INTRODUCTION. Berkeley, CA 2 PHFE-WIC Program, Irwindale, CA

Transcription

1 Research Article Favorable Impact of Nutrition Education on California WIC Families Lorrene D. Ritchie, PhD, RD 1 ; Shannon E. Whaley, PhD 2 ; Phil Spector, PhD 1 ; Judy Gomez, MPH, RD 2 ; Patricia B. Crawford, DrPH, RD 1 ABSTRACT Objective: To explore the impact of coordinated statewide nutrition education on Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) family behavior regarding fruits and vegetables, whole grains, and lower-fat milk. Design: Survey of different cross-sectional samples of WIC families before and after education. Setting: Intervention delivered at all WIC sites in California from April-September Participants: Random samples of pregnant or postpartum women and/or caregivers of children enrolled in WIC: 3,015 before and 3,004 after coordinated nutrition education. Intervention: Education directed at families to eat more, and a greater variety of, fruits and vegetables; eat more whole grains; and drink lower-fat milk instead of whole milk. Main Outcome Measures: Data collected by phone interview of women and caregivers on recognition of education messages, intention to consume, and change in family consumption of target food items. Analysis: Means and frequencies computed. Chi-square and t tests used to compare responses before and after education and for Spanish and English subgroups. Results: Following nutrition education, women and caregivers reported increased recognition of education messages, positive movement in stage of change for target food items, increased family consumption of fruits and whole grains, and replacement of whole milk with lower-fat milk. Impacts were similar for Spanish and English speakers. Conclusions and Implications: Coordinated nutrition education in WIC can significantly influence consumption toward more healthful food choices. Key Words: WIC, nutrition education, behavior change, dietary intake, women, fruits, vegetables, whole grains, lower-fat milk (J Nutr Educ Behav. 2010;42:S2-S10.) INTRODUCTION The Special Supplemental Nutrition Program for Women, Infants and Children (WIC), funded by the United States Department of Agriculture (USDA), provides nutritious food and education on nutrition and breastfeeding to low-income pregnant and postpartum women, and children up to age 5. WIC is the only Food and Nutrition Services program required to provide nutrition education; the program focuses considerable resources on developing and conducting highquality, participant-centered learning. About half of all children in the United States (US) receive services from WIC at some point between birth and age 5. 1 Nationwide 8.2 million women and children receive WIC benefits, 1 with over 1.4 million in California alone. 2 Given the broad reach of the WIC program to lowincome families, it is imperative that nutrition education efforts be evaluated to demonstrate impact on nutrition-related behaviors. 1 Dr. Robert C. and Veronica Atkins Center for Weight and Health, University of California, Berkeley, CA 2 PHFE-WIC Program, Irwindale, CA The authors have no conflict of interest to report with the sponsor of this supplement article or products discussed in this article. Corresponding Author: Lorrene D. Ritchie, PhD, RD, Dr. Robert C. and Veronica Atkins Center for Weight and Health, 119 Morgan Hall, University of California, Berkeley, CA 94720; Phone: (510) ; lorrene_ritchie@sbcglobal.net Ó2010 SOCIETY FOR NUTRITION EDUCATION doi: /j.jneb The year 2009 marked an historic change to align the WIC food package with the 2005 Dietary Guidelines for Americans. For the first time, fruits, vegetables, and whole grains were included in the WIC food package nationwide, and milk purchases were restricted to lower-fat milk for all women and all children over 2 years of age. 3 Prior to these changes, many WIC programs embarked on intensive nutrition education efforts to focus on fruits, vegetables, whole grains, and lower-fat milk. There have been relatively few published results of WIC interventions that focus specifically on increasing consumption of fruits and vegetables, even fewer focusing on lower-fat dairy, and, to date, no published studies to increase whole grain consumption among WIC participants. Although recent studies reveal that interventions can be effective in increasing fruit and vegetable intake, 4-6 these studies included more intensive nutrition education than is mandated by S2 Journal of Nutrition Education and Behavior Volume 42, Number 3S, 2010

2 Journal of Nutrition Education and Behavior Volume 42, Number 3S, 2010 Ritchie et al S3 federal WIC regulations (eg, more frequent contacts and/or use of incentives). This study describes the California statewide coordinated nutrition education that took place from April- September 2009, before the October food package change. The goal of the study was to evaluate the impact of a unique, 3-topic nutrition education curriculum (fruits and vegetables, whole grains, lower-fat milk), designed to support participants awareness and intention, and to target dietary behaviors that are the focus of the revised WIC food package. Developed at the state level and implemented at all local agency WIC sites, the curriculum incorporated education methods familiar to WIC staff and participants and was administered at the same time throughout the state of California. The authors hypothesized that the coordinated statewide nutrition education would increase participant recognition of target messages, and that this nutrition education would result in significant positive movement in stage of change and consumption of targeted food items. METHODS Description of Intervention WIC delivers nutrition education to mothers who themselves are the recipients of WIC food vouchers, as well as to the caregivers of children 0-5 years of age who receive WIC food. Families who participate in WIC are mandated to receive at least 2 nutrition education contacts every 6 months. In California, caregivers receive the education from a dietitian or paraprofessional WIC educator, either in a group ( class ) or individual (one-on-one counseling) format. The 6-month statewide nutrition education curriculum (Healthy Habits Every Day) was delivered in 3 2-month blocks, each with a different nutrition focus: fruits and vegetables, lower-fat milk, and whole grains. The curriculum was designed specifically for the California WIC population, which is largely Latino (>75%). 2 Cultural relevance was achieved by piloting the key messages with WIC families during the curriculum development phase. The session focusing on fruits and vegetables (April and May), titled Get Healthy Now, had 2 key messages: eat a rainbow of fruits and vegetables, and eat more anytime food and less sometimes food. For the lower-fat milk session (June and July), the key message was: Lose the Fat, Keep the Vitamins... Drink Lower Fat Milk. The session on whole grains (August and September) focused primarily on identifying whole-grain foods and learning the health benefits of grains. Examples of activities for sessions included card sorts, taste tests, and label reading. All local WIC agency programs in California (n ¼ 82, with monthly caseloads ranging from 75 to over 300,000 participants, averaging 18,800) were mandated to participate in the statewide coordinated nutrition education. The designers of the curriculum trained lead local agency nutrition education staff, who then trained their front-line staff. In September 2009, all local agency WIC programs (primarily agency directors or nutrition education coordinators) completed an on-line survey designed to describe the delivery of the 6-month, statewide nutrition education to WIC participants. The survey included questions about training and implementation of the nutrition education messages, including questions on when (which months) and how (group vs individual) the education was delivered. The questionnaire also assessed agency adoption of the curricula on the 3 topics, and whether agencies modified the curricula substantially. Study Design This study used a pre-post, cross-sectional comparison of results. Identical survey methodology was used with random samples of WIC participants before (Time 1) and at the conclusion (Time 2) of the nutrition education. The study was reviewed by the California State Institutional Review Board in March 2009 and determined to be exempt because of reliance on anonymous survey procedures with adults. Data Collection Surveys were conducted by Field Research Corporation (San Francisco, CA), an independent public opinion research organization. To maintain WIC client confidentiality, autodialers were used such that interviewers knew only the first name of the person they were calling. Prior to the start of each data collection period, Field s interviewers were trained in study-specific interviewing techniques. Retraining sessions were held to ensure that interviewers followed consistent procedures between Time 1 and Time 2. Survey Development Surveys were developed by the lead authors, with extensive input from the state WIC division and WIC local agency staff who developed the nutrition education curricula. The primary focus of the surveys was to capture WIC participants reported behaviors in the areas targeted by the nutrition education. Questions from relevant tools were adapted to include recognition of key messages of the education, intention to consume target food items, and reported change in consumption of target food items. 4,5,7-10 Questions were written in English and translated into Spanish, pilottested, and revised accordingly. Surveys were minutes in length. Prior to the start of data collection, questionnaires were programmed into a computer-assisted telephone interviewing system to provide greater consistency by controlling skip patterns, branches, randomization of items in a battery, and other control features during the call. A respondent could decline to answer any question. Sociodemographic variables were assessed with 13 questions. Outcome measures on each of the 3 nutrition education topics (fruits and vegetables, whole grains, and lower-fat milk) included 3 constructs: recognition of education messages, stage of change, and intake. Recognition of education messages was assessed by asking respondents if they had seen or heard information about each topic in the past 6 months ( yes or no ). Respondents who reported affirmatively were further asked for the source of the information ( WIC or someplace else ). Three questions asked about recognition of the fruit and vegetable

3 S4 Ritchie et al Journal of Nutrition Education and Behavior Volume 42, Number 3S, 2010 message: (1) eating a rainbow of different colors; (2) eating sometimes or anytime food items; and (3) the importance of eating more fruits and vegetables. Two questions asked about recognition of the whole-grain message: (1) seeing an ingredient label; and (2) the importance of eating more whole grains. One question assessed recognition of the lower-fat milk message: importance of drinking lower-fat milk. The Transtheoretical Model plays an important role in understanding the readiness of a population for behavior change. Drawing on this model, the stage of change construct was assessed on behaviors for each nutrition education topic, namely, fruits and vegetables, whole grains, and lower-fat milk. Response options were: not thinking about doing right now, thinking about doing in next 6 months, or already doing. Stage of change was assessed on eating more fruits and vegetables each day, eating more 100% whole-wheat bread, eating brown rice instead of white rice, eating wholewheat instead of plain flour tortillas, and drinking 2% reduced fat milk, 1% low-fat milk, and fat-free or nonfat milk. For milk, since different family members may drink different types, the stage of change questions addressed each milk type and participants could respond to all types. Fruit and vegetable intake was assessed in 2 ways: frequency of respondent intake in the past week, and reported change in intake for the respondent and her family compared to 6 months earlier. Frequency of the past week s intake was assessed separately for 100% fruit juice, other fruits, and vegetables, with answer options ranging from 0, 1-2, 3-4, and 5-6 times per week to 1, 2, 3, and 4þ times per day. Answer options for change in intake of respondent and family were: more, less, or about the same. Whole-grain intake was assessed with a single question on intake of respondent and family compared to 6 months ago, with answer options of more, less, or about the same. Milk intake was assessed with 2 items: type of milk respondent consumes most of time, and, if respondent had a 2- to 5-year-old child on WIC, type of milk child usually consumes. Answer options were: whole or vitamin D milk, 2% or reduced-fat milk, 1% or low-fat milk, fat-free or nonfat milk, don t drink milk, or other milk. Nonfat, 1%, and 2% were combined to constitute lower-fat milk. Participant Sampling A random sample of 9,000 participants was selected from all participants who received WIC services (n > 800,000) in January 2009 (Time 1) and in July 2009 (Time 2). Participants were either pregnant, postpartum, or a parent of a child (0-5 years old) enrolled in WIC. Data collection was considered complete at each time point as soon as 3,000 surveys were concluded. Only participants who could complete the survey in English or Spanish, reported that they or a child in the household was enrolled in the WIC program, and were at least 18 years of age were eligible for the survey. Stratification of the sample to oversample from small WIC agencies was ruled out so that results would be representative of the state as a whole. Therefore, the smallest agencies were not well represented in the sample. Up to 8 attempts were made to reach each respondent at each time point. The cooperation rate, the ratio of those interviewed to those contacted, was 90.0% at Time 1 and 92.6% at Time 2. The response rate, the percentage of those interviewed as a proportion of the total sampled, was 51.6% and 45.5% and Time 1 and Time 2, respectively. A total of 3,015 pregnant or postpartum women, or caregivers of children enrolled in WIC were interviewed at Time 1; 3,004 participants were interviewed at Time 2. Data Analysis Data were analyzed by the Center for Weight and Health at the University of California, Berkeley. Assuming a 5% type 1 error rate and 90% power, a sample size of 3,000 per time point was estimated to be necessary to detect a 2%-3% change in the proportion of participants reporting a behavior change. Means, standard deviations, and frequencies were computed. Time 1 to Time 2 comparisons of means were made by t tests and of frequencies by chi-square analyses. Sociodemographic characteristics were compared between Spanish- and English-speaking subgroups using analysis of variance (for means) and logistic regression (for frequencies) with time point in the model. Because of differences in sociodemographic variables between subgroups, separate analyses were performed according to language preference. Because sociodemographic characteristics were similar by language preference, separate adjustment for sociodemographic variables was not performed. Analyses were considered significant at P <.05. RESULTS Intervention Implementation by Local WIC Agencies All WIC agencies reported adopting the 3 lesson plans without any major modifications, with the exception of 1 small agency (monthly caseload < 5,000) that did not use the milk lesson plan. Nearly all (96%) of the WIC agencies provided the education in the designated months. Most agencies (84%) provided the education in both group and individual formats. The remainder used 1 format only. Survey Sample Description The majority (99%) of respondents were female, 9% were pregnant, and 14% were breastfeeding (Table 1). Pregnant and breastfeeding women were not separated in the analyses after determining that there were no detectible differences in their responses (data not shown). Compared to English speakers, respondents whose preferred language was Spanish were older, had higher rates of current breastfeeding, had larger household sizes, had completed fewer years of education, were less likely to work outside of the home or attend school, and had lower monthly family incomes. Impact of Intervention Fruit and vegetable behaviors. Between Times 1 and 2, significant increases were observed for recognition of education messages on fruits and vegetables: eating a rainbow, eating fewer sometimes and more anytime food items, and the importance

4 Table 1. Sample Characteristics by Time Point and Language Preference All Spanish English Spanish vs English Characteristic Time 1 (n ¼ 3,015) Time 2 (n ¼ 3,004) P a Time 1 (n ¼ 1,731) Time 2 (n ¼ 1,760) P a Time 1 (n ¼ 1,284) Time 2 (n ¼ 1,244) P a P b Female 99.4% 99.4% % 99.7% % 99.1%.613 * Age (y) <.001 * Pregnant currently 9.6% 8.9% % 8.8% % 8.9%.760 NS Breastfeeding currently 13.6% 14.8% % 15.6% % 13.6%.127 ** Number of people * in household Number of children NS (<18 y) in household Number in household NS currently on WIC Duration (y) of NS household on WIC Spanish language 57.4% 58.6%.342 c c c c c preference Race/ethnicity *** Hispanic 82.0% 80.4% 99.4% 99.2% 58.3% 55.5% Non-Hispanic white 9.2% 9.1% 0.0% 0.2% 21.7% 21.0% Non-Hispanic black 5.0% 6.1% 0.2% 0.5% 11.5% 13.5% Other race/ethnicity 3.8% 4.3% 0.4% 0.1% 8.5% 10.0% Highest level of education *** Not higher than eighth grade 24.0% 23.2% 41.1% 39.1% 1.0% 1.0% 9 th 12 th grade but not high 23.3% 22.5% 29.2% 27.5% 15.5% 15.6% school graduate High school graduate 27.6% 29.1% 20.8% 23.4% 36.8% 37.0% Some college or trade school 19.6% 20.3% 6.2% 7.0% 37.7% 39.2% College graduate or post-graduate 5.3% 4.7% 2.6% 3.1% 9.1% 7.2% Working or attending school 42.7% 40.1% % 31.3% % 52.6%.458 *** Family monthly income *** <$1, % 43.4% 46.3% 44.5% 37.2% 41.8% $1,000-$1, % 45.3% 48.4% 48.9% 42.6% 40.0% $$2, % 11.2% 5.4% 6.5% 20.2% 18.2% NS indicates not significant; WIC, Special Supplemental Nutrition Program for Women, Infants, and Children; y, years. Note: Time 1, before the statewide coordinated nutrition education; Time 2 at end of the statewide coordinated nutrition education. When mean is provided, standard deviation is also included. Values in columns may not sum to 100% because of rounding. a By t test for means and by c 2 test for percentages; b By analysis of variance for means and by logistic regression models for percentages with time included in the model: * P <.05; ** P >.01, ***P <.001; c By definition, 100% for Spanish group, 0% for English group. Journal of Nutrition Education and Behavior Volume 42, Number 3S, 2010 Ritchie et al S5

5 S6 Ritchie et al Journal of Nutrition Education and Behavior Volume 42, Number 3S, 2010 of eating more fruits and vegetables (Table 2). Increases in message recognition were significant for both language subgroups with 1 exception, the importance of eating more fruits and vegetables message for Spanish speakers. The proportion of respondents identifying or hearing nutrition education messages from WIC increased significantly from Time 1 to Time 2 for all 3 fruit and vegetable messages. For the sample as a whole, as well as for Spanish- and English-speaking subgroups, there was also a significant shift in intention to consume more fruits and vegetables. Mean frequency of intake by the respondent in the past week decreased significantly between Time 1 and Time 2 for fruit juice intake. There were no differences in frequency of intake of other fruits or vegetables. Likewise, changes in family intake compared to 6 months earlier did not differ for vegetables. However, the proportion of respondents who reported their family eating more fruits increased between Time 1 and Time 2. Whole-grain behaviors. For whole grains, there was no difference between time points in the proportion of respondents who reported looking at ingredient labels on packaged food (Table 3). However, there was a significant increase in the proportion of respondents who, in the past 6 months, had heard about the importance of eating whole grains. This increase was significant for English speakers only. Among those who recognized this message, there was nearly a tripling from Time 1 to Time 2 in the proportion who had heard this message from WIC among both English and Spanish speakers. Stage of change was assessed on the following whole-grain behaviors: eating more 100% whole-wheat bread, eating brown rice instead of white rice, and eating whole-wheat instead of plain flour tortillas. Significant improvements were observed between time points for the entire sample and the Spanish language subgroup for all 3 whole grain behaviors. The English-language subgroup demonstrated significant change for wholegrain tortillas. There was a small, but significant, increase in the proportion of respondents who reported eating more whole grains compared to 6 months earlier; this increase was significant for English speakers, but not for Spanish speakers. Lower-fat milk behaviors. The importance of drinking lower-fat milk (2%, 1%, or nonfat) was a message heard by a higher proportion of Time 2 versus Time 1 respondents of both language subgroups (Table 4). Positive movement in stage of change was observed for 2%, 1%, and nonfat milk for the total sample as well as both language subgroups with 1 exception; the changes for nonfat milk were not significant for English speakers. Significant changes between time points in the type of milk usually consumed by respondents were found for both language subgroups, with less whole milk and more lower-fat milk usually consumed. A similar trend was observed for children. DISCUSSION To the authors knowledge, this is the first published study to document the impact of WIC nutrition education on participant recognition of key education messages, stage of change, and consumption patterns. The nutrition education was designed to prepare participants for subsequent WIC food package modifications, providing an unprecedented opportunity to isolate the impact of nutrition education, separate from other WIC services. Although several intervention studies have been designed to augment nutrition education in WIC, this intervention is unique in that delivery format and frequency were consistent with usual practice, meaning they were offered on the same schedule and of similar duration as usual nutrition education, were piloted with multicultural WIC participants, and were not enhanced with added participant incentives. The only change was that specific topics were delivered simultaneously using the same curriculum in all 82 California WIC agencies. With the exception of reading ingredient labels on packaged food, there was a consistent increase in nutrition education message recognition for all targeted behaviors. Considering that a given participant was likely to hear each message at only a single nutrition education session over the course of the 6-month intervention, the degree of increased recognition suggests that the key messages were delivered as intended in a culturally sensitive fashion that resonated well with California WIC participants. Following the nutrition education, a higher percentage of respondents reported eating more fruits (compared to 6 months earlier). It appeared that respondents shifted intake toward a greater variety rather than larger amounts of the same kinds of fruits. The trend for vegetables, although not significant, was similar. It is possible that WIC participants more readily embraced the eat a rainbow message than the message about the importance of eating more fruits and vegetables. Indeed, of the 3 fruit and vegetable target messages, the largest increase was observed for the eating a rainbow message. Others have also found nutrition education to have a greater impact on fruit rather than vegetable intake. 8 In contrast to respondents report of a change in intake, there was no significant change in intake of fruits and vegetables as assessed by food frequency questions. Capturing the likely small change in intake of fruits and vegetables achievable with a relatively brief exposure to nutrition education may require more sensitive dietary assessment methods. Study respondents reported a mean intake of 6.4 fruits and vegetables per day, which is approximately 50% higher than national 11 and California 12 average estimates. In California, seasonal variation in fruit and vegetable intake is minimal and not a likely explanation for the relatively high reported intakes. 13 Less than one-third of participants reported cost to be a barrier to eating fruits and vegetables at either time point; thus, the high reported intakes by the study participants may be due to a tendency to provide socially desirable responses, particularly since increasing fruit and vegetable intakes was a key message of the education. Inasmuch as this over-reporting is systematic, comparisons between time points remain meaningful; however, absolute intakes in this study must be interpreted with caution.

6 Table 2. Fruit and Vegetable Message Recognition, Intention to Change, and Consumption by Time Point and Language Preference (n ¼ 3,015) All Spanish English (n ¼ 3,004) P a (n ¼ 1,731) (n ¼ 1,760) P a (n ¼ 1,284) (n ¼ 1,244) P a Measure Education messages heard (past 6 mo) Eating a rainbow of different colors < < <.001 From WIC < < <.001 Sometimes, anytime food items < < <.001 From WIC < < <.001 Importance of eating more fruits and vegetables < <.001 From WIC < < <.001 Stage of change for fruits and vegetables (next 6 mo) Not thinking about eating more < Thinking about eating more Already eating more Reported change in intake for respondent and family (compared to 6 mo earlier) Eating more fruits < < Greater variety Larger amounts < Eating more vegetables Greater variety Larger amounts Frequency of intake for respondent (times per d over past week) Mean juice intake < < Mean fruit intake Mean vegetable intake d indicates day; mo, months; WIC, Special Supplemental Nutrition Program for Women, Infants, and Children. Note: Time 1, before the statewide coordinated nutrition education; Time 2 at end of the statewide coordinated nutrition education. When mean is provided, standard deviation is also included. Values in columns may not sum to 100% because of rounding. a By t test for means and by c 2 test for percentages. Journal of Nutrition Education and Behavior Volume 42, Number 3S, 2010 Ritchie et al S7

7 S8 Ritchie et al Journal of Nutrition Education and Behavior Volume 42, Number 3S, 2010 Table 3. Whole-grain Message Recognition, Intention to Change, and Consumption by Time Point and Language Preference All Spanish English Measure (n ¼ 3,015) (n ¼ 3,004) P a (n ¼ 1,731) (n ¼ 1,760) P a (n ¼ 1,284) (n ¼ 1,244) P a Education messages heard (past 6 mo) Seen ingredient label on packaged food From WIC < < <.001 Importance of eating <.001 more whole grains From WIC < < <.001 Stage of change (past 6 mo) Eating more 100% wholewheat bread Not thinking about Thinking about Already Eating brown rice instead of white rice Not thinking about < < Thinking about Already Eating whole-wheat instead of plain flour tortillas Not thinking about < < <.001 Thinking about Already Reported change in intake for respondent and family (compared to 6 mo earlier) Eating more whole grains mo indicates months; WIC, Special Supplemental Nutrition Program for Women, Infants, and Children. Note: Time 1, before the statewide coordinated nutrition education; Time 2 at end of the statewide coordinated nutrition education. Values in columns may not sum to 100% because of rounding. a By t test for means and by c 2 test for proportions. Whole-grain education is complicated by the inherent difficulty in identifying whole versus refined grains. Although considerable WIC education time was devoted to learning about nutrition labels for this purpose, there was no change in reported label reading as a result of the education. However, there was movement in intent to use whole grains (100% whole-wheat bread, brown rice, and whole-wheat tortillas) when using stage of change assessment. The fact that more English than Spanish speakers reported reading ingredient labels and eating more whole grains after the education compared to before suggests that recognition of whole grains by reading labels may require more than a single education session for Spanish speakers. In contrast to teaching about whole grains, the milk messaging is straightforward: switch to lower-fat milk. This simpler message may explain the consistent and significant changes related to nearly every measured milk behavior. Movement toward switching from whole to lower-fat milk was evident in improvements in stage of change and usual consumption patterns for both respondents and their children, and for both Spanish- and English-speaking subgroups. A randomized, controlled trial is the strongest design for testing the effectiveness of an intervention. However, because the new nutrition education was delivered simultaneously across the state, this study design was not possible. Potential bias in the sample was reduced by random selection of WIC participants, by multiple attempts to reach participants, and by achieving relatively high cooperation and response rates. Comparing the demographics of survey respondents to the California WIC population demographics on language preference, ethnicity, and education, the random samples appeared to be representative. 2 Further, with the exception of small differences in age and duration of household on WIC for English-speaking respondents, there were no systematic differences in sample characteristics between survey administrations. Although it is unlikely that sample selection was biased, measurement bias is a potential study limitation since data were collected by selfreport rather than by objective measures. Finally, it must be acknowledged that the authors were unable to adjust

8 Journal of Nutrition Education and Behavior Volume 42, Number 3S, 2010 Ritchie et al S9 Table 4. Milk Message Recognition, Intention to Change, and Consumption by Time Point and Language Preference All Spanish English Measure (n ¼ 3,015) (n ¼ 3,004) P a (n ¼ 1,731) (n ¼ 1,760) P a (n ¼ 1,284) (n ¼ 1,244) P a Education messages heard (last 6 mo) Importance of drinking < < <.001 lower-fat milk From WIC < < <.001 Stage of change (next 6 mo) Drinking 2% milk Not thinking about Thinking about Already Drinking 1% milk Not thinking about < < <.001 Thinking about Already Drinking non-fat milk Not thinking about < < Thinking about Already Type respondent usually consumes Whole milk < < Lower-fat milk Other None Type child usually consumes Whole milk < < <.001 Lower-fat milk Other None mo indicates months; WIC, Special Supplemental Nutrition Program for Women, Infants, and Children. Note: Time 1, before the statewide coordinated nutrition education; Time 2 at end of the statewide coordinated nutrition education. Lower-fat refers to 2%, 1%, or nonfat milk. Values in columns may not sum to 100% because of rounding. a By t test for means and by c 2 test for percentages. the analyses for intervention exposure. Nonattendance at WIC nutrition sessions is a concern. 14 Some WIC intervention studies have invested considerable energy into encouraging attendance (eg, repeated letters, invitations, and phone calls) and have demonstrated a relationship between attendance at sessions and change in consumption patterns. 4,5 It is possible that had the authors been able to quantify and adjust for exposure to nutrition education topics, the results would have been even more compelling. However, inasmuch as this study reflects typical attendance rates of WIC participants, the results are similar to intention-totreat analyses and reflect the impact likely to be achieved with nutrition education as routinely delivered at WIC. Other intervention studies with WIC populations suggest that the impact of nutrition education is augmented in the presence of environmental support for behavior change. 6,7 A repeat survey administration is scheduled for spring 2010 to assess the impact on dietary behaviors of this nutrition education in conjunction with the addition of fruits, vegetables, whole grains, and lower-fat milk to the WIC food package. It is believed that linking the WIC nutrition education with the food provision will go even further toward improving the dietary intakes of low-income women, infants, and children. IMPLICATIONS FOR RESEARCH AND PRACTICE Coordinated statewide nutrition education was associated with consistent increases in respondent recognition of key education messages, improved intention to change behavior, and reported increases in family consumption of fruits, whole grains, and lowerfat milk. These changes were observed for both Spanish- and Englishspeaking respondents. This study provides evidence that nutrition education in WIC is influential in changing participant knowledge, attitudes, and behaviors supporting the consumption of a more healthful diet.

9 S10 Ritchie et al Journal of Nutrition Education and Behavior Volume 42, Number 3S, 2010 NOTES Activities and other supporting materials and lesson plans can be found at Survey questions can be accessed at ACKNOWLEDGMENTS The authors gratefully acknowledge Linnea Sallack, Eloise Jenks, Nancy Crocker, Nelly Mallo, Mark DiCamillo, Laurie True, May Wang, Dena Herman, and Mike Whaley for their unique and significant contributions to this work and the WIC participants for their willingness to complete the survey. Funding for this study was provided by the US Department of Agriculture, Food and Nutrition Service through a grant for WIC administered by the California Department of Public Health. REFERENCES 1. Oliveira V, Frazao E. The WIC Program: Background, Trends, and Economic Issues, 2009 Edition. Economic Research Report No. 73, US Department of Agriculture, Economic Research Service; April California WIC Association. WIC Facts & Figures. Accessed December 1, Institute of Medicine. Committee to Review the WIC Food Packages and Food and Nutrition Board. WIC Food Packages: Time for a Change. Washington, DC: National Academies Press; April Havas S, Anliker J, Damron D, Langenberg P, Ballesteros M, Feldman R. Final results of the Maryland WIC 5-A-Day Promotion Program. Am J Public Health. 1998;88: Havas S, Anliker J, Greenberg D, et al. Final results of the Maryland WIC Food for Life Program. Prev Med. 2003;37: Herman DR, Harrison GG, Afifi AA, Jenks E. Effect of a targeted subsidy on intake of fruits and vegetables among low-income women in the Special Supplemental Nutrition Program for Women, Infants, and Children. Am J Public Health. 2008;98: Anderson JV, Bybee DI, Brown RM, et al. 5-A-Day fruit and vegetable intervention improves consumption in a low-income population. J Am Diet Assoc. 2001;101: Heneman K, Block-Joy A, Zidenberg- Cherr S, et al. A Contract for Change increases produce consumption in lowincome women: a pilot study. J Am Diet Assoc. 2005;105: Tessaro I, Rye S, Parker L, Mangone C, McCrone S. Effectiveness of a nutrition intervention with rural low-income women. Am J Health Behav. 2007;31: PHFE-WIC Survey of WIC Participants. DataMiningNew/Survey/FinalSurvey 2008.pdf. Accessed November 10, Guenther PM, Dodd KW, Reedy J, Krebs-Smith SM. Most Americans eat much less than recommended amounts of fruits and vegetables. J Am Diet Assoc. 2006;106: Oppen M, Sugerman S, Foerster SB. Fruit and Vegetable Consumption in California Adults Ten-Year Highlights from the California Dietary Practices Surveys California Department of Health Services, Cancer Prevention and Nutrition Section; DiSogra C, Hudes M. California Fruit & Vegetable Intake Calibration Study. Award No , Cancer Research Program, California Department of Health Services; August 9, Fox M, Burstein N, Golay J, Price C. WIC Nutrition Education Assessment Study; Final Report. Prepared by Abt Associates Inc. for US Department of Agriculture, Food and Nutrition Service; September 1998.