Levels and trends in child malnutrition

UNICEF WHO World Bank Group joint child malnutrition estimates

Key findings of the 2015 edition

The data in this 2015 edition supersede all historical data previously published by UNICEF, WHO and World Bank Group.
UNICEF/GHAA2015-01436/Quarmyne

Stunting rates are dropping but 159

million children around the world are
still affected.
There are 41 million overweight
children in the world; about 10 million
more than there were 2 decades ago.

The ultimate goal:

for all children to be

free of malnutrition in
all its forms.

Wasting still threatens the lives of 50

million children across the globe.

Global overview
Stunting
The global trend in
stunting prevalence
and numbers of
children affected is
decreasing...

Between 1990 and 2014, stunting prevalence declined

from 39.6 per cent to 23.8 per cent...

- 96M

and numbers
affected declined
from 255 million to

159 million.

1990
2014

...but not fast enough.

Overweight
The global trend
in overweight
prevalence and
numbers of children
affected is
rising.

Overweight prevalence has gone up slightly between

1990 and 2014, from 4.8 per cent to 6.1 per cent

+ 10M

and numbers
affected have risen
from 31 million to

41 million.

1990
2014

Wasting
In 2014, the global
wasting rate was
7.5 per cent.

50M

Approximately 1 out of every 13 children in the world

was wasted in 2014.
2014
Nearly a third of all wasted children were severely wasted,
with a global prevalence in 2014 of 2.4 per cent.

In 2014, there were 667 million

children under 5 in the world.
An estimated:

Globally, 50 million
children under 5
were wasted, of
which 16 million
were severely
wasted in 2014.

159 million
were stunted
41 million
were overweight

20M

50 million
were wasted
(each pair of children represents 20 million children)

Notes on the updated joint malnutrition estimates

In September 2015, UNICEF, WHO and World Bank Group
released updated joint child malnutrition estimates for
the 1990 to 2014 period, which represent the most recent
global and regional figures after adding 62 new surveys from
57 countries to the joint dataset. This key findings report
summarizes the new numbers, main messages and identifies
some minor changes in methodology.

Additional materials include:

(i) the latest country-level joint
malnutrition dataset; and, (ii)
interactive dashboards, which
allow users to visualize and
export the global and regional
estimates.

UNICEF
<uni.cf/jmedashboard2015>
WHO
<www.who.int/nutgrowthdb/estimates>
World Bank Group
<data.worldbank.org/child-malnutrition>

Regional overview prevalence

Africa has seen slow progress in reducing stunting
Percentage of children under 5 stunted and percentage of children under 5 overweight, by United Nations region, 1990 2014

Asia

70

Africa

Asia*

Latin America
and
Caribbean
Latin
America

Oceania*

Oceania**

and Caribbean

60

per cent

50

47.6

42.3

40

32.0

25.1

30

stunting
overweight
95% confidence interval

38.1

35.9

24.5

20

11.7

8.9

6.6

3.4

2010

2005

2000

1995

1990

2014

2010

2005

2000

7.2
1995

1990

2010

2005

2000

1995

1990

2014

5.7

4.9

2014

2010

2005

2000

1995

5.4

4.0
1990

10

2014

Africa

*Asia (excluding Japan); **Oceania (excluding Australia and New Zealand) 1990 and 1995 estimates had consecutive low population coverage.
Source: UNICEF, WHO, World Bank Group joint malnutrition estimates, 2015 edition.

Unequal progress in
stunting reduction
since 1990

-24%
Africa

...progress among
subregions has
been uneven.

While Asia as a
whole has cut
stunting by almost
half...

-47%
Asia

Eastern
Asia

- 82%

2x
reduction

- 41%

Southern
Asia

In 2014, one subregion was above the public health emergency line for wasting
Percentage of children under 5 wasted, by United Nations subregion, 2014

3.9

1.1

Poor
5 - < 10%
Serious
10 - < 15%
Critical
15%
No data

Public
health
emergency
range

Central
America

3.1
Caribbean
1.4
South
America

2.1

Central
Asia 14.2

7.5

Acceptable
< 5%

Not even one

subregion in
Africa has an
acceptable level
of wasting.

Eastern
4.0
Asia*
Northern Africa
Southern
Western
9.3
9.0
Asia
Asia
6.7
Western
7.9
Southeastern
Africa
Eastern
Asia
Middle
Africa
Africa
5.4

9.0
Oceania

Three subregions
are approaching
the public health
emergency line.

Southern
Africa
*Eastern Asia, excluding Japan
**Oceania, excluding Australia and New Zealand
Source: UNICEF, WHO, World Bank Group joint malnutrition estimates, 2015 edition.

This map is stylized and not to scale. It does not reflect a position by UNICEF,
WHO or World Bank Group on the legal status of any country or territory or
the delimitation of any frontiers.

Forms of malnutrition* highlighted in this key findings report

Stunting refers to a child who is too
short for his/her age. Stunting is the
failure to grow both physically and
cognitively and is the result of chronic or
recurrent malnutrition. Its effects often
last a lifetime.

Wasting refers to a child who is too thin

for his/her height. Wasting is the result of
sudden or acute malnutrition, where the
child is not getting enough calories from
food and faces an immediate risk of death.

*Note it is possible for a child to show combinations of malnutrition, such as be stunted and overweight or stunted and wasted.

Overweight refers to a child who is

too heavy for his/her height. This form
of malnutrition results from expending
too few calories for the amount
consumed, and increases the risk of
noncommunicable disease later in life.

Regional overview numbers affected

In Africa, the number of stunted
children is rising

The number of overweight children is

on the rise in all regions

Number of children under 5 stunted, by United Nations

region, 1990 and 2014

Number of children under 5 overweight, by United

Nations region, 1990 and 2014
+22%

200

1990

2014

1990

increase

15

+23%

100

2014

nearly 3x

+91%

20

-52%

150

25

+67%

-57%

10

+5%

50

190

47

91

58

14

0.5

0.3

16.0 19.6

10.3

3.9

0.03 0.12

*Asia (excluding Japan); **Oceania (excluding Australia and New Zealand) 1990 estimate had consecutive low population coverage.
Source: UNICEF, WHO, World Bank Group joint malnutrition estimates 2015 edition.

In 2014, more than half of all stunted children under 5

lived in Asia and more than one third lived in Africa.

In 2014, almost half of all overweight children under 5

lived in Asia and one quarter lived in Africa.

Asia 57%

Asia 48%

Africal.37%

Three out of five sub-regions in Africa, Eastern

Africa, Middle Africa and Western Africa, have
rising numbers of stunted children under 5.

Africal.25%

The number of overweight children under 5 in

Africa has nearly doubled since 1990.

The majority of children under 5 suffering from wasting live in Asia

(each child silhouette represents 1 million children)

In 2014, almost all wasted

children under 5 lived in Asia
and Africa.

Asia
34.3 M
Africa
13.9 M
Latin America
and Caribbean 0.7 M

Oceania
0.1 M

Southern Asia is home to more than

half of all wasted children under 5
globally.

Asia 68%

Africal.28%

Strengths and weaknesses of malnutrition data

Prevalence estimates
for stunting and
overweight are
relatively robust.
Hence it is possible
to generate reliable
time trends.

Trends are not informative for

wasting and severe wasting
given that these are acute
conditions, which can change
rapidly. Thus only 2014 global
and regional estimates are
presented.

The underlying data for global and regional estimates are from
country-level household surveys. Such country data are collected
infrequently and measure malnutrition at one point in time. This
makes it difficult to capture the rapid fluctuations of wasting and
severe wasting over time. For programme purposes, incidence data
(i.e., the number of new cases that occur during an entire calendar
year) would be ideal, however, these currently do not exist.

Country income groupings overview

Low-income countries have made the least progress towards stunting
reductions since 1990
Percentage of children under 5 stunted and percentage of children under 5 overweight,
by country income classification, 1990 2014

Low-income

70
60

54.6

55.3

Lower-middleincome

Upper-middle-income
stunting
overweight
95% confidence interval

37.6

40

33.5

32.6

Only a 32% decrease in

low-income countries...

30
20

7.5

4.9

Source: UNICEF, WHO, World Bank Group joint malnutrition estimates, 2015 edition.

2010

2005

2000

6.8
1995

1990

2014

2010

2005

2000

1995

6.8

2014

2.7
1990

2014

2010

2005

2000

3.4

3.1
1995

10

1990

per cent

50

Unequal progress in
stunting reduction
since 1990

-32%

-77%

... and a 77% decrease

in upper-middle-income
countries.

Overweight numbers have doubled in lower-middle-income countries since 1990

The number of overweight children in lowermiddle-income countries has more than doubled
since 1990, from 7.5 million to 15.5 million.

1990

2x
as high

2014

Low-income and lower-middle-income countries now account for almost all stunted children
worldwide
The share of all stunted children that live in lowincome and lower-middle-income countries has
shifted from 7 in 10 to 9 in 10 between 1990 and
2014.

1990

2014

Lower-income countries bear a disproportionate share of stunted children relative to the total
population distribution
Less than half of all children under 5 lived in lowermiddle-income countries in 2014, yet these countries
accounted for two thirds of all stunted children globally.
47%
26%

66%

24%

15%

8%

Share of under-5
population in 2014

Share of stunted children

under 5 in 2014

Low-income countries only accounted for 15 per cent

of the global under-5 population in 2014, but nearly one
quarter of all stunted children live in these countries.
In 2014, one quarter of all children under 5 lived in uppermiddle-income countries, yet these countries only
accounted for 8 per cent of all stunted children globally.
Note: the numbers do not add up to 100 per cent;
the residual is for high-income countries.

Notes on methodology
The analysis methods have remained unchanged from the 2012 report, except for some minor refinements detailed below:
1. Year assigned to each survey
When data collection begins in one calendar year and continues into the next, the survey year assigned
is the one in which most of the fieldwork took place. For example, if a survey was conducted between
1 September 2009 and 28 February 2010, the year 2009 would be assigned, since the majority of data
collection took place in that year (i.e., four months in 2009 versus two months in 2010). This method
has been used since the 2013 edition (prior to that, the latter year was used by default e.g., 2010 in
the example above).

FINAL

Spain

Niue

Suriname Zambia

Belgium

The former Yugoslav Republic of Macedonia

Saint Vincent and the Grenadines

Madagascar

Iran (Islamic Republic of)

Brazil

Antigua and Barbuda

Libya

Uzbekistan

China

Senegal
Morocco Saudi Arabia Samoa Holy See Mozambique Mongolia
Slovakia
Malaysia
Bolivia (Plurinational State of) Namibia
Jordan
State of Palestine Uganda Democratic Republic of the Congo
Kuwait United Arab Emirates Nepal Peru
Mauritania
Nauru Saint Lucia Croatia
Mali
Nicaragua Bahrain
Costa Rica
Cuba
Albania
Sri Lanka
Turkmenistan
Montenegro
Saint Kitts and Nevis
Hungary
Portugal
Bulgaria Afghanistan
United Kingdom Togo
Myanmar
Slovenia Republic of Moldova
Tunisia
Pakistan Honduras
Kenya
Zimbabwe Belize Algeria Bosnia and Herzegovina
Finland Tajikistan Sao Tome and Principe Italy
Rwanda
South Sudan Micronesia (Federated States of)
Tonga
Australia
Qatar
Djibouti
Timor-Leste Estonia
Thailand Kazakhstan
Paraguay
Armenia
Guinea Bangladesh
Azerbaijan
Cabo Verde
Cameroon Kiribati Gabon
Kyrgyzstan
Ghana
South Africa
Uruguay Switzerland Sudan Malawi
Palau

United States

Nigeria

India

Mexico

Jamaica

Benin Belarus

Russian Federation

Romania
Cte d'Ivoire

Ireland

Singapore
Burundi Haiti Canada Ethiopia Indonesia
Israel
Austria Guatemala
Guyana
Equatorial Guinea
Iraq Chad
Chile Angola
Trinidad and Tobago Ukraine
Norway
Sweden
Tuvalu

Bahamas Yemen

Monaco

Germany Eritrea

Lithuania

Turkey

Barbados

Marshall Islands

Cyprus

Mauritius

Fiji Dominica

Iceland

Lebanon
Netherlands
Niger

Oman

Botswana
Central African Republic

Cambodia Democratic People's Republic of Korea

Egypt Cook Islands
Brunei Darussalam Poland El Salvador
Greece
Papua New Guinea
Burkina Faso
Grenada
Philippines
Denmark Seychelles San Marino
Viet
Nam
Liberia
Ecuador Vanuatu Luxembourg
Dominican Republic
Guinea-Bissau Lesotho Latvia Colombia
Lao People's Democratic Republic
Somalia

Japan

Bhutan

Argentina

Swaziland

Gambia

Maldives

SEPT

OCT

NOV

DEC

2010
JAN

FEB

2. Final reports only

As of the 2014 edition, the dataset used to generate the global and regional estimates is based only on final survey
results. Preliminary survey results are no longer included in the dataset due to situations where they had been
cancelled or significantly changed before release.

V 0.1

Sierra Leone

2009

Panama

Venezuela (Bolivarian Republic of)

Andorra

Georgia

Serbia

Comoros

Liechtenstein

France Solomon Islands

Czech Republic
Congo Republic of Korea

United Republic of Tanzania

New Zealand

Syrian Arab Republic

3. Updated data sources

i. The updated joint dataset which includes:
778 national surveys (62 new)
data from 150 countries and territories (representing more
than 90 per cent of all children under 5 globally (population
coverage varies by regions and periods)).
ii. The under 5 population estimates
The United Nations World Population Prospects, 2015
Revision, were used as weighting factors for each country
survey to derive the regional and global prevalence estimates
and calculate the numbers affected.
iii. Regional and country income classifications as per
July 2015
countries with data

countries without data

4. Footnotes on population coverage

As in the 2014 edition, a separate exercise was conducted to assess population coverage. This was important in order to alert the reader,
via footnotes, to instances where the data should be interpreted with caution due to low population coverage (defined as less than 50
per cent). A conservative method was applied looking at available data within mutually exclusive five-year periods around the projected
years. Population coverage was calculated as:
the sum of country five-year average populations (for which surveys are available in the dataset)
the total of country five-year average population for all countries in the region

This brochure was prepared by the Data and Analytics Section of the Division of Data, Research and Policy, UNICEF New York, the
Department of Nutrition for Health and Development, WHO Geneva and the Development Data Group, World Bank Group Washington
DC. September 2015.
Email: data@unicef.org
data.unicef.org

Email: nutrition@who.int
www.who.int/nutrition

Email: data@worldbank.org
data.worldbank.org