Boy Scout 5-a-Day Badge: Outcome results of a troop ... - PubAg - USDA

Preventive Medicine 49 (2009) 518–526

Contents lists available at ScienceDirect

Preventive Medicine j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / y p m e d

Boy Scout 5-a-Day Badge: Outcome results of a troop and Internet intervention Debbe Thompson a,⁎, Tom Baranowski a, Janice Baranowski a, Karen Cullen a, Russell Jago b, Kathy Watson a, Yan Liu a a b

USDA/ARS Children's Nutrition Research Center, Baylor College of Medicine, Department of Pediatrics, Houston, TX, USA University of Bristol, Department of Exercise, Nutrition and Health Sciences Centre for Sport, Exercise and Health Bristol, England, UK

a r t i c l e

i n f o

Available online 16 September 2009 Keywords: Fruit Vegetable Adolescents Internet Interactive multimedia Boy Scouts Intervention study Availability Self-efficacy

a b s t r a c t Objectives. The effects of a Boy Scout Five-A-Day Badge program on fruit juice (FJ) and low-fat vegetable (LV) consumption were evaluated using a two-condition (treatment, active-attention-placebo-control) group randomized trial, with three data collection periods (baseline, immediate post, 6-month post). Methods. Forty-two Boy Scout troops (n = 473, 10- to 14-year-old Scouts) in Houston, TX, were randomly assigned to condition. The 9-week program included approximately 30 min of weekly troop time, plus ∼ 25 min of weekly Internet programming. The intervention was delivered in two waves (Spring and Fall). Data were collected in 2003–2004, and analyses were completed in 2008. Main outcomes were FJ and LV consumption (validated food frequency questionnaire). FV self-efficacy, preferences, and home availability were also measured. Results. Significant increases in FJ consumption (p = .003), FJ home availability (p = .009), and LV selfefficacy (p = .004) were observed among the intervention group immediately following the intervention but were not maintained 6 months later. Conclusion. A Boy Scout troop-plus-Internet intervention promoting FJ and LV consumption resulted in short-term changes in FJ consumption among U.S. Boy Scouts. Future research should investigate ways to extend these results to LV and maintain the increases over time. Published by Elsevier Inc.

Introduction Fruit and vegetable (FV) consumption has been associated with decreased risk for chronic diseases (Van Duyn and Pivonka, 2000), such as certain cancers (World Cancer Research Fund and American Institute for Cancer Research, 2007) and cardiovascular disease (Bazzano et al., 2002). National dietary guidelines recommend 9- to 13-year-olds consume 7–10 daily FV servings based on calorie needs (USDHHS and USDA, 2005), but many do not (Guenther et al., 2006); b1% consume ≥10 servings a day, b4% consume ≥7 servings, and b20% consume ≥ 5 servings (Guenther et al., 2006). Because adolescent eating behavior is somewhat stable into young–adulthood (Lien et al., 2001), effective methods for enhancing youth FV consumption are needed to reduce adult risk of chronic disease. Along with home availability and preference (Jago et al., 2007; Rasmussen et al., 2006), gender appears to be a key determinant of FV consumption (Rasmussen et al., 2006), and its effects may be mediated by preference (Bere et al., 2008). A recent review of youth obesity treatment and prevention programs found few specifically designed for boys (Flynn et al., 2006). Tailored dietary change interventions are more effective than generic ones (Kroeze et al., ⁎ Corresponding author. E-mail address: [email protected] (D. Thompson). 0091-7435/$ – see front matter. Published by Elsevier Inc. doi:10.1016/j.ypmed.2009.09.010

2006), and gender-specific prevention-oriented interventions may be preferred over those that are non-gender specific (Flynn et al., 2006). Gender-specific interventions may be an effective method for promoting boys' FV consumption (Perry et al., 1998). Multi-component interventions appear to be more effective (Blanchette and Brug, 2005; French and Stables, 2003; Knai et al., 2006; Perry et al., 2004). Schools are a particularly attractive channel for promoting FV consumption (French and Stables, 2003; Knai et al., 2006; Perez-Rodrigo et al., 2005; Mangunkusumo et al., 2007), but they have limited time. Out-of-school programs are needed to complement school-based programs. A previous 9-week troopbased pilot intervention with urban Boy Scouts attained a 0.8 serving increase in FJV intake (Baranowski et al., 2002). Although the troopbased intervention was effective, it was time intensive, requiring ≥ 45 min of weekly troop time, thus attenuating troop leader enthusiasm. An Internet-based complement to an in-troop program would use less troop time. Youth Internet use is high with 87% of U.S. teenagers using the Internet, many daily, with slightly more than half reporting high-speed home Internet connections (Lenhart et al., 2005). The Internet is familiar, available, and accessible, thus providing an attractive adjunct to in-person programs. Internet-based genderspecific programs have increased FV consumption among girls (Thompson et al., 2008a). Therefore, a gender-specific Internet-

D. Thompson et al. / Preventive Medicine 49 (2009) 518–526

based program, combined with a troop-based program, may be an effective method for increasing FV consumption among boys. This paper reports the outcome of the “5-a-Day Achievement Badge Program” (5AD), an innovative nine-session troop-plusInternet FV intervention, conducted with Boy Scout troops in Houston, TX. The treatment group was hypothesized to have greater increases in FV consumption and psychosocial mediators than the control group. This study is innovative in that it targets Boy Scouts for primary prevention using a theory-based approach that combines personal contact, peers, and the Internet. Methods

519

(Baranowski et al., 2002) and approximately 10 Scouts per troop, the minimum number of participants needed was 445. Inclusionary criteria Troop inclusionary criterion was a high likelihood of Scouts having a home computer with Internet access. Scout inclusionary criteria included participating troop membership, a home computer with Internet access, and written consent/assent. Troop leaders received a $1,000 incentive for troop use following post-2 data collection. Recruitment Permission to conduct the study was obtained from Houston Boy Scouts of America. Presentations were then made to troop leaders; Scouts were recruited from troops expressing interest in the study.

Study design A two-condition (treatment, active-attention-placebo-control) group randomized trial with three data collection periods (baseline, immediate post, and 6-month post), conducted in two waves (i.e., cohorts) was used: Spring (March–baseline; 16 troops) and Fall (August–baseline; 26 troops) 2003. After baseline, troops were randomized to condition-within-wave using a coin toss by investigators. The CONSORT statement guided reporting of results (Moher et al., 2001). The Baylor College of Medicine institutional review board approved the study protocol (November 5, 2001). Sample size and power The study was powered to detect a difference of 0.5 FV servings a day. A conservative estimate for sample size was calculated (Cohen, 1988) for the two-way analysis of variance (ANOVA) with the change from baseline to post-1 representing the dependent variable and group and wave as factors (Cohen, 1988). With alpha of 0.01 and a moderate effect size (d = 0.50), 204 participants were needed to achieve 80% power. This number was doubled to include the change from baseline to post-2 and multiplied by a variance inflation factor (Donner and Klar, 2000) to account for the nesting of subjects within troops. Given an intra-class correlation of 0.01 associated with troop

Intervention The treatment group participated in an intervention to increase FV consumption, while the control group participated in a mirror-image intervention to increase physical activity (PA) (Jago et al., 2006). Each condition followed the same general structure, with approximately 55 min of weekly programming (∼ 30 min in-troop; ∼ 25 min Internet). The theoretical model driving the treatment group intervention is depicted in Fig. 1. Participants, data collectors, nor interventionists were blinded to condition. Treatment intervention In-troop activities Based on the Social Cognitive Theory (Bandura, 1986), trained staff (5AD troop leaders) led weekly in-troop sessions that taught functional knowledge and skills (Table 1) to enhance self-efficacy. They also made outcome-expectancy-related comments to enhance motivation to consume FV. In-troop preparation of simple recipes taught recipe preparation skills. Recipe tasting was included to enhance preference (Birch et al., 1987).

Fig. 1. Theoretical model.

520

Table 1 Components of the “5-a-Day“ Badge program, and how each was addressed in each week of the program. Troop meeting components Component

Web site

Week 1

Week 2

Week 3

Week 4

Week 5

Week 6

Week 7

Week 8

Week 9

Troop meeting knowledge activity

NA

Intro to Web, afterschool veggie snack

Identifying and eating fast food fruit and veggies Fast food alternatives to make at home

Troop meeting taste testing recipes

NA

Fresh veggies and pre-packaged low-fat dips (ranch dressings, onion dips, salsa) for at home snacks

Fruit smoothies: Tropical Dream & and Power-Paced Fruit Smoothie

Mixed up garden salad and pepper it up! - —veggie recipes

Twisted soda and pineapple blast drink

Outcome expectancy

NA

Veggies taste good prepared the “right” way

Fruit/100% juice at breakfast gives boys good energy to start the day

Score with Fruit and Power Pudding Dip: pre-packaged ready to eat pudding/fruit dips and fruit for snacks; and Super Bowl Sundae Fruit makes boys feel better about themselves than if they ate a non-FJV snack (less guilt)

Eating 5 A day for weekend meals Easy FJV meals and snacks to prepare when parents are not home All that salsa and calico bean salad

Badge award ceremony

Quick after-school snacks using pre-packaged veggies and dips

Asking for veggies for dinner Veggie side dishes boys can make for dinner

Eating 5 A day, every day

NA

Making easy after-school after School fruit snack Fast fruit and fruit dips for snacks

Choosing veggie for lunch

Troop meeting recipe prep skills

Fast and easy fruit/100% juice for breakfast Fast and easy ways to add fruit/100% juice to their breakfast

Eating veggies from the school cafeteria can be a cool thing to do

Eating more veggies for dinner is good for scouting (because it makes them more alert and healthy) and helping achieve their merit badges

FJV can taste good if they find a place that prepares them the way they like them

Settings

Weekday/outdoors— with a friend Decision making: choosing a favorite non-veggie snack vs. a favorite veggie snack

Weekday/ indoors—alone Asking skills: effective ways to ask parents to have favorite fruit/100% juice available/ accessible for breakfast Too lazy to prepare FJV snack to eat.

Weekday/outdoors— with friends Decision making: choosing a no-food prep fruit snack vs. usual afterschool after school snack

Weekend—with friends Problem solving: brainstorm solutions to eating veggies with peers at NSL/snack bar

Weekday/ indoors—alone Asking skills : ask parents to have veggies they prefer available and accessible for dinner

Weekday/ indoors—alone Decision making: choosing an FJV to eat with their meals at their favorite restaurant

Peer pressure that it is n't not cool to eat FJV at school from cafeteria

Parents do not have available nor prepare veggies boys like for dinner

Boys like to eat at fast food restaurants which have limited FJV choices, and when they do go to restaurants they do n't not choose FJV because the alternatives taste better

Parents are not always home to make meals for boys

Skills presented in comic book

Make your mark! Problem solving poll for comic book cliffhanger

Comic book cliffhangers = barriers encountered and solutions presented in “make your mark”

Not enough time to eat balanced or any breakfast

Modified favorite snack recipes

Wrap it my way!

NA

Making their own FJV meals is fun, tastes good and helps them meet BS merit badges

Making their own FJV meals is fun, tastes good and helps them get 5 A Day, every day

NA

Weekends— family Problem solving: planning FJV recipes to prepare for weekend meals when parents are not home Learning to choose smart snacks; portion sizing of snack choices

Weekday/ outdoors/or PA Decision making: choosing smarter snack foods

NA

NA

NA

NA


Internet components Comic book —real times 5

Easy veggie sack lunch recipes if do n't not like choices at school lunch Pick a pocket and That's a wrap!

Table 1 (continued) Troop meeting components Web site

Week 1

Week 2

Week 3

Week 4

Week 5

Goal setting

Challenge yourself!

Eat one more serving of veggies than you normally eat for after-school snack (M–F) 3× this week. Prepare one 5-a-Day Badge Web web recipe this week.

Ask for and eat one more fruit/ 100% juice than they normally eat for breakfast (M–F) 4× this week. Prepare one 5-a-Day Badge Web web recipe this week.

Eat one more fruit than you normally eat for a snack (M–F) 5× this week. Prepare one 5-a-Day Badge Web recipe this week.

Eat one more veggie than you normally eat for lunch (M–F) 5× this week. Prepare one 5-a-Day Badge Web web recipe this week.

Ask for and eat one more veggie than you normally eat for dinner (M–F) 5× this week. Prepare one 5-a-Day Badge Web web recipe this week.

Maintain at least 5 F and V a day.

Goal review— problem solving “solve it” Knowledge games

A-MAZE-ing combinations

Eat at their Eat 5-a-Day for Eat 5-a-Day for favorite fast both days this 7 day days this food restaurants weekend. week. Prepare and choose at Prepare one one 5-a-Day least 1 serving 5-a-Day Badge Badge Web web of veggies to eat Web web recipe this week. each time with recipe this their meal (M–F). weekend. Prepare one 5-a-Day Badge Web recipe this week. Each week the Scout is given the option to look at why they were unable to complete their goal. They go through a series of problem solving solutions to help them achieve their goal (and needed points for the badge) for next week.

Are you game?

What is a veggie? Drop catch game

What's a fruit? Drop catch game DROP CATCH GAME

What counts as a serving of fruit, veggie or juice? Trivia game

Identify FV choices at school. Drag and drop game DRAG & DROP GAME

Choosing a balanced dinner using the American Plate Guidelines. Make a meal drop game

NA

Web activity

Snack down Web recipes

Each week a different “Greek Dip”, “Onion Dip”, “Fiesta Salsa”, “Pancho Bean Dip”, “Spinaccoli Cheese Dip”

5-A-Day Badge Program Components (attached as a separate file).

Week 6

Identify FJV choices at fast food restaurants. Drag and drop game

Week 7

Identify ways to add veggies to weekend meals or any meal when they cook. Trivia game

set of recipes they could make at home were featured in this section. They also set goals to prepare these recipes for “Blazin' Raisin Sandwich recipes “Fruity Coleslaw”, “Burrito Fix”, “Bean Soup”, “Jack Stack” with “Pasta Primavera”, “Spread the Fruit” Tortillas”, “Power with veggies. “Hearty Rice”, “Mama Mia “Veggie Mac & —fruit spreads to Pudding Dip”, “Cool Sandwich “French Fry Pizza”, “Mexican add to pancakes, “Cool Fruit Kabobs”, Stackers” Fanatic”, Tostada”, “Veggie and Cheese”, “Vegetable waffles; “Fruit Freezers”, “Greek “Sunshine Burger” Stuffed Spuds“ “Strawberry “Fruit Pizza” Sandwich Carrots” Yogurt Breakfast Stackers“, Split”, “Microwave “Mexican Fruit Crisp” Sandwich Stackers“, “PNB Sandwich Stackers”, “Roll-N”

Week 8

Smart snacks— Identify ways to limit the amount of chips, candy, sodas, cookies, and other high-fat/ high-calorie snacks. Drop catch game “Peanut Butter Dip”, “Black Bean Salsa”, “Sticks-NStones Trail Mix, “Veggie Quesadillas”

Week 9

NA Stone Soup; Veggie Kabobs—grilled; Veggie Chili; Easy Veggie Pie.


Component

521

522


Scouts received a recipe booklet containing troop-prepared recipes to enhance knowledge of healthy recipes and to encourage home preparation. Recipe preparation skills were designed to enhance home availability. Together, self-efficacy and availability were emphasized to enhance consumption. Online activities Scouts were encouraged to log-on to the study Web site at least twice a week; first, to participate in the behavior change program and set goals, and second, to report goal attainment. Personal goal setting, goal reporting/ self-monitoring, and problem solving were included to enhance selfregulatory skills. Scout comic characters modeled how to use selfregulatory and asking skills to meet FJV goals. Comics ended with a “cliffhanger” (an unresolved problem likely keeping Scout players from eating FJV) to entice Scouts to return to the Web site. Comics were followed by a problem solving poll where players chose a solution to the cliffhanger. Scouts who did not meet their weekly goals participated in online and in-person problem solving. The in-person problem solving was conducted by the 5AD troop leader. Online mini-games refined functional knowledge (e.g., appropriate portions) (Thompson et al., 2008b). The 5AD troop leader monitored weekly log-ons. Points were awarded for goal attainment. At program end, Scouts who obtained sufficient points (≥ 70% of maximum) received 5AD badges. Measures Demographics Scouts' ethnicity and highest household educational attainment were obtained by parental self-report at time of consent. Anthropometrics Anthropometric data (height, weight, and body mass index) were collected. Method is detailed elsewhere (Jago et al., 2006). FV consumption F, juice (J), and V consumption were measured using a modified Food Frequency Questionnaire validated against 24-h dietary recalls (r = .92) with urban Boy Scouts (Cullen et al., 1999). It included four 100% J, 17 F, and 17 V. The response scale represented the non-averaged number of servings consumed in the previous 7 days. FJ and low-fat vegetables (LV) were analyzed separately. FJ consumption was computed by summing servings of the 4 J and 17 F. LV consumption was determined by removing three high-fat V (i.e., French fries, potato salad, other potatoes) and computing the servings of the remaining 14 lower-fat items (e.g., carrots, broccoli). Psychosocial variables FV self-efficacy was measured with a modified version of a questionnaire with acceptable internal consistency (.72–.87) and two-week test–retest reliability (.52–.67) (Domel et al., 1996). The modified scale contained 6 F/ 100% J and 7 V items, measured on a 5-point scale (“disagree a lot” to “agree a lot”). Items were summed, with higher scores indicating higher self-efficacy. FJV preferences were measured with a modified version of a scale demonstrating acceptable internal consistency (Cronbach's alpha = .67–.78) and test–retest reliability (.67–.72) (Domel et al., 1993). It assessed 4 J, 17 F, and 18 V using a 3-point scale (“I do not like this” to “I like this a lot”). Items were summed, with higher scores indicating higher preference. Home FJV availability was assessed using a scale having acceptable internal consistency (Cronbach's alpha = .77) and 12-week test–retest reliability (ICC = .50) (Cullen et al., 2004). The scale assessed 4 J, 17 F, and 17 V on a 2-point yes/no response scale. Scores for each subscale were summed, with higher scores indicating greater availability. Social desirability Social desirability of response was assessed using the 9-item “Lie Scale” from the Revised Children's Manifest Anxiety Scale (Reynolds and Paget, 1983). The scale has a 5-item response format (“never” to “always”). “Lie” score was determined by summing the responses. The instrument has shown good reliability and validity in children across a variety of ethnic groups (Dadds et al., 1998). Statistical analysis No data were available for troops, or Scouts-within-troops, not participating in the study. Missing data status was initially assessed by stratifying

participants into those who did and did not provide some data at all three collection periods (Fig. S1). Missing data were assessed at two levels: the collection period and each variable/scale with specific emphasis on FV consumption. Missing data status was defined as participants who did not provide FV consumption data at all collection periods. Differences in demographic characteristics by group and missing data status were examined. Baseline demographic characteristics were described using means, standard deviations, frequencies, and percentages. Two-way ANOVAs or chi-square tests for 3-way tables were used to test for differences in baseline characteristics among group and missing data status. The analyses yielded results for missing data status and group main effects and a missing data status × group interaction. A significant interaction effect would indicate the association between missing data status and the baseline characteristic was different between the groups. Appropriate followup tests were conducted to determine where differences occurred. The primary outcomes were analyzed using nested repeated measures within the Proc Mixed (Littlell et al., 1996) procedure in SAS (Version 9.1) to detect group differences over three time periods. Main effects for group (treatment, control), time (baseline, post-1, post-2), and wave (Spring, Fall) were treated as fixed effects, while the troop effect was treated as random. All two- and three-way interactions were included. Global effects identified as significant (p ≤ .05) or exhibiting a trend toward significance (.05 b p ≤ .10) were further examined to determine more specifically where the difference(s) occurred. Specifically, the regression parameters of the linear mixed model provided the a priori contrast estimates representing the change at post-1 (from baseline) and change at post-2 (from baseline). A priori contrasts were used to minimize inflated Type I error from multiple testing and the alpha was apportioned so that the change immediately following the intervention (post-1) received slightly higher priority with a 60/40% split of the overall family-wise error rate of 0.05 (Tabachnick and Fidell, 2001). Main effects were not interpreted in the presence of significant interactions. Separate analyses were performed for each consumption (FJ, LV) and psychosocial (availability, preference, self-efficacy) measure. All analyses included social desirability as a covariate to control for self-report response bias. To examine moderating effects, analyses were rerun with the addition of demographic and anthropometric characteristics and corresponding interactions. Due to skewness, both FJ and LV were log-transformed for analyses. Outliers were determined as cases exceeding the median-plus-3-times the interquartile range. The possibility of confounding between parallel experiments was assessed to determine if there was covariance in change in the two behaviors. Changes at post-1 and post-2 relative to baseline (Thompson et al., 2004) were calculated. Pearson correlations determined if changes in FJ and LV consumption covaried with changes in PA.

Results Four-hundred seventy-three 10- to 14-year-olds were recruited from 42 Houston-area troops. Participant characteristics are reported elsewhere (Jago et al., 2006). Parental education, group, and wave were significantly related (p = .007), with more participants from the control group in the Fall wave living in households in which at least one parent had a college or postgraduate degree (Jago et al., 2006). Average troop attendance was 81%; 76% (78% 5AD and 75% control) logged-on to the study Web site at least once a week. Participant-flow through the study phases was previously reported (Jago et al., 2006). Of the 867 Scouts in 42 troops, 736 Scouts were eligible. The participation rate among eligible Scouts was 64% (473/736). Non-participation was primarily due to routine nontroop attendance. The average participation rate among qualified Scouts, aggregated at the troop level, was 75% (11%–100%). Regardless of treatment or control group, only the highest level of household educational attainment was significantly higher in the group not missing data at the three time periods (p = .020). No significant group × missing data status interactions were observed, indicating no participation bias. The percentage of missing data was very small (0.4%–1.9%) (Fig. S1). Average baseline FJ and LV consumption did not significantly differ between groups. A significant group × time interaction was observed for FJ consumption (p = .003) (Table 2). Regression estimates representing


523

Table 2 Means (M), standard errors (SE), and significant results from fruit/juice consumption and psychosocial measures stratified by intervention group (5AD, control). Response variables

Consumption Baseline Post-1 Post-2 ICC = .04 Availability Baseline Post-1 Post-2 ICC = .01 Self-efficacy Baseline Post-1 Post-2 ICC = .01 Preferences Baseline Post-1 Post-2 ICC = .02

5AD

Control

Significant interaction

N

M (SE)

N

M (SE)

224 235 213

2.5 (0.1) 3.5 (0.1) 2.8 (0.1)

228 234 219

2.3 (0.1) 2.9 (0.1) 3.0 (0.1)

224 235 213

9.0 (0.4) 10.9 (0.4) 10.0 (0.4)

228 234 219

9.1 (0.3) 9.7 (0.3) 10.0 (0.3)

224 235 213

27.1 (0.5) 28.4 (0.5) 28.0 (0.5)

228 234 219

27.2 (0.4) 27.0 (0.4) 27.0 (0.5)

224 235 213

45.4 (0.7) 46.5 (0.6) 45.9 (0.7)

228 234 219

45.3 (0.6) 45.3 (0.6) 44.4 (0.6)

F(2,817) = 5.90, p = .003a a Significant difference in mean change Post-1 vs. baseline 5AD: 0.94 (0.0) Control: 0.56 (0.0) F(2,814) = 4.76, p = .009b b Significant difference in mean change Post-1 vs. baseline 5AD: 1.87 (0.3) Control: 0.58 (0.3) F(2,804) = 2.70, p = .068c c Significant difference in mean change Post-1 vs. baseline 5AD: 0.68 (0.5) Control: − 0.77 (0.5)

Note. Although analyses were performed on log-transformed consumption data [ln(Fruit/Juice+ 1)], estimates presented were back-transformed [exp(ln(Fruit/Juice + 1)− 1] to servings. Presented estimates for differences in consumption are calculated from the back-transformed estimates. Proportion of variation between troops (intra-class correlation; ICC). Study was conducted in Houston, TX. Data were collected 2003–2004, and analyses were completed in 2008. a Group × Time interaction; significant group difference between change at post-1 (post-1–baseline): t(817) = 2.20, p = .028. b Group × Time interaction; significant group difference between change at post-1 (post-1–baseline): t(814) = 2.39, p = .017. c Group × Time interaction; Significant group difference between change at post-1 (post-1–baseline): t(804) = 3.02, p = .003.

the change at post-1 yielded significant group differences in FJ consumption (p = .028). FJ consumption at post-1 increased by nearly one serving, with a mean increase (and standard error) of 0.94 (0.0) servings in the 5AD group compared to a mean increase of 0.56 (0.0) servings in the control group. The post-1 changes translated into a positive effect (effect size,0.4 servings) for the 5AD group. However, the improvement was not maintained. Although not significant, the changes

at post-2 from baseline translated into a positive effect (effect size,0.4 servings) for the control group. There was a significant group × time × wave interaction for LV consumption (p = .014) (Table 3). Regression estimates yielded a significant (p = .005) group difference in LV consumption between baseline and post-2 in the Spring wave, with the control group reporting a mean increase of 0.85 (0.1) servings and the 5AD reporting a slight

Table 3 Means (M), standard errors (se), and significant results from low-fat vegetable consumption and psychosocial measures stratified by intervention group (5AD, Control). Response variables

Consumption Baseline Post-1 Post-2 ICC = .04 Availability Baseline Post-1 Post-2 ICC = .00 Self-efficacy Baseline Post-1 Post-2 ICC = .00 Preference Baseline Post-1 Post-2 ICC = .01

5AD

Control

Significant interaction

N

M (SE)

N

M (SE)

224 230 213

2.0 (0.1) 2.5 (0.1) 2.1 (0.1)

228 234 219

1.6 (0.0) 1.9 (0.0) 2.2 (0.0)

224 230 213

8.2 (0.3) 9.4 (0.3) 8.9 (0.3)

228 234 219

8.2 (0.3) 8.7 (0.3) 9.0 (0.3)

224 230 213

22.8 (0.4) 24.6 (0.4) 24.0 (0.4)

228 234 209

22.6 (0.4) 22.5 (0.4) 23.1 (0.4)

224 230 213

32.5 (0.6) 33.4 (0.6) 33.1 (0.6)

228 234 219

31.3 (0.5) 32.0 (0.5) 31.9 (0.5)

F(2,811) = 4.27, p = .014a Significant difference in mean change Spring–post-2 vs. baseline 5AD: − 0.14 (0.1) Control: 0.85 (0.1)

F(2,802) = 5.59, p = .004b Significant difference in mean change Post-1 vs. baseline 5AD: 1.82 (0.4) Control: − 0.03 (0.4)

Note. Although analyses were performed on log-transformed consumption data [ln(low-fat vegetable + 1)], estimates presented were back-transformed [exp(ln(Low-fat vegetable + 1)) − 1] to servings. Presented estimates for differences in consumption are calculated from the back-transformed estimates. Percent of variation between troops (intra-class correlation; ICC). Study was conducted in Houston, TX. Data were collected 2003–2004, and analyses were completed in 2008. a Wave × Group × Time interaction; significant group difference in Spring between change at post-2 (post-2–baseline): t(811) = − 2.82, p = .005. b Group × Time interaction; significant group difference between change at post-1 (post-1–baseline): t(802) = 2.17, p = .030.

524


mean decrease of −0.14 (0.1) servings. Although not significant, the changes at post-1 translated to a positive effect (effect size, 0.2 servings) for the 5AD group whereas changes at post-2 translated into a positive effect (effect size, 0.5 servings) for the control group. A significant group × time interaction was observed for home FJ availability (p = .009), and a trend toward significance was observed for FJ self-efficacy (p = .068) (Table 2). The 5AD group reported a mean increase of 1.87 (0.3) available FV items, as compared to a mean increase of 0.58 (0.3) items in the control group. The 5AD group reported a slight increase in mean self-efficacy score of 0.68 (0.5) points, while the control reported a slight decrease of −0.77 (0.5) points. No other significant differences in FJ psychosocial measures were observed. There was a significant group × time interaction for LV self-efficacy (p = .004) (Table 3). The 5AD group had a mean increase in LF selfefficacy of 1.82 (0.4) points, while the control group remained nearly unchanged, with a mean decrease of −0.03 (0.4) points. Significant time main effects were observed for LV home availability (p b .001) and preferences (p = .029). Regardless of group or wave, LV home availability significantly increased from baseline by one item at post-1 and post-2 with a total mean score of 9.0 (0.2) LV items available in the home. LV preferences significantly (p = .010) increased from baseline to post-1 by one unit. No other significant differences were observed. The troop-associated intra-class correlations (ICC) for all of the primary and secondary outcomes were between .00 and .04 (Tables 2 and 3). Neither household education nor psychosocial characteristics moderated the intervention effect, and change in total FJV, FJ, or LV did not significantly covary with changes in PA. Discussion The 5AD badge program, with troop and Internet components, demonstrated a significant treatment effect at program end for FJ, but not LV, consumption. FJ intake in the treatment group decreased from post-1 to post-2. Paralleling the increased FJ consumption were increases in home FJ availability and self-efficacy in the 5AD group at post-1 which decreased at post-2, adding credence to this effect. Unexpectedly, LV consumption increased in the control group. However, LV self-efficacy increased in the 5AD group but remained nearly unchanged in the control group. Regardless of group or wave, statistically significant increases were observed at both post-1 and post-2 for home LV availability and at post-1 for LV preferences. While enhanced FJ and/or LV consumption has been promoted to youth using in-person (Baranowski et al., 2002, 2000; Reynolds et al., 2000) or Internet-based programs alone (Thompson et al., 2008c) or one followed by the other (Baranowski et al., 2003a), few other youthbased interventions have combined the two (Goreley et al., 2009; Mangunkusumo et al., 2007; Perez-Rodrigo et al., 2005). Because youth are heavy users of the Internet (Lenhart et al., 2005), Internetbased programs hold promise for broad dissemination to approximately 1.2 million Boy Scouts (The Boy Scouts of America) at a relatively low cost per participant, thereby warranting further study. Research is needed to identify ways in which to maintain these effects over time and to increase and maintain LV consumption. Other youth intervention studies reported statistically significant group difference in F but not V (Perry et al., 1998, 2004; te Velde et al., 2008), marginally significant increase in V but not F (Baranowski et al., 2002), and statistically significant increases in both F and V separately (Baranowski et al., 2003b; Reynolds et al., 2000). It is not clear why these outcomes differed across studies. Possible explanations are variations in theoretical frameworks, intervention components, measurement, intervention dose, analysis methods, populations, geographic regions, or delivery channels. Consistency in intervention design, measurement, evaluation and reporting would help reduce the difficulty in comparing the results across studies and should be a priority for the field (Baranowski et al., 2009).

It is possible Scouts viewed consuming FJ or LV as an “either/or” proposition, not realizing it was important to increase both. Alternatively, since youth prefer FJ over LV (Cullen et al., 1998; Domel et al., 1993), it may have been easier to meet FJ goals. Additionally, the control group received a PA intervention (Jago et al., 2006) which could have increased their desire to make healthier dietary choices. However, there was no covariability of changed diet and PA behaviors in this or an earlier study (Thompson et al., 2004). It is possible that outliers could have influenced LV consumption results. However, results were unchanged when outliers were removed and analyses rerun. Control group Scouts may have over-stated their LV consumption to please the interviewers, but the results were corrected for social desirability of response, thereby decreasing this possible response bias. Finally, the change in control group LV behavior occurred with no corresponding group- and wave-specific change in home availability, preference, or self-efficacy, which makes it even more of an unexplained effect. Perhaps greater emphasis should have been placed on environmental strategies, such as increasing home FV availability (Jago et al., 2007), home recipe preparation (Cullen et al., 2007), or on enhancing motivation to consume FV (Ryan and Deci, 2000). Although in-troop recipe preparation and tasting to enhance home availability and outcomeexpectancy statements by 5AD troop leaders were designed to enhance motivation to eat FV or to prepare recipes, these strategies were not enough to enhance consumption. Future interventions should investigate other ways in which to motivate Scouts to consume FV. Lastly, the intervention was designed around Social Cognitive Theory. Because behavior is influenced by multiple factors (Baranowski et al., 1997), a more complex framework may be needed (Thompson et al., 2008b). Other studies have demonstrated a reduction towards baseline in the target behavior following intervention removal (Nader et al., 1999; Skender et al., 1996; te Velde et al., 2008; Bere et al., 2007). It is possible the observed post-1 changes resulted from extrinsic motivation (Ryan and Deci, 2000). When removed, there may have been a decrease in motivation to continue the behavior. However, badge attainment is an integral component of Scout culture. Thus, this issue needs to be better understood and procedures developed to maintain extrinsically motivated change or to simultaneously develop internal motivation for the behavior. Designing badge interventions around basic needs that motivate behavior may enhance internal motivation, as well as behavioral maintenance; i.e., offering Scouts more choice (autonomy), placing greater emphasis on building FV consumption competence, or connecting goal setting to personal values (relatedness) (Ryan and Deci, 2000). Increases (5AD group) were observed in home FJ availability and self-efficacy at post-1 and post-2. The previous Boy Scout 5-a-Day program (Baranowski et al., 2002) also emphasized asking/negotiation skills and revealed a significant increase in home FJ availability, suggesting that equipping youth with skills to influence parents to increase home FJ availability may be a critical factor in enhancing consumption. Increasing functional knowledge and self-regulatory skills enhance self-efficacy (Bandura, 1986), both of which were heavily emphasized. Goal setting and problem solving were included to increase personal mastery through setting attainable goals that gradually increased in difficulty and developing plans to overcome barriers. The Web-based comics also provided observational learning. Procedures for maintaining behavior over time need to be identified (Rothman, 2000). Our findings raise a number of issues for future research, including how to (a) better integrate the in-troop and Internet activities into troop activities, (b) enhance dose, (c) involve parents, and (d) maintain behavior change. This study achieved acceptable log-on rates (76%), but the approach was labor intensive and likely infeasible in other environments. Less costly methods might include automatic e-mail reminders or a hotlink to the Web site (Thompson et al., 2008c).


Study limitations and strengths Strengths included a reasonably large sample, the innovative combination of in-troop and Internet activities, and using an activeplacebo-control group. The study has limited generalizability due to the predominantly Euro-American, middle-class sample. Diet was assessed with a food frequency questionnaire, which has recognized limitations (Tefft and Boniface, 2000; Thompson and Subar, 2008). Conclusion Results suggest a combined in-person and Internet-based program promoting FJ and LV may be an effective method of enhancing shortterm youth FJ consumption. Additional research is needed to identify methods to increase LV intake and to maintain changes over time. Conflict of interest statement The authors declare that there are no conflicts of interests.

Acknowledgments This research was largely funded by a grant from the American Cancer Society (TURSG-04). This work is also a publication of the United States Department of Agriculture (USDA/ARS) Children's Nutrition Research Center, Department of Pediatrics, Baylor College of Medicine, Houston, Texas, and had been funded in part with federal funds from the USDA/ARS under Cooperative Agreement No. 586250-6001. The contents of this publication do not necessarily reflect the views or policies of the USDA nor does mention of trade names, commercial products, or organizations imply endorsement from the U.S. government. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.ypmed.2009.09.010. References Bandura, A, 1986. Social Foundations of Thought and Action: A Social Cognitive Theory. Prentice Hall, Englewood Cliffs, NJ. Baranowski, T., Baranowski, J., Cullen, K.W., et al., 2003a. The Fun, Food and Fitness Project (FFFP): The Baylor GEMS pilot study. Ethn. Dis. 13, S1–S13. Baranowski, T., Baranowski, J., Cullen, K., et al., 2003b. “Squire's Quest!” dietary outcome evaluation of a multimedia game. Am. J. Prev. Med. 24, 52–61. Baranowski, T., Baranowski, J., Cullen, K.W., et al., 2002. 5 A Day Achievement Badge for African American Boy Scouts: pilot outcome results. Prev. Med. 34, 353–363. Baranowski, T., Cerin, E., Baranowski, J., 2009. Steps in the design, development and formative evaluation of obesity prevention-related behavior change trials. Int j. of behav. Nutr. Phys. Act. 6 (6) (21 January 2009). Baranowski, T., Davis, H., Hearn, M., et al., 2000. Gimme 5 fruit and vegetables for fun and health: outcome evaluation. Health Educ. Behav. 27, 96–111. Baranowski, T., Lin, L., Wetter, D., Resnicow, K., Hearn, M., 1997. Theory as mediating variables: why aren't community interventions working as desired? Ann. Epidemiol. S7, S89–S95. Bazzano, L.A., He, J., Ogden, L.G., et al., 2002. Fruit and vegetable intake and risk of cardiovascular disease in US adults: the first National Health and Nutrition Examination Survey Epidemiologic Follow-up Study. Am. J. Clin. Nutr. 76, 93–99. Bere, E., Brug, J., Klepp, K.I., 2008. Why do boys eat less fruit and vegetables than girls? Public Health Nutr. 11, 321–325. Bere, E., Veierod, M., Skare, O., Klepp, K., 2007. Free school fruit—sustained effect three years later. Int j. of behav. Nutr. Phys. Act. 4 (5) (19 February 2007). Birch, L.L., McPhee, L., Shoba, B.C., Pirok, E., Steinberg, L., 1987. What kind of exposure reduces children's food neophobia? Looking vs. tasting. Appetite 9, 171–178. Blanchette, L., Brug, J., 2005. Determinants of fruit and vegetable consumption among 6–12-year-old children and effective interventions to increase consumption. J. Hum. Nutr. Diet. 18, 431–443. Cohen, J., 1988. Statistical Power Analysis for the Behavioral Sciences 2nd ed. Academic Press, New York. Cullen, K., Watson, K., Zakeri, I., Baranowski, T., Baranowski, J., 2007. Achieving fruit, juice, and vegetable recipe preparation goals influences consumption by 4th grade students. Int j. of behav. Nutr. Phys. Act. 4 (28) (29 June 2007). Cullen, K.W., Baranowski, T., Baranowski, J., Hebert, D., de Moor, C., 1999. Pilot study of

525

the validity and reliability of brief fruit, juice and vegetable screeners among inner city African-American boys and 17 to 20 year old adults. J. Am. Coll. Nutr. 18, 442–450. Cullen, K.W., Baranowski, T., Baranowski, J., et al., 1998. “5 A Day” achievement badge for urban boy scouts: formative evaluation results. J. Cancer Educ. 13, 162–168. Cullen, K.W., Klesges, L.M., Sherwood, N.E., et al., 2004. Measurement characteristics of diet-related psychosocial questionnaires among African-American parents and their 8- to 10-year-old daughters: results from the girls' health enrichment multisite studies. Prev. Med. 38, S34–S42 Suppl. Dadds, M.R., Perrin, S., Yule, W., 1998. Social desirability and self-reported anxiety in children: an analysis of the RCMAS lie scale. J. Abnorm. Child Psychol. 26, 311–317. Domel, S., Baranowski, T., Davis, H., Leonard, S., Riley, P., Baranowski, J., 1993. Measuring fruit and vegetable preferences among 4th and 5th grade students. Prev. Med. 22, 866–879. Domel, S., Thompson, W., Davis, H., Baranowski, T., Leonard, S., Baranowski, J., 1996. Psychosocial predictors of fruit and vegetable consumption among elementary school children. Health Educ. Res. Theory Pract. 11, 299–308. Donner, A., Klar, N., 2000. Design and Analysis of Cluster Randomization Trials in Health Research. Oxford University Press, New York. Flynn, M.A., McNeil, D.A., Maloff, B., et al., 2006. Reducing obesity and related chronic disease risk in children and youth: a synthesis of evidence with ‘best practice’ recommendations. Obes. Rev. 7 (Suppl. 1), 7–66. French, S., Stables, G., 2003. Environmental interventions to promote vegetable and fruit consumption among youth in school settings. Prev. Med. 37, 593–610. Goreley, T., Nevill, M., Morris, J., Stensel, D., Nevill, A., 2009. Effect of a school-based intervention to promote healthy lifestyles in 7–11 year old children. Int. J. of Behav. Nutr. Phys. Act. 6 (5) (21 January 2009). Guenther, P.M., Dodd, K.W., Reedy, J., Krebs-Smith, S.M., 2006. Most Americans eat much less than recommended amounts of fruits and vegetables. J. Am. Diet. Assoc. 106, 1371–1379. Jago, R., Baranowski, T., Baranowski, J., et al., 2006. Fit For Life Boy Scout badge: outcome evaluation of a troop and Internet intervention. Prev. Med. 42, 181–187. Jago, R., Baranowski, T., Baranowski, J.C., 2007. Fruit and vegetable availability: a micro environmental mediating variable? Public Health Nutr. 10, 681–689. Knai, C., Pomerleau, J., Lock, K., McKee, M., 2006. Getting children to eat more fruit and vegetables: a systematic review. Prev. Med. 42, 85–95. Kroeze, W., Werkman, A., Brug, J., 2006. A systematic review of randomized trials on the effectiveness of computer-tailored education on physical activity and dietary behaviors. Annals Behav. Med. 31, 205–223. Lenhart, A., Madden, M., Hitlin, P., 2005. Pew Internet & American Life Project: Teens and Technology. Pew Internet, Washington, DC. available: http://www. pewinternet.org/pdfs/PIP_Teens_Tech_July2005web.pdf. Lien, N., Lytle, L.A., Klepp, K.I., 2001. Stability in consumption of fruit, vegetables, and sugary foods in a cohort from age 14 to age 21. Prev. Med. 33, 217–226. Littlell, R.C., Milliken, G.A., Stroup, W.W., Wolfinger, R.D., 1996. SAS System for Mixed Models. SAS Institute, NC. Mangunkusumo, R.T., Brug, J., de Koning, H.J., van der Lei, J., Raat, H., 2007. Schoolbased Internet-tailored fruit and vegetable education combined with brief counselling increases children's awareness of intake levels. Public Health Nutr. 10, 273–279. Moher, D., Schulz, K.F., Altman, D., 2001. The CONSORT statement: revised recommendations for improving the quality of reports of parallel-group randomized trials. JAMA 285, 1987–1991. Nader, P., Stone, E., Lytle, L., et al., 1999. Three-year maintenance of improved diet and physical activity: The CATCH cohort. Arch. Pediatr. Adolesc. Med 153, 695–704. Perez-Rodrigo, C., Wind, M., Hildonen, C., et al., 2005. The Pro Children intervention: applying the intervention mapping protocol to develop a school-based fruit and vegetable promotion programme. Ann. Nutr. Metab. 49, 267–277. Perry, C., Bishop, D., Taylor, G., et al., 2004. A randomized school trial of environmental strategies to encourage fruit and vegetable consumption among children. Health Educ. Behav. 31, 65–76. Perry, C., Bishop, D., Taylor, G., et al., 1998. Changing fruit and vegetable consumption among children: the 5-a-Day Power Plus program in St. Paul, Minnesota. Am. J. Public Health 88, 603–609. Rasmussen, M., Krolner, R., Klepp, K.I., et al., 2006. Determinants of fruit and vegetable consumption among children and adolescents: a review of the literature. Part I: quantitative studies. Int. J. Behav. Nutr. Phys. Act. 3 (22) (11 August 2006). Reynolds, C.R., Paget, K.D., 1983. National normative and reliability data for the Revised Children's Manifest Anxiety Scale. School Psychol. Rev. 12, 324–336. Reynolds, K., Franklin, F., Binkley, D., et al., 2000. Increasing the fruit and vegetable consumption of fourth-graders: results from the High 5 Project. Prev. Med. 30, 309–319. Rothman, A., 2000. Toward a theory-based analysis of behavioral maintenance. Health Psychol. 19, 64–69. Ryan, R., Deci, E., 2000. Self-determination theory and the facilitation of intrinsic motivation, social development, and well being. Am. Psychol. 55, 68–78. Skender, M.L., Goodrick, G.K., Del Junco, D.J., et al., 1996. Comparison of 2-year weight loss trends in behavioral treatments of obesity: diet, exercise, and combination interventions. J. Am. Diet. Assoc. 96, 342–346. Tabachnick, B.G., Fidell, L.S., 2001. Using Multivariate Statistics, 4th edition. Allyn and Bacon, Boston. te Velde, S., Brug, J., Wind, M., et al., 2008. Effects of a comprehensive fruit- and vegetable-promoting school-based intervention in three European countries: the Pro Children Study. Br. J. Nutr. 99, 893–903.

526


Tefft, M., Boniface, D., 2000. Estimating food and nutrient intake from food frequency questionnaire data by reference to a standard weighted diet survey. J. Human Nutr. Diet. 13, 219–224. The Boy Scouts of America, http://www.scouting.org/Media/AnnualReports/2007/ 11memtotals.aspx; downloaded 09/08/08. Thompson, D., Baranowski, T., Cullen, K., et al., 2008a. Food, Fun, & Fitness Internet Program for Girls: pilot evaluation of an eHealth Youth Obesity Prevention Program Examining Predictors of Obesity. Prev. Med. 47, 494–497. Thompson, D., Baranowski, T., Buday, R. et al., 2008b. Serious Video Games for Health: How Behavioral Science Guided the Development of a Serious Video Game. Simulation & Gaming. doi:10.1177/1046878108328087 (ahead of print). Thompson, D., Baranowski, T., Cullen, K., et al., 2008c. Food, Fun, & Fitness Internet Program for girls: influencing log-on rate. Health Educ. Res. 23, 228–237.

Thompson, D., Jago, R., Baranowski, T., et al., 2004. Covariability in diet and physical activity in African-American girls. Obes. Res. 12, 46S–54S. Thompson, F.E., Subar, A., 2008. Chapter 1. Dietary assessment methodology, In: Coulston, A., Boushey, C. (Eds.), Nutrition in the Treatment and Prevention of Disease, 2nd Edition. Academic Press, St. Louis, MO, pp. 3–39. USDHHS, USDA, 2005. Dietary Guidelines for Americans, 2005 6th Edition. US Government Printing Office, Washington, DC. January 2005. Available: http:// www.health.gov/dietaryguidelines/dga2005/document/pdf/DGA2005.pdf. Van Duyn, M.A., Pivonka, E., 2000. Overview of the health benefits of fruit and vegetable consumption for the dietetics professional: selected literature. J. Am. Diet. Assoc. 100, 1511–1521. World Cancer Research Fund and American Institute for Cancer Research, 2007. Food, Nutrition, Physical Activity, and the Prevention of Cancer: a Global Perspective. AICR, Washington, DC.