Scholarly article on topic 'Improving Nutrition by Increasing Supplemental Nutrition Assistance Program Benefits'

Improving Nutrition by Increasing Supplemental Nutrition Assistance Program Benefits Academic research paper on "Agriculture, forestry, and fisheries"

CC BY-NC-ND
0
0
Share paper
Keywords
{}

Abstract of research paper on Agriculture, forestry, and fisheries, author of scientific article — Ann M. Collins, Jacob A. Klerman

The diets of Americans fall far short of recommended dietary guidelines, and those who live in low-income households have even poorer diets than higher-income households. Many low-income Americans rely on the Supplemental Nutrition Assistance Program (SNAP). The program’s dual goals are to improve food security and nutrition. Among the possible strategies to address dietary shortfalls among low-income Americans is to increase the SNAP benefit. This article uses data from the random assignment evaluation of the Summer Electronic Benefit Transfer for Children demonstration to add new insights on the impact of SNAP on diet quality for households receiving SNAP who also received SNAP-like benefits through Summer Electronic Benefit Transfer for Children. Households received $60 each month per eligible school-aged child. The objective of the evaluation was to see if Summer Electronic Benefit Transfer for Children improved children’s food security and nutrition. The evaluation surveyed these households to collect information about food expenditures, food security, and children’s diets. For households receiving SNAP in sites that used the SNAP Electronic Benefit Transfer delivery system, the analysis showed increases in food expenditures and decreases in levels of food insecurity. The analysis also indicates improvements in dietary quality among school-aged children, but the impacts were modest.

Academic research paper on topic "Improving Nutrition by Increasing Supplemental Nutrition Assistance Program Benefits"

american journal of

Preventive Medicine

|specialarticle|

Improving Nutrition by Increasing Supplemental Nutrition Assistance Program Benefits

Ann M. Collins, MA, Jacob A. Klerman, MA1

The diets of Americans fall far short of recommended dietary guidelines, and those who live in low-income households have even poorer diets than higher-income households. Many low-income Americans rely on the Supplemental Nutrition Assistance Program (SNAP). The program's dual goals are to improve food security and nutrition. Among the possible strategies to address dietary shortfalls among low-income Americans is to increase the SNAP benefit. This article uses data from the random assignment evaluation of the Summer Electronic Benefit Transfer for Children demonstration to add new insights on the impact of SNAP on diet quality for households receiving SNAP who also received SNAP-like benefits through Summer Electronic Benefit Transfer for Children. Households received $60 each month per eligible school-aged child. The objective of the evaluation was to see if Summer Electronic Benefit Transfer for Children improved children's food security and nutrition. The evaluation surveyed these households to collect information about food expenditures, food security, and children's diets. For households receiving SNAP in sites that used the SNAP Electronic Benefit Transfer delivery system, the analysis showed increases in food expenditures and decreases in levels of food insecurity. The analysis also indicates improvements in dietary quality among school-aged children, but the impacts were modest.

Am J Prev Med 2017;52(2S2):S179-S185. © 2016 American Journal of Preventive Medicine. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

CrossMark

INTRODUCTION

The diets of Americans fall far short of the recommended dietary guidelines, leading to poor health outcomes.1,2 Those living in low-income households have even poorer diets than higher-income households.3-7 The goal of the Supplemental Nutrition Assistance Program (SNAP) is "to alleviate hunger and malnutrition ... by increasing food purchasing power for all eligible households who apply for participation," as stated in the Food Stamp Act of 1977, as amended (P.L. 108-269). SNAP provides supplementary resources to be spent on food, with the expectation that households spend 30% of non-SNAP income on food.

Theoretically, the additional purchasing power provided by SNAP could have positive, negative, or neutral effects on the diet quality of participants. To the degree that "healthy" foods (e.g., fresh fruits and vegetables, whole grains) are more expensive than "unhealthy" foods (e.g., foods containing high levels of carbohydrates, sugars, and fats)8—and to the degree that SNAP participants find them more desirable—participants could use the additional resources to increase food purchases and resulting consumption that move their diets closer to recommended dietary guidelines. However, households receiving SNAP

may not desire more-expensive healthy foods and might instead use increases in SNAP to purchase additional quantities of the unhealthy foods9 that they already purchase, resulting in poorer diets. Finally, households might substitute for more-expensive unhealthy foods with the less-expensive options that they previously purchased (e.g., prepackaged baked goods with fresh ones), resulting in neither a positive nor a negative effect on diet quality.

Doubtful that raising SNAP benefits alone will address the nutrition gap for low-income households, policymakers and others have suggested restructuring SNAP in several ways, which include:

1. prohibiting the purchase of unhealthful foods with

SNAP benefits;

From the 1Abt Associates Social and Economic Policy Division, Cambridge, Massachusetts

Ann M. Collins is an independent consultant

Address correspondence to: Jacob A. Klerman, MA, Abt Associates, Social and Economic Policy Division, 55 Wheeler Street, Cambridge, MA 021387. E-mail: jacob_klerman@abtassoc.com.

This article is part of a supplement issue titled The Supplemental Nutrition Assistance Program's Role in Addressing Nutrition-Related Health Issues.

0749-3797/$36.00

http://dx.doi.org/10.10167j.amepre.2016.08.032

© 2016 American Journal of Preventive Medicine. Published by Elsevier Inc. Am J Prev Med 2017;52(2S2):S179-S185 S179

This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

2. making SNAP more like the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) by only allowing (some portion of) SNAP to be used for the purchase of healthful foods; and

3. providing incentives to encourage the purchase of healthful foods.10

These approaches are described in more detail in other articles that are part of this supplement. However, some worry that adding restrictions on foods that can be purchased with SNAP will reduce its use because of lack of availability of these foods or stigma. Others have expressed concern that some of these approaches will further increase the stigma of using SNAP, discouraging its use and worsening food security.9,11

The question of the relationship between SNAP participation and healthy eating has been the focus of research for many years. Many studies have attempted to determine the association between SNAP and diet quality. The results are mixed. In 2004, Fox and colleagues12 conducted a review of 26 studies, published between 1978 and 2002, on the impact of the Food Stamp Program (FSP, the former name of SNAP) on diets of FSP recipients. The authors determined that there was little evidence that participation in FSP "consistently affects" dietary intakes. The authors also noted that only a few of the studies reviewed looked at the impact of FSP on carbohydrates, fat, saturated fat, sodium, or fiber.

A more recent literature review was conducted by Andreyeva and colleagues,13 using 25 relevant studies published between 2003 and 2014. Sixteen of these studies compared the diet and nutrition of SNAP participants with income-eligible nonparticipants. Though there were some exceptions, most studies found no differences between the two groups in energy intake or fruit and vegetable consumption. Two studies that looked at whole grain intake found it to be lower in SNAP participants compared with income-eligible nonparticipants. Of the ten studies that looked at the consumption of sugar-sweetened beverages, six did not report a difference and four studies found higher rates of consumption among SNAP participants.

The majority of the papers reviewed by Andreyeva et al.13 relied on nationally representative data sets, such as the National Health and Nutrition Examination Survey. All but three studies attempted to adjust systematic variation between SNAP recipients and non-recipients. Three of the reviewed papers did more than regression correction for unobservable characteristics. Gregory and colleagues4 and Todd and Ver Ploeg14 both used instrumental variables based on interstate variation in SNAP policies. One article used maximum likelihood methods.15 Even these more methodologically sophisticated approaches may fail to address unobservable characteristics that are associated with both the choice

to participate in SNAP and healthy eating (Bitler16 came to a similar conclusion).

These unavoidable limitations may confound the studies' findings on the impact of additional SNAP assistance on nutrition. By contrast, the evaluation of the Summer Electronic Benefit Transfer for Children (SEBTC) provides random assignment evidence on the direction and magnitude of the impact of unrestricted food assistance on nutrition.17 As such, its design takes account of both observable and unobservable individual characteristics that also could affect impacts. The demonstration and the evaluation methodology are described below. The article then presents the findings for a subsample of the participating households—those who reported receiving SNAP at baseline and in sites that used the SNAP Electronic Benefit Transfer (EBT) system to deliver the summer benefits—to provide additional evidence about the potential impact of SNAP on food expenditures, food security, and more generally on nutritional outcomes.

SUMMER ELECTRONIC BENEFIT TRANSFER FOR CHILDREN DEMONSTRATION

Concerned about the food security of low-income children in the summer, when they did not have access to the National School Lunch Program or the School Breakfast Program, the U.S. Congress and the U.S. Department of Agriculture created the SEBTC demonstration, designed to address children's food security issues in the summer, when school was not in session. During the summers of 2011-2013, approximately 100,000 households in 16 sites were randomly assigned to receive SEBTC or be in a control group. Households were eligible to participate in the evaluation if they had school-aged children who were certified for the National School Lunch Program, School Breakfast Program, or both in the prior school year. These households had incomes < 185% of the federal poverty limit.

Grantees provided SEBTC in the form of an EBT card for the summer months. Grantees could choose to deliver the SEBTC benefits through either WIC or SNAP EBT systems. Households receiving SEBTC through the SNAP EBT system followed the SNAP rules and could purchase all SNAP-allowable foods (i.e., most foods with the exception of alcohol, nutrition supplements, hot food, and food to be eaten in the store). Conversely, households that received SEBTC through the WIC EBT system were limited in their choices of foods to a subset of items from the regular WIC food package, selected by the U.S. Department of Agriculture to be appropriate for school-aged children.

METHODS

Although the SEBTC impact study occurred over a 3-year period, this analysis used data from 2012 only from a subset of the 14 sites that participated in 2012. In that year, households were randomly

assigned to receive a summer monthly benefit of $60 per eligible child or be in a no-benefit control group.

For the impact analysis, the evaluation's data collection team surveyed the study households in the spring (before the school year ended) and again in the summer. The summer survey response rate was 80.3%. The final total sample (all study participants from all 14 sites) included 27,092 households. There were no cross-overs. Unless otherwise noted, this article only discusses impacts that are significant at p< 0.001. To measure household food security, the evaluation used the U.S. Department of Agriculture's 18-item U.S. Household Food Security Survey Module with a 30-day reference period, which reflects the food security status of adults and children in the household.18,19

This paper focuses on analyses of dietary outcomes for children, which were measured using questions on children's food consumption developed by the National Cancer Institute. These questions, drawn from the 2009-2010 National Health and Nutrition Examination Survey Multifactor Diet Screener,20 assess the intake of specific dietary factors associated with the 2010 Dietary Guidelines for Americans. Scoring procedures developed by the National Cancer Institute were used to convert the reports into estimated amounts of fruits and vegetables, whole grains, dairy items, and added sugars consumed per day.21,22

Finally, the impact analysis also assessed the demonstration's impact on household food expenditures and participation in food assistance programs (e.g., SNAP, WIC, Summer Food Service Program). All analyses used weights that account for deviations from simple random sampling and differential non-response. Details can be found in Collins and colleagues.17

For both continuous and binary outcomes, the analysis computed estimates using weighted least squares, with robust SEs accounting for non-normal residuals induced by the analysis of binary outcomes and the nested sample structure. Differences in average household outcomes between random assignment arms were considered statistically significant evidence of a non-zero impact if p< 0.05 in a two-tailed test. For more information about all years of the evaluation, refer to Collins and colleagues.17

Eight of the 14 sites participating in 2012 delivered benefits through the SNAP EBT model. The other six sites used the WIC EBT model and are not considered here. Although the full study describes the SEBTC's impacts on the entire sample (i.e., all households in both SEBTC-SNAP and WIC sites), this article restricts the sample to households who reported receiving SNAP at baseline and who received SEBTC-SNAP benefits. This subsample includes 9,124 households (4,510 in the $60 benefit group and 4,614 in the no-benefit group, SE=0.64). The treatment and control groups of this subgroup are comparable in baseline household characteristics (Table 1). Though previously published reports describe impacts on all outcomes of some subgroup,17,23-25 this article describes impacts on household food security, household food expenditures, and children's nutrition for this specific population—households receiving SNAP at baseline in SEBTC—most appropriate for the considered policy question. As would be expected if random assignment was properly implemented, in this subsample (and in the full data), there is no evidence of treatment/control imbalance.

RESULTS

For further context, it is helpful to note that households receiving SNAP had approximately 1.5 school-aged

children and therefore were issued approximately $90 in SEBTC benefits each summer month. This is roughly equivalent to one quarter of their reported SNAP benefits. Therefore, the SEBTC benefit was roughly analogous to increasing SNAP by 25%.

The theory of change posits that increased food expenditure results in improved food security or spending on more-expensive foods. For the subsample that is the focus of this article, the analyses showed that SEBTC resulted in an increase in food expenditures, the first step in the theory of change. As shown in Table 2, the average SEBTC benefit amount was $107, which resulted in a net increase in expenditures between the treatment and control group of $60 (SE=$1.01), implying that every dollar of SEBTC assistance increased household food expenditures by 56 cents.

The benefit was successful at improving the food security status of households (Table 3). For instance, SEBTC benefits reduced very low food security among children by 34%; the control group rate was 10.6% compared to 6.9% in the treatment group (Table 3; SE=0.64). Similarly, very low food security for adults was reduced by 36%, and for any member of the household by 35%.

In addition to increases in food expenditures and food security, among SNAP recipients in sites that used the SNAP model, SEBTC also resulted in moderate improvements in three of eight measured child nutrition outcomes, including an increase in children's fruit and vegetable consumption (with and without fried potatoes) and consumption of dairy products (Table 4). There was no impact on the consumption of whole grains, whether or not children usually drank nonfat or low-fat milk, or consumption of added sugars.

More specifically, SEBTC's impact on fruit and vegetable consumption without fried potatoes increased from 2.8 servings per day without SEBTC to 3.0 servings with SEBTC (SE=0.04, the size of the difference was the same for fruit and vegetable consumption with fried potatoes). The difference of 0.2 servings a day of fruits and vegetables excluding fried potatoes is roughly equivalent to one fifth of a cup of raw fruit or two fifths of a cup of salad greens. Neither the SEBTC group nor the control group was close to the recommended dietary intake of five daily servings of fruits and vegetables without fried potatoes.

Children in the SEBTC group consumed 0.10 more servings of dairy per day compared with the control group (2.4 servings versus 2.3 servings, SE=0.04). Both groups met the dietary recommendations of 2-3 cups of dairy per day (the recommended number of cups varies by children's age).

DISCUSSION

The analysis also showed that, among households reporting participating in SNAP, SEBTC SNAP did not

Source: SEBTC, Spring and Summer Survey, 2012.

Note: Boldface indicates statistical significance (p<0.05). The p-values are based on a test of the difference between treatment group households and control group households. The null hypothesis being tested is that the treatment-control difference is zero. Very low food security: the food intake of household members is reduced and their normal eating patterns are disrupted because the household lacks money and other resources for food. Low food security: Household members experience reduced quality, variety, or desirability of diet. Little or no indication for reduced food intake. Food insecurity: Household members experience either very low food security or food insecurity.

SEBTC, Summer Electronic Benefits for Children; SNAP, Supplemental Nutrition Assistance Program; WIC, Special Supplemental Nutrition Program for Women, Infants, and Children.

Table 1. Baseline Characteristics of Households Receiving SNAP at Baseline in SEBTC-SNAP Sites

Characteristics Control (sample size) Treatment (sample size) Difference SE p-value

Household composition

Single female-headed households 57.21% (4,485) 56.59% (4,600) 0.62 1.23 0.62

Single male-headed households 3.68% (4.485) 2.80% (4,600) -0.88 0.45 0.05

Two or more adults in the household 39.11% (4.485) 40.61% (4,600) 1.50 1.21 0.21

Household size 4.37 (4,510) 4.40 (4,615) 0.02 0.04 0.55

Number of children in the household 2.53 (4,510) 2.55 (4,615) 0.02 0.03 0.54

Age of oldest child Race/ethnicity of respondent Hispanic 12.22 (4,502) 12.33 (4,612) 0.11 0.11 0.28

27.62% (4,479) 27.64% (4,576) 0.02 1.17 0.99

Black non-Hispanic 25.23% (4,479) 25.87% (4,576) 0.65 0.80 0.42

White non-Hispanic 40.97% (4,479) 40.59% (4,576) -0.38 1.15 0.74

Other non-Hispanic 6.18% (4,479) 5.90% (4,576) -0.29 0.70 0.68

Education level of respondent

Less than high school 29.90% (4,488) 30.50% (4,596) 0.60 1.11 0.59

Completed high school Some college Completed college 31.55% (4,488) 32.93% (4,488) 5.61% (4.488) 31.53% (4,596) 32.24% (4,596) 5.73% (4,596) -0.02 —A ßQ 1.19 1.26 0.56 0.58 0.58 0.84

-0.69 0.11

Employment, income, and program participation

Employed adult in the household 61.12% (4,498) 60.03% (4,605) -1.09 1.23 0.38

Prior month's income as proportion of federal poverty line 0.66 (4,454) 0.65 (4,555) 0.02 0.01 0.12

Participation in WIC Household food security Very low food security 25.02% (4,509) 26.41% (4,609) 1.40 1.07 0.19

30.62% (4,510) 30.02% (4,615) -0.60 1.17 0.61

Food insecure (i.e., very low food security + low food security) 63.79% (4,510) 64.90% (4,615) 1.11 1.28 0.39

result in children's increased consumption of sugar-sweetened foods and beverages, considered to be unhealthy. Put differently, the evaluation did not indicate that

SEBTC SNAP's additional food assistance led households to purchase these foods with their increased benefits, as some policymakers and others have feared.

Table 2. Impact of SEBTC on Household Food Expenditures

Control (sample size) Treatment (sample size) Difference SE p-value % change9

Out-of-pocket monthly food expenditures Reported monthly SNAP assistance SEBTC benefits redeemed $239.06 (4,338) $371.24 (4,338) $0 (4,510) $202.18 (4,356) $362.37 (4,356) $106.82 (4.356) $37.78 5.18 <0.0001 <0.05 <0.0001 15.7 2.4 NA

-$8.86 $106.82 4.64 1.01

Total difference in food expenditures $611.38 (4,338) $671.55 (4,356) $60.17 6.01 <0.0001 9.8

Source: SEBTC, Spring and Summer Survey, 2012.

Note: Boldface indicates statistical significance (p<0.001). The p-values are based on a test of the difference between treatment group households and control group households. The null hypothesis being tested is that the treatment-control difference is zero. a"% change" is impact as a percentage of control group level.

NA, not applicable; SEBTC, Summer Electronic Benefits for Children; SNAP, Supplemental Nutrition Assistance Program.

Collins and Klerman / Am J Prev Med 2017;52(2S2):S179-S185 Table 3. Impact of SEBTC on Household Food Security

Source: SEBTC, Spring and Summer Survey, 2012.

Note: Boldface indicates statistical significance (p< 0.001). The p-values are based on a test of the difference between treatment group households and control group households. The null hypothesis beingtested is that the treatment-control difference is zero. Very lowfood security: The food intake of household members is reduced and their normal eating patterns are disrupted because the household lacks money and other resources for food. Lowfood security: Household members experience reduced quality, variety, or desirability of diet. Little or no indication for reduced food intake. Food insecurity: Household members experience either very low food security or food insecurity. a"% change" is impact as a percentage of control group level. SEBTC, Summer Electronic Benefits for Children.

Control (4,510) Treatment (4,614) Difference SE p-value % change'

Very low food security among children 10.55% 6.94% -3.61 0.64 <0.0001 -34.2%

Food insecurity among children 48.12% 37.47% -10.65 1.03 <0.0001 22.1%

Very low food security among adults 30.77% 19.47% -11.3% 0.89 <0.0001 36.7%

Food insecurity among adults 57.17% 42.63% -14.54 1.03 <0.0001 25.4%

Very low food security for any household member (children or adult) 32.13% 20.76% -11.37% 0.89 <0.0001 35.4%

Food insecurity for any household member (children or adults) 61.87% 48.66% -13.22% 1.04 < 0.0001 21.4%

Extrapolating to the SNAP program as a whole, the SEBTC evaluation suggests that an increase of 25% in SNAP may improve only some children's nutrition outcomes, and those only by a modest 6%-7%. In real terms, this change amounts, for instance, to an average one-fifth cup of fruits and vegetables daily and closes the gap between actual and recommended intake by about 7%. SEBTC SNAP resulted in an increase in children's consumption of less than a tenth of a serving of dairy each day. However, children's intake of dairy

(irrespective of whether it is low- or non-fat) does not appear to be a concern for this sample, as the control group appeared to consume amounts of dairy that were within the recommended dietary guidelines. In terms of the impacts of SEBTC SNAP on the consumption of sugar-sweetened foods, although the demonstration did no "good," it also did no "harm."

Even though the positive impacts on healthful foods were quite modest in size—especially compared with those achieved in the SEBTC WIC sites—a straight

Table 4. Impact of SEBTC on Children's Nutrition Among Households Receiving SNAP in SEBTC SNAP Sites

Control (sample size) Treatment (sample size) Difference SE p-value % changea

Fruits and vegetables (servings per day)b 2.95 (4,349) 3.14 (4,265) 0.19 0.04 <0.0001 6.4%

Fruits and vegetables without fried potatoes (servings per day)b 2.83 (4,358) 3.02 (4,269) 0.19 0.04 < 0.0001 6.7%

Whole grains (servings per day)b 1.78 (4,396) 1.94 (4,304) 0.16 0.08 0.0509 9.0%

Dairy products (servings per day)b 2.27 (4,410) 2.41 (4,309) 0.14 0.04 < 0.0001 6.2%

Usually drank nonfat or low-fat milk (%) 17.00 (4,256) 16.58 (4,238) -0.42 1.08 0.6953 -2.5%

Added sugars (teaspoons per day) 17.73 (4,324) 17.94 (4,229) 0.21 0.28 0.4599 1.2%

Added sugars excluding cereals (teaspoons per day) 16.49 (4,353) 16.53 (4,255) 0.04 0.24 0.8629 0.20%

Sugar-sweetened beverages (teaspoons per day) 7.40 (4,400) 7.22 (4,306) -0.18 0.25 0.4674 -2.4%

Source: SEBTC, Spring and Summer Survey, 2012.

Note: Boldface indicates statistical significance (p< 0.001). The p-values are based on a test of the difference between treatment group households and control group households. The null hypothesis beingtested is that the treatment-control difference is zero. a"% change" is impact as a percentage of control group level.

bDaily servings of fruits and vegetables and dairy are measured in cup equivalents and in ounce equivalents for whole grains, as defined by the 2010 Dietary Guidelines for Americans. One fruit and vegetable serving is 1cup raw or cooked fruit or vegetables, vegetable juice, or fruit juice; 2 cups leafy green vegetables; or V cup dried fruit. One dairy serving is 1 cup milk, fortified soy beverage, or yogurt; 1V ounces natural cheese; or 2 ounces of processed cheese.

SEBTC, Summer Electronic Benefits for Children; SNAP, Supplemental Nutrition Assistance Program.

increase in SNAP as a means to improve nutrition may be preferable to some policymakers and researchers, who oppose limiting SNAP purchases to healthful foods or restricting the use of SNAP for unhealthful foods on both practical and ethical grounds.10,11,26

Limitations

Unlike other studies of the impact of SNAP on nutrition, this study's random assignment design and large samples allow a rigorous estimate of food security and nutrition impacts. The study avoids the internal validity issues that impede the interpretation of the numerous previous studies reviewed by Fox et al.12 and Andreyeva and colleagues.13

However, the study collected data on only a limited number of nutrition outcomes and these were only in reference to school-aged children. Although there was a similar pattern in improvements in food security among children and adults in households that received SEBTC, there is no evidence of impacts on adult nutrition.

As with most other studies that assess nutrition outcomes and their relationship with SNAP, the SEBTC analysis relied on household self-reports of SNAP receipt. This reliance is part of a standard approach to understanding the effect of SNAP on nutrition, comparing nutrition between reported households receiving SNAP and those eligible who do not, controlling for observ-ables.12,13 Other authors have noted concerns that observable characteristics may not sufficiently control for differences between the two groups, potentially confounding findings about the causal impact of SNAP. There is a second problem with this strategy: SNAP is known to be substantially underreported in survey data.27 As a result, the "no SNAP" group includes SNAP recipients and further biases the estimated impacts.

Although the SEBTC evaluation also relied on self-report of SNAP participation, in this case, this critique is a second-order concern. In the standard analysis, under-reporting affects a key covariate, that is, the impact of SNAP on food choices.27 In this analysis, under-reporting only affects the sample on which the impacts are estimated. Specifically, the authors only use SNAP receipt to define which households to include in the sample for the analysis used for this article. The general direction of the nonreporting bias is under-reporting of SNAP receipt; this sample is likely to be almost all true SNAP recipients. Some households that actually receive SNAP are likely to be excluded from the analytic sample.

Finally, the study collected information about a short-term benefit. It is not clear whether nutrition impacts would increase, decrease, or stay the same with a longer-term intervention.

CONCLUSIONS

The SEBTC SNAP demonstration provided SNAP-like benefits in the summer to households <185% of the federal poverty limit with children receiving the school food programs during the school year. In 2012, the $60 monthly per child benefit amounted to approximately $90 per month per household, analogous to a 25% increase in SNAP. The evaluation was implemented with fidelity and had a large sample size.

This article restricted the analysis sample to households in sites using the SNAP EBT system and who reported receiving SNAP at baseline, prior to the SEBTC intervention. The analysis showed that although the SEBTC benefit resulted in increases in food expenditures and food security for this group, the SEBTC SNAP-like benefit resulted in only modest improvements in three of five nutrition outcomes collected by the study that are mostly considered healthy (fruits and vegetables with and without fried potatoes and dairy products).

ACKNOWLEDGMENTS

Publication of this article was supported by the Physicians Committee for Responsible Medicine. The findings and conclusions in this article are those of the authors and do not necessarily represent the official position of the Physicians Committee for Responsible Medicine.

This paper draws extensively on the Summer Electronic Benefit Transfer for Children Demonstration. We acknowledge funding for those efforts from the U.S. Department of Agriculture (USDA), Food and Nutrition Service.

Partial funding for preparation of this article was provided by the Physicians Committee for Responsible Medicine and by Abt Associates internal funds. The discussion reflects the position of the authors only. It does not represent the position of USDA or Abt Associates.

No financial disclosures were reported by the authors of this paper.

REFERENCES

1. Rehm CD, Monsivais P, Drewnowski A. Relation between diet cost and Healthy Eating Index 2000 scores among adults in 2007-2010. Prev Med. 2015;73:70-75. http://dx.doi.org/10.10167j.ypmed.2015.01.019.

2. U.S. DHHS, U.S. Department of Agriculture. 2015-2020 Dietary Guidelines for Americans. 8th ed. http://health.gov/dietaryguidelines/ 2015/guidelines/. Published December 2015. Accessed April 15, 2016.

3. Condon E, Drilea S, Jowers K, et al. Diet Quality of Americans by SNAP Participation Status: Data From the National Health and Nutrition Examination Survey, 2007-2010. Washington, DC: U.S. Department of Agriculture. www.fns.usda.gov/sites/default/files/ops/NHANES-S NAP07-10.pdf. Published 2015. Accessed April 15, 2016.

4. Gregory C, Ver Ploeg M, Andrews M, Coleman-Jensen A. Supplemental Nutrition Assistance Program (SNAP) Participation Leads to Modest Changes in Diet Quality. Washington, DC: U.S. Department of Agriculture, Economic Research Service. www.ers.usda.gov/media/ 1083334/err147.pdf. Published 2013. Accessed April 14, 2016.

17. Collins AM, Briefel R, Klerman JA, et al. Summer Electronic Benefit Transfer for Children (SEBTC) Demonstration: Summary Report. Alexandria, VA: U.S. Department of Agriculture, Food and Nutrition Service; forthcoming; 2016.

18. Bickel G, Nord M, Price C, Hamilton W, Cook J. Guide to Measuring Household Food Security. Alexandria, VA: U.S. Department of Agriculture, Food and Nutrition Service; 2000.

19. Coleman-Jensen A. Commemorating 20 years of U.S. food security measurement. Amber Waves. www.ers.usda.gov/amber-waves/2015-octo ber/commemorating-20-years-of-us-food-security-measurement. aspx.V17EmbEpCik. Published October 5, 2015. Accessed June 13, 2016.

20. National Cancer Institute. Dietary Screener Questionnaire (DSQ) in the NHANES 2009-10: DSQ & SAS programs. http://epi.grants.cancer. gov/nhanes/dietscreen/questionnaires.html. Published 2015. Accessed May 13, 2016.

21. National Cancer Institute. Dietary Screener Questionnaire (DSQ) in the NHANES 2009-10: data processing & scoring procedures. http://epi.grants.cancer.gov/nhanes/dietscreen/scoring/. Published 2015. Accessed April 14, 2016.

22. National Cancer Institute. Dietary screener questionnaire in the NHANES 2009-10: background. http://epi.grants.cancer.gov/nhanes/ dietscreen/. Published 2015. Accessed April 14, 2016.

23. Collins AM, Briefel R, Klerman JA, et al. Summer Electronic Benefits Transfer for Children (SEBTC) Demonstration: Evaluation Findings for the Full Implementation Year. Alexandria, VA: U.S. Department of Agriculture, Food and Nutrition Service. www.fns.usda.gov/sites/ default/files/SEBTC2012.pdf. Published 2013. Accessed April 14, 2016.

24. Collins AM, Briefel R, Klerman JA, et al. Summer Electronic Benefits Transfer for Children (SEBTC) Demonstration: Evaluation Findings for the Proof-of-Concept Year. Alexandria, VA: U.S. Department of Agriculture, Food and Nutrition Service. www.fns.usda.gov/sites/ default/files/SEBTC_Year1Findings.pdf. Published 2012. Accessed July 25, 2016.

25. Collins AM, Briefel R, Klerman JA, et al. Summer Electronic Benefits Transfer for Children (SEBTC) Demonstration: Findings for the Third Implementation Year. Alexandria, VA: U.S. Department ofAgriculture, Food and Nutrition Service. www.fns.usda.gov/sites/default/files/ops/ sebtc2013.pdf. Published 2014. Accessed July 25, 2016.

26. Barnhill A, King K. Evaluating equity critiques in food policy: the case of sugar-sweetened beverages. J Law Med Ethics. 2013;41(1):301-309. http://dx.doi.org/10.1111/jlme.12020.

27. Meyer BD, Mok WKC, Sullivan JX. The Under-reporting of Transfers in Household Surveys: Its Nature and Consequences. Cambridge, MA: National Bureau of Economic Research. www.nber.org/papers/ w15181. Published 2009. Accessed April 22, 2016.

5. Bhattacharya J, Currie J, Haider S. Poverty, food insecurity, and nutritional outcomes in children and adults. J Health Econ. 2004;23 (4):839-862. http://dx.doi.org/10.1016Zj.jhealeco.2003.12.008.

6. Han E, Powell LM, Isgor Z. Supplemental Nutrition Assistance Program and body weight outcomes: the role of economic contextual factors. Soc Sci Med. 2012;74(12):1874-1881. http://dx.doi.org/ 10.1016/j.socscimed.2012.02.032.

7. Ludwig DS, Blumenthal SJ, Willett WC. Opportunities to reduce childhood hunger and obesity: restructuring the Supplemental Nutrition Assistance Program (the Food Stamp Program). JAMA. 2012;308 (24):2567-2568. http://dx.doi.org/10.1001/jama.2012.45420.

8. Drewnowski A. The cost of U.S. foods as related to their nutritional value. Am J Clin Nutr. 2010;92(5):1181-1188. http://dx.doi.org/ 10.3945/ajcn.2010.29300.

9. Gundersen C. SNAP and obesity. In: Bartfeld J, Gunderson G, Smeeding TM, Ziliak JP, eds. SNAP Matters: How Food Stamps Affect Health and Well-Being. Palo Alto, CA: Stanford University Press; 2015.

10. Richards MR, Sindelar JL. Rewarding healthy food choices in SNAP: behavioral economic applications. Milbank Q. 2013;91(2):395-412. http://dx.doi.org/10.1111/milq.12017.

11. Schanzenbach DW. Proposals to ban purchase of sugary drink with food stamps won't work. Christ Sci Monitor. www.csmonitor.com/ Commentary/0pinion/2013/0313/Proposals-to-ban-purchase-of-su gary-drink-with-food-stamps-won-t-work. Published March 13, 2013. Accessed April 14, 2016.

12. Fox MK, Hamilton W, Biing-Hwan L, eds. Effects of Food Assistance and Nutrition Programs on Nutrition and Health: Volume 3, Literature Review. Alexandria, VA: U.S. Department of Agriculture, Economic Research Service. www.ers.usda.gov/media/873018/fanrr19-3_002.pdf. Published 2004. Accessed April 21, 2016.

13. Andreyeva T, Tripp AS, Schwartz MB. Dietary quality of Americans by Supplemental Nutrition Assistance Program participation status: A systematic review. Am J Prev Med. 2015;49(4):594-604. http://dx.doi. org/10.1016/j.amepre.2015.04.035.

14. Todd JE, Ver Ploeg M. Caloric beverage intake among adult Supplemental Nutrition Assistance Program participants. Am J Public Health. 2014;104(9):e80-e85. http://dx.doi.org/10.2105/AJPH.2014.301970.

15. Yen ST. The effects of SNAP and WIC programs on nutrient intakes of children. Food Policy. 2010;35(6):576-583. http://dx.doi.org/10.1016/j. foodpol.2010.05.010.

16. Bitler M. The health and nutrition effects of SNAP: selection into the program and a review of the literature on its effects. In: Bartfeld J, Gunderson G, Smeeding TM, Ziliak JP, eds. SNAP Matters: How Food Stamps Affect Health and Well-Being. Palo Alto, CA: Stanford University Press; 2015.