WIC Households’ Bread and Cold Cereal Purchases: When They Use Benefits Versus Paying Out of Pocket

Abstract

Purpose:

The US Department of Agriculture’s Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) seeks to improve the health of participating women and children by providing nutrition education and a monthly package of supplemental foods including whole grain bread and cereal. While some studies confirm that participants consume more whole grains, others find no effect. In this study, we hypothesize that the positive association between WIC and whole grains is being reduced in size and consistency by several factors.

Design/Setting/Participants:

American households were surveyed about their food purchases. Overall response rate was 45.6%. A total of 4826 households completed the survey including 471 WIC households.

Measures:

The survey recorded households’ purchases of refined and whole grains in bread and cereal over 1 week.

Analysis:

T tests were used to compare the bread and cereal purchases of WIC and eligible, non-WIC households. Probit models were also estimated to assess a WIC household’s likelihood to choose whole grain foods when using benefits versus other payment methods.

Results:

On average, WIC households acquired more whole grains in bread than eligible, non-WIC households (1.33 vs 0.72 ounce equivalents per household member aged 1 year or older; P < .05). No difference is found for cereal (P > .10). Moreover, when using payment methods other than WIC benefits, WIC participants are 19% less likely than other households to choose whole grain bread (P < .05) and 20% less likely to choose a whole grain cold cereal (P < .05), which suggests that WIC-provided foods may replace some whole grains participants would otherwise buy for themselves.

Conclusion:

WIC is positively associated with whole grains. However, the association is stronger for bread than cereal. Moreover, foods provided through the program may partially replace whole grains that WIC households would otherwise buy for themselves.

Keywords

WIC whole grains bread cereal food assistance nutrition interventions

Purpose

Among US food assistance programs, the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) has a unique potential to shape the diets of its participants.^1,2 Participating individuals, who include pregnant and postpartum women, infants, and children, receive supplemental foods and nutrition education at no personal cost. Supplemental foods include many staple items desired by young families such as infant formula, fluid milk, and breakfast cereal. Moreover, within each of these categories, the US Department of Agriculture (USDA) has selected individual food products that are high in nutrients inadequately consumed by the target population and low in overconsumed nutrients. For example, WIC allows only breakfast cereals high in iron and low in sugar. WIC is also the only US food assistance program mandated to provide nutrition education as a benefit to participants (eg, opportunities to meet with a nutrition counselor or take classes). Thus, WIC like no other US program has the potential to positively shape the diet quality of its participants, including over a quarter of all pregnant and postpartum women, over half of all infants, and over a quarter of all children less than 5 years of age nationwide.

WIC is administered at the federal level by USDA and at the local level by WIC State agencies. In 2009, following the recommendations of an ad hoc committee sponsored by the National Academies of Sciences, Engineering, and Medicine (NASEM),³ USDA revised its WIC food packages. Included in the revisions were changes aimed at increasing whole grains consumption among participating women and children. The 2015-2020 Dietary Guidelines for Americans stipulate that the majority of an individual’s grain consumption should be whole grains, which can help individuals maintain a healthy weight and lower their risk to develop type 2 diabetes and cardiovascular disease.^4
-6 However, Americans typically underconsume whole grains and overconsume refined grains.^4,7,8

USDA’s efforts to promote whole grains among WIC participants have focused on bread and breakfast cereal. For bread, USDA added 2 pounds of whole grain bread to its monthly WIC food packages for children aged 1 through 4 years and 1 pound to its monthly food packages for pregnant and breastfeeding women.⁹ Nearly all WIC State agencies also allow whole grain rolls and buns, brown rice, and corn or whole wheat tortillas as alternatives to bread.¹⁰ For the 36 ounces of breakfast cereal provided to all participating women and children, USDA stipulated that at least half of all cereals allowed by a WIC State agency must list whole grains as the primary ingredient, while all must be high in iron and low in sugar. All WIC-prescribed products deemed whole grain must furthermore meet US Food and Drug Administration requirements for labeling the product a “whole grain food with moderate fat content.”⁹

A large of body of research investigates whether WIC participants have been consuming more whole grains since the 2009 food package revisions, but study results are mixed. Some researchers find significant increases in WIC participants’ whole grains consumption, purchases, or expenditures.^11

-15 Other studies find no significant change or find that any significant changes are limited to a only subsample of the WIC population.^16
-18 That researchers do not find a strong and consistent association between WIC and whole grains is particularly curious, given that a majority of WIC participants redeem these benefits.^19,20

In this study, we hypothesize that the positive association between WIC and whole grains is being reduced in size and consistency by several factors. First, we hypothesize that the association may be stronger for bread than for cereal since USDA regulations allow WIC participants to redeem their benefits for refined grain cereals, but not for refined grain breads. Second, we hypothesize that WIC households may be less likely than other households to purchase whole grains when buying food “out of pocket” (defined to include using any payment method besides WIC benefits). WIC-provided foods are only supplementary. They do not meet 100% of an individual’s dietary requirements; rather a WIC participant’s overall intake of whole and refined grains depends on both foods acquired through the program and those purchased through other means. It is possible that WIC supplies most of the whole grain products that some households desire. If so, these WIC households may be less inclined to choose whole grains than other households when paying with cash or Supplemental Nutrition Assistance Program (SNAP) benefits. WIC-provided, supplemental foods might even replace some whole grains participants would otherwise buy themselves (assuming they had bought any) and thereby offset some of the program’s overall positive impact.

Methods

Design

The National Household Food Acquisition and Purchase Survey (FoodAPS) is a cross-sectional, nationally representative survey of US households’ food purchases and acquisitions, including foods acquired through USDA food assistance programs.²¹ Mathematica Policy Research (Mathematica) administered FoodAPS under contract with USDA. The survey took place between April 2012 and January 2013 with an overall response rate of 45.6%. Sample weights were adjusted to reduce the potential for nonresponse bias. Data quality and accuracy were independently assessed by Westat, also under contract with USDA.

FoodAPS was conducted under the Confidential Information Protection and Statistical Efficiency Act of 2002, which requires federal agencies that use the data to guard the confidentiality of survey respondents. The data have been previously used to investigate the cost of cold cereals purchased by WIC households²² and to broadly compare the food expenditure patterns of WIC and eligible, non-WIC households.^10,23

Sample

Because of USDA’s special interest in households eligible for and participating in its food assistance programs, FoodAPS oversampled low-income Americans. Among all 4826 households who completed the survey, 1581 (32.8%) participated in SNAP and 471 (9.8%) either self-identified as participating in WIC or reported using WIC benefits during the survey period. Also included among all 4826 households are another 266 (5.5%) who did not report participating in WIC and did not use WIC benefits during the survey period, but were income eligible (income <185% of poverty thresholds) and categorically eligible (include a female who is pregnant or postpartum and/or children aged less than 5 years old).

That WIC participants represent less than 10% of all FoodAPS households is not surprising, given that they account for only a small portion of the overall US population at any given time. In fiscal 2017, WIC participation averaged 7.3 million people per month² or approximately 2% to 3% of all 330 million Americans alive at the time. Indeed, because low-income Americans were oversampled, FoodAPS data include more WIC participants than a random sample of US households likely would. However, the data are not necessarily representative of WIC households nationwide since that was not a goal of the sample design.

Measures

Primary outcome measures include each household’s whole grain bread and cereal acquisitions. Surveyed households kept a diary in which they reported all foods acquired by all members over a 7-day period. Mathematica and USDA then matched each acquired food to USDA data sets including the Food and Nutrient Database for Dietary Studies (2011-2012) and the Food Patterns Equivalents Database (FPED 2011-2012). For this study, we defined bread to include loaves, buns, rolls, bagels, and English muffins. Cereals were defined to include cold cereals. Using the FPED (2011-2012), we then converted the weight of all acquired bread and cereal products into grain ounce equivalents. Finally, we classified each bread and cereal product purchased by each FoodAPS households as either a refined grain or a whole grain food if over half the grams from grains were whole grain.

A key independent variable of interest along with WIC participation status is the method of payment used in observed bread and cereal acquisitions. FoodAPS households reported all payment methods used on each shopping trip, including WIC benefits, cash, check, credit or debit card, gift card, or SNAP benefits (which work like cash at grocery stores except that they cannot be used to buy nonfoods, hot foods, or foods sold for on-premises consumption). However, this information is not specific to food products. On a single shopping trip, a household may have used both cash and WIC benefits. Following a past study,²² we assume that if a household used WIC benefits, those benefits financed all purchases of WIC-approved foods.

Analysis

Our main empirical analysis involved (1) comparing the amount of whole and refined grains acquired in bread and cold cereal by eligible, non-WIC and WIC households when using benefits versus paying out of pocket and (2) modeling the propensity of non-WIC and WIC households to choose whole grain products when using different methods of payment.

Some preliminary analysis of the data was also performed prior to our main empirical analysis. Because FoodAPS data are not necessarily representative of WIC households nationwide, we first compared our sample to USDA administrative data.²⁴ Since 1988, USDA has produced biennial reports on the characteristics of WIC participants. These reports include information about participant type (ie, mother, infant, or child), participant race and ethnicity, and the size of the household in which participants live. We then compared the WIC households in our data to all FoodAPS households and specifically eligible, nonparticipants. Finally, we examined the number of bread and cold cereal products each of these 3 groups of households purchased.

Comparing households’ bread and breakfast cereal purchases

For the first part of our main empirical analysis and to test our first research hypothesis, we compared the amounts of whole and refined grains acquired in bread and cereal by WIC and eligible, non-WIC households. USDA regulations allow WIC participants to redeem their benefits for refined grain cereals, but not for refined grain breads. Any positive association between WIC and whole grains may likewise be stronger for bread than for cereal. To control for differences in household sizes, we further divided each household’s grain acquisitions by the number of household members aged 1 year and older. Finally, we calculated average values over our 471 WIC households and our 266 eligible, non-WIC households using the software package Stata 14 with the “svyset” command and “over” option to incorporate sample weights that account for survey design. T tests were conducted to evaluate the statistical significance of any differences between the 2 groups.

Modeling a household’s propensity to choose whole grain products

For the second part of our main empirical analysis and to test our second research hypothesis, we modeled a household’s propensity to choose whole grain products. Because the program provides these foods to participating women and children at no personal cost, we expect WIC households are more likely than other households to buy whole grains when using benefits. However, we hypothesize that the opposite is true when WIC households pay out of pocket. It is possible that WIC benefits supply most of the whole grain products that some households desire. If so, these WIC households may be less inclined to choose whole grains than other households when paying with cash or SNAP benefits. WIC-provided, supplemental foods might even replace some whole grains participants would otherwise buy for themselves (assuming they had bought any) and thereby offset some of the program’s overall positive impact.

For both bread and cold cereal, we separately modeled the probability that a household chose a whole grain product, given that it bought any bread or cold cereal during the survey week (ie, a conditional probit model).²⁵ About 56% of households (2718/4826) bought bread and 26% (1275/4826) bought cold cereal. Of course, some households bought more than one bread or cold cereal product. Our models predict whether at least one of those purchases were a whole grain product (ie, a majority of all grams from grains were whole grain). Predictor variables included in the model capture a household’s participation in WIC and whether the household listed WIC benefits among the payment methods it used that week. Control variables are also included to account for potential confounders. Specifically, based on past research on Americans’ consumption of refined and whole grains,⁸ we control for a household’s income relative to poverty thresholds, level of education, knowledge of nutrition, and use of food labels.

Using data on all 4826 FoodAPS households, we estimated our model for both products using the software package Stata 14. Specifically, we employed Stata’s “heckprobit” and “svy” commands to incorporate sample weights into model estimation and generate robust standard errors that account for survey design. Model fit was judged by a Wald test for overall statistical significance and by evaluating the statistical significance of each individual predictor variable. However, in order to more easily interpret the relationship between those variables and the probabilities of interest, it was necessary to further calculate marginal effects: the change in the probability being modeled given a unit change in the predictor, holding all other predictors constant. For each binary predictor, such as WIC participation, we seek to measure the difference in the probability being modeled when the variable assumes different values (ie, WIC household = 1 and WIC household = 0).

Results

Preliminary analysis of the data revealed no substantial demographic differences between all WIC participants nationwide and WIC participants who completed FoodAPS (Table 1). For example, USDA reports that children accounted for 53.3% of all WIC participants in 2014 followed by women (23.6%) and infants (23%).²⁴ By contrast, in our 2012 to 2013 FoodAPS data, these proportions are 54.4%, 28.8%, and 16.8%, respectively.

Table 1.

Characteristics of WIC Households in the 2012 to 2013 National Household Food Acquisition and Purchase Survey (FoodAPS) Versus All WIC Households Nationwide, 2014.

Characteristic	USDA Administrative Data	FoodAPS
Participant type
Women (mothers)	23.6%	28.8%
Infants (age <1 year)	23.0%	16.8%
Children	53.3%	54.4%
Participant race and ethnicity
Hispanic	41.6%	39.3%
Non-Hispanic, black	20.3%	18.3%
Non-Hispanic, white	58.7%	63.0%
Other household characteristics
Household size	4.1 people	4.4 people

Abbreviations: USDA, US Department of Agriculture; WIC, Special Supplemental Nutrition Program for Women, Infants, and Children.

^a The 471 WIC households that completed FoodAPS included 669 individuals who reported participating in WIC. Some households had multiple beneficiaries (eg, both mother and child received benefits). USDA administrative data are from biennial reports on the characteristics of WIC participants (available at: https://fns-prod.azureedge.net/sites/default/files/ops/WICPC2014.pdf).

Preliminary analysis of the data also confirmed that our 471 WIC and 266 eligible, non-WIC households who participated in FoodAPS are demographically similar, although the WIC households had somewhat higher incomes (P < .05; Table 2). Households in both groups are larger and have less income as compared with all 4826 FoodAPS households (P < .05). The primary respondent in both types of household is also less apt to have completed college and is less likely to use the nutrition facts panel which lists the number of calories per serving, among other nutrition information, on most packaged food products (P < .05).

Table 2.

Characteristics of National Household Food Acquisition and Purchase Survey (FoodAPS) Participants, 2012 to 2013.^a

Characteristic	All Households (N = 4826)	WIC Households^b (n = 471)	Eligible, Non-WIC Households^c (n = 266)
Income relative to poverty threshold	381.7^d (16.61)	169.3^e (8.304)	108.2^f (5.885)
WIC household (0/1)	0.039^d (0.003)	1.0^e (0.0)	0.0^f (0.0)
Used WIC benefits during survey week (0/1)	0.014^d (0.002)	0.352^e (0.034)	0.0^f (0.0)
SNAP household (0/1)	0.136^d (0.009)	0.516^e (0.034)	0.521^e (0.046)
Used SNAP benefits during survey week (0/1)	0.083^d (0.005)	0.317^e (0.029)	0.349^e (0.04)
Meal planner finished college (0/1)	0.323^d (0.020)	0.066^e (0.012)	0.089^e (0.024)
Meal planner uses nutrition facts panel at least sometimes (0/1)	0.711^d (0.012)	0.543^e (0.033)	0.499^e (0.05)
Meal planner aware of MyPlate (0/1)	0.256^d (0.010)	0.297^d (0.03)	0.216^d (0.053)
Total number of people in the household	2.442^d (0.045)	4.609^e (0.122)	4.522^e (0.151)

Abbreviations: SNAP, Supplemental Nutrition Assistance Program; WIC, Special Supplemental Nutrition Program for Women, Infants, and Children.

^a Means are weighted. Standard errors in parentheses. Means that have different superscript letters (d, e, and f) are significantly different at the 5% level.

^b FoodAPS includes 471 households who either self-identified as receiving WIC benefits or reported using WIC benefits during the survey period.

^c FoodAPS includes 266 non-WIC households who are income eligible (income <185% of poverty thresholds) and categorically eligible (include a pregnant female and/or children aged less than 5 years).

Descriptive statistics further revealed that a similar percentage of WIC and eligible, non-WIC households reported buying bread and cold cereal (Table 3). Among the 471 WIC households who participated in FoodAPS, about 58% (273 /471) bought bread and 32% (153/471) bought cold cereal. Among the survey’s 266 eligible, non-WIC households, about 57% (151/266) bought bread and 38% (100/266) bought cold cereal.

Table 3.

Count of Participants in the National Household Food Acquisition and Purchase Survey (FoodAPS), 2012 to 2013, and Their Purchases of Bread and Cold Cereal.^a

Total number of households who participated in FoodAPS	4826
Number who purchased bread	2718
Bread transactions financed with WIC benefits^b	67
Bread transactions financed out of pocket	5653
Number who purchased cold cereal	1275
Cold cereal transactions financed with WIC benefits^b	112
Cold cereal transactions financed out of pocket	2394
Number of WIC households who participated in FoodAPS^c	471
Number who purchased bread	273
Bread transactions financed with WIC benefits^b	67
Bread transactions financed out of pocket	533
Number who purchased cold cereal	153
Cold cereal transactions financed with WIC benefits^b	112
Cold cereal transactions financed out of pocket	256
Number of eligible, non-WIC households who participated in FoodAPS^d	266
Number who purchased bread	151
Bread transactions financed with WIC benefits^b	0
Bread transactions financed out of pocket	376
Number who purchased cold cereal	100
Cold cereal transactions financed with WIC benefits^b	0
Cold cereal transactions financed out of pocket	200

Abbreviations: USDA, US Department of Agriculture; WIC, Special Supplemental Nutrition Program for Women, Infants, and Children.

^a Simple count of FoodAPS households and their purchases of bread and cold cereal. Sample weights not used.

^b FoodAPS reports all payment methods used on a shopping trip. Following Dong et al,²² if WIC benefits were among the reported payment methods, we assume that those benefits covered all WIC-allowed foods. In order to identify WIC-allowed foods, we further used Internet Archive’s Wayback Machine to examine food lists published by WIC State agencies between April 2012 and January 2013. For a small number of state agencies, we were unable to locate an historical food list. In these cases, foods were classified as WIC-allowed if they were allowed by any state agency, since those foods were at least on USDA’s list of WIC-authorized foods.

^c FoodAPS includes 471 households who either self-identified as receiving WIC benefits or reported using WIC benefits during the survey period.

^d FoodAPS includes 266 non-WIC households who are income eligible (income <185% of poverty thresholds) and categorically eligible (include a pregnant female and/or children aged less than 5 years).

Despite being largely similar except for their participation in the program, WIC households acquired 1.33 ounce equivalents of whole grains in bread per member aged 1 year or older, on average, versus 0.72 for eligible, non-WIC households. As shown in Table 4, this difference is statistically significant (P < .05) and reflects purchases of bread through the WIC program. As expected, all breads acquired by WIC households through the program were whole grain, but not all cold cereals. WIC households acquired 0.90 ounce equivalents of whole grains in cold cereal per member aged 1 year or older, on average, versus 0.80 for eligible, non-WIC households. However, this difference is not statistically significant (P > .10).

Table 4.

Grain Ounce-Equivalents Acquired in Bread and Cold Cereal Per Household Member Aged 1 Year or Older, National Household Food Acquisition and Purchase Survey (FoodAPS), 2012 to 2013.^a

	WIC Households^b (n = 471)	Eligible, Non-WIC Households^c (n = 266)	Difference^d
Bread purchased out of pocket
Refined grain	4.19 (0.47)	5.20 (0.53)
Whole grain	0.65 (0.15)	0.72 (0.17)
Bread acquired with WIC benefits^e
Refined grain	0 (0)
Whole grain	0.69 (0.17)
Total amount of bread acquired
Total refined grain	4.19 (0.47)	5.20 (0.53)	−1.01 (0.71)
Total whole grain	1.33 (0.22)	0.72 (0.17)	0.61^f (0.28)
Cereal purchased out of pocket
Refined grain	0.55 (0.09)	1.05 (0.20)
Whole grain	0.48 (0.11)	0.80 (0.14)
Cereal acquired with WIC benefits^e
Refined grain	0.34 (0.11)
Whole grain	0.42 (0.12)
Total amount of cereal acquired
Total refined grain	0.89 (0.14)	1.05 (0.20)	−0.16 (0.24)
Total whole grain	0.90 (0.17)	0.80 (0.14)	0.09 (0.22)

Abbreviations: USDA, US Department of Agriculture; WIC, Special Supplemental Nutrition Program for Women, Infants, and Children.

^a Means are weighted. Standard errors in parentheses. Numbers are rounded so that total grains may not precisely equal the sum of refined and whole grains.

^b FoodAPS includes 471 households who either self-identified as receiving WIC benefits or reported using WIC benefits during the survey period.

^c FoodAPS includes 266 non-WIC households who are income eligible (income <185% of poverty thresholds) and categorically eligible (include a pregnant female and/or children aged less than 5 years).

^d Statistical significance based on T tests. Samples are assumed to be independent. The standard error reported in parentheses was calculated as the square root of the sum of the 2 sample variances.

^e FoodAPS reports all payments methods used on a shopping trip. Following Dong et al,²² if WIC benefits were among the reported payment methods, we assume that those benefits covered all WIC-allowed foods. In order to identify WIC-allowed foods, we further used Internet Archive’s Wayback Machine to examine food lists published by WIC State agencies between April 2012 and January 2013. For a small number of state agencies, we were unable to locate an historical food list. In these cases, foods were classified as WIC-allowed if they were allowed by any state agency, since those foods were at least on USDA’s list of WIC-authorized foods.

^f Statistically significant at 5% level.

Although WIC provides bread and cereal to its participants at no personal cost, these foods are only supplementary. A program participant’s overall intake of grains depends on both foods acquired through the program and those purchased through other means. Results for our probit analysis in Table 5 reveal that during weeks when WIC households pay for all food purchases out of pocket, they are 19% less likely than other households to choose a whole grain bread (P < .05) and 20% less likely than other households to choose a whole grain cold cereal (P < .05). Control variables for education (P < .05), use of food labels (P < .05), and awareness of USDA’s MyPlate campaign to promote federal nutrition recommendations (P < .10) are also significant for bread. Controls for income relative to poverty thresholds (P < .10) and awareness of MyPlate (P < .05) are significant for cold cereal.

Table 5.

Marginal Effects for Probit Model of a Household’s Conditional Decision to Buy At Least 1 Whole Grain Product, National Household Food Acquisition and Purchase Survey (FoodAPS), 2012 to 2013.^a

Conditional Probability to Select a Whole GrainProduct	Bread	Cold Cereal
Income relative to poverty threshold	−0.00003 (0.00004)	0.0001^b (0.00006)
WIC household	−0.187^c (0.055)	−0.199^c (0.087)
Used WIC benefits during survey week	0.374^c (0.067)	0.102^c (0.038)
Meal planner finished college	0.069^c (0.034)	0.04 (0.036)
Meal planner uses nutrition facts panel at least sometimes	0.085^c (0.041)	−0.007 (0.038)
Meal planner aware of MyPlate	0.071^b (0.037)	0.065^c (0.027)
Results of Wald test for overall model fit	P = .0001	P = .0047

Abbreviations: SNAP, Supplemental Nutrition Assistance Program; WIC, Special Supplemental Nutrition Program for Women, Infants, and Children.

^a Since we only observe a household’s decision to choose whole grains if it made a purchase while participating in FoodAPS, we use a conditional probit model. A household’s likelihoods to have bought bread and cold cereal were hypothesized to depend on its income relative to poverty thresholds, household size, participation in WIC, whether the household reported using WIC benefits, participation in SNAP, and whether the household reported using SNAP benefits. Sample weights were used in estimation. Robust standard errors are reported in parentheses.

^b Statistically significant at 10% level.

^c Statistically significant at 5% level.

Discussion

Explaining Inconsistencies in the Existing Literature

Taken together, our results help to explain why studies do not consistently find a strong, positive association between WIC and whole grain consumption. As noted above, some studies find a positive association,^11

-15 but others find limited or no association.^16
-18 This may reflect in part USDA policies that allow WIC State agencies to include refined grain cereals on their food lists. WIC households appear to use their benefits to buy both types of cold cereal. Another reason why studies do not consistently find a strong, positive association is that increases in whole grain purchases during weeks when households use their benefits are somewhat offset by a reduced probability to choose whole grain products during other weeks when households pay for all foods out of pocket.

Obstacles to Increased Whole Grain Consumption

Our results underscore the perennial need for nutrition education. Low interest in and knowledge of nutrition are negatively associated with whole grain consumption (Table 5). Controlling for participation in WIC and use of WIC benefits, we find that bread-buying households are 9% and 7% more likely to choose a whole grain product, if their main meal planner uses the nutrition facts panel at least sometimes and is aware of USDA’s MyPlate campaign, respectively.

Practitioners may also want to consider behavioral economic approaches. Specifically, nutritionists may want to consider approaches that encourage households to choose the same WIC-approved foods when using their benefits and paying out of pocket, since WIC-eligible items within a category of food emphasize nutrients inadequately consumed by the target population (eg, WIC-eligible breads must be whole grain). Households, for example, might be encouraged to precommit to always buying the same, WIC-approved foods on all shopping trips, since research shows that households are likely to make healthier choices when they commit to buying foods ahead of time as opposed to making purchasing decisions on the spot.²⁶

New Recommendations

USDA is presently considering a new round of changes to its WIC food packages. In 2014, USDA again asked NASEM to conduct a review of its WIC food packages. NASEM responded by creating the Committee to Review WIC Food Packages which, in turn, released a 2017 report, Review of WIC Food Packages: Improving Balance and Choice: Final Report.¹⁰ Included in that report are additional recommendations to allow only whole grain breakfast cereals, tightening current restrictions that allow WIC State agencies to offer some refined grain products. In this study, we find no significant difference in the amounts of whole grain cold cereal acquired by WIC and eligible, non-WIC households (Table 4).

In its 2017 report, NASEM’s Committee to Review WIC Food Packages also recommends that WIC provide children aged 1 through 4 years, pregnant women, and breastfeeding women with 16 to 24 ounces of whole wheat bread.¹⁰ This would relax current regulations that effectively require WIC-authorized stores to stock 16-ounce loaves. NASEM found that whole wheat bread sold in 16-ounce loaves costs more money per pound than whole wheat bread in other package sizes. This is largely a cost containment issue for USDA. WIC participants receive a monthly food instrument—commonly, a paper voucher or electronic benefits—which they exchange at authorized retail stores for their prescribed food items at no personal cost. However, NASEM’s finding may still have implications for household behavior. In this study, we found that WIC households are less likely to choose whole grain bread when spending their own money as compared with all other households (Table 5). Tastes and preferences for refined grain products are one likely reason. The price issue identified by NASEM may also contribute. That is, if WIC-authorized bread is relatively expensive, then households will not likely get in the habit of buying the same bread products on all shopping occasions regardless of how they pay.

Limitations

National Household Food Acquisition and Purchase Survey (FoodAPS) is a cross-sectional survey of 4826 US households including 471 WIC and 266 eligible, non-WIC households. The overall response rate was 45.6%. Sample weights were adjusted to reduce the potential for nonresponse bias. When analyzed using these sample weights, FoodAPS is economically and demographically representative of US households. However, the data are not necessarily representative of WIC households. A comparison of USDA administrative data and WIC households who participated in FoodAPS revealed no substantial demographic differences (Table 1). Moreover, we note that other researchers have examined FoodAPS and found the data suitable for an analysis of WIC households.^10,22,23 These include NASEM’s Committee to Review WIC Food Packages which used FoodAPS to compare the food expenditure patterns of WIC and eligible, non-WIC households,^10,23 much like we use FoodAPS to compare grain acquisitions between these same 2 household groups in the current study. However, we cannot entirely dismiss the possibility that our results are influenced by data problems including nonrepresentativeness of the sample.

Directions for Future Research

That WIC households are less likely than other households to choose whole grain foods when paying out of pocket suggests WIC-provided supplemental foods may replace some whole grains participants would otherwise buy for themselves. But because our analysis describes the behavior of WIC households during weeks when they use and do not use benefits, our results are not necessarily a causal measure of the program’s effect on household behavior. This is because WIC participation is not randomly assigned. For example, it is possible that mothers who are more concerned about nutrition are more likely to join the WIC program and are also more likely to choose whole grains for themselves and for their children. If so, our means in Table 4 and our marginal effects in Table 5 may confound the effects of those omitted covariates with program effects. To fully guard against this possibility, future research might use an instrumental variable procedure or other corrective econometric method. Among studies of WIC households’ purchases of whole grains before and after 2009, we know of only one analysis to do so.¹¹ Of course, neither that study nor any previous study to our knowledge accounts for the differential behavior of non-WIC and WIC households when they use benefits and pay out of pocket.

So What? Implications for Health Promotion Practitioners and Researchers

What is already known on this topic?

Existing studies reach mixed results about whether WIC participation is associated with consuming more whole grains even though participants receive these foods at no personal cost.

What does this article add?

This is the first study to compare the bread and cereal purchases of non-WIC and WIC households when they use benefits and pay out of pocket. Our results help to explain why studies do not consistently find a strong, positive association between participation in WIC and whole-grain consumption. One explanation is that WIC households are less likely than other households to choose whole grains when they pay out of pocket. This, in turn, suggests that foods provided through the program partially replace whole grains that WIC households would otherwise purchase with other financial resources. Another contributing explanation may be USDA’s decision to allow WIC State agencies to offer refined grain cold cereals.

What are the implications for health promotion practice or research?

There is clearly a need for enhanced nutrition education to grow WIC households’ acceptance of whole grains. Practitioners may also want to consider behavioral economic approaches which take advantage of the psychological and emotional underpinnings of an individual’s food choices along with prices and income that economists have traditionally considered.

Footnotes

Authors’ Note

Any opinions, findings, recommendations, or conclusions are those of the authors and do not necessarily reflect the views of the Economic Research Service of the US Department of Agriculture (USDA-ERS).

Acknowledgments

The authors thank John Kirlin and Xingyou Zhang of the USDA-ERS for their assistance with the National Household Food Acquisition and Purchase Survey (FoodAPS).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Oliveira

Frazão

. The WIC Program: Background, Trends, and Economic Issues, 2015 Edition. Economic Information Bulletin no. 134. Washington, DC: USDA, Economic Research Service. 2015. https://www.ers.usda.gov/publications/pub-details/?pubid=43927. Accessed May 2018.

US Department of Agriculture, Food and Nutrition Service. About WIC. https://www.fns.usda.gov/wic/about-wic. 2017. Accessed October 2017.

Committee to Review the WIC Food Packages and Food and Nutrition Board. WIC Food Packages: Time for a Change. Washington, DC: Institute of Medicine, National Academies Press. 2005. https://www.fns.usda.gov/wic/wic-food-packages-time-change. Accessed October 2017.

US Department of Health and Human Services (HHS) and USDA. 2015-2020 Dietary Guidelines for Americans. 8th ed. Washington, DC. 2015. http://health.gov/dietaryguidelines/2015/. Accessed May 2018.

Chacko

Chou

Kugizaki

Liu

. Greater whole-grain intake is associated with lower risk of type 2 diabetes, cardiovascular disease, and weight gain. J Nutr. 2012;142(7):1304–1313.

Jonnalagadda

Harnack

Liu

. Putting the whole grain puzzle together: health benefits associated with whole grains—summary of American Society for Nutrition 2010 Satellite Symposium. J Nutr. 2010;141(5):1011S–1022S.

O’Neil

Nicklas

Zanovec

Cho

. Whole-grain consumption is associated with diet quality and nutrient intake in adults: the National Health and Nutrition Examination Survey, 1999-2004. J Am Diet Assoc. 2010;110(10):1461–1468.

Lin

Yen

. The US Grain Landscape: Who Eats Grain, In What Form, Where, and How Much? Economic Research Report no. 50. Washington, DC: USDA, Economic Research Service. 2007. https://www.ers.usda.gov/publications/pub-details/?pubid=45922. Accessed May 2018.

US Department of Agriculture, Food and Nutrition Service. Special Supplemental Nutrition Program for Women, Infants and Children (WIC): Revisions in the WIC food packages. Final Rule, 7 C.F.R. § 246.0. 2014. https://www.fns.usda.gov/wic/final-rule-revisions-wic-food-packages. Accessed October 2017.

10.

Committee to Review WIC Food Packages, Food and Nutrition Board, Health and Medicine Division. Review of WIC Food Packages: Improving Balance and Choice: Final Report. Washington, DC: National Academies of Science, Engineering, and Mathematics, National Academies Press. 2017. http://nationalacademies.org/hmd/Reports/2017/review-of-wic-food-packages-improving-balance-and-choice.aspx. Accessed October 2017.

11.

Jensen

Rahkovsky

. Did revisions to the WIC program affect household expenditures on whole grains? Appl Econ Perspect Policy. 2016;38(4):578–598.

12.

Ishdorj

Capps

. The effect of the revised WIC food packages on native American children. Am J Agric Econ. 2013;95(5):1266–1272.

13.

Andreyeva

Luedicke

. Federal food package revisions: effects on purchases of whole-grain products. Am J Prev Med. 2013;45(4):422–429.

14.

Chiasson

Findley

Sekhobo

. Changing WIC changes what children eat. Obesity (Silver Spring). 2013;21:1423–1429.

15.

Whaley

Ritchie

Spector

Gomez

. Revised WIC food package improves diets of WIC families. J Nutr Educ Behav. 2012;44(3):204–209.

16.

Tester

Leung

Crawford

. Revised WIC food package and children’s diet quality. Pediatrics. 2016;137(5):e20153557.

17.

Kong

Odoms-Young

Schiffer

. The 18-month impact of special supplemental nutrition program for women, infants, and children food package revisions on diets of recipient families. Am J Prev Med. 2014;46(6):543–551.

18.

Odoms-Young

Kong

Schiffer

. Evaluating the initial impact of the revised Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) food packages on dietary intake and home food availability in African-American and Hispanic families. Public Health Nutr. 2014;17(1):83–93.

19.

Phillips

Bell

Morgan

. Transition to EBT in WIC: Review of Impact and Examination of Participant Redemption Patterns. Ann Arbor, MI: Altarum Institute. 2014. https://altarum.org/publications/transition-to-ebt-in-wic-review-of-impact-and-examination-of-participant-redemption-patterns. Accessed May 2018.

20.

Ritchie

Whaley

Crocker

. Satisfaction of California WIC participants with food package changes. J Nutr Educ Behav. 2014;46(3S):S71–S78.

21.

US Department of Agriculture, Economic Research Service. FoodAPS National Household Food Acquisition and Purchase Survey. 2017. https://www.ers.usda.gov/data-products/foodaps-national-household-food-acquisition-and-purchase-survey.aspx. Accessed May 2018.

22.

Dong

Stewart

Frazão

. WIC Household Food Purchases Using WIC Benefits or Paying Out of Pocket: A Case Study of Cold Cereal Purchases. Economic Research Report no. 207. Washington, DC: USDA, Economic Research Service; 2016. https://www.ers.usda.gov/publications/pub-details/?pubid=45536. Accessed October 2017.

23.

Committee to Review WIC Food Packages, Food and Nutrition Board, Institute of Medicine. Review of WIC Food Packages: Proposed Framework for Revisions: Interim Report. Washington, DC: National Academies of Science, Engineering, and Mathematics, National Academies Press; 2015. http://nationalacademies.org/hmd/reports/2015/review-of-wic-food-packages-interim-report.aspx?_ga=2.155841595.1831752444.1517412413-1560374825.1514471028. Accessed January 2018.

24.

Thorn

Tadler

Huret

. WIC Participant and Program Characteristics 2014. Prepared by Insight Policy Research under Contract No. AG-3198-C-11-0010. Alexandria, VA: USDA, Food and Nutrition Service; 2015. https://fns-prod.azureedge.net/sites/default/files/ops/WICPC2014.pdf. Accessed October 2017.

25.

Van de Ven

Van Praag

. The demand for deductibles in private health insurance. J Econ. 1981;17(2):229–252.

26.

Sadoff

Samek

Sprenger

. Dynamic Inconsistency in Food Choice: Experimental Evidence from a Food Desert. Working Paper no. 2572821. Chicago, IL: University of Chicago, Becker Friedman Institute for Research in Economics; 2015.