FAO Statistics Working Paper Series

FAO Statistics Working Paper Series

Issue 21/25

THE RELATIONSHIP BETWEEN FOOD INSECURITY AND DIETARY

OUTCOMES

AN ANALYSIS CONDUCTED WITH NATIONALLY REPRESENTATIVE DATA FROM

KENYA, MEXICO, SAMOA, AND THE SUDAN

FAO Statistics Working Paper Series / 21-25

THE RELATIONSHIP BETWEEN FOOD INSECURITY AND DIETARY

OUTCOMES

AN ANALYSIS CONDUCTED WITH NATIONALLY REPRESENTATIVE DATA

FROM KENYA, MEXICO, SAMOA, AND THE SUDAN

Cristina Alvarez-Sanchez, Ana Moltedo, Nathalie Troubat, Talent Manyani, Firas Yassin,

Anne Kepple and Carlo Cafiero

Food and Agriculture Organization of the United Nations

Rome, 2021

Required citation: Alvarez-Sanchez, C., Moltedo, A., Troubat, N., Manyani, T., Yassin, F., Kepple, A.

and Cafiero, C. 2021. The relationship between food insecurity and dietary outcomes – An analysis

conducted with nationally representative data from Kenya, Mexico, Samoa, and the Sudan. FAO

Statistics Working Paper 21-25. Rome, FAO. https://doi.org/10.4060/cb6217en

The designations employed and the presentation of material in this information product do not imply the expression of any opinion

whatsoever on the part of the Food and Agriculture Organization of the United Nations (FAO) concerning the legal or development

status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. The

mention of specific companies or products of manufacturers, whether or not these have been patented, does not imply that these

have been endorsed or recommended by FAO in preference to others of a similar nature that are not mentioned.

The views expressed in this information product are those of the author(s) and do not necessarily reflect the views or policies of

FAO.

ISBN 978-92-5-134819-2

© FAO, 2021

Some rights reserved. This work is made available under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 IGO

licence (CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-nc-sa/3.0/igo/legalcode/legalcode).

Under the terms of this licence, this work may be copied, redistributed and adapted for non-commercial purposes, provided that the

work is appropriately cited. In any use of this work, there should be no suggestion that FAO endorses any specific organization,

products or services. The use of the FAO logo is not permitted. If the work is adapted, then it must be licensed under the same or

equivalent Creative Commons licence. If a translation of this work is created, it must include the following disclaimer along with the

required citation: “This translation was not created by the Food and Agriculture Organization of the United Nations (FAO). FAO is

not responsible for the content or accuracy of this translation. The original [Language] edition shall be the authoritative edition.”

Disputes arising under the licence that cannot be settled amicably will be resolved by mediation and arbitration as described in

Article 8 of the licence except as otherwise provided herein. The applicable mediation rules will be the mediation rules of the World

Intellectual Property Organization http://www.wipo.int/amc/en/mediation/rules and any arbitration will be conducted in accordance

with the Arbitration Rules of the United Nations Commission on International Trade Law (UNCITRAL).

Third-party materials. Users wishing to reuse material from this work that is attributed to a third party, such as tables, figures or

images, are responsible for determining whether permission is needed for that reuse and for obtaining permission from the copyright

holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with the user.

Sales, rights and licensing. FAO information products are available on the FAO website (www.fao.org/publications) and can be

purchased through [email protected]. Requests for commercial use should be submitted via: www.fao.org/contact-

us/licence-request. Queries regarding rights and licensing should be submitted to: [email protected]

https://doi.org/10.4060/cb6217en

https://creativecommons.org/licenses/by-nc-sa/3.0/igo/legalcode

mailto:[email protected]

http://www.fao.org/contact-us/licence-request

http://www.fao.org/contact-us/licence-request


iii

Abstract

Households and individuals that experience moderate or severe food insecurity may have poorer diets, in

quantity or in quality, than their food secure counterparts. Much of the evidence highlighting the

associations between food insecurity and diet comes from North America. However, far less research has

been conducted on the association of food insecurity, particularly at the moderate level, and dietary

consumption in low- and middle-income countries. This study expands on previous work by considering

cross-country comparable measures of food insecurity that are calibrated against the global Food

Insecurity Experience Scale (FIES). We use three household consumption and expenditure surveys, two

from lower-middle-income countries (Kenya and the Sudan), and one from an upper-middle-income

country (Samoa); as well as an individual dietary intake survey from Mexico (upper-middle-income

country). The four surveys are nationally representative and include the FIES (Kenya, Samoa and the

Sudan) or a similar experience-based food insecurity measure (Mexico). Estimates of the average usual

consumption of foods from ten food groups, dietary energy, and macronutrients are computed by food

insecurity class. Overall, the analysis reveals that people who experience moderate or severe food

insecurity consume less meat and dairy products (in all four countries) and less fruits and vegetables

(Kenya and the Sudan) than those who are food secure or mildly food insecure. Consumption of cereals,

roots, tubers and plantains, and pulses, seeds and nuts either decreases slightly, remains similar, or

increases, resulting in a higher proportional contribution of these food groups to the total diet.

Consequently, the more food insecure people are, the larger the share of staples in their diet. This holds

true even if food insecure people in Kenya and the Sudan reduce their consumption of staples, because

they reduce consumption of other food groups even more.

Keywords

ELCSA, food insecurity, FIES, Food Insecurity Experience Scale, food consumption, diet quality, Kenya,

Mexico, Samoa, Sudan.

v

Contents

Abstract .............................................................................................................................................. iii

Figures ................................................................................................................................................ vi

Tables ................................................................................................................................................. vi

Acknowledgements ........................................................................................................................... vii

Introduction ......................................................................................................................................... 1

Methods .............................................................................................................................................. 3

Datasets .................................................................................................................................................... 3

Food insecurity data analysis .................................................................................................................... 5

Food consumption data analysis .............................................................................................................. 6

Dietary statistics according to food insecurity class ................................................................................. 8

Analysis with HCES food consumption data ........................................................................................... 10

Analysis with individual-level dietary intake data .................................................................................. 10

Results ............................................................................................................................................... 11

Discussion .......................................................................................................................................... 25

Limitations and strengths ....................................................................................................................... 26

Conclusion ......................................................................................................................................... 27

References ......................................................................................................................................... 28

Annex A1: Food Insecurity Experience Scale (FIES) survey module included in the Kenya Integrated

Household Budget Survey, Samoa Household Income and Expenditure Survey, and the Sudan

Consumption Patterns and Nutrition Study ........................................................................................ 31

Annex A2: Latin American and Caribbean Scale (ELCSA) module included in Mexico’s National Health

and Nutrition Survey (ENSANUT) 2012 ............................................................................................... 33

Annex A3: Food group classification used .......................................................................................... 35

vi

Figures

Figure 1. Daily per capita consumption of selected food groups in each food insecurity class. Kenya and the Sudan. ................................................................................................................................................... 12

Figure 2. Daily per capita dietary energy (kcal) at-home (excluding FCAH) and total (including FCAH), by food insecurity class. Kenya and the Sudan. ............................................................................................... 13

Figure 3. Daily per capita carbohydrates (available), fats, protein, and total fibre (grams) by food insecurity class. Kenya and the Sudan. ........................................................................................................................ 15

Figure 4. Proportion of total macronutrients to total dietary energy consumption by food insecurity class. Kenya and the Sudan. ................................................................................................................................. 16

Figure 5. Daily per capita consumption of selected food groups in each food insecurity class. Mexico and Samoa. ........................................................................................................................................................ 19

Figure 6. Daily per capita dietary energy (kcal) at-home (excluding FCAH) for Samoa and total (including FCAH) for Mexico and Samoa, by food insecurity class. ............................................................................. 20

Figure 7. Daily per capita carbohydrates (available), fats, protein, and total fibre (grams) by food insecurity class. Mexico and Samoa. ........................................................................................................................... 22

Figure 8. Proportion of total macronutrients to total dietary energy consumption by food insecurity

class. Mexico and Samoa……………………………………………………………………………………………………………………….23

Tables

Table 1. Survey description........................................................................................................................... 4

Table 2. Descriptive statistics of the analytical samples analysed ............................................................... 5

Table 3. Percentage of the survey foods (in total foods matched with FCTs/FCDBs) by Food Composition

Tables/Databases used for the three HCES. ................................................................................................. 7

Table 4. Dietary statistics produced by data source ..................................................................................... 8

Table 5. Percentage of individuals living in households reporting consumption of food groups during the

reference period, by food insecurity class. Kenya and the Sudan. ............................................................. 17

Table 6. Percentage of individuals reporting (Mexico) or living in households reporting (Samoa)

consumption of food groups during the reference period, by food insecurity class. ................................ 24

vii

Acknowledgements

This working paper is authored by Cristina Alvarez-Sanchez, Ana Moltedo, Nathalie Troubat, Talent

Manyani, Firas Yassin, Anne Kepple and Carlo Cafiero. The authors gratefully acknowledge: Dr. Teresa

Shama and her team, from Mexico’s National Public Health Institute, for sharing processed microdata and

for their expert advice; Mr. Josiah Waithaka Kaara and his team, from the Kenya National Bureau of

Statistics, for giving access to the microdata and their expert advice, and the Poverty Analysis Unit at the

Kenya National Bureau of Statistics for their expert advice on county level statistics; Mr. Hamza Abdalla

Siror Osman, from the Food Security Technical Secretariat at the Ministry of Agriculture and Forests of the

Sudan, for sharing microdata and for his expert advice; and Dr. Aliimuamua Maleafono Taua-T. Faasalaina,

Government Statistician at the Samoa Bureau of Statistics, for providing data and staff needed to perform

the analysis for Samoa and Ms. Edith Faaola, Assistant Chief Executive Officer, of the Samoa Bureau of

Statistics, for assisting with the analysis.

1

Introduction

Worldwide, 26.4 percent, or 2 billion people, are food insecure at moderate or severe levels (FAO et al.,

2019). Food insecurity means not having access to enough nutritionally adequate and safe foods on a

continuous basis for an active and healthy life (FAO et al., 2019). Households and individuals that

experience moderate or severe food insecurity may have difficulties in physically accessing enough safe

and nutritious food on a continuous basis if food markets are far or hard to reach, or if food is not available

– particularly nutritious foods. They may also experience economic constrains such that the quality,

diversity or quantity of the food they are able to purchase is compromised. As a consequence, they may

have poorer diets, in quantity or in quality, than their food secure counterparts.

Studies have found that exposure to food insecurity may contribute to micronutrient deficiencies in

children (Mundo-Rosas et al., 2014), anaemia in women of reproductive age, child stunting and adult

obesity (FAO et al., 2019). In high-income countries, the experience of food insecurity has also been

associated with cardiovascular disease, stroke, diabetes, cancer and depression (Gundersen and Ziliak,

2015; Martin et al., 2016; Maynard et al., 2018)

Much of the evidence highlighting the association between food insecurity and diet comes from North

America, where household food insecurity has been routinely monitored for decades (Nord, 2014). A large

number of studies in adults from the United States of America and Canada, using validated experience-

based food insecurity measures,1 have found that, in general, food insecurity leads to the “substitution”

of higher-quality food (such as fruits, vegetables and dairy) with highly processed energy-dense foods that

are cheaper on a per-calorie basis (Hanson and Connor, 2014; Johnson et al., 2018; Leung et al., 2014).

However, in spite of the substitution effect, in those studies total energy consumption did not vary largely

across levels of severity of food insecurity. Studies have also found evidence of lower fruit consumption

in food insecure children compared to food secure children (Hanson and Connor, 2014).

Far less research has been conducted on the association between food insecurity, particularly at the

moderate level, and dietary intake in low- and middle-income countries, with the exception of Mexico2

and Brazil,3 where household food security has also been monitored for a long time. In Mexico, dietary

diversity has been found to be inversely related to the severity of food insecurity in children, adolescents,

and adults (FAO, 2012; Mundo-Rosas et al., 2014; Vega-Macedo et al., 2014). A study by Rodríguez and

colleagues showed that dietary quality (as assessed with the Healthy Eating Index-2010) among Mexican

children and adolescents deteriorated as food insecurity became more severe, compared to their

counterparts living in food secure households (Rodríguez et al., 2017). In particular, food insecure children

and adolescents ate less fruits, protein foods,4 and dairy. Pinheiro de Toledo-Vianna and colleagues (2012)

found that in the Brazilian State of Paraíba, household food insecurity was positively associated with daily

sugar consumption and inversely associated with daily consumption of fruits, vegetables, dairy and bread.

1 Such as the Household Food Security Survey Module (HFSSM) and, to a lesser extent, its predecessor, the Radimel/Cornell questionnaire. 2 Food insecurity was assessed with the Latin American and Caribbean Food Security Scale (ELCSA - Escala Latinoamerica y Caribeña de Seguridad Alimentaria). 3 Food insecurity was assessed with the Brazilian Food Insecurity Scale (EBIA - Escala Brasileira de Insegurança Alimentar). 4 In the Healthy Eating Index, the “total protein foods” includes animal-source foods, nuts, seeds, soy products (other than beverages), and legumes (beans and peas).

2

In Ethiopia, Bangladesh, and Viet Nam, researchers found that the more food insecure households were,5

the less likely children achieved minimum dietary diversity (Ali et al., 2013).

The FAO Statistics Division has been publishing estimates of the prevalence of food insecurity, based on

the FIES, since 2017 (FAO et al., 2019). The FIES is the first standardized measure of people's direct

experiences of food insecurity, appropriate for application on a global scale. The prevalence of moderate

or severe food insecurity based on the FIES is one of the official Sustainable Development Goals indicators

(2.1.2). No studies in the literature examine the nexus between dietary intake and food insecurity based

on the FIES or a similar experience-based food insecurity measure calibrated to the FIES global reference

scale developed by FAO (FAO, 2016).

The objective of this study was to explore the relationship between the severity of food insecurity, as

measured with the FIES (or an analogous experience-based food insecurity scale calibrated to the global

reference scale), and dietary intake using microdata from four middle-income countries from different

world regions: Kenya, Mexico, Samoa and the Sudan.

Household Consumption and Expenditure Surveys (HCES) typically collect information on food acquired

or consumed at the household level. The three HCES used in this analysis (from Kenya, Samoa and the

Sudan) collected information on food consumed at the household level (i.e. apparently consumed). This

is in contrast to the individual dietary intake survey from Mexico that collected information on individuals’

food intake (i.e. food ingested). However, for simplicity, herein, we use the term “consumption” to refer

to both household and individual-level food consumption/intake data. This is different to the use of the

same term by economists in welfare analyses, who refer to “consumption” as food and non-food

expenditures (World Bank, 2019).

Due to differences in survey design, particularly between the three HCES and the individual-level survey,

and between the food consumption modules in the three HCES, the average food/nutrient consumption

statistics, for a given food insecurity class, should be compared with caution. Comparison should be

focused across food insecurity classes within a given country.

5 Food security was assessed with the Household Food Insecurity Access Scale (HFIAS).

3

Methods

Datasets We used microdata from surveys that included both data collected with an experience-based food

insecurity scale and either household or individual food consumption data.

Kenya Integrated Household Budget Survey (KIHBS) 2015/16. This survey was implemented by the Kenya

National Bureau of Statistics. It was conducted over a 12-month period to obtain data on a range of

socioeconomic indicators used to monitor the implementation of national development initiatives. The

survey is representative at the national and county levels, as well as place of residence (rural and urban).

The analytical sample includes 92 825 individuals in 21 756 households.

Mexico National Health and Nutrition Survey (ENSANUT, by its acronym in Spanish) 2012. The ENSANUT

is a survey designed to assess the health and nutritional status of adults and children in Mexico. It is

implemented by the National Institute of Public Health. Food consumption data was collected using two

rounds of 24-hour recalls: in the first round, 13 percent of the individuals surveyed in the ENSANUT were

interviewed; the second round was administered to a subsample of 8 percent of individuals who had

completed the first round. Data collection was performed between October 2011 and May 2012. The

analytical sample consisted of 9 908 individuals one year of age or older. The data are representative at

the national level, urban-rural location, and four geographic regions.

Samoa Household Income and Expenditure Survey (HIES) 2018. This survey was designed to assess the

hardship and incidence of poverty throughout Samoa, obtain expenditure weights to rebase Samoa’s

Consumer Price Index, and to assist in compiling official estimates of household accounts in the system of

National Accounts. It was implemented in four rounds, from March to November, by the Samoa Bureau

of Statistics. The data are representative at the national level, urban-rural location, and four geographic

regions. The analytic sample covers 20 741 individuals in 2 997 households.

The Sudan Consumption Patterns and Nutrition Study (CPNS) 2018. This survey was the third national

household survey implemented in the Sudan covering 18 Sudanese states. It was conducted in 2018 by

the Food Security Technical Secretariat of the Ministry of Agriculture and Forestry between April and May.

The analytical sample includes 114 039 individuals in 16 311 households.

4

Table 1 presents a summary of key characteristics of the four datasets and Table 2 presents the descriptive

statistics of the analytical samples analysed.

Table 1. Survey description

Characteristics

Kenya

KIHBS 2015/16

Mexico

ENSANUT 2012

Samoa

HIES 2018

Sudan

CPNS 2018

Unit of analysis of

food consumption Household Individual Household Household

Level of assessment of

food security Household Household Household Household

Food module type Consumption

and acquisition Intake

Consumption

and acquisition Consumption

Data processed for

this analysis Consumptiona Intake

Consumption

and acquisitionb Consumptionc

Reference period 7 days

Past 24 hours

(administered

twice)

14 days 7 days

Food data collection

method Recall Recall Diary Recall

Number of food items

included/covered 210 1 190 504 169

Experience-based

food insecurity scale

included

FIES ELCSA FIES FIES

Food consumed away

from home reported

as quantities (e.g.

grams) or monetary

values

Monetary values Quantities Monetary values Monetary values

Source: authors’ preparation, 2021.

Note: ELCSA: Escala Latinoamericana y Caribeña de Seguridad Alimentaria (Latinamerican and Caribbean Food Security Scale);

FIES: Food Insecurity Experience Scale. a Kenya’s HCES collected information on food acquired and the amount of food consumed from the acquisitions. For this analysis,

we processed data on food consumed (from purchases, own production, gifts/food for work, etc.). b Samoa’s HCES collected information on food acquisition from purchases and on food consumption from other sources such as

own production or received in kind. For this analysis, we processed both types of data. c The Sudan’s HCES collected information on food consumption and their related cost. Therefore, for this analysis, we processed

data on food consumed.

5

Table 2. Descriptive statistics of the analytical samples analysed

Kenya

KIHBS 2015/16

Mexico

ENSANUT 2012a

Samoa

HIES 2018

Sudan

CPNS 2018

Sample size households, n 21 756 N/A 2 997 16 311

Sample size individuals, n 92 825 9 908 20 741 114 039

Sex, n (percent)

Male 45 899 (49) 4 792 (48) 10 602 (51) 58 334 (51)

Female 46 926 (51) 5 116 (52) 10 139 (49) 55 705 (49)

Age group, n (percent)

Pre-school children (0–4.9 y) 12 685 (14) 2 066 (21) 2 837 (14) 27 516 (32)

School childrena (5–11.9 y) 19 580 (21) 2 704 (27) 3 687 (18) 23 247 (27)

Adolescents (12–19.9 y) 18 493 (20) 2 021 (20) 3 625 (17) 22 882 (26)

Adults (20 y and over) 42 067 (45) 3 117 (31) 10 592 (51) 40 394 (47)

Source: authors’ own calculations, 2021. With data from the KIHBS 2015, Mexico’s ENSANUT 2012, the Samoa HIES 2018 and the

Sudan CPNS 2018.

Note: N/A: not applicable.

a The ENSANUT 2012 excludes children less than 1 year of age.

Food insecurity data analysis

Analysis of food insecurity data with the Rasch Model

The Rasch model is based on the Item Response Theory, a branch of statistics that permits the

measurement of unobservable traits through analysis of responses to surveys and tests. The Rasch model

provides a theoretical base and a set of statistical tools to assess the suitability of a set of survey questions

(“items”) for constructing a measurement scale and to compare a scale’s performance across different

populations and survey contexts. The analysis involves the following steps (FAO, 2017): (1) Parameter

estimation. Calculation of the severity of food insecurity associated to each survey item and each

respondent.6 (2) Statistical validation. Assessment of the quality of the data collected by testing their

consistency with the assumptions of the Rasch model. This analysis involves the interpretation of several

statistics that reveal: a) items that do not perform well in a given context, b) respondents with highly

erratic response patterns, c) pairs of items that may be redundant, and d) the proportion of total variance

in the population that is accounted for by the measurement model. 3) Calculation of measures of food

insecurity (respondents’ probabilities). For each sampled household (each case in the data), the

probability of the household experiencing food insecurity above a given level of severity was calculated,

based on their responses to the questions.

6 In the four surveys food insecurity was assessed at the household level. Thus, each respondent represents a household.

6

In the Kenya, Samoa and Sudan HCES, food security was assessed at the household level using the FIES

with a recall period of 12 months (the module is presented in Annex 1). Mexico’s ENSANUT 2012 survey

measured household food security using the Latin American and Caribbean Food Security Scale (ELCSA -

Escala Latinoamericana y Caribeña de Seguridad Alimentaria) with a recall period of three months.

The ELCSA has 15 items: the first eight items are common to all households (i.e. respondents), while the

remaining items are asked of households with children. For the parameter estimation and statistical

validation steps, we analysed separately data from households with and without children, that is, 8 and

15 items, respectively (the module is presented in Annex 2). The items in both groups showed sufficient

consistency with the Rasch model assumptions to produce reliable measures of food insecurity in the

country.

In the Sudan, the eight FIES items showed sufficient consistency with the Rasch model assumptions to

produce reliable measures of food insecurity. In the case of Kenya, one item (“you were worried you

would not have enough food to eat”) failed to fit the Rasch model; in the case of Samoa, one item (“you

were unable to eat healthy and nutritious food”) did not pass the Rasch model either.

Equating: calibration of the scales on a common metric

To have standardized food insecurity classes across countries, we needed to calibrate the food insecurity

scales by using the common items between the scales and the FIES global standard, which is a set of item

parameter values based on results from over 140 countries covered by the Gallup World Poll in 2014,

2015 and 2016.

The calibration to the FIES global scale was performed with eight items for the Sudan and seven items for

Kenya and Samoa (i.e. all the items that showed consistency with the Rasch model were common to the

items in the FIES global scale). In Mexico, despite all items showing consistency with the Rasch model, the

calibration (of both groups) to the FIES global scale was performed with six items. The reason is that two

of the ELCSA items (“you were worried that food would run out in your home” and “you or another adult

in your home ate only one time a day or went without eating for a whole day”) did not appear to be

common to the items in the FIES global scale.

From the calibration procedure, applied independently to each country, we obtained two probabilities for

each raw score: the probability of being moderately or severely food insecure (Prob mod+sev), and the

probability of being severely food insecure (Prob sev). Then, for Kenya, Mexico and the Sudan, each

household was classified as food secure or mildly food insecure if Prob mod+sev<0.5, moderately food

insecure if Prob mod+sev>=0.5 and Prob sev <0.5, and severely food insecure if Prob sev>=0.5. In the case

of Samoa, very few households were classified as being severely food insecure; therefore, to have reliable

estimates of food consumption by level of food insecurity, the moderately and severely food insecure

groups were combined (households were classified as moderately or severely food insecure if the Prob

mod+sev>=0.5).

Food consumption data analysis

With HCES data from Kenya, Samoa and the Sudan

Food consumption data from the HCES were prepared following the approach outlined by Moltedo and

colleagues (Moltedo, A. Troubat, N. Lokshin, M. Sajaia, 2014), which involves: standardization of

7

quantities into gram equivalents (when quantities were reported in non-standard units they were

converted into grams using the regional median cost of one gram of product); detection and imputation

of food quantity missing values and outliers; and adjustment of food quantities for non-edible portions.

Food items were matched to available food composition data from several Food Composition Tables (FCT)

and Databases (FCDB) (see Table 3), following FAO and INFOODS food matching guidelines (FAO and

INFOODS, 2012). Matching was conducted considering foods as purchased (i.e. considering, in most cases,

the content of nutrients in the raw form of the food).

Table 3. Percentage of the survey foods (in total foods matched with FCTs/FCDBs) by food composition

tables/databases used for the three HCES

Country Kenya

FCT

United

States of

America

FCDB

West African

FCT

Pacific

Nutrient

Database

Other

FCT/FCDB

Kenya 72 16 5 0 7

Samoa 0 0 0 100 0

Sudan 11 26 62 0 1

Source: authors’ preparation, 2021.

Note: This applies only to foods reported in terms of quantities. Food items with only expenditures attached were not matched

with foods in FCTs/FCDBs. Food Composition Tables and Databases Sources: Kenya (FAO & Government of Kenya, 2018), United

States of America (U.S. Department of Agriculture, 2019), West Africa (FAO, INFOODS, The West African Health Organization and

Bioversity International, 2012), Pacific (Pacific Community, FAO and University of Wollongong, 2020).

Lastly, the at-home and away from home dietary energy and nutrient consumption was estimated. Some

foods consumed away from home (FCAH) only had a monetary value attached (i.e. no quantity), for those

particular food items (2 percent in Kenya, 0.8 percent in Samoa, and 3 percent in the Sudan), the

calculation of amounts of dietary energy and nutrients provided by these foods was performed using

median at-home calorie and nutrient unit values, respectively (Moltedo, Álvarez-Sánchez, Troubat,

Cafiero, et al., 2018). The at-home median values were obtained at the regional income quintile

urban/rural level.

With individual-level data from Mexico

Data on individuals’ food quantities consumed (net edible portions in grams), dietary energy and nutrient

intake derived from the ENSANUT 2012 were prepared at Mexico’s National Public Health Institute and

shared with us. A description of the processing of the dietary data from the ENSANUT 2012 is provided by

López-Olmedo et al. (2016). In contrast to the HCES food consumption data, these data are all based on

quantities (e.g. grams, millilitres) reported.

8

Dietary statistics according to food insecurity class

The dietary statistics produced using each of the data sources are presented in Table 4. All statistics were

produced by food insecurity class.

Table 4. Dietary statistics produced by data source

Dietary statistic

Kenya

KIHBS

2015/16

Mexico

ENSANUT

2012

Samoa

HIES 2018

Sudan

CPNS 2018

Food group consumption

(g/capita/day)a ✓ ✓ ✓ ✓

Percentage of individuals or

individuals living in households with

access to different food groups

✓ ✓ ✓ ✓

Total dietary energy

(kcal/capita/day) consumption ✓b ✓ ✓b ✓b

At-home dietary energy

(kcal/capita/day) consumption ✓ ✓ ✓

Total fats, protein and carbohydrates

(available)c consumption

(g/capita/day)

✓b ✓ ✓b ✓b

Total dietary fibre consumption

(g/capita/day) ✓b ✓ ✓b ✓b

Contribution of total carbohydrates,

fats and protein to dietary energy

(percent)

✓b ✓ ✓b ✓b

Source: authors’ preparation, 2021. a Food group consumption estimates presented for 11 selected food groups. b The energy and nutrient content of FCAH was estimated using at-home median calorie and nutrient, respectively, unit values at

the regional-income quintile-urban/rural level. c Carbohydrates here refers to “available carbohydrates”, calculated as total carbohydrates minus total fibre.

Details about each of the statistics produced are presented below.

Food group consumption

Foods were classified into 19 groups on the basis of their nutritional relevance following the criteria used

in the FAO/WHO Global Individual Food consumption data (GIFT) Tool (FAO and WHO, 2020). Annex A1

presents the food groups and subgroups. For this analysis, we considered 11 out of the 19 groups: cereals

and their products; roots, tubers, plantains and their products; pulses, seeds and nuts and their products;

dairy products; eggs and their products; fish, shellfish and their products; meat and their products;

vegetables and their products; fruits and their products; fats and oils; and sweets and sugars. The average

consumption (g/capita/day) of each food group was estimated for each food insecurity class.

Eight food groups from the GIFT classification were not considered – insects, grubs and their products;

foods for particular nutritional uses; food supplements and similar; food additives; composite dishes;

9

savoury snacks; beverages; and spices and condiments – due to the very low/zero consumption or

negligible contribution to usual energy and nutrient consumption, or because they were not well captured

by HCES.

Percentage of individuals (or individuals living in households) that reported consumption from

different food groups

In the case of the three HCES, we estimated the percentage of individuals living in households that

reported consumption of at least one item from the food group during the reference period. In the case

of Mexico’s individual dietary intake survey, the statistic is based on the number of individuals who

reported consumption of an item, in the food group, during the reference period (previous 24 hours).

Dietary energy consumption

From the at-home consumption

This statistic was only computed from HCES food consumption data. All foods consumed at home were

reported in terms of quantities; therefore, estimates of the at-home dietary energy consumption were

based on the caloric content in foods (calculated using macronutrient values from food composition tables

and databases and applying the Atwater extensive general factor system of coefficients).

Total dietary energy consumption

In the case of the three HCES, the energy provided by FCAH was estimated using median at-home calorie

unit values (see description in section 2.3). This was added to the energy provided by food consumed at

home. In the case of Mexico’s dietary intake survey, the dietary energy contribution from all foods was

calculated from the reported quantities (Lopez-Olmedo et al., 2016).

Carbohydrates, fats, protein, and total fibre consumption

As for total energy, the available carbohydrates (i.e. total carbohydrates - total fibre), fats, protein, and

total fibre consumed away from home were estimated using the corresponding median at-home nutrient

unit value and then added to the at-home nutrient consumption estimate. In the case of Mexico’s dietary

intake survey, the nutrient contribution from all foods was calculated from the reported quantities (Lopez-

Olmedo et al., 2016).

Contribution of total carbohydrates, fats and protein to dietary energy (percent)

First, the total energy contributed by protein and fats was calculated multiplying the average consumption

of protein and fats by the number of kilocalories per gram: 4 for protein and 9 for fat (these are referred

to as Atwater coefficients) (FAO, 2003). Second, the proportion of energy contributed by protein and fats

was calculated; and the proportion of energy contributed by total carbohydrates was obtained by

substraction (100 - proportion of energy contributed by protein and fats).

10

Analysis with HCES food consumption data

The statistics on usual food quantities consumed by food group, as well as those for usual dietary energy

and nutrient consumption – by food insecurity level – were obtained using the ADePT-FSM software

(Moltedo, Álvarez-Sánchez, Troubat and Cafiero, 2018; Moltedo, Troubat, Lokshin and Sajaia, 2014).

Regression analyses, followed by Tukey’s pairwise post-hoc tests (family-wise error rate of 5 percent) in

the case of Kenya and the Sudan, were used to determine whether there was a difference between the

mean of all possible pairs. No pairwise post-hoc comparison tests were conducted for Samoa because

only two groups (food secure and moderately or severely food insecure) were involved.

Analysis with individual-level dietary intake data

Estimates of usual food quantities consumed by food group, and usual dietary energy and nutrient

consumption – by food insecurity level – were produced applying the National Cancer Institute (NCI)

method for usual intake (Tooze et al., 2010), implemented through the Mixtran and Distrib SAS macros

(National Cancer Institute, 2009).

Dietary energy and the selected nutrients were consumed by almost everyone. Thus, we used the 1-part,

amount-only, model. However, all food groups were found to be episodically consumed by individuals in

each of the food insecurity levels. Therefore, we used the correlated model for all food groups.

The Indivint SAS macro (National Cancer Institute, 2009) was used to predict individuals’ consumption.

The results were used in a subsequent regression model followed by Tukey’s pairwise post-hoc tests

(family-wise error rate of 5 percent) to determine whether there was a difference between the mean of

all possible pairs.

11

Results

In Kenya and the Sudan, those who experience moderate food insecurity have a lower food consumption

of all food groups (with the exception of cereals in both countries) and a lower dietary energy intake than

those who are food secure or mildly food insecure (from here on, referred to as “food secure”) (Figure 1

and Figure 2). The diets of people experiencing severe food insecurity contain even lower quantities of

roots, tubers and plantains, dairy, vegetables, fats and oils, and sweets and sugars (Kenya and the Sudan),

and cereals, fruits, eggs and fish (Kenya), compared to those in the moderately food insecure class. In

Kenya, severely food insecure people consume more fish than those who are food secure.

12

Figure 1. Daily per capita consumption of selected food groups in each food insecurity class. Kenya and

the Sudan

Source: authors’ own calculations, 2021. With data from the KIHBS 2015/16 and the Sudan CPNS 2018.

Note: Food consumption estimates shown for selected food groups only. Differences across groups were assessed with regression

analysis followed by Tukey’s pairwise post-hoc tests (family-wise error rate of 5 percent). a Significant difference between the moderately food insecure and the food secure/mildly food insecure groups (p-value < .05). b Significant difference between the severely food insecure and the food secure/mildly food insecure groups (p-value < .05). c Significant difference between the moderately and severely food insecure groups (p-value < .05).

b,c

a,b,c

a,b

a,b,c

a,b,c a,b,ca,b

a,b,c

a,b,c

a,b,ca,b,c

0

50

100

150

200

250

300

350

400

Cerealsand theirproducts

Roots,tubers,

plantainsand theirproducts

Pulses,seeds andnuts and

theirproducts

Dairyproducts

Eggs andtheir

products

Fish,shellfishand theirproducts

Meat andmeat

products

Vegetablesand theirproducts

Fruits andtheir

products

Fats andoils

Sweets andsugars

gram

s/ca

pit

a/d

ay

Kenya

Food secure or mildly food insecure Moderately food insecure Severely food insecure

a,b,c

a,ba,b,c

a,b aa,b

a,b,ca,b

a,b,c

a,b,c

0

50

100

150

200

250

300

350

400


Roots,tubers,



theirproducts

Dairyproducts

Eggs andtheir

products


Meat andmeat

products


Fruits andtheir

products

Fats andoils

Sweets andsugars

gram

s/ca

pit

a/d

ay

Sudan


13

Figure 2. Daily per capita dietary energy (kcal) at-home (excluding FCAH) and total (including FCAH), by

food insecurity class. Kenya and the Sudan


Note: For HCES, at-home dietary energy was calculated excluding FCAH; total dietary energy was calculated considering both at-

home and away-from home foods. Differences across groups were assessed with regression analysis followed by Tukey’s pairwise

post-hoc tests (family-wise error rate of 5 percent). a Significant difference between the moderately food insecure and the food secure/mildly food insecure groups (p-value < .05). b Significant difference between the severely food insecure and the food secure/mildly food insecure groups (p-value < .05). c Significant difference between the moderately and severely food insecure groups (p-value < .05).

a,b,ca,b,c

0

500

1 000

1 500

2 000

2 500

Dietary energy excluding FCAH Total dietary energy

kcal

/cap

ita/

day

Kenya


a,b,ca,b

0

500

1 000

1 500

2 000

2 500

3 000


kcal

/cap

ita/

day

Sudan


14

As a consequence of the reduction in food consumption, the amount of available carbohydrates (i.e. total

carbohydrates minus total fibre), protein and fats consumed by the moderately and severely food

insecure groups is lower compared to that of their food secure counterparts (Figure 3). Nevertheless, all

food insecurity classes (both in Kenya and the Sudan) meet the acceptable macronutrient distribution

ranges (55–75 percent of daily calories from carbohydrates, 15–30 percent from fats and 10–15 percent

from protein) (Figure 4).

In both countries, total fibre consumption across the three food insecure classes is much higher (close to

or above 50 g/capita/day) than the recommended minimum of 25 grams/capita/day (WHO and FAO,

2003), nevertheless, it decreases with the severity of food insecurity (Figure 3).

15

Figure 3. Daily per capita carbohydrates (available), fats, protein, and total fibre (grams) by food insecurity

class. Kenya and the Sudan


Note: For HCES, total nutrient consumption was calculated considering both at-home and away-from home foods. Differences

across groups were assessed with regression analysis followed by Tukey’s pairwise post-hoc tests (family-wise error rate of

5 percent). a Significant difference between the moderately food insecure and the food secure/mildly food insecure groups (p-value < .05). b Significant difference between the severely food insecure and the food secure/mildly food insecure groups (p-value < .05). c Significant difference between the moderately and severely food insecure groups (p-value < .05).

a,b,c

a,b,ca,b,c a,b,c

0

50

100

150

200

250

300

350

400

Available carbohydrates Protein Fats Total fibre

gram

s/ca

pit

a/d

ay

Kenya


a,b

a,ba,b,c a,b

0

50

100

150

200

250

300

350

400

450


gram

s/ca

pit

a/d

ay

Sudan


16

Figure 4. Proportion of total macronutrients to total dietary energy consumption by food insecurity class.

Kenya and the Sudan


In both countries, most households across all food insecurity classes report consuming cereals, fats and

oils, and sweets and sugars (Table 5). For all food groups, the proportion of households reporting their

consumption decreases with the severity of food insecurity, the exception being fish consumption in

Kenya. In both Kenya and the Sudan, the number of food insecure households reporting consumption of

eggs and fruits is much lower than for food secure households. In Kenya, the number of households

reporting meat consumption is much lower than for food secure households.

0

10

20

30

40

50

60

70

80

Carbohydrates Protein Fats

Shar

e o

f d

ieta

ry e

ner

gy c

on

trib

ute

d

Kenya

Minimum recommended Food secure or mildly food insecure Moderately food insecure Severely food insecure Maximum recommended

0

10

20

30

40

50

60

70

80


Shar

e o

f d

ieta

ry e

ner

gy c

on

trib

ute

d

Sudan


17

Table 5. Percentage of individuals living in households reporting consumption of food groups during the

reference period, by food insecurity class. Kenya and the Sudan

Item group Food insecurity class Kenya Sudan

Cereals and their

products

Food secure or mildly food insecure 99 99

Moderately food insecure 99 99

Severely food insecure 99 98

Roots, tubers,

plantains and their

products




Pulses, seeds and

nuts and their

products




Dairy products




Eggs and their

products




Fish, shellfish and

their products




Meat and their

products




Vegetables and their

products




Fruits and their

products




Fats and oils




Sweets and sugars




Value between 25–50 percent Value ≤ 25 percent


Note: The reference period was 7 days for both Kenya and the Sudan.

18

In Mexico and Samoa (both of which are upper-middle-income countries), notable differences in the diet

are also observed between the food secure and the food insecure groups,7 but they follow a different

pattern compared to Kenya and the Sudan (Figure 5 and Figure 6). As food insecurity becomes more

severe, dietary energy intake remains similar in Samoa, and declines less markedly in Mexico, in

comparison to the Sudan and Kenya. There is a reduction in the consumption of some animal-source foods

(such as dairy and meat), but minimal change (or even an increase) in the consumption of some plant-

based foods (such as cereals, roots, tubers and plantains, pulses, seeds and nuts, and vegetables) and

sweets and sugars. In Mexico, fruit consumption decreases as food security worsens, whereas in Samoa

it increases. Conversely, consumption of eggs in Mexico is higher in food insecure classes than in the food

secure.

7 In Samoa, the number of cases with severe food insecurity was extremely low, therefore, these cases were combined with those presenting moderate food insecurity, so as to provide reliable estimates of food consumption by class of food insecurity. This combined group is then referred as “moderately or severely food insecure”.

19

Figure 5. Daily per capita consumption of selected food groups in each food insecurity class. Mexico and

Samoa

Source: authors’ own calculations, 2021. With data from Mexico’s ENSANUT 2012 and the Samoa HIES 2018.

Note: Food consumption estimates shown for selected food groups only. For Mexico, differences across groups were assessed

with regression analysis followed by Tukey’s pairwise post-hoc tests (family-wise error rate of 5 percent). For Samoa, differences

across groups were assessed with regression analysis. a Significant difference between the moderately food insecure and the food secure/mildly food insecure groups (p-value < .05). b Significant difference between the severely food insecure and the food secure/mildly food insecure groups (p-value < .05). c Significant difference between the moderately and severely food insecure groups (p-value < .05).

a,b,c

a,b,c

a,b,c

a,b,c

a,b,c

a,b,c

a,b,ca,b,c

a,b,c

a,b

0

50

100

150

200

250


Roots,tubers,



theirproducts

Dairyproducts

Eggs andtheir

products


Meat andmeat

products


Fruits andtheir

products

Fats andoils

Sweets andsugars

gram

s/ca

pit

a/d

ay

Mexico


a

a

aa

aa

a

0

50

100

150

200

250

300

350

400

450

500


Roots,tubers,



theirproducts

Dairyproducts

Eggs andtheir

products


Meat andmeat

products


Fruits andtheir

products

Fats andoils

Sweets andsugars

gram

s/ca

pit

a/d

ay

Samoa

Food secure or mildly food insecure Moderately or severely food insecure

20

Figure 6. Daily per capita dietary energy (kcal) at-home (excluding FCAH) for Samoa and total (including

FCAH) for Mexico and Samoa, by food insecurity class


Note: For Mexico, only total dietary energy was calculated, and it considered all food items reported; differences across groups

were assessed with regression analysis followed by Tukey’s pairwise post-hoc tests (family-wise error rate of 5 percent). For

Samoa, at-home dietary energy was calculated excluding FCAH; total dietary energy was calculated considering both at-home

and away-from home foods; differences across groups were assessed with regression analysis. a Significant difference between the moderately food insecure and the food secure/mildly food insecure groups (p-value < .05). b Significant difference between the severely food insecure and the food secure/mildly food insecure groups (p-value < .05). c Significant difference between the moderately and severely food insecure groups (p-value < .05).

a,b,c

0

500

1 000

1 500

2 000

2 500


kcal

/cap

ita/

day

Mexico


a

0

500

1 000

1 500

2 000

2 500

3 000


kcal

/cap

ita/

day

Samoa

Food secure or mildly food insecure Moderately or severely food insecure

21

In Mexico, the total amount of available carbohydrates (i.e. total carbohydrates minus total fibre), protein,

and fats consumed by those who are food insecure is slightly lower than the consumption by their food

secure counterparts (Figure 7). In Samoa, consumption of protein is lower in the food insecure class

compared to the food secure, whereas consumption of both available carbohydrates and fats is similar.

Overall, in Mexico and Samoa, consumption of dietary energy from total carbohydrates, protein and fats

is not within the recommended range for both the food secure and food insecure groups (Figure 8). The

percentage of dietary energy provided by total carbohydrates is below the lower recommended threshold

(55 percent), with the exception of food insecure individuals in Mexico, and the contribution from fats is

above the upper threshold (30 percent). Total fibre consumption in Mexico is slightly below the

recommended minimum of 25 g/capita/day for all food insecurity classes, whereas in Samoa it is slightly

above (Figure 7).

22

Figure 7. Daily per capita carbohydrates (available), fats, protein, and total fibre (grams) by food insecurity

class. Mexico and Samoa


Notes: For Mexico, total nutrient consumption considered all food items reported; differences across groups were assessed with

regression analysis followed by Tukey’s pairwise post-hoc tests (family-wise error rate of 5 percent). For Samoa, total nutrient

consumption was calculated considering both at-home and away-from home foods; differences across groups were assessed with

regression analysis. a Significant difference between the moderately food insecure and the food secure/mildly food insecure groups (p-value < .05). b Significant difference between the severely food insecure and the food secure/mildly food insecure groups (p-value < .05). c Significant difference between the moderately and severely food insecure groups (p-value < .05).

a,b,c

a,b,c a,b,c

a,c

0

50

100

150

200

250

300


gram

s/ca

pit

a/d

ay

Mexico

Food secure or mildly food insecure Moderately food insecure Severely food insecure Label

a

0

50

100

150

200

250

300

350

400


gram

s/ca

pit

a/d

ay

Samoa

Food secure or mildly food insecure Moderately or severely food insecure label

23

Figure 8. Proportion of total macronutrients to total dietary energy consumption by food insecurity class.

Mexico and Samoa


In Samoa, a majority of households in the two food insecurity classes report consuming cereals, fats and

oils, sweets and sugars, roots, tubers and plantains, fish, meat, and vegetables (Table 6). In Mexico, only

cereals are reported by a majority of individuals. The fact that in Mexico the percentages of individuals

reporting consumption of food groups is lower compared to the other three countries may be related to

differences in the design of the food consumption module; additionally, the statistics for Mexico reflect

individuals’ consumption based on the first 24-hour recall, while for the other countries, they reflect

households’ usual consumption.

0

10

20

30

40

50

60

70

80


Shar

e o

f d

ieta

ry e

ner

gy c

on

trib

ute

d

Mexico


0

10

20

30

40

50

60

70

80


Shar

e o

f d

ieta

ry e

ner

gy c

on

trib

ute

d

Samoa

Minimum recommended Food secure or mildly food insecure Moderately or severely food insecure Maximum recommended

24

Table 6. Percentage of individuals reporting (Mexico) or living in households reporting (Samoa)

consumption of food groups during the reference period, by food insecurity class

Item group Food insecurity class Mexicoa Samoa

Cereals and their

products



Severely food insecure 93

Roots, tubers,

plantains and their

products




Pulses, seeds and

nuts and their

products




Dairy products




Eggs and their

products




Fish, shellfish and

their products




Meat and meat

products





products




Fruits and their

products




Fats and oils




Sweets and sugars




Value between 25–50 percent Value ≤ 25 percent


Notes: The reference period was: 24 hours for Mexico and 14 days for Samoa. a On the first 24-hour recall.

25

Discussion

The analysis reveals that people who experience moderate or severe food insecurity consume less meat,

dairy products, vegetables (for Kenya and the Sudan) and fruits (except for Samoa) and a higher

proportion of other plant-based foods (such as cereals, roots, tubers and plantains, and pulses, seeds and

nuts) that are typically cheaper on a per-calorie basis. Overall, the more food insecure people are, the

larger the share of staples in their diet. Moreover, the reduction in nutritious food groups becomes more

pronounced as the severity of food insecurity increases.

These findings are consistent with the underlying theoretical constructs of food insecurity: people

experiencing moderate food insecurity face uncertainties about their ability to obtain food and have been

forced to compromise on the nutritional quality or quantity of the food they consume, whereas people

experiencing severe food insecurity have typically run out of food and, at worst, gone one or more days

without eating (Ballard, 2013), thereby markedly reducing the quantities of food they consume.

One reason for food insecure households in Kenya having a higher increased fish consumption compared

to food secure households could be that subsistence fishing is practiced by some of the poorest and most

food insecure local communities in the country (Abila, 2003).

Another important finding of this analysis is that the ways moderately food insecure people modify their

diets vary according to the income level of the country. In the two lower-middle-income countries studied

(Kenya and the Sudan), there is a marked decrease in the consumption of most food groups with an

increase in the share of staples in the diet. On the other hand, in Mexico and Samoa, which are upper-

middle-income countries, the consumption of cheaper foods increases with more expensive foods being

consumed in lesser amounts. Mexico particularly shows a decrease in fruit and dairy consumption. This is

in line with Muhammad et al. (2017), who found that purchases of fruits and milk are vulnerable to

changes in income and prices. In the case of Samoa, the increase in fruit consumption with an increase in

food insecurity is explained by the fact that a high proportion of the total fruit consumed is not purchased,

but rather comes from own production (58 percent and 77 percent for food secure, and moderate or

severe food insecure, respectively; results not shown): as a result, consumption is less susceptible to

prices.

There are several plausible reasons why food insecurity, as measured by experience-based scales like the

FIES, may contribute to different dietary outcomes in lower-middle- and upper-middle-income countries,

to the extent that these countries may be exemplary of other countries in the same income level groups.

First, healthy diets may generally be less affordable in lower-middle-income countries than in upper-

middle-income countries (Hirvonen et al., 2020). Second, social protection programmes may receive less

funding in lower-middle-income countries (World Bank, 2018). Lastly, vulnerable people’s access to food,

especially perishable nutritious foods, may be more compromised in lower-middle-income countries than

in upper-middle-income countries, due to a lack of physical infrastructure and food processing and storage

technology, as well as food safety issues (Burlingame and Dernini, 2019; Jaffee et al., 2019).

This study also found that in Mexico and Samoa, the decrease in dietary energy consumption among

people who are moderately food insecure, compared to those who are food secure, is less marked than

in Kenya and the Sudan. A related finding is that, in Samoa and Mexico, no group met the recommended

ranges of energy intake from macronutrients (the share of dietary energy provided by fats is too high and

26

the share from carbohydrates is too low); whereas in Kenya and the Sudan all classes met the

recommendations for macronutrient distribution. This may be because both Mexico and Samoa are well

into the nutrition transition, which is characterized by a rapid shift in diet composition towards a higher

consumption of energy-dense foods high in fat, sugar or salt that are low-cost and widely available

(Popkin, Adair and Ng, 2012).

Limitations and strengths

As with every study, our analysis has limitations that should be considered when interpreting results. First,

we used two different sources of food consumption data, which hampers comparability of results across

data sources. While individual-level dietary food consumption surveys provide detailed quantitative

information at the individual level, a small number of recent nationally representative surveys are

available, and an even smaller number of surveys including an assessment of food security with an

experience-based food insecurity scale. For this reason, we relied on an additional data source, i.e. HCES.

Second, HCES are not purposefully designed to capture food consumption, therefore derived estimates

of food and nutrient consumption may be biased and not even comparable across different HCES due to

different survey designs. In addition, they do not provide information about the food and nutrient intake

of household members. However, we attempted to make food consumption statistics comparable across

countries to the extent possible.

In spite of these limitations, we found coherent trends in changes in consumption across food insecurity

classes within countries, according to what had been hypothesized. Furthermore, this is the first study of

the relationship of food insecurity and diet with multiple datasets where all food insecurity measures

were calibrated to the global reference scale following the FIES methodology. This allowed us to obtain

cross-country comparable classes of food insecurity.

27

Conclusion

Overall, the analysis revealed that the more food insecure people are, the larger the share of staples in

their diet. In the two lower-middle-income countries studied (Kenya and the Sudan), there is a marked

decrease in consumption of most food groups, except for staple foods. In the two upper-middle-income

countries examined (Mexico and Samoa), people who are moderately food insecure consume more foods

that are typically cheaper on a per-calorie basis (cereals, roots, tubers and plantains), and consume lesser

amounts of expensive foods (such as meat), compared with those who are food secure. All four countries

show a decrease in dairy consumption with the increase of food insecurity.

28

References

Abila, R. 2003. Fish trade and food security: are they reconcilable in lake Victoria? Rome: FAO.

http://www.fao.org/3/Y4961E/y4961e0d.htm

Ali, D., Saha, K. K., Nguyen, P. H., Diressie, M. T., Ruel, M. T., Menon, P., & Rawat, R. 2013. Household Food

Insecurity Is Associated with Higher Child Undernutrition in Bangladesh, Ethiopia, and Vietnam, but the

Effect Is Not Mediated by Child Dietary Diversity. The Journal of Nutrition, 143(12), 2015–2021.

https://doi.org/10.3945/jn.113.175182

Ballard, T.J., Kepple, A.W. & Cafiero, C. 2013. The Food Insecurity Experience Scale. Development of a

Global Standard for Monitoring Hunger Worldwide. FAO. http://www.fao.org/3/a-as583e.pdf

Burlingame, B. & Dernini, S. (Eds.). 2019. Sustainable diets: linking nutrition and food systems.

https://doi.org/10.1079/9781786392848.0000

FAO. 2003. Food Energy: Methods of Analysis and Conversion Factors. FAO Food and Nutrition Paper 77.

Rome: FAO. http://www.fao.org/uploads/media/FAO_2003_Food_Energy_02.pdf

FAO. 2012. Panorama de la seguridad alimentaria y la nutrición. Mexico City. http://www.fao.org/3/a-

i3269s.pdf

FAO. 2016. Methods for estimating comparable rates of food insecurity experienced by adults throughout

the world. Rome, FAO. http://www.fao.org/3/c-i4830e.pdf

FAO. 2017. The Food Insecurity Experience Scale: Measuring food insecurity through people’s experiences.

Rome, FAO. https://doi.org/I7835EN/1/09.17

FAO & Government of Kenya. 2018. Kenya Food Composition Tables. Nairobi. http://www.fao.org/

3/I8897EN/i8897en.pdf

FAO, IFAD, UNICEF, WFP & WHO. 2019. The State of Food Security and Nutrition in the World 2019.

Safeguarding against economic slowdowns and downturns. http://www.fao.org/3/ca5162en/

ca5162en.pdf

FAO & INFOODS. 2012. Guidelines for Food Matching: Version 1.2. Rome. http://www.fao.org/docrep/

017/ap805e/ap805e.pdf

FAO, INFOODS, The West African Health Organization, & Bioversity International. 2012. West African Food

Composition Table. Rome. http://www.fao.org/3/a-i2698b.pdf

FAO & WHO. 2020. FAO/WHO Global Individual Food consumption data (GIFT) Tool. http://www.fao.org/

gift-individual-food-consumption/en/

Gundersen, C., & Ziliak, J. P. 2015. Food insecurity and health outcomes. Health Affairs, 34(11), 1830–

1839. https://doi.org/10.1377/hlthaff.2015.0645

Hanson, K. L., & Connor, L. M. 2014. Food insecurity and dietary quality in US adults and children: a

systematic review. The American Journal of Clinical Nutrition, 100(2), 684–692.

https://doi.org/10.3945/ajcn.114.084525

http://www.fao.org/3/Y4961E/y4961e0d.htm

https://doi.org/10.3945/jn.113.175182

http://www.fao.org/3/a-as583e.pdf

https://doi.org/10.1079/9781786392848.0000

http://www.fao.org/uploads/media/FAO_2003_Food_Energy_02.pdf

http://www.fao.org/3/a-i3269s.pdf

http://www.fao.org/3/a-i3269s.pdf

http://www.fao.org/3/c-i4830e.pdf

https://doi.org/I7835EN/1/09.17

http://www.fao.org/3/I8897EN/i8897en.pdf

http://www.fao.org/3/I8897EN/i8897en.pdf

http://www.fao.org/3/ca5162en/ca5162en.pdf

http://www.fao.org/3/ca5162en/ca5162en.pdf

http://www.fao.org/docrep/017/ap805e/ap805e.pdf

http://www.fao.org/docrep/017/ap805e/ap805e.pdf

http://www.fao.org/3/a-i2698b.pdf

http://www.fao.org/gift-individual-food-consumption/en/

http://www.fao.org/gift-individual-food-consumption/en/

https://doi.org/10.1377/hlthaff.2015.0645

https://doi.org/10.3945/ajcn.114.084525

29

Hirvonen, K., Bai, Y., Headey, D., & Masters, W. A. 2020. Affordability of the EAT–Lancet reference diet: a

global analysis. The Lancet Global Health, 8(1), e59–e66. https://doi.org/10.1016/S2214-109X(19)30447-

4

Jaffee, S., Henson, S., Unnevehr, L., Grace, D., & Cassou, E. 2019. The Safe Food Imperative: Accelerating

Progress in Low- and Middle-Income Countries. Washington, DC: World Bank.

https://openknowledge.worldbank.org/handle/10986/30568

Johnson, C. M., Sharkey, J. R., Lackey, M. J., Adair, L. S., Aiello, A. E., Bowen, S. K., Fang W., Flax V.L. &

Ammerman, A. S. 2018. Relationship of food insecurity to women’s dietary outcomes: a systematic review.

Nutrition Reviews, 76(12), 910–928. https://doi.org/10.1093/nutrit/nuy042

Leung, C. W., Epel, E. S., Ritchie, L. D., Crawford, P. B., & Laraia, B. A. 2014. Food Insecurity Is Inversely

Associated with Diet Quality of Lower-Income Adults. Journal of the Academy of Nutrition and Dietetics,

114(12), 1943-1953.e2. https://doi.org/10.1016/j.jand.2014.06.353

López-Olmedo, N., Carriquiry, A. L., Rodríguez-Ramírez, S., Ramírez-Silva, I., Espinosa-Montero, J.,

Hernández-Barrera, L., Campirano F., Martínez-Tapia B. & Rivera, J. A. 2016. Usual Intake of Added Sugars

and Saturated Fats Is High while Dietary Fiber Is Low in the Mexican Population. The Journal of Nutrition,

146(9), 1856S-1865S. https://doi.org/10.3945/jn.115.218214

Martin, M. S., Maddocks, E., Chen, Y., Gilman, S. E., & Colman, I. 2016. Food insecurity and mental illness:

Disproportionate impacts in the context of perceived stress and social isolation. Public Health, 132, 86–

91. https://doi.org/10.1016/j.puhe.2015.11.014

Maynard, M., Andrade, L., Packull-McCormick, S., Perlman, C. M., Leos-Toro, C., & Kirkpatrick, S. I. 2018.

Food insecurity and mental health among females in high-income countries. International Journal of

Environmental Research and Public Health. MDPI AG. https://doi.org/10.3390/ijerph15071424

Moltedo, A., Álvarez-Sánchez, C., Troubat, N., & Cafiero, C. 2018. Optimizing the use of ADePT-Food

Security Module for Nutrient Analysis: ADePT-FSM Version 3. (S. D. FAO, Ed.). Rome.

http://www.fao.org/fileadmin/templates/ess/foodsecurity/Optimizing_the_use_of_ADePT_FSM_for_nu

trient_analysis.pdf

Moltedo, A., Troubat, N., Lokshin, M., & Sajaia, Z. 2014. Analyzing Food Security Using Household Survey

Data: Streamlined Analysis with ADePT Software. Washington, DC: World Bank.


Muhammad, A., D’Souza, A., Meade, B., Micha, R., & Mozaffarian, D. 2017. How income and food prices

influence global dietary intakes by age and sex: Evidence from 164 countries. BMJ Global Health, 2(3),

e000184. https://doi.org/10.1136/bmjgh-2016-000184

Mundo-Rosas, V., De La Cruz-Góngora, V., Jiménez-Aguilar, A., & Shamah-Levy, T. 2014. [Dietary diversity

and nutrient intake in children 24-59 months old and their association with food insecurity]. Salud Publica

Mex, 56 (supplement 1). http://www.scielo.org.mx/pdf/spm/v56s1/v56s1a7.pdf

National Cancer Institute. 2009. Usual dietary intakes: SAS macros for the NCI method. National Cancer

Institute. [online] [cited 20 March 2020] https://epi.grants.cancer.gov/diet/usualintakes/macros.html

Nord, M. 2014. What have we learned from two decades of research on household food security? Public

https://doi.org/10.1016/S2214-109X(19)30447-4

https://doi.org/10.1016/S2214-109X(19)30447-4


https://doi.org/10.1093/nutrit/nuy042

https://doi.org/10.1016/j.jand.2014.06.353

https://doi.org/10.3945/jn.115.218214

https://doi.org/10.1016/j.puhe.2015.11.014

https://doi.org/10.3390/ijerph15071424

http://www.fao.org/fileadmin/templates/ess/foodsecurity/Optimizing_the_use_of_ADePT_FSM_for_nutrient_analysis.pdf

http://www.fao.org/fileadmin/templates/ess/foodsecurity/Optimizing_the_use_of_ADePT_FSM_for_nutrient_analysis.pdf


https://doi.org/10.1136/bmjgh-2016-000184

http://www.scielo.org.mx/pdf/spm/v56s1/v56s1a7.pdf

https://epi.grants.cancer.gov/diet/usualintakes/macros.html

30

Health Nutrition. Cambridge University Press. https://doi.org/10.1017/S1368980013003091

Pacific Community, FAO & University of Wollongong. 2020. Pacific Nutrient database Version 2 - A tool to

facilitate poverty, nutrition and food security analysis in the Pacific region.

Pinheiro de Toledo-Vianna, R., Hromi-Fiedler, A., Segall-Correa, A., & Pérez-Escamilla, R. 2012. Household

food insecurity in small municipalities in Northeastern Brazil. In International Scientific Symposium on

Food and Nutrition Security information: from Valid Measurement to Effective Decision Making. Rome.

Popkin, B. M., Adair, L. S., & Ng, S. W. 2012. Global nutrition transition and the pandemic of obesity in

developing countries. Nutr Rev, 70(1), 3–21. https://doi.org/10.1111/j.1753-4887.2011.00456.x

Rodríguez, L. A., Mundo-Rosas, V., Méndez-Gómez-Humarán, I., Pérez-Escamilla, R., & Shamah-Levy, T.

2017. Dietary quality and household food insecurity among Mexican children and adolescents. Maternal

& Child Nutrition, 13(4), e12372. https://doi.org/10.1111/mcn.12372

Tooze, J., Kipnis, V., Buckman, D., Carroll, R., Freedman, L., Guenther, P., Krebs-Smith S.M., Subar A.F. &

Dodd, K. 2010. A mixed-effects model approach for estimating the distribution of usual intake of nutrients:

the NCI method. Statistics in Medicine, 29(27), 2857–2868. https://doi.org/10.1002/sim.4063

U.S. Department of Agriculture, Agricultural Research Service. 2019. FoodData Central.

https://fdc.nal.usda.gov/

Vega-Macedo, M., Shamah-Levy, T., Peinador-Roldan, R., Mendez-Gomez, I., & Melgar-Quiñonez, H. 2014.

[Food insecurity and dietary diversity in Mexican households with children less than five years of age].

Salud Publica Mex, 56 (supplement 1), 1:S21-S30. http://www.scielo.org.mx/pdf/spm/v56s1/v56s1a5.pdf

WHO & FAO. 2003. Diet, nutrition and the prevention of chronic diseases: report of a joint WHO/FAO

expert consultation. World Health Organization technical report series (Vol. 916). Geneva.

https://apps.who.int/iris/bitstream/handle/10665/42665/WHO_TRS_916.pdf?sequence=1

World Bank. 2018. The State of Social Safety Nets 2018. Washington, DC: World Bank.


World Bank. 2019. Global Consumption Database. [online] http://datatopics.worldbank.org/

consumption/

https://doi.org/10.1017/S1368980013003091

https://doi.org/10.1111/j.1753-4887.2011.00456.x

https://doi.org/10.1111/mcn.12372

https://doi.org/10.1002/sim.4063

https://fdc.nal.usda.gov/

http://www.scielo.org.mx/pdf/spm/v56s1/v56s1a5.pdf

https://apps.who.int/iris/bitstream/handle/10665/42665/WHO_TRS_916.pdf?sequence=1


http://datatopics.worldbank.org/consumption/

http://datatopics.worldbank.org/consumption/

31

Annex A1: Food Insecurity Experience Scale (FIES) survey module

included in the Kenya Integrated Household Budget Survey, Samoa

Household Income and Expenditure Survey, and the Sudan

Consumption Patterns and Nutrition Study

The Food Insecurity Experience Scale Survey Module (FIES-SM) is composed of eight questions (i.e. items)

with dichotomous yes/no responses. Together, the FIES-SM items compose a statistical scale designed to

cover a range of severity of food insecurity and should be analysed together as a scale, not as separate

items.

GLOBAL FOOD INSECURITY EXPERIENCE SCALE Household Referenced

Now I would like to ask you some questions about food. During the last 12 MONTHS, was there a time when: Q1. You or others in your household worried about not having enough food to eat because of a lack of money or other resources?

0 No 1 Yes 98 Don’t know 99 Refused

Q2. Still thinking about the last 12 MONTHS, was there a time when you or others in your household were unable to eat healthy and nutritious food because of a lack of money or other resources?

0 No 1 Yes 98 Don’t know 99 Refuse

Q3. Was there a time when you or others in your household ate only a few kinds of foods because of a lack of money or other resources?


Q4. Was there a time when you or others in your household had to skip a meal because there was not enough money or other resources to get food?


Q5. Still thinking about the last 12 MONTHS, was there a time when you or others in your household ate less than you thought you should because of a lack of money or other resources?


Q6. Was there a time when your household ran out of food because of a lack of money or other resources?


Q7. Was there a time when you or others in your household were hungry but did not eat because there was not enough money or other resources for food?


32

Q8. Was there a time when you or others in your household went without eating for a whole day because of a lack of money or other resources?


33

Annex A2: Latin American and Caribbean Scale (ELCSA) module

included in Mexico’s National Health and Nutrition Survey

(ENSANUT) 2012

Latin American and Caribbean Scale

Household Referenced

1. En los últimos 3 meses, por falta de dinero u otros recursos, alguna vez ¿Usted se preocupó de que los alimentos se acabaran en su hogar?

1 SI 2 NO 99 No Sabe 88 No Responde

2. En los ultimos 3 meses, por falta de dinero u otros recursos, alguna vez ¿En su hogar se quedaron sin alimentos?


3. En los ultimos 3 meses, por falta de dinero u otros recursos, alguna vez ¿En su hogar dejaron de tener una alimentacion (saludable, nutritiva, balanceada, equilibrada)?


4. En los ultimos 3 meses, por falta de dinero u otros recursos, alguna vez ¿Usted o algun adulto en su hogar tuvo una alimentacion basada en poca variedad de alimentos?


5. En los ultimos 3 meses, por falta de dinero u otros recursos, alguna vez ¿Usted o algun adulto en su hogar dejo de desayunar, (comer, almorzar) o cenar?


6. En los ultimos 3 meses, por falta de dinero u otros recursos, alguna vez ¿Usted o algun adulto en su hogar comio menos de lo que debia comer?


7. En los ultimos 3 meses, por falta de dinero u otros recursos, alguna vez ¿Usted o algun adulto en su hogar sintio hambre pero no comio?


8. En los ultimos 3 meses, por falta de dinero u otros recursos, alguna vez ¿Usted o algun adulto en su hogar solo comio una vez al dia o dejo de comer todo un dia?


¿En su hogar viven personas menores de 18 años? (Si, pasa a la pregunta 9; No, fin de la sección)

SI NO

9. En los últimos 3 meses, por falta de dinero u otros recursos, alguna vez ¿Algún menor de 18 años en su hogar dejó de tener una alimentación (saludable, nutritiva, balanceada, equilibrada)?


34

10. En los últimos 3 meses, por falta de dinero u otros recursos, alguna vez ¿Algún menor de 18 años en su hogar tuvo una alimentación basada en poca variedad de alimentos?


11. En los últimos 3 meses, por falta de dinero u otros recursos, alguna vez ¿Algún menor de 18 años en su hogar dejó de desayunar, (comer, almorzar) o cenar?


12. En los últimos 3 meses, por falta de dinero u otros recursos, alguna vez ¿Algún menor de 18 años en su hogar comió menos de lo que debía?


13. En los últimos 3 meses, por falta de dinero u otros recursos, alguna vez ¿Tuvieron que disminuir la cantidad servida en las comidas a algún menor de 18 años en su hogar?


14. En los últimos 3 meses, por falta de dinero u otros recursos, alguna vez ¿Algún menor de 18 años en su hogar sintió hambre pero no comió?


15. En los últimos 3 meses, por falta de dinero u otros recursos, alguna vez ¿Algún menor de 18 años en su hogar solo comió una vez al día o dejó de comer todo un día?


35

Annex A3: Food group classification used

For these analyses, we considered 11 out of a total of 19 food groups: cereals and their products; roots,

tubers, plantains and their products; pulses, seeds and nuts and their products; dairy products; eggs and

their products; fish, shellfish and their products; meat and their products; vegetables and their products;

fruits and their products; fats and oils; and sweets and sugars.

Food group classification

Food groups Sub-groups

Cereals and their

products

Rice and rice-based products

Maize and maize-based products

Wheat and wheat-based products

Sorghum and sorghum-based products

Millet and millet-based products

Other cereals, mixed cereals or unspecified cereals and their products

Roots, tubers, plantains

and their products

Potato, sweet potato and their products

Cassava and similar roots (excluding taro) and their products

Taro and taro-based products

Yam and yam-based products

Other and unspecified starchy roots and tubers (excluding sugary roots and

tubers) and their products

Plantain and plantain-based products

Pulses, seeds and nuts

and their products

Pulses (excluding soybeans) and their products

Soybean and soy-based products

Nuts, seeds and their products

Dairy products

Milk: fresh and processed (excluding fermented milk products, cream, whey,

cheese and other milk products)

Fermented milk products

Cream, whey and any other milk products excluding fermented milk products

and cheese

Cheese

Eggs and their products Eggs: fresh and processed

Fish, shellfish and their

products

Freshwater fish (excluding offal): fresh and processed (excluding dried)

Diadromous fish (excluding offal): fresh and processed (excluding dried)

Marine fish (excluding offal): fresh and processed (excluding dried)

Offal - fish and shellfish: fresh and processed (excluding dried)

Shellfish (excluding offal) - all types: fresh and processed (excluding dried)

Fish and shellfish - mixed or unspecified: fresh and processed (excluding

dried)

Fish and shellfish (including offal) - all types: dried

36

Food groups Sub-groups

Meat and their

products

Offal - all types: fresh and processed (excluding dried)

Mammals, reptiles and amphibians (excluding offal): fresh and processed

(excluding dried)

Birds (excluding offal): fresh and processed (excluding dried)

Meat - mixed or unspecified: fresh and processed (excluding dried)

Meat - all types: dried


products

Leafy vegetables: fresh

Yellow and orange vegetables: fresh

Vegetables (excluding leafy vegetables and including fresh legumes): fresh

Vegetables - all types: dried

Vegetables - all types, mixed and unspecified: processed (excluding dried)

Vegetables - mixed and unspecified: fresh

Fruits and their

products

Yellow and orange fruits: fresh

Fruits: fresh

Fruits: dried

Fruits: processed (excluding dried and candied)

Fats and oils

Vegetable fat and oil (excluding red palm oil)

Red palm oil

Animal fat and oil

Sweets and sugars

Dough-based sweets

Chocolate-based sweets

Fruit and nut-based sweets

Other sweets

Sugars

Dairy- or dairy imitate-based sweets

Eight food groups were not considered (insects, grubs and their products; foods for particular nutritional

uses; food supplements and similar; food additives; composite dishes; savoury snacks; and spices and

condiments) due to the very low/zero consumption or negligible contribution to usual energy and nutrient

consumption, or because they were not well captured by HCES.

Contact:

Statistics Division – Economic and Social Development

[email protected]

www.fao.org/food-agriculture-statistics/resources/publications/working-papers/en/

Food and Agriculture Organization of the United Nations

Rome, Italy


http://www.fao.org/food-agriculture-statistics/resources/publications/working-papers/en/

Documents

FAO Statistics Working Paper Series