Estimating the Health-Related Costs of Hunger

Full text

WWW.HUNGERREPORT.ORG • 2016 HUNGER REPORT 247PB APPENDIX 2 • BREAD FOR THE WORLD INSTITUTE
Appendix 2:
Estimating the Health-Related Costs of
Food Insecurity and Hunger
Exhibit 1 Number and percent of people living in food-insecure
households in the US, 2007-2014
Source: Coleman-Jensen, et al., 20152.
Year
Total Number of Individuals
Food Insecure (1000s)
Percent of Individuals
Food Insecure
2007 36,229 12.2%
2008 49,108 16.4%
2009 50,162 16.6%
2010 48,832 16.1%
2011 50,120 16.4%
2012 48,966 15.9%
2013 49,078 15.8%
2014 48,135 15.4%
John T. Cook, PhD, MAEd, Principal Investigator, Associate Professor of Pediatrics, Boston University School of Medicine
Ana Paula Poblacion, MSc, Project Manager & Research Assistant, Universidade Federal de São Paulo
Introduction
Hunger is a health issue. This report is primarily about health-related costs attributable to food insecurity and
hunger in the United States in 2014. The report also includes other kinds of costs associated with food insecurity,
but its focus is health-related costs. Our charge is to update information on costs of food insecurity in the United
States published in 2011,1 employing the most recently available data on prevalence of food insecurity in 2014
with the most valid estimation procedures available, and to expand on the health-related costs attributable to food
insecurity in the United States.
Executive Summary
Each September the Economic Research Service of the U.S. Department of Agriculture (USDA) reports esti-
mates of the number and prevalence of people living in food insecure households by various demographic char-
acteristics and levels of severity of food insecurity. Data for this report come from the December implementation
by the Census Bureau of the Current Population Survey, a nationally representative survey of the U.S. population.
In 2014, there were 48.135 million people (15.4 percent of the total population) living in households that were food
insecure at some level of severity (Exhibit 1). The number of food-insecure people in the United States in 2014
was 11.906 million higher than in 2007, the year the Great Recession began, and only 0.697 million lower than
in 2010. Between 2010 and 2014 the
nation’s food security situation did
not improve appreciably.
The most recent prior estimates
of the cost of food insecurity to the
nation by researchers at Brandeis
University1 addressed costs within
three domains: illness costs, educa-
tion and related costs, and charity
costs. The total illness costs esti-
mated for calendar year 2010 within
these three areas was $130.5 Billion.
We surveyed empirical food
security research literature pub-
lished in peer-reviewed academic
journals between 2005 and 2015,

WWW.HUNGERREPORT.ORG • 2016 HUNGER REPORT 249248 APPENDIX 2 • BREAD FOR THE WORLD INSTITUTE
Exhibit 2 Estimated Costs Attributable to Food Insecurity and
Hunger in the US, 2014
Sources described in document text.
Source of Cost
Costs
($Billion 2014 Dollars)
Direct health-related costs in 2014 based on new
research evidence
$29.68
Non-overlapping direct health-related costs reported by
Brandeis researchers in 2011, continued in 2014 and
expressed in 2014 dollars
$124.92
Indirect costs of lost work time due to workers’ illnesses
or workers providing care for sick family members based
on new research evidence
$5.48
Total direct and indirect 2014 health-related costs
$160.07
Indirect costs of special education in public primary and
secondary schools, based on new research evidence
$5.91
Total costs of dropouts reported by Brandeis research-
ers in 2011, continued in 2014 and expressed in 2014
dollars
$12.94
TOTAL ESTIMATED COSTS $178.93
and based our estimates on relationships identiable
in that literature. Using information from the research
literature reviewed, and from the 2011 Brandeis report,
we estimate the health-related costs attributable to food insecu-
rity to be $160.07 Billion in 2014 (Exhibit 2).
Domains of Costs Addressed in this Report
The cost estimates described in this report address the
following domains:
1. Direct costs of treatment of specic disease or
health conditions that are plausibly attributable to
household food insecurity.
2. Direct costs of special education in public primary
and secondary schools plausibly attributable to
food insecurity.
3. Indirect costs of lost work productivity resulting
from:
a. Workers’ own illnesses or other health prob-
lems attributable to food insecurity,
b. Workers providing care to a family member
whose illness is attributable to food insecurity.
Methods
To estimate the direct health-related costs attribut-
able to food insecurity in 2014, we reviewed empirical
research literature published in peer-reviewed journals
from approximately 2005 to 2015, searching for quan-
titative ndings of associations between food insecurity
and health outcomes. We specically searched for quan-
titative ndings that involved either odds ratios (most
often), likelihood ratios, or relative risk ratios expressing
the differences in likelihood of a person living in a food-
insecure household having a disease or disease condition
compared to a person living in a food-secure household
(food security status is the exposure variable).
Those probability ratios were then translated into
population attributable fractions (PAFs) expressing the
proportion of the total prevalence of the disease in the
population attributable to food insecurity (i.e., the excess
fraction attributable to food insecurity). As noted above,
this process requires the assumption that food insecurity
is causally related to the disease conditions.
In case-control studies, if adjusted odds ratios (ORs)
are available, they can be trans-
formed into relative risk ratios
using formula 1 below3:
1. RR = OR/[(1-Po)+(Po*OR)],
where RR is the relative risk
ratio,
OR is the odds ratio, and
Po is the proportion of the
unexposed (food secure)
who develop the outcome, or
become cases.
This adjustment is desirable
since, though the OR is an accept-
able estimate of the Relative Risk
ratio (RR) in case-control studies,
and approaches RR in the situation
of rare diseases in which very few
of the unexposed develop the dis-
ease, the higher the prevalence of
the disease in the unexposed popu-

WWW.HUNGERREPORT.ORG • 2016 HUNGER REPORT 249248 APPENDIX 2 • BREAD FOR THE WORLD INSTITUTE
lation (e.g., the food-secure population), the greater the
deviation of the RR from the OR.
With the relative risk ratios thus calculated (or if they
are available), they can be used to calculate estimates
of the excess population attributable fractions (PAF) of
the diseases arising due to exposure to the predictor,
food insecurity, using formula 2 below4:
1. PAF = Pe (RR - 1) / [Pe (RR - 1) + 1] * 100%, where
PAF is the excess population attributable fraction
of disease in the population considered to result
from the presence of the exposure variable or
condition (i.e., food insecurity),
RR is the relative risk ratio calculated as above,
and
Pe is the proportion of controls (those who do not
have the outcome or disease) who were exposed
(live in a food-insecure household).
A complete table of all the conditions for which we
found new studies providing the information needed
to calculate attributable fractions can be found in
Appendix Exhibit A1. For most of the health condi-
tions, the attributable fraction (AF) is relatively small,
10 percent or less. For a few conditions we found
research results leading to more than one AF for a con-
dition. In those cases, we either used the average of the
AFs, or used the one which was more reliable for the
specic age group and condition under consideration.
And for a few conditions, we were either unable to nd
data on the prevalence and number of people in the
relevant sub-population with the condition, or data
on the cost of treating cases of the condition. In those
few instances, we were unable to estimate the disease
burden or the costs. This was particularly true when
the condition was failure to receive recommended or
prescribed treatment, or treatment foregone due to
inability to pay as a result of food insecurity.
For a couple of conditions (e.g., PEDS concerns;
parents report of developmental concerns about their
child), we had to add an additional link to the chain of
logic such as obtaining positive predictive value of the
indicator (PEDS concerns) and the outcome (special
education). With a few conditions for which we could
not nd needed prevalence data, we relied on data
from the U.S. Census Bureau on relationships between
reported health status and health services utilization.5
Using the information in Exhibit 1A, together with
data from the Agency for Healthcare Research and
Quality’s Medical Expenditure Panel Survey (MEPS, or
other national survey data) on the number of cases of
each disease condition in the population in 2014 (when
available), we estimated the fraction (proportion) of cases
of each health condition attributable to food insecurity.
Combining the results of these calculations with data
on annual expenditures for treatment of individuals
with the condition (from MEPS or other national health
surveys), we estimated the total annual direct costs of
treatment for all individuals with the condition.
Data on numbers of hospitalizations, and average
costs of hospital stays were obtained from the Agency
for Healthcare Research & Quality’s Healthcare Cost &
Utilization Project public access data obtained via the
HCUPnet online query system (http://hcupnet.ahrq.
gov/). Data were obtained from both the HCUP National
Inpatient Database and the HCUP Kids’ Inpatient Data-
base. Several price index series were used to adjust the
price of various healthcare services. These price indices
were taken from the Bureau of Labor Statistics’ online
databases (http://www.bls.gov/cpi/). Resulting estimated
costs for each condition are presented in Appendix
Exhibit 2.
The Brandeis researchers estimated the cost of the
private food assistance system at $17.8 Billion in 2010
($19.52 Billion in 2014 dollars), and we calculated
the total cost of the public food assistance system to
be $103.55 Billion in 2014. However discussions with
healthcare colleagues and others led us to the position
that the costs of these two complementary food assis-
tance systems are more accurately viewed as the costs
of prevention of food insecurity, not as a cost of food
insecurity itself. The costs of these two food assistance
systems are the costs of the vaccine that prevents food
insecurity and hunger from occurring in the nation’s
households, families and children. Thus the costs of
these two systems are not included as costs attributable
to food insecurity.

WWW.HUNGERREPORT.ORG • 2016 HUNGER REPORT 251250 APPENDIX 2 • BREAD FOR THE WORLD INSTITUTE
Background and Context
A Note on Hunger
Hunger is probably a more complex phenomenon
than most people imagine. The term is used to mean
several different things, and its scope varies depending
on its intended meaning. First, hunger is part of
humans’ “creatureliness,” arising from of our nature as
living systems that require regular intake of food to live,
act, grow, develop, and be healthy. We all experience
hunger every day; we know when we are hungry, and
we can tell someone how hungry we are; i.e., we can
“self-report” our hunger and its severity.6
At its most basic level, hunger is a neurochemical
feedback loop: a reinforcing feedback loop that leads to
more food intake the hungrier we are. The hunger feed-
back loop involves transmission of information to the
brain as the stomach empties and its biochemical state
changes. The time required for this emptying process is
approximately 2-4 hours, depending on the contents of
the stomach, activity levels, and other factors. It coincides
generally with humans’ customary schedule of eating
three meals per day. When a person’s normal pattern of
food intake is interrupted by a lack of food, she becomes
hungry. If she doesn’t eat, she becomes even hungrier.6
Hunger can be described and measured in several
ways. It is a drive to nd and consume food, and the
intensity of this drive depends partly on the amount
of food eaten during, and length of time since, the last
episode of food intake. Hunger also is a state, with
physical and mental components; it is the opposite
of satiety. When we are hungry, and food is readily
available, and accessible, we eat until we are sated, or
no longer hungry, and normally then we stop eating.
Satiety is also a neurochemical feedback loop; a bal-
ancing feedback loop that leads to less food intake as
the stomach lls and sends neurochemical signals to
the brain causing the feeling of satiety to increase, and
the feeling of hunger to decrease. Healthy people, with
no eating issues, stop eating when they become sated.
But the “processes” of hunger and satiety are neither
mechanistic nor completely regular. And they are not
isolated within an individual. They occur within and are
strongly inuenced by social contexts, because humans
are social beings. Each of us is a set of body systems
living and acting within concentrically larger and more
complex social systems. And we experience hunger as
both a personal and a social condition. Our very ear-
liest social interactions involve being fed, and nurtured.
And as we grow, food, hunger, eating together, sharing
food, being fed, nourished and nurtured, and nour-
ishing and nurturing others, are fundamental social
processes through which we learn to trust, respect, and
care for each other.
We learn through social interactions around hunger,
food, and eating that we depend on others, and that
others depend on us. We learn etiquette: basic social
rules that form a foundation on which we build ethics,
and moral values. We celebrate important life-cycle
events, such as birthdays, graduations, marriages, reli-
gious and civil holidays, and deaths, by enjoying and
sharing food. Food and satisfying hunger are at the
base of Maslow’s hierarchy of needs,7 and until their
food and hunger needs are met, humans cannot fulll
other higher-order needs. But food and hunger are also
social, and they permeate our social lives. We employ
food and hunger, and satisfying hunger, in pursuit of
higher-order needs.
So hunger is an individual set of feelings and sen-
sations, grounded in individuals’ neurochemical feed-
back loops, but it is even more a set of social feelings
and sensations, grounded in humans’ social nature. We
live in relationships, some intimate, some casual, some
formal, some informal, but all fundamental to our
nature as social beings. Hunger is both an individual
and a social process, experienced and responded to
in social contexts through social interactions and pro-
cesses. And when hunger cannot be satised, for what-
ever reasons, it affects our social beings, our social lives,
social relationships, and social interactions.
Hunger becomes problematic when it cannot be
reduced, or when we cannot respond to it appropriately,
because we lack the wherewithal or resources necessary
to obtain and consume food in socially acceptable ways.
The reinforcing feedback loop of hunger can become
out of control, and cause the system to collapse, liter-
ally, if the balancing feedback loop of satiety is not able
to operate. But neither of these feedback loops operates

WWW.HUNGERREPORT.ORG • 2016 HUNGER REPORT 251250 APPENDIX 2 • BREAD FOR THE WORLD INSTITUTE
in isolation; both also are social processes operating
within social contexts. And they involve and depend on
social interactions to reestablish balance.
Hunger becomes a social policy issue when the social
context, and all the social relationships it involves, fail to
provide socially acceptable ways for individual or family
systems to obtain the food needed to address hunger
in socially acceptable ways. When this occurs, those
systems are placed at risk for toxic stresses. And toxic
stress, intense acute stress or less intense chronic stress,
can be very corrosive and destructive. It damages both
child and adult health, and is especially pernicious in
young children. Toxic stress can damage the architec-
ture of children’s developing brains8, 9 and place signi-
cant constraints on their human capital development,
impairing the trajectories of their entire lives.10
The toxic stress of socially ignored or tolerated hunger
damages physical and mental health, but it also erodes
basic trust in and respect for social relationships, institu-
tions, and the people within them. Our health, well-being,
and prosperity depend on a strong functional base of
trust, respect, and compassion in all our relationships.
These are the glue that binds the public together and
makes it healthy and strong. And without a healthy,
strong public, none of us can really be healthy and strong
or prosperous, either as individuals or in relationships.
Humans are social, inter-dependent beings, and our
health, strength, well-being and prosperity depend on the
public welfare and strong public infrastructure. As trivial
as it can sometimes sound, we very literally are all in
this together. There is no “us” and “them,” there is only
us. And when some of us experience food insecurity or
hunger, it harms and diminishes us all.
Food Insecurity and Hunger
“Food security—access by all people at all times to
enough food for an active, healthy life—is one of several
conditions necessary for a population to be healthy and
well nourished.”11 Food insecurity and hunger are mea-
sured in the US with a household survey administered
each December by the U.S. Census Bureau. The U.S.
Food Security Survey Module and the Food Security
Scales it contains were developed in the 1990s under
the Food Security Measurement Study, a multi-agency
collaborative effort involving scientists and academics,
government analysts and policy experts, and individuals
from for-prot and not-for-prot private entities.6 The
primary food security scale development activities were
implemented through a competitive contracting process
sponsored and overseen by the USDA and the National
Center for Health Statistics (NCHS), with Abt Associ-
ates, Inc. as the prime contractor.
The food security and hunger scales developed by the
Abt team were incorporated into the ongoing national
Current Population Survey (CPS) implemented by the
Census Bureau annually. Data from administration of
the scales in the CPS are delivered by the Census Bureau
to the USDA Economic Research Service (ERS) for
summary analysis, estimation of prevalence in different
socio-demographic subgroups, tabulation and reporting
in its annual reports on food security in the US.
A Note on Causality
Establishing causation is correctly the ideal of all sci-
entic endeavor, but it is seldom achieved, especially in
the health and social sciences. The experimental design
considered by most scientists, and many non-scientists,
to be the “gold standard” for determining causality is
the randomized controlled trial or “RCT,” in which
randomization can “control for” unobserved potentially
confounding factors that might lead researchers to erro-
neously infer causation in relationships, by rendering
those confounders random in the studied samples. Yet
as good as they are, RCTs are not perfect, nor are they
immune from various kinds of error.12
Moreover, many of the phenomena and conditions of
interest in both health sciences and social sciences are
not amenable to randomization. It would be unethical,
for example, to randomly assign subjects to conditions
of food insecurity or hunger, or to randomly assign
food-insecure households to receive or not receive
food assistance or other interventions. Consequently,
food security research almost always relies on creative
quasi-experimental designs, and efforts to control for
unobserved confounders statistically.
Thus, conclusive, unassailable evidence that food
insecurity causes the multitude of illnesses and adverse
health conditions that a very large body of research liter-

WWW.HUNGERREPORT.ORG • 2016 HUNGER REPORT 253252 APPENDIX 2 • BREAD FOR THE WORLD INSTITUTE
ature indicates it is strongly related to most likely cannot
be produced. Yet, as with the relationships between
smoking tobacco and lung, throat, and mouth cancers,
the evidence of relationships between food insecurity and
these health outcomes is so strong, and the expected con-
sequences of not treating the relationships as causal are
so grave that we are justied in acting on strong evidence
even if it is not absolutely conclusive and unassailable.
A Groundbreaking Study Helps Provide A Path
Forward
An extremely important recent study of the relation-
ships between food insecurity and health care costs in
Ontario, Canada, where health insurance is univer-
sally available, achieves a major breakthrough toward
providing conclusive evidence of causal relationships
between food insecurity and adverse health outcomes.
Since health insurance is universally available in
Ontario, the intractable obstacle of adverse selection
bias is virtually eliminated in this study. Successfully
merging administrative data on health services utiliza-
tion and costs in Ontario with data on food security
status of Ontario households from the Canadian Com-
munity Health Survey, the researchers come closer
than any yet to demonstrating that food insecurity
causes bad health outcomes.
Results from this path-breaking research show
a monotonic dose-response relationship between
severity of food insecurity and total health care costs
per person, after adjusting for a number of potential
confounders known to be social determinants of health,
even after excluding prescription drug costs which are
only covered for a subset of the population.13 Moreover,
food insecurity was strongly and signicantly related to
healthcare costs, whereas income quintile of patients’
neighborhood was not.13
While this study does not connect food insecurity
causally with specic diseases, results are described as
consistent with ndings from other research of strong
associations between food insecurity and poorer self-
reported health status, increased likelihood of chronic
disease diagnoses, poorer management of disease, and
increased healthcare costs. The study’s authors also
note that “the extreme levels of material deprivation
associated with household food insecurity, and severe
food insecurity in particular, have been associated with
extensive dietary compromise, higher levels of stress,
and compromises across a broad spectrum of basic
needs, all of which diminish individuals’ abilities to
manage health problems and potentially increase the
need for health care.13
So while the presence of causal relationships between
food insecurity and specic diseases and adverse health
outcomes remains to be conclusively established, this
study comes closer than any previous research to estab-
lishing conclusive causal relationships between food
insecurity and higher health services utilization and
health related costs. It is, therefore, a breakthrough,
and provides strong support for the cost estimates pro-
duced in this current study.
Updating the October 2011 Hunger in America
Cost Estimates
In October 2011, researchers at Brandeis Univer-
sity published a set of estimates of national-level costs
Exhibit 3 Estimated costs of food insecurity and hunger in the US, 2007 and 2010.
Source: Recreated from Shepard, et al., 20111.
2007
($Billions)
2010
($Billions)
Amount of
Change, 2007-
2010 ($Billions)
Percent Change,
2007-2010
Illness Costs $98.4 $130.5 $32.1 33%
Education and Related Costs $13.9 $19.2 $5.3 38%
Charity Costs $13.2 $17.8 $4.6 35%
Total Hunger Bill $125.5 $167.5 $42.0 33%

WWW.HUNGERREPORT.ORG • 2016 HUNGER REPORT 253252 APPENDIX 2 • BREAD FOR THE WORLD INSTITUTE
attributable to food insecurity and hunger in 2010.1
Those estimates (Exhibit 3) comprised an update of an
earlier set published in 2007.14 The authors concluded
that costs attributable to food insecurity and hunger in
2010 conservatively amounted to a total of $167.5 Bil-
lion spread over illness-related costs, education-related
costs, and charity costs (Exhibit 3). The costs estimates
produced for 2010 ranged from 33 percent to 38 percent
higher than the 2007 estimates across these categories.
As described in the remainder of this section, there is
little evidence that economic conditions in 2014 were
sufciently better than those in 2010 to suggest signi-
cant reductions in the costs attributable to food security
over that period.
Over the period 2007-2010, food insecurity increased
dramatically, mainly due to the Great Recession and the
massive increases in unemployment during the recession
and after it ofcially ended (Exhibit 4). In Exhibit 4, the
red vertical arrow indicates the month the Great Reces-
sion began (December 2007), and the green vertical arrow
the month it was determined by the National Bureau of
Economic Research (NBER) Business Cycle Dating Com-
mittee to have ended (June 2009). The horizontal blue
arrow marks the level of unemployment in the month
before the recession began (November 2007). As the
chart shows, the number unemployed in January 2013
was above 12.3 million, but declined steadily throughout
the year, ending at just over 10.3 million. However, more
than six years after the end of the recession (July 2015), the
number of unemployed people in the U.S. labor force had
not returned to its pre-recession level.
In July 2015 there were still more than a million more
unemployed workers than in the month prior to the start
of the recession (November 2007). Unemployment more
than doubled during the recession, going from 7.24 mil-
lion in November 2007 to 14.71 million in June 2009,
the month the recession ended. And it continued to
increase, surpassing 15 million in September 2009 and
Exhibit 4 Number of unemployed workers in the US labor force by month, from January 2007
through July 2015.
Source: US Bureau of Labor Statistics (http://data.bls.gov/pdq/SurveyOutputServlet;jsessionid=AE49BA7CEF85EEB690DE95D4FC5D758F.tc_instance5).
0
2000
4000
6000
8000
10000
12000
14000
16000
18000
Jan-07 Sep-07 May-08 Jan-09 Sep-09 May-10 Jan-11 Sep-11 May-12 Jan-13 Sep-13 May-14 Jan-15

WWW.HUNGERREPORT.ORG • 2016 HUNGER REPORT 255254 APPENDIX 2 • BREAD FOR THE WORLD INSTITUTE
staying above 15 million until May 2010. The recovery of
jobs since the recession ended has been extraordinarily
slow, with ups and downs as Exhibit 4 shows.
Among the most harmful aspects of the very high
unemployment levels during and after the Great Reces-
sion was the unparalleled expansion of the number
of long-term unemployed, workers who had been
unemployed for 27 weeks or longer. The number of
long-term unemployed reached a record high of 6.7
million, 45.1 percent of all the unemployed in the
second quarter of 2010. In addition, the proportion of
unemployed workers who had been unemployed for 52
weeks or longer reached a record high of 31.9 percent
in the second quarter of 2011, and the proportion who
had been unemployed for 99 weeks or longer reached
a record high of 15.1 percent in the fourth quarter of
2011.15 And while all three of these measures of long-
term employment have declined over the past several
years, they remain high by historical standards.
Another extraordinary characteristic of the very slow
job recovery from the Great Recession has been the large
numbers of people withdrawing from the labor force;
some for non-economic reasons, but others because they
could not nd suitable work, or any work at all. Between
the end of the recession in June 2009, and December
2010, nearly 6 million people (5.999 million) withdrew
from the labor force. By the end of 2013, an additional
6.6 million had withdrawn. Workers have continued to
withdraw from the labor force since the end of 2013, but
the rates of withdrawal have slowed and been nearly
offset by new entrants. Even so, in July 2015, there were
12.6 million more workers not in the labor force than
when the recession ended in June 2009.16
Among the 12.6 million people who withdrew from
the labor force since the recession ended, nearly half
chose to attend or return to school, or to engage in
other non-labor force activities voluntarily. However,
just over half reported they were available to work and
wanted a job, but were not nding any. In addition
to these labor-force leavers, the number of so-called
“discouraged workers,” who had looked for work some-
time within the past year, but recently stopped looking
because they believed there were no jobs available for
them, went from 363,000 to 793,000 during the reces-
sion, and reached 1.318 million by December 2010. The
number of “discouraged workers” remained close to
1.0 million over 2012-2014, but had declined to 668,000
by July 2015, still nearly double the number when the
recession began.
In addition to the very large increases in numbers
of unemployed, long-term unemployed, and those who
withdrew from the labor force for economic reasons,
the Great Recession also led to major increases in the
number of “involuntary part time workers,” people who
wanted to be working full time but were only able to nd
part-time work. From November 2007, the month before
the recession began, to when it ended in June 2009, the
number of involuntary part-time workers doubled,16
increasing from 4.494 million to 9.024 million. And
as with unemployment, this number remained little
changed through December 2010 when it was 8.935
million. By the end of 2013 the number of involuntary
part time workers had fallen to 7.776 million, and in July
2015, at 6.325 million it was still 41 percent higher than
in the month before the recession began.16
Thus in terms of labor market conditions, the unprec-
edented high levels of unemployment during and fol-
lowing the Great Recession have slowly declined over
the past six years, but labor markets and the employ-
ment situation has by no means returned to normal,
unless this is the “new normal.” While the number of
unemployed per month over the period January 2008
to December 2010 averaged 12.683 million workers,
during the period January 2011 to December 2013,
most of the period over which we are updating the
estimates of costs attributable to food insecurity and
hunger (indicated by the black vertical arrow in Exhibit
4), the average number of unemployed each month was
12.563 million, less than 1.0 percent lower (0.95 percent)
than the average over 2008-2010. Thus on the basis of
unemployment, under-employment, long-term unem-
ployment, labor force withdrawals, and other labor force
conditions, there is no reason to expect food insecurity,
or its costs, to be signicantly lower in 2014 than in 2010,
and several reasons to expect them to be higher.
While the recovery has been very robust in terms
of growth in GDP and corporate prots, with GDP
growing at an average annual rate of 3.28 percent, and

WWW.HUNGERREPORT.ORG • 2016 HUNGER REPORT 255254 APPENDIX 2 • BREAD FOR THE WORLD INSTITUTE
corporate prots increasing by an average of nearly
10 percent per year over the period 2010-2014 in the
non-nancial sector of the economy (which includes
manufacturing, transportation, utilities, wholesale and
retail trade, and information), average weekly earn-
ings for workers in private non-agricultural industries
only increased in real (ination-adjusted) terms over
that period, by an average of 0.08 percent per year.
The unavoidable implication of these numbers is that
many people who have been able to nd jobs during
the recovery are earning less and less in real, ination-
adjusted terms, while corporate prots have increased
at unprecedented rates.17 These stagnant weekly earn-
ings resulted in median annual income levels in real
2014 dollars for households declining from 2007-2010
by -6.7 percent. And while median income levels did
not decline further from 2010-2014, they only increased
by 0.28 percent, i.e., by less than three tenths of a per-
centage point in real 2014 dollars over the ve years. It
is worth noting that these trends in real average weekly
earnings and real median income are unprecedented in
the history of the U.S. economy since the Great Depres-
sion ended.
The unprecedented increase in food insecurity
during the rst year of the Great Recession is apparent
in the data on food insecurity levels and prevalence
in Exhibit 5, as is the persistence of high prevalence
of all levels of severity of household food insecurity
throughout the period 2008-2010, as well as 2011-
2014. The economic context underlying the dramatic
increases in food insecurity prevalence at all levels
of severity was characterized primarily by massive
increases in job losses and unemployment.* The eco-
nomic context underlying the persistence of resulting
*The bursting of the housing bubble and collapse of the nancial institutions whose unfettered speculative gambling with contrived “bundled instruments” of
questionable legality was responsible for the subprime mortgage debacle, and ultimately for both the housing bubble and its bursting, led to unprecedented losses
of wealth held in the form of owner-occupied residential real estate. That huge loss of wealth together with the large debt loads many homeowners had accumulated
through “equity lines of credit” supported by the homes whose mortgages they were no longer able to afford, and the massive devaluation of residential real estate
that followed bursting of the bubble, all contributed to the complex, multi-faceted market failures accompanying the nancial market collapse. And all these market
failures worked to shut down activities that had been employing millions of workers, thus playing a major role in initiation of the Great Recession. While the “too
big to fail” banks and other nancial institutions who were propped up and bailed out with public revenues quickly recovered and are among the corporations now
earning unprecedented prots, the millions of homeowners, and other people who lost their homes, their wealth and their jobs are still struggling to recover. And
they are among the millions of Americans still suffering from food insecurity. However, as relevant, interesting and important as this larger story is, its telling is
beyond the scope of this project.
Exhibit 5 Numbers and percents of people in the United States living in Food-Insecure households
by food security status of the household, 2007-2014.
Source: Coleman-Jensen, et al., 20152.
Year
Total Number
of Individuals
Food Insecure
(1000s)
Percent of
Individuals
Food Insecure
Number of
Individuals In
Households With
Low Food Security
(1000s)
Percent of
Individuals In
Households
With Low Food
Security
Number of
Individuals in
Households with
Very Low Food
Security (1000s)
Percent of
Individuals in
Households with
Very Low Food
Security
2007 36,229 12.2% 24,287 8.2% 11,942 4.0%
2008 49,108 16.4% 31,824 10.6% 17,284 5.8%
2009 50,162 16.6% 32,499 10.8% 17,663 5.9%
2010 48,832 16.1% 32,777 10.8% 16,055 5.3%
2011 50,120 16.4% 33,232 10.9% 16,888 5.5%
2012 48,966 15.9% 31,787 10.3% 17,179 5.6%
2013 49,078 15.8% 31,974 10.3% 17,104 5.5%
2014 48,135 15.4% 30,922 9.9% 17,213 5.5%

WWW.HUNGERREPORT.ORG • 2016 HUNGER REPORT 257256 APPENDIX 2 • BREAD FOR THE WORLD INSTITUTE
high prevalence of food insecurity in the years since the
recession ended was one of declining weekly earnings,
declining then stagnant real median income levels,
major increases in the numbers of people engaging in
involuntary part-time work, extraordinary numbers of
workers withdrawing from the labor force for economic
reasons, mainly because they could not nd jobs, and
the large increase and persistence of high numbers of
long-term unemployed and “discouraged workers” over
these two periods. Unfortunately there are few reasons
to expect these conditions to change for the better in
the near term.
The effects of these labor market dynamics on food
insecurity are depicted graphically in Exhibits 6 and
7. While the increase in household food insecurity was
rapid and extensive for adults and children, it was less
pronounced among people living in households with
elderly (Exhibit 6). However, while the number of food
insecure adults stabilized at its higher level over the
period 2010-2014, and the number of food-insecure chil-
dren declined slightly from its peak in 2009, the number
of food-insecure people in households with elderly con-
tinued to increase throughout the period 2010-2013,
offsetting the decline in the number of food-insecure
children. The net result of these subgroup changes was
a fairly stable plateau of the total number of people
living in food-insecure households at a level 12-14 mil-
lion higher than its pre-recession level. Most notably, in
spite of the supposed recovery from the recession, and
signicant declines in the total number of people unem-
ployed over the period 2010-2013, economic conditions
persisted that prevented food insecurity from declining.
Though the absolute numbers are comparatively
smaller, the number of people living in households
with very low food security, or severe food insecurity
(previously food insecurity with hunger), increased in
a pattern very similar to low food security (Exhibit 6).
A notable difference between the trends in low food
Exhibit 6 Numbers of people in the United States living in food-insecure households by age group,
2000-2014.
Source: Coleman-Jensen, et al., 20152. (People in households with elderly can be of any age.)
0
10,000
20,000
30,000
40,000
50,000
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
All People Adults Children People in Hhlds with Elderly
(1,000s)

WWW.HUNGERREPORT.ORG • 2016 HUNGER REPORT 257256 APPENDIX 2 • BREAD FOR THE WORLD INSTITUTE
security (Exhibit 6) and those for very low food secu-
rity (Exhibit 7) is that the prevalence of very low food
security had been on an upward trajectory since 2000,
especially among adults, but also to a lesser degree
among children.
The fall in prevalence of very low food security
over 2009-2010 (Exhibit 7) partially reects the across
the board 13 percent increase in SNAP (Supplemental
Nutrition Assistance Program) benets and enhanced
eligibility for single adults who had lost jobs, instituted
under the American Recovery and Reinvestment Act
(ARRA).18 SNAP is the largest federal food assistance
program, and also an entitlement program, making it
the most important “counter-cyclical” support program
the United States has. Since it is an entitlement, SNAP
must be provided to all eligible applicants. Therefore in
economic downturns that occur periodically as part of
the usual business cycle, when jobs are lost and unem-
ployment increases, more families and individuals
become eligible for SNAP, and SNAP enrollment
increases. When a recovery gets underway and jobs are
created, unemployment falls, and the number of fami-
lies eligible for SNAP, and SNAP enrollment decline.
That makes this food assistance program the only real
counter-cyclical program in the United States. Relative
to low food security, very low food security appears to
have responded more noticeably to the higher SNAP
benet levels.
The persistence of high levels of food insecurity into
2014 is thus largely due to underlying weakness in the
recovery from the Great Recession of 2007-2009, espe-
cially the extraordinarily slow recovery of jobs in the
economy. It is also the result of changes in the structure
of labor markets, work, and job stability. Emergence of
“contingent labor,” companies ability and willingness
to rely on contract labor and temporary jobs that do
not provide benets, and to adjust their demand for
labor practically in real time by notifying workers on
Exhibit 7 Numbers of people in the United States living in households with very low food insecurity
on the adult or household scale, 2000-2013.
Source: Coleman-Jensen, et al., 201411. (People in households with elderly can be of any age.)
0
2,000
4,000
6,000
8,000
10,000
12,000
14,000
16,000
18,000
20,000
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
All People Adults Children People in Hhlds with Elderly
(1,000s)

WWW.HUNGERREPORT.ORG • 2016 HUNGER REPORT 259258 APPENDIX 2 • BREAD FOR THE WORLD INSTITUTE
a daily basis as to whether they are needed, all have
made work, earnings, and income less stable. Volatility
in earnings for wage workers may be the “new normal,”
and its effects can be seen in persistent poverty and
food insecurity (Exhibit 8).
Effects of efforts to reduce or eliminate SNAP
benets, and other social infrastructure that provide
support for U.S. working families are likely reected in
the reductions in both the number of people receiving
SNAP and the average SNAP benets per person
from 2013 to 2014 (Exhibit 9). These declines in SNAP
benets and participation are, in turn, likely a factor
in the persistence of high food insecurity levels from
2013 to 2014.
Conclusion
Food insecurity in the US was at an unacceptably
high level in 2010, and remained so through 2014.
The costs attributable to food insecurity are also unac-
ceptably high. The extraordinarily slow recovery of
employment from the Great Recession is a key factor in
persistent food insecurity in the United States, however
changes in labor market structures and practices also
play a role.
The health-related costs associated with food insecu-
rity are clearly high. Though we estimated costs related
to several disease conditions that are plausibly attribut-
able to food insecurity, there are others that we did not
nd sufcient evidence to estimate. What is clear is that
the health-related costs of food insecurity and hunger
are high, and are likely to increase unless addressed.
The Affordable Care Act has provided several windows
of opportunity for the healthcare system to engage with
and contribute to viable solutions to food insecurity
and hunger, and these need to be implemented and
supported.
The public and private social infrastructures that
have emerged in response to food insecurity and
hunger in the United States have very large associated
costs, but it is important to acknowledge that both the
public and private food assistance systems meet mul-
tiple objectives, some of which are not directly related
to reducing food insecurity. SNAP is our largest and
Exhibit 8
Numbers of people in the United States living in food-insecure households by age group, with
the numbers of all people and children in households with incomes below poverty, 2000-2013
*Though data on poverty in the US in 2014 will be released by the Census Bureau later this month, they are currently only available through 2013.
0
10,000
20,000
30,000
40,000
50,000
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013
All People Adults Children
People in Hhlds with Elderly All People in Poverty All Children in Poverty

WWW.HUNGERREPORT.ORG • 2016 HUNGER REPORT 259258 APPENDIX 2 • BREAD FOR THE WORLD INSTITUTE
most effective counter-cyclical program to offset the
inevitable downturns in economic activity and avail-
ability of jobs that is systemically built into the U.S.
economy. WIC provides nutrition education and
medical services in addition to food targeted speci-
cally to pregnant and lactating mothers, and infants
and children.
In addition to providing much needed food and
other services for low-income and food-insecure fami-
lies and individuals, the private food assistance system
also provides opportunities for corporations to remove
unprotable product from their inventories, reduce
their tax burdens, and improve public perceptions of
their degree of social responsibility. In addition, both
the public and private food assistance systems provide
much-needed jobs, many of which pay very well.
It is also extremely important to note that the public
and private food assistance systems comprise comple-
mentary systems for dealing with food insecurity and
hunger, with overlap and interaction between the two
systems. And it is necessary to state the obvious fact
that the two systems combined are still far from ade-
quate solutions to the problems of food insecurity and
hunger. Food insecurity and hunger, like poverty, their
main proximal cause, are systemic problems that result
from numerous market, policy, and leadership failures.
And they will not be eliminated until those systemic
failures are acknowledged, addressed, and resolved.
Exhibit 9 Average monthly number of SNAP participants, and average monthly per person benefit
level, 2000-2014.
Source: USDA Food and Nutrition Service; SNAP program data (http://www.fns.usda.gov/pd/supplemental-nutrition-assistance-program-snap)
$0
$20
$40
$60
$80
$100
$120
$140
$160
0
10,000
20,000
30,000
40,000
50,000
60,000
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014
Thousands
(Average Monthly
Participation; Bars,
Lft Axis)
(Average Monthly Benefit
per Person; Line, Rt Axis)

WWW.HUNGERREPORT.ORG • 2016 HUNGER REPORT 261260 APPENDIX 2 • BREAD FOR THE WORLD INSTITUTE
References
1. Shepard D, Setren E, Cooper D. Hunger in America, Suffering We All Pay ForCenter for American Progress;
2011.
2. Coleman-Jensen A, Rabbitt M, Gregory C, Singh A. Household Food Security in the United States in 2014, ERR-
194: U.S. Department of Agriculture, Economic Research Service; September 2015.
3. Zhang J, Yu KF. What’s the relative risk? A method of correcting the odds ratio in cohort studies of common
outcomes. JAMA 1998;280(19):1690-1.
4. Hennekens C, Buring J. Epidemiology in Medicine. Boston/Toronto: Little Brown & Company; 1987.
5. O’Hara B, Caswell K. Health Status, Health Insurance, and Medical Services Utilization: 2010: US Census
Bureau;July 2013.
6. Hamilton W, Cook J, Thompson W, Buron L, Frongillo Jr E, Olson C, et al. Household Food Security in the
United States in 1995: Summary Report of the Food Security Measurement Project. In. Alexandria, VA; 1997.
7. Maslow A. Atheory of human motivation. Psychological Review 1943;50:26.
8. Bourre JM. Effects of nutrients (in food) on the structure and function of the nervous system: update on dietary
requirements for brain. Part 1: micronutrients. J Nutr Health Aging 2006;10(5):377-85.
9. Rice D, Barone S. Critical periods of vulnerability for the developing nervous system: evidence from humans and
animal models. Environ Health Perspect 2000;108 Suppl 3:511-33.
10. Shonkoff J, Phillips D. From Neurons to Neighborhoods: The Science of Early Childhood Development.; Avail-
able from: http://www.nap.edu/openbook.php?record_id=9824&page=R1
11. Coleman-Jensen A, Gregory C, Singh A. Household Food Security in the United States in 2013: Statistical Supple-
ment: U.S. Department of Agriculture, Economic Research Service2014.
12. Grossman J, Mackenzie FJ. The randomized controlled trial: gold standard, or merely standard? Perspect Biol
Med 2005;48(4):516-34.
13. Tarasuk V, Cheng J, de Oliveira C, Dachner N, Gundersen C, Kurdyak P. Association between household food
insecurity and annual health care costs. CMAJ 2015.
14. Brown J, Shepard D, Martin T, Orwat J. The Economic Cost of Domestic Hunger;Estimated Annual Burden
to the United States: An analysis commissioned by The Sodexho Foundation, in partnership with The Public
Welfare Foundation and Spunk Fund, Inc2007.
15. Kosanovich K, Sherman E. Trends in Long-term Unemployment: US Bureau of Labor Statistics; 2015.
16. US Bureau of Labor Statistics. Available from: http://data.bls.gov/pdq/SurveyOutputServletjsessionid=AE49BA
7CEF85EEB690DE95D4FC5D758F.tc_instance5.
17. Economic Report of the President. 2015: US Government Publishing Ofce.
18. Nord M, Prell M. Food Security Improved Following the 2009 ARRA Increase in SNAP Benets: U.S. Depart-
ment of Agriculture, Economic Research Service; 2011.
19. Cook JT, Frank DA, Berkowitz C, Black MM, Casey PH, Cutts DB, et al. Food Insecurity is associated with
adverse health outcomes among human infants and toddlers. The Journal of Nutrition 2004;134(6):1432-1438.
20. Rose-Jacobs R, Black MM, Casey PH, Cook JT, Cutts DB, Chilton M, et al. Household Food Insecurity: Associa-
tions with at-risk infant and toddler development. Pediatrics 2008;121(1):65-72.

WWW.HUNGERREPORT.ORG • 2016 HUNGER REPORT 261260 APPENDIX 2 • BREAD FOR THE WORLD INSTITUTE
21. Cook JT, Black M, Chilton M, Cutts D, Ettinger de Cuba S, Heeren TC, et al. Are food insecurity’s health
impacts underestimated in the U.S. population? Marginal food security also predicts adverse health outcomes in
young U.S. children and mothers. Adv Nutr 2013;4(1):51-61.
22. Skalicky A, Meyers AF, Adams WG, Yang Z, Cook JT, Frank DA. Child food insecurity and iron deciency anemia
in low-income infants and toddlers in the United States. Maternal and Child Health Journal 2006;10(2):177-185.
23. Black MM, Quigg AM, Cook J, Casey PH, Cutts DB, Chilton M, et al. WIC participation and attenuation of
stress-related child health risks of household food insecurity and caregiver depressive symptoms. Arch Pediatr
Adolesc Med 2012;166(5):444-51.
24. Leung CW, Epel ES, Willett WC, Rimm EB, Laraia BA. Household food insecurity is positively associated with
depression among low-income supplemental nutrition assistance program participants and income-eligible non-
participants. J Nutr 2015;145(3):622-7.
25. Ma CT, Gee L, Kushel MB. Associations between housing instability and food insecurity with health care access
in low-income children. Ambul Pediatr 2008;8(1):50-7.
26. Park CY, Eicher-Miller HA. Iron deciency is associated with food insecurity in pregnant females in the United
States: National Health and Nutrition Examination Survey 1999-2010. J Acad Nutr Diet 2014;114(12):1967-73.
27. Parks CG, D’Aloisio AA, DeRoo LA, Huiber K, Rider LG, Miller FW, et al. Childhood socioeconomic fac-
tors and perinatal characteristics inuence development of rheumatoid arthritis in adulthood. Ann Rheum Dis
2013;72(3):350-6.
28. Tayie FA, Zizza CA. Food insecurity and dyslipidemia among adults in the United States. Prev Med
2009;48(5):480-5.
29. Seligman HK, Bindman AB, Vittinghoff E, Kanaya AM, Kushel MB. Food insecurity is associated with diabetes
mellitus: results from the National Health Examination and Nutrition Examination Survey (NHANES) 1999-
2002. J Gen Intern Med 2007;22(7):1018-23.
30. Berkowitz SA, Gao X, Tucker KL. Food-insecure dietary patterns are associated with poor longitudinal glycemic
control in diabetes: results from the Boston Puerto Rican Health study. Diabetes Care 2014;37(9):2587-92.
31. Beydoun MA, Wang Y. Pathways linking socioeconomic status to obesity through depression and lifestyle factors
among young US adults. J Affect Disord 2010;123(1-3):52-63.
32. Carmichael SL, Yang W, Herring A, Abrams B, Shaw GM. Maternal food insecurity is associated with increased
risk of certain birth defects. J Nutr 2007;137(9):2087-92.
33. Chi DL, Masterson EE, Carle AC, Mancl LA, Coldwell SE. Socioeconomic status, food security, and dental caries
in US children: mediation analyses of data from the National Health and Nutrition Examination Survey, 2007-
2008. Am J Public Health 2014;104(5):860-4.
34. Fitzgerald N, Hromi-Fiedler A, Segura-Pérez S, Pérez-Escamilla R. Food insecurity is related to increased risk of
type 2 diabetes among Latinas. Ethn Dis 2011;21(3):328-34.
35. Whitaker RC, Phillips SM, Orzol SM. Food insecurity and the risks of depression and anxiety in mothers and
behavior problems in their preschool-aged children. Pediatrics 2006;118(3):e859-e868.
36. Laraia BA, Siega-Riz AM, Gundersen C. Household food insecurity is associated with self-reported pregravid
weight status, gestational weight gain, and pregnancy complications. J Am Diet Assoc 2010;110(5):692-701.
37. Berkowitz SA, Baggett TP, Wexler DJ, Huskey KW, Wee CC. Food insecurity and metabolic control among U.S.
adults with diabetes. Diabetes Care 2013;36(10):3093-9.

WWW.HUNGERREPORT.ORG • 2016 HUNGER REPORT 263262 APPENDIX 2 • BREAD FOR THE WORLD INSTITUTE
Exhibit A1
Health conditions for which information was available to calculate population attributable
fractions indicating the proportion of cases in the population attributable to food insecurity.
Relationship AOR* RR* AF* Source
1) HFI & Child non-perinatal hospitalization (yes-no): 1.31 1.23 4.55% Cook, et al., J Nut, 200419
2) HHLD FI & Caregivers’ report of child health status fair/poor: 1.90 1.73 12.47% Cook, et al., J Nut, 200419
3) HFI & Caregivers’ report of PEDS 1 concerns: 1.76 1.60 10.87% Rose-Jacobs, et al., Peds,
200820
4) HHLD FI & Caregivers’ report of PEDS 2 concerns: 1.46 1.43 9.09% Cook, et al., Adv Nut, 201321
5) CFI & Iron deficiency Anemia: 2.40 2.01 8.25% Skalicky, et al., J MCH, 200622
6) HFI & Caregivers’ self-reported health status fair/poor: 2.28 1.91 6.81% Cook, et al., Adv Nut, 201321
7) HFI & Caregivers’ self report of Positive Depressive Symptoms: 3.06 2.28 10.96% Cook, et al., Adv Nut, 201321
8) HFI + PDS & Caregivers’ report of child health status fair/poor: 2.45 2.12 8.45% Black, et al., Arch Ped Adoles
Med, 201223
9) HFI + PDS & Child non-perinatal hospitalization (yes-no): 1.35 1.25 2.10% Black, et al., Arch Ped Adoles
Med, 201223
10) HFI + PDS & Caregivers’ report of PEDS 1. 2.49 2.26 9.83% Black, et al., Arch Ped Adoles
Med, 201223
11) HVLFS % Adults’ Depression 3.42 2.97 31.69% Leung, et al., J Nutr, 201524
12) FI (based on subset of 4 of the 18 USFSSM questions) & failure of
children, 3-5 yrs & 11-17 yrs, to receive recommended well-child
visits (postponed recommended care)
1.40 1.09 7.44% Ma, et al., Ambul Pediatr,
200825
13) FI (based on subset of 4 of the 18 USFSSM questions) & failure of
children, 3-5 yrs & 11-17 yrs, to receive needed health care (fore-
gone needed care)
1.61 1.58 17.66% Ma, et al., Ambul Pediatr,
200825
14) FI (based on subset of 4 of the 18 USFSSM questions) & failure
of children, 3-5 yrs & 11-17 yrs, to receive prescribed medication
(foregone needed care)
2.48 2.42 34.07% Ma, et al., Ambul Pediatr,
200825
15) FI and iron deficiency in pregnant women ages 13-54 yrs, based
on Ferritin <12 ug/L reported in a 24 hr dietary recall and a 30-day
supplement question; NHANES 1999-2010.
2.9 2.05 12.90% Park; Eicher-Miller J Acad Nutr
Diet, 201426
16) FI, based on 1 ad lib question; “When you were growing up, were
there times your family didn’t have enough to eat?”, and Rheuma-
toid arthritis (self-reported with any current or past DMARD (disease
modifying antirheumatic drugs) use and bilateral swelling, or steroid
use and bilateral swelling, in the absence of another autoimmune
disease), in women 35-74 yrs old.
1.50 1.49 4.33% Parks, et al., Ann Rheum Dis,
201327
17) MFS & LDL cholesterol in males & females 18-50 yrs; NHANES
1999-2002
1.85 1.30 3.68% Tayie; Zizza Prev Med, 200928
18) MFS & TRG/HDL ratio in males & females 35-50 yrs; NHANES
1999-2002
1.98 1.33 4.05% Tayie; Zizza Prev Med, 200928
19) H LFS & Triglycerides in males & females 35-50 yrs; NHANES 1999-
2002
1.91 1.31 3.64% Tayie; Zizza Prev Med, 200928
20) H Severe FI (6-10 Adult Scale items affirmed) & Diabetes in Adults
ages >20 yrs, NHANES 1999-2002.
2.20 1.89 7.89% Seligman, et al., J Gen Inter
Med, 200729
21) HFI & poor Diabetes Control in adults ages >21 yrs w DM, from
clinics in Boston.
1.97 1.40 5.00% Berkowitz, et al, Diabetes Care,
201430
22) FI w/o Hunger (HLFS) & Major Depressive Disorder in Women
20-39 yrs old in a subsample of NHANES 1999-2004 receiving MDD
measurement.
2.76 2.43 10.32% Beydoun; Wang J Affect
Disord, 201031

WWW.HUNGERREPORT.ORG • 2016 HUNGER REPORT 263262 APPENDIX 2 • BREAD FOR THE WORLD INSTITUTE
*Abbreviations: AOR=Adjusted Odds Ratio; CFI=Child food insecurity; DMARD=Disease modifying antirheumatic drugs; DM=Diabetes mellitus; FI=Food insecurity; HDL=High-
density lipoprotein; GAD=Generalized anxiety disorder; HFI=Household food insecurity; HVLFS=Household very low food security; LDL=Low-density lipoprotein; LFS=Low
food security; MDD=Major depressive disorder; MDE=Major depressive episode; MFS=Marginal food security; NHANES=National Health and Nutrition Examination Survey;
NTD=Neural tube defects; PAF=Population attributable fraction; PEDS=Parents’ evaluation of developmental status; PDS=Positive depression screen; RR=Relative risk;
SES=Socio-economic status; T2D=Type two diabetes; TRG=Triglycerides; USFSSM=US Food Security Survey Module; VLFS=Very low food security.
Relationship AOR* RR* AF* Source
23) HFI & Birth Defects (NTD, Orofacial Clefts, Conotruncal Heart
Defects) in newborns.
1.41 1.12 1.11% Carmichael, et al., J Nutr,
200732
24) HFI, SES, & Dental Caries in Children 5-17 yrs in the NHANES,
2007-2008.
2.51 2.01 15.34% Chi, et al., Am J Public Health,
201433
25) VLFS & T2D in Latina Women, 35-60 yrs old 3.33 1.61 7.79% Fitzgerald, et al., Ethn Dis,
201134
26) MFS & MDE in Mothers age >18 yrs in the Fragile Families data,
1998-2000.
1.40 1.32 5.53% Whitaker, et al., Pediatrics,
200635
27) FI & MDE in Mothers age >18 yrs in the Fragile Families data, 1998-
2000.
2.20 1.88 9.10% Whitaker, et al., Pediatrics,
200635
28) MFS & GAD in Mothers age >18 yrs in the Fragile Families data,
1998-2000.
1.70 1.66 11.13% Whitaker, et al., Pediatrics,
200635
29) FI & GAD in Mothers age >18 yrs in the Fragile Families data, 1998-
2000.
2.30 2.20 13.93% Whitaker, et al., Pediatrics,
200635
30) MFS & Either MDE or GAD in Mothers age >18 yrs in the Fragile
Families data, 1998-2000.
1.40 1.32 5.46% Whitaker, et al., Pediatrics,
200635
31) FI & Either DME or GAD in Mothers age >18 yrs in the Fragile Fami-
lies data, 1998-2000.
2.20 1.86 8.70% Whitaker, et al., Pediatrics,
200635
32) MFS & Aggression in 3-yr-old Children of Mothers age >18 yrs in
the Fragile Families data, 1998-2000.
1.50 1.45 7.53% Whitaker, et al., Pediatrics,
200635
33) FI & Aggression in 3-yr-old Children of Mothers age >18 yrs in the
Fragile Families data, 1998-2000.
1.90 1.68 8.11% Whitaker, et al., Pediatrics,
200635
34) MFS & Anxiety/Depression in 3-yr-old Children of Mothers age >18
yrs in the Fragile Families data, 1998-2000.
1.80 1.68 10.75% Whitaker, et al., Pediatrics,
200635
35) FI & Anxiety/Depression in 3-yr-old Children of Mothers age >18 yrs
in the Fragile Families data, 1998-2000.
2.20 1.99 10.97% Whitaker, et al., Pediatrics,
200635
36) MFS & Inattention/Hyperactivity in 3-yr-old Children of Mothers age
>18 yrs in the Fragile Families data, 1998-2000.
1.60 1.53 8.89% Whitaker, et al., Pediatrics,
200635
37) FI & Inattention/Hyperactivity in 3-yr-old Children of Mothers age
>18 yrs in the Fragile Families data, 1998-2000.
1.90 1.77 9.29% Whitaker, et al., Pediatrics,
200635
38) MFS & Any of the Three Behavior Problems in 3-yr-old Children of
Mothers age >18 yrs in the Fragile Families data, 1998-2000.
1.60 1.45 7.12% Whitaker, et al., Pediatrics,
200635
39) FI & Any of the Three Behavior Problems in 3-yr-old Children of
Mothers age >18 yrs in the Fragile Families data, 1998-2000.
2.10 1.77 8.01% Whitaker, et al., Pediatrics,
200635
40) FI & Poor Glycemic Control in Adult Diabetics in the Immigration,
Culture & Healthcare Study, San Francisco, CA, 2008-2009.
1.46 1.27 10.17% Seligman, et al., J Gen Inter
Med, 200729
41) FI & severe obesity in pregnant women ≤400% poverty level in the
Pregnancy, Infection, and Nutrition (PIN) cohort in NC, 2001-2005.
2.97 2.07 7.17% Laraia, et al, J Am Diet Assoc,
201036
42) HFI and poor glycemic control among diabetics ≥20 yrs old in the
NHANES 1999-2008.
1.53 1.42 4.16% Berkowitz, et al., Diabetes
Care, 201337
43) HFI and poor LDL control among diabetics ≥20 yrs old in the
NHANES 1999-2008.
1.86 1.32 2.37% Berkowitz, et al., Diabetes
Care, 201337

WWW.HUNGERREPORT.ORG • 2016 HUNGER REPORT PB264 APPENDIX 2 • BREAD FOR THE WORLD INSTITUTE
Exhibit A2 Detailed description of costs attributable to food insecurity by condition
Sources of Costs, 2014 Report
Costs Based on New
Evidence ($Billions
2014 Dollars)
Types of Costs, 2010
Report
Costs From 2010
Report ($Billion
2010 Dollars)
Costs From 2010 Re-
port Inflated to 2014
Dollars (% Change in
CPI-U for medical care,
1010-2014=9.674%)
TOTAL
Cost of additional non-neonatal
hospital stays among children ages
<18 years
$1.82 Hospitalizations $16.10 $17.66 (Estimate based on
new evidence was
used)
Cost of additional hospital stays
among adults ages 18+ years
$8.19
Cost of additional ambulatory visits
among people all ages
$1.51
Migraine $2.20 $2.41
Cost of additional dental care visits
among people all ages
$0.79
Colds $0.80 $0.88
Cost or treatment of mental health
problems in children ages <18
years
$1.22
Depression $29.20 $32.03
Cost of treatment of mental health
problems in adults ages 18-64
years
$4.75
Anxiety $17.40 $19.08
Cost of treatment of anemias and
other deficiencies in people all
ages
$0.85 Iron Deficiency $0.50 $0.55 (Estimate based on
new evidence was
used)
Suicide $19.70 $21.61
Treatment of osteoarthritis and
other inflammation in joints among
adults
$3.37
Upper GI Disorders $5.70 $6.25
Treatment of diabetes mellitus in
people all ages
$4.90
Health Status $38.90 $42.66
Treatment of hyperlipidemia $1.41
Treatment of endocrine system
problems related to poor control of
diabetes mellitus
$0.81
Treatment of congenital defects
and complications of pregnancy
and birth
$0.06
Indirect costs of lost work time
due to workers’ illnesses or work-
ers providing care for sick family
members
$5.48
TOTAL health costs $35.16 $124.92 $160.07
Expenditures for special education
in public primary and secondary
education
$5.91 Special Education $6.40 $7.02 (Estimate based on
new evidence was
used)
Dropout due to Reten-
tion
$6.00 $6.58
Dropout due to Absen-
teeism
$5.80 $6.36
TOTAL education & food
assistance
$5.91 $12.94 $18.85
TOTAL health, education & food
assistance
$178.92