food insecurity is linked to a food environment promoting obesity in households with children

5
Research Brief Food Insecurity Is Linked to a Food Environment Promoting Obesity in Households With Children Lisa M. Nackers, PhD, MPH; Bradley M. Appelhans, PhD ABSTRACT Objective: To determine the extent to which the presence and accessibility of healthful and less healthful foods in children’s homes vary with level of food security. Methods: A total of 41 parents or primary caregivers who had at least 1 child ages 2–13 and resided in a low-income area with limited food access completed a home food inventory and a validated measure assessing household food security. Results: Compared with food-secure participants, marginal or low/very low food-secure caregivers re- ported significantly more obesity-promoting foods in the home, more microwavable or quick-cook frozen foods, and greater access to less healthful foods in the kitchen (all Ps < .05). Conclusions and Implications: Given the greater presence and accessibility of less healthful foods, tar- geting home food environment may improve diet quality and health status in children of low-income, food insecure households. Key Words: food security, home environment, home food inventory, childhood obesity (J Nutr Educ Behav. 2013;45:780-784.) INTRODUCTION Food insecurity, or inaccessibility to nutritionally adequate foods owing to nancial or other resource limita- tions, 1 has emerged as an important public health concern. United States (US) Department of Agriculture esti- mates from 2011 indicate that 14.9% of households (17.9 million) qualied as food insecure, an increase from 11.9% in 1995. 2 Of households with children, 10% were characterized by child food insecurity in 2011. 2 Al- though food insecurity is inconsis- tently associated with child obesity risk, 3 evidence suggests that children in food-insecure households experi- ence poor diet quality, lower health status, decreased cognitive develop- ment, emotional and behavioral prob- lems, and subsequent accumulation of body fat in adulthood. 4 An association exists among food security, income, and diet quality, with lower food security and income associated with decreased healthful food intake and variety. 3,5 When assessing food-purchasing decisions in high-poverty neighborhoods, food- insecure families identied price as the most salient factor inuencing their purchases. 6 Nutrient-dense foods (eg, fruits, vegetables, whole grains, nonfat or low-fat milk, and lean meats) cost signicantly more per calorie than energy-dense foods (eg, soft drinks, salty and sugary snacks, pastries, and packaged and frozen foods). 5,7-9 In addition to price, low-income house- holds with children place greater im- portance on preparation convenience and shelf-life. 10 Consequently, food- insecure individuals more frequently consume sugar-sweetened beverages, fast food, and energy-dense foods, and report lower fruit and vegetable intake compared with those who expe- rience food security. 5,11 Home food availability serves as an important determinant of eating behavior and diet quality for food- insecure children. 12 The literature re- ports that food-insecure individuals consume poorer-quality foods and demonstrate difculty obtaining nutrient-dense healthy foods for the home, but a paucity of in-home food environment assessments exist. It re- mains unclear how the availability of healthful and less healthful foods in the home varies across levels of food security. Therefore, the present study compared high, marginal, and low/ very low food-security households with children on the availability and accessibility of healthful and less healthful foods derived from a vali- dated home food inventory. METHODS Participants and Recruitment Participants were adults who self- identied as the primary household food shopper, the individual who makes at least 50% of food purchases for the household. Recruitment oc- curred through yers posted on Rush University Medical Center campus, re- search fairs conducted at local Boys and Girls Clubs, referrals from a large primary care pediatric medicine prac- tice, and word of mouth. Inclusion criteria consisted of being a parent or primary caregiver with at least 1 child ages 213 and living within a 4.3- square-mile documented food desert region, a predominantly low-income Department of Preventive Medicine, Rush University Medical Center, Chicago, IL Address for correspondence: Bradley M. Appelhans, PhD, Department of Preventive Med- icine, Rush University Medical Center, 1700 W Van Buren St, Ste 470, Chicago, IL 60612; Phone: (312) 942-3477; Fax: (312) 942-8119; E-mail: [email protected] Ó2013 SOCIETY FOR NUTRITION EDUCATION AND BEHAVIOR http://dx.doi.org/10.1016/j.jneb.2013.08.001 780 Journal of Nutrition Education and Behavior Volume 45, Number 6, 2013

Upload: bradley-m

Post on 27-Jan-2017

212 views

Category:

Documents


0 download

TRANSCRIPT

Page 1: Food Insecurity Is Linked to a Food Environment Promoting Obesity in Households With Children

Research BriefFood Insecurity Is Linked to a Food Environment PromotingObesity in Households With ChildrenLisa M. Nackers, PhD, MPH; Bradley M. Appelhans, PhD

DepartmenAddress foicine, RushPhone: (31�2013 SOhttp://dx.d

780

ABSTRACT

Objective: To determine the extent to which the presence and accessibility of healthful and less healthfulfoods in children’s homes vary with level of food security.Methods: A total of 41 parents or primary caregivers who had at least 1 child ages 2–13 and resided ina low-income area with limited food access completed a home food inventory and a validated measureassessing household food security.Results: Compared with food-secure participants, marginal or low/very low food-secure caregivers re-ported significantlymore obesity-promoting foods in the home, moremicrowavable or quick-cook frozenfoods, and greater access to less healthful foods in the kitchen (all Ps < .05).Conclusions and Implications: Given the greater presence and accessibility of less healthful foods, tar-geting home food environment may improve diet quality and health status in children of low-income,food insecure households.Key Words: food security, home environment, home food inventory, childhood obesity (J Nutr EducBehav. 2013;45:780-784.)

INTRODUCTION

Food insecurity, or inaccessibility tonutritionally adequate foods owingto financial or other resource limita-tions,1 has emerged as an importantpublic health concern. United States(US) Department of Agriculture esti-mates from 2011 indicate that 14.9%of households (17.9 million) qualifiedas food insecure, an increase from11.9% in 1995.2 Of households withchildren, 10% were characterized bychild food insecurity in 2011.2 Al-though food insecurity is inconsis-tently associated with child obesityrisk,3 evidence suggests that childrenin food-insecure households experi-ence poor diet quality, lower healthstatus, decreased cognitive develop-ment, emotional and behavioral prob-lems, and subsequent accumulationof body fat in adulthood.4

An association exists among foodsecurity, income, and diet quality,with lower food security and incomeassociated with decreased healthful

t of Preventive Medicine, Rush Unr correspondence: Bradley M. AppelhUniversity Medical Center, 1700 W2) 942-3477; Fax: (312) 942-8119; E-CIETY FOR NUTRITION EDUCoi.org/10.1016/j.jneb.2013.08.001

food intake and variety.3,5 Whenassessing food-purchasing decisionsin high-poverty neighborhoods, food-insecure families identified price asthe most salient factor influencingtheir purchases.6 Nutrient-dense foods(eg, fruits, vegetables, whole grains,nonfat or low-fatmilk, and leanmeats)cost significantlymore per calorie thanenergy-dense foods (eg, soft drinks,salty and sugary snacks, pastries, andpackaged and frozen foods).5,7-9 Inaddition to price, low-income house-holds with children place greater im-portance on preparation convenienceand shelf-life.10 Consequently, food-insecure individuals more frequentlyconsume sugar-sweetened beverages,fast food, and energy-dense foods,and report lower fruit and vegetableintake comparedwith those who expe-rience food security.5,11

Home food availability serves asan important determinant of eatingbehavior and diet quality for food-insecure children.12 The literature re-ports that food-insecure individuals

iversity Medical Center, Chicago, ILans, PhD, Department of Preventive Med-Van Buren St, Ste 470, Chicago, IL 60612;mail: [email protected] AND BEHAVIOR

Journal of Nutrition Education and Beh

consume poorer-quality foods anddemonstrate difficulty obtainingnutrient-dense healthy foods for thehome, but a paucity of in-home foodenvironment assessments exist. It re-mains unclear how the availability ofhealthful and less healthful foods inthe home varies across levels of foodsecurity. Therefore, the present studycompared high, marginal, and low/very low food-security householdswith children on the availability andaccessibility of healthful and lesshealthful foods derived from a vali-dated home food inventory.

METHODSParticipants and Recruitment

Participants were adults who self-identified as the primary householdfood shopper, the individual whomakes at least 50% of food purchasesfor the household. Recruitment oc-curred through flyers posted on RushUniversity Medical Center campus, re-search fairs conducted at local Boysand Girls Clubs, referrals from a largeprimary care pediatric medicine prac-tice, and word of mouth. Inclusioncriteria consisted of being a parent orprimary caregiver with at least 1 childages 2–13 and living within a 4.3-square-mile documented food desertregion, a predominantly low-income

avior � Volume 45, Number 6, 2013

Page 2: Food Insecurity Is Linked to a Food Environment Promoting Obesity in Households With Children

Journal of Nutrition Education and Behavior � Volume 45, Number 6, 2013 Nackers and Appelhans 781

area with limited food access,13 nearthe Rush University Medical Center.Exclusion criteria included an inabil-ity to provide written consent inEnglish, ineligibility to work in theUS, living with a household memberwith major dietary restrictions or his-tory of bariatric surgery, or living intemporary housing (ie, homelessshelter or transitional housing). Allincluded parents and primary care-givers provided written informedconsent. The Rush University MedicalCenter Institutional Review Boardapproved this study.

Procedures

Included parents and primary care-givers attended an in-person assess-ment at the research facility tocomplete surveys on food securityand socioeconomic and demographicfactors. Study staff then provided in-structions on completing the homefood inventory and asked parentsand primary caregivers to completethe form at home within 2 days. Par-ents and primary caregivers returnedthe inventories during a second visitto the research facility, at which time$25 compensation was dispensed.

Measures

Parent and primary caregiver age, gen-der, race and ethnicity, education level,household income, marital status,number of household members, andcurrent food assistance receipt were as-sessed via self-report. The 2008 updateof the US Department of Agriculture'sHousehold Food Security Survey14 wasadministered to assess perceived restric-tion in the quantity, quality, or desir-ability of diet resulting from financiallimitations. The survey consisted of 18items. The sumof affirmative responsesprovided a total food security scorethat classified households into 1 of 4categories: high (0), marginal (1–2),low (3–7), and very low food security(8–18). Because only 4 parents and pri-mary caregivers were categorized asvery low food security, the categoriesof low and very low food security werecollapsed into a single category.

The Home Food Inventory of Ful-kerson and colleagues15 was used todocument the presence of 190 foodsin the home. Parents and primary

caregivers indicated whether a fooditem was present using a checklist for-mat with yes/no response options.The scale provided scores for homeavailability of 30 food groups andthe accessibility of healthful and lesshealthful foods in the kitchen and re-frigerator.15 A summative score of 71obesity-promoting foods (ie, discre-tionary caloric beverages, regular-fatdairy, snacks, desserts, candy, and mi-crowavable or quick-cook frozenfoods, but not including frozen vege-tables or fruits) yielded an obesogenichome food availability score, withhigher scores representing greateravailability. The authors evaluatedwhether food security status was asso-ciated with the obesogenic home foodavailability score, the presence offoods in 7 obesity-relevant food cate-gories, and access to healthful andless healthful foods in the kitchen(ie, visible and readily accessible onthe countertop, on top of the refriger-ator, and on the table) and refrigera-tor. Previous assessments determinedthat participant-completed invento-ries show high agreement with staff-completed inventories, and thereforeserve as valid and convenient mea-surements of home food environ-ment.15 In addition, food categoryscores are significantly and positivelycorrelated with dietary intake withincorresponding food groups.15

Data Analysis

Descriptive statistics (means and SDs)characterized the study sample andhome food availability. Fisher's exacttest compared household food securitygroups on income and food assistance.Linear regression models compared in-dividualswithmarginal and, separately,low/very low household food securitywith those with high food security onhome food availability. The primary de-pendent variable, the obesogenic homefood availability score, was tested ata ¼ .05. Food security groups were alsocompared on 11 other food outcomes,with an adjusted a of P < .02 (totalfamily-wise a ¼ .22). All continuousvariable distributions were examinedfor normality and none exhibitedpotentially problematic skewness orkurtosis. Models were adjusted forpotential confounding variables, in-cluding household income, race andethnicity, marital status, number of

household members, and food assis-tance receipt. Data were analyzed usingStata 11 statistical software (StataCorpLP, College Station, TX, 2009).

RESULTS

A total of 41 parents and primary care-givers contributed data to the analyses(Table 1). The mean household foodsecurity score was 3.5 (SD 3.4) of a pos-sible 18, with parents and primarycaregivers categorized into high(n ¼ 10), marginal (n ¼ 7), and low/very low (n ¼ 24) food securitygroups. Food security groups differedon household income (P ¼ .02), withlower income among marginal andlow/very low food security groups.Groups did not differ on food assis-tance receipt (P ¼ .73), with 2 of 10(20%) high, 3 of 7 (43%) marginal,and 11 of 24 (46%) low/very lowfood security participants receivingfood assistance.

Overall, parents and primary care-givers reported the presence of 32.7(SD 11.6) of the 71 items contributingto the obesogenic home food avail-ability score. Lower obesogenic homefood availability scores were associ-ated with smaller household size(b ¼ �2.2, t ¼ 2.3, P ¼ .04) and His-panic ethnicity (b ¼ 12.0, t ¼ �2.8,P < .01, compared with African Amer-ican race). Hispanics also reported sig-nificantly smaller household sizescompared with African Americans(3.1 vs 4.4 household members,P ¼ .03, respectively). In covariate-adjusted models, compared with par-ents and primary caregivers withhigh food security, significantlygreater obesogenic home food avail-ability scores were reported by parentsand primary caregivers with marginal(b ¼ 14.3, t ¼ 2.3, P ¼ .03) and low/very low (b ¼ 12.4, t ¼ 2.4, P ¼ .03)food security. A larger number of mi-crowaveable or quick-cook frozenfoods were indicated by marginal(b ¼ 3.5, t ¼ 2.7, P < .01) and low/very low (b ¼ 3.5, t ¼ 3.2, P < .01)food-secure parents and primary care-givers than by those with high foodsecurity. In addition, greater accessto less healthful foods in the kitchenwas reported by marginal (b ¼ 2.6,t ¼ 2.7, P < .01) and low/very low(b ¼ 2.3, t ¼ 2.9, P < .01) food-secureparents and primary caregivers com-pared with those with high food

Page 3: Food Insecurity Is Linked to a Food Environment Promoting Obesity in Households With Children

Table 1. Individual Demographic and Household Characteristics Among Parents and Primary Caregivers (n ¼ 41)

Overall High Marginal Low/Very Low

Parent/primary caregiver age, y (mean [SD]) 37.5 (9.4) 39.7 (10.1) 33.3 (7.7) 37.8 (9.5)

Index child age, y (mean [SD]) 6.7 (3.4) 5.7 (1.8) 7.4 (4.1) 6.9 (3.7)

Household members, n (mean [SD]) 3.6 (1.7) 3.1 (1.2) 4.4 (2.6) 3.5 (1.5)Adults 1.6 (0.8) 1.4 (0.8) 1.7 (0.8) 1.7 (0.8)Children 2.0 (1.3) 1.7 (0.9) 2.7 (2.2) 1.8 (1.1)

Female gender, n (%) 32 (78) 7 (70) 6 (86) 19 (79)

Race/ethnicity, n (%)Black/African American 17 (42) 2 (20) 4 (57) 11 (46)Hispanic/Latino 17 (42) 6 (60) 2 (29) 9 (38)Non-Hispanic White/Caucasian 7 (17) 2 (20) 1 (14) 4 (17)

Marital status, n (%)Single 22 (54) 4 (40) 5 (71) 13 (54)Married/living with partner 12 (29) 4 (40) 2 (29) 6 (25)Divorced/widowed 7 (17) 2 (20) 0 5 (21)

Household income (USD), n (%)$0–$19.999 13 (32) 1 (10) 2 (29) 10 (42)$20,000–$39,999 20 (49) 3 (30) 4 (57) 13 (54)$ $40,000 8 (20) 6 (60) 1 (14) 1 (4)

Education, n (%)High school or below 14 (34) 1 (10) 4 (57) 9 (38)Some college 10 (24) 1 (10) 2 (29) 7 (29)2-year degree or technical school 7 (17) 3 (30) 0 4 (17)Baccalaureate or master’s degree 10 (24) 5 (50) 1 (14) 4 (17)

Receiving food assistance, n (%) 14 (34) 2 (20) 3 (42) 9 (38)

USD indicates US dollars.

782 Nackers and Appelhans Journal of Nutrition Education and Behavior � Volume 45, Number 6, 2013

security. Food security groups did notdiffer in healthful kitchen access,healthful or unhealthful refrigeratoraccess, or the availability of fruits, veg-etables, prepared desserts, snacks, dis-cretionary caloric beverages, or candy.Table 2 provides detailed home foodavailability in high, marginal, andlow/very low food-secure parents andprimary caregivers.

DISCUSSION

Previous research suggests that food-insecure individuals overconsumeenergy-dense, nutrient inadequatefoods because of the high costs ofhealthier nutrient-dense foods.3,5,7-9

Food purchasing decisions by food-insecure families depend mainly onprice, with home food supplies reflect-ing these price constraints.6 In thepresent study, marginal and low/verylow food-secure parents and primarycaregivers reported significantlygreater obesogenic home food avail-ability than the high food securityparents and primary caregivers, whichsupports the idea that in low-income,

food-insecure homes, financial con-straints may hinder healthy eating.In addition, preparation ease andshelf life serve as important food prod-uct attributes in food-insecure house-holds.10 Marginal and low/very lowfood-secure parents and primary care-givers in the present study identifiedmore microwavable and quick-cookfrozen foods, which often contributelittle nutrition to total daily dietary in-take.16 Because eating behavior andenergy and nutritional intake corre-late with home food availability,12

children within food-insecure homesare at risk for adverse outcomes as a re-sult of the greater presence of obeso-genic foods.

Overall, 37% of fruit and 54% ofvegetable items assessed on the homefood inventory were reported as beingpresent. Availability of fruits and vege-tables in the home did not vary byfood security status. Although a num-ber of previous studies observed lowerhousehold availability and decreasedfruit and vegetable consumption asfood security status worsened,5,11,17

others reported no difference in fruit

and vegetable intake across foodsecurity status when controlling fordemographic variables.18 Specifically,children from food-insecure house-holds appear to consume fruits andvegetables in similar amounts asfood-secure children.19 Because ap-proximately one third of the currentsample received food assistance, andfood security groups did not differ onfood assistance receipt, it is possiblethat parents and primary caregiversused this funding to purchase fruitsand vegetables. According to the Na-tional Food Stamp Program Survey,19.6% of foods purchased by food as-sistance recipients were fruits and veg-etables.20

Smaller household size and His-panic ethnicity, compared with Afri-can American race, were associatedwith lower obesogenic food availabil-ity scores. Notably, household sizeappeared to drive this ethnic differ-ence, as Hispanics reported fewer av-erage household members than didAfrican Americans (3.1 vs 4.4, respec-tively). Household size affects theamount of money spent on food

Page 4: Food Insecurity Is Linked to a Food Environment Promoting Obesity in Households With Children

Table 2. Home Food Availability for Parents and Primary Caregivers Reporting High, Marginal, and Low/Very Low Food Security(n ¼ 41)

PossibleScoring Range(Low to High)

Overall (n ¼ 41)Mean (SD)

High(n ¼ 10)

Mean (SD)

Marginal(n ¼ 7)

Mean (SD)

Low/VeryLow (n ¼ 24)Mean (SD)

Obesogenic home food availability scorec 0–71 32.7 (11.6) 27.1 (9.6) 35.9 (13.9)a 34.1 (11.4)a

Vegetables 0–20 10.8 (3.9) 10.0 (2.4) 12.0 (3.7) 10.8 (4.5)

Fruits 0–26 9.5 (5.3) 7.7 (3.5) 9.7 (5.4) 10.2 (5.9)

Snacks 0–18 5.3 (3.3) 5.4 (3.0) 6.0 (3.4) 5.1 (3.5)

Microwaveable or quick-cook frozen foods 0–8 3.6 (2.4) 1.9 (1.5) 4.1 (2.5)b 4.1 (2.5)b

Discretionary caloric beverages 0–6 2.8 (1.6) 2.5 (1.4) 2.1 (1.8) 3.1 (1.5)

Prepared desserts 0–8 2.0 (1.7) 1.2 (1.0) 2.0 (2.0) 2.3 (1.8)

Candy 0–5 1.9 (1.6) 1.5 (1.1) 1.9 (1.6) 2.1 (1.9)

Unhealthful kitchen accessd 0–6 3.5 (1.9) 2.5 (2.2) 4.0 (2.2)b 3.8 (1.6)b

Healthful kitchen accessd 0–6 1.7 (0.9) 1.3 (0.9) 1.6 (1.0) 1.9 (0.9)

Unhealthful refrigerator accessd 0–6 3.1 (1.4) 2.9 (1.5) 2.9 (1.3) 3.3 (1.5)

Healthful refrigerator accessd 0–9 3.4 (1.4) 4.1 (1.3) 3.3 (1.5) 3.2 (1.4)

aSignificantly different from high food-security households in adjusted linear regression models at P< .05; bSignificantly differ-ent from high food-security households in adjusted linear regression models at P < .01; cThe summative obesogenic homefood availability score was composed of 71 obesity-promoting foods (ie, discretionary caloric beverages, regular-fatdairy, snacks, desserts, candy, and microwavable or quick-cook frozen foods, but not including frozen vegetables or fruits);dUnhealthful and healthful kitchen and refrigerator access were based on whether foods in these subscales were visible andreadily accessible in the kitchen and refrigerator, respectively.Note: Estimates shown are unadjusted means and SDs. Significance levels are for group differences in fully adjusted linear re-gression models controlling for household income, food assistance receipt, household size, and race or ethnicity, with highfood-security households as the reference group. Except for the obesogenic home food availability score, which was testedat a ¼ .05, dependent variables were tested at adjusted a ¼ .02.

Journal of Nutrition Education and Behavior � Volume 45, Number 6, 2013 Nackers and Appelhans 783

within the home21 and adequacy ofdietary intake.22 Smaller householdsize has also been associated withlower snack intake in Mexican Amer-icans.23 In addition, compared withAfrican Americans, Hispanic familiesreported moderately fewer sweetsnacks available in the home.24

Regardless of food availability,environment serves as an importantcue for eating.25 In the present study,marginal and low/very low food-secure households reported greateraccessibility of less healthful foodsin the kitchen. Greater proximity,accessibility, and visibility of foodswithin the immediate environmentincrease the frequency and amountof food consumed by children.26

Therefore, food-insecure individualsmay be at risk for unhealthy eatingnot only because of greater availabil-ity, but also because of greater accessi-bility of obesogenic foods within thehome environment.

This study had notable strengths.Parents and primary caregivers com-pleted a validated, comprehensive

home food inventory assessing a vari-ety of foods in the household. In ad-dition, the project focused onchildren from predominantly low-income households in an area of lim-ited food access, a group particularlyvulnerable to obesity and diet-related illnesses. Several limitationsshould also be considered. This studyinvolved a small convenience sampleof parents and primary caregivers liv-ing within a specific geographicallydefined region, which limited statisti-cal power and generalizability to thelarger community. In addition, thecross-sectional design only permittedexamination of associations at a singletime point, and home food availabil-ity may change across weeks. Forexample, the Food and Nutrition Ser-vice recently reported that 53% ofhouseholds receiving food assistanceredeemed all or most of theirmonthly benefits by the secondweek.27 Finally, this study did not as-sess food intake or acquisition offoods from different sources (eg,breakfast consumption or fast-food

intake), which contribute to dietquality and have been associatedwith food security status.5,11

IMPLICATIONS FORRESEARCH ANDPRACTICE

In this study of households with chil-dren, greater availability and accessi-bility of obesity-promoting foodswere observed in marginal and low/very low food-secure homes relativeto food secure households. No groupdifferences were found for fruit andvegetable availability, which suggeststhat food insecurity may influencediet quality and health through anoverabundance of less healthful foodsin the home, rather than througha lack of healthful foods. Becausechildren's eating habits are shaped bythe foods available within thehome,28 further research is warrantedto develop intervention strategiesthat effectively modify the homefood environment in disadvantaged

Page 5: Food Insecurity Is Linked to a Food Environment Promoting Obesity in Households With Children

784 Nackers and Appelhans Journal of Nutrition Education and Behavior � Volume 45, Number 6, 2013

households, many of which rely onfood assistance interventions. Strate-gies that teach children and parentsor primary caregivers healthy cookingskills, encourage collaboration be-tween children and parents or primarycaregivers when developing grocerylists and grocery shopping, and edu-cate parents or primary caregivers onfood stamp management28,29 mayserve as potential targets to improvediet quality and overall health statusin children of low-income, food-inse-cure households.

ACKNOWLEDGMENTS

The study was funded by the Consor-tium to Lower Obesity in ChicagoChildren, a coalition of researchers,clinicians, community stakeholders,and public health officials dedicatedto reversing the epidemic of childobesity.

REFERENCES

1. Life Sciences Research Office. Core in-dicators of nutritional state for difficult-to-sample populations. J Nutr. 1990;120(suppl 11):1559-1600.

2. Coleman-Jensen A, Nord M,Andrews M, Carlson S. HouseholdFood Security in the United States in2011. Washington, DC: US Depart-ment of Agriculture, Economic Re-search Service; 2012.

3. Franklin B, Jones A, Love D, Puckett S,Macklin J, White-Means S. Exploringmediators of food insecurity and obe-sity: a review of recent literature. J Com-mun Health. 2012;37:253-264.

4. Kaiser LL, TownsendMS. Food insecu-rity among US children: implicationsfor nutrition and health. Top ClinNutr. 2005;20:313-320.

5. Drewnowski A, Darmon N. Foodchoices and diet costs: an economicanalysis. J Nutr. 2005;135:900-904.

6. Dachner N, Ricciuto L, Kirkpatric SI,Tarasuk V. Food purchasing and foodinsecurity among low-income familiesin Toronto. Can J Diet Pract Res. 2010;71:e50-e56.

7. Aggarwal A, Monsivais P,Drewnowski A.Nutrient intakes linkedto better health outcomes are associatedwith higher diet costs in the US. PLoSONE. 2012;7:e37533.

8. Rehm CD, Monsivais P,Drewnowski A. The quality and mon-

etary value of diets consumed by adultsin the United States. Am J Clin Nutr.2011;94:1333-1339.

9. Drewnowski A, Specter SE. Povertyand obesity: the role of energy densityand energy costs. Am J Clin Nutr.2004;79:6-16.

10. Bruening M, MacLehose R, Loth K,Story M, Neumark-Sztainer D. Feed-ing a family in a recession: food insecu-rity among Minnesota parents. Am JPublic Health. 2012;102:520-526.

11. Morton JF, Guthrie JF. Diet-relatedknowledge, attitudes, and practices oflow-income households with childrenin the household. Fam Econ Nutr Rev.1997;10:2-15.

12. Rasmussen M, Krolner R, Klepp KI,et al. Determinants of fruit and vegeta-ble consumption among children andadolescents: a review of the literature.Part I: quantitative studies. Int J BehavNutr Phys Act. 2006;3:22.

13. The Chicago Food Desert ProgressReport. Gallagher M. http://marigallagher.com/site_media/dynamic/project_files/ChicagoFoodDesProg2009.pdf. Ac-cessed September 22, 2012.

14. Bickel G, Nord M, Price C,Hamilton W, Cook J. Guide to Measur-ing Household Food Security: Revised2000. Alexandria, VA: US Departmentof Agriculture, Food and NutritionService; 2000.

15. Fulkerson JA, Nelson MC, Lytle L,Moe S, Heitzler C, Pasch KE. The val-idation of a home food inventory. Int JBehav Nutr Phys Act. 2008;5:55.

16. Eicher-Miller HA, Fulgoni VL,Keast DR. Contributions of processedfoods to dietary intake in the US from2003-2008: a report of the Food andNutrition Science Solutions Joint TaskForce of the Academy of Nutrition andDietetics, American Society for Nutri-tion, Institute of Food Technologists,and International Food InformationCouncil. J Nutr. 2012;142:2065S-2072S.

17. Kendall A, Olson CM, Frongillo EA.Relationship of hunger and food inse-curity to food availability and con-sumption. J Am Diet Assoc. 1996;96:1019-1024.

18. Mello JA, Gans KM, Risica PM,Kirtania U, Strolla LO, Fournier L.How is food insecurity associated withdietary behaviors? An analysis withlow-income, ethnically diverse partici-pants in a nutrition intervention study.J Am Diet Assoc. 2010;110:1906-1911.

19. Grutzmacher S, Gross S. Householdfood security and fruit and vegetable in-

take among low-income fourth-graders.J Nutr Educ Behav. 2011;43:455-463.

20. Cohen B, Ohls J, Andrews M, et al.Food Stamp Participants’ Food Securityand Nutrient Availability. Alexandria(VA): US Department of Agriculture,Food and Nutrition Service, Office ofAnalysis and Evaluation; 1999.

21. Morgan KJ, Johnson SR, Burt J.Household size and the cost of nutri-tionally equivalent diets. Am J PublicHealth. 1983;73:530-537.

22. Kucera B,McIntoshWA. Family size asa determinant of children’s dietary in-take: a dilution model approach. EcolFood Nutr. 1991;26:127-138.

23. Ayala GX, Baquero B,Arrendondo EM, Campbell N,Larios S, Elder JP. Association betweenfamily variables andMexican Americanchildren’s dietary behaviors. J Nutr EducBehav. 2007;39:62-69.

24. Skala K, Chuang RJ, Evans A,Hedberg AM,Dave J, Sharma S. Ethnicdifferences in the home food environ-ment and parental food practicesamong families of low-income His-panic and African-American pre-schoolers. J Immigr Minor Health. 2012;14:1014-1022.

25. Sobal J, Wansink B. Kitchenscapes, ta-blescapes, platescapes, and foodscapes:influences of microscale built environ-ments on food intake. Environ Behav.2007;39:124-142.

26. Musher-Eizenman DR, Young KM,Laurene K, Galliger C, Hauser J, Wag-ner Oehlhof M. Children’s sensitivityto external food cues: how distance toserving bowl influences children’s con-sumption. Health Educ Behav. 2010;37:186-192.

27. Caster L, Henke J. Benefit RedemptionPatterns in the Supplemental NutritionAssistance Program. http://www.fns.usda.gov/ora/MENU/Published/snap/FILES/ProgramOperations/ARRASpendingPatterns.pdf. Accessed October11, 2012.

28. Dhokarh R, Himmelgreen DA,Peng YK, Segura-P�erez S, Hromi-Fiedler A, P�erez-Escamilla R. Food in-security is associated with acculturationand social networks in Puerto Ricanhouseholds. J Nutr Educ Behav. 2011;43:288-294.

29. Dammann K, Smith C. Food-relatedattitudes and behaviors at home, school,and restaurants: perspectives fromracially diverse, urban, low-income 9-to 13-year-old children in Minnesota.J Nutr Educ Behav. 2010;42:389-397.