A youth mentor-led nutritional intervention in urban recreation centers: a promising strategy for childhood obesity prevention in low-income neighborhoods

Abstract

B’More Healthy Community for Kids (BHCK) is an ongoing multi-level intervention to prevent childhood obesity in African-American low-income neighborhoods in Baltimore city, MD. Although previous nutrition interventions involving peer mentoring of youth have been successful, there is a lack of studies evaluating the influence of cross-age peers within interventions targeting youth. This article evaluates the implementation of the BHCK intervention in recreation centers, and describes lessons learned. Sixteen youth leaders delivered bi-weekly, interactive sessions to 10- to 14-y olds. Dose, fidelity and reach are assessed, as is qualitative information regarding what worked well during sessions. Dose is operationalized as the number of interactive sessions, and taste tests, giveaways and handouts per session; fidelity as the number of youth leaders participating in the entire intervention and per session and reach as the number of interactions with the target population. Based on a priori set values, number of interactive sessions was high, and number of taste tests, giveaways and handouts was moderate to high (dose). The number of participating youth leaders was also high (fidelity). Of the 14 planned sessions, the intervention was implemented with high/moderate reach. Data suggest that working with cross-age peers is a promising nutritional intervention for recreation centers.

Introduction

The rate of childhood obesity has skyrocketed in the past decades and has become a public health priority. One-third of children in the United States are currently overweight or obese. The prevalence is much higher among certain groups, with 41.6% of African-American children and 44% of youth living under the federal poverty level classified as obese [1]. Childhood obesity has both physical and psychosocial repercussions and leads to short- and long-term health burdens [2]. Also, obese children are more likely to become obese teenagers and adults—two studies conducted in the United States with representative samples found that 77–92% of obese adolescents became obese adults [3].

Most nutrition interventions focusing on childhood obesity have centered on schools. However, those programs have yielded conflicting results [4], while some meta-analysis described effectiveness in reducing the children’s body mass index (BMI) [5,6], others reported no BMI improvement [7,8]. One possible reason is that students may exhibit positive changes in their behavior during school hours, but revert to an unhealthful diet once outside of school. Expanding the intervention to other settings where youth spend their time would reinforce exposure to healthy eating messages [9].

Time constraints during the school day and competing priorities have also been ascribed to the uneven success seen in school interventions [10]. Thus, after-school programs are a promising venue for health promotion and youth obesity prevention interventions in cities with such centers [11]. Despite this potential, few interventions have worked with after-school programs, including those held in recreation centers. Many recreation centers are open to new curriculum enhancements and are considered a safe place for youth to spend time [12].

An increasing number of interventions have implemented peer mentoring approaches [13]. Peer mentoring has the potential to make concepts relevant and engaging to other youth, serving an influential role during adolescence [14]. Peer-mentors may be particularly beneficial when peers are slightly older (‘cross-age’) than the target population [13], but little attention has been given to interventions with cross-age peers. In addition while many nutritional interventions have focused on peer mentoring for children, very few have targeted older youth and early adolescents [15].

Baltimore Healthy Eating Zones (BHEZ) was an environmental nutrition intervention targeting early adolescents during the years 2009–10. In BHEZ, peers were trained to assist interventionists in delivering a nutrition curriculum in recreation centers to adolescents. The program was successful but delivery of the peer component was only moderately implemented [11]. Based on their experience with BHEZ and existing literature, Shin et al. [16] suggest that the delivery of peer mentoring in recreation centers could be improved by having cross-age peers with more experience.

It has been recognized that obesity has a complex, multi-factorial etiology, thus nutritional education should be implemented in conjunction with other approaches (e.g. environmental, political) in order to promote sustainable dietary changes [9]. Obesity in low-income urban areas has been associated with environmental factors such as low availability and high prices of healthy foods, especially when compared with the low prices of high-fat and high-sugar foods [12]. A study conducted in Baltimore observed that 43% of African-American neighborhoods had low availability of healthy foods [17]. Despite this population vulnerability to obesity, there has been a lack of obesity prevention studies in low-income urban populations.

The B’More Healthy Community for Kids (BHCK) trial is a multi-level intervention to prevent childhood obesity in 10- to 14-y-old African-American youth living in low-income neighborhoods of Baltimore, MD. The intervention involves multiple components at different levels (policy makers, food retailers, youth leaders in recreation centers and caregivers). The youth-targeted nutrition education component was delivered in recreation centers by 18- to 22-y-old youth leaders. The trial was guided by Social Cognitive Theory (SCT), Social Ecology and Systems Theory [8]. SCT offers a theoretical underpinning for the facilitation of behavior changes through the use of role models in health-related interventions [18, 19], while Social Ecology and Systems Theory offer an understanding of how the individual’s behaviors interact with multiple other levels of the environment [20].

Process evaluation is a vital step in evaluating the impact of an intervention [21]. However, there is currently no published literature on process evaluation relating to nutrition programs in recreation centers delivered by cross-age peers. Therefore, the purpose of this article is to present and evaluate the implementation of the BHCK nutrition curriculum by youth leaders in recreation centers. We describe the nutrition changes promoted by BHCK in recreation centers and address three research questions: (i) How was BHCK’s nutrition curriculum delivered in recreation centers? (ii) Which education practices worked best? (iii) Was nutrition mentoring with cross-age peers in urban recreation centers a successful approach?

Materials and methods

Setting

Baltimore has a pattern of racial and economic segregation. The majority of the population in Baltimore city is African-Americans (63.7%) and 23.8% of the population lives below the poverty line [22]. Gittelsohn et al. [11] observed that 43% of African-American youth in Baltimore were obese or overweight. In Baltimore, most low-income, African-American neighborhoods have a recreation center [12]. Baltimore city recreation centers are typically staffed by a director and 2–3 staffs. They are generally comprised 1–2 large gym spaces, a kitchen and ancillary rooms for classes and other activities [23].

Recreation centers offer after school programming for elementary and middle school-aged youth that typically involves physical activity classes (i.e. karate and dance) and free play, homework time, arts and crafts, access to a computer lab, and other activities. Youth in the after school program receive a dinner meal and snacks provided by Baltimore social service agencies. The vast majority of youth who attend these centers are African-Americans, from low-income households, receive free or reduced-cost lunch and purchase foods from corner stores and carry-outs surrounding the centers, before and after school.

Sample

Neighborhoods eligible for the intervention had a recreation center, were located in food deserts [24] and were predominantly low-income, African American population [12]. Once we established an agreement with the Department of Recreation and Parks, our team obtained a sample of 30 recreation centers. For the first wave of the trial, 14 neighborhoods were randomized. After project approval by the Baltimore City Recreation and Parks Department, recreation center directors in the eligible neighborhoods were invited to participate in the intervention. Then centers were randomized to intervention (n = 7) and control (n = 7) zones.

Youth leaders

The project utilized social modeling via cross-age peer-leaders. Youth leaders were recruited from local high schools, colleges, youth organizations and General Educational Development programs. Sixteen youth (age, 18–22 y) were selected and trained to deliver the BHCK nutrition curriculum to 10- to 14-y olds in intervention recreation centers. To become a youth leader, interested youth had to successfully complete an application, an in-person interview and a 12-session training program. Selected youth leaders were paid for the completion of the training program and for their time conducting nutrition sessions at recreation centers [12].

Formative research

Extensive formative research with community partners and young people from intervention neighborhoods was conducted to guide program planning and material development. In-depth interviews (n = 64) were used to identify appropriate information channels and material content.

The selection of foods and beverages promoted during the intervention was based on a two-stage process. First, study staff observed the inventory at local wholesalers to create lists of available and affordable healthy foods and beverages. Food and beverage items successfully promoted in previous retail food interventions [25, 26] were preferentially selected from the foods lists. Then four focus groups (with a combined total of 43 youth participants) were conducted with children in Baltimore city recreation centers to select the most acceptable items. Youth were also involved in selecting materials to promote (e.g. images, videos, posters, slogans and messages).

The curriculum for the 14-week nutrition curriculum to be delivered by the youth leaders was developed in partnership with a youth-led community-based organization in Baltimore city. The curriculum drew elements from pre-existing nutrition resources, e.g. the Food & Fun After-School created by Harvard School of Public Health Prevention Research Center. The pre-existing resources were then culturally tailored for Baltimore youth, pilot tested for feasibility and acceptability and the final curriculum was adjusted based on lessons learned from piloting.

Process evaluation

A formal process evaluation of an intervention can assess whether the program was properly delivered and if the target population was sufficiently exposed to the program [21]. Prior to the intervention, implementation standards were defined based on the research group’s previous experience with program implementation in Baltimore, one of which (BHEZ) was implemented in recreation centers [11, 27]. Standards addressed three constructs normally used in process evaluation of health intervention programs: dose, fidelity and reach [25]. Dose refers to the extent to which intervention components were delivered to target population. Fidelity indicates how well delivering components of the intervention followed the initial plan and reach measures the extent to which the target population was exposed to the intervention [26]. Standards and variables for dose, fidelity and reach were set a priori to the intervention and are presented in Table I.

Table I.

BHCK recreation center intervention process evaluation standards

Intervention standarda	Low	Medium	High
Dose
Number of visits to recreation centers	<47	47–70	>70
Number of food samples given per session	<7	7–12	>12
Number of handouts given per session (when applicable)	<7	7–12	>12
Number of giveaways given per session (when applicable)	<7	7–12	>12
Fidelity
Number of youth leaders participating in the entire intervention	<5	5–10	>10
Number of visits by youth leaders to recreation centers	<7	7–10	>10
Number of youth leaders involved in each session	<2	2–4	>4
Reach
Number of 10–14 y old per session	<5	5–10	>10

^aStandards were set before the program started and monitored every 2 mo during the intervention.

Instruments

Data on the process of the intervention implementation were collected using an interventionist form. BHCK staff members supervised recreation center sessions and completed the interventionist form in real time during each session. Information collected included average duration of interactive sessions, number of youth leaders present in each session, number of posters displayed at the recreation center, description of activities held, number of participants attending the sessions, type and number of taste tests, handouts and giveaways delivered and feedback on taste tests (liked, somewhat liked and disliked). Descriptive analyses were performed using SPSS version 17.0 (IBMI, Chicago, IL) and values were compared with standards.

Qualitative data were also collected via an interventionist-completed form. BHCK staff recorded comments on aspects that worked and that did not work in each session, and activities children enjoyed the most. Qualitative data on taste tests were collected using a rating sheet with participant’s comments on how they would change the food to be more appealing to them. All qualitative data collected were organized by themes and summarized [28]. Comments that were repeated in several interactive sessions and taste tests are presented in the Results section.

Results

Promotion of nutrition education in recreation centers

Nutrition education in recreation centers was divided into four phases: beverages, snacks, breakfast and cooking. The topics, activities and taste tests held at recreation centers mirrored the topics discussed in other components of the program (corner stores, carry-out, social media and wholesaler). All phases were delivered in four different sessions with the exception of breakfast, to which two sessions were devoted (total of 14 sessions). Each session had a theme and educational content was constructed to address that theme. Sessions were delivered bi-weekly by youth leaders, from July 2014 to January 2015. Average session length was 61.4 min (SD 13.09), ranging from 30 to 98 min. Every session started with an introduction/ice-breaker, followed by an opening discussion on the day’s topic. Youth leaders had a script of questions to probe during the discussion. After the discussion, youth were engaged in different educational activities (e.g. video, game and skit), followed by an experiential learning activity which was often in the form of an active game. In the cooking phase (last four sessions), educational and experimental learning activities were replaced by hands-on cooking classes.

During the session, youth had the opportunity to taste test food samples related to the educational component and received a giveaway. In the cooking phase, giveaways were replaced by a handout with the day’s recipe. Table II describes sessions and their components. At the end of each session, youth leaders and participants summarized the messages of the day to help consolidate what was learned.

Table II.

BHCK’s nutritional curriculum—themes, educational content, activities, experimental learning activities, taste tests, handouts and giveaways^a

Session	Theme	Educational content	Activity	Experimental learning activity	Taste test	Giveaway
1	Rethink your drink	Amount of sugar we consume in one year The amount of sugar in common drinks Why should we avoid too much sugar Low-sugar drink options	Cards matching game (drinks and amount of sugar)	Sweetened beverage tag with beverage chains	Sugar-free drink mix	BHCK sunglasses
2	Stoplight beverages	Introduce stoplight method for selecting food Categorizing a selection of beverages	Red, yellow and green light beverages	Healthy beverage bowling	Lemonade and diet half and half	BHCK drawstring bags
3	Why water?	Function of water in the body Hydration and dehydration Effects of dehydration	Tracing body and coloring percentage of it made up of water	‘Drink water’ obstacle course	Water with fruit	BHCK water bottle
4	Fruit imposters	Fruit has sugar in it, but it is okay to eat What is fiber? How does it affect your body? Slow vs fast carbs	Food for thought video Reading ingredients label of fruit beverages	Differentiate fruit juices and ‘fruit imposters’ beverages	Oranges and grapefruit	Grape or orange stress balls
5	The purpose of snacks and healthy snacking behavior	Discuss the purpose of snacks Discuss healthy snack options Discussing of healthy snaking behavior Provide tips on healthy snacking behaviors	Healthy battle video	Smart snack Jeopardy game	Low-fat string cheese	BHCK pens
6	Snack alternatives	Explain why it is unhealthy to eat too much fat and salt Explain why it is bad to eat too much sugar Facts about African American diet and health	Measuring amount of sugar/fat in snacks Choose what you chew video	Skits about making healthy food choice in different scenarios	Low-fat granola bars	BHCK sunglasses
7	Snack sleuth	Food companies spend big bucks to get you to buy food Advertising tricks Some ads are so powerful we recognize them without even seeing anything about the product	Design a healthy food ad	Food advertisement Family Feud	Baked chips	Grape or orange stress balls
8	Stoplight snacks	Review the stoplight method for selecting food Review the purpose of snacks and things to avoid Discuss how to identify snacks in each category at the corner store	Classifying snacks into ‘green’, ‘yellow’ and ‘red’	Red light, green light and eat right game	Carrots	Shopping bags
9	Breakfast as an on-the-go snack	Discuss breakfast benefits Ideas for fast, healthy, portable snacks	Identifying examples of foods for a healthy breakfast	Healthy house game	Low-sugar cereal	Portion plates
10	Stoplight breakfasts	Review the stoplight method for selecting food Review the purpose and benefits of breakfast Identify food in the corner store that could serve as a healthy breakfast	Classifying breakfast into ‘green’, ‘yellow’ and ‘red’	Breakfast benefits obstacle course	Bagel with peanut butter	BHCK pens
11	Cooking 1	How to create a veggie omelet and omelet with lean meat How to crack eggs and use a whisk	Nutritional lesson on vegetables	Cooking ‘Power-up Omelets’	Omelets	Power-up Omelet recipe card
12	Cooking 2	How to create crunchy parmesan chicken strips without deep frying the chicken How to cut boneless chicken breast into strips Being safe when handling raw meat, including sanitizing and proper cooking temperatures	Healthy cooking video	Cooking ‘What’s Clucking Crispy Chicken’	Crispy chicken nuggets	What’s Clucking Crispy Chicken recipe card
13	Cooking 3	How to boil noodles How to reduce the sodium when eating ramen noodles	Nutritional lesson on sodium	Noodle re-design	Ramen noodles with vegetables	High Energy Noodle recipe
14	Cooking 4	How to use cooking spray instead of higher fat alternatives when cooking meals	Nutritional lesson on full-fat products versus light products	Cooking ‘Crazy Quesadillas’ and home-made salsa	Quesadillas and salsa	Crazy Quesadilla and salsa recipe card

^aMore information on the curriculum and intervention components can be found on www.healthystores.org.

At the end of each phase, interventionists gave relevant posters to each recreation center director. The total number of posters distributed was 11: three for the first and fourth phases each, four for the second phase and one for the third phase. Posters provided information on the phase’s theme, e.g. for the snacks phase, the content of one of the poster’s was ‘Hungry for a Salty Snack: try popcorn, pretzels or baked chips – choose snacks with 6 grams of fat or less’.

Evaluation of nutrition education activities in recreation centers

By the end of the intervention, recreation center directors had received posters from phases one, two and three (n = 8); posters from phase four were given at the last day of intervention and were therefore not included in this evaluation. The mean observed number of posters displayed at the end of the intervention was 1.66 (SD 1.96) per recreation center.

Dose

The youth leaders completed a total of 98 visits to recreation centers over the course of 7 mo, which met our ‘high’ standard (>70 visits). During these visits, a total of 1085 food samples, 307 handouts and 739 giveaways were distributed. When analyzed per session, achievement of distribution of six taste tests (43%), six giveaways (60%) and two handouts (50%) was high. No items had low distribution. Averages of food samples, handouts and giveaways per session are presented in Table III.

Table III.

Mean numbers of BHCK food samples, handouts, giveaways, youth leaders and youth per interactive session in recreation centers

Session	Number of taste tests		Number of giveaways/handouts distributed		Number of youth leaders delivering the curriculum		Number of youth (10–14 y old) attending the sessions
	Mean	SD	Mean	SD	Mean	SD	Mean	SD
1	15.13a	8.34	12.86a	8.07	5.14a	0.45	15.57a	5.96
2	17.83a	5.38	17.43a	4.86	4.43a	0.68	14.00a	6.95
3	16.57a	7.41	14.33a	5.61	4.14a	0.50	13.86a	7.01
4	15.00a	3.63	14.00a	6.39	4.00	0.18	11.86a	5.49
5	10.71	6.67	14.20a	7.04	3.17	0.47	8.00	4.12
6	8.00	3.84	11.00	4.69	3.43	0.48	8.86	3.33
7	10.00	3.83	9.00	4.35	3.14	0.50	7.14	4.70
8	9.83	4.30	11.50a	8.24	3.33	0.21	6.71	4.19
9	9.00	4.12	10.43	5.47	2.29	0.47	7.17	2.99
10	8.86	2.61	7.50	4.18	2.57	0.42	10.29a	5.28
11	11.25a	5.03	11.00	5.26	2.57	0.36	8.29	2.28
12	7.60	3.05	13.86a	3.93	3.71	0.42	9.14	4.22
13	14.80a	0.83	11.75	4.99	3.00	0.44	8.71	5.55
14	9.00	4.78	14.33a	2.73	2.43	0.29	10.14a	5.87

^aValues that meet pre-defined standards as high.

A total of 1049 taste tests were distributed and rated, with an average of 10.7 taste tests per session. Overall participants liked the taste tests. Crispy chicken was prepared on site by youth with the help of youth leaders and was the taste test with the highest number of ‘liked’ rate (mean 76%; SD 20.5). Baked chips and oranges were also well received by the participants, with a mean rating of liking of 71% (SD 13.8) and 71% (SD 33.0), respectively. Fruit flavored water (water flavored with pieces of fruit such as strawberries and limes) was the food sample least liked by the participants (28%; SD 21.3). To improve the food sample component, participants suggested adding more fruit to fruit flavored water, adding sugar to fruits, cereal and beverages and to adding dressing and seasoning in vegetables and savory food preparations. Acceptance rates for each food sample are included in Table IV.

Table IV.

Rates of youth who liked taste tests

Session/taste test	Number of youth who reported liking the taste test		Acceptance rate (%)
	Mean	SD
1. Sugar Free drink mix (Hawaian Punch)	9.57	6.29	46.12
2. Diet lemonade/half and half	9.28	4.46	60.34
3. Fruit flavored water	5.00	3.57	27.55
4. Orange slices	11.57	5.79	70.55
5. Low-fat string cheese	8.85	6.17	62.42
6. Low fat, low-sugar granola bar	7.42	3.35	66.11
7. Baked chips	8.28	2.81	71.06
8. Baby carrots	4.42	2.43	43.10
9. Low-sugar cereal	6.16	2.92	50.53
10. Whole wheat bagel with peanut butter	5.85	3.28	43.32
11. Veggie omelet	6.71	2.81	54.22
12. Baked crispy chicken	10.85	3.97	76.43
13. Healthy ramen noodles	6.57	3.86	50.90
14. Homemade salsa	3.75	2.98	57.25
14 Whole wheat veggie quesadillas	8.83	4.44	31.50

Fidelity

The number of youth leaders actively participating in the entire intervention was high (n = 12/16; 75%). The number of visits by youth leaders to recreation centers was also high; all recreation centers received 14 visits. The mean number of youth leaders present in each session was 3.39 (SD 1.34). The number of youth leaders per session was high in the first three sessions. No sessions achieved a ‘low’ standard for mean number of youth leaders. The number of youth leaders per session is presented in Table III.

Reach

Interactive sessions reached a total of 1412 interactions with people of different ages, with an average of 14.56 people per session (SD 5.96). The highest number of attendees in one session was 36, and the lowest was 3. Mean number of younger children (6–10 y) per session was 3.01 (SD 3.59), mean number of older teens (15–17 y) was 0.46 (SD 1.42) and mean number of adults was 1.07 (SD 1.64). Total of interactions with target population was 971 (this number includes repeated interactions with youth that came to multiple sessions). The mean number of target-population participants (10–14 y) per session was 10.01 (SD 5.44), which is high by our standards. When divided by session, six interactive sessions met high standards, while eight sessions met medium standards. Mean numbers of children aged 10–14 per session are presented in Table III.

Open-ended comments recorded on the process evaluation form showed that youth liked and were very engaged in most of the games. Games in which they had to try to get answers right, such as Family Feud and Jeopardy-style trivia games were the most successful. Cooking lessons produced the most comments reflecting youth engagement; children were particularly excited about cutting vegetables and mixing ingredients. Discussions were the most challenging activity in which to engage youth. Children were very participative in some discussions (e.g. discussion about hydration), but not as interested in others (e.g. discussion about healthy snacks). One successful strategy used to engage youth in the discussions was using a poster board for them to write on the topics that they discussed about during the session.

Discussion

This is the first study to evaluate a nutrition intervention entirely delivered by cross-age peers in low-income urban recreation centers. Measures of fidelity were high overall, supporting the feasibility of using youth leaders to implement this and other interventions in recreation centers. In addition, reach measures for our target population were moderate to high, suggesting that cross-age mentoring is a promising nutrition intervention approach for early adolescents.

This study complements BHEZ’s findings on implementing programs in recreation centers and suggests directions for the expansion of nutrition interventions beyond schools. For example, although same-age peer mentoring is a suitable approach within schools, BHEZ, a recreation center intervention, faced some challenges in this new setting, such as scheduling conflicts during the school year, and reliance on public transportation [11]. In contrast, BHCK’s college youth leaders had more experience using public transportation and more flexible schedules, which facilitated their participation. Moreover, youth leaders had prior knowledge about health or youth development, as most of them were majoring in nutrition, nursing or elementary education.

Further, in BHEZ, same-age mentors were not responsible for delivering the entire curriculum; their role was assisting interventionists and helping with intervention activities. Eventually they were expected to conduct activities independently. However, mentors dropped out progressively and did not achieve self-sufficiency in intervention delivery. Only 61.5% of the youth selected to be youth leaders in BHEZ attended all introductory training sessions and only 38.4% remained involved until the end of the intervention [11]. In BHCK, peers delivered the entire educational component throughout the intervention. They all completed the training by attending each session or completing a make-up session, and 75% of them actively participated throughout the entire intervention. This study corroborates Shin et al. [16], who suggested that interventions in recreation centers could be improved by using older peer mentors, as in other studies.

Training youth to teach other youth about nutrition and healthy eating without relying on adult interventionists or teachers is an important step toward sustainability of this or similar programs. However, our program was not fully sustainable. One approach to make the program sustainable in the future would be partnering with the Baltimore City Department of Recreation and Parks. Recreation center directors and staff would receive training to manage the program and the junior youth leaders would be integrated as a part of the recreation centers’ regular programming. Other strategies such as maintaining nutritional curriculum after the intervention are over and making changes in recreation center food environments, as well as implementing wellness policy could also help improve the changes promoted by BHCK. Studies exploring strategies to increase sustainability of nutrition interventions, especially outside of schools, are needed.

Dose delivered was moderate to high, and giveaways achieved high dose delivered. Taste tests had mostly moderate dose delivered, but the mean number of food taste tests slightly increased from that achieved within last recreation center nutrition intervention performed by this research group (BHEZ), from 10 to 11.68 food samples per session. Youth were sometimes hesitant to consume new foods. However, studies have shown that youth’s preferences for new tastes increase with repeated exposure [29] and that they are more willing to try new foods in social situations, such as those created as part of this intervention [29–31].

Low-sugar sweets are important food to promote but could be challenging due to personal food preferences. Our participants seemed to prefer high-sugar foods, as they often suggested adding more sugar to our food samples. High consumption of sugar by American youth in Baltimore City was observed by Kolahdooz et al. [17], who found that 71% of 9- to 13-y olds reported drinking sugary drinks in a 24-h dietary recall. It is important for interventions to identify and promote foods that are acceptable replacements to the food/beverage currently being consumed.

Vegetables were not very well accepted. Youth opinions on how to improve taste tests suggested that incorporating low-fat sauces and dressings could make vegetables more palatable to them. A study showed that improving culinary skills in youth could also improve their acceptance of vegetables, as it was observed to increase consumption of fruits, vegetables and whole grains and to promote long-term health [32, 33]. As described elsewhere [11], the BHEZ trial observed that recreation centers had the structure for intervention activities to be expanded, and to include cooking classes and games. BHEZ’s cooking lessons were well received, but reached a small number of children [11]. In the BHCK trial, reach of youth participating was improved, in the four cooking sessions mean number of 10–14 y olds per session was moderate (during three sessions) and high (during one session).

Despite the potential of culinary education, a majority of such programs have been minimally effective, possibly because of the lack of influential role models to serve as educators. Peer-based ways to improve teenagers’ culinary skills could therefore be a promising future direction. Peers can collaborate to create a positive social affective context to cooking lessons and increase adolescents’ healthy foods familiarity and acceptance [14]. Further, in the current multilevel trial, teaching adolescents how to select, handle and prepare healthy foods have the potential to enhance the effect of other program components by enabling youth to choose and cook the healthy foods brought and promoted in corner-stores.

Overall, the intervention was successful, but some of the taste tests and activities could be improved. Substituting fruit-flavored water for a low-sugar version of a drink youth already know could improve acceptance. One limitation of the intervention was the low number of posters placed and maintained in recreation centers, which limited message exposure; one way to improve this might be for the interventionists to place posters themselves and to replace them in case they were taken down.

This study addresses one possible way to deliver a nutrition curriculum to youth that goes further than just education. According to the SCT the youth leader component has the potential to be more than only an educational component, but to promote observational learning and social modeling [34]. We suggest that after-school programs can complement schools and other community messages about nutrition, as part of multi-level interventions. In BHCK, we coordinated the messages promoted in the recreation centers with intervention components in corner stores, carry-outs and on social media. Thus by simultaneously promoting food environment changes, the project aims not only to increase availability of healthy foods in those areas but also to allow application of the nutrition lessons taught and support the behavioral changes proposed by the program. Future work could expand on the number and types of components in the intervention to include health classes provided by the school, and the school environment. Future practice should coordinate this nutrition curriculum with health courses in high school, so students receive course credit.

This study has limitations in terms of assessment of the impact of intervention sessions on the target population. As part of a larger study, it was not our goal to measure the impact of the interactive sessions in recreation centers alone, but to investigate how the intervention as a whole could impact African-American youth in Baltimore city. For future evaluations, it might be useful to incorporate data collection to observe which parts of the intervention in recreation centers were most meaningful to children that participated, as well as the impact of the recreation center intervention and youth leader components alone.

Conclusion

Cross-age mentors were able to successfully implement the BHCK nutrition education component in recreation centers, supporting the use of this population for intervention with youth, and the promise of the recreation center setting. Future work exploring sustainability is needed. Cooking demonstrations were successful in engaging youth and could be used to improve vegetable consumption. New strategies to engage youth in discussions should be developed and tested. Process evaluation results from this intervention can be used to inform best practices for future interventions.

Funding

The project described was supported by the Johns Hopkins Global Center on Childhood Obesity, Grant Number U54HD070725 from the National Institute of Child Health & Human Development (NICHD). The project is co-funded by the NICHD and the Office of Behavioral and Social Sciences Research (OBSSR). The content is solely the responsibility of the authors and does not necessarily represent the official views of the NICHD or OBSSR. The manuscript was supported by Sciences without Borders (SwB) program, CAPES.

Conflict of interest statement

None declared.