelated_deaths_dataset/ (accessed 8 July 2015).
Uppsala Conflict Data Program. UCDP Non-State Conflict Dataset v.5-2015, 1989-2014. Oslo, Uppsala
University, 2015. Available at: http://www.pcr.uu.se/research/ucdp/datasets/ucdp_nonstate_conflict_dataset_/ (accessed 8 July 2015).
Uppsala Conflict Data Program. UCDP One-Sided Violence Dataset v.5-2015, 1989-2014. Oslo, Uppsala
University, 2015. Available at: http://www.pcr.uu.se/research/ucdp/datasets/ucdp_onesided_violence_dataset/ (accessed 8 July 2015).
World Health Organization
Page 39
22. Human Security Report Project. Estimating Battle Deaths: A Challenging Exercise. Vancouver, BC, 17
September 2012. Available at: http://www.hsrgroup.org/docs/Publications/AdditionalPublications/HSRP_Est.Battle_Deaths.pdf (accessed 1 December 2015).
23. Burnham G, Lafta R, Doocy S, Roberts L. Mortality after the 2003 invasion of Iraq: a cross-sectional cluster
sample survey. Lancet 2006; 368 (9545): 1421-28.
24. Iraq Family Health Survey Study Group. Violence-Related Mortality in Iraq from 2002 to 2006. N Engl J
Med, 2008, NEJMsa0707782.
25. Iraq Body Count. Iraqi deaths from violence 2003–2015. Available at: http://www.iraqbodycount.org/
26. Gakidou E, King G. Death by survey: estimating adult mortality without selection bias from sibling survival
data. Demography 2006; 43: 569-585.
27. Office for the Coordination of Humanitarian Affairs (OCHA) - Occupied Palestinian Territory (OPT).
http://www.ochaopt.org
28. B’Tselem – The Israeli Information Center for Human Rights in the Occupied Territories. Statistics on
injuries and deaths suffered by both sides in the conflict. http://www.btselem.org/statistics
29. Price M, Klingner and BallPreliminary Statistical Analysis of Documentation of Killings in Syria. UN OHCHR
commissioned report, January 2013. Available
athttp://www.ohchr.org/Documents/Countries/SY/PreliminaryStatAnalysisKillingsInSyria.pdf (accessed 17
January 2014).
30. UN Secretary General Ban Ki-moon. Statement by the Secretary-General on Fulfilling our Collective
Responsibility on Syria. 12 Mar 2015. http://www.un.org/sg/statements/index.asp?nid=8457
31. Syrian Observatory for Human Rights. 7 Feb 2015. http://www.syriahr.com/en/2015/02/about-2-millionskilled-and-wounded-in-47-months-and-it-is-still-not-enough/
32. Wikipedia. Casualties of the Syrian Civil War.
https://en.wikipedia.org/wiki/Casualties_of_the_Syrian_Civil_War (accessed 30 Nov 2015).
33. International Campaign to Ban Landmines. Landmine monitor. Available at http://www.the-monitor.org/
(accessed 30 Aug 2015).
34. National Consortium for the Study of Terrorism and Responses to Terrorism. Global Terrorism Database
http://www.start.umd.edu/gtd/ (accessed 30 Aug 2015).
35. Hoeffler A. Dealing with the consequences of violent conflicts in Africa. Background Paper for the African
Development Bank, 2008. Available at: http://users.ox.ac.uk/~ball0144/consequences.pdf
36. World Health Organization. European Programme for Intervention Epidemiology Training. Retrospective
mortality survey among the internally displaced population, Greater Darfur, Sudan, August 2004. Geneva,
World Health Organization, 2004. Available at: http://www.who.int/disasters/repo/14652.pdf
37. UN Population Division (2015). World Population Prospects - the 2015 revision. New York, United Nations.
38. Krug EG, et al. World Report on violence and health. Geneva: World Health Organization, 2002.
39. World Health Organization. World Health Statistics 2007. Geneva: World Health Organization, 2007.
40. World Health Organization 2013. WHO methods and data sources for global causes of death 2000-2011
(Global Health Estimates Technical Paper WHO/HIS/HSI/GHE/2013.3)
41. Lozano R, et al. Global and regional mortality from 235 causes of death for 20 age groups in 1990 and
2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet, 2012, 380(9859):2095128.
42. Naghavi M, Wang H, Lozano R, et al. Global, regional, and national age–sex specific all-cause and causespecific mortality for 240 causes of death, 1990–2013: a systematic analysis for the Global Burden of
Disease Study 2013. Lancet, 2015, 385(9963): 117-171.
43. Murray C, Lopez AD, Wang H. Mortality estimation for national populations: methods and applications.
Seattle, University of Washington Press, 2012.
World Health Organization
Page 40
Annex C: Estimated completeness of death registration data
Annex Table C: Estimated completeness of death registration data, by country and year, 1985-2015.
Albania
Argentina
Armenia
=--=-===="---=--
1.0 1 .0 -
1.0 -
0.8 -
0.8 -
0.8 -
0.6 -
0.6 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 -
0.0 -
0.0 2005
1H85
1.0 -
2015
=-:....
0.0 -
198.
1995
2005
2015
1985
19H5
2005
2015
--===---1 0-
Australia
Azerbaijan
Austria
0.8 -
1 .0 - 0.8 -
0.6 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 -
0 0-
I
1oes
I
I
1005
-
1.0 -
2005
I
2015
-=--
-
-
- - 08 0.6 -
oo 1
1085
I
1005
I
2005
I
2015
1.0 -
o o-
1
1095
Belgium
0.6 -
0.6 -
0.6 -
0.0 -
0.6-
0.4 -
0.4 -
0.4 -
0.2 0.0-
0.2 0.0 -
0.2 0.0 -
I
2005
I
2015
I
1965
Brazil
1995
I
2005
2015
1985
Brunei Darussalam
===::::::;:;;::::::::::::::::=­
.c:::
1 .0 -
-:=--:->,.-...-----.:::::::::::::========­
08 -
0.6 -
0.6 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 -
0.0 -
1995
2005
2015
1985
I
2005
I
2015
Bulgaria
0.8 -
1985
1995
1.0-
1.0 - "":>
0.8-
0.0 -
I
2015
Bosnia and Herzegovina
0.8 -
I
1995
I
2005
1.0-
Belarus
1985
I
1905
0.0 1995
2005
2015
1985
1995
2005
2015
Key - IHME - WHO (previous) - WHO (current)
World Health Organization
Page 41
Annex Table C (continued): Estimated completeness of death registration data, by country and year, 1985-2015.
Canada
1.0 -
---====---
1.0 -
Chile
=---=-
Colombia
1.0 -
0.8 -
0.6 -
0.8 - -----
0.6 -
0.6 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 -
0.0 -
0.0 -
1
1985
1.0 0.8 -
-
I
1995
2005
2015
1995
Costa Rica
2005
2015
1985
1.0 -
0.8 -
0.8 -
0.6 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 -
I
1985
0.0 -
I
1995
2005
2015
Czech Republic
1.0--
--
0.0 -
I
1985
1995
2005
2015
- - -
--------=-
0.6 -
0.6 -
0.6 -
0.4 -
0.4 -
04-
0.2 -
0.2 -
0.2 -
0.0 -
1
0.0 -
I
1995
2005
I
2015
1985
1995
2005
Ecuador
I
2015
1985
Egypt
1.0 -
0.
0.8 -
0.8 -
0.6 -
0.6 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 0.0 -
0.2 0.0 -
0.2 0.0 -
I
2005
World Health Organization
I
2015
I
1985
I
1995
2015
ElSalvador
1.0 -
1995
2005
:::;::::>
1.0 - ;:::::---::...-
I
1985
1995
-
.........
'<:::::
8-
2015
2005
0.8 -
0.6 -
1985
1995
Dominican Republic
0.8 -
0.0 -
I
1985
Denmark
1.0 -
2015
2005
Cuba
.-
1.0 -
0.6 -
0.0 -
1995
Croatia
s
--==-
0.0 -
I
1985
I
2005
2015
Key -IHME - WHO (previous) - WHO (current)
-
1985
1995
I
2005
2015
Page 42
Annex Table C (continued): Estimated completeness of death registration data, by country and year, 1985-2015.
Estonia
Finland
France
1.0 -
1 .0 -
1.0 -
0.8 -
0.6 -
0.8 -
0.6-
0.6 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 -
0.0 -
0.0 -
I
0.0 -
I
1985
1995
2005
I
2015
1985
Georgia
1995
2005
2015
1985
Germany
1o -
0.8 -
0.8 -
0.6 -
0.6 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 -
.,...--...,
="":::::.
<::::::::?'
0.8 -
0.0-
0.0 -
I
1985
0.0 -
I
1995
2005
I
2015
1985
1995
Guatemala
1.0 -
2005
I
2015
1985
Guyana
-::-'"='"
::::;;;;;:::; ;:;= ;;:::::
0.8 -
0.6 -
0.6 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 -
0.0 -
1
2005
I
2015
1985
1995
Iceland
200
198
Ireland
----==:<:!!!!!!!!!:::=-
1.0 0.6 -
0.6 -
0.6 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 -
0.0 -
0.0 -
0.0 -
I
1
1995
2005
I
2015
I
1985
1995
2005
2015
Israel
0.8 -
1985
-
I
2015
1.0 1.0 -
-- - -
0.0 -
I
1995
2015
2005
Hungary
1.0 -
0.8 -
1985
1995
1 .0 -
0.8 -
0.0 -
2015
2005
Greece
1.0 -
1.0 -
1995
========::::::::-__
...,.,
0.8 -
I
1995
I
2005
I
2015
1985
1995
2005
2015
Key - IHME - WHO (previous) - WHO (current)
World Health Organization
Page 43
Annex Table C (continued): Estimated completeness of death registration data, by country and year, 1985-2015.
Italy
Japan
Kazakhstan
1.0 -
1.0 -
0.8 -
0.8 -
0.6 -
0.6 -
0.8 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 -
0.0 -
0.0 -
0.0 -
I
1985
1.0 -
I
1995
2005
I
2015
1985
Kyrgyzstan
1995
2005
2015
1985
0.8 -
0.8 -
0.6 -
0.5 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 -
1.0 0.8 -
I
0.0 -
I
1995
2005
2015
Lu embourg
1.0 -
2005
2015
1.0 -
1.0 -
0.6 -
0.8 -
0.6 -
0.0 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 -
0.0 1995
2005
1985
========:-
1995
200
198
Mexico
10 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 -
I
2015
I
1985
2015
0.6 -
0.0 -
I
2005
0.8 -
0.5 -
2005
1995
__
0.6 -
1995
--
Mongolia
0.8 -
1
-- -
1.0 -
=="""'-:a.- --
0.8 -
0.0 -
2015
I
2015
--
1985
-
I
2015
Mauritius
1.0 -
2005
0.0 -
I
1
1985
1995
Malta
- ------=-""-c'======-
0.8 -
0.0 -
I
1985
Maldives
- -===:::====:_
-
1995
2015
---
....
0.0 -
I
1985
2005
Lithuania
1.0 -
1985
1995
Latva
i
10 -
0.0 -
---.......
0.0 I
1995
I
2005
I
2015
1985
1995
2005
2015
Key -IHME - WHO (previous) - WHO (current)
World Health Organization
Page 44
Annex Table C (continued): Estimated completeness of death registration data, by country and year, 1985-2015.
7"7"'"«"'>0::::.. oo:;:<:;::===-1.0 -
1.0
-
Montenegro
-
- -
-
1.0 -
-------""===-
Netherlands
New Zealand
0 80.6 -
0.6 -
0.6 -
0.6 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 -
0.0-
0.0 -
0.0 -
1
1985
I
1995
2005
I
2015
1985
Nicaragua
1995
2005
2015
1985
Norway
0.8 -
0.8 -
0.8 -
0.6 -
0.6 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 -
-- -
-
0.0 -
1
0.0 -
I
1995
2005
I
2015
1985
Peru
1995
2005
I
2015
1985
Philippines
0.8- --------
::?-:-:::::::==:.. -
0.6 -
0.6 -
0.4 -
0.4 -
0.2-
0.2 -
0.2 -
1
0.0 -
I
1995
2005
0.0 -
I
2015
1985
Portugal
1.0-
--==--=======-
1995
200
I
2015
198
Puerto Rico
1.0 -
0.8 -
0.6 -
0.6 -
0.6-
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2-
0.0 -
1
I
1995
2005
I
2015
I
1985
2005
2015
1.0 -
0.6 -
1985
1995
Republic of Korea
-;;;;=-='"""==--===-
0.8 -
0.0 -
2015
0.8 -
0.4 -
1985
2005
1.0 -
0.6 -
0.0 -
1995
Poland
1 .0 - ----------
0.6 -
2015
Panama
1.0 - -
1985
2005
1.0 -
1.0 - -----
0.0 -
1995
0.0 I
1995
I
2005
I
2015
1985
1995
2005
2015
Key -IHME - WHO (previous) - WHO (current)
World Health Organization
Page 45
Annex Table C (continued): Estimated completeness of death registration data, by country and year, 1985-2015.
Republic of Moldova
Romania
Russian Federation
1.0 -
1.0 -
0.8 -
0.8 -
0.8 -
0.6 -
0.6 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 -
0.0 -
0.0 -
1.0 -
...............
I
1985
I
1995
2005
2015
1985
Serbia
0.0 -
I
1995
2005
2015
1985
Slovakia
1.0 -
08 -
0.6 -
0.8 -
0.6 -
0.6 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 -
0.0-
I
1985
0.0 -
I
1995
2005
2015
1985
Spain
1.0 -
----------=-
0.8 -
0.0 -
I
1995
2005
2015
1.0 -
08 -
0.8 -
0.4 -
0.2 -
0.2 -
0.2 -
0.0 1995
2005
1985
- - --
.._
Switzerland
1
1995
2005
I
2015
-
____
1.0 .0 - ---
1985
Tajikislan
0.6 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 0.0 -
0.0 -
I
1995
2005
I
2015
I
1985
2015
0.8 -
0.6 -
1
2005
The fonner Yugoslav Republic of Macedona
0.6 -
0.0 -
1995
1.0 -
0.8 -
1985
--==------
I
2015
2015
0.0 -
I
1
2005
0.6 -
0.4 -
1985
1995
Sweden
0.4 -
0.0 -
2015
I
1985
Suriname
0.6 -
0.6 -
2005
Slovenia
1.0 -
-:::-
1.0 -
1995
I
1995
I
2005
I
2015
1985
1995
2005
2015
Key -IHME - WHO (previous) - WHO (current)
World Health Organization
Page 46
Annex Table C (continued): Estimated completeness of death registration data, by country and year, 1985-2015.
Turkmenistan
1.0 0.8 -
Ukraine
==--
1.0 -
United Kingdom
-===----
1.0 -
0.8 -
0.8 -
0.6 -
0.6 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 -
0.0 -
0.0 -
I
0.0 -
I
1985
1995
2005
I
2015
1985
United States of America
1995
2005
2015
1985
Uruguay
1.0 -
1.0 -
0.8 -
0.8 -
0.8 -
0.6 -
0.6 -
0.6 -
0.4 -
0.4 -
0.4 -
0.2 -
0.2 -
0.2 -
0.0 -
I
1985
2005
---------
I
2015
1985
2015
0.0 -
I
1995
2005
Uzbekistan
1.0 -
0.0 -
1995
1995
2005
I
2015
1985
1995
2005
2015
Venezuela (Bolivarian Republic of)
1.0 - ---------0.6 - 0.6 0.4 0.2 0.0 -
1
1985
I
1995
2005
2015
Key - IHME - WHO (previous) - WHO (current)
World Health Organization
Page 47
Annex D: Estimated completeness of death registration data for most recent year
Country
Year
Albania
2009
Argentina
Armenia
Completeness (%)
Country
Year
Completeness (%)
76.1
Kyrgyzstan
2013
95.8
2013
98.7
Latvia
2012
97.5
2012
101.1
Lithuania
2013
88.2
Australia
2012
97.7
Luxembourg
2013
91.5
Austria
2014
98.8
Maldives
2011
99.3
Azerbaijan
2011
95.9
Malta
2014
98.9
Bahamas
2012
93.5
Mauritius
2014
97.4
Barbados
2012
75.7
Mexico
2013
100.8
Belarus
2012
97.0
Mongolia
2010
94.5
Belgium
2012
98.5
Montenegro
2010
92.0
Belize
2013
80.2
Netherlands
2013
99.1
Bosnia and Herzegovina
2011
94.7
New Zealand
2011
97.0
Brazil
2013
100.0
Nicaragua
2013
72.0
Brunei Darussalam
2013
100.0
Norway
2013
96.2
Bulgaria
2012
97.8
Panama
2013
93.2
Canada
2011
96.1
Peru
2013
61.7
Chile
2013
100.0
Colombia
2012
76.2
Philippines
2009
86.0
Poland
2013
100.0
Costa Rica
2013
87.2
Portugal
2013
97.7
Croatia
2013
99.1
Puerto Rico
2013
99.8
Cuba
2013
100.0
Republic of Korea
2013
96.9
Czech Republic
2013
100.0
Republic of Moldova
2013
85.7
Denmark
2012
95.3
Romania
2012
99.1
Dominican Republic
2012
54.8
Russian Federation
2011
95.6
Ecuador
2013
83.0
Saint Lucia
2012
81.3
Egypt
2013
95.4
Serbia
2013
89.7
El Salvador
2012
83.6
Slovakia
2014
96.8
Estonia
2012
97.7
Slovenia
2010
98.7
Finland
2013
97.8
Spain
2013
95.4
France
2012
99.0
Suriname
2012
77.6
Georgia
2014
100.0
Sweden
2014
98.8
Germany
2013
100.0
Switzerland
2012
98.5
Greece
2012
98.4
Tajikistan
2005
82.8
Guatemala
2013
92.0
TFYR Macedonia
2010
101.1
Guyana
2011
93.3
Turkmenistan
2013
76.4
Hungary
2013
97.6
Ukraine
2012
95.3
Iceland
2012
97.8
United Kingdom
2013
98.5
Ireland
2012
98.4
United States of America
2013
97.7
Israel
2014
99.5
Uruguay
2013
101.0
Italy
2012
100.0
Uzbekistan
2005
89.5
Japan
2013
99.9
2012
89.4
Kazakhstan
2012
95.0
Venezuela
(Bolivarian Republic of)
World Health Organization
Page 48
Annex E: Comparison of 45q15 estimatesWPP2015 and GHE2016
Annex Table E: GH£2016 and WPP2015 estimates of 45q15, by country, sex and year, 1985-2015.
Afghanistan
.
450-
Albania
1-emale
Male
1-emale
Male
-
: =·-. ........•.
. ·.
300
••••
:250-
•••
100-
•••
••
. ••
•••
50 -
I
I
I
201!1196!1
I
I
1 995
, ._
Nale
19&
Female
1990
""'
2015
1905
2C05
Algeria
:::
Angola
fOO -
:,,._= ·········.. ·····.....
100-
-
1085
Fen1ale
Male
"'"
1005
2.)151985
·······--
::!00-
2005
2015
'''"
••. •• •••
,..
,.,
2005
I
0 .
2015
I
2COS
Argentina
Ant1gua and Barbuda
Male
Female
Male
160 - ·--
200-
20 0-
··-·--
:=:
140-
.•
Femael
-
•.
0 0 00
120100-
• ••
I
1985
I
1990
2005
I
20151985
World Health Organization
Armenia
'9:)5
I
2005
···...
100-
I
2015
I
1005
1995
I
2C05
I
20151385
I
I
I
1995
2005
2015
M
ale
Page 49
1-emale
A
u
s
t
r
a
l
i
a
140-
k
M
a
l
e
1
e
m
a
l
e
120-
·.• •
100-
I
I
I
19&5
1990
2005
I I
I
I
2015 19&5
" 990
2000
I
2015
••
:=
40
-I
199!1
1905
2C05
Austria
1995
2005
2015
AzerbaiJan
260_-: •
150- '
Female
Male
100 -
00-
·
1SO-
50-
I
1
085
201!1 19&5
---------
1
I
I
'0;15
2005
2005
20151985
Female
•
100-
•
'''"
series - WHO • • ·· WHO. HIV-and shock
World Health Organization
Male
•
I
2015
1005
•
1095
2005
4015185
1QQS
2005
2015
+ WPP
Page 50
Annex Table E (continued): GH£2016 and WPP2015 estimates of 45q15, by country, sex and year, 1985-2015.
Bahrain
Bahamas
Male
Ma'e
Female
female
22 -
"12"0--
·· ...
100-
eoeo 2005
1005
2015 Ui8!i
2005
1005
2015
1t5
2005
201 51065
Bangladesh
11105
--
: JL
Ma'a
1>0-
Female
·········· ····· .
100-
.. .
1- ,
199)
200:>
201:, IS8t
2J15
Barbados
Mako
1<:;8:>
2005
199:>
lOOt
·..
...
2l1:>
201!1
an1s
Belgium
Female
Ma'e
::= .
Female
10 0 -
H0 -
120 -
2-
100-
2>l-
eo-
100 -
t<J-
1>l1085
1005
2005
20151G85
2005
1005
2016
I GQS
1 E5
20151985
2005
Female
2)15
Female
Ma'a
·····.
·.. ···
2 -
2005
Benin
Bel1 ze
Mako
250-
1.. 5
350 -
..·
300-
,,.,_
151)-
..../
..
•
..........
.
.
··...
_
1>l1
,
"""
lU1t>1 :>
'""
111\/o
Bhutan
Bolivia (Piurinational State of )
Male
Female
::=
Ma'e
Female
35<)-
D-
•
2J(j 2
•
-
.
I
I
1085
1905
I
2006
I
I
2015 1G86
I
1006
I
2005
I
2016
1;>t5
series - WHO · · ·· WHO, HIV-and shock
World Health Organization
1G06
2006
20161006
1.. 6
2005
2J15
• WPP
Page 51
Annex Table E (continued): GH£2016 and WPP2015 estimates of 45q15, by country, sex and year, 1985-2015.
Bosnia and Herzegovina
=
Botswana
Female
Male
Ferrale
Male
== _A
400-
200• 0.
300-
•
200-
•.. .• . •• • . • •• • .•
0.
···········.....
100 -
too- I
?(t1'11M5
701!\
160
Brazil
Male
:l>O250 -
Female
• . •..
199S
Brunei Darussalam
'-!ale
un -
. • • • . .•••
200-
1 1'ifi
150-
1001 20-
100-
M-
Ferrale
•
.
-
•
I
I
196
199
199!:i
2000
201:i
196
I
199:i
200:i
Bulgaria
'-!ale
2«> -
100-
I
I
I
1995
I
1985
2005
••• .
I
I
1995
2005
Burundi
Mala
I
ferrale
::
150-
1965
I
2:005
Burkina Faso
Female
Male
I
201:iH8G
I I
·
......····
I
201 5 H85
I
1995
2005
4015
Cabo Verde
'-!ale
Female
. ::
Ferrale
250 200-
•.•
ISO-
..
..
•,•
• •,
·· ··....
250-
··..
100-
••• • •.
I
"""
2005
10'>5
2C15t08S
10'>5
2015
1!)85
"'"
2005
Cambodia
Mal
:=
-
;5
)
3
<( >)-
250-
200-
201 51)85
1005
2005
2015
Cameroon
Female
'-!ale
Ferrale
400-
.•
••• • -
350- .... .... .
.• . • • .. ' •
•
.· -.
150-
• .. ••
,.,
"""
•
. .• •
•••
...
2:,0-
lU1
series -WHO ·· ·· WHO, HIV-and shock
World Health Organization
·
lOO-
19\1)
:ltl lHIS
,,.,
:.:ot
• WPP
Page 52
Annex Table E (continued): GH£2016 and WPP2015 estimates of 45q15, by country, sex and year, 1985-2015.
. ,. , ,
Cenlral African Repul>ic
Canada
Female
M
Female
Male
='"='- /.
ale
120-
\
.... . .....
350-
•• .
•• .
·.
100 -
• •.••• •
:=
300-
1965
zco
""'
I
201:5191!!5
I
I
I
1990
2005
201!5
• • . ••••• . . • ••••••
250- 1
199S
""'
199!5
2005
Chad
2005
201
Ch1le
Male
female
Male
Female
2<0 150-
-.
•
...
350-
..
100 -
300-
• ..
50-
I
1985
1995
2COS
201519€5
199S
2005
2015
199S
1985
2005
i!:01S 198S
Female
Male
''"'
Female
::=
=:· · .
140-
1:0:01CO-
., -
100 -
80-
60-
19815
1...
2011519E5
1...
2006
2015
1
1086
2006
"'"
Color11bia
Comoms
Fcmab
Mule
200-
2005
China: Province of Taiwan only
China
Male
250 -
1995
.
Mole
Fcmolc
.
.. ••
·._..·....
150-
­
·
100 1 .5
1005
200S
201519ES
..
,.
2005
2015
..
,,
1085
zo1stgss
2005
Congo
Male
19{15
2005
2015
Costa Rica
Femal9
Malo
140 -
6007
1:::0-
"1,., A
A
Female
'
...- "'f"#'
500-
::= · ··-·-"· ...
200-
lCO80..•
. •••
I
?01!iiNF5
series
World Health Organization
""'
?015
- WHO ·· · · WHO, HIV-and shock
'"'"
• WPP
I
I
I
?00
Page 53
Annex Table E (continued): GH£2016 and WPP2015 estimates of 45q15, by country, sex and year, 1985-2015.
C te d'lvoire
Croatia
Female
Male
Female
Male
2004 -
400350- . ... .
.2CO-
.
/ • .··"· ··.
• • • • •• .. .... . ..
...
...............
300-
,
150 -
•
1(0-
• • .•..
so - I
1965
zco
""'
201:5191!!5
1990
201!5
""'
'""'
2005
Cyprus
Cuba
Male
:::=
1(0 -
so-
100 -
60-
w1985
female
Male
Female
40 1995
2COS
199S
201519€5
2005
2015
1985
Czech Repubilc
2005
'""'
i!:01S 198S
1995
2005
''"'
Democratic People's Republic of Korea
Male
Female
Male
Female
200 -
150100-
-----------
50- I
19815
1
1...
2011519E5
I
1
I
.
I
2006
2015
Oen1oaatic Republic of the Congo
Denmar k
Fcmab
Mule
Mole
Fcmolc
1
W-HO
400-
---
350- ···········...
300-
••• .
250-
1 .5
I:iO-
• ...••
• •• ••
1005
200S
• •••••••
201519ES
..
,.
10060-
••
00 2005
2015
1985
..
,,
Djibouti
Male
3002:i<l-
..•
Malo
• ••••
·••
• • • .• ·• •
•...
·.
250-
... ·. .
2(0 -
:...•':
2015
.
:.
10.
1(0-
1
'"'"
?015
1QR!l
series - WHO ·· · · WHO, HIV-and shock
World Health Organization
2005
Female
150 -
200-
?01!iiNF5
19{15
Dominican Republic
Femal9
...
350-
zo1stgss
2005
'"'"
"'
/
..
·..···.
?00
• WPP
Page 54
Annex Table E (continued): GH£2016 and WPP2015 estimates of 45q15, by country, sex and year, 1985-2015.
Egypt
Ecuador
=-.
Female
Male
Female
Male
25()- , · - 160 -
100-
,.
I
2011911!J
""'"
"""
I
I
I
201l!:ftl!l
"'"
El Salvador
Equalorial Gu1nea
Male
400400300 -
·--
·.
350-
.•• • •
300-
•• ••.•
.
.
• • • ••
•
200?SO100-
I
'9 5
I
1995
I
2005
20151985
1995
I
2(05
2015
19!15
1995
2005
20151985
Eritrea
2005
015
Estona
Female
Male
Female
Malo
=A=
400-
:m-
:= :- "
JOO200
19f5
.
-
-1
I
I
200-
·••
•,
100-
..
I
., _,
Mole
Female
Male
2015193.5
·QM
1 5
2005
20151985
1995
2COS
2015
19!15
"'"
2005
015
2005
1995
Fiji
Female
Ethiopia
260-
..,._ .. .... •
300-
•••
.
.
300-
.• .
• •• •
25<)-
200-
• • • ••
•••
•
• •• • • •
200-
• • • '.
1. .5
2005
2()151085
1"'5
2COS
20Hi
15()1085
1.. 5
2005
Finland
Male
World Health Organization
2015101)5
100-
Female
200
Frarc:e
M•le
·--
Page 55
Female
50 -
50-
I
1995
2005
20151985
1995
2COS
20Hi
1965
series -WHO ····WHO, HIV-and shock
World Health Organization
1995
2005
20151985
19 5
2005
01·!
• WPP
Page 56
Annex Table E (continued): GH£2016 and WPP2015 estimates of 45q15, by country, sex and year, 1985-2015.
Gabon
Gambia
Male
f
::
i
....
'"'-
Ma'e
Female
female
3>0-
300 -
--
250-
·.
· - ..
2002005
1005
2015 Ui8!i
2005
1005
1t5
2015
2005
Georgia
201 51065
11105
2005
2J15
Genmany
Ma'a
Mako
2>0-
Female
16 0 -
uo -
2 -
·..
120-
•
100-
eo1J<l -
eo I
199)
200:>
I
I
I
""'
201:,0158:,
2l1:>
Greece
Ghana
Female
3J<J25<-
· -.. .. . ... _ _ _
.•
Ma'e
."'
0
··-···...
2J<l­
::=
120 -
Female
.. .
100-
eo60 40- I
I
1005
1085
2005
2015 1G85
1005
2005
1E5
I
I GQS
2005
Grenada
201 51985
I
I
I
1.. 5
2005
2)1 5
Guatemala
Mako
Female
Ma'a
250-
::
Female
3002 -
-
15<)-
..
,,
..
,,
"""
·.
· . ·...
1&0-
lU1t>1 :>
'""
Guinea-Bissau
Guinea
Male
Female
Ma'e
35<)- \
3J<l-
V·- "·...··
Female
3>0-
·..··
.···.
0 -
······. . .
300-
-: .. ....
•
25
2 1085
---· ·
1005
2006
2015 1G86
1006
·.
2005
.· .·.
2016
200-
1;>t5
series - WHO · · ·· WHO, HIV-and shock
World Health Organization
1G06
2006
201 61006
1.. 6
2005
2J15
• WPP
Page 57
Annex Table E (continued): GH£2016 and WPP2015 estimates of 45q15, by country, sex and year, 1985-2015.
30
Guyana
. ..
Ha1ti
700- J
.
Female
Male
2:>0-
.•
Female
Male
:=
• ··
400200-
300-
·g:s
• • . .. • • •
200-
tso- I
""'"
"""
2011911!J
···········
•
·..• • • .•. .• . • . -
11 9tl!l
201
"'"
Honduras
Hungary
Male
:=-
300-
7 -
:.=_
200-
A
100- I
100-
I
150-
·...........
160-
'9 5
•·• • • · •• ··•
1995
2005
·
20151985
I
1995
I
2(05
2015
Iceland
19!15
1995
2005
20151985
19f5
2005
015
India
=·. .
20 0-
••
100oo-
Female
Male
Female
Malo
60-
•.
150-
40 -
I
1995
I
20151985
2005
1995
I
2015
2COS
19!15
1995
2005
Indonesia
Mole
2015193.5
"'"
2005
015
Iran ! slamic Republc of)
Female
Male
400-
Female
•
?40-
220-
JOO •
0
200-
• •
••
180160-
1. .5
200S
2()151085
10<:S
2COS
20Hi
50-
Iraq
JOO- \l}V
Male
World Health Organization
2
200-
•••. • .
Page 58
1SO-
Ire and
• '·• • • •• •
Female
Female
M•le
100- I
::=
100120-
80-
60-
·ga
1995
200S
20151985
series
World Health Organization
199S
2COS
20Hi
1965
-WHO · · · ·WHO, HIV-and shock
1995
+
2005
20151985
19 5
2005
01· !
WPP
Page 59
Annex Table E (continued): GH£2016 and WPP2015 estimates of 45q15, by country, sex and year, 1985-2015.
140-
Israel
Female
Male
::=
:=
Italy
120 -
·
·
.
•. • •• .
:=
·
I
""'"
•. ••• .
100 -
40-
Female
Male
•, '• . -----
40-
I
I
·
I
I
"""
2011911!J
I
"'"
I
I
I
201l!:ftl!l
Jamaica
Japan
Male
=-
120 100-
1 -
"'-
••
.
$0100-
40-
I
'9 5
I
1995
2005
I
20151985
1995
2(05
I
2015
19!15
I
1995
2005
Jordan
19f5
2005
015
Kazakhstan
Female
Male
20151985
Female
Malo
100-
19!15
I
1995
2005
Kenya
I
2015193.5
I
"'"
I
2005
I
015
Kiribati
35
< 00Moe
F emale
400- · ._
/\
..
A
300-
250- . . . . • . . • • . • • . • . . • • . • •
Female
Male
•
200 -
. ..••• •• •••••.•• • •
tso- I
1. .5
2005
6L
0-
2()151085
1"'5
2COS
20Hi
1085
1.. 5
Kuwait
2005
2015101)5
200
Kyrgyzstan
Female
Male
M•le
Female
=-
500-
..,._
100 -
.
300 -Health Organization
World
Page 60
300 2
1995
2005
20151985 191l5
2COS
-
150100-
20Hi
1965
1995
series
World Health Organization
2005
20151985
19 5
2005
-WHO · · · ·WHO, HIV-and shock
01· !
+
WPP
Page 61
Annex Table E (continued): GH£2016 and WPP2015 estimates of 45q15, by country, sex and year, 1985-2015.
Lao People's Democratic Republic
1{)85
Latvia
Female
Male
1005
1006
20151085
"'''
Female
Male
2005
1!)86
:101f
20)5
100E
:=
:=
Female
...
0
-
so -
I
I
1985
1995
I
2005
I
·.
4'X-
I
3ClC-
• •
I
I
1995
2005
• •. , ., ,
20C-
I
2015
I
I
1985
1995
········
I
20)5
2o1s1ess
Female
•• • ••
I
1995
2005
2015
Libya
liberia
25<-
:1016
...
-.
20151985
::= _ j A,
::=.V
- \--J .
2005
Female
Male
>X-
1:>(.)-
1096
Lesotho
Lebanon
Male
2)()-
20161!:85
Female
Male
:::
BC-
A
q-!:':':(..-·-A
• . ...·• ·.-.
•.
.
10C-
. •
14C-
12C-
2'JfJ19&5
199S
Z005
20151985
1995
2005
01!
19&5
1995
20)5
Li: uania
1995
2005
:1015
Luxembourg
rem!.le
Male
2015tE85
remele
Male
35<JJ<l-
25<2 J<l-
15< 1J<l-
50-
1""
Mal91wi
Madagascar
Female
Male
Female
3>0 -
3-
·•
,._
19&:5
.
,.,
-
201:51985
.......
,..,
•
..
,,
2000
2005
19&5
201:5H:85
2)()- I
21JC-
·········
'
series
World Health Organization
WHO · · ·· WHO, HIV-and shock
• WPP
Page 62
Annex Table E (continued): GH£2016 and WPP2015 estimates of 45q15, by country, sex and year, 1985-2015.
Madives
Ma'aysia
Male
2)()-
ale
Fem.:!le
...
· ·. .
15< -
Female
300-
:
,,.,_ :=
11)0 -
••
!:0-
I
I
005
1085
1095
2005
2015
1085
:i005
Mah
20151,8$
1095
2Co05
2015
Malta
Femal
=
::
::=
25< -
-10-
,,.,
199
2005
,,..,
'''"
Mauritania
Mauritius
Male
::=
?4r• -
N'ale
Female
200 - •.
·.-..
22C'-
Female
'• •
?00-
. ..
100-
'"'-
13£1-1
1085
·
..
,
;;oos
1095
...
2005
100 4:015
1015
4005
Mexico
1985
1!195
4005
2015Hi85
1.. 5
2(()5
2015
Micronesia (Federated States of)
,,..,
2005
2015
1985
2005
Mongolia
20151 85
1995
2015
Montenegro
femele
Male
ale
::= '-.
::=
140-
::=
20151,85
Female
- -
120 100-
1SC-
1085
_./
1...
:mos
I
series
World Health Organization
•o-
-
I
I
I
2005
:'t:015
1085
-WHO · · ··WHO, HIV-and shock
1005
:mos
2015 Ui85
1.. 5
2(()5
2015
• WPP
Page 63
Annex Table E (continued): GH£2016 and WPP2015 estimates of 45q15, by country, sex and year, 1985-2015.
Mozambique
Morocco
Female
Male
Female
Male
-·
400-
••
35()300260 2011911!J
""'"
=..
"""
201l!:ftl!l
"'"
Myanmar
Namibia
Male
== -./\.'v. J\
•
L·
=200 -
• • • • ..
•.
··...
I
'9 5
I
1995
2005
200 -
. ·•·• • • ··• ·.•• •. .
I
20151985
1995
---- . ..
•
·····
I
2(05
2015
19!15
1995
2005
Nepal
20151985
19f5
2005
015
Netherlerds
Male
'
150 -
I
1995
2005
I
20151985
1995
I
2015
2COS
19!15
1995
2005
Female
Male
=
:= ' ; ·. ·.- -..
::=
:=
120-
100-
60-
I
15
I
2005
I
I
20151085
I
I
1Q0S
2COS
I
20Hi
,..,
2005
015
Female
·- ;"":"""
2005
1...
Niger
2015101)5
200
Nigeria
Female
Male
=· ·· ·. . . ··· . .
"'"
Nicaragua
New Zealand
Mole
I
·o E-
2015193.5
Female
M•le
: )¥;
·..
·..
_
200 -
•••
1995
2005
20151985
1995
series
World Health Organization
2COS
20Hi
1965
1995
-WHO · · · ·WHO, HIV-and shock
WPP
2005
20151985
19 5
2005
01· !
+
Page 64
Annex Table E (continued): GH£2016 and WPP2015 estimates of 45q15, by country, sex and year, 1985-2015.
Occupted Palestine Territories
Norway
Female
Male
Female
Male
1200 100 80 60 -
zco
""'
201:5191!!5
1990
201!5
""'
199!5
2005
'"'"
Oman
Male
:=:
201
Pakistan
Female
Male
Female
Male
female
Papua New Guinea
Panama
Male
2005
.A
•.teo-
Female
o-
190140-
•.
····..• .
>«>-
-
·,··
120 -
· ····
.
100 2!0-
so19815
1...
2011519E5
1...
2006
. • •• •
....
2015
2006
"'"
PaJC:tguay
-
150 -
Mole
::-
200 1SO -
• •.
Fcmolc
..·.
·. :
1&>-
•••:
•••••••• ••••• •• •••
·. ···..
1(0 -
120- I
1 .5
,..,
200S
201519ES
. .,
2005
2015
1085
• ••
...,
2005
Philippines
Male
zo1stgss
19{15
• •
2005
0
2015
Poland
Femal9
300-
••
Peru
Fcmab
Mule
2016 ISS'
0
2005
1995
Malo
Female
2!0-
2SO -
zooIW-
200-
1(0-
1"' -
----------1
'"'"
series
World Health Organization
- WHO ·· · · WHO, HIV-and shock
I
I
I
?00
• WPP
Page 65
Annex Table E (continued): GH£2016 and WPP2015 estimates of 45q15, by country, sex and year, 1985-2015.
Portugal
Puerto Rico
Female
Male
Female
Male
25<>200 150100-
00-
I
1
·g:s
""'"
11 9tl!l
201
"'"
2011911!J
I
I
I
Republic of Korea
Qatar
?5<1-·
20015<>100-
!0-
I
I
'9 5
1995
I
2005
20151985
1995
2(05
I
2015
19!15
1995
Republic of Moldova
::,
•
.,
I
I
19f5
2005
I
015
Female
Malo
25<>-
250 -
20151985
Romania
Female
Male
I
2005
·; \..
. /\._
200 -.......
200-
·
15<>-
150100 -
100-
I
1995
I
20151985
2005
11X>5
2COS
2015
19!15
I
11X>5
2005
400 -
Female
200-
2005
015
Fcmt lc
Male
'
•• • •
"'"
Rwanda
Ru5sian Federa1ion
Mole
2015193.5
1600000-600400-
•
•
+ • ..••.• • •. •. •. • .
200100-
I
·o E-
1. .5
2()151085
200S
101:S
2COS
1-65
20Hi
Saint Lucia
:·=·.
1&0-
..
•
2015 Hi85
1005
2005
Saint Vincent and the Grenadines
Female
Male
2005
1WS
Female
M•le
200-
.
180-
•• • •
• • • • •...• .•
·.I..
0
160 -
••• •
100140-
- --
140-
1£0-
1995
200S
20151985
series
World Health Organization
199S
2COS
20Hi
1965
-WHO · · · ·WHO, HIV-and shock
1995
+
2005
20151985
19 5
2005
01·!
WPP
Page 66
Annex Table E (continued): GH£2016 and WPP2015 estimates of 45q15, by country, sex and year, 1985-2015.
Samoa
Sao Tome and Principe
Male
emale
P..lale
.
::·=. . .
3
0025
220-
·.
200-
200 1
.
100 -
m-
"'"
1995
201.51!ii8S
..
.
,,
'"''
1602015
1 985
...
,
"'"
'''"
2005
2015
Senegal
Saudi Arabia
Female
Male
:=
:101.51985
remate
P..le le
:::
300-
1601JI O-
120-
············· ····
100-
w-
,,.,
I
150- I
I
2:01$ 168
20CO
1990
2 )00
199
2010
1 9M
I
1 990
lCO'
I
201!19M
1 99
2005
2:01'
2005
2015
I
I
Seychelles
Serbia
M•la
200160-
200-
•
150100-
--.-
1001085
1995
2015 HiSS
I
1995
2)05
I
I
2015
1 085
1995
2C05
201!1085
I
Sierra Leone
..
,.
Sln apore
F motle
Male
1"'0 0--
w
o55 0-
•••
...
·....
.
500-
•.
450-
·•
400-
,..,,
120.
·..
"""
.
..
..
.
100 Ill-
· ..
·.
.
0
0
•
..
•
60•
40-
,,.,
lUbllits!l
ll'll)
Slovakia
Slovenia
fl.lale
Female
Male
Femate
::
250- 200 -
150 -
150100100-
soI
1005
1995
20C5
I
2015Hi8S
1995
2)05
l
2015
1 985
series -WHO ··· · WHO, HIV-and shock
World Health Organization
I
""'
2COS
201!1985
l OSS
I
I
2Co05
2015
• WPP
Page 67
Annex Table E (continued): GH£2016 and WPP2015 estimates of 45q15, by country, sex and year, 1985-2015.
Solon-en Islands
Somal1a
Female
Male
Ferrale
Male
:lOO -
45
5
00-
250-
::·············
-
.....
1- 1
?(t1'11M5
701!\
199S
South Africa
South Sudan
..
:: . . .. . . . .
-
Male
::=
········...
Male
Female
200-
Ferrale
• •• • •••. .• •
'• ,
450 -
JOQ-
.
••
2!i0-
l
196
199
199!:i
2000
201:i
196
199:i
Spain
140- ··· ••
120 -
2005
SriLanka
Female
Male
201:iH8G
200:i
Male
• • ••
" ·v.\_
250-
100 -
••.,
200-
so-
ferrale
l A
wo-
·• ·
150-
60100 40-
I
I
I
1965
1995
I
1985
2005
=· ·. .
Sudan
Mala
350-
JOO-
Female
• .. • •••• ••
?50-
·
• •
200••
• •
1065
2005
I
I
201 5 H85
I
I
I
1995
200S
4015
Suriname
Male
·-
······
I
I
1995
Ferrale
0
.
••••
·•
·.
2005
10'>5
2C15t08S
10'>5
2015
1!)85
2005
"'"
1005
2005
2015
Sweden
Swaziland
Mal
20151)85
Female
Male
Ferrale
:::=p
_/'(
::
. ............... ... ............ ::=
World Health Organization
Page 68
I
I
"""
11.1>
lU1
,,.,
series -WHO ·· ·· WHO, HIV-and shock • WPP
World Health Organization
Page 69
Annex Table E (continued): GH£2016 and WPP2015 estimates of 45q15, by country, sex and year, 1985-2015.
•Ol- j
Switzerland
Syrian Arab Republic
Female
1
lJO -
- •• 0-·
•
+
WJ-
soso-
200-
,, 1!)86
Female
Male
,....
·oo 2006
1!)(}6
20161085
'0)5
2015
•.
1(85
"'"
2005
:2011: 1!)85
•
1005
2016
Tl1ailand
Tajikrstan
Mole
Fem3le
..
M3le
::2- J·· .
•
,,_ _
Fell'l31e
2())- ····.••
·····
·s.J -
··.....
150'() -
IJO - I
1935
I
I
1995
2005
I
20151985
I
19;15
I
I
21)()5
2015
I
1 65
2005
1995
1995
2015
1995
2015
Ttmor-Lesta
The former Yugoslav Republic of Macedonia
t-emare
Male
201!: 1985
Male
ISO 140 120-
'
•3o0oJ---
-
2()) -
801905
1995
I
zoo
I
I
I
201:1
2015196:;
""'
2005
1995
Togo
Tonga
Meale
Female
20) - --........__ _
J>O3:>0-
•• •• • • • • • • .
•
'"'1035
2005
A
-.-/.... ······ \_
20151985
19 5
2t)()5
2015
Female
-
...,_
'M-
'41)-
120'0) -
""'
1995
2005
250-
·········
2 -
150-
·...
2000
201!1198!5
series
World Health Organization
20119&5
1995
20)5
I
I
21)15
Tunisia
Trinidad and Tobago
19&
201!: 1985
19 5
2000
::
2015
- WHO · · · ·WHO, HIV-and shock
2005
I
20119&
""'
I
2•)1!1
+ WPP
Page 70
Annex Table E (continued): GH£2016 and WPP2015 estimates of 45q15, by country, sex and year, 1985-2015.
Turkey
Turkmenistan
Fem.:!le
Male
ale
Female
300250£00150-
I
1085
20151,8$
:i005
Uganda
1095
2015
Ukraine
Male
:·-=
2CoOS
Femal
::·="".
30!'1200 -
........
••
..•
200-
3:><- "········• . . ··......
.. •. ... ···• •• • •
1&0 -
,,., _
100-
I
198>
I
1,..,
199
1''"
1911!1
2005
11l<J>
United Arab Emirates
Unrted Kingdom
Male
Female
140 -
uo-
16<-
:::=
.
100-
1X• -
•o-
eo
,.,_
1985
Female
N'ale
--..........
co ;;oos
1005
1995
2005
4:015
.
Z\
-
1015
Male
35(J- . - ..
JJt.l -
•
0 ..... ••
•
180- ·
160-
"'"
• ••
0
•••••
140-
• • • . • ••
.
120-
100-
·.
,,.,_
2(()5
Female
•
4X' -
1.. 5
United States of America
United Republic of Tanzania
45()- ·
20151,85
4005
eo ­
I
1985
1W5
4005
2015Hi85
1""5
2005
2015
19115
2005
Uruguay
Uzbekistan
femele
Male
2:><15(1 -
•
•
·.
ale
.
200-
150-
1J()-
1085
Female
1005
:mos
2015 "85
1095
2005
:'t:015
I
I
I
1085
1005
:mos
2015 Ui85
...
,
(()5
I
2015
•
2
World Health Organization
Page 71
series
World Health Organization
-WHO ····WHO, HIV-and shock
+ WPP
Page 72
Annex Table E (continued): GH£2016 and WPP2015 estimates of 45q15, by country, sex and year, 1985-2015.
Vanuatu
Vanezuela (Boilvarian Republic of)
Male
P,.lale
emale
260200-
::
• • • ••
200-
..
-
1:;() -
100-
.;
·
100-
I
1965
199
201;'i 1585
2:>05
1995
2015
19&5
1995
2(05
Viet Nam
1995
2«lS
2015
Yemen
Male
emale
P,.lale
1-emate
200- "-------.
3001502SO100-
7001965
1005
2XS
20151505
1995
-1.\..
2)05
2015
1905
1995
2COS
Zambia
Pl.lale
Female
800-
. ..... . . .
'•,
200-
I
2XS
2015
Femate
•
•
400-
2001005
2(()5
600 -
ooo-
,.,.,
19SS
ZimbabY/e
Male
..,._
2011905
20151GSS
...
,
·····
I
2)05
2015
1085
series - WHO ··· · WHO, HIV-and shock
World Health Organization
·········
,..,.
•
I
2C()S
········
I
201E 1085
1()!)5
2(<)5
I
201!
WPP
Page 73