Final Project Milestone Two: Literature Review

T h e S u p p l e m e n t a l N u t r i t i o n A s s i s t a n c e P r o g r a m , F o o d I n s e c u r i t y , D i e t a r y Q u a l i t y , a n d O b e s i t y A m o n g U S A d u l t s I B in h T. N g u y e n , P h D , K e r e m S h u v a l, P h D , F a rry l B e r t m a n n , P h D , a n d A m y L Y a r o c h , P h D Objectives. W e e x a m in e d w h e t h e r S u p p le m e n ta l N u t r i t i o n A s s is ta n c e P r o ­ g r a m (S N A P ) p a r t i c i p a t i o n c h a n g e s a s s o c i a t io n s b e t w e e n f o o d i n s e c u r i t y , d i e t a r y q u a l i t y , a n d w e i g h t a m o n g U S a d u lt s .

Methods. W e a n a ly z e d a d u lt d ie t a r y In ta k e d a ta (n = 8333) f r o m t h e 2003 to 2 010 N a t io n a l H e a lth a n d N u t r i t i o n E x a m in a t io n S u r v e y . B iv a ria te a n d m u l t i v a r i ­ a b le m e t h o d s a sse sse d a s s o c ia t io n s o f S N A P p a r t i c ip a t i o n a n d 4 le v e ls o f fo o d s e c u r it y w i t h d ie t a n d w e i g h t . M e a s u r e s o f d ie ta r y q u a li t y w e r e t h e H e a lth y E a tin g In d e x 2010, t o t a l c a lo r ic In ta k e , e m p t y c a lo r ie s , a n d s o li d fa t ; w e i g h t m e a s u r e s w e r e b o d y m a s s in d e x (B M I), o v e r w e i g h t , a n d o b e s it y .

Results. S N A P p a r t ic ip a n t s w i t h m a r g in a l f o o d s e c u r it y h ad lo w e r B M I (1.83 k g / m 2; P < .01 ) a n d l o w e r p r o b a b i l i t y o f o b e s it y (9 p e r c e n ta g e p o in t s ; P < . 05).

S N A P p a r t ic ip a n t s w i t h m a r g in a l (3.46 p o in t s ; P < .0 1 ) , l o w (1.98 p o in t s ; P < .05), a n d v e r y l o w (3 .8 4 p o in t s ; P < .01 ) f o o d s e c u r it y h a d b e tt e r d ie ts , as illu s t r a t e d b y t h e H e a lth y E a tin g In d e x . A s s o c ia tio n s b e tw e e n S N A P p a r t i c ip a t i o n a n d i m ­ p r o v e d d ie t a n d w e i g h t w e r e s t r o n g e r a m o n g W h ite s t h a n B la cks a n d H is p a n ic s .

C o n c lu s io n s . O u r r e s e a r c h h i g h l i g h t s t h e r o l e o f S N A P In h e l p i n g i n ­ d i v i d u a l s w h o a r e a t r is k f o r f o o d i n s e c u r i t y t o o b t a i n a h e a l t h i e r d i e t a n d b e t t e r w e i g h t s t a t u s . {A m J P u b lic H ealth. 2 0 1 5 ; 1 0 5 : 1 4 5 3 - 1 4 5 9 . d o i : 1 0 .2 1 0 5 / A J P H . 2 0 1 5 .3 0 2 5 8 0 ) Food insecurity, broadly defined as having limited access to ad equate food,1 is associated w ith increased stress levels and red u ced overall well-being.2 In addition, food insecurity has b e e n show n to dim inish dietary quality and affect nutritional intake and has b een associ­ ated with chronic m orbidity (e.g., type 2 di­ abetes, hypertension) and w eight gain.1,3' 5 In 20 1 2 , approxim ately 14.5% o f US households (17.6 million households) experienced food insecurity, o f w hom 5 .7 % (7.0 million hou se­ holds) e x p e rie n c e d v e ry low food se c u rity (i.e., red u ctio n in food intake).6 T h e Supple­ m ental N utrition Assistance Program (SNAP), formerly know n as food stamps, is th e largest g overnm ent assistance p rogram in th e United States a n d seeks to alleviate food insecurity in US households.7 SNAP has th e potential to mitigate th e adverse effects of food insecurity on health outcom es n o t only through attenu­ ating food insecurity b u t also by enhancing th e dietary quality of its participants.8'9 A lthough cross-sectional studies have found no significant differences in food insecurity levels betw een SNAP participants and n o n p ar­ ticip an ts,10'11 in a lo n g itu d in a l study, N o rd observed a 2 8 % reduction in th e odds fo r very low fo o d se c u rity a m o n g h o u s e h o ld s th a t rem ained on SNAP throu g h o u t th e y ear rela­ tive to those w ho left before th e last 3 0 days o f th e year.12 In addition, studies by Leung and V illam or13 and W eb b et al.14 found th a t in­ d e p e n d e n t of food insecurity, SNAP participa­ tion is associated w ith th e increased likelihood o f obesity, and o th e r studies h av e observed low er dietary quality specifically am ong SNAP p a rtic ip a n ts .15,16 K re id e r e t al. u se d p a rtia l identification b o unding m ethods to take into account th e endogenous selection and misre- porting o f SNAP enrollm ent and found that SNAP red u ced th e prevalence o f food insecu­ rity, p o o r genera] health, an d obesity among children.17 Thus, th e interrelationships am ong SNAP participation, food insecurity, dietary quality,and weight status w a rra n t further investigation to inform SNAP programming, policy, and outreach to ultimately im prove th e h ealth and well-being o f SNAP participants. W e explored these relationships in d ata from th e National H ealth and N utrition E xamination Survey (NHANES) over multiple years. W e aim ed to determ ine mitigating effects SNAP participa­ tion m ight h av e on th e association o f food insecurity w ith dietary quality and obesity am ong a nationally representative sam ple of US adults.

M E T H O D S T he NHANES is a multistage, cross-sectional, nationally representative survey conducted by th e National C enter for H ealth Statistics to explore th e health and nutritional status of US children and adults.18 O ur participants came from 4 waves: 2 0 0 3 to 2 0 0 4 , 2 0 0 5 to 2 0 0 6 , 2 0 0 7 to 2 0 0 8 , a n d 2 0 0 9 to 2 0 1 0 . T h e NHANES collects inform ation on dem ographic and socioeconomic characteristics and healthfrom an in-home questionnaire, as well as obtaining dietary d a ta and medical and phys­ iological m easurem ents and perform ing labo­ ratory tests and a physical examination in mobile examination centers.18 W e focused on participants with family income less than 2 0 0 % of the federal poverty level in 2 0 0 3 to 2 0 1 0 to reduce residual confounding in the sample, especially between the high-income food security group and other groups. W e did not limit our sample to SNAP-eligible participants with incomes lower than 1 3 0 % of the federal poverty level because we aimed at capturing both the marginal food security and food insecurity population, many of whom are not eligible for SNAP.6 ' 9 Thus o ur sample consisted o f 8 3 3 3 nonpregnant adults aged 2 0 years and older who had completed day 1 dietaiy interviews.

M e a s u r e s W e d eterm ined participation in th e SNAP program by an affirmative response to the question, “In th e last 12 months, did [you, o r any m em b er of y o u r household] receive food July 2 0 1 5 , Vol 1 0 5 , No. 7 | A m eric a n Journal o f P ublic Health Nguyen e t al. \ Peer Reviewed | Research an d P ractice | 1 4 5 3 stamp benefits?”18 W e derived 4 levels of food security from responses to the NHANES Food Security Survey Module questionnaires, details of which are available online.20 Households with high food security reported no food access problems or limitations; households with mar­ ginal food security may have had anxiety over food sufficiency or shortage of food in the house; households with low food security generally reported reduced quality, variety, or desirability of diet without changes in diet or food intake; and households with very low food security generally reported multiple indications of disrupted eating patterns and reduced food intake. W e considered households in the high and marginal categories to be food secure.

Our key outcome variables were (1) diet (the Healthy Eating Index 2 0 1 0 [HEI-2010]21 and intake of added sugar, solid fat, empty calories, and total calories) and (2) weight (body mass index [BMI], defined as weight in kilograms divided by the square of height in meters; overweight; and obesity). From the first-day dietary recall data (24 hours), we computed HEI-2010 as well as other dietary indicators, such as empty calorie, solid fat, and added- sugar intake, with National Cancer Institute methodology.22 W e used HEI-2010, a tool that aims to determine compliance with the 2010 Dietary Guidelines for Americans, to assess overall dietary quality.21 W e calculated BMI with the standard formula and objectively measured height and weight. W e used World Health Organization criteria to categorize par­ ticipants’ BMI as underweight (< 18.5 kg/m 2), normal weight (18.5—< 25 kg/m 2), overweight (> 2 5 - < 30 kg/m 2), or obese (> 3 0 kg/m 2).22 S t a t i s t i c a l A n a ly s is W e used the first-day 24-hour dietary recall data to document participants’ sociodemo­ graphic characteristics by participation status in the SNAP program. W e also examined the differences in H E I-2010 score and intake of added sugar, solid fat, empty calories, and total calories among those with full food security versus all others (i.e., participants with mar­ ginal, low, and very low food security). In addition, we examined the differences in per­ centage of underweight, overweight, and obe­ sity and in BMI between these 2 groups.

To examine the combined effect of SNAP participation and food security, we estimatedan ordinary least squares model with the in­ teraction coefficient of SNAP and food security.

Our formula was ( 1 ) Y ij = a ¡j + bSNAPj + yFoodlnsecj + aSNAPj x Foodlnsecj + SyXy + e,y W here T,y, the dependent variables, denoted outcomes of individual i in household a ¡y was the intercept; b was a parameter estimate for the baseline difference between SNAP partici­ pants and SNAP nonparticipants; and y was an estimate for th e difference between 4 levels of food security. The main parameter of interest, a, was an estimate of the cross-level interac­ tions of a household’s SNAP status and food security.

Other control covariates (Xy) were age; gender; race/ethnidty (non-Hispanic White, non-Hispanic Black, Hispanic, other); education ( college); marital status (married, never married, divorced or separated, widowed); poverty-to-income ratio; Women, Infants, and Children program p ar­ ticipation in the past year; health insurance status (insured or not); employment status (employed or not); w hether the survey was completed on a weekday or w eekend24; and interview wave ( 2 0 0 3 - 2 0 0 4 , 2 0 0 5 - 2 0 0 6 , 2 0 0 7 - 2 0 0 8 , 2 0 0 9 - 2 0 1 0 ) .

We conducted all statistical analyses with STATA version 1325 and accounted for the NHANES complex, multistage probability sam­ pling design of households and individuals to enable nationally representative estimates.26 Because individuals in the mobile examination centers sample provided the dietary recall data, we used the centers’ sample weights (provided by NHANES) in all analyses. W e computed HEI-2010 scores with SAS software version 9.3.27 R E S U L T S The characteristics of the study population are shown in Table 1. T he study sample consisted of 8 3 3 3 adults. Participants had a m ean age of 4 5 .5 years; 5 5 .4 % were women, 5 5 .9 % w ere non-Hispanic Whites, 16.4% were non-Hispanic Blacks, 2 1 .9 % were Hispanics, 51 .1 % were married, 64 .5 % had health insurance, and 4 9 .2 % were employed. Food security was high in 59 .1 %of respondents’ households, marginal in 13.2%, low in 17.2%, and very low in 10.5%.

The bivariate relationship of food security status to dietary quality and weight status is presented in Table 2. Participants with any level of food insecurity had a significantly lower H E I-2010 score than those with full food security (43.7 vs 46.6), higher intake of added sugar (22.0 vs 18.7 teaspoons), and higher intake of empty calories (787.9 vs 731.5 kilocalories; P< .05 for all). Furthermore, in­ dividuals living in households without food security h ad significantly higher BMIs and w ere likelier to be obese th a n those with food security (3 8 .4 % vs 33.7% ; P < .0 1 ) . How­ ever, we observed no significant differences in solid fat consumption o r the probability of being underweight.

The association of SNAP participation and food security status with dietary quality and weight status among low-income respondents is presented in Table 3. SNAP participants had a poorer nutrient profile (lower H E I-2010 score, higher consumption of added sugar, solid fat, and empty calories) than nonpartidpants.

HEI-2010 scores were lowest among partici- pants who reported living in households with very low food security (2.59 points lower than in the reference group, partidpants with high food security), followed by those with marginal (-2 .2 7 points), and low ( -1 .6 3 points) food security.

Table 3 also shows the interaction between SNAP partidpation and food security (i.e., whether SNAP partidpation may change the assodations between food insecurity, dietary quality, and weight status among US adults). Partidpation in SNAP was associated with higher H E I-2010 scores (better nutrient profile) among individ­ uals in households with marginal (+3.46 points), low (+1.98 points), and very low (+3.84 points) food security than among re ­ spondents with corresponding food insecurity who did n o t receive SNAP benefits. For participants with low food security, partici­ pating in SNAP was only associated with lower added-sugar ( - 3 .8 8 teaspoons) and empty calorie (—6 7 .5 6 kcal) intake. Although SNAP participants and respondents experi­ encing food insecurity each independently had a higher BMI and higher probability of being obese, the combined association of SNAP participation and food insecurity 1 4 5 4 I Research and P ractice | Peer Reviewed | Nguyen e t al.Am eric a n Journal o f P ublic H ealth | July 2 0 1 5 , Vol 1 0 5 , No. 7 TABLE 1 -S u m m a ry Statistics of Low- Income Adults: National Health and Nutrition Examination Survey, 2 0 0 3 - 2010 VariableFull Sample (n * 8 3 3 3 ) Women, % 5 5 .4 Age, y, mean 4 5 .5 SNAP particip atio n, % 2 7 .3 Household food security,3 % Full 5 9 .1 Marginal 1 3 .2 Low 1 7 .2 Very low 1 0 .5 R ac e/e th n lcity, % Non-Hispanic White 5 5 .9 Non-Hispanic Black 1 6 .4 Hispanic 2 1 .9 Other 5 .9 M a rital status, % Married 5 1 .1 Widowed 9 . 6 D ivo rc ed /s ep a rated 1 7 .1 Never married 2 2 .2 Education, % < high school 3 5 .7 High school 2 8 .3 Some college 2 6 .6 > college 9 .4 Health insurance, % 6 4 .5 Currently employed, % 4 9 .2 Received WIC benefits in 2 1 .1 past year, % Poverty-to-income ratio, FPL, % 0 - 5 0 1 2 .3 5 1 - 1 0 0 2 7 .1 1 0 1 - 1 3 0 2 0 .4 1 3 1 - 2 0 0 4 0 .2 Household size, mean 3 .3 Survey on weekend, % 3 9 .0 Continued ap p e ared to decrease BMI across all 3 food-insecure groups and reduce the likelihood of obesity among participants with marginal food security (9 percentage points).

Table 4 presents the associations of SNAP participation and food security with dietary quality and weight status, stratified by race/ ethnicity. These results indicated that SNAPTABLE 1 - C o n t i n u e d Wave 1 ( 2 0 0 3 - 2 0 0 4 ) 2 4 .1 2 ( 2 0 0 5 - 2 0 0 6 ) 2 3 .0 3 ( 2 0 0 7 - 2 0 0 8 ) 2 6 .2 4 ( 2 0 0 9 - 2 0 1 0 ) 2 6 .7 Note. FPL = f ed eral poverty level; S N A P 'S u p p le m e n ­ tal Nutrition Assistance Program; WIC = Women, Infants, an d Children program. Results ta k e survey weights into ac count. Respondents were aged 2 0 years or older an d had family income under 2 0 0 % of th e FPL.

R e s p o n d e n ts from households with children younger th a n 1 8 years were asked 1 8 questions from the US Food Security Survey Module; respondents from households w ithout children were asked 1 0 questions.

The food insecurity variable, with 4 response levels, was derived from affirm ative responses. Household full food security » z e r o affirm ativ e responses; marginal food s e c u r ity * 1 - 2 affirm ativ e responses; low food security * 3 - 5 affirm ativ e responses fo r households without ch ildren 3 - 7 affirm ative responses for households with children; very low food s e c u rity * 6 - 1 0 affirm ativ e responses fo r households without children and 8 - 1 8 affirm ative responses fo r house­ holds with children.

participation had limited effect on dietary quality and weight status among food-insecure non-Hispanic Black adults. By contrast, SNAP participation among food-insecure non-Hispanic Whites was associated with a higher H EI-2010 score for respondents with marginal (+5.29 points), low (+3.92 points), and very low (+4.83 points) food security as well as with lower overall BMI among participants with marginal (-2 .5 9 kg/m 2) and very low (-2 .0 3 kg /m 2) food security. Among Hispanic adults, SNAP participation was related to lower added-sugar consumption ( -3 .1 5 teaspoons) lower BMI ( -1 .5 4 kg /m 2), and lower likeli­ hood of obesity ( - 1 2 percentage points) among the marginal food security group.

DISCUSSION W e analyzed nationally representative data to determ ine w hether SNAP participation modified the associations between food inse­ curity and individuals’ dietary quality and weight. Consistent with the literature, we found that food insecurity and SNAP participation, independently, were associated with lower di­ etary quality and a higher prevalence of obesity among adults.15,16 In addition, we augmentedprevious research with our finding that SNAP participation among those who do not have full food security might protect against a less healthful diet and obesity. Specifically, we found that the interaction between SNAP par­ ticipation and marginal, low, or very low food security was associated with higher dietary quality and lower BMI. The interaction b e­ tween SNAP participation and food insecurity was significantly associated with a lower likeli­ hood of obesity only among the marginal food security group. This result aligns with Hanson et al., who studied the interaction between food insecurity, marital status, and body weight and found th a t food insecurity was related to a greater likelihood of obesity among married women with marginal food security.28 Because the recession of 2 0 0 7 to 2 0 0 9 was associated with a record high rate of job loss, low rate of reemployment, and substantial earnings losses,29 the population of persons temporarily experiencing marginal food secu­ rity is expected to grow.30 SNAP could play a prominent role in ensuring that this popula­ tion has the necessary resources to obtain a nutritionally adequate diet during difficult times.31 On the other hand, it is important to understand the reasons some participants are still unable to consume healthy food, whether it is because of inadequate SNAP benefit, insuf­ ficient time to shop for and prepare nutritious meals, or lack of nutrition knowledge and budgeting skills.

Our results also showed th a t adults with­ out full food security had a higher intake of total calories, added sugar, and empty calo­ ries than those with full food security. Re­ search shows th at food insecurity, often a cyclic phenom enon, is associated with preferences for energy-dense foods, because adults who anticipate future food scarcity often overconsume w hen food is available.32 Moreover, food-insecure persons, who are often low income, may be hesitant to p u r­ chase nutrient-rich foods such as fruits and vegetables, which cost m ore p er calorie than energy-dense foods with minimal nutritional values.33-35 Those who experience food in­ security may also n o t have the means to travel to buy food frequently and may opt to purchase nonperishable o r canned products or energy-dense foods th at are less healthy, y et less costly.36 July 2 0 1 5 , Vol 1 0 5 , No. 7 | Am eric a n Journal o f P ublic Health Nguyen et ai. | Peer R eviewed ¡ R esearch an d P ra ctice ¡ 1 4 5 5 TABLE 2 - F o o d S e c u r ity , D ie ta r y Q u a lity , a n d W e ig h t S t a tu s A m o n g U S L o w -In c o m e A d u lts : N a t io n a l H e a l t h a n d N u tr itio n E x a m in a tio n Survey, 2 0 0 3 - 2 0 1 0 F ull S a m p le (n = 8 3 3 3 ) , N o. o r %Fu ll Food S e c u r it y (n = 4 6 4 5 ) , No. o r %Food In s e c u rity (n = 3 6 8 8 ) , No. o r %Pa D ie ta r y q u a lity H e a lth y E a tin g In d e x 2 0 1 0 , t o t a l s c o r e4 5 . 44 6 . 64 3 . 7< . 0 0 1 A d d e d s ug ar, te a s p o o n s2 0 .11 8 .72 2 . 0< . 0 0 1 S o lid f a t , g3 9 9 . 53 9 8 . 14 0 1 . 5.7 6 8 E m pty c a lo r ie s , k c al7 5 4 . 67 3 1 . 57 8 7 . 9< . 0 0 1 T o ta l c a lo r ie s , k c a l2 1 2 8 . 82 1 0 3 . 02 1 6 6 . 2. 0 8 4 W e ig h t (B M I) C o n tin u o u s2 8 . 82 8 .52 9 . 2< . 0 0 1 U n d e rw e ig h t ( < 1 8 .5 k g / m 2)2 . 32 . 32 . 4. 6 0 7 O v e rw e ig h t ( < 2 5 - < 3 0 k g / m 2)3 1 . 63 2 . 63 0 . 2. 1 7 8 O be se ( > 3 0 k g / m 2)3 5 . 63 3 . 73 8 . 4< . 0 0 1 Note. BM I = b o d y m a s s in d e x . R es ults t a k e su rv e y w e ig h ts in to a c c o u n t. N o t f u ll fo o d s e c u r ity g r o u p in c lu d e s p e o p le liv in g in h o u s e h o ld s w ith m a rg in a l fo o d s e c u rity , lo w f o o d s e c u r ity a n d very lo w f o o d s e c u rity .

’ D iffe r e n c e b e tw e e n f u l l f o o d s e c u r ity a n d a n y c a te g o ry o f f o o d in s e c u r ity d e riv e d fr o m S t u d e n t t te s t .

O ur subgroup analysis revealed th a t SNAP might affect ra c ial/eth n ic groups differentially:

interactions betw een SNAP participation and food insecurity benefited dietary quality and weight status am ong W hites (all food insecurity groups) to a m uch g reater extent th an am ong Blacks. Among Hispanics, SNAP participation was associated with im proved diet quality andw eight status only in households with marginal food security.

One possible explanation o f SNAP’s differen­ tial modification o f the association of food in­ security to dietary intake and weight status is neighborhood disparities in access to healthy food.3 7 Although low-income W hites te n d to live in neighborhoods with other socioeconomicgroups, low-income Blacks and Hispanics often live in segregated neighborhoods, especially in in n er cities.3 8 Studies have found th a t residents o f mixed-race o r solely Black neighborhoods (regardless of income) are less likely than those in predom inantly W hite communities to have access to healthy food choices,3 9 even if they have SNAP benefits.3 7 Many studies have TABLE 3 - M u l t i v a r i a b l e R e g re s s io n A n a ly s is on A s s o c ia tio n s o f S u p p l e m e n t a l N u tr itio n A s s is ta n c e P r o g ra m P a r t i c i p a t i o n a n d Food In s e c u r it y W it h D ie ta r y Q u a lit y a n d W e ig h t S t a t u s A m o n g U S L o w -In c o m e A d u lts : N a t io n a l H e a l t h a n d N u tr itio n E x a m in a tio n S u rvey, 2 0 0 3 - 2 0 1 0 D ie ta r y Q u a lit y W e ig h t S ta tu s H e a lth y E a tin g IndexA d d e d S u g a rS o lid FatE m pty C alo rie sB M I ( c o n tin u o u s ; O v e rw e ig h t Obese (n - 8 1 7 4 ) , (n - 8 3 3 3 ) , b (SE)(n - 8 3 3 3 ) , b (SE) (n = 8 3 3 3 ) , b (SE)(n = 8 3 3 3 ) , b (SE)n - 8 1 7 4 ) , b (SE) (n - 8 1 7 4 ) , b (SE) b (SE) SNAP- 3 . 1 8 * * ( 0 . 5 3 )2 . 5 3 * * ( 0 . 7 4 )1 8 . 6 5 ( 1 1 . 7 0 )5 3 . 3 4 * * ( 1 9 . 7 0 ) 2 . 1 0 ” ( 0 . 2 8 )- 0 . 0 3 ( 0 . 0 2 )0 . 1 2 * * ( 0 . 0 2 ) H o u s e h o ld fo o d s e c u rity M a rg in a l- 2 . 2 7 * * ( 0 . 5 5 )0 . 4 8 ( 0 . 7 7 ) - 1 4 . 3 1 ( 1 2 . 1 0 )- 1 5 . 7 9 ( 2 0 . 3 6 )0 . 6 3 * ( 0 . 2 9 )- 0 . 0 1 ( 0 . 0 2 )0 . 0 4 * ( 0 . 0 2 ) Low- 1 . 6 3 * * ( 0 . 5 3 )2 . 3 5 * * ( 0 . 7 3 )- 3 . 1 9 ( 1 1 . 5 2 )1 7 . 4 4 ( 1 9 . 3 9 )0 . 4 7 ( 0 . 2 8 )- 0 . 0 3 ( 0 . 0 2 )0 . 0 4 * ( 0 . 0 2 ) Very low- 2 . 5 9 * * ( 0 . 6 7 )4 . 9 4 * * ( 0 . 9 4 )1 3 .3 1 ( 1 4 . 7 7 ) 1 0 2 . 4 8 * * ( 2 4 . 8 7 )1 . 0 2 ” ( 0 . 3 5 )- 0 . 0 2 ( 0 . 0 2 )0 . 0 5 * ( 0 . 0 2 ) SNAP x m a rg in a l fo o d s e c u rity 3 . 4 6 * * ( 0 . 9 9 )- 1 . 4 2 ( 1 . 3 7 )1 5 . 8 4 ( 2 1 . 6 4 )0 . 5 4 ( 3 6 . 4 3 )- 1 . 8 3 ” ( 0 . 5 2 )0 . 0 0 ( 0 . 0 3 )- 0 . 0 9 * ( 0 . 0 3 ) SNAP x lo w f o o d s e c u rity 1 . 9 8 * ( 0 . 8 8 )- 3 . 8 8 * * ( 1 . 2 2 )- 2 1 . 5 0 ( 1 9 . 3 6 )- 6 7 . 5 6 * ( 3 2 . 5 8 )- 0 . 9 8 * ( 0 . 4 6 )0 . 0 0 ( 0 . 0 3 )- 0 . 0 5 ( 0 . 0 3 ) SNAP x very lo w f o o d s e c u rity3 . 8 4 * * ( 1 . 0 4 )- 2 . 9 9 * ( 1 . 4 4 )- 1 1 . 5 9 ( 2 2 . 7 6 )- 6 5 . 2 4 ( 3 8 . 3 2 )- 1 . 1 7 * ( 0 . 5 5 )0 . 0 2 ( 0 . 0 4 )- 0 . 0 6 ( 0 . 0 4 ) Note. B M I - b o d y m a s s in d e x ; FPL - f e d e r a l p o v e rty le v e l; SNAP - S u p p le m e n ta l N u t r it io n A s s is ta n c e P rog ram ; WIC « W o m e n , In fa n ts , a n d C h ild r e n p ro g ra m . R e s p o n d e n ts w e re ag e d 2 0 ye a rs o r o ld e r a n d h a d f a m ily in c o m e u n d e r 2 0 0 % o f t h e FPL. V a lu e s a r e c o e ffic ie n ts d e riv e d fr o m o r d in a r y le a s t s q u a re s re g re s s io n s . D e p e n d e n t v a r ia b le s w e re h e a lth y e a t in g in d e x (m a x im u m s c o r e - 1 0 0 ) , a d d e d s u g a r ( te a s p o o n s ) , s o lid f a t (g ra m s ) , e m p ty c a lo r ie s ( k ilo c a lo r ie s ) , B M I ( c o n t in u o u s v a lu e ) , o v e rw e ig h t (d u m m y v a r ia b le ) , a n d o b e s ity (d u m m y v a r ia b le ) . T hese d e p e n d e n t v a r ia b le s w ere re gresse d on v a r ia b le s in d ic a t in g SNAP p a r t ic ip a t io n , h o u s e h o ld s ’ f o o d s e c u r ity c a te g o r ie s ( d u m m y v a r ia b le s ) , a n d SNAP p a r t ic ip a t io n in te r a c te d w ith fo o d s e c u r ity c a te g o r ie s . C o n tro l v a r ia b le s w ere ag e , r a c e / e t h n ic it y , in c o m e (v ia p o v e rty in c o m e r a tio g r o u p s ) , m a r ita l s ta t u s , e d u c a tio n , in s u r a n c e s t a t u s , WIC p a r t ic ip a t io n , an d e m p lo y m e n t s ta t u s . R e s u lts t a k e su rv e y w e ig h ts i n t o a c c o u n t.

* P < . 0 5 ; ” P < . 0 1 .

1 4 5 6 I Research an d P ractice ¡ Peer R eviewed | Nguyen e t al.A m eric a n Journal o f Public H ealth | July 2 0 1 5 , Vol 1 0 5 , No. 7 TABLE 4 - M u l t i v a r i a b l e R e g re s s io n A n a ly s is on R a c i a l / E t h n i c D i f f e r e n c e s in A s s o c ia tio n s o f S u p p l e m e n t a l N u tr itio n A s s is ta n c e P ro g ra m P a r t i c i p a t i o n a n d Food In s e c u r it y W it h D ie ta r y Q u a lity a n d W e ig h t S t a tu s A m o n g U S L o w -In c o m e A d u lts : N a t io n a l H e a l t h a n d N u tr itio n E x a m in a tio n Su rv e y , 2 0 0 3 - 2 0 1 0 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ D i e t a r y Q u a l i t y _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ W e i g h t S t a t u s H e a l t h y E a t i n g A d d e d S u g a r , S o l i d F a t , E m p ty C a l o r ie s , O v e r w e ig h t , O b e s e , I n d e x , b ( S E ) b ( S E ) b ( S E ) b (S E ) B M I , b ( S E ) b ( S E ) b (S E ) N o n - H i s p a n i c B la c k s ( n - 1 7 6 8 ) S N A P x m a r g i n a l f o o d s e c u r i t y - 0 . 2 9 ( 1 . 7 9 ) - 2 . 8 5 ( 2 . 5 0 ) - 2 2 . 0 8 ( 4 1 . 8 1 ) S N A P x l o w f o o d s e c u r ity 0 . 0 0 ( 1 . 6 7 ) - 2 . 1 3 ( 2 . 3 3 ) 2 2 . 6 1 ( 3 8 . 8 5 ) S N A P x v e r y lo w f o o d s e c u r ity 3 . 0 4 ( 1 . 9 0 ) - 2 . 4 5 ( 2 . 6 5 ) - 1 . 4 8 ( 4 4 . 3 4 ) N o n - H i s p a n i c W h i t e s (n = 3 3 9 8 ) S N A P x m a r g i n a l f o o d s e c u r ity 5 . 2 9 * * ( 1 . 7 1 ) 1 . 5 5 ( 2 . 5 5 ) 2 9 . 1 5 ( 3 7 . 5 4 ) S N A P x l o w f o o d s e c u r ity 3 . 9 2 * ( 1 . 5 3 ) - 7 . 9 3 * * ( 2 . 2 9 ) - 3 4 . 3 8 ( 3 3 . 6 9 ) S N A P x v e r y l o w f o o d s e c u r ity 4 . 8 3 * * ( 1 . 7 6 ) - 5 . 0 1 ( 2 . 6 3 ) - 3 7 . 6 3 ( 3 8 . 7 3 ) H i s p a n i c s ( n - 2 8 0 6 ) S N A P x m a r g i n a l f o o d s e c u r ity 2 . 5 5 ( 1 . 6 0 ) - 3 . 1 5 ( 1 . 7 3 ) 4 8 . 4 5 ( 3 3 . 1 6 ) S N A P x l o w f o o d s e c u r ity 1 . 5 3 ( 1 . 4 2 ) - 1 . 2 6 ( 1 . 5 3 ) - 3 6 . 1 3 ( 2 9 . 4 8 ) S N A P x v e r y l o w f o o d s e c u r ity 1 . 4 3 ( 1 . 7 8 ) - 1 . 3 1 ( 1 . 9 3 ) 3 . 2 5 ( 3 7 . 0 1 )- 3 3 . 4 1 ( 6 9 . 7 8 ) - 0 . 6 9 ( 1 . 1 5 ) 0 . 0 3 ( 0 . 0 6 ) - 0 . 0 6 ( 0 . 0 7 ) - 1 6 . 6 9 ( 6 4 . 8 4 ) - 0 . 5 5 ( 1 . 0 7 ) 0 . 0 5 ( 0 . 0 6 ) - 0 . 0 8 ( 0 . 0 6 ) - 9 . 8 6 ( 7 4 . 0 1 ) - 0 . 3 9 ( 1 . 2 2 ) 0 . 0 7 ( 0 . 0 7 ) - 0 . 0 6 ( 0 . 0 7 ) 7 4 . 6 2 ( 6 4 . 8 9 ) - 2 . 5 9 * * ( 0 . 8 8 ) - 0 . 0 6 ( 0 . 0 6 ) - 0 . 0 8 ( 0 . 0 6 ) ■ 1 4 3 . 9 0 * ( 5 8 . 2 3 ) - 1 . 0 7 ( 0 . 7 9 ) - 0 . 0 7 ( 0 . 0 5 ) - 0 . 0 0 ( 0 . 0 5 ) - 1 2 7 . 0 6 ( 6 6 . 9 4 ) - 2 . 0 3 * ( 0 . 9 1 ) 0 . 0 4 ( 0 . 0 6 ) - 0 . 0 9 ( 0 . 0 6 ) - 3 4 . 1 2 ( 5 0 . 8 1 ) - 1 . 5 4 * ( 0 . 7 2 ) 0 . 1 4 * ( 0 . 0 6 ) - 0 . 1 2 * ( 0 . 0 6 ) - 3 8 . 2 7 ( 4 5 . 1 7 ) - 0 . 6 4 ( 0 . 6 4 ) 0 . 0 6 ( 0 . 0 5 ) - 0 . 0 5 ( 0 . 0 5 ) - 5 2 . 4 4 ( 5 6 . 7 1 ) - 0 . 2 4 ( 0 . 8 0 ) - 0 . 0 1 ( 0 . 0 7 ) - 0 . 0 1 ( 0 . 0 6 ) N o te .

FPL = f e d e r a l p o v e r t y le v e l; S N A P = S u p p l e m e n t a l N u t r i t i o n A s s is ta n c e P r o g r a m ; W I C - W o m e n , I n f a n t s , a n d C h ild r e n p r o g r a m . R e s p o n d e n t s w e r e a g e d 2 0 y e a r s o r o l d e r a n d h a d f a m i l y in c o m e u n d e r 2 0 0 % o f t h e FPL. V a l u e s a r e c o e f f i c i e n t s d e r i v e d f r o m o r d i n a r y le a s t s q u a r e s r e g r e s s io n s . D e p e n d e n t v a r i a b l e s w e r e h e a l t h y e a t i n g in d e x ( m a x i m u m s c o r e - 1 0 0 ) , a d d e d s u g a r ( t e a s p o o n s ) , s o l i d f a t ( g r a m s ) , e m p t y c a l o r i e s ( k i l o c a l o r i e s ) , B M I ( c o n t i n u o u s v a l u e ) , o v e r w e i g h t ( d u m m y v a r i a b l e ) , a n d o b e s it y ( d u m m y v a r i a b l e ) . T h e s e d e p e n d e n t v a r i a b l e s w e r e r e g r e s s e d o n v a r i a b l e s i n d i c a t i n g S N A P p a r t i c i p a t i o n , h o u s e h o l d s ’ f o o d s e c u r i t y c a t e g o r i e s ( d u m m y v a r i a b l e s ) , a n d S N A P p a r t i c i p a t i o n i n t e r a c t e d w it h f o o d s e c u r ity c a t e g o r i e s . C o n t r o l v a r i a b l e s w e r e a g e , r a c e / e t h n i c i t y , in c o m e (v ia p o v e r t y in c o m e r a t i o g r o u p s ) , m a r i t a l s t a t u s , e d u c a t i o n , i n s u r a n c e s t a t u s , W I C p a r t i c i p a t i o n , a n d e m p l o y m e n t s t a t u s . R e s u lt s t a k e s u r v e y w e ig h ts in t o a c c o u n t .

* P < . 0 5 ; * * / > < . 0 1 .

assessed food deserts (low-income areas with limited access to fresh, healthy, and affordable food) 4 0 '41 and policies, such as th e Healthy Food Financing Initiative, th a t have b e e n d e­ veloped to address these inequities.4 2 F ur­ therm ore, dietary intake disparities may turn into discrepancies in th e incidence and m a n ­ agem ent o f obesity, hypertension, diabetes, and o th e r diet-sensitive chronic diseases.4,3 2 Thus, fu rth e r research is need ed to b e tte r understan d why and how g overnm ent nutrition assistance program s such as SNAP can affect varying food-insecure populations differently and to assess th e possible solutions.

T h e D epartm ent o f Agriculture has continued to make im provem ents to th e SNAP-Education (SNAP-Ed) program, specifically to enhance the quality o f SNAP participants’ diets.16'4 3 ,4 4 T he m ain aim o f SNAP-Ed is to im prove th e likeli­ hood th a t SNAP participants will make healthier food choices within a limited b u d g e t4 3 Research show s th a t nutrition education program s can le a d to h e a lth ie r food choices a m o n g low- incom e households participating in SNAP,4 5 an d thus SNAP-Ed could aid SNAP participantsin meeting th e challenge o f consuming a health­ ful diet on a limited b u d g e t4 6 ,4 7 In addition, SNAP-Ed interventions have begun addressing environmental factors affecting dietary intake, such as providing access to more healthful foods (e.g., fruits and vegetables) in local c o m er or convenience stores, which are m ore prevalent in low-income neighborhoods.4 8 O ur findings are consistent with th e litera­ tu re suggesting th a t racial/eth n ic groups may differ in taking advantage o f governm ent as­ sistance programs, so interventions encourag­ ing increasing utilization of SNAP and SNAP-Ed and incorporating promotion of more healthful food consumption should be tailored to ethnic minority subgroups. These programs will help n o t only food-insecure populations, b u t also th e m arginally food secure, to achieve a m ore health fu l diet an d conseq u en tly im prove h e a lth outcom es an d well-being.

L i m i t a t i o n s As a self-reported dietary recall d ata set, NHANES may b e p ro n e to overestim ation of portion size and dietary intake and may notrep resen t longer-term dietary intake pat­ t e r n s4 9 " 51 In addition, we could n o t establish a causa] relationship because o f th e cross- sectional n a tu re o f th e data. Furtherm ore, we could n o t control for self-selection into SNAP, w hich could have b een affected b y such u n ­ observed factors as personal preferences and underlying health conditions. T herefore, o ur identification strategy, similar to m any studies in th e literature, 11 did n o t perm it us to identify th e causal effects o f SNAP participation on health behaviors and outcomes.

O ur results w ere derived from self-reported food security and SNAP participation status, which may reflect an individual’s own p ercep­ tion ra th e r than th e actual situation, and the re p o rte d n um bers w ere subject to possible m easu rem en t e rr o r (e.g., m isreporting o r mis- classification bias) . 5 2 For example, w eight sta­ tus may influence re p o rted food insecurity:

obese individuals may b e m o re likely to re p o rt food insecurity because of th e ir habits and perceptions ab o u t food consum ption.2 8 ,5 3 However, although o u r m easure o f food in­ security relied on self-report, 6 1 7 it a d h ered to July 2 0 1 5 , Vol 1 0 5 , No. 7 | Am eric a n Journal o f Public Health Nguyen e t al. | Peer Reviewed | R esearch an d P ractice | 1 4 5 7 the Department of Agriculture classification, which is regarded as the gold standard.3,6 Future research should aim to enhance un ­ derstanding of the interrelationship of SNAP and food insecurity with health outcomes and account for both selection bias and measure­ ment error, as in a few works on related topics.17,54 Conclusions W e found that among the food-insecure population, SNAP participation appears to buffer against poor dietary quality and obesity, particularly among non-Hispanic Whites and marginally food-secure Hispanics. Most impor­ tant, our research highlights the role that SNAP may play in helping individuals who are at risk for food insecurity to obtain a healthful diet and better weight status.

SNAP, food insecurity, obesity, dietary pat­ terns, food availability and access, and other factors should be considered together rather than separately, because these factors may interact in a complex relationship. ■ A bout t h e Authors Birth T. Nguyen and Kerem Shuval are with the Intramural Research Department, Economic and Health Policy Re­ search Program, American Cancer Society, Atlanta, GA.

Farryl Bertmann is a public health nutrition and dietetics research consultant, South Hero, VT. A m y L. Yaroch is with the Gretchen Swanson Center f o r Nutrition, Omaha, NE.

Correspondence should be sent to Binh T. Nguyen, PhD, American Cancer Society, 2 5 0 William St, Atlanta, GA 3 0 3 0 3 (e-mail: [email protected]). Reprints can be ordered at http://www.ajph.org by clicking the ‘Reprints" link.

This article was accepted January 15, 2015.

C on trib u to rs B. T. Nguyen conceptualized and designed the study, with contributions from the other authors, analyzed the data, and drafted the article. All authors interpreted the data and reviewed and revised the article.

Human P a rtic ip a n t Pro te c tio n No protocol approval was required because the study used de-identified, publicly available data.

R eferences 1. US Dept of Agriculture, Economic Research Service.

Definitions of food security. 2014. Available at: h t tp :// www.ers.usda.gov/topics/food-nutrition-assistance/food- security-in-the-us/definitions-of-food-security .aspx#.

U7VjvPldUSV. Accessed July 3, 2014.

2. Holben DH; American Dietetic Association. Position of the American Dietetic Association: food insecurity in the United States. J A m Diet Assoc. 2 0 1 0 ; 1 10(9): 1 3 6 8 - 1377.3. Coleman-Jensen A, Nord M, Andrews M, Carlson S.

Household food security in the United States in 2010.

2011. Available at: h ttp://ssm .com /abstra ct= 2116606 o r http://dx.doi.org. Accessed July 3, 2014.

4. Seligman HK, Laraia BA, Kushel MB. Food in­ security is associated with chronic disease among low- income NHANES participants./Nutr. 2 0 1 0 ; 140(2):

3 0 4 - 3 1 0 .

5. Laraia BA. Food insecurity and chronic disease. Adv Nutr. 2 0 1 3 ;4 (2 ):2 0 3 -2 1 2 .

6. Coleman-Jensen A, Nord M, Singh A. Household Food Security in the United States in 2012. Washington, DC: US Dept of Agriculture, Economic Research Service; 2013.

ERR-155.

7. US Dept of Agriculture. Supplemental Nutrition Assistance Program (SNAP). 2 0 1 3 . Available at: h ttp :// www.fhs.usda.gov/snap. Accessed July 10, 2014.

8. Larson NI, Story MT. Food insecurity and weight status among U.S. children and families: a review of the literature. A m J Prev Med. 2 0 1 1;40(2): 1 6 6 -1 7 3 .

9. Berkowitz SA, Seligman HK, Choudhry NK. Treat or eat: food insecurity, cost-related medication underuse, and unm et needs. A m J Med. 2 0 1 4 ;1 2 7 (4 ):3 0 3 -3 1 0 .e3 .

10. Gundersen C, Oliveira V. The food stamp program and food insufficiency. A m J AgrEcon. 2001 ;83(4):875- 887.

11. Dinour LM, Bergen D, Yeh M-C. The food insecurity-obesity paradox: a review of the literature and the role food stamps may play. J A m Diet Assoc. 2 0 0 7 ; 107(11 ): 1952—1961.

12. Nord M. How much does the Supplemental Nutri­ tion Assistance Program alleviate food insecurity? Evi­ dence from recent programme leavers. Public Health Nutr. 2 0 12;15(5):811 -8 1 7 .

13. Leung CW, Villamor E. Is participation in food and income assistance programmes associated with obesity in California adults? Results from a state-wide survey. Public Health Nutr. 2 0 1 1;14(4):645-652.

14. W ebb AL, Schiff A, Currivan D, Villamor E. Food Stamp Program participation but not food insecurity is associated with higher adult BMI in Massachusetts resi­ dents living in low-income neighbourhoods.

Public Health Nutr. 2 0 0 8 ;1 1 (1 2 ):1 2 4 8 -1 2 5 5 .

15. Leung CW, Ding EL, Catalano PJ, Villamor E, Rimm EB, Willett WC. Dietary intake and dietary quality of low-income adults in the Supplemental Nutrition Assis­ tance Program. A m J Clin Nutr. 2 0 1 2 ;9 6 (5 ):9 7 7 -9 8 8 .

16. Nguyen BT, Shuval K, Njike VY, Katz DL. The Supplemental Nutrition Assistance Program and dietary quality among U.S. adults: findings from a nationally representative survey. Mayo Clin Proc. 2 0 1 4;89(9):

1 2 1 1 -1 2 1 9 .

17. Kreider B, Pepper JV, Gundersen C, Jolliffe D.

Identifying the effects of SNAP (food stamps) on child health outcomes when participation is endogenous and m isreported./Am Stat Assoc. 2 0 1 2 ;1 0 7 (4 9 9 ):9 5 8 -9 7 5 .

18. Centers for Disease Control and Prevention Na­ tional Center for Health Statistics. National Health and Nutrition Examination Survey. 2 0 1 3 . Available at:

http://www.cdc.gov/nchs/nhanes.htm. Accessed May 12, 2013.

19. Gundersen C, Kreider B, Pepper J. The economics of food insecurity in the United States. Appl Econ Perspect Pol. 2011 ;33(3):281-303.20. US Dept of Agriculture. Food security in the U.S.— survey tools. 20 1 4 . Available at: http://www.ers.usda.

gov/topics/food-nutrition-assistance/food-security-in- the-us/survey-tools.aspx#.U7qhOPldUSV. Accessed July 7, 2014.

21. National Cancer Institute. Applied research: cancer control and population sciences. Healthy Eating Index.

2012. Available at: http://appliedresearch.cancer.gov/ tools/hei. Accessed February 4, 2014.

22. Guenther PM, Casavale KO, Kirkpatrick SI, et al.

Update of the Healthy Eating Index: H E I-2010. J Acad Nutr Diet. 2 0 1 3 ;1 1 3 (4 ):5 6 9 -5 8 0 .

23. World Health Organization. Global database on body mass index. 20 1 4 . Available at: http://apps.who.

int/bmi/index.jsp?introPage=intro_3.html. Accessed July 7, 2014.

24. Haines PS, Hama MY, Guilkey DK, Popkin BM.

Weekend eating in the United States is linked with greater energy, fat, and alcohol intake. Obes Res. 2003; 1 1(8):945-949.

25. Stata, Version 13 [computer program]. College Station, TX: StataCorp LP; 2013.

26. Centers for Disease Control and Prevention. Na­ tional Health and Nutrition Examination Survey. Over­ view of NHANES survey design and weights. Available at: http://www.cdc.gov/Nchs/tutorials/environm ental/ orientation/sample_design/index.htm. Accessed July 7, 2014.

27. SAS, Version 9.3 [computer program]. Cary, NC:

SAS Institute Inc; 2011.

28. Hanson KL, Sobal J, Frongillo EA. Gender and marital status clarify associations between food insecurity and body weight. J Nutr. 2 0 0 7 ;1 3 7 (6 ):1 4 6 0 -1 4 6 5 .

29. Färber HS. Job Loss in the Great Recession: Historical Perspective From the Displaced Workers Survey, 1 9 8 4 - 2010. Cambridge, MA: National Bureau of Economic Research; 20 1 1 . Working paper 17040.

30. Ganong P, Liebman JB. The Decline, Rebound, and Further Rise in SNAP Enrollment: Disentangling Business Cycle Fluctuations and Policy Changes. Cambridge, MA:

National Bureau of Economic Research; 2013. Working paper 19363.

31. Schmidt L, Shore-Sheppard L, Watson T. The effect of safety net programs on food insecurity. University of Kentucky Center for Poverty Research discussion paper series, D P2012-12. Available at: http://www.ukcpr.org/ Publications/DP2012-12.pdf. Accessed July 1, 2014.

32. Seligman HK, Schillinger D. Hunger and socioeco­ nomic disparities in chronic disease. N Engl J Med 2010; 3 6 3 ( l):6 - 9 .

33. Monsivais P, Drewnowski A. The rising cost of low- energy-density foods. J A m Diet Assoc. 2 0 0 7 ; 107(12):

2 0 7 1 - 2 0 7 6 .

34. Maillot M, Darmon N, Darmon M, Lafay L, Drewnowski A. Nutrient-dense food groups have high energy costs: an econometric approach to nutrient pro­ filing. J Nutr. 2 0 0 7 ; 137(7): 1 8 1 5 - 1 8 2 0 .

35. Carlson A, Frazäo E. Are healthy foods really more expensive? It depends on how you measure the price.

20 1 2 . US Dept of Agriculture, Economic Research Service. Economic information bulletin 96. Available at:

http://www.ers.usda.gov/m edia/6 0 0 4 7 4 / eib96_ 1 _.pdf.

Accessed August 8, 2014.

36. Coveney J, O’Dwyer LA. Effects of mobility and location on food access. Health Place. 2 0 0 9 ; 15(1):45—55.

1 4 5 8 j Research an d P ractice | Peer Reviewed | Nguyen e t al.Am eric a n Journal o f P ublic H e a lth | July 2 0 1 5 , Vol 1 0 5 , No. 7 37. Larson NI, Story MT, Nelson MC. Neighborhood environments: disparities in access to healthy foods in the U.S. A m JP revM ed. 2 0 0 9 ;3 6 (1 ):7 4 -8 1 .

38. Galvez MP, Morland K, Raines C, et al. Race and food store availability in an inner-city neighbourhood.

Public Health Nutr.

2 0 0 8 ; 1 1(6):624-631.

39. Baker EA, Schootman M, Bamidge E, Kelly C. The role of race and poverty in access to foods that enable individuals to adhere to dietary guidelines. Prev Chronic Dis. 2006;3(3):A76.

40. US Dept of Agriculture, Agricultural Marketing Service. Food deserts. 2014. Available at: http://apps.

ams.usda.gov/fooddeserts/foodDeserts.aspx. Accessed February 6, 2014.

41. Walker RE, Keane CR, Burke JG. Disparities and access to healthy food in the United States: a review of food deserts literature. Health Place. 2 0 1 0 ;1 6 (5 ):8 7 6 - 884.

42. Administration for Children and Families, Office of Community Services. Healthy Food Financing Initiative.

2011. Available at: http://www.acf.hhs.gov/programs/ ocs/resource/healthy-food-financing-initiative-O.

Accessed August 5, 2 0 1 4 .

43. US Dept of Agriculture. Supplemental Nutrition Assistance Program education guidance: Nutrition Education and Obesity Prevention Grant Program.

Available at: http://snap.nal.usda.gov/snap/Guidance/ FinalFY2015SNAP-EdGuidance.pdf. Accessed August 9, 2014.

44. Guthrie JF, Frazào E, Andrews M, Smallwood D.

Improving food choices—can food stamps do more?

Amber Waves. 2007;5 (2 ):2 2 -2 8 .

45. Long V, Cates S, Blitstein J, et al. Supplemental Nutrition Assistance Program Education and Evaluation Study (Wave II). Washington, DC: Dept of Agriculture, Food and Nutrition Service; 2013.

46. McLaughlin C, Tarasuk V, Kreiger N. An examina­ tion of at-home food preparation activity among low- income, food-insecure women. J A m Diet Assoc. 2003; 103(11):.

47. Rose D. Food Stamps, the Thrifty Food Plan, and meal preparation: the importance of the time dimension for US nutrition policy. J Nutr Educ Behav. 2007;39(4):

2 2 6 - 2 3 2 .

48. US Dept of Agriculture. SNAP-ED strategies and interventions: an obesity prevention toolkit for states.

Available at: http://snap.nal.usda.gov/snap/SNAP- EdlnterventionsToolkitpdf. Accessed December 9, 2014.

49. Faggiano F, Vineis P, Cravanzola D, et al. Validation of a method for the estimation of food portion size.

Epidemiology. 19 9 2 ;3 (4 ):3 7 9 -3 8 2 .

50. Mertz W, Tsui JC, Judd J, et al. W hat are people really eating? The relation between intake derived from estimated diet records and intake determined to maintain body weight. A m J Clin Nutr. 1 9 9 1;54(2):291 -2 9 5 .

51. Briefel RR, Sempos CT, McDewell MA, Chien S, Alaimo K. Dietary methods research in the Third Na­ tional Health and Nutrition Examination Survey: under­ reporting of energy intake. A m ] Clin Nutr. 1997; 65(4 suppl): 12 0 3 S -1 2 0 9 S .

52. Gundersen C, Kreider B. Food stamps and food insecurity what can be learned in the presence of non- classical measurement error? J Hum Resour. 2008; 4 3 ( 2 ):3 5 2 -3 8 2 .53. Kaiser LL, Townsend MS, Melgar-Quiñonez HR, Fujii ML, Crawford PB. Choice of instrument influences relations between food insecurity and obesity in Latino women. A m J Clin Nutr. 2004;80(5): 1 3 7 2 -1 3 7 8 .

54. Gundersen C, Kreider B. Bounding the effects of food insecurity on children’s health outcomes. J Health Econ. 2 0 0 9 ;2 8 (5 ):9 7 1 -9 8 3 .

Chronic Disease, Epidemiology, and Control, Third Edition B y P a tr ic k L. R e m in g to n , M D , M P H ; R o ss C. B ro w n s o n , P h .D ; a n d M a r k V. W egner, M D , M P H “This book is an indispensable tool f o r the practitioner in charge o f de­ veloping and implementing chronic disease programs. It describes the latest developments in the science o f chronic diseases, and puts a t your fingertips solid, evidence-based strategies th at can be p u t to use to com­ bat the growing chronic disease epidemic." — Victor D. Sutton, PhD, M PPA Director, Office o f Preventive Health — Mississippi Department o f Health an d President, National Association o f Chronic Disease Directors T h e t h i r d e d itio n o f C hronic Disease, Epidemiology, a n d C ontrol p r o ­ vides t h e r e a d e r w ith u p - t o - d a t e i n f o rm a tio n a b o u t c h r o n ic diseases.

L eading sc h o la rs a n d p r a c titio n e rs have c o n s o lid a te d i n f o rm a tio n f ro m c o u n tle s s rese a rc h s tu d ie s —p r o v id in g clear a n d con cise in f o rm a tio n a b o u t c h r o n ic disease causes, t h e i r c o n s eq u en ces , th e g r o u p s at h ig h e s t risk, a n d effective m e th o d s o f p r e v e n tio n . T h is b o o k p ro v id e s all th e clues a b o u t ways to r e d u c e th e b u r d e n o f c h r o n ic diseases, c o n d itio n s , a n d ris k factors.

©APHA PRESS O R D E R TODAY!

imprint o, NMERicNN public »..LT» .«oc,.T,o« g 5 9 p a g e s , s o f tc o v e r , 2 0 1 0 , IS B N 9 7 8 ‘ 0 ‘ 8 7 5 5 3 - 1 9 2 - 2 O R D E R O N L IN E : w w w . a p h a b o o k s t o r e . o r g E -M A IL A P H A @ P B D .C O M TEL: 8 8 8 .3 2 0 .A P H A FAX: 8 8 8 .3 6 1 .A P H A July 2 0 1 5 , Vol 1 0 5 , No. 7 | A m eric a n Journal o f P ublic Health Nguyen e t al. | Peer R eviewed | R esearch and Practice | 1 4 5 9 Copyright ofAmerican JournalofPublic Health isthe property ofAmerican PublicHealth Association anditscontent maynotbecopied oremailed tomultiple sitesorposted toa listserv without thecopyright holder'sexpresswrittenpermission. However,usersmayprint, download, oremail articles forindividual use.