Spice MyPlate - SAGE Journals

0 downloads 10 Views 269KB Size Report
Measures: Dietary intake reported on 3-day food records and healthy eating attitudes questionnaires was .... ponderance of ''food deserts'' in which access to healthy foods ...... Feinberg E, Kavanagh PL, Young RL, Prudent N. Food insecurity.

Nutrition/Youth

Spice MyPlate: Nutrition Education Focusing Upon Spices and Herbs Improved Diet Quality and Attitudes Among Urban High School Students

American Journal of Health Promotion 2016, Vol. 30(5) 346-356 ª The Author(s) 2016 Reprints and permission: sagepub.com/journalsPermissions.nav DOI: 10.1177/0890117116646333 ajhp.sagepub.com

Christopher R. D’Adamo, PhD1, Patrick F. McArdle, PhD2, Lyssa Balick, MS3, Erin Peisach, BS, RD, LDN4, Tenaj Ferguson, BS5, Alica Diehl, BA5, Kendall Bustad, MS, PhD(c)6, Brandin Bowden, BA7, Beverly A. Pierce, MA, MLS, RN7, and Brian M. Berman, MD8

Abstract Purpose: To determine whether an experiential nutrition education intervention focusing on spices and herbs (‘‘Spice MyPlate’’) is feasible and improves diet quality and healthy eating attitudes among an urban and predominantly African-American sample of adolescents more than standard nutrition education alone. Design: A nonrandomized controlled trial compared standard nutrition education in U.S. Department of Agriculture MyPlate guidelines (control group) with standard nutrition education plus adjuvant Spice MyPlate curriculum (intervention group). Data were collected at baseline and after 3, 6, and 10 weeks. Setting: Study setting was two public high schools in Baltimore, Maryland. Subjects: A total of 110 students in grades 9 to 12 participated. Intervention: The 6-week school-based intervention conducted during health class focused on cooking using spices and herbs to eat healthier diets according to MyPlate. Measures: Dietary intake reported on 3-day food records and healthy eating attitudes questionnaires was analyzed. Analysis: Differences in diet quality and healthy eating attitudes between study groups were estimated by t-tests, WilcoxonMann-Whitney tests, and covariate-adjusted regression models. Results: Spice MyPlate was feasible and there were modest but significant improvements (p  .05) in the Spice MyPlate group compared with control in whole grains (31.2 g/wk) and protein foods (13.2 ounces per week) intake, and attitudes toward eating vegetables, whole grains, lean protein, and low-fat dairy. Conclusions: Although randomized trials are needed, experiential nutrition education focusing on spices and herbs may help urban and predominantly African-American adolescent populations eat healthier diets. Keywords nutrition education, diet quality, usda myplate, spices and herbs, low-income population, adolescents, prevention research

1

The Departments of Family and Community Medicine, and Epidemiology and Public Health, and with the Center for Integrative Medicine, University of Maryland, Baltimore, MD, USA 2 The Departments of Medicine, and Epidemiology and Public Health, University of Maryland, Baltimore, MD, USA 3 The Nutrition and Wellness at Mind Body Well, Catonsville, MD, USA 4 The Center for Integrative Medicine, University of Maryland, Baltimore, MD, USA 5 HealthCorps, NY, USA 6 Morgan State University, Baltimore, MD, USA 7 The Institute for Integrative Health, Baltimore, MD, USA 8 The Department of Family and Community Medicine, and Center for Integrative Medicine, University of Maryland, Baltimore, MD, USA Corresponding Author: Chris D’Adamo, PhD, University of Maryland School of Medicine, 520 West Lombard St, East Hall, Baltimore, MD 21201. Email: [email protected]

D’Adamo et al.

Introduction The prevalence of childhood and adolescent obesity has been increasing rapidly in the United States. Data from the Centers for Disease Control and Prevention’s National Health and Nutrition Examination Surveys (NHANES) revealed an increase in the prevalence of obesity among 12- to 19-yearolds, from 6.1% in the first NHANES in 1971–1974 to 18.4% in NHANES 2009–2010.1 Although the prevalence of obesity among 2- to 11-year-olds has oscillated during the past decade, there has been a persistent increase in obesity among 12- to 19year-olds nationwide.1 These data suggest a particular need for health promotion interventions among high school students. The repercussions of the increase in childhood and adolescent obesity are severe and pose a substantial public health problem. Obese children and adolescents are more likely to develop type 2 diabetes,2 possess cardiovascular disease risk factors,3 and suffer from discrimination, low self-esteem, and other social and psychologic problems.4 Although obesity is multifactorial in etiology, poor diet quality has been identified as a particularly profound risk factor.5,6 Diet quality is also an important predictor of chronic disease, and the importance of healthy eating habits in childhood for health across the lifespan is well recognized.7-10 Consequently, many interventions aimed at improving the diet quality of children and adolescents have been conducted. Although the results of these interventions have been mixed,11-15 a number of factors associated with diet quality have been identified, including general nutrition knowledge, cooking skills, involving children in food preparation, recording dietary intake, growing healthy foods, experiential learning through group participation, and exposure to healthy foods at markets.15,16 Despite these advances in the understanding of factors associated with healthy eating habits, certain subgroups of children and adolescents face unique challenges to improving the quality of their diets. Children and adolescents living in urban areas, racial minorities, and those of lower socio-economic status often face limited access to healthy foods.17,18 Cultural factors, including differences in cooking and flavor preferences, may further limit the effectiveness of the standardized nutrition education typically offered in urban schools with students from predominantly minority backgrounds. Thus, novel nutrition education programs appear to be warranted for these settings. One nutrition education intervention that effectively espoused a more experiential and culturally tailored approach to improve diet quality and healthy eating attitudes among children was ‘‘LA Sprouts.’’15,19 Inspired in part by the success of LA Sprouts, the purpose of this study was to develop and evaluate an experiential nutrition education program tailored for students in urban and predominantly African-American high schools. There was a dire need for such a program, because the prevalence of obesity among African-American 12- to 19-year-olds is the highest of all racial groups.1 The setting for the implementation and evaluation of this nutrition education program was Baltimore,

347 Maryland, a city rife with socioeconomic hurdles and a preponderance of ‘‘food deserts’’ in which access to healthy foods is limited.20 Community stakeholder engagement with parents, students, and school faculty revealed that experiential nutrition education featuring cooking skills and an emphasis on flavorful foods that students enjoy would help promote sustainable dietary improvements in our target population. Accordingly, a nutrition education intervention focusing on experiential learning through cooking and seasoning healthy foods with spices and herbs, entitled ‘‘Spice MyPlate,’’ was developed. Spice MyPlate emphasized the flavor enhancement and simple cooking applications of spices and herbs to encourage healthy dietary changes. The primary goal of the Spice MyPlate program was to provide urban and predominantly African-American high school student populations with the knowledge, skills, and motivation to eat healthier diets in accordance with U.S. Department of Agriculture (USDA) MyPlate.21 This pilot study evaluated the hypothesis that the Spice MyPlate curriculum would be feasible as an adjuvant to standard nutrition education and would improve indicators of diet quality and healthy eating attitudes among urban and predominantly African-American high school student populations more effectively than standard nutrition education alone.

Methods Design A controlled, two-arm, school-based intervention was conducted comparing standard nutrition education (control group) with standard nutrition education plus the adjuvant Spice MyPlate intervention (intervention group).

Setting and Participants The study was conducted in the spring of 2013 at two public high schools in Baltimore, Maryland. The two schools were generally similar with respect to socioeconomic status (free or reduced-price school meal participation: School A, 75%; School B, 74%), racial minority composition (School A, 88%; School B, 99%), location (both located in East Baltimore), and the relatively high prevalence of ‘‘food deserts’’ in the neighborhoods in which students reside. All students in grades 9 through 12 from both participating high schools who could read and write in English were eligible to participate in this study. Students who did not agree to provide assent and obtain consent from a parent or guardian were excluded from the study. Potential participants were recruited by flyers posted at each school. Study enrollment was limited to 55 participants obtained through a convenience sample at each school (1:1 intervention to control ratio, 110 total participants) in order to maintain intervention session sizes of fewer than 30 students. Participants in both the intervention and control groups received gift cards to one of two local grocery stores ($25 for each complete set of food logs and study surveys returned at baseline, intervention midpoint, intervention end,

348 and follow-up, and $50 if all food logs and study surveys were completed [potential for a total of $150]), the 12 core spices and herbs used in the intervention (approximate value $50), and a variety of kitchen tools (hot plate, slow cooker, measuring cups, etc. [approximate value $50]) for their participation in the study. The Institutional Review Board of the University of Maryland School of Medicine and the administration of the participating high schools approved this study. Informed written consent from parents or legal guardians and written assent from all participating students were obtained prior to the beginning of the study.

Description of the Control Group Participants in the control group received 1 hour of lecturebased nutrition education in the USDA MyPlate guidelines. This nutrition education session was representative of the type of nutrition education typically offered in Baltimore high schools and featured general instruction on the five food groups of USDA MyPlate, examples of healthy foods that fit each of the food groups of MyPlate, practical ways to eat in accordance with the MyPlate guidelines, healthy serving sizes, and how to track and record food and beverage intake using 3-day food records (the method of dietary assessment used in this study). This nutrition education session was provided to study participants at School A (control group) during their regularly scheduled health education curriculum time during the school day. The School A coordinator for HealthCorps,22 a national program aimed at reducing the prevalence of childhood and adolescent obesity through healthy lifestyle programming in schools, led the nutrition education session. A wait list control design was employed, because all participants in the control group were invited to receive the Spice MyPlate nutrition education intervention offered at School A after the conclusion of the study.

Development of the Spice MyPlate Intervention In light of the multifaceted challenges to improving diet quality among high school students in Baltimore, an interdisciplinary team from the academic, dietetic, public health, communitybased nonprofit, and public school system communities sought feedback from community stakeholders and collaborated in the development of the Spice MyPlate nutrition education program for students in urban and predominantly African-American high schools. Stakeholder engagement with students (conducted during health education class), teachers (through interviews), and community-based health professionals (through interviews) who worked with the students in the schools in this study was also conducted to help develop an intervention that would be feasible, engage the students, and identify the most important needs for intervention. The stakeholder engagement process revealed that flavorful food, affordable ingredients, convenience, and time-efficient cooking and meal preparation were important factors for healthy eating in this student population. Spices and herbs were hypothesized by the research team, and confirmed by community stakeholders, to be an

American Journal of Health Promotion 30(5) effective program focus to help motivate the students to achieve improvements in diet quality. This hypothesis was based on the belief that spices and herbs could help improve the flavor of, and therefore the willingness of students to consume, a variety of healthier foods.

Description of the Spice MyPlate Intervention The Spice MyPlate intervention was composed of six weekly nutrition education sessions that focused on using spices and herbs to eat a diet in accordance with the USDA MyPlate guidelines. The intervention was provided to participants at School B (Spice MyPlate group) during health education class during the school day. In order to accommodate the schedules of all participating students and maintain the desired student to teacher ratio of less than 15:1, the six weekly Spice MyPlate intervention sessions were split into two groups at different times in the school day. The School B coordinator for HealthCorps led the Spice MyPlate intervention sessions. Assistance in leading each intervention session was provided by the Community Programs Coordinator from The Institute for Integrative Health,23 a Baltimore-based nonprofit with established relationships in the Baltimore City Public School system. The two Spice MyPlate instructors allowed for a student to teacher ratio of less than 15:1. Each intervention session was 1 hour in duration, and an overview of the 6-week Spice MyPlate nutrition education curriculum is presented in Table 1. Twelve spices and herbs—cinnamon, black pepper, red pepper, basil, garlic, oregano, thyme, nutmeg, ginger, turmeric, rosemary, and cumin—were chosen as the core focus of the Spice MyPlate curriculum. These spices and herbs were chosen on the basis of their accessibility at many food markets, cultural acceptability, relative affordability, general palatability, cooking versatility, health benefits,24-27 and a desire for a combination of familiarity and novelty for participating students. For instance, spices and herbs that were believed to be prohibitively expensive (such as saffron), not readily available in underserved communities (such as whole vanilla beans), or featured a taste that might be overpowering for adolescent palates (such as ground cloves) were not included in the intervention. In light of the feedback received from community stakeholders, an emphasis on experiential education was incorporated into each session. Cooking and meal preparation, creation of novel recipes and flavor combinations, engaging group activities, interactive educational games, and colorful ‘‘flavor profiles’’ included in a notebook for each participant detailing the sensory, culinary, and health properties for each of the 12 core spices and herbs were the primary vehicles of nutrition education. In addition to the 1 hour of standard nutrition education that preceded the Spice MyPlate curriculum and emphasized the USDA MyPlate guidelines (identical curriculum to the control group), participants in the intervention group received the following: six classroom-based sessions of the Spice MyPlate curriculum that were each 1 hour in length, a 1-hour tour of a local grocery store (located approximately 1 mile from School B) aimed at teaching the students how to choose healthier foods

D’Adamo et al.

349

Table 1. Overview of Spice MyPlate Intervention. Session

Nutrition Topics

1

 Overview: Making healthy food choices using U.S. Department of Agriculture MyPlate  Introduction to flavor enhancement  Growing fresh spices and herbs  Food on the Go! (Part I): eating a healthy diet away from home  Creating healthy meals and snacks  Healthy alternatives to favorite foods  Incorporating spices and herbs for taste  Food on the Go! (Part II): eating a healthy diet away from home  Spices, herbs, and healthy MyPlate choices  ‘‘Spice Taboo’’: spices and herbs facts game  Spices and herbs to create healthy recipes  Make a Meal: cooking and eating together  Using spices and herbs for easy salads, soups, and stews  Food preparation skills and safety  Healthy cooking made easy: using slow cookers and hot plates  Using spices and herbs to increase the flavor of healthy foods  Healthy food shopping strategies  Where do my foods fit into MyPlate?  Review: using spices and herbs to eat according to MyPlate  Preparing meals with friends and family  Spice MyLife: reflection on healthy eating, favorite spices and herbs, flavor combinations, recipes, and cooking techniques

2

3

4

5

6

while on a budget, and 2 hours of cooking sessions with a professional chef that featured kitchen safety, a variety of simple cooking methods (stovetop, slow cooker, hot plate, etc.), knife skills, and using spices and herbs to enhance the flavor of healthy foods. The Spice MyPlate intervention was designed to be timeefficient in light of the limited amount of school curriculum time that is generally available for health education. Thus, although each session focused on two particular spices or herbs, experiential reinforcement of each lesson was achieved through recipe preparation incorporating the spices and herbs focused upon during other intervention sessions.

Dietary Assessment Three-day food records (including a weekend day) were used to assess the dietary intake of participating students during the study. This validated28 method of dietary assessment was selected because of the consensus among the researchers, registered dietitians, and stakeholders in the schools and community that 3-day food records would be the most feasible and culturally appropriate method of dietary assessment for this population and study setting. Research staff instructed participants to complete their 3-day food records in real time during the 3-day period. Recognizing that this was not always feasible, students were instructed to complete any food record entries missed during the day at the end of each day. Participants in both the intervention and control groups provided 3day food records prior to the standardized nutrition education session received by both study groups (baseline), and at 3 weeks (intervention midpoint), 6 weeks (intervention end), and 10 weeks (follow-up). The 3-day food record data were

Spices and Herbs in Focus

Recipes

Rosemary

 Veggie dip I

Oregano

    

Basil Thyme

Cinnamon yogurt parfait Spicy trail mix Spiced egg scramble Stuffed chicken pita Cinnamon-raisin oatmeal

Cinnamon Turmeric

Create your own healthy recipes  Trail mix, savory veggie sauce, fruit fondue, etc.

Black pepper Ginger

 Zesty crockpot chicken  Homemade salad dressings

Nutmeg Cumin

      

Red pepper Garlic

Homemade marinara sauce Healthy chicken tenders Veggie dip II MyPlate Salsa Turkey Sloppy Joes Hot plate chili Fruit smoothie

entered into an electronic database and all data were reviewed by both a registered dietitian and the researchers as a quality control measure. Data that the research team determined to be unreasonable or incomplete (for instance, no serving size information) were treated as missing data and not included in the analysis. USDA SuperTracker was used to quantify the food record data into amounts of each of the five groups of USDA MyPlate. SuperTracker is a publicly available tool that has been described previously in the literature29-31 and is commonly employed in the dietetic community and among individual consumers to assess how the foods they consume fit within the USDA MyPlate guidelines. A registered dietitian entered the data from the 3-day food records into SuperTracker to compute quantities of food consumed within each of the five groups of USDA MyPlate. All meals, snacks, and beverages for each of the 3 days for each participant were entered into SuperTracker, which produced total quantities of vegetables, fruits, dairy, whole grains, and protein foods for each of the assessment points (baseline, intervention midpoint, intervention end, and follow-up). The primary outcomes in this study were intake of each of the metrics of diet quality (vegetables [cups], fruits [cups], dairy [cups], whole grains [ounces], and protein foods [ounces]) at each of the study assessment points.

Assessment of Attitudes Toward Healthy Eating Secondary outcomes in this study were attitudes toward healthy eating at each of the study assessment points. Two questions were asked—‘‘How likely are you to eat the following foods?’’ and ‘‘Would you be more likely to eat the following foods if

350 flavored with spices and herbs?’’—for each of the following five food groups of emphasis in MyPlate: (1) vegetables, (2) fruits, (3) low-fat dairy, (4) whole grains, and (5) lean protein. Each question was scored on a three-point Likert scale, with potential responses of ‘‘Not at all likely’’ (score of 1), ‘‘Somewhat likely’’ (score of 2), and ‘‘Highly likely’’ (score of 3). These questions were not previously validated, but novel questions assessing attitudes toward healthy eating that are unique to the intervention using the Likert scale have been successfully evaluated in previous work.19

Data Collection All study data were collected from participants on-site at the two schools under study by Institutional Review Boardapproved research staff from the University of Maryland School of Medicine. Research staff that collected data were not involved in the analysis of study data.

Statistical Analysis Descriptive statistics were computed to characterize the study population and describe the distributions of the diet quality and healthy eating attitudes outcomes throughout the course of the study. Differences in demographic characteristics between the Spice MyPlate and control groups at baseline were assessed via w2 or t-tests. The Shapiro-Wilk test of normality was conducted for all study outcomes. Potential differences in changes in the two attitudes toward healthy eating, as measured by three-point Likert scale questionnaires, between the Spice MyPlate and control groups were determined by Wilcoxon-Mann-Whitney tests. Potential differences in changes in indicators of diet quality, as measured by intake of each of the five food groups in MyPlate (vegetables, fruits, dairy, whole grains, and protein foods), between the Spice MyPlate and control groups were evaluated using t-tests. In addition, Generalized Estimating Equations (GEE) multivariable regression modeling was performed to control for confounding introduced by potential differences in key covariates between the Spice MyPlate and control groups at baseline. Regression models were run for weekly intake of each of the 5 food groups of MyPlate (vegetables, fruits, dairy, whole grains, and protein foods) and the variables included in each model were chosen a priori and included study group allocation, study visit, baseline intake of the food group being modeled, race, age, and gender. Although the investigators expected that the absolute variations in both age and race would be relatively small in this urban and predominantly African-American sample of high school students, these covariates were included in the regression analysis because they were believed to be potential confounders of the relationship between the intervention and diet quality. Many Baltimore neighborhoods are largely segregated by race, which can introduce confounding due to the differences in the availability of healthy foods in different neighborhoods. Despite the fact that all participating students were teenagers, the research team suspected that there could be substantial differences in the

American Journal of Health Promotion 30(5) eating habits and food purchasing and preparation abilities between 13- and 18-year-olds. Age and school grade were highly correlated in both study groups. Consequently, school grade was not included in the regression models. It was also expected that dietary intake could be correlated within individuals throughout the study. GEE regression modeling accounts for within-subject correlation in outcome measures over time32 and was used in these analyses for this purpose. Within-group changes in diet quality throughout the study in the Spice MyPlate and control groups were also modeled using GEE regression models and were adjusted for race, age, and gender. Statistical significance was defined as twotailed p  .05, and all analyses were performed using SAS (version 9.2; SAS Institute Inc., Cary, North Carolina).

Results A total of 110 participants enrolled in the study (55 participants in both the Spice MyPlate and control groups), and the characteristics of the study population at baseline are presented in Table 2. More girls than boys participated in the study (68% girls in Spice MyPlate group, 56% girls in control group) and there was no difference in gender distribution between the Spice MyPlate and control groups (p > .2). Participants in the Spice MyPlate group were on average 1 year younger than those in the control group (age 17.1 vs. 16.1 years, p ¼ .002), and although both groups were composed primarily of racial minorities, there was a higher percentage in the Spice MyPlate group than in the control group (98% vs. 86% racial minority; p ¼ .01). There was no significant difference in intake of the five indicators of diet quality between the study groups at baseline, with the exception of whole grains. Participants in the control group reported greater baseline intake of whole grains than those in the Spice MyPlate group (31.2 vs. 8.5 g/wk; p ¼ .02). There were also no significant differences in responses to either questionnaire assessing healthy eating attitudes at baseline, with the exception of participants in the Spice MyPlate group reporting that they were more likely to eat fruit than participants in the control group (2.7 vs. 2.4 points; p ¼ .02).

Three-Day Food Record and Healthy Eating Attitudes Questionnaires Compliance Participants in both study groups were highly compliant in completing and returning the 3-day food record and healthy eating attitudes questionnaires data at all study assessment points: baseline (96% Spice MyPlate, 98% control), intervention midpoint (96% Spice MyPlate, 95% control), intervention end (95% Spice My Plate, 94% control), and follow-up (85% Spice MyPlate, 76% control).

Changes in Diet Quality Figure 1 illustrates the changes in vegetable, fruit, and dairy intake in the Spice MyPlate and control groups throughout the

D’Adamo et al.

351

Table 2. Baseline Demographic and Dietary Characteristics of Study Sample.

Characteristic

Spice MyPlate Group

Control Group

No. of participants 55 55 Age, y (SD) 16.2 (1.6) 17.1 (1.3) School grade Senior 17 29 Junior 16 12 Sophomore 0 6 Freshman 20 5 Gender Male 16 20 Female 37 34 Race African-American 48 35 White 1 7 Hispanic 0 2 Asian or Pacific Islander 0 2 Native American 2 0 Other 1 5 Diet quality, mean weekly intake (SD) Vegetables, cups 3.7 (4.6) 4.3 (4.6) Fruits, cups 3.5 (6.6) 2.9 (5.5) Whole grains, ounces 0.4 (1.7) 1.1 (3.6) Protein foods, ounces 31.9 (21.8) 33.7 (26.8) Dairy, cups 5.5 (6.6) 5.5 (5.4) Healthy eating attitudes, no. of responses on Likert scale How likely are you to eat the following foods? Vegetables Fruits Whole grains Lean protein Low-fat dairy Would you be more likely to eat the following foods if flavored with spices and herbs? Vegetables Fruits Whole grains Lean protein Low-fat dairy

p Value* 1.0 0.002 0.003

0.2 Figure 1. Changes in Mean Fruit, Vegetable, and Dairy Intake. 0.01

0.3 0.4 0.02 0.5 0.9

Questiony Questiony 1

2

3

1

2

3

2 0 8 12 20

32 16 31 31 30

20 38 15 11 4

4 4 9 8 18

30 24 30 27 29

19 26 14 19 7

0.7 0.02 1.0 0.2 0.6

10 20 17 13 28

24 11 24 20 19

20 22 13 20 7

4 16 8 5 16

24 16 31 21 28

25 20 14 28 8

0.2 0.5 0.1 0.09 0.08

y

Likert scale question responses were: 1, not at all likely; 2, somewhat likely; and 3, highly likely. *The p values were calculated using t-tests (continuous variables) and w2 tests (categorical variables).

study. There were no significant differences in the change of intake of vegetables, fruit, and dairy between the Spice MyPlate and control groups at the intervention end or at follow-up. There was a decrease in fruit intake in both groups from baseline to the intervention midpoint, and this decrease was greater in the Spice MyPlate group than in the control group (1.5 vs. 0.3 cups per week; p ¼ .03). The difference in rate of decrease was mitigated by the end of intervention and at follow-up, because participants in the Spice MyPlate group

Figure 2. Change in Mean Whole Grain Intake. *Statistically significant (p  .05) between-group difference in changes from baseline in favor of Spice MyPlate group. **Statistically significant (p  .05) within-group change from baseline in Spice MyPlate group.

experienced favorable but nonsignificant changes in mean fruit intake relative to the control group. Figure 2 shows that the significantly higher intake of whole grains at baseline in the control group compared with the Spice MyPlate group was reversed by the end of the follow-up period. There was a significant difference in mean change in whole grain intake between participants in the Spice MyPlate and control groups (19.8 vs. 14.2 g/wk; p ¼ .004) from baseline to follow-up. The significant between-group difference in change in whole grain intake from baseline to follow-up was robust to adjustments for the difference in baseline whole grain intake, age, race, and gender in the regression model. The covariate-adjusted regression model revealed a 31.2 g/wk difference in mean change in whole grain intake from baseline to follow-up between the Spice MyPlate group and the control group (95% confidence interval [CI], 4.8, 55.8; p ¼ .02). There was also a significant within-group mean increase of 22.7 g/wk of whole grain intake from baseline to the end of the follow-up period among participants in the Spice MyPlate group (95% CI, 5.1, 37.4; p ¼ .01). There were no significant within-group

352

Figure 3. Change in Mean Protein Food Intake. *Statistically significant (p  .05) between-group difference in changes from baseline in favor of Spice MyPlate group. **Statistically significant (p  .05) within-group change from baseline in Spice MyPlate group.

American Journal of Health Promotion 30(5) question ‘‘How likely are you to eat the following foods?’’ demonstrate that participants in the Spice MyPlate group reported that they were more likely to eat lean protein (p ¼ .02) at the intervention midpoint than at baseline and that they were more likely to eat vegetables (p ¼ .03), low-fat dairy (p ¼ .01), and lean protein (p ¼ .006) at follow-up than at baseline. The results for the question ‘‘Would you be more likely to eat the following foods if flavored with spices & herbs?’’ illustrate that participants in the Spice MyPlate group reported that they would be more likely to eat vegetables from baseline to the intervention midpoint (p ¼ .04), from baseline to the intervention end (p ¼ .01), and from baseline to follow-up (p ¼ .01) if flavored with spices and herbs. Participants in the Spice MyPlate group also reported that they would be more likely to eat whole grains from baseline to follow-up (p ¼ .05) if flavored with spices and herbs. There were no significant changes in any of the questions assessing healthy eating attitudes in the control group.

Discussion changes in whole grain intake from baseline to any of the study assessment points in the control group. Figure 3 shows that the largest difference in the changes in diet quality between the Spice MyPlate and control groups was in the intake of protein foods. Participants in the Spice MyPlate group experienced a positive unadjusted difference in mean change in protein food intake compared with the control group from baseline to the end of the intervention (Spice MyPlate, 6.8 ounces per week; control, 4.3 ounces per week; p ¼ .001) and from baseline to follow-up (Spice MyPlate, 7.7 ounces per week; control, 4.7 ounces per week; p ¼ .0004). These significant differences between the Spice MyPlate and control groups persisted after adjustment for covariates in regression models. There was a 14.3 ounce per week difference in change in mean protein food intake from baseline to the end of the intervention in favor of the Spice MyPlate group compared with the control group (95% CI, .8, 22.8; p ¼ .001) and a 13.2 ounce per week difference from baseline to follow-up in favor of the Spice MyPlate group (95% CI, 4.5, 22.3; p ¼ .004) after adjustment for covariates. There were also significant within-group increases in mean protein food intake among participants in the Spice MyPlate group from baseline to end of intervention (6.7 ounces per week; 95% CI, 0.2, 13.2; p ¼ .04) and from baseline to follow-up (7.7 ounces per week; 95% CI, 2.4, 13.0; p ¼ .004) after adjustment for covariates in regression models. There were no significant within-group changes in protein food intake from baseline to any of the study assessment points in the control group.

Changes in Attitudes Toward Healthy Eating Table 3 provides data describing the changes in healthy eating attitudes among participants in the Spice MyPlate and control groups throughout the course of the study. The results for the

The Spice MyPlate nutrition education intervention and evaluation were feasible among this sample of urban and predominantly African-American high school students in Baltimore. Furthermore, this adjuvant intervention was associated with modest improvements in several indicators of diet quality and attitudes toward healthy eating compared with standard nutrition education alone. The positive effects of the time-efficient (less than 10 hours) and school-based (offered during health education class) Spice MyPlate curriculum were not apparent at the 3-week intervention midpoint, but they were evident at the end of the 6-week intervention and persisted at the followup assessment visit 1 month after the end of the intervention. These findings suggest that more than 3 weeks of 1-hour weekly sessions of this adjuvant nutrition education intervention are necessary to promote dietary habit changes and attitudes toward healthy eating in this setting. Spice MyPlate was the first experiential and school-based nutrition education intervention conducted in an urban and predominantly African-American sample of adolescents. Although few studies have evaluated the impact of experiential nutrition education on diet quality and healthy eating attitudes among urban populations of children and adolescents of any minority ethnicity, the feasibility and generally positive preliminary results of the Spice MyPlate intervention corroborate the findings from one notable previous study that inspired this intervention. The 12-week LA Sprouts program incorporated experiential nutrition education lessons, cooking, and gardening targeted to the cultural food preferences of a sample of predominantly Latino fourth and fifth graders in Los Angeles. The participants in LA Sprouts experienced improvement in their intake of vegetables as well as their attitudes toward healthy eating. Moreover, overweight participants in the LA Sprouts program demonstrated a reduction in body mass index. Although assessing changes in weight and body mass index were beyond the scope of the Spice MyPlate pilot, future

D’Adamo et al.

353

Table 3. Changes in Attitudes Toward Healthy Eating.* Intervention Midpoint Responses (Change From Baseline)y

Baseline Responsesy Spice MyPlate 1

2

Spice MyPlate

Control

3

1

2

3

1

How likely are you to eat the following foods? Vegetables 2 32 20 4 30 19 2

2

Intervention End Responses (Change From Baseline)y Spice MyPlate

Control 3

2

3

2

3

1

2

3

1

2

Control 3

1

2

3

3 23 27 (p ¼ 0.1) 1 26 26 (p ¼ 0.7) 9 26 18 (p ¼ 0.5) 9 23 20 (p ¼ 0.9) 17 28 8 (p ¼ 0.8)

1 18 27 2 26 19 (p ¼ 0.03) (p ¼ 0.5) 2 12 32 4 17 26 (p ¼ 0.8) (p ¼ 0.6) 3 27 16 4 29 14 (p ¼ 0.2) (p ¼ 0.4) 3 24 19 5 23 19 (p ¼ 0.006) (p ¼ 0.5) 10 24 12 18 18 11 (p ¼ 0.01) (p ¼ 0.8)

Would you be more likely to eat the following foods if flavored with spices and herbs? Vegetables 10 24 20 4 24 25 4 17 26 2 21 29 3 19 31 (p ¼ 0.04) (p ¼ 0.3) (p ¼ 0.01) Fruit 20 11 22 16 16 20 18 9 20 14 14 25 18 14 21 (p ¼ 0.9) (p ¼ 0.4) (p ¼ 0.9) Whole grains 17 24 13 8 31 14 10 19 18 5 26 22 7 31 15 (p ¼ 0.1) (p ¼ 0.09) (p ¼ 0.1) Lean protein 13 20 20 5 21 28 7 13 26 6 23 24 9 22 22 (p ¼ 0.06) (p ¼ 0.5) (p ¼ 0.5) Low-fat dairy 28 19 7 16 28 8 26 9 12 17 23 13 25 16 12 (p ¼ 0.7) (p ¼ 0.6) (p ¼ 0.4)

5 22 26 (p ¼ 0.9) 15 18 20 (p ¼ 0.9) 7 28 18 (p ¼ 0.4) 7 23 23 (p ¼ 0.3) 16 27 10 (p ¼ 0.8)

3 15 28 5 20 22 (p ¼ 0.01) (p ¼ 0.9) 14 12 20 14 16 17 (p ¼ 0.6) (p ¼ 0.9) 7 22 17 5 24 18 (p ¼ 0.05) (p ¼ 0.2) 6 16 23 5 17 25 (p ¼ 0.08) (p ¼ 1.0) 19 16 11 18 18 10 (p ¼ 0.2) (p ¼ 1.0)

0 16 38

4 24 26

Whole grains

8 31 15

9 30 14

Lean protein

12 31 11

8 27 19

Low-fat dairy 20 30

4 18 29

7

2 23 28 (p ¼ 0.07) 3 20 29 (p ¼ 0.4) 6 25 22 (p ¼ 0.1) 8 27 18 (p ¼ 0.9) 15 27 11 (p ¼ 0.3)

1

Spice MyPlate

Control

2 25 27 (p ¼ 0.2) 3 12 39 (p ¼ 0.9) 6 32 16 (p ¼ 0.7) 7 28 19 (p ¼ 0.06) 14 29 11 (p ¼ 0.07)

Fruit

21 24 (p ¼ 0.2) 2 15 30 (p ¼ 0.4) 5 29 13 (p ¼ 0.8) 4 26 17 (p ¼ 0.02) 14 24 9 (p ¼ 0.2)

1

Follow-Up Responses (Change From Baseline)y

y Likert scale question responses were as follows: 1, not at all likely; 2, somewhat likely; and 3, highly likely. *The p values were calculated in comparison to baseline scores using Wilcoxon-Mann-Whitney tests. Bold italicized type indicates statistically-significant change from baseline (p < 0.05).

studies may consider assessing changes in these important outcomes. There were several key strengths of this pilot intervention and evaluation. The Spice MyPlate intervention and evaluation were both found to be feasible in the urban high school settings under study. The success in the challenging school environments was likely due to diverse contributions from the multidisciplinary study team of collaborators from the academic, dietetic, public health, community-based, nonprofit, and public school system communities that collectively designed, implemented, and evaluated the Spice MyPlate intervention. Key programmatic and study design elements, such as the decisions to offer the intervention as a part of health education curriculum time during the school day, focusing the content on spices and herbs, using 3-day food records to feasibly assess diet quality in this demographic, and emphasizing experiential nutrition lessons and participatory cooking, were informed through stakeholder engagement with teachers, students, and parents/guardians. School administrators embraced the intervention, partly because of the fact that the Spice MyPlate intervention was designed to be time-efficient, minimally disruptive to the school day, required less than 10 hours of curriculum time spread during 6 weeks, and was both culturally tailored and based on accepted USDA MyPlate guidelines that

complemented the standard health education curriculum in their schools. The feasibility of the Spice MyPlate intervention was manifested in the successful delivery of every intervention session, and the exceptionally high participant compliance with the completion and returning of the 3-day food records and healthy eating attitude questionnaires (94% in both study groups at intervention end) suggests that the evaluation strategy was also feasible in this setting. An additional strength of the study was the follow-up outcomes assessment conducted 1 month after the conclusion of the intervention. This allowed the investigators to assess whether any changes in diet quality or healthy eating attitudes that occurred during the intervention persisted in the shortterm. Compliance at this follow-up assessment 1 month later was also relatively high (85% Spice MyPlate, 76% control) and likely would have been higher had data collection not occurred during the final 2 weeks of the school year, when many seniors had finished their academic requirements and were no longer at school. Future studies would be well served to perform all data collection several weeks in advance of the conclusion of the school year. There were also several notable limitations to this study. Participants were not randomly assigned to the Spice MyPlate or control groups. Furthermore, the Spice MyPlate group was

354 drawn entirely from one school and the control group was drawn entirely from another school. Although this likely created some bias due to inherent differences between the students at the two schools that could not be adjusted for in regression models, the research team believed that this design would be the best way to prevent ‘‘contamination’’ of the school-based intervention content between students in the Spice MyPlate and control groups if both intervention arms were offered at the same school. Staff at the intervention schools advised the research team that there would be interaction and likely sharing of information between students in the two arms of this study that would occur outside of health class during the remainder of the school day if both groups were offered at the same school. This was believed to be a particularly important consideration because the cooking and food preparation sessions with a professional chef, and other Spice MyPlate intervention components were highly visible to other students during the school day. Nevertheless, although the Baltimore high schools in this study were intentionally chosen for their relative similarities with respect to socioeconomic status, racial minority composition, and geographic proximity, there were some differences between the study groups at baseline by age and race. The improvements in diet quality in the Spice MyPlate group compared with the control group withstood adjustment for age, race, gender, and baseline dietary intake in the regression models, but there may still have been some residual differences between the study populations. Feasibility and preliminary evidence of benefit of Spice MyPlate were established in this pilot study, but randomized allocation will increase the rigor of future studies. Another limitation is that there was the potential for recall bias in this study because the 3-day food records used to assess diet quality were self-reported. However, 3-day food records have been validated for usual dietary intake28 and were believed to be a more feasible method of dietary assessment in this setting than 24-hour dietary recalls or Food Frequency Questionnaires. Nevertheless, although every 3-day food record collected in this study was reviewed and entered into a standardized database (USDA SuperTracker) by a registered dietitian and subject to collaborative secondary review with other investigators on the research team, there was likely some degree of misreporting of food quantities and other inaccuracies in the food records by participants in both study groups. An additional limitation is that changes in indicators of diet quality were represented by quantities of dietary intake of the five groups of USDA MyPlate (vegetables, fruits, whole grains, protein foods, and dairy) that were consumed and were not adjusted for total energy intake. It is possible that the improvements in diet quality noted in this study may have been due to participants in the Spice MyPlate group eating substantially more food than the control group during the study. However, the authors do not believe that this is a likely explanation for the findings during this relatively short intervention period, in which the potential to change the quantity of food that students consumed was limited by several factors.

American Journal of Health Promotion 30(5) A related limitation is that there were likely restrictions to the changes in diet quality that were possible in this study. Approximately 75% of students in both of the high schools participating in this study were on the standardized free and reduced-price breakfast and lunch programs at schools. Therefore, the relative standardization of dietary intake in two primary meals during the school week likely incurred a ceiling on the ability to change the foods that the participants consumed during the study period. Access to healthy foods was also generally limited in the neighborhoods in which most participating students lived in both study groups. In light of these barriers that likely restricted the potential for improvement in objective measures of diet quality at this point in the students’ lives, the investigators also assessed various subjective attitudes toward healthy eating that were hoped to become ingrained in the students. Although improvements were noted in these outcomes, there were several limitations to the healthy eating attitude assessments. The healthy eating attitudes questionnaires had not been previously validated and the participants

SO WHAT? Implications for Health Promotion Practitioners and Researchers What is already known on this topic? Previous nutrition education interventions among children and adolescents have revealed mixed results.11-15 However, experiential and culturally tailored nutrition education has been shown to improve the diet quality and healthy eating attitudes of younger children and some racial minorities, particularly Latinos.15,19

What does this article add? To date, no studies have evaluated the feasibility and effectiveness of an experiential nutrition education intervention tailored to urban, African-American adolescents. This article begins to fill an important gap because African-American adolescents have been shown to have lower diet quality33 and suffer the highest prevalence of obesity of all racial groups.1

What are the implications for health promotion practice or research? Although randomized trials in larger and more generalizable samples are needed to more rigorously assess the effectiveness of the Spice MyPlate intervention as an adjuvant to standard nutrition education, the preliminary results of this pilot study suggest that this school-based intervention is feasible and may offer modest improvement in some indicators of diet quality and healthy eating attitudes among urban and predominantly AfricanAmerican adolescent populations.

D’Adamo et al. were also asked to report their attitudes toward food groups that may not have been entirely clear to them. However, one of the primary points of emphasis in the standard nutrition education that participants in both the control and Spice MyPlate groups received was providing specific examples of the types of foods that fit into each food group of MyPlate. The investigators believed that this was essential not only to help participating students improve healthy eating habits, but also to properly categorize the food group questions that were asked in the study. Despite the limitations of the healthy eating questionnaires, the standardized meals and barriers in access to healthy foods may partially explain the finding that participants in the Spice MyPlate group reported that they would be more likely to eat vegetables if flavored with spices and herbs at all study points, yet did not report an actual increase in dietary intake of vegetables at any study point. In light of the potential restrictions on the ability to improve diet quality in this setting, the effectiveness of the Spice MyPlate intervention in improving diet quality may have been underestimated in this study. Although the feasibility and modest improvements in diet quality and healthy eating attitudes observed in this pilot study of a novel nutrition education intervention are encouraging, further research in larger, more generalizable, and randomized study samples is needed to confirm these findings. Acknowledgments The authors would like to thank the staff and administration of the University of Maryland School of Medicine Center for Integrative Medicine and The Institute for Integrative Health for their assistance with the study; the administration at the participating high schools for supporting the Spice MyPlate intervention; the administration of HealthCorps for offering their Coordinators to deliver the Spice MyPlate intervention; and the study participants and their families for their involvement.

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

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

References 1. Centers for Disease Control and Prevention. Prevalence of obesity among children and adolescents: United States, trends 1963–1965 through 2009–2010. Available at: http://www.cdc.gov/nchs/data/ hestat/obesity_child_09_10/obesity_child_09_10.htm. Accessed April 12, 2014. 2. Hannon TS, Rao G, Arslanian SA. Childhood obesity and type 2 diabetes mellitus. Pediatrics. 2005;116:473-480. 3. Freedman DS, Mei Z, Srinivasan SR, et al. Cardiovascular risk factors and excess adiposity among overweight children and adolescents: the Bogalusa Heart Study. J Pediatr. 2007;150:12-17.e2. 4. Schwartz MB, Puhl R. Childhood obesity: a societal problem to solve. Obes Rev. 2003;4: 57-71.

355 5. Dehghan M, Akhtar-Danesh N, Merchant AT. Childhood obesity, prevalence and prevention. Nutr J. 2005;4:24. 6. Boggs DA, Rosenberg L, Rodrı´guez-Bernal CL, Palmer JR. Long-term diet quality is associated with lower obesity risk in young African American women with normal BMI at baseline. J Nutr. 2013;143:1636-1641. 7. Ness AR, Maynard M, Frankel S, et al. Diet in childhood and adult cardiovascular and all cause mortality: the Boyd Orr cohort. Heart. 2005;91:894-898. 8. Ainuki T, Akamatsu R, Hayashi F, Takemi Y. Association of enjoyable childhood mealtimes with adult eating behaviors and subjective diet-related quality of life. J Nutr Educ Behav. 2013; 45:274-278. 9. Paulozzi LJ. Does inadequate diet during childhood explain the higher high fracture rates in the Southern United States? Osteoporos Int. 2010;21:417-423. 10. Mikkila¨ V, Ra¨sa¨nen L, Raitakari OT, et al. Longitudinal changes in diet from childhood into adulthood with respect to risk of cardiovascular diseases: The Cardiovascular Risk in Young Finns Study. Eur J Clin Nutr. 2004;58:1038-1045. 11. Perry CL, Bishop DB, Taylor G, et al. Changing fruit and vegetable consumption among children: the 5-a-Day Power Plus program in St. Paul, Minnesota. Am J Public Health. 1998;88: 603-609. 12. Spence AC, McNaughton SA, Lioret S, et al. A health promotion intervention can affect diet quality in early childhood. J Nutr. 2013;143:1672-1678. 13. Herman A, Nelson BB, Teutsch C, Chung PJ. ‘‘Eat Healthy, Stay Active!’’: a coordinated intervention to improve nutrition and physical activity among Head Start parents, staff, and children. Am J Health Promot. 2012;27: e27-e36. 14. Lytle LA, Murray DM, Perry CL, et al. School-based approaches to affect adolescents’ diets: results from the TEENS study. Health Educ Behav. 2004;31:270-287. 15. Davis JN, Ventura EE, Cook LT, et al. LA Sprouts: a gardening, nutrition, and cooking intervention for Latino youth improves diet and reduces obesity. J Am Diet Assoc. 2011;111:1224-1230. 16. Wansink B, Kranz S. Who’s using MyPlate? J Nutr Educ Behav. 2013;45:728-732. 17. Karpyn A, Young C, Weiss S. Reestablishing healthy food retail: changing the landscape of food deserts. Child Obes. 2012;8: 28-30. 18. Feinberg E, Kavanagh PL, Young RL, Prudent N. Food insecurity and compensatory feeding practices among urban black families. Pediatrics. 2008;122: e854-e860. 19. Gatto NM, Ventura EE, Cook LT, et al. LA Sprouts: a gardenbased nutrition intervention pilot program influences motivation and preferences for fruits and vegetables in Latino youth. J Acad Nutr Diet. 2012;112:913-920. 20. D’Angelo H, Suratkar S, Song HJ, et al. Access to food source and food source use are associated with healthy and unhealthy foodpurchasing behaviours among low-income African-American adults in Baltimore City. Public Health Nutr. 2011;14:1632-1639. 21. US Dept of Agriculture. Choose MyPlate–food groups. Available at: http://www.choosemyplate.gov/food-groups/. Accessed May 24, 2014.

356 22. HealthCorps. Who we are – all about HealthCorps. Available at: http://healthcorps.org/who-we-are/about/. Accessed May 27, 2014. 23. The Institute for Integrative Health. The Institute for Integrative Health–about us. Available at: http://tiih.org/about-us. Accessed May 27, 2014. 24. Allen RW, Schwartzman E, Baker WL, et al. Cinnamon use in type 2 diabetes: an updated systematic review and metaanalysis. Ann Fam Med. 2013;11:452-459. 25. Borek C. Garlic reduces dementia and heart-disease risk. J Nutr. 2006;136(suppl 3):810S-812S. 26. Grzanna R, Lindmark L, Frondoza CG. Ginger – an herbal medicinal product with broad anti-inflammatory actions. J Med Food. 2005;8:125-132. 27. Srinivasan K. Antioxidant potential of spices and their active constituents. Crit Rev Food Sci Nutr. 2014;54:352-372.

American Journal of Health Promotion 30(5) 28. Jorgensen LM, Isaksson B, Schroll M. Reproducibility and validity of 3-day food records. Eur J Clin Nutr. 1992;46: 729-734. 29. Post RC, Maniscalco S, Herrup M, Chang S. What’s new on MyPlate?: a new message, redesigned web site, and SuperTracker debut. J Acad Nutr Diet. 2012;112:18-22. 30. Post RC, Herrup M, Chang S, Leone A. Getting plates in shape using SuperTracker. J Acad Nutr Diet. 2012;112:354-358. 31. Post RC, Haven J, Chang S, Bard S. Making SuperTracker work for you. J Acad Nutr Diet. 2012;112:1520-1522. 32. Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous outcomes. Biometrics. 1986;42:121-130. 33. Wang Y, Jahns L, Tussing-Humphreys L, et al. Dietary intake patterns of low-income urban African-American adolescents. J Am Diet Assoc. 2010;110:1340-1345.