4

Nutrient and Food Group Intakes of WIC Participants

Three criteria for food package revisions required that the committee evaluate nutrient and food intakes of the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC)-eligible population and assess their adequacy relative to dietary guidance: (1) the package contributes to reduction of the prevalences of inadequate nutrient intakes and of excessive nutrient intakes; (2) the package contributes to an overall dietary pattern that is consistent with the 2015–2020 Dietary Guidelines for Americans (DGA)¹ for individuals 2 years of age and older, and; (3) the package contributes to an overall diet that is consistent with established dietary recommendations for infants and children less than 2 years of age, including encouragement of and support for breastfeeding. In addition, the committee was tasked with evaluating the health needs of this population.

To address these criteria and requirements of the task, the committee collected and reviewed published evidence on key nutrition-related health priorities of relevance to the WIC population, reviewed published evidence describing WIC participants’ nutrient and food intakes, and conducted independent data analyses (examining nutrient intake, food intake, and diet quality) using data from the National Health and Nutrition Examination Survey (NHANES).

This chapter describes the committee’s subsequent findings and related challenges. The chapter concludes with a summary of findings from the committee’s evaluation. A description of how the committee applied this information to determine priorities for changes to the food packages is presented in Chapter 5.

__________________

¹ References to the DGA in this chapter are specific to 2015–2020 unless otherwise noted.

TABLE 4-1 Selected Nutrition-Related Health Risks Relevant to the WIC Population

NOTES: Relevant findings of the Dietary Guidelines for Americans, including shortfall nutrients identified for individuals 2 years of age and older, are reviewed in Chapter 3.

SOURCES: As noted in the Reference column.

LITERATURE AND REPORT REVIEW RELATED TO THE WIC POPULATION AND WIC PARTICIPANTS’ NUTRIENT AND FOOD GROUP INTAKES

Nutrition-Related Health Priorities Relevant to the WIC Population

The committee’s review of nutrition-related health risks and the corresponding prevalence of these risks among WIC participants (as available) was part of the phase I review (see Chapter 6 of NASEM, 2016). The review focused on health risks of population groups relevant to the WIC program that are not covered comprehensively in the DGA, namely pregnant women, breastfeeding women, infants, and children less than 2 years of age. The key findings of this review are summarized in Table 4-1. Relevant findings in the DGA, including shortfall nutrients identified for individuals 2 years of age and older (i.e., children and women who are not pregnant or breastfeeding), are reviewed in Chapter 2.

Food Safety Considerations Relevant to the WIC Population

In addition to reviewing nutrition-related health risks as part of its phase I study, the committee also reviewed health risks related to food safety that were relevant to the WIC-eligible population (see Chapter 6 of NASEM, 2016). This review was conducted with the understanding that the safety of the U.S. food supply is ensured by the U.S. Food and Drug Administration (FDA). Table 4-2 presents a summary of the key findings and recommended actions related to food-borne illness, pharmaceutical residues, and environmental contaminants (i.e., recommended not by this committee, but instead by referenced authorities or experts listed in the table).

Since the release of the phase I report, the FDA proposed an action level for inorganic rice in infant cereal (FDA, 2016). The announcement stated that “the majority of infant rice cereal currently on the market either meets, or is close to, the proposed action level” and offered guidance for parents

TABLE 4-2 Summary of Food Safety-Related Health Risks Relevant to the WIC Population

* Recommended not by this committee, but by referenced authorities or experts listed in the table.

SOURCE: As noted in the Reference column. See Chapter 6 of NASEM, 2016, for additional details of the committee’s health risk review.

and caregivers of infants (see Box 4-1). This guidance supports varying grain intake for both infants and pregnant women.

Nutrient Intakes of WIC Participants Before and After the 2009 Food Package Changes

As part of its first step toward evaluating whether the food packages meet the three criteria outlined in the introduction of this chapter, the committee evaluated the scientific literature and reports on nutrient intakes

among WIC participants. The committee identified three reports comparing nutrient intakes by infants or children before versus after the 2009 WIC food package revisions (see Table 4-3). Together, these studies indicate that there were some beneficial changes in food intake after the introduction of the new food packages, but specific findings were inconsistent from study to study. It is noteworthy that the committee was unable to identify any nationally representative “pre-post” (before and after the 2009 food package changes) studies of nutrient intake by WIC-participating women apart from the Diet Quality of Young American Children study (USDA/FNS, 2015a), for which sample sizes for both women and infants were too small to be reliable. As described later in this chapter, there are several challenges with collecting and comparing pre-post data in this context.

TABLE 4-3 Pre/Post Studies of 2009 Food Package Revision Effects on Nutrient Intake of WIC Participants

NOTES: Nationally representative data examining the effects of food package changes on nutrient intakes of WIC participants were not identified.

SOURCES: As noted in the Reference column.

Food Group Intakes of Women and Children Participating in WIC

The Effect of the 2009 Food Package Revisions on Food Group Intakes

Except for studies on breastfeeding, data characterizing the effect of the 2009 WIC food package changes on participants’ food intake or health are sparse. The data that do exist are from pre-post regional studies, and the outcomes are summarized in Table 4-4. These data indicate that the food package revisions were generally associated with improved intake of key food groups. In their systematic review, Schultz et al. (2015) examined the same studies and reported that there was an overall improvement in dietary intake after the 2009 food package changes, although the body of evidence was limited. The WIC food package changes were associated

TABLE 4-4 Regional Pre/Post Studies of 2009 Food Package Revision Effects on Food Intake of WIC Participants

SOURCES: As noted in the Reference column.

with a positive effect on the purchase of healthful foods targeted by WIC in WIC-participating households compared to WIC-eligible, nonparticipating households in two studies (Andreyeva and Tripp, 2016; Oh et al., 2016). Oh et al. (2016) reported an increase in household purchases of whole grain products using a national dataset. Andreyeva and Tripp (2016) reported an increase in purchases of healthful foods (categorized by sodium, saturated fat, and added sugars content) in two states. These two studies suggest potential positive effects of the food package changes. After its own independent review, the committee likewise concluded that the 2009 food package changes likely had some positive effects on intake.

The Effect of Racial and Ethnic Differences on Food Purchasing and Consumption Among WIC Participants

Findings from several reports suggest that food purchasing and consumption patterns may be strongly connected to culture, race, or ethnicity (Dubowitz et al., 2007, 2008; Bermúdez-Millán et al., 2009; Kong et al., 2013; Pooler and Gleason, 2014; Chaparro et al., 2015; Di Noia et al., 2016). This is evident in the WIC population. For example, Kong et al. (2013) reported that the diets of Hispanic mothers and children who participated in WIC were lower in the proportion of calories from fat, added sugars, sodium, and sweetened beverages and higher in whole grains, fruits, and dairy foods compared to their African-American counterparts. Reported differences in intake among and between racial and ethnic groups, however, are not always consistent (Faith et al., 2006; Odoms-Young et al., 2014; Chaparro et al., 2015; Cho et al., 2015).

Geographical Differences in Food Intakes Among WIC Participants

The committee identified one cross-sectional study on geographic differences in food intake. In a comparison of fruit and vegetable consumption between urban and rural African-American WIC participants in Texas, Ettienne-Gittens et al. (2013) found that urban African-American women consumed a wider variety of fruits compared to their rural counterparts. Urban children were provided with a wider variety of vegetables and consumed them more frequently than rural children.

Complementary Food Intakes of Infants Participating in WIC

It is recommended that the transition to intake of food begin at around 6 months of age (AAP, 2014), although parents often offer solid foods earlier than this time. The transition to the family diet lasts until a child is about 24 months of age (AAP, 2014).

As part of its phase I review (NASEM, 2016), the committee relied on food intake data from three large contemporary datasets for its evaluation of complementary food intakes among WIC-participating infants: (1) Infant Feeding Practices Study II (IFPS II) (Grummer-Strawn et al., 2008; CDC, 2014),² (2) 2008 Feeding Infants and Toddlers Study (FITS 2008)¹ (Briefel et al., 2010; Deming et al., 2014), and (3) NHANES (Grimes et al., 2015). A summary of the study designs and key results of these three studies are outlined in Appendix I, Tables I-1 through I-3.³ Details of the committee’s review are available in the phase I report (NASEM, 2016), and a summary of key findings is presented here. Additionally, for this report, the committee reviewed limited results from the currently underway WIC Infant and Toddler Feeding Practices Study (WIC ITFPS-2) (Personal communication, K. Castellanos-Brown, USDA-FNS, April 27, 2016).

Areas of Concern for Complementary Feeding

The committee’s review of these studies led to the identification of four areas of concern about complementary feeding: (1) early introduction of complementary foods; (2) insufficient intake of iron-fortified foods and supplements among older infants; (3) early introduction of cow’s milk; and (4) consumption of foods of poor nutritional value. Details of the committee’s evaluation leading to these areas of concern are provided in the phase I report (NASEM, 2016).

It should be reiterated that data collection for IFPS II, FITS 2008, and most of the NHANES analysis in Grimes et al. (2015) occurred before full implementation of the 2009 WIC food package revisions. Some of the changes in the food packages, such as not issuing complementary foods before an infant reached 6 months of age, have the potential to affect some of these areas of concern.

ANALYSIS OF THE NATIONAL HEALTH AND NUTRITION EXAMINATION SURVEY DATA: NUTRIENT ADEQUACY

As a second step toward assessing the adequacy of nutrient and food intakes of the WIC-eligible population, the committee estimated nutrient intake adequacy based on recommended Dietary Reference Intakes (DRIs)

__________________

² These data were collected prior to the 2009 food package changes.

³ The IFPS II analysis combined WIC with WIC-eligible, nonparticipating infants, and the results reflect all consumption in the 7 days before the survey. The FITS 2008 analysis described in this section, in contrast, compared WIC participants and eligible, non-WIC participants, and the data collected were for food intake only during the 24 hours before the interview.

(IOM, 1997, 1998, 2000a, 2001, 2002/2005, 2005, 2011). This section describes the methods and results of this analysis.

Analytical Methods in Brief

The U.S. Department of Agriculture’s Food and Nutrition Service (USDA-FNS) tasked the committee with two comparisons: (1) nutrient and food group intake of WIC participants compared to WIC-eligible nonparticipants and (2) nutrient and food group intake of WIC participants before the 2009 food package changes compared to after the 2009 food package changes. For this task, the committee analyzed the 2005–2012 data from the NHANES (USDA/ARS, 2005–2012). NHANES was used because it is the most comprehensive national nutrition survey and also because it captures reported household income and WIC participation. Portions of these data were used depending on the subgroup and comparisons of interest to this review (see Table 4-5). Full details on the methodology applied are described in Appendix J, and a description of the portion of data presented in this chapter can be found below in the section titled, “Nutrient and Food Group Intake Data Presented in This Chapter.”

This section highlights the following:

• The committee’s rationale for including in this chapter only nutrient and food intake data for WIC participants after the 2009 food package changes (other results are available in Appendix J);
• The analytical subgroups and NHANES survey years used for each subgroup analysis;
• The challenge of assessing nutrient intake of women who are breastfeeding at unknown intensity;
• The committee’s relaxed 5 percent threshold for identifying nutrient intake as being either inadequate or excessive.

Nutrient and Food Group Intake Data Presented in This Chapter

Results presented in this chapter are limited to the dietary intakes of WIC participants.⁴ Although intake estimates were generated for both WIC participants and WIC-eligible nonparticipants, interpretation of any differences is complicated by the potential for underlying differences between the two groups or selection bias. Selection bias results from the likelihood that differences in intake, even if statistically significant, result not from WIC participation, but from factors that caused an individual to participate in WIC. These comparisons could also be affected by challenges that the

__________________

⁴ Dietary supplement intake was not included in the analysis.

TABLE 4-5 NHANES Survey Years Applied for Each Analytical Subgroup

NOTES: NHANES = National Health and Nutrition Examination Survey.

^a The pre-2009 food package change results (NHANES 2005–2008) are available in Appendix J.

^b This group includes some WIC participants receiving the pre-2009 food package because the committee was unable to divide the 2009–2010 NHANES survey data set. As a result, the 2009–2010 NHANES release was included in the post-2009 food package change analysis to ensure adequacy of sample sizes.

committee experienced with correct identification of survey respondents as participating in WIC. Challenges related to selection bias are described in more detail later in this chapter.

Similarly, statistical comparisons of pre- to post-2009 intake data were considered inappropriate. For women and breastfed infants, small sample sizes required the committee to collapse multiple survey years (see Table 4-5); therefore, presented results do not uniquely represent pre- or post-2009 intake data. For other subgroups, results presented in this chapter are limited to nutrient intakes after the 2009 food package changes (that is, NHANES 2011–2012 data) because any detected differences before and after 2009 cannot necessarily be attributed to changes in the food packages. For example, as discussed in more depth in Chapter 2, adoption of the new food package in 2009 took place at the end of a recession and at a time when families were facing the worst labor market since the recession of the early 1980s. The American Recovery and Reinvestment Act of 2009 provided the funds necessary to increase the maximum benefit level of the Supplemental Nutrition Assistance Program (SNAP) by about 15 percent (EOPUS, 2014). SNAP recipients who meet the requirements for age and physiological state for WIC are automatically income-eligible for WIC. As

a result, because many WIC participants also receive SNAP benefits, food expenditures and consumption may have changed among those who were receiving both benefits. Additionally, the NHANES design is a repeated cross-sectional survey that does not allow for longitudinal analysis at any level (i.e., individual, state, or locality).

NHANES Analytical Subgroups

Using information available in NHANES, population subgroups were defined based on income, age, and physiological state required for WIC eligibility (i.e., women must be pregnant, breastfeeding, or postpartum). The committee finalized the subgroups as follows:

1. Women 19 to 50 years of age
   1. WIC participants⁵
   2. WIC-eligible (income ≤185 percent of poverty, pregnant, breastfeeding or postpartum) nonparticipants
   3. Low-income, WIC-ineligible (income ≤185 percent of poverty but neither pregnant, nor breastfeeding, nor postpartum) nonparticipants⁶
2. Formula-fed infants 0 to less than 6 months of age
   1. WIC participants
   2. WIC-eligible (income ≤185 percent of poverty) nonparticipants
3. Formula-fed infants 6 to less than 12 months of age
   1. WIC participants
   2. WIC-eligible (income ≤185 percent of poverty) nonparticipants
4. Breastfed infants⁷ 6 to less than 12 months of age
   1. WIC participants
   2. WIC-eligible (income ≤185 percent of poverty) nonparticipants
5. Children 1 to less than 2 years of age
   1. WIC participants
   2. WIC-eligible (income ≤185 percent of poverty) nonparticipants
6. Children 2 to less than 5 years of age
   1. WIC participants
   2. WIC-eligible (income ≤185 percent of poverty) nonparticipants

__________________

⁵ Survey respondents for which current WIC participation was reported in the NHANES survey.

⁶ The purpose of this subgroup is to anticipate needs of women who might become eligible at a later time with a change in physiological status, and therefore are potential WIC participants.

⁷ In the NHANES datasets, only intakes of formula and foods were quantified. Breastmilk intakes were not quantified for infants coded as breastfed, which posed a challenge for assessment of total nutrient intakes for these infants. Given that iron and zinc intakes are a concern for breastfed infants over 6 months of age, the committee analyzed intakes of these nutrients only for infants coded as breastfed.

NHANES Survey Years Applied

Table 4-5 summarizes the survey year data analyzed for each population subgroup and provides the rationale for each decision. All results from the analyses outlined in Table 4-5, including mean usual intakes and intake distributions of all subgroups, both before and after the 2009 food package changes (when possible) for both WIC and WIC-eligible subgroups, are available in Appendix J.

Challenges with Dietary Intake Assessment of Pregnant and Breastfeeding Women

For this report, nutrient intakes of all women coded as “pregnant” in NHANES were compared to the DRIs for pregnant women and to a 2,600-kcal food pattern.⁸ This corresponds to the energy needs of a pregnant woman in the second trimester of pregnancy and, thus, overestimates the needs of some pregnant women and underestimates the needs of others. The WIC pregnant women in NHANES were distributed among the trimesters of gestation. As a result, the estimates of food intakes below that recommended for pregnant women are not biased in one direction or another, but they are more imprecise than if trimester-specific estimates could have been generated. Unfortunately, the available sample sizes were inadequate for this purpose.

Intakes of all women coded as “breastfeeding” in NHANES were compared to the DRIs for exclusively breastfeeding women and to a 2,600-kcal food pattern. For breastfeeding women, NHANES identifies which women are breastfeeding, but not the intensity of their breastfeeding or the amount of milk they are producing. Therefore, in the tables in this report, the “BF” subgroup includes a mix of women who are exclusively and partially breastfeeding. For women who are partially breastfeeding and producing less milk than exclusively breastfeeding women, a 2,600-kcal diet is likely to be an overestimate of their caloric need. Given that only a minority of breastfeeding women in WIC are exclusively breastfeeding, the proportion of breastfeeding women whose nutrient intakes are inadequate and the proportion with food group intakes below that recommended are both overestimated.

__________________

⁸ The food patterns were selected because they most closely matched the estimated energy expenditure (EER) for the corresponding subgroups. To calculate the EER in this report, a “low-active” physical activity level was assumed. Additional details on the methodology applied for all NHANES analyses are presented in Appendix J.

Estimates of Inadequate and Excessive Nutrient Intakes: Using a Relaxed 5 Percent Threshold

The tables in this chapter show both the estimated prevalence of inadequate nutrient intakes and the estimated prevalence of excessive nutrient intakes for WIC subpopulations. As described in Appendix J (detailed methodology), the prevalence of dietary nutrient inadequacy is the estimated proportion of reported intakes below the Estimated Average Requirement (EAR); the prevalence of excessive intakes⁹ is the estimated proportion of the population with intakes above the Tolerable Upper Intake Level (UL) (IOM, 2000b). Inadequate or excessive intake of a nutrient typically becomes a concern when it occurs in 2.5 percent or more of the population of interest (IOM, 2003). However, for this report, a 5 percent threshold was applied. This slightly relaxed standard accounts for some of the uncertainty in setting the EARs as well as some of the generally accepted errors associated with dietary assessment. Inasmuch as inadequacy was 5 percent or greater for many nutrients, the committee considered two additional levels of inadequacy: (1) 10 to less than 50 percent and (2) inadequacy of 50 percent or greater. Nutrients with proportions of inadequacy in the latter category were considered of greatest concern.

Nutrient Intake Adequacy of Women Participating in WIC

Micronutrient Adequacy

As shown in Table 4-6, vitamin E intakes were inadequate in between 96 and 100 percent of women across subgroups. Additional inadequacies are noted below by physiological stage.

More than 50 percent of pregnant women participating in WIC reported intakes below the EAR for iron. The prevalence of inadequacy was between 10 and 50 percent for magnesium, folate, zinc, and vitamins A, C, and B6. Calcium intakes were below the EAR in only 6 percent of pregnant women.

More than 50 percent of breastfeeding women participating in WIC had intakes below the EAR for vitamin A. Inadequacy was between 10 and 50 percent for magnesium, zinc, vitamin C, vitamin B6, folate, copper, and calcium. Thiamin intakes were also below the EAR in approximately 10 percent of breastfeeding women.

__________________

⁹ Not all nutrients have ULs and, for four nutrients (folate, vitamin E, niacin, and magnesium), the UL is based on intake from supplements, fortificants, or pharmacological agents only (IOM, 1997, 1998, 2000a). Thus, the probability of exceeding the UL was determined only for retinol, vitamins C and B6, calcium, iron, phosphorous, zinc, copper, choline, and selenium in this report.

TABLE 4-6 Estimated Prevalence of Inadequacy of Selected Nutrients, WIC-Participating Women 19 to 50 Years of Age, NHANES 2005–2012

NOTES: αTOC = α-tocopherol; DFE = dietary folate equivalents; EAR = Estimated Average Requirement; N = sample size; RAE = retinol activity equivalents; SE = standard error.

^a Inadequacy, % = percentage of individuals with usual intake below the EAR.

^b Values represent the P/BF/PP groups. One value indicates that the EAR is the same across groups.

Subgroup definitions are as follows:

^c P = Pregnant WIC-participating women.

^d BF = Breastfeeding (any intensity), nonpregnant WIC-participating women.

^e PP = WIC-participating women who are 6 months postpartum, not pregnant, and not breastfeeding; data were sourced only from 2007–2012 NHANES because no postpartum variable is available in the 2005–2006 dataset.

SOURCES: IOM, 1997, 1998, 2000a, 2001, 2002/2005, 2005, 2011; USDA/ARS, 2005–2012.

More than 50 percent of postpartum women in WIC had intakes of magnesium, vitamin A, and calcium below the EAR. Intakes were inadequate for between 10 and 50 percent of postpartum women for many nutrients: vitamin C, folate, copper, zinc, thiamin, vitamin B6, vitamin B12, iron, and riboflavin.

Serum vitamin D concentrations are discussed later in this chapter and presented in Table 4-22.

Intakes of Micronutrients with an AI

For nutrients with only an Adequate Intake (AI) value, although the proportion of individuals with an inadequate intake cannot be determined, mean intakes at or above the AI imply that there is a low probability of inadequacy in the group (IOM, 2000b). As shown in Table 4-7, mean usual intakes of potassium and choline fell below the AI for all three subgroups of women. Intakes of sodium were far above the AI.

Macronutrients and Energy

The prevalence of inadequate dietary protein intakes ranged from 21.1 percent to 42.6 percent across the three groups of women (see Table 4-8). Mean intakes of fiber were below the AI in all subgroups of women. Approximately 68 to 78 percent of women had intakes of added sugars above recommended limits, whereas 50 to 79 percent of women had intakes of saturated fat above recommended limits.

Mean reported energy intake was below the median Estimated Energy Requirement (EER) by 10, 15, and 22 percent for pregnant, breastfeeding, and postpartum women, respectively (see Table 4-9).

TABLE 4-7 Reported Mean Usual Intakes of Selected Nutrients Compared to the Adequate Intake (AI) Value, WIC-Participating Women 19 to 50 Years of Age, NHANES 2005–2012

NOTES: AI = Adequate Intake; N = sample size; SE = standard error.

^a Values represent the AI for P/BF/PP groups. One value indicates that the AI is the same across groups.

Subgroup definitions are as follows:

^b P = Pregnant WIC-participating women.

^c BF = Breastfeeding (any intensity), nonpregnant WIC-participating women.

^d PP = WIC-participating women who are 6 months postpartum, not pregnant, and not breastfeeding; data were sourced only from 2007–2012 NHANES because no postpartum variable is available in the 2005–2006 dataset.

SOURCES: IOM, 1998, 2005; USDA/ARS, 2005–2012.

TABLE 4-8 Reported Intakes of Selected Macronutrients Compared to Recommended Intakes, WIC-Participating Women, 19 to 50 Years of Age, NHANES 2005–2012

NOTES: AI = Adequate Intake; DRI = Dietary Reference Intake; EAR = Estimated Average Requirement; g/d = grams per day; g/kg/d = grams per kilogram of body weight per day; kcal = kilocalories; SE = standard error; tsp-eq = teaspoon-equivalents.

^a Values represent a DRI. One value indicates recommendation is the same across P/BF/PP groups.

Subgroup definitions are as follows:

^b P = Pregnant WIC-participating women.

^c BF = Breastfeeding (any intensity), nonpregnant WIC-participating women.

^d PP = WIC-participating women who are 6 months postpartum, not pregnant, and not breastfeeding; data were sourced only from 2007–2012 NHANES because no postpartum variable is available in the 2005–2006 dataset.

^e Added sugars data were generated as part of the food group analysis. See methodology in Appendix J.

^f Tsp-eq and gram limits shown for added sugars and saturated fat, respectively, are based on energy intakes of 2,600 (pregnant and breastfeeding) and 2,300 (postpartum) kcals. The percent above 10 percent of energy is based on the limits corresponding to the actual energy intake reported in NHANES.

SOURCES: IOM, 2002/2005; USDA/ARS, 2005–2012; USDA/HHS, 2016.

TABLE 4-9 Estimated Energy Requirement and Reported Energy Intake, WIC-Participating Women, 19 to 50 Years of Age, NHANES 2005–2012

NOTES: kcal = kilocalories; N = sample size; SE = standard error.

Subgroup definitions are as follows:

^a P = Pregnant WIC-participating women.

^b BF = Breastfeeding (any intensity), nonpregnant WIC-participating women.

^c PP = WIC-participating women who are 6 months postpartum, not pregnant, and not breastfeeding; data were sourced only from 2007–2012 NHANES because no postpartum variable is available in the 2005–2006 dataset.

^d EERs were calculated assuming a low-active physical activity level. For pregnant women, EER calculations assumed the second trimester. For breastfeeding women, EER calculations assumed the first 6 month period postpartum.

SOURCES: USDA/ARS, 2005–2012; EERs calculated according to the method in IOM, 2002/2005.

Micronutrient Excess

The prevalence of excessive sodium intakes was greater than 50 percent across all three subgroups of women. Nearly all pregnant and breastfeeding women had excessive sodium intakes (see Table 4-10). Prevalences of intakes above the UL were less than 0.01 across subgroups for calcium, copper, iron, phosphorus, zinc, selenium, vitamin C, vitamin B6, retinol, and choline, and, thus, are not included in the table.

Nutrient Intake Adequacy of Formula-Fed WIC-Participating Infants

This section applies exclusively to infants in WIC who were coded as “formula-fed” in the NHANES datasets. This designation means that they received infant formula but is uninformative about the amount of breast milk that they may also have received. No infants in the dataset were characterized as “partially breastfed” even though this behavior is sufficiently common to have detected. As a result, the amount of breast milk that these

TABLE 4-10 Estimated Prevalence of Micronutrient Excess Compared to Tolerable Upper Intake Level (UL), WIC-Participating Women, 19 to 50 Years of Age, NHANES 2005–2012

NOTES: N = sample size; SE = standard error; UL = Tolerable Upper Intake Level.

Subgroup definitions are as follows:

^a P = Pregnant WIC-participating women.

^b BF = Breastfeeding (any intensity), nonpregnant WIC-participating women.

^c PP = WIC-participating women who are 6 months postpartum, not pregnant, and not breastfeeding.

SOURCES: IOM, 2005; USDA/ARS, 2005–2012.

infants may have received in addition to formula is unknown. The nutrient intakes of formula-fed infants were analyzed in two age groups: (1) from birth to less than 6 months of age and (2) from 6 to less than 12 months of age. These groups align with the recommended age for introduction of complementary feeding (about 6 months [AAP, 2014]) and also the current age categories for the WIC food packages for infants (see Chapter 1, Tables 1-2 and 1-3).

Formula-Fed Infants 0 to Less Than 6 Months of Age

Micronutrient intake compared to AIs Only AI levels (no EARs) are available for infants from birth to less than 6 months of age. These AIs are presented in Table 4-11 along with mean usual intakes for each nutrient. With the exception of choline, mean usual intakes for all nutrients exceeded these AIs.

Macronutrient and energy intake Macronutrient and energy intakes of infants up to 6 months of age are presented in Tables 4-12 and 4-13. Mean protein intakes exceeded the AI for this nutrient. Fiber, saturated fat, and added sugars were not evaluated for infants.¹⁰

__________________

¹⁰ There are no DRIs or dietary guidance for intake of fiber, added sugars, or saturated fat by infants. In addition, the AAP (2014) advises whole fat dairy products for children 1 to less than 2 years of age, and restriction of saturated fat generally begins at 2 years of age. Therefore, it would be difficult to assess intake adequacy or excess for these nutrients. In addition, infants begin complementary feeding at 4 to 6 months of age; it would be difficult to evaluate “mean” intake of nutrients from foods for a population subgroup that spans 0 to less than 6 months of age since some of these infants are not yet consuming any solid foods.

TABLE 4-11 Reported Mean Usual Intakes of Selected Micronutrients Compared to Adequate Intake (AI) Values, Formula-Fed, WIC-Participating Infants 0 to 12 Months of Age, NHANES 2011–2012

NOTES: αTOC = α-tocopherol; AI = Adequate Intake; DFE = dietary folate equivalents; EAR = Estimated Average Requirement; N = sample size; RAE = retinol activity equivalents; SE = standard error.

^a Values represent AI or EAR for infants <6 months and 6 to <12 months; iron and zinc values for infants 6 to <12 months are EARs; all other values are AIs.

^b Iron inadequacy was 4.2 percent.

^c Zinc inadequacy was less than 1 percent.

SOURCES: IOM, 1997, 1998, 2000a, 2001, 2002/2005, 2005, 2011; USDA/ARS, 2011–2012.

TABLE 4-12 Reported Mean Intakes of Macronutrients, Formula-Fed, WIC-Participating Infants 0 to 12 Months of Age, NHANES 2011–2012

NOTES: AI = Adequate Intake; EAR = Estimated Average Requirement; g/kg = grams per kilogram of body weight; kcal = kilocalories; N = sample size; NA = not applicable; SE = standard error.

* Values represent AI/EAR for infants <6 months and 6 to <12 months.

SOURCES: IOM, 2002/2005; USDA/ARS, 2011–2012.

TABLE 4-13 Estimated Energy Requirement and Reported Energy Intake, Formula-Fed, WIC-Participating Infants 0 to 12 Months of Age, NHANES 2011–2012

NOTES: kcal = kilocalories; N = sample size; SE = standard error.

SOURCES: USDA/ARS, 2011–2012; EERs calculated according to the method in IOM, 2002/2005.

Mean usual energy intake of WIC-participating infants less than 6 months of age was 692 kcal per day, which is slightly higher than the median EER of 617 kcal per day (see Table 4-13).

Micronutrient excess The prevalence of excessive micronutrient intakes compared to the UL for formula-fed infants are presented in Table 4-14. UL values for this age group have been defined only for calcium, iron, selenium, retinol, and zinc. Excess zinc intakes occurred in 88 percent and

TABLE 4-14 Estimated Prevalence of Micronutrient Excess Compared to the Tolerable Upper Intake Level (UL), Formula-Fed, WIC-Participating Infants 0 to 12 Months of Age, NHANES 2011–2012

NOTES: N = sample size; SE = standard error; UL = Tolerable Upper Intake Level.

* Values represent UL for infants <6 months and 6 to <12 months. One value indicates that the UL is the same across groups.

SOURCES: IOM, 1998, 2001, 2011; USDA/ARS, 2011–2012.

excess retinol intake in 40 percent of the formula-fed infants in this analysis. However, zinc and retinol intakes above the established ULs are not considered of concern because the method used to set the UL resulted in a narrow margin between the RDA and the UL and because retinol toxicity is rare unless from supplemental sources (IOM, 2001). Infant formula (and the zinc and other nutrients provided in it) is tightly regulated for safety by the Food and Drug Administration.

Formula-Fed Infants 6 to Less Than 12 Months of Age

Micronutrient adequacy Micronutrient EARs for this age group have been established only for zinc and iron (see Table 4-11). The prevalence of inadequate zinc and iron intake was low in formula-fed infants of this age group.

Intake of nutrients with an AI Mean usual intakes fell below the AI for choline but above the respective AIs for all other nutrients (see Table 4-11).

Macronutrient and energy intake Nearly 100 percent of the infants in this age group exceeded the EAR for protein (see Table 4-12). For other macronutrients, no DRI value (fiber, saturated fat) or only an AI value (i.e. carbohydrate) is published, and therefore the basis for evaluation of intakes is limited. As noted above, fiber, saturated fat, and added sugars were not evaluated for infants.

The mean energy intake of WIC-participating infants 6 to less than 12 months of age was 928 kcal per day, which is 28 percent higher than

the median EER of 727 kcal per day for this population subgroup (see Table 4-13).

Micronutrient excess The prevalence of excessive micronutrient intakes compared to the UL for infants in this age subgroup are presented in Table 4-14. Excess zinc intakes occurred in 85 percent and excess retinol intake in 28 percent of formula-fed infants 6 months and older. However, zinc and retinol intakes above the UL are not considered of concern for formula-fed infants for the reasons noted previously.

Evaluation of Energy and Iron Provided in the WIC Food Packages for Fully Formula-Fed Infants

Formula-fed infants ages 0 to 3 months of age receive 806 fluid ounces per month (537 kcal per day) as the full nutrition benefit¹¹ or, if the powder form is provided, 870 fluid ounces as the maximum monthly allowance (MMA).¹² Infants 4 to less than 6 months of age receive 884 fluid ounces per month (589 kcal per day) as the full nutrition benefit. These quantities of formula provide slightly less energy than the median EER determined for the WIC subgroup of infants ages 0 to less than 6 months (617 kcal per day [see Table 4-15]). WIC-participating infants 0 to less than 6 months of age who consume infant formula as their sole source of nutrition would be provided with 9.5 to 10.5 mg of iron per day, a quantity that is above the AI (0.27 mg per day)¹³ but below the UL (40 mg per day) for this population subgroup. WIC formula provided to infants ages 6 to less than 12 months provides 57 percent of energy needs, based on the EER for children participating in WIC, and slightly more than the EAR for iron. The committee presumed that infants begin to receive complementary foods between 4 and 6 months of age and that those complementary foods meet the balance of their needs for both energy and nutrients. Some additional detail related to the committee’s consideration of iron and energy provided by infant formulas is presented in Appendix Q.

__________________

¹¹ The full nutrition benefit (FNB) is the minimum amount of reconstituted fluid ounces of amounts for infant formula (based on a 13-ounce can), which formed the basis of substitution rates for other physical forms of infant formula (i.e., powder and ready-to-feed infant formula). The FNB is defined to ensure that participants receive a comparable nutritional benefit no matter which physical form of infant formula they receive.

¹² Formula provided to infants in any form may not exceed the maximum monthly allowance.

¹³ The Adequate Intake (AI) was set based on iron intake of fully breastfed infants. The iron bioavailability from human milk differs from that of formula. Healthy infants born to iron-sufficient mothers have iron stores adequate to meet their needs in the first 6 months of life.

Iron and Zinc Intakes of Breastfed Infants Participating in WIC

For older breastfed infants (6 to less than 12 months of age) only zinc and iron intake were examined because human milk contains low levels of these nutrients and, thus, they are of particular concern. This fact is recognized by the American Academy of Pediatrics (AAP), which indicates that breastfed infants 6 to 12 months of age who are not consuming the recommended amount of iron from formula and complementary foods may benefit from oral iron supplementation (AAP, 2014). To increase the sample sizes, data for breastfeeding infants were merged across NHANES 2009–2012 (see Table 4-5). As shown in Table 4-16, at least 10 percent of these infants had inadequate intakes of iron and zinc.

Nutrient Intake Adequacy of Children Participating in WIC, Ages 1 to Less Than 2 Years

Micronutrient Adequacy

For WIC-participating children 1 to less than 2 years of age, the estimated prevalence of inadequacy was less than 5 percent for all nutrients, with the exception of vitamin E (see Table 4-17).

Intakes of Micronutrients with an AI

Mean potassium intakes were below the AI for children 1 to less than 2 years of age (see Table 4-18). Mean intakes of sodium and choline were above the AI values.

Macronutrient and Energy Intake

Selected macronutrient intakes for this age group are summarized in Table 4-19. Protein intakes were adequate for all children, but fiber intakes fell below the AI. Added sugars and saturated fat intakes are not reported for children ages 1 to less than 2 years because the DGA limits for these nutrients apply only to children 2 years of age and older. Eighty percent of children ages 2 to less than 5 years of age exceeded the DGA limit of 10 percent of kilocalories from added sugars. Intake of saturated fat averaged 10.9 percent of kilocalories for children 2 to less than 5 years of age. Seventy percent of these children had saturated fat intakes above the recommended limit of 10 percent of kilocalories.

Mean usual energy intakes and the corresponding EER values are presented in Table 4-20. Estimated mean energy intake of children participating

TABLE 4-15 Energy and Iron Provided to Fully Formula-Fed WIC-Participating Infants in the Current WIC Food Packages, Compared to the EER and DRI

NOTES: AI = Adequate Intake Level; DRI = Dietary Reference Intake; EAR = Estimated Average Requirement; EER = Estimated Energy Requirement; FNB = full nutrition benefit; NA = not applicable; UL = Tolerable Upper Intake Level.

^a Unless otherwise indicated.

^b Based on the USDA-FNS Final Rule.

^c Values were based on reconstituted powdered formula, which is the most commonly issued formula in the WIC program; total energy therefore differs slightly from those presented in the phase I report.

^d Based on the baseline nutrient profile calculations, as detailed in Appendix R.

^e EER was calculated for WIC-participating infants using NHANES 2011–2012.

^f Based on the average amount of iron across formulas from the three major manufacturers that contain 20 kcal/oz: 1.76 mg iron per 100 kcal.

^g This volume is the maximum monthly allowance, which aligns with available sizes of powdered formula. Powdered formula is the form most commonly provided for infants ages 0 to 3 months.

^h Based on formula-fed infants ages 0 to less than 6 months in NHANES 2011–2012, n = 93. ⁱ Based on formula-fed infants ages 6 to less than 12 months in NHANES 2011–2012, n = 98.

^j An Adequate Intake value (a population mean intake exceeding this value implies a low prevalence of inadequacy).

^k An Estimated Average Requirement value (used to estimate the prevalence of inadequacy).

SOURCES: USDA/ARS, 2011–2012; USDA/FNS, 2014; EERs calculated according to method in IOM, 2002/2005.

TABLE 4-16 Iron and Zinc Intake of Breastfed WIC-Participating Infants 6 to Less Than 12 Months of Age, NHANES 2009–2012

NOTES: N = sample size; SE = standard error.

* Inadequacy, % = percentage of individuals with usual intake below the EAR.

SOURCES: IOM, 2001; USDA/ARS, 2009–2012.

TABLE 4-17 Estimated Prevalence of Inadequacy of Selected Nutrients, WIC-Participating Children, NHANES 2011–2012

NOTES: αTOC = α-tocopherol; DFE = dietary folate equivalents; EAR = Estimated Average Requirement; N = sample size; RAE = retinol activity equivalents; SE = standard error.

^a Inadequacy, % = percentage of individuals with usual intake below the EAR.

^b Values represent EAR for children 1 to 3 years and 4 to <5 years of age. One value indicates that the EAR is the same across groups.

SOURCES: IOM, 1997, 1998, 2000a, 2001, 2002/2005, 2005, 2011; USDA/ARS, 2011–2012.

TABLE 4-18 Reported Mean Intakes of Selected Micronutrients Compared to the Adequate Intake (AI) Value, WIC-Participating Children 1 to Less Than 5 Years of Age, NHANES 2011–2012

NOTES: AI = Adequate Intake; N = sample size; SE = standard error.

* Values represent AI for children 1 to 3 years and 4 to <5 years of age. One value indicates that the AI is the same across groups.

SOURCES: IOM, 1998, 2005; USDA/ARS, 2011–2012.

in WIC (1,314 kcal per day) was 43 percent higher than the median EER for WIC participants of this age group (917 kcal per day).

Micronutrient Excess

Among WIC-participating children 1 to less than 2 years of age, the prevalence of nutrient intakes exceeding the UL was more than 5 percent for sodium, zinc, copper, and retinol (see Table 4-21). As is the case with infants, zinc and retinol intakes above the UL are not of concern for children in this age group because of the way these values were derived. The UL for copper is an extrapolation of the UL for adults, and adult intakes of up to 12 mg of copper per day from food have not resulted in adverse effects (IOM, 2001). Therefore, copper intakes above the UL in this age group are likewise not considered to be of concern. Fewer than 0.01 percent of children ages 1 to less than 2 years exceeded the UL for iron, phosphorus, vitamin B6, and choline.

Nutrient Adequacy of WIC-Participating Children, Ages 2 to Less Than 5 Years

Micronutrient Adequacy

For children 2 to less than 5 years of age, there was a high prevalence of inadequate vitamin E intake (see Table 4-17). Intake inadequacy of other nutrients was below the threshold of 5 percent.

TABLE 4-19 Reported Intakes of Selected Macronutrients Compared to Recommended Intakes, WIC-Participating Children 1 to Less Than 5 Years of Age, NHANES 2011–2012

NOTES: AI = Adequate Intake; g/kg = grams per kilogram of body weight; kcal = kilocalories; N = sample size; NA = not applicable; SE = standard error; tsp-eq = teaspoon equivalents.

^a Numbers represent an EAR, AI, or upper limit as noted, for ages 1–3/age 4.

^b Added sugars data were generated as part of the food group analysis. See methodology in Appendix J.

^c The DGA limits for added sugars and saturated fat apply only to children 2 years of age and older. Tsp-eq and gram limits shown for added sugars and saturated fat, respectively, are based on an energy intake of 1,300 kcal. The percent above 10 percent of energy is based on the limits corresponding to the actual energy intake reported.

SOURCES: IOM, 2002/2005; USDA/ARS, 2011–2012; USDA/HHS, 2016.

Intakes of Micronutrients with an AI

For children ages 2 to less than 5 years of age, mean potassium and choline intakes were below the AI (see Table 4-18).

Macronutrient and Energy Intakes

Protein intakes were adequate for all children in this age group (see Table 4-19). Mean fiber intake was approximately half the AI. Usual mean energy intakes and the corresponding EER values are presented in Table 4-20. Reported mean energy intake (1,509 kcal) was comparable to the median EER (1,517 kcal).

TABLE 4-20 Estimated Energy Requirement and Reported Energy Intake, WIC-Participating Children 1 to Less Than 5 Years of Age, NHANES 2011–2012

NOTES: kcal = kilocalories; N = sample size; SE = standard error.

* EERs were calculated assuming a low-active physical activity level.

SOURCES: USDA/ARS, 2011–2012; estimated energy requirements calculated according to IOM, 2002/2005.

Micronutrient Excess

More than 5 percent of children in this age category exceeded the UL for a number of micronutrients evaluated for this report: retinol, zinc, copper, selenium, and sodium (see Table 4-21). As noted previously, intakes above the UL for retinol, zinc, or copper are not of concern for children. Similar to copper, the UL for selenium is extrapolated from the selenium UL for adults (IOM, 2000a). In addition, no cases of selenosis have been reported even in high-selenium areas of the United States (IOM, 2000a). For these reasons, intakes of selenium above the UL were not considered of concern. Fewer than 0.01 percent of children ages 2 to less than 5 years exceeded the UL for iron, phosphorus, vitamin B6, and choline.

Special Case: Vitamin D Status Across Age Categories

As described in Appendix J, because an individual’s vitamin D status is determined by both dietary intake and sun exposure, serum 25(OH)D concentrations were used to assess vitamin D status of all population subgroups, except infants.

The data presented in Table 4-22 indicate a low prevalence of inadequacy (less than 5 percent) for both subgroups of children when compared to the serum value linked to the EAR (40 nmol/L [IOM, 2011]). However,

TABLE 4-21 Estimated Prevalence of Micronutrient Excess, WIC-Participating Children 1 to Less Than 5 Years of Age, NHANES 2011–2012

NOTES: N = sample size; SE = standard error; UL = Tolerable Upper Intake Level.

* Values represent the UL for children 1 to 3 years/4 to <5 years of age.

SOURCES: IOM, 1997, 1998, 2000a, 2001, 2002/2005, 2005, 2011; USDA/ARS, 2011–2012.

TABLE 4-22 Serum 25-Hydroxy Vitamin D of WIC Participants, NHANES 2005–2006

NOTES: 25(OH)D = 25-hydroxy-vitamin D; BF = breastfeeding (any intensity), nonpregnant WIC-participating women; N = sample size; P = pregnant WIC-participating women; PP = WIC-participating women who are 6 months postpartum, not pregnant, and not breastfeeding. Serum data for 25-hydroxy vitamin D were only available in NHANES 2005–2006.

* A serum 25(OH)D level of 40 nmol/L was established by the Institute of Medicine (2011) as an average requirement that meets the needs of approximately half the population, used to establish EARs for dietary intake of vitamin D.

SOURCE: USDA/ARS, 2005–2006.

the prevalence of inadequacy was greater than 5 percent among pregnant women and greater than 10 percent among breastfeeding and postpartum women. Percentile data are presented in Appendix J.

Dietary vitamin D intakes of infants are presented separately in Table 4-23. The AI for vitamin D in infants is 10 µg per day. Mean intake

TABLE 4-23 Mean Vitamin D Intakes of Formula-Fed WIC-Participating Infants 0 to Less Than 12 Months of Age, NHANES 2011–2012

NOTES: AI = Adequate Intake; N = sample size; SE = standard error.

SOURCES: IOM, 2011; USDA/ARS, 2011–2012.

among formula-fed WIC-participating infants less than 12 months of age falls slightly below the AI.

ANALYSIS OF THE NATIONAL HEALTH AND NUTRITION EXAMINATION SURVEY DATA: FOOD GROUP INTAKES COMPARED TO RECOMMENDATIONS

In addition to evaluating NHANES findings reported in the literature (described earlier in this chapter), the committee conducted its own analysis of food intake among WIC-eligible individuals using the same NHANES dataset, analytical subgroups, and survey years used for its nutrient intake analysis. Because dietary intakes were evaluated relative to the DGA food patterns, which are available only for adults and children 2 years of age and older, the committee’s analysis of food intake by infants was limited to an assessment of intake of “meat, poultry, and seafood” (as a key source of the key nutrients iron and zinc) among older breastfed infants. The analysis of food intake in children ages 1 to less than 2 years was limited to mean intakes (i.e., did not include assessment of the percent of the population below recommended intakes). For women and children 2 years of ages and older, priority concern was given to food groups for which the prevalence of intakes below recommended amounts was 75 percent or greater, followed by food groups for which the prevalence of intakes below recommended amounts was 50 to less than 75 percent.

As was done with nutrients, this section presents data for WIC participant subgroups only, with pre- and post-2009 survey years for women and breastfed infants collapsed into a single analysis and only post-2009 data presented for the other subgroups. Detailed methodology, mean usual intake data, and intake distributions for all of the analyzed population subgroups are presented in Appendix J.

Food Group Intake of Women Participating in WIC

Intakes of food groups and subgroups for women compared to recommendations are presented in Table 4-24. For all food groups except refined grains, approximately 50 percent or more of women participating in WIC had intakes below recommended amounts. For total vegetables, nearly 100 percent of pregnant or postpartum women had intakes lower than the recommended amount and 50 percent of breastfeeding women had intakes lower than the recommended amount. Too few women reported intake of several food subgroups to produce estimates of those foods.

Intake of whole fruit was less than half of recommended fruit intakes in 69 to 90 percent of women. The 1 cup-equivalent of fruit juice included in Table 4-24 represents the DGA recommendation that not more than 50 percent of fruit intake come from juice and is, therefore, an upper limit rather than a target (USDA/HHS, 2016).

Between 70 and 77 percent of women had intakes of solid fat¹⁴ exceeding 10 percent of energy. Data in Table 4-8 indicate that women consume approximately 15 (pregnant and breastfeeding) to 17 (postpartum) percent of energy from added sugars. On average, women consume 13 (postpartum) to 14 (pregnant and breastfeeding) percent of energy from solid fat. Together, this equates to 29 to 30 percent of total energy as “calories for other uses” apart from any other dietary contributions (i.e., excesses of other food groups such as refined grains) to this value.¹⁵ The DGA recommend a limit of 14 and 15 percent of energy be allotted to these calories for a 2,300- or a 2,600-kcal diet, respectively.

Food Group Intake of Breastfed WIC-Participating Infants

Because there are no recommended food group patterns by which to assess the adequacy of infants’ intake, the committee did not assess overall food intake in the infant subgroups. The AAP generally recommends introduction of a variety of complementary foods at approximately 6 months of age (while also recommending exclusive breastfeeding for the first 6 months of life). Inasmuch as iron and zinc intake are of concern for breastfed infants who begin complementary feeding, intake of meat (a source of bioavailable iron as well as zinc) was assessed for older breastfed infants only (6 to 12 months of age). Of the 39 identified infants participating in WIC who were breastfed in NHANES 2009–2012, 10 percent consumed infant

__________________

¹⁴ Most solid fats are high in saturated fats and/or trans fats and have less monounsaturated or polyunsaturated fats.

¹⁵ Calories for other uses (COU) includes added sugars, added refined starches, solid fats, alcohol, or calories from consuming more than the recommended amount of food in a food group (USDA/HHS, 2016). Chapter 2 includes a more detailed discussion of COU.

TABLE 4-24 Food Group Intakes Compared to the DGA Recommendations: Pregnant, Breastfeeding, or Postpartum WIC-Participating Women 19 to 50 Years, NHANES 2005–2012

NOTES: c-eq = cup-equivalents; d = day; g-eq = gram-equivalents; N = sample size; NA = data not available because an inadequate number of individuals reported intake; oz-eq = ounce-equivalents; SE = standard error; tsp-eq = teaspoon-equivalents; wk = week. Kilocalorie (kcal) patterns determined from NHANES assessment of usual intake for each group. Recommended intakes are indicated for P and BF/PP women; one value indicates that the recommendation is the same across groups. See Appendix J, Table J-4, for a description of the foods that compose each food group.

Population subgroups are as follows:

^a P = Pregnant WIC-participating women; kcal pattern = 2,600 kcal.

^b BF = Breastfeeding (any intensity), nonpregnant WIC-participating women; kcal pattern = 2,600 kcal.

^c PP = WIC-participating women who are 6 months postpartum, not pregnant, and not breastfeeding; data were sourced only from 2007–2012 NHANES because no postpartum variable is available in the 2005–2006 dataset.; kcal pattern = 2,300 kcal.

^d The DGA include the recommendation that not more than 50 percent of fruit intake should come from juice.

^e Limits for solid fats and added sugars are based on an upper limit of 10 percent of energy for the 2,600 and 2,300 kcal patterns.

SOURCES: USDA/ARS, 2005–2012; USDA/HHS, 2016.

food meats (specifically, meat, poultry or seafood) over the two survey days (see Table 4-25). For the few who consumed infant meats alone (not as a component of mixed dinners), intake was reported to be 1.9 ounces on the first day observed, an amount that is close to the AAP recommended maximum intake per day of 2 ounces per day (AAP, 2014), though only for the 10 percent of infants who consumed the infant food meats.

Food Group Intake of Children Ages 1 to Less Than 2 Years Participating in WIC

As was the case for infants, the DGA do not include recommended food patterns for children 1 to less than 2 years of age. Inasmuch as there is no comparator for adequacy, mean usual food group and subgroup intakes for children of these ages are presented in Table 4-26. Intake of juice was 0.83 cup-equivalents per day (6.3 ounces), which exceeds the AAP recommended maximum of 6 ounces per day. Solid fat intake in this subgroup was estimated to be 26.5 g-equivalents per day, whereas reported added sugars intake was 9.59 tsp-equivalents per day.

TABLE 4-25 Consumption of Infant Food Meat and Other Sources of Meat by WIC-Participating Breastfed Infants Ages 6 to Less Than 12 Months, NHANES 2009–2012

NOTES: Population weights were applied. N = sample size for WIC-participating breastfed infants. Data from two days of reported intakes.

^a A total of 68 breastfed infants were identified in NHANES 2009–2012.

^b Includes 3 infants with reported intake on day 1.

SOURCE: USDA/ARS 2009–2012.

Food Group Intake of WIC-Participating Children Ages 2 to Less Than 5 Years

The proportions of children ages 2 to less than 5 years with food group and subgroup intakes below the DGA food pattern recommendations are presented in Table 4-27. (As with the other age groups, mean intakes and intake distributions are presented in Appendix J.) The prevalence of inadequate intake was greater than 75 percent of this population subgroup for total vegetables (including dark green and total red and orange); whole grains; seafood; and nuts, seeds, and soy. It was between 50 and 75 percent for beans and peas, total starchy vegetables, other vegetables, total protein foods, and total dairy. The prevalence of excess intake was greater than 75 percent for added sugars and solid fats. Only intakes of total fruit and whole fruit; total grains; refined grains; and meat, poultry, and eggs met or exceeded the DGA recommended amounts in more than 50 percent of these children.

As with WIC-participating women (see Table 4-24), the recommended 0.63 cup-equivalent of fruit juice included in Table 4-27 is an upper limit rather than a target amount (USDA/HHS, 2016). Juice intake among these children averaged 0.71 cup-equivalents per day.

TABLE 4-26 Mean Usual Food Group Intakes of WIC-Participating Children 1 to Less Than 2 Years, NHANES 2011–2012

NOTES: c-eq = cup-equivalents; d = day; g-eq = gram-equivalents; N = sample size; NA = data not available because an inadequate number of individuals reported intake; oz-eq = ounce-equivalents; SE = standard error; tsp-eq = teaspoon-equivalents; wk = week. There is no recommended food pattern for this age group in the 2015–2020 Dietary Guidelines for Americans (USDA/HHS, 2016). See Appendix J, Table J-4, for a description of the foods that comprise each food group.

SOURCE: USDA/ARS, 2011–2012.

Intakes of solid fats and added sugars exceeded 10 percent of energy in over 90 and 80 percent of children in this age group, respectively. On average, children in this age group consume 18 percent of their energy from solid fats and 16 percent of their energy from added sugars. Together, this means that approximately 34 percent of energy is derived from “calories

TABLE 4-27 Food Group Intakes Compared to the DGA Recommendations: WIC-Participating Children 2 to Less Than 5 Years, NHANES 2011–2012

NOTES: c-eq = cup-equivalents; d = day; g-eq = gram-equivalents; N = sample size; NA = data not available because an inadequate number of individuals reported intake; oz-eq = ounce-equivalents; SE = standard error; tsp-eq = teaspoon-equivalents; wk = week.

^a Reference values are the USDA food patterns from the 2015–2020 Dietary Guidelines for Americans (DGA) for a 1,300 kilocalorie (average of 1,200 and 1,400 kilocalorie) food pattern. See Appendix J, Table J-4, for a description of the foods that comprise each food group.

^b The DGA include the overall recommendation that not more than 50 percent of fruit intake should come from juice.

^c Limits for solid fats and added sugars are based on an upper limit of 10 percent of energy for the 1,300-kcal food pattern.

SOURCES: USDA/ARS, 2011–2012; USDA/HHS, 2016.

from other uses” (COU) apart from other dietary contributions (COUs are discussed earlier in this chapter and in Chapter 2). The DGA recommend a limit of 8 percent of energy be allotted to COU for diets ranging between 1,200 and 1,600 kcal.

QUALITY OF WIC PARTICIPANTS’ DIETS

The committee was tasked with applying two different indexes to evaluate the diet quality of WIC participants: the healthy eating index–2010 (HEI–2010) and a second index of the committee’s choosing. As described in the phase I report (NASEM, 2016), the committee developed a nutrient-based diet quality (NBDQ) index to serve as the second index. Both the HEI–2010 and the NBDQ index have maximum scores of 100. The HEI–2010 measures conformance to the DGA. The NBDQ score is based on the probability of adequacy of the shortfall nutrients, as identified in the DGA. The rationale for development of the NBDQ index and methodologies for generating these two indexes are available in Appendix J. The results of this evaluation are summarized here, beginning with calculated HEI–2010 values.

Healthy Eating Index–2010

Mean HEI–2010 index scores ranged from 52 to 61 for women (Table 4-28). For children 2 to less than 5 years of age, the mean HEI–2010 index score was 65 (see Table 4-29). These scores are comparable to the HEI–score estimates of 54 to 58 for the total U.S. population ages 2 years and older, which were based on data from NHANES 2005 to 2010 (USDA/HHS, 2016). As with the NBDQ results, HEI–2010 scores for the subgroups of women were highest among breastfeeding women and lowest among postpartum women, and the score for children was higher than the score for any subgroup of women. Inasmuch as the HEI–2010 measures conformance to the DGA and because the DGA do not provide recommendations for infants or children less than 2 years of age, the HEI–2010 index scores could not be calculated for children in these age groups.

Nutrient-Based Diet Quality Index

Median nutrient-based diet quality (NBDQ) index scores ranged from 40 to 54 among the three subgroups of WIC-participating women (see Table 4-30); median NBDQ index scores were 66 for children ages 1 to less than 2 years and 69 for children ages 2 to less than 5 years (see Table 4-31). Scores for women were highest among breastfeeding women and lowest among postpartum women. The NBDQ could not be calculated for infants because there are so few nutrient EAR values have been determined.

TABLE 4-28 Healthy Eating Index: Results for WIC-Participating Women, 19–50 Years of Age, NHANES 2005–2012

NOTES: HEI = Healthy Eating Index; SE = standard error. For postpartum women, data were sourced from 2007–2012 NHANES. Sample sizes may differ from the nutrient and food group analyses because the methodology uses reported intake on day 1 only.

^a Includes any beans and peas not counted as Total Protein Foods.

^b Includes 100 percent fruit juice.

^c Includes all forms except juice.

^d Includes all milk products, such as fluid milk, yogurt, and cheese, as well as fortified soy beverages.

^e Beans and peas are included here (and not with vegetables) when the Total Protein Foods standard is otherwise not met.

^f Includes seafood, nuts, seeds, soy products (other than beverages) as well as beans and peas counted as Total Protein Foods.

^g Ratio of poly- and monounsaturated fatty acids (PUFAs and MUFAs) to saturated fatty acids (SFAs).

^h Calories from solid fats, alcohol, and added sugars; threshold for counting alcohol is more than 13 grams/1,000 kcal.

ⁱ Breastfeeding intensity was not known.

^j WIC-participating women who are 6 months postpartum, not pregnant, and not breastfeeding; data were sourced only from 2007–2012 NHANES because no postpartum variable is available in the 2005–2006 dataset.

SOURCE: USDA/ARS, 2005–2012.

TABLE 4-29 Healthy Eating Index: Results for WIC-Participating Children, 2 to Less Than 5 Years of Age, NHANES 2011–2012

NOTES: HEI = Healthy Eating Index; SE = standard error. N = 263.

^a Includes any beans and peas not counted as Total Protein Foods.

^b Includes 100 percent fruit juice.

^c Includes all forms except juice.

^d Includes all milk products, such as fluid milk, yogurt, and cheese, and fortified soy beverages.

^e Beans and peas are included here (and not with vegetables) when the Total Protein Foods standard is otherwise not met.

^f Includes seafood, nuts, seeds, soy products (other than beverages) as well as beans and peas counted as Total Protein Foods.

^g Ratio of poly- and monounsaturated fatty acids (PUFAs and MUFAs) to saturated fatty acids (SFAs).

^h Calories from solid fats and added sugars.

SOURCE: USDA/ARS, 2011–2012.

CONSIDERATIONS FOR INTERPRETATION OF FINDINGS FROM THE LITERATURE AND THE COMMITTEE’S ANALYSES

The potential limitations to interpreting the data presented in this chapter are summarized here. First, the challenges the committee faced when evaluating WIC-specific data are discussed. Then, the challenges faced by relying on the NHANES dataset are discussed. These limitations

TABLE 4-30 Nutrient-Based Diet Quality Index: Results for WIC-Participating Women, 19–50 Years of Age, NHANES 2005–2012

NOTES: N = sample size. The index is based on nine nutrients considered to be of some concern in the diets of Americans: calcium, fiber, folate, iron, magnesium, potassium, vitamin A, vitamin C, and vitamin E. It is computed as the average percent adequacy of these nine nutrients for each individual (see Appendix J for a description of the methodology).

* Data were sourced only from 2007–2012 NHANES because no postpartum variable is available in the 2005–2006 dataset.

SOURCE: USDA/ARS, 2005–2012.

are linked directly to the recommendations for data collection presented in Chapter 11.

Evaluating WIC-Specific Data: The Challenge of Selection Bias

Since the creation of the WIC program, it has been difficult to evaluate the effect of participation on any outcome. Because random assignment of individuals to study groups that either receive or do not receive WIC benefits is not considered ethical, experimental study design options suitable for causal inference are limited (e.g., random assignment of a WIC service area, delayed start of a new benefit). In the 1980s, David Rush and his colleagues used studies of several different designs (e.g., historical, longitudinal cohort, and cross-sectional), each with different weaknesses, to provide a

TABLE 4-31 Nutrient-Based Diet Quality Index: Results for WIC-Participating Children, 1 to Less Than 5 Years of Age, NHANES 2011–2012

NOTES: N = sample size. The index is based on nine nutrients considered to be of some concern in the diet of Americans: calcium, fiber, folate, iron, magnesium, potassium, vitamin A, vitamin C, and vitamin E. It is computed as the average percent adequacy of these nine nutrients for each individual (see Appendix J for a description of the methodology).

SOURCE: USDA/ARS, 2011–2012.

comprehensive assessment of the WIC program (Rush et al., 1988a,b,c,d). Such a large and comprehensive set of studies has not been repeated. Nearly all studies reviewed for this report either compare WIC participants to a group of nonparticipants or compare WIC participation before and after the 2009 food package changes. The study designs for these comparisons are not only insufficient for causal inference, but they are likely also confounded by factors that result in the decision to participate in WIC. This phenomenon, known as selection bias, occurs when individuals who choose to participate in a program differ from eligible individuals who choose not to participate.

Differences that contribute to selection bias can be either observable or unobservable. With many social assistance programs, participants are likely to be negatively selected (e.g., participants likely have lower educational achievement or lower wage income than nonparticipants). Negative selection leads to results that make it appear as though the program is not as effective as it really is. Conversely, participants may be positively selected, again for either observable or unobservable characteristics (e.g., parents’ motivation or eagerness to keep their children healthy [Besharov and Germanis, 2001]). Positive selection leads to results that make it appear that a program, such as WIC, has more positive effects than it really does. For WIC specifically, positively biased effects could also result from longer-lasting pregnancies which increase the chances that WIC-eligible women will enter the program and also provide for a longer period of time during which women can benefit from the program.¹⁶ However, there is little reason to expect that there is solely a positive bias in reported WIC program effects, given the likely cumulative effect of negative selection on both observable and unobservable factors (Altonji et al., 2005; Altonji and Elder, 2008).There are several examples in the published literature in which WIC participants differed from WIC-eligible nonparticipants based on negatively selected observable factors. And when there is negative selection on observable factors, as shown in Bitler and Currie (2005) and Currie and Rajani (2015), it seems likely that there is also negative selection on at least some unobservable factors (e.g., the woman’s propensity to have negative birth outcomes outside of any conditions that can be measured by the researcher).

As a result of selection bias, there are critical problems with any

__________________

¹⁶ One important possible source of bias that is prominent in the recent WIC literature is gestational age bias. For example, suppose two women are similar on every dimension, but for idiosyncratic reasons unrelated to WIC use one gives birth at 7.5 months and the other at 9 months. Suppose further that the woman with the premature birth did not enroll in WIC, but would have enrolled at 8 months had her pregnancy lasted to 8 months, and that the second woman does enroll at 8 months. In this case, a comparison of prenatal WIC use and gestation would lead to the mistaken conclusion that WIC participation caused longer gestation.

comparison of WIC participants to nonparticipants. Any observed differences in nutrient or food intake adequacy may misrepresent the intake adequacy of either group. For observed differences in nutrient or food intake to be real, the data cannot be confounded by differences that led to an individuals’ participation (or nonparticipation) in WIC.

Methods for Addressing Selection Bias

Researchers have used several approaches to address selection bias. These methods are briefly reviewed in this section. In one example, researchers use variations in state WIC and Medicaid rules as instrumental variables that predict use of WIC but are assumed not to be correlated with any outcomes other than via the changes they induce in participation (e.g., Chatterji et al., 2002). These studies report some negative effect of WIC on breastfeeding. Other studies have used propensity score matching or other propensity score approaches, which ensure that WIC participants and non-participants also are comparable on other dimensions.¹⁷ This allows the comparison to more closely estimate the effect of WIC on breastfeeding and the health of newborns (Foster et al., 2010; Jiang et al., 2010) and food purchases (Oh et al., 2016). A third approach uses variation in the timing of WIC program rollout in specific counties (e.g., Hoynes et al., 2011) to examine its effects on birth weight. These studies generally find positive effects of having WIC implemented in a particular county. The methods used by Kreider et al. (2016) offer another alternative approach. They used nonparametric, partial identification methods to account for selection and measurement problems jointly and evaluated the impacts of WIC on food insecurity in children with NHANES data. Other possible methods include difference-in-differences approaches (interrupted time-series approaches) that leverage changes affecting one group but not another (e.g., a state-level policy change, also in Chatterji et al., 2002, or Hoynes et al., 2011). Some studies have resolved the challenge of selection bias by using a panel-data approach (see, e.g., Odoms-Young et al., 2014). However, this approach may miss effects of the program on new participants.

Strengths and Limitations of NHANES Data for This Analysis

NHANES is the most comprehensive source of national survey data of individuals with information on both food intake and WIC participation

__________________

¹⁷ Propensity score matching controls for observed confounding variables that are included in the analysis. It does not address differences in observed factors that are not included nor unobservable factors that cannot be accounted for (Cook et al., 2008; Steiner et al., 2010).

that was available to the committee. It provides consistent data collection over time, so survey rounds can be combined or compared. NHANES is nationally representative of women by age groups, as well as for children less than 5 years of age. Despite these important strengths, it was difficult for the committee to create a sample population that accurately represented the national WIC population with data from NHANES. This was because NHANES is not designed to provide a nationally representative sample of several population subgroups served by the WIC program. For example, the numbers of pregnant, breastfeeding, or postpartum women who are captured in NHANES are relatively small and the number eligible for and served by WIC is smaller still. Major challenges that were experienced by the committee included small sample sizes for some subgroups of the WIC population, limitations imposed by combining NHANES samples, inaccurate reporting of WIC participation and household income as well as caveats to interpreting micronutrient adequacy, energy intake, food intake, and diet quality.

Sampling

As noted above, NHANES was not designed to be representative of the WIC population or some of the committee’s key subgroups of interest. As a result, extracting from NHANES only data on low-income individuals of specific ages and in specific physiological states resulted in some sample sizes that were quite small (for women, WIC subgroups ranged between 27 and 165). This problem was exacerbated by restricting the sample to reported WIC participation. These limitations 0required the committee to combine data over several 2-year survey rounds. Although this approach makes the committee’s results more robust because of the larger sample sizes achieved, merging older with more recent data compromises inferences to contemporary dietary intakes. As explained earlier, the need to merge survey rounds to achieve adequate sample sizes for its primary analyses made it impossible for the committee to conduct pre-post analyses of the effects of the 2009 food package changes.

Reported WIC Participation in NHANES and Evaluating Pre-Post 2009 Changes

In NHANES, as in most other surveys, there is substantial misclassification of program use, which likely extends to WIC (Bitler et al., 2003; Celhay et al., 2015; Meyer et al., 2015). When WIC use is underreported, data from participants may erroneously be combined with those of nonparticipants. Conversely, some individuals who are income-eligible or of slightly higher incomes and who did not report WIC use may actually

be WIC participants. Some women also reported being WIC participants although they did not meet the income (income may have changed since the interview) or physiological criteria. Along with selection bias, this misclassification confounds the challenge of comparing WIC participants and nonparticipants. As noted above, some women reported participation in WIC and also reported not being pregnant, breastfeeding, or postpartum. It is unclear whether these women are actually receiving WIC benefits themselves. Reported WIC participation by these women may indicate instead that they have a child who is receiving WIC benefits.

Income Reporting

Identifying which low-income individuals in NHANES are eligible for WIC poses another challenge. State-level income requirements for WIC eligibility do not correspond to the measure of income reported in NHANES (at the household level and annually). Additionally, some individuals may legitimately participate in WIC if adjunctively or automatically eligible because of their participation in Medicaid, Temporary Assistance for Needy Families, or SNAP. These are programs with different eligibility thresholds, as well as different rules, than what WIC uses regarding how income is calculated and compared to the respective program-specific thresholds. For example, SNAP does not include the fetus in the count of individuals in the household unit, but WIC does in some states. Finally, although the NHANES population weights include an income-adjustment that accounts for the national income distribution with adjustment for low-income persons defined as those at or below 130 percent of the federal poverty level, it does not account for the distribution of income characteristic of the WIC population. The most recently available national data (from 2014) indicate that more than 65 percent of WIC participants have household incomes that are less than 100 percent of the poverty-to-income ratio (CDC, 2013; USDA/FNS, 2015b).

Estimating Micronutrient Adequacy

For certain nutrients, estimates of prevalence of inadequacy based on NHANES data must be interpreted carefully. For iron specifically, requirements are not normally distributed for women, mostly because of menstrual losses of iron. As a result, the EAR cut-point method cannot be used to estimate its prevalence of inadequacy. Inasmuch as most of the women in the NHANES analytical sample were either pregnant or breastfeeding and the sample size was small (for WIC-participating and breastfeeding women, N = 27), the EAR cut-point method was implemented nonetheless. This limitation was considered when interpreting the data.

Several additional micronutrient intake estimates should be interpreted with caution because of small sample sizes. The committee determined that calculating a mean usual intake within 3 percent of the true value (95 percent confidence interval) required a minimum of 25 to 30 individuals, depending on the nutrient and age group. However, this minimum is not adequate for accurate calculation of population-level intake adequacy. Although the statistical method applied by the committee gives relatively reliable numbers around the median and mean even with small sample sizes, it provides less reliable numbers at the tails of distributions. Given that inadequacy is defined by the proportion of individuals in the lower tail of the distribution, this source of unreliability makes it difficult to determine the proportion of individuals with inadequate intakes. Thus, the committee applied a less stringent threshold of 5 percent in the assessment of nutrient intake adequacy and, furthermore, focused its attention on those nutrients with evidence of still higher proportions of inadequate intakes (see Chapter 5).

Estimating Energy Requirements and Energy Intake

The Estimated Energy Requirements (EERs) in this report were calculated based on established equations developed by the IOM (2002/2005). Recently, however, Butte et al. (2014) proposed that these IOM equations overestimate energy expenditure for toddlers because they are based on incorrect physical activity assumptions. This possibility was considered by the committee in the evaluation of the energy expenditure of WIC-participating children.

Self-reported energy intake is of limited value as a measure of true energy intake (Archer et al., 2013; Subar et al., 2015). In general, underreporting is more pervasive than overreporting (Murakami and Livingstone, 2015), especially among overweight and obese women (Briefel et al., 1997; Macdiarmid and Blundell, 1998; McKenzie et al., 2002). CDC data indicate higher levels of obesity in lower-income women (CDC, 2010b), and overweight is commonly reported as a nutritional risk factor in WIC participants. A recent evaluation of reporting accuracy in NHANES 2002–2012 indicated that 25 percent of adults (ages 20 and older) were likely to underreport energy intake and that respondents were more likely to underreport if they were female, non-Hispanic black, had lower education or income, or were overweight or obese (Murakami and Livingstone, 2015). Underreporting could exaggerate the estimated micronutrient inadequacies for women identified in this report. As noted in Subar et al. (2015), even if the discrepancy between reported and recommended intakes is large, concern that actual intakes are low may still be warranted.

Assessing energy (and dietary) intake in any age is challenging (as described in more detail in the phase I report [NASEM, 2016]), but measuring energy (and dietary) intake of infants and young children can be particularly problematic. Overreporting of energy intake for children in NHANES has been documented (Murakami and Livingstone, 2015). The energy intake of infants and children in the NHANES dataset was estimated based on reports given by the caregiver who accompanied the child to the interview. However, given that multiple people may be responsible for the care of the child, collecting an accurate estimate of intake often requires combining parental reports with observations from other caregivers (Foster and Adamson, 2014); such multiple reports are not available in NHANES.

Estimating Food Intake

The sample sizes were smaller in the food intake analysis compared to those for nutrient intake. For example, the sample size for pregnant women was 165 for nutrient intake and 139 for intake of food groups. The already low sample size for breastfeeding women was 27 for nutrients but just 25 for food groups. These smaller samples sizes resulted from the fact that the software used to estimate food intake¹⁸ required 2 days of reported food intake per survey respondent. Reported intake could be zero on one or both days.

Evaluating Diet Quality

Although the reliability and consistency of the HEI–2010 has been validated for prediction of diet quality (Guenther et al., 2014), the index has a few limitations. Specifically, consumers of beans and peas may have lower scores for “seafood and plant proteins” or “total vegetables” because beans and peas are counted towards other groups first, with only “leftovers” contributing to these groups. Additionally, the HEI–2010 does not account for physical activity or the appropriateness of energy intake. Therefore, an individual who consumes too much energy may have a higher HEI–2010 score than one who consumes an appropriate level of energy but who, as a result, has difficulty meeting the recommended food pattern. Moreover, even if the recommended food pattern is met, individuals may have difficulty meeting nutrient requirements. For example, individuals over the age of 8 with energy needs less than 1,600 kcal will have difficulty meeting nutrient requirements (Guenther et al., 2013). Although consuming amounts of food groups according to the 2010 DGA food patterns would

__________________

¹⁸ The only software that does not require equal number of observations per person is the National Cancer Institute (NCI) software, but it failed to converge in several cases in these analyses.

result in a perfect score, neither the 2010 food patterns (Guenther et al., 2013) nor the 2015–2020 food patterns (USDA/HHS, 2016) provide the recommended amounts of vitamins D or E, or potassium or choline. However, the HEI–2010 does provide a validated way to compare diet quality among population groups.

Comparison of the Committee’s Analyses to Other Published Results

Nutrient Intake

In accordance with the results presented in the DGA (USDA/HHS, 2016), the committee found that intakes of vitamins A, D, E, C, folate, calcium, and magnesium were below recommended amounts and intakes of fiber and potassium were below the AI in more than 5 percent of all subgroups of WIC-participating women. In addition, intakes of iron and zinc, as well as several B vitamins, were below recommended amounts in more than 5 percent of subgroups of women. Saturated fat, sodium, and added sugars intakes exceeded recommended amounts in a high proportion of WIC-participating subgroups, again in accordance with what was reported in the 2015–2020 DGA for the overall U.S. population of individuals ages 2 years and older (USDA/HHS, 2016).

The low prevalence of nutrient inadequacies reported here for infants and young children are similar to those reported by Ahluwalia et al. (2016) using NHANES data from 2009–2012. These authors also reported similarly low intakes of vitamin E, fiber, and potassium, and high intakes of vitamin A in children less than 2 years of age. Breastfed infants participating in WIC showed inadequacies of iron and zinc intake above the committee’s threshold of 5 percent, which is commonly the case given the low iron and zinc content of human milk (AAP, 2014).

Food Group Intake

Overall, results from comparisons of actual intakes to recommended food patterns presented in this report are similar to what has been reported in other studies. Most recently, in their comparison of food group intakes of the U.S. population to federal dietary recommendations, Krebs-Smith et al. (2010) applied an approach similar to the committee’s approach. They used 2001–2004 NHANES data and found that most individuals in the U.S. population did not meet the recommended intakes for any food group except “total grains” and “meat and beans” (food groups were categorized differently then). Additionally, similar to the findings reported here and in the DGA (USDA/HHS, 2016), Krebs-Smith et al. (2010) found that energy intake from solid (saturated) fats and added sugars was excessive.

Diet Quality

HEI–2010 results for WIC-participating children reported herein are higher than those reported by Tester et al. (2016), who evaluated the HEI–2010 among children ages 2 to 4 years old using NHANES 2011–2012. Tester et al. (2016) identified WIC participants by response to the NHANES question, “In the last 12 months, did you [the child] or any member of the household receive benefits from the WIC program, that is, the Women, Infants and Children program?”; whereas, for the committee’s analysis, “Is [child] now receiving benefits from the WIC program” was the determinant for reported WIC participation. The Tester et al. (2016) question is not a robust identifier of WIC participation, because many children do not participate after 1 year of age and therefore may not have been participants at the time the caregiver was surveyed. In addition to reporting lower HEI–2010 results overall, Tester et al. (2016) also found that the HEI–2010 was significantly higher for children participating in WIC after the 2009 food package changes compared to before. However, as discussed earlier in this chapter, the committee found it impossible to use NHANES data to link pre- versus post-2009 nutrient and food intake differences to changes in the WIC food packages.

Considerations for the Design of Studies to Evaluate Effects of the WIC Food Packages

In Appendix K, the committee proposes study designs that would facilitate evaluation of the effects of the WIC program on various outcomes. In its deliberations, the committee considered three common limitations of existing research studies of the WIC program.

Issues of Measurement of Dietary and Nutritional Status Outcomes

Measurement issues related to dietary intake (whether of nutrients or food groups) and nutritional status include identifying the best method to use, deciding on the timing of baseline measure or the reference period, and determining the amount of time required to ensure a measurable effect of the WIC services on outcomes. The methodological challenges and limitations of self-reported dietary assessment instruments are well-recognized. The National Cancer Institute has developed a Dietary Assessment Primer intended to assist researchers with determining the best way to assess the dietary intake of a study group (NIH, 2016). Although biomarkers of nutritional status are generally superior indicators compared to self-reported methods, factors such as genetic variability, lifestyle or physiologic factors (e.g., smoking), biological sample variability, and analytical methodology could affect biomarker

measurements (Jenab et al., 2009). Thus, research studies that use biomarkers of nutritional status need to consider these factors in study design or analyses.

Selection Bias

As discussed earlier in this chapter, because the majority of research studies on WIC are observational, selection bias is particularly problematic for research on the WIC program (as it is in evaluations of many other social programs). Several types of observational designs have been used to compare WIC-participants with some set of nonparticipants:

• Regression discontinuity
• Comparison group or comparisons over time (difference in differences or interrupted time series)
• Instrumental variables (an approach that deals with selection bias by finding variables that affect WIC use and that plausibly do not affect the outcomes of interest, such as program rules)
• Fixed effects for families (comparisons among siblings or individuals across time)

These designs are defined in Appendix K. Observational studies will continue to be important for evaluating the differences between WIC participants and nonparticipants on dietary intake or nutrition-related health outcomes as well as for understanding numerous aspects of WIC program implementation.¹⁹

Lack of Generalizability to the National Level

Most research studies on the WIC program either have small sample sizes or use only regional data because of the lack of comprehensive national data. This is true of both observational studies and the few randomized or nonrandomized quasi-experimental studies that have been conducted (e.g., examinations of the efficacy of breastfeeding support or nutrition education programs in WIC settings). It is likely that experimental studies in

__________________

¹⁹ If possible to conduct, randomized controlled trials would be useful because they remove the problem of selection bias. For example, a pilot study in which the CVV for children is increased and effects on child retention are evaluated; or USDA allowing an agency or state to match clinics on demographics and breastfeeding rates and then randomly assign clinics to implement various models of breastfeeding support. If testing of the food packages in this way is prohibited, providing a waiver to states allowing design of the food packages in the first 30 days would provide information about whether the choice to breastfeed is influenced by the food package. Such studies would be conducted in alignment with 7 C.F.R. § 246.26 which states that all participant or applicant information is confidential.

WIC settings will continue to be small in scale and with potentially limited generalizability. Although larger sample size and/or better geographic coverage would increase external validity, large studies are not necessarily better than smaller ones for internal validity (the extent to which causality can be inferred). Pooled analyses of administrative data across regions would enhance generalizability to the national level as long as WIC participation is not affected by program changes currently under consideration.

SUMMARY OF FINDINGS

In this chapter, the committee describes the data sources used for its analyses as well as its strengths and limitations for this purpose. In addition, nutrient- and food safety-related health concerns as well as nutrient, food intake, and diet quality of WIC participants were evaluated. The next chapter describes how the committee used a decision tree to (1) evaluate the data presented here, and (2) develop proposed actions for revising the food packages. Important findings from this chapter’s evaluation are summarized here:

• Changes to the current WIC iron specifications for infant formula are not warranted at this time, given that fully formula-fed infants ages 0 to less than 6 months are provided with an amount of iron that falls above the AI but below the UL for this group.
• Both women and children (ages 2 to less than 5 years) had several nutrient inadequacies that potentially could be addressed with changes to the food packages. These inadequacies may be linked to food intakes that were below recommended levels.
• Women, infants, and children also had excessive intakes of several nutrients, including sodium, saturated fat, and added sugars. In some cases, these excessive intakes may be addressed with changes to the food packages.
• Diet quality as measured by the HEI–2010 was similar to HEI–2010 results for the total U.S. population ages 2 years and older.
• There are many challenges to studying the effects of participation in WIC (and other similar programs). Strategies for dealing with selection bias and other methodological challenges were identified and considered by the committee when developing its recommendations for research to enhance future WIC program evaluation (which are outlined in Chapter 11).

REFERENCES

AAP (American Academy of Pediatrics). 2014. Pediatric nutrition. 7th ed. Edited by R. E. Kleinman and F. R. Greer. Elk Grove Village, IL: American Academy of Pediatrics.

AAPD (American Academy of Pediatric Dentistry). 2012. Policy on dietary recommendations for infants, children, and adolescents. Pediatric Dentistry 30(7 Suppl):47–48.

Ahluwalia, N., K. A. Herrick, L. M. Rossen, D. Rhodes, B. Kit, A. Moshfegh, and K. W. Dodd. 2016. Usual nutrient intakes of U.S. infants and toddlers generally meet or exceed Dietary Reference Intakes: Findings from NHANES 2009–2012. American Journal of Clinical Nutrition. doi: 10.3945/ajcn.116.137752.

AHRQ (Agency for Healthcare Research and Quality). 2014. Vitamin D and calcium: A systematic review of health outcomes (update). Rockville, MD: U.S. Department of Health and Human Services. http://effectivehealthcare.ahrq.gov/ehc/products/537/1953/vitamind-calcium-report-140902.pdf (accessed December 21, 2016).

AHRQ. 2015. Routine iron supplementation and screening for iron deficiency anemia in pregnant women: A systematic review to update the U.S. Preventive Services Task Force recommendation. Rockville, MD: U.S. Department of Health and Human Services. http://www.ncbi.nlm.nih.gov/books/NBK285986 (accessed December 21, 2016).

AHRQ. 2016. Omega-3 fatty acids and maternal and child health: An updated systematic review. Rockville, MD: U.S. Department of Health and Human Services, https://www.ncbi.nlm.nih.gov/books/NBK395921/pdf/Bookshelf_NBK395921.pdf (accessed August 30, 2016).

Altonji, J. G., and T. E. Elder. 2008. Using selection on observed variables to assess bias from unobservables when evaluating Swan-Ganz catheterization. American Economic Review 98(2):345–350.

Altonji, J. G., T. E. Elder, and C. R. Taber. 2005. Selection on observed and unobserved variables: Assessing the effectiveness of catholic schools. Journal of Political Economy, University of Chicago 113(1):151–184.

AND (Academy of Nutrition and Dietetics). 2014. Position of the Academy of Nutrition and Dietetics: Nutrition and lifestyle for a healthy pregnancy outcome. Journal of the Academy of Nutrition and Dietetics 114:1099–1103.

Andreyeva, T. and A. S. Tripp. 2016. The healthfulness of food and beverage purchases after the federal food package revisions: The case of two New England States. Preventive Medicine 91:204–210.

Archer, E., G. A. Hand, and S. N. Blair. 2013. Validity of U.S. nutritional surveillance: National Health and Nutrition Examination Survey caloric energy intake data, 1971–2010. PLoS One 8(10):e76632.

Aune, D., O. Saugstad, T. Henriksen, and S. Tonstad. 2014. Maternal body mass index and the risk of fetal death, stillbirth, and infant death: A systematic review and meta-analysis. Journal of the American Medical Association 311(15):1536–1546.

Berglund, S., and M. Domellöf. 2014. Meeting iron needs for infants and children. Current Opinion in Clinical Nutrition and Metabolic Care 17(3):267–272.

Bermúdez-Millán, Á., A. Hromi-Fiedler, G. Damio, S. Segura-Pérez, and R. Pérez-Escamilla. 2009. Egg contribution towards the diet of pregnant Latinas. Ecology of Food and Nutrition 48(5):383–403.

Besharov, D. J., and P. Germanis. 2001. Rethinking WIC: An evaluation of the women, infants, and children program: Washington, DC: AEI Press.

Bitler, M. P., and J. Currie. 2005. Does WIC work? The effects of WIC on pregnancy and birth outcomes. Journal of Policy Analysis and Management 24(1):73–91.

Bitler, M., J. Currie, and J. K. Scholz. 2003. WIC eligibility and participation. Journal of Human Resources 38:1139–1179.

Briefel, R. R., C. T. Sempos, M. A. McDowell, S. Chien, and K. Alaimo. 1997. Dietary methods research in NHANES III: Under-reporting of energy intake. American Journal of Clinical Nutrition 65(S):1203S–1209S.

Briefel, R. R., L. M. Kalb, E. Condon, D. M. Deming, N. A. Clusen, M. K. Fox, L. Harnack, E. Gemmill, M. Stevens, and K. C. Reidy. 2010. The Feeding Infants and Toddlers Study 2008: Study design and methods. Journal of the American Dietetic Association 110(12, Suppl. 1):S16–S26.

Butte, N. F., W. W. Wong, T. A. Wilson, A. L. Adolph, M. R. Puyau, and I. F. Zakeri. 2014. Revision of dietary reference intakes for energy in preschool-age children. American Journal of Clinical Nutrition 100(1):161–167.

CDC (Centers for Disease Control and Prevention). 2010a. CDC grand rounds: Additional opportunities to prevent neural tube defects with folic acid fortification. Morbidity and Mortality Weekly Report 59(30):980–984.

CDC. 2010b. Obesity and socioeconomic status in adults: United States, 2005–2008. http://www.cdc.gov/nchs/data/databriefs/db50.pdf (accessed December 21, 2016).

CDC. 2013. National Health and Nutrition Examination Survey: Estimation procedures, 2007–2010. http://www.cdc.gov/nchs/data/series/sr_02/sr02_159.pdf (accessed September 27, 2016).

CDC. 2014. Infant feeding practices study ii and its year six follow-up, chapter 2: Neonatal survey. http://www.cdc.gov/ifps/pdfs/data/ifps2_tables_ch2.pdf (accessed December 20, 2016).

Celhay, P., B. D. Meyer, and N. Mittag. 2015. Measurement error in program participation (forthcoming).

Chaparro, M. P., B. A. Langellier, M. C. Wang, M. Koleilat, and S. E. Whaley. 2015. Effects of parental nativity and length of stay in the US on fruit and vegetable intake among WIC-enrolled preschool-aged children. Journal of Immigrant and Minority Health 17(2):333–338.

Chatterji, P., K. Bonuck, S. Dhawan, and N. Deb. 2002. WIC participation and the initiation and duration of breastfeeding. Madison, WI: Institute for Research on Poverty. http://www.ssc.wisc.edu/irpweb/publications/dps/pdfs/dp124602.pdf (accessed August 30, 2016).

Chiasson, M. A., S. E. Findley, J. P. Sekhobo, R. Scheinmann, L. S. Edmunds, A. S. Faly, and N. J. McLeod. 2013. Changing WIC changes what children eat. Obesity (Silver Spring) 21(7):1423–1429.

Cho, S. H., K. L. Chang, J. Yeo, L. Wounded Head, M. Zastrow, C. Zdorovtsov, L. Skjonsberg, and S. Stluka. 2015. Comparison of fruit and vegetable consumption among Native and non-Native American populations in rural communities. International Journal of Consumer Studies 39(1):67–73.

Cook, T.D., W. R. Shadish, and V. C. Wong. 2008. Three conditions under which experiments and observational studies produce comparable causal estimates: New findings from within study comparisons. Journal of Policy Analysis and Management 27(4): 724–750.

Currie, J., and I. Rajani. 2015. Within-mother estimates of the effects of WIC on birth outcomes in New York City. Economic Inquiry 53(4):1691–1701.

Deming, D. M., R. R. Briefel, and K. C. Reidy. 2014. Infant feeding practices and food consumption patterns of children participating in WIC. Journal of Nutrition Education and Behavior 46(3 Suppl):S29–S37.

Di Noia J., D. Monica, K. W. Cullen, R. Pérez-Escamilla, H. L. Gray, and A. Sikorskii. 2016. Differences in fruit and vegetable intake by race/ethnicity and by Hispanic origin and nativity among women in the Special Supplemental Nutrition Program for Women, Infants, and Children. Preventing Chronic Disease 25(13):E115. doi: 10.5888/pcd13.160130.

Dubowitz, T., S. A. Smith-Warner, D. Acevedo-Garcia, S. V. Subramanian, and K. E. Peterson. 2007. Nativity and duration of time in the United States: Differences in fruit and vegetable intake among low-income postpartum women. American Journal of Public Health 97(10):1787–1790.

Dubowitz, T., S. V. Subramanian, D. Acevedo-Garcia, T. L. Osypuk, and K. E. Peterson. 2008. Individual and neighborhood differences in diet among low-income foreign and U.S.-born women. Womens Health Issues 18(3):181–190.

Endres, L. K., H. Straub, C. McKinney, B. Plunkett, C. S. Minkovitz, C. D. Schetter, S. Ramey, C. Wang, C. Hobel, T. Raju, and M. U. Shalowitz. 2015. Postpartum weight retention risk factors and relationship to obesity at 1 year. Obstetrics Gynecology 125(1):144–152.

EOPUS (Executive Office of the President of the United States). 2014. The economic impact of the American Recovery and Reinvestment Act five years later: Final report to Congress. Washington, DC: Executive Office of the President of the United States. https://www.whitehouse.gov/sites/default/files/docs/cea_arra_report.pdf (accessed December 21, 2016).

EPA (U.S. Environmental Protection Agency). 1994. Arsine. https://cfpub.epa.gov/ncea/iris2/chemicalLanding.cfm?substance_nmbr=672 (accessed December 21, 2016).

Ettienne-Gittens, R., E. L. McKyer, M. Odum, C. S. Diep, Y. Li, A. Girimaji, and P. S. Murano. 2013. Rural versus urban Texas WIC participants’ fruit and vegetable consumption. American Journal of Health Behavior 37(1):130–140.

Faith, M. S., B. A. Dennison, L. S. Edmunds, and H. H. Stratton. 2006. Fruit juice intake predicts increased adiposity gain in children from low-income families: Weight status-by-environment interaction. Pediatrics 118(5):2066–2075.

FAO/WHO (Food and Agriculture Organization/World Health Organization). 2011. Joint FAO/WHO expert consultation on the risks and benefits of fish consumption. Geneva, Switzerland. http://www.fao.org/docrep/014/ba0136e/ba0136e00.pdf (accessed December 21, 2016).

FDA (U.S. Food and Drug Administration). 2016. FDA proposes limit for inorganic arsenic in infant rice cereal. http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm493740.htm (accessed August 30, 2016).

Foster, E., and A. Adamson. 2014. Challenges involved in measuring intake in early life: Focus on methods. Proceedings of the Nutrition Society 73(2):201–209.

Foster, E. M., M. Jiang, and C. M. Gibson-Davis. 2010. The effect of the WIC program on the health of newborns. Health Services Research 45(4):1083–1104.

Geyer, H. J., K. W. Schramm, E. A. Feicht, A. Behechti, C. Steinberg, R. Bruggemann, H. Poiger, B. Henkelmann, and A. Kettrup. 2002. Half-lives of tetra-, penta-, hexa-, hepta-, and octachlorodibenzo-p-dioxin in rats, monkeys, and humans—a critical review. Chemosphere 48(6):631–644.

Grimes, C. A., E. A. Szymlek-Gay, K. J. Campbell, and T. A. Nicklas. 2015. Food sources of total energy and nutrients among U.S. infants and toddlers: National Health and Nutrition Examination Survey 2005–2012. Nutrients 7(8):6797–6836.

Grummer-Strawn, L. M., K. S. Scanlon, and S. B. Fein. 2008. Infant feeding and feeding transitions during the first year of life. Pediatrics 122(Suppl 2):S36–S42.

Guenther, P. M., K. O. Casavale, J. Reedy, S. I. Kirkpatrick, H. A. Hiza, K. J. Kuczynski, L. L. Kahle, and S. M. Krebs-Smith. 2013. Update of the Healthy Eating Index: HEI–2010. Journal of the Academy of Nutrition and Dietetics 113(4):569–580.

Guenther, P. M., S. I. Kirkpatrick, J. Reedy, S. M. Krebs-Smith, D. W. Buckman, K. W. Dodd, K. O. Casavale, and R. J. Carroll. 2014. The Healthy Eating Index-2010 is a valid and reliable measure of diet quality according to the 2010 Dietary Guidelines for Americans. Journal of Nutrition 144(3):399–407.

Hoynes, H., M. Page, and A. H. Stevens. 2011. Can targeted transfers improve birth outcomes? Evidence from the introduction of the WIC program. Journal of Public Economics 95(7–8):813–827.

IARC (International Agency for Research on Cancer). 1998. Arsenic and arsenic compounds. http://www.inchem.org/documents/iarc/suppl7/arsenic.html (accessed December 21, 2016).

IOM (Institute of Medicine). 1997. Dietary Reference Intakes for calcium, phosphorus, magnesium, vitamin D, and fluoride. Washington, DC: National Academy Press.

IOM. 1998. Dietary Reference Intakes for thiamin, riboflavin, niacin, vitamin B6, folate, vitamin B12, pantothenic acid, biotin, and choline. Washington, DC: National Academy Press.

IOM. 2000a. Dietary Reference Intakes for vitamin C, vitamin E, selenium, and carotenoids. Washington, DC: National Academy Press.

IOM. 2000b. Dietary Reference Intakes: Applications in dietary assessment. Washington, DC: National Academy Press.

IOM. 2001. Dietary Reference Intakes for vitamin A, vitamin K, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon, vanadium, and zinc. Washington, DC: National Academy Press.

IOM. 2002/2005. Dietary Reference Intakes for energy, carbohydrate, fiber, fat, fatty acids, cholesterol, protein and amino acids. Washington, DC: The National Academies Press.

IOM. 2003. Dioxins and dioxin-like compounds in the food supply: Strategies to decrease exposure. Washington, DC: The National Academies Press.

IOM. 2005. Dietary Reference Intakes for water, potassium, sodium, chloride, and sulfate. Washington, DC: The National Academies Press.

IOM. 2006. WIC food packages: Time for a change. Washington, DC: The National Academies Press.

IOM. 2009. Weight gain during pregnancy: Reexamining the guidelines. Washington, DC: The National Academies Press.

IOM. 2011. Dietary Reference Intakes for calcium and vitamin D. Washington, DC: The National Academies Press.

Ishdorj, A., and O. Capps. 2013. The effect of revised WIC food packages on Native American children. American Journal of Agricultural Economics 95(5):1266–1272.

Jenab, M., N. Slimani, M. Bictash, P. Ferrari, and S. A. Bingham. 2009. Biomarkers in nutritional epidemiology: Applications, needs and new horizons. Human Genetics 125(5–6):507–525.

Jiang, M., E. M. Foster, and C. M. Gibson-Davis. 2010. The effect of WIC on breastfeeding: A new look at an established relationship. Children and Youth Services Review 32(2):264–273.

Kong, A., A. M. Odoms-Young, L. A. Schiffer, M. L. Berbaum, S. J. Porter, L. Blumstein, and M. L. Fitzgibbon. 2013. Racial/ethnic differences in dietary intake among WIC families prior to food package revisions. Journal of Nutrition Education and Behavior 45(1):39–46.

Kong, A., A. M. Odoms-Young, L. A. Schiffer, Y. Kim, M. L. Berbaum, S. J. Porter, L. B. Blumstein, S. L. Bess, and M. L. Fitzgibbon. 2014. The 18-month impact of Special Supplemental Nutrition Program for Women, Infants, and Children food package revisions on diets of recipient families. American Journal of Preventive Medicine 46(6):543–551.

Krebs, N. F., J. E. Westcott, N. Butler, C. Robinson, M. Bell, and K. M. Hambidge. 2006. Meat as a first complementary food for breastfed infants: Feasibility and impact on zinc intake and status. Journal of Pediatric Gastroenterology Nutrition 42(2):207–214.

Krebs-Smith, S. M., P. M. Guenther, A. F. Subar, S. I. Kirkpatrick, and K. W. Dodd. 2010. Americans do not meet federal dietary recommendations. Journal of Nutrition 140(10):1832–1838.

Kreider, B., J. V. Pepper, and M. Roy. 2016. Identifying the effects of WIC on food insecurity among infants and children. Southern Economic Journal 82(4):1106–1122.

Macdiarmid, J., and J. Blundell. 1998. Assessing dietary intake: Who, what and why of underreporting. Nutrition Research Reviews 11(2):231–253.

Marchi, J., M. Berg, A. Dencker, E. K. Olander, and C. Begley. 2015. Risks associated with obesity in pregnancy, for the mother and baby: A systematic review of reviews. Obesity Reviews 16(8):621–638.

McKenzie, D. C., R. K. Johnson, J. Harvey-Berino, and B. C. Gold. 2002. Impact of interviewer’s body mass index on underreporting energy intake in overweight and obese women. Obesity Research and Clinical Practice 10(6):471–477.

Meiqari, L., L. Torre, and J. A. Gazmararian. 2015. Exploring the impact of the new WIC food package on low-fat milk consumption among WIC recipients: A pilot study. Journal of Health Care for the Poor and Underserved 26(3):712–725.

Meyer, B. D., W. K. C. Mok, and J. X. Sullivan. 2015. Household surveys in crisis. National Bureau of Economic Research Working Paper Series No. 21399. Cambridge, MA: NBER.

Murakami, K., and M. B. Livingstone. 2015. Prevalence and characteristics of misreporting of energy intake in US adults: NHANES 2003–2012. British Journal of Nutrition 114(8):1294–1303.

NASEM (National Academies of Sciences, Engineering, and Medicine). 2016. Review of WIC food packages: Proposed framework for revisions: Interim report. Washington, DC: The National Academies Press. doi: 10.17226/21832.

NIH (National Institutes of Health). 2016. Dietary assessment primer. https://dietassessmentprimer.cancer.gov/citation.html (accessed October 19, 2016).

Odoms-Young, A. M., A. Kong, L. A. Schiffer, S. J. Porter, L. Blumstein, S. Bess, M. L. Berbaum, and M. L. Fitzgibbon. 2014. 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 Nutrition 17(1):83–93.

Oh, M., H. Jensen, and I. Rahkovsky. 2016. Did revisions to the WIC program affect household expenditures on whole grains? Applied Economic Perspectives and Policy 38(4):578–598. doi: 10.1093/aepp/ppw020.

Pooler, J., and S. F. Gleason. 2014. Comparison of WIC benefit redemptions in Michigan indicates higher utilization among Arab American families. Journal of Nutrition Education and Behavior 46(3 Suppl):S45–S52.

Rasmussen, K. M., and C. L. Kjolhede. 2004. Prepregnant overweight and obesity diminish the prolactin response to suckling in the first week postpartum. Pediatrics 113(5):e465–471.

Rush, D., J. M. Alvir, D. A. Kenny, S. S. Johnson, and D. G. Horvitz. 1988a. The National WIC Evaluation: Evaluation of the Special Supplemental Food Program for Women, Infants, and Children. III. Historical study of pregnancy outcomes. American Journal of Clinical Nutrition 48(2 Suppl):412–428.

Rush, D., M. R. Kurzon, W. B. Seaver, and D. S. Shanklin. 1988b. The national WIC evaluation: Evaluation of the Special Supplemental Food Program for Women, Infants, and Children. VII. Study of food expenditures. American Journal of Clinical Nutrition 48(2 Suppl):512–519.

Rush, D., J. Leighton, N. L. Sloan, J. M. Alvir, and G. C. Garbowski. 1988c. The National WIC Evaluation: Evaluation of the Special Supplemental Food Program for Women, Infants, and Children. II. Review of past studies of WIC. American Journal of Clinical Nutrition 48(2):394–411.

Rush, D., N. L. Sloan, J. Leighton, J. M. Alvir, D. G. Horvitz, W. B. Seaver, G. C. Garbowski, S. S. Johnson, R. A. Kulka, M. Holt, J. W. Devore, J. T. Lynch, M. B. Woodside, D. S. Shanklin. 1988d. The National WIC Evaluation: Evaluation of the Special Supplemental Food Program for Women, Infants, and Children. V. Longitudinal study of pregnant women. American Journal of Clinical Nutrition 48(2 Suppl):439–483.

Sabin, M. A., and W. Kiess. 2015. Childhood obesity: Current and novel approaches. Best Practice & Research Clinical Endocrinology & Metabolism 29(3):327–338.

Schultz, D. J., C. Byker Shanks, and B. Houghtaling. 2015. The impact of the 2009 Special Supplemental Nutrition Program for Women, Infants, and Children food package revisions on participants: A systematic review. Journal of the Academy of Nutrition and Dietetics 115(11):1832–1846.

Shin, D., and W. O. Song. 2014. Prepregnancy body mass index is an independent risk factor for gestational hypertension, gestational diabetes, preterm labor, and small- and large-for-gestational-age infants. Journal of Maternal-Fetal and Neonatal Medicine:1–8.

Steiner, P.M., T.D. Cook, W. R. Shadish, M. H. Clark. 2010. The importance of covariate selection in controlling for selection bias in observational studies. Psychological Methods 15(3):250.

Subar, A. F., L. S. Freedman, J. A. Tooze, S. I. Kirkpatrick, C. Boushey, M. L. Neuhouser, F. E. Thompson, N. Potischman, P. M. Guenther, V. Tarasuk, J. Reedy, and S. M. Krebs-Smith. 2015. Addressing current criticism regarding the value of self-report dietary data. Journal of Nutrition 145(12):2639–2645.

Tester, J. M., C. W. Leung, and P. B. Crawford. 2016. Revised WIC food package and children’s diet quality. Pediatrics 137(5).

Thornton, H. E., S. H. Crixell, A. M. Reat, and J. A. Von Bank. 2014. Differences in energy and micronutrient intakes among central Texas WIC infants and toddlers after the package change. Journal of Nutrition Education and Behavior 46(3 Suppl):S79–S86.

Turcksin, R., S. Bel, S. Galjaard, and R. Devlieger. 2014. Maternal obesity and breastfeeding intention, initiation, intensity and duration: A systematic review. Maternal and Child Nutrition 10(2):166–183.

USDA/ARS (U.S. Department of Agriculture/Agricultural Research Service). 2005–2006. What we eat in America, NHANES 2005–2006. Beltsville, MD: USDA/ARS. http://www.ars.usda.gov/services/docs.htm?docid=13793 (accessed December 21, 2016).

USDA/ARS. 2005–2012. What we eat in America, NHANES 2005–2012. Beltsville, MD: USDA/ARS. http://www.cdc.gov/nchs/nhanes/wweia.htm (accessed December 21, 2016).

USDA/ARS. 2009–2012. What we eat in America, NHANES 2009–2012. Beltsville, MD: USDA/ARS. http://www.ars.usda.gov/services/docs.htm?docid=13793 (accessed December 21, 2016).

USDA/ARS. 2011–2012. What we eat in America, NHANES 2011–2012. Beltsville, MD: USDA/ARS. http://www.ars.usda.gov/services/docs.htm?docid=13793 (accessed December 21, 2016).

USDA/FNS (U.S. Department of Agriculture/Food and Nutrition Service). 2014. Special Supplemental Nutrition Program for Women, Infants and Children (WIC): Revisions in the WIC food packages; Final Rule, 7 C.F.R. § 246.

USDA/FNS. 2015a. Diet quality of American young children by WIC participation status: Data from the National Health and Nutrition Examination Survey, 2005–2008. Alexandria, VA: USDA/FNS. http://www.fns.usda.gov/sites/default/files/ops/NHANES-WIC05-08.pdf (accessed December 21, 2016).

USDA/FNS. 2015b. WIC participant and program characteristics report. Alexandria, VA: USDA/FNS. http://www.fns.usda.gov/sites/default/files/ops/WICPC2014.pdf (accessed September 22, 2016).

USDA/HHS (U.S. Department of Agriculture/U.S. Department of Health and Human Services). 2016. Dietary Guidelines for Americans 2015. Washington, DC: U.S. Government Printing Office. https://health.gov/dietaryguidelines/2015/ (accessed August 29, 2016).

Vinturache, A. E., S. McDonald, D. Slater, and S. Tough. 2015. Perinatal outcomes of maternal overweight and obesity in term infants: A population-based cohort study in Canada. Scientific Reports 5:9334.

Whaley, S. E., L. D. Ritchie, P. Spector, and J. Gomez. 2012. Revised WIC food package improves diets of WIC families. Journal of Nutrition Education and Behavior 44(3):204–209.

WHO (World Health Organization). 2014. Dioxins and their effects on human health. Geneva, Switzerland: World Health Organization. http://www.who.int/mediacentre/factsheets/fs225/en (accessed August 30, 2016).

Williams, J., C. T. Mai, J. Mulinare, J. Isenburg, T. J. Flood, M. Ethen, B. Frohnert, and R. S. Kirby. 2015. Updated estimates of neural tube defects prevented by mandatory folic acid fortification—United States, 1995–2011. Morbidity and Mortality Weekly Report 64(1):1–5.

Yan, J. 2015. Maternal pre-pregnancy BMI, gestational weight gain, and infant birth weight: A within-family analysis in the United States. Economics and Human Biology 18:1–12.

Zeisel, S. H. 2013. Nutrition in pregnancy: The argument for including a source of choline. International Journal of Women’s Health 5:193–199.