Changing WIC changes what children eat

Authors


  • Disclosure: The authors declared no conflict of interest.

  • Funding source: This project was supported by grants from The Robert Wood Johnson Foundation Statewide Evaluations of Childhood Obesity Prevention Policies and the New York State Health Foundation.

Correspondence: M.A. Chiasson (machiasson@healthsolutions.org)

Abstract

Objective

This study assessed the impact of revisions to the USDA Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) food packages on nutritional behavior and obesity in children 0- to 4-years-old participating in the New York State (NYS) WIC program. In January 2009, NYS was the first to implement these revisions, which added fruits, vegetables, and whole grains and replaced whole milk with low(1%)-/nonfat milk for children 2- to 4-year-old.

Design and Methods

In this cross-sectional study, >3.5 million administrative records in the NYS WIC Statewide Information System (WICSIS) were analyzed at 6-month intervals from July to December 2008 (pre-implementation) through July to December 2011. Behavioral data in WICSIS were obtained from parent interview by WIC staff at mandatory certification and recertification visits.

Results

Comparing July to December in 2008 and 2011, increases were observed in breastfeeding initiation (72.2-77.5%); delaying introduction of solid foods until after 4 months of age (90.1-93.8%); daily fruit (87.0-91.6%), vegetable (78.1-80.8%), and whole grain consumption (59.0-64.4%) by children aged 1-4 years; and switches from whole milk to low-/nonfat milk by children aged 2-4 years (66.4-69.4%). In 1-year-old children, the proportion ≥95th percentile weight-for-recumbent length decreased from 15.1 to 14.2%; the proportion of children 2- to 4-year-old with body mass index (BMI) ≥95th percentile decreased from 14.6 to 14.2%.

Conclusions

These findings demonstrate that positive changes in dietary intake and reductions in obesity followed implementation of the USDA-mandated cost-neutral revisions to the WIC food package for the hundreds of thousands of young children participating in the NYS WIC program.

Introduction

Childhood overweight is highly prevalent in the United States, with 31.8% of children 2- to 19-years-old overweight or obese (BMI ≥85th percentile) in 2009-2010 ([1]). Overweight begins early in life, with 26.7% of 2- to 5-year-olds already overweight or obese ([1]). Children growing up in poverty are at particularly high risk for overweight from a very early age ([2]). Furthermore, there is evidence for a cumulative impact of weight gain and attendant increases in hypertension, diabetes and other cardiovascular risk factors ([3, 4]), and fatty liver ([5]). Excessive weight gain before 6-12 months of age is associated with overweight at the age of 3 years ([6]), and childhood overweight tracks into adulthood ([9, 10]). Therefore, the Institute of Medicine recently urged the development of more obesity prevention programs that would target infants and young children ([11]).

WIC, the Special Supplemental Nutrition Program for Women, Infants, and Children, is just such a program. WIC provides nutritious foods to supplement diets, nutrition education, breastfeeding promotion and support, education around healthy lifestyles with young children, and referrals to health care. Each month it reaches nearly half of all infants born in the United States and serves approximately nine million low-income pregnant and postpartum women, infants, and children below the age of 5 years ([12]). WIC is well-positioned to reach the low-income children at greatest risk for early weight gain because only those with family income ≤185% of the Federal poverty guidelines are eligible to participate.

WIC was originally established to prevent under-nutrition and micronutrient deficiencies in those consuming inadequate calories and continues to support adequate nutrition. However, in 2005, the Institute of Medicine (IOM) report WIC Food Packages: Time for a Change challenged WIC to offer a more balanced approach to malnutrition, namely to reduce the elements of the food package that contribute to overweight and obesity without compromising key nutrients/micronutrients often absent in the diet ([13]). In 2008, the United States Department of Agriculture (USDA) proposed its first major revisions to food packages provided by WIC since the program's inception in the 1970s ([14]). The changes reflected recommendations from the 2005 IOM report almost in their entirety and aligned with the American Academy of Pediatrics recommendations and the US Dietary Guidelines ([15]). For the first time, the food packages were required to include fruits, vegetables, and whole grains. For children 2-4 years of age, low (1%)-nonfat milk replaced whole milk. WIC programs were also encouraged to offer healthy lifestyle educational programs focused on increasing physical activity and reducing television viewing for children and their families in addition to WIC's core nutrition education function. The new food packages were designed to be cost neutral by balancing additions with a reduction in allotments of some dairy products and fruit juices ([16]).

On January 5, 2009, New York (NYS) was the first state to implement the new WIC food package. Fruits, vegetables, and whole grains were added to the package, and children 2-4 years old were limited to low-(1%) or nonfat milk. At the same time, NYS introduced breastfeeding peer counseling and client-centered counseling as part of the WIC Healthy Lifestyle Initiative. This multi-faceted, comprehensive initiative provides caregivers participating in WIC throughout the state with education on ways to help their families increase healthy eating and physical activity and reduce screen time. The Healthy Lifestyle Initiative was built on programs that have been progressively implemented by NYS WIC since 1997 as part of the statewide initiative for childhood obesity prevention ([17, 18]).

This study was undertaken as part of a comprehensive evaluation of the innovative NYS WIC program obesity prevention policies and implementation of the new WIC food package. Using cross-sectional data abstracted from the NYS WIC administrative data system, the objective of the study was to examine trends over time in prevalence of infant feeding practices; daily consumption of fruits, vegetables, whole grains, and low-/nonfat milk; screen time; and obesity in children 1-4 years of age.

Methods and Procedures

This study consisted of multiple cross-sections of records for infants and children through 4 years of age who were enrolled in the NYS WIC program between July 1, 2008 and December 31, 2011. Participant records were obtained from data extracts of WICSIS, an automated database housed within the NYS Department of Health in Albany that captures administrative data for the NYS WIC program. Behavioral data were obtained by WIC staff through a parent/caregiver interview during mandatory certification (initial enrollment) and mid-certification visits and entered into WICSIS. There was, on average, a 6-month interval between each child certification and mid-certification visit. Records for infants and children enrolled during the study period were extracted quarterly from the WICSIS database. All personal identifiers were removed, potentially identifying data elements were re-coded to further protect anonymity of participants, and each record was assigned a unique code prior to delivery of data to research staff. Data were then combined into seven 6-month intervals that started with July 1-December 31, 2008 and ended with July 1-December 31, 2011. If more than one data measurement existed for an individual infant/child in any 6-month period, the first data record was retained to ensure that an infant/child appeared only once in a 6-month period. However, because the data were cross-sectional, the same child may have been included in more than one 6-month period.

Each infant/child record contained the following demographic information: gender, age, and race/ethnicity [categorized as Hispanic, non-Hispanic Black (Black), non-Hispanic White (White), Asian, and Other]. Height or length and weight were measured by trained WIC staff according to standard protocol ([19]) or at physicians' offices within 60 days of WIC certification. Height or length was recorded to the nearest 1/4 inch and weight to the nearest 1/4 pound. For children ≥2 years of age, recorded measures of height and weight were converted to metric equivalents, and BMI was computed as weight in kilograms divided by height in meters squared. Age- and sex-specific percentiles for BMI were computed based on the reference population for the 2000 CDC growth charts for US children. Overweight was defined as a sex-specific-BMI-for-age ≥85th and <95th percentile and obesity as sex-specific-BMI-for-age ≥95th percentile ([20]). For infants and children <2 years of age, sex-and age-specific percentiles for weight-for-recumbent length were computed based on the 2000 CDC growth charts for US children ([21]).

Parents/caregivers were asked about the following behavioral outcomes during certification and recertification visits: the child's daily consumption of any fruits excluding juice (Y, N), any vegetables (Y, N), and whole grains (Y, N); if the child was ever breastfed or fed breast milk (Y, N); the type of milk the child drank most often (15 choices, collapsed into four categories: low (1%)-/non fat, whole milk, other, and none); age infant was introduced to solid foods (<4, ≥4 months-for infants >4 and <12 months only); the daily number of hours the caregiver /child sat and watched TV, videos, and DVDs (categorical 0-5+ h/day); the daily number of hours the caregiver /child sat and used the computer and/or played computer video games (categorical 0-5+ h/day). Screen time was calculated as time spent watching TV, videos, and DVDs or using the computer. Screen time was presented as the proportion of children meeting the age-specific guidelines for screen time: ≤2 h/day for 2-4-year-olds and no screen time for 0<2-year-olds. Breastfeeding questions were asked until the child stopped breastfeeding. Physical activity measures were excluded from this analysis because data were collected by family activity and not at the individual level.

Data were analyzed using SPSS 14 and SAS version 9.2. Biologically implausible values for height or weight (∼1% of the sample) based on the CDC criteria ([22]) were excluded. The analysis primarily involved the calculation of prevalence proportions for behaviors of interest and BMI categories. Proportions (%) reporting behaviors of interest and in specific BMI categories were calculated for children in each of the seven 6-month intervals. Relative (not absolute) percent change was calculated by comparing values for the 6-month period before implementation of the new WIC food program and Healthy Lifestyles Initiative on January 5, 2009, with values in the July 1-December 31, 2011, interval. Statistical significance tests were not conducted because bi-annual WIC data represented a census of all children who participated in WIC.

This study was approved by the Institutional Review Boards of Public Health Solutions, Columbia University, and the NYS Department of Health.

Results

A total of 3,562,184 participant records were included in the analysis. WIC enrollment fluctuated in the seven 6-month study intervals, averaging 508,883 infants and children per interval. The distribution of demographic characteristics remained nearly constant over time (Table 1). The proportion of participants by gender was 51% male and 49% female throughout. More than one-third of participants (36%) were infants (<1 year) with the proportion by age dropping to 12% for four-year-olds. Nearly 38% of participants were Hispanic followed by 28% White and 24% Black. The proportion of Asian participants increased slightly during the study period from 7.5% to 8.3%.

Table 1. Demographic characteristics of children enrolled in New York State WIC
Characteristics2008200920102011
July- DecemberJanuary- JuneJuly- DecemberJanuary- JuneJuly- DecemberJanuary- JuneJuly- December
  1. aIndividual N's vary due to missing data.
  2. bNon-Hispanic
Na494,311506,660515,360511,405512,429509,012513,007
 (%)(%)(%)(%)(%)(%)(%)
Gender
Male50.750.850.850.850.850.850.8
Female49.349.249.249.249.249.249.2
Age (Year)
<136.436.836.636.536.336.235.8
1 (≥ 1 < 2)21.421.020.720.520.520.220.1
2 (≥ 2 < 3)16.316.516.816.816.816.716.8
3 (≥ 3 < 4)14.014.214.314.614.714.814.9
4 (≥ 4 < 5)11.911.511.611.611.712.112.4
Race/ethnicity
Black, NHb24.824.724.424.424.324.224.3
White, NHb27.927.928.028.027.927.927.8
Hispanic37.537.637.737.837.837.637.5
Asian7.57.57.67.67.88.18.3
Other2.32.32.32.32.22.22.1

Between July-December 2008 (pre-implementation of the WIC food package) and July-December 2011 positive changes, as measured by relative percent increase, were observed in all reported behaviors (Table 2). The largest increases in most behaviors occurred in the first year after implementation but continued throughout the study period. The proportions of mothers who initiated breastfeeding increased by 7.3% from 72.2 to 77.5%, and proportions of mothers who waited to introduce solid foods to infants until after 4 months of age increased by 4.1%.

Table 2. Trends in behavior and weight among WIC participants following implementationa of a revised WIC food Package in New York State
Behavior2008a200920102011% Changeb 2008–11
July- DecemberJanuary- JuneJuly- DecemberJanuary- JuneJuly- DecemberJanuary- JuneJuly- December
  1. a6 months before January 5, 2009, implementation
  2. bComparing July-December 2011 and July-December 2008 by relative percent change.
  3. cFor all infants > 4 months old
  4. dTime spent watching TV, videos, DVDs or using the computer
Breastfeeding(%)(%)(%)(%)(%)(%)(%)(%)
Initiation72.273.073.974.875.776.677.57.3
Introduction of solids
After 4 months of agec90.191.292.793.293.093.293.84.1
Low-/nonfat milk consumption
≥ 2–4 year olds66.469.470.770.069.669.469.44.5
Daily consumption: 1–4 year olds
Fruits87.088.490.491.291.591.291.65.3
Vegetables78.179.480.380.580.580.380.83.5
Whole/grains59.059.162.163.864.364.464.49.1
Screen timed ≤2 h
≥ 2–4 years olds68.268.770.270.670.769.869.92.5
No screen timed        
0 < 2 year olds23.426.128.129.130.030.631.233.3
Weight
BMI ≥ 2–4 year olds
85th −< 95th percentile16.817.116.717.216.416.916.3− 3.0
≥ 95th percentile14.614.714.514.814.314.614.2− 2.7
Weight for recumbent length
≥ 1–2 year olds
≥ 95th percentile15.115.615.515.914.815.214.2− 6.0

Overall, 2-4-year-olds increased consumption of low-/nonfat milk by 4.5% (66.4-69.4%) (Table 2), but changes varied by year of age. While 2-year-olds showed no increase in consumption of low-/nonfat milk over time, 3 and 4-year-olds did. Four-year-olds showed the largest increase from 75.2 to 83.8% (Figure 1). Low-/nonfat milk consumption in children aged ≥2 years also varied by race/ethnicity in 2011 from a low of 64.8% in Asians to a high of 71.7% in Whites (data not shown). Approximately one-quarter of children aged ≥2 years continued to drink whole milk and <1% of children drank no milk (data not shown).

Figure 1.

Prevalence of low-/nonfat milk consumption by year of age before and after changes in the New York State WIC program.

Steady increases in daily consumption of any fruits (5.3%), vegetables (3.5%), and whole grains (9.1%) were observed in 1- to 4-year-olds (Table 2). The increases over time were similar across all age groups (data not shown). However, proportions varied by race/ethnicity with the highest levels of daily consumption in 2011 of fruits (93.8%), vegetables (87.4%), and whole grains (77.5%) reported by White children, while the lowest levels of fruit consumption (89.1%) were reported by Black, of vegetables (76.1%) by Hispanic, and of whole grains (38.9%) by Asian children.

In addition to healthier dietary behaviors, the proportion of children who met age-specific screen-time guidelines during the study period also gradually increased. Proportions of 2- to 4-year-olds who reported ≤2 h of screen time increased by 2.5%, while the proportion of children <2 years who reported no screen time increased by 33.3% from 23.4 to 31.2% (Table 2).

Proportions of children overweight or obese (BMI ≥85th percentile) gradually decreased by 2.9% over the study period. Overall 31.4% of children aged 2-4 years were overweight/obese at pre-implementation, compared to 30.5% in the last half of 2011. Proportions of overweight children (BMI ≥85th and < 95th percentile) fluctuated during the study period but declined by 3.0% from 16.6% at pre-implementation to 16.3% in the last half of 2011. The proportion of obese children (BMI ≥95th percentile) showed similar fluctuations over time, declining by 2.7% from 14.6% at pre-implementation to 14.2% in the last half of 2011 (Table 2).

These patterns differed when trends in overweight/obesity were examined separately by age group and race/ethnicity. Overall 4-year-olds were more likely to be overweight/obese than 2- or 3-year-olds, but declines in proportions of children overweight and obese were observed in all three age groups. On comparing the last 6 months of 2008 with the last 6 months of 2011, the largest relative percent decrease in overweight/obesity was 4.2% in 2-year-olds followed by 3.4% in 4-year-olds and 3.0% in 3-year-olds (Figure 2). A larger decrease of 6.0% from 15.1% to 14.2% was observed in proportions of 1-year-old children with weight-for-recumbent length ≥95th percentile (Table 2).

Figure 2.

Declines in overweight and obesity by age after changes in the New York State WIC program. Before: July-December 2008; After: July-December 2011

Variations were also seen in decreases in prevalence of overweight/obesity by race/ethnicity comparing the two time periods (Figure 3). Overweight/obesity was most prevalent in Hispanic, similar in White and Black, and least prevalent in Asian children. More than one-third of Hispanic children 2- to 4-years-old were overweight/obese, and they showed the smallest relative percent decline (1.6%) between the last 6 months of 2008 and the last 6 months of 2011. In 2008, 29% of Black and White children were overweight/obese. The largest relative percent decrease in prevalence over time, 4.5%, was observed in Black children, followed by a 3.8% decrease in White children. Nearly one-quarter of Asian children were overweight/obese, but that proportion decreased by 3.3% by the last half of 2011.

Figure 3.

Prevalence of overweight and obesity in children 2 to 4 years of age by race/ethnicity before and after changes to the New York State WIC program. NH = Non-Hispanic; before: July-December 2008; after: July-December 2011

Discussion

Positive changes in behavior and reductions in weight were observed in WIC participants following implementation of the revised NYS WIC food packages and the Healthy Lifestyles Initiative in January 2009. Analyses at 6-month intervals from July-December 2008 (pre-implementation) through July-December 2011 of >3.5 million active records in the NYS WIC administrative database showed rapid, consistent increases in breastfeeding initiation; delayed introduction of solid foods to infants; daily fruit, vegetable, and whole grain consumption by children aged 1-4 years; low (1%)-/nonfat milk consumption by children aged 2-4 years; and meeting age-specific screen-time guidelines. The magnitude of the relative percent changes between the last half of 2008 and the last half of 2011 varied from 2.5 to 33.3%. However, even when the magnitude of change was small, the number of children positively impacted was large given that ∼500,000 children from birth through age 4 years participated in the NYS WIC program in any given study year, and nearly one-half of all infants born in NYS were reached by the WIC program. We estimate, for example, that in the last half of 2011 ∼8,200 more children participating in WIC were consuming vegetables daily and 18,000 more were consuming whole grains than would have occurred if consumption rates had remained at 2008 levels.

Changes in dietary behavior were directly related to specific changes to the WIC food package. Increases in whole grain, fruit, and vegetable consumption began after these items were added to the food package, whereas decreases in whole milk consumption occurred after it was eliminated from the food package for 2- to 4-year-olds. An increasing proportion of mothers delayed introduction of solid foods until after 4 months of age when eligibility for infant cereal in the infant food package was changed from 4 months to 6 months of age.

In accordance with Dietary Guidelines for Americans ([15]), daily consumption of fruits and vegetables by 1- to 4-year-olds increased following the addition to the NYS WIC program of a $6 fruit and vegetable voucher per month per child enrolled. These WIC population changes confirm the findings of previous intervention studies that targeted subsidies increased the availability ([23]) and consumption of fruits and vegetables ([24, 25]). Nevertheless, while daily consumption of any fruits and vegetables increased following implementation of the new voucher system, nearly 10% of children in WIC were not eating fruit daily and 20% were not eating vegetables daily in 2011. This indicates the need for continued exploration of new ways to introduce healthy foods to families enrolled in WIC. The success of targeting the family environment for the promotion of healthy eating behaviors has been well described ([26]), and population-level interventions in settings, such as WIC, are likely to be the most cost-effective ([27]).

By limiting children 2- to 4-years-old to low-/nonfat milk checks, the NYS WIC Program achieved a 4.5% relative increase in the proportion of children consuming low-/nonfat milk with a concomitant reduction in consumption of whole milk. However, this overall number obscured age-specific differences: increased low-/nonfat milk consumption was observed in 3- and 4-year-olds but not in 2-year-olds. These differences in age-specific behavior illustrate the importance of examining milk consumption by year of age. A transition period appears to be needed for children to shift from whole milk at the age of 1 year to low-/nonfat milk at the age of 2 years. Switching from whole milk to low-fat milk is important because it results in a 30% reduction in calories (from 146 to 102) and a 75% reduction in fat (from 8 grams to 2 grams) per 8 ounce serving of milk ([28]).

The findings of our NYS WIC study, in which nearly 70% of children 2- to 4-year-olds consumed low-/nonfat milk, stand in sharp contrast to 2007-2008 National Health and Nutrition Survey (NHANES) data. Among a nationally representative sample of children, only 13% of those aged 2-5 years reported usually consuming low-/nonfat milk ([29]). Additional interventions, similar to the NYS WIC program restrictions and removal of whole milk from all public school cafeterias in New York City ([28]), that affect children at the population level are needed to reduce daily fat and calorie consumption.

In contrast to stabilizing national obesity trends at ∼12% in children 2-5 years of age seen in NHANES in 2009-2010 ([1]), obesity in 1- to 4-year-old NYS WIC participants plateaued in 2006-2007 ([30]) followed by a proportional decline over the study period. The largest decrease (6%) in obesity was seen in 1-year-olds, but the proportion of children 2- to 4-year-old who were obese also declined by 2.7% to 14.2% over the seven 6-month intervals of the study. Nationally, 26.7% of children aged 2-5 years were overweight or obese in 2009-2010 ([1]). While the proportion of children 2- to 4-years-old participating in NYS WIC who were overweight or obese in the last half of 2011 (30.5%) was higher than the national level, it was 0.9% lower than it was in the last half of 2008 (31.4%). This means that ∼2,300 fewer children in NYS WIC were overweight or obese in 2011 compared to 2008.

The goal of preventing overweight and obesity in young children is particularly important given that childhood overweight tracks into adulthood ([6, 7, 9, 10]) and in light of growing evidence that excess BMI years are associated with an increased incidence of diabetes among ever younger US birth cohorts ([31]). Of the many chronic diseases associated with overweight and obesity in children, the development of type 2 diabetes is of particular concern given the results of a recent study showing that, unlike adults, a majority of youth with type 2 diabetes may require combination treatment or insulin therapy within a few years after diagnosis ([32, 33]).

The changes in diet and behavior observed in this study were likely supported and enhanced by breastfeeding ([34]) and parental education counseling available through the WIC Healthy Lifestyles Initiative as evidenced by the increase in the proportion of NYS WIC participants meeting age-specific screen-time guidelines ([17]). Excessive screen time, primarily television viewing, in young children has been associated with a variety of negative consequences, including cognitive and language delays ([35]), higher measures of body fatness, and reduced physical activity ([36, 37]). Excessive screen time is one of many factors contributing to the current obesity epidemic either directly, by limiting time available for physical activity, or indirectly, by exposing young children to aggressive food and beverage marketing ([38]) and/or promoting consumption of energy dense snacks ([36]).

The findings from our cross-sectional study demonstrated a strong temporal relationship between NYS WIC program changes, implemented in January 2009, and positive changes in dietary behaviors, weight, and screen time at the NYS WIC population level. However, these findings should be interpreted within their limitations. While use of the WICSIS administrative database, which included information on all infants and children actively participating in NYS WIC, had many advantages including responses to a broad array of questions asked in exactly the same way of all parents/caregivers at 6-month intervals, the data collected were limited to those required for program administration and evaluation. Other than heights and weights, most other information was self-reported by the parent/caregiver. Only crude measures of dietary intake for a limited number of items, such as type of milk and daily fruit and vegetable consumption, breastfeeding initiation, other infant-feeding practices, and screen time, were collected. Collection of more detailed information on these behaviors may have demonstrated an even greater programmatic impact. Lastly, no information was available on the intensity of exposure of individual parents/caregivers to nutritional, breastfeeding, or other health behavior educational messages from WIC, healthcare providers, media, or other sources. Additional studies which compare behavior change in WIC and non-WIC participants such as NHANES are needed to confirm and quantify the impact of changes to the WIC food package on a national level.

Nonetheless, the findings from this study suggest that, within the context of the WIC Healthy Lifestyle Initiative which includes breastfeeding peer and client-centered counseling on ways parents can help their families increase healthy eating and reduce screen time, the USDA-mandated changes to the WIC food package have resulted in positive changes in dietary intake, weight, and screen time for the hundreds of thousands of young children participating in the NYS WIC program. While our findings are limited to NYS, they imply that WIC has enormous potential to positively influence nutrition and diet, both now and in the future ([39]), through revenue neutral changes to a nutritional program that reaches one-half of all children born in the United States. A national focus on evidence-based, population-level obesity prevention programs, such as WIC, Let's Move! Child Care, various components of the Affordable Care Act 9 (Public Law 111-248) and the Healthy, Hunger-Free Kids Act (Public Law 111-296) that target children from the earliest ages is essential to combat the obesity epidemic in the United States ([40]).

Ancillary