Increase Font Size Decrease Font Size View as PDF Print

What is the effect of nutrition education delivered via digital media and technology on children’s dietary intake-related behaviors?

Conclusion

Moderate evidence shows that nutrition education delivered via digital media/technology (computer- and internet-based programs) may be effective for improving dietary intake-related behaviors among children and adolescents.
 

Grade

Moderate

 

Evidence Summary Overview

Overall, the ability to draw strong conclusions as to the effect of nutrition education delivered via digital media/technology on children’s dietary intake-related behaviors is limited by the small number of relevant studies and large degree of variation in intervention design and study characteristics, and by the fact that digital media/technology is not only rapidly evolving, but its use as an educational tool in the field of nutrition is novel. In general, use of digital media/technology can ensure that interventions are delivered to a wide range of children in a variety of settings with high fidelity. In addition, using a theoretical framework that targets specific behaviors, and providing frequent doses of education over longer periods of time can help ensure that the intervention is successful. However, keeping students engaged is a challenge, and therefore, more research is needed to better understand the utility and effectiveness of these types of programs.

 

Description of the Evidence

The literature search for studies that tested the effects of nutrition education delivered to children and adolescents using digital media and/or technology identified 1,850 articles, 83 of which were selected for review (Fig. 4-A.1). Of these 83 articles, 10 were selected for inclusion. An additional 14 articles were identified via hand search. Therefore, a total of 24 articles were included in this systematic review. A detailed description of literature search results, including the databases searched and the number of articles identified using each database, articles identified using hand search, a list of citations for all included articles, and a table that lists excluded studies with rationale for exclusion can be found in the sort list. 
 
Click Figure 4-A.1. to see Flow chart of literature search results for studies examining the effects of nutrition education delivered to children and adolescents using digital media and/or technology.
 

Kindergarten Through Grade Five

This evidence review includes 11 studies that examined the effects of nutrition education delivered via digital media/technology on dietary intake-related behaviors in children in kindergarten through grade five. Ten studies were randomized controlled trials (RCT) [Bannon, 2006 (neutral quality); Baranowski, 2003a (positive quality); Baranowski, 2003b (neutral quality); Cullen, 2005 (positive Quality); Haire-Joshu, 2010 (neutral quality); Horne, 2004 (positive quality); Mangunkusumo, 2007 (positive quality); Pempek, 2009 (neutral quality); Thompson, 2008 (neutral quality); Turnin, 2001 (positive quality)] and one study was a non-randomized controlled trial [Moore, 2009 (neutral quality)]. Five studies received a positive quality rating, and six studies received a neutral quality rating. Eight of the studies were conducted in the United States (US), one in the United Kingdom, one in the Netherlands, and one in France. Sample sizes of these studies ranged from 30 to 1,876 (four studies had <100 subjects, three studies had 100 to 500 subjects, one study had 500 to 1,000 subjects, and three studies had >1,000 subjects). Nine of the studies included both boys and girls, while two studies included only girls. Subjects in these studies ranged in age from five to 12 years. Four studies included only African American subjects, and seven studies included a predominantly white or mixed ethnicity subject population. Few of the studies provided information regarding the socio-economic status of subjects.
 
The digital media/technology programs used were either computer- or internet-based. Seven studies used a multimedia computer game, three used an interactive website that provided nutrition information, goal-setting opportunities, and individually-tailored feedback, and two used educational videos.
 
All eleven studies targeted children’s dietary intake, with a primary focus on behaviors that would improve compliance with the Dietary Guidelines for Americans. While most of the studies aimed to increase fruit and vegetable intake, some also focused on increasing the following: 100% fruit juice, fiber, water, protein, dairy products, breakfast, and snacks. Several studies also aimed to decrease the following: Fat, calories, soft drinks, fruit drinks (not 100%), sodium, snacks, sweets, and sugar.
 
Overall, nine of the 11 studies found that nutrition education delivered via digital media/technology led to improved dietary intake-related behaviors (Bannon, 2006; Baranowski, 2003a; Cullen, 2005; Haire-Joshu, 2010; Horne, 2004; Moore, 2009; Pempek, 2009; Thompson, 2008; Turnin, 2001). Of these, six studies found that nutrition education delivered via digital media/technology was more effective than the no-intervention control (Baranowski, 2003a; Cullen, 2005; Horne, 2004; Thompson, 2008) or a comparison intervention with another type of nutrition education method (Pempek, 2009; Turnin, 2001), while two studies found that two different strategies for delivering nutrition education using digital media/technology were both effective (Bannon, 2006; Moore, 2009). Two of the twelve studies found no significant differences in dietary intake following nutrition education delivered via digital media/technology compared to a control or comparison intervention (Baranowski, 2003b; Mangunkusumo, 2007). However, Baranowski et al. (2003b) did report “clinically significant” changes in dietary intake following a digital intervention, but the differences were not statistically significant, and it is likely the study was not adequately powered. In addition, Mangunkusumo et al. (2007) measured outcomes three months after the completion of the intervention, and may have missed any short-term changes in dietary intake that occurred immediately subsequent to the intervention.
 

Table 4-A.1. Summary of the results from studies examining the effects of nutrition education delivered to children in kindergarten–grade 5 using digital media and/or technology
Study
Outcomes
Type of Digital Media/Technology
Bannon, 2006
N = 50
(+) The nutrition message videos led more children to choose apples as a snack (56%) than those who saw the control condition (33%; p<0.01).
(Ø) Snack choice did not differ between the gain-framed compared to the loss-framed videos
Video with either gain-framed (i.e., stated benefits of doing the behavior) or loss-framed (i.e., stated benefits that will be missed if behavior is not done) message compared to non-nutrition video
 
Baranowski, 2003a
N = 1,489
(+) Intervention compared to control → increased fruit, 100% juice, and vegetable intake (1 serving/d, mainly from f/v; p=0.002)
 Computer game (Squire's Quest) vs. no intervention control
Baranowski, 2003b
N = 35
Ø  Intervention compared to comparison→ decreased calories (-231 kcal), percent calories from fat (-1.6%), sweetened beverages (-0.8 servings/d); increased water (1.4 servings/d), increased fruit/ 100% juice/vegetables intake (1.2 servings/d) (NS)
Interactive, tailored website vs. basic website
 
Cullen, 2005
N = 1,578
(+) Intervention compared to control → increased fruit (+0.26 servings; p<0.001) and 100% fruit juice (+0.06 servings, p<0.05) at snacks and regular vegetables at lunch (+0.16 servings; p<0.01)
 Computer game (Squire's Quest) vs. no intervention control
Haire-Joshu, 2010
N = 451
(+) Overweight/Obese children: Intervention compared to control → decreased calories from high fat foods (p<0.05)
Computer storybooks vs. usual care
Horne PJ, 2004
N = 435
(+) Intervention compared to control →lunchtime fruit and vegetable intake was higher at intervention and follow-up than baseline (P<0.001), while snack intake was higher at intervention than baseline (P<0.001). There were also significant increases in fruit/veg intake at home (P<0.05).
Videos vs. No intervention control
Mangunkusumo, 2007
N = 486
Ø Intervention compared to control → no difference in fruit and vegetable intake (did not report change in intake from pre- to post-intervention)
Interactive, tailored website vs. no intervention control
Moore, 2009
N = 126
(+)/Ø Intervention compared to comparison → all improved nutrition self-care practices (p<0.05) with no difference between groups
Website/game (MyPyramid Blast-Off): small group vs. larger group
Pempek, 2009
N = 30
(+) Healthier advergame compared to the less healthy advergame → increased healthy snacks intake (1.4 servings vs. 0.20 servings; p=0.007); the control group fell in between (0.90 servings)
Advergames: healthier advergame vs. less healthy advergame
Thompson, 2008
N = 73
(+) Intervention compared to control → increased fruit, 100% juice, and vegetable intake (1 serving/d; p=0.002)
Interactive, tailored website vs. no intervention control (delayed group)
Turnin, 2001
N = 1,876
 
(+) Intervention compared to comparison → better dietary intake post-intervention: more carb (46.4% vs. 45%; p<0.05), less fat (37.1% vs. 37.6%; p<0.05), less protein (16.5% vs. 16.7%; p<0.05), less sugar (11.5% vs. 12.2%; p<0.001), more calcium (p<0.001), and more fiber (p<0.05)
Computer game vs. conventional teaching
 
(+): indicates a positive association.
(-): indicates an inverse association.
Ø: indicates a non-significant difference.

Older Than Grade Five

The literature search identified 13 studies that examined the effects of nutrition education delivered via digital media/technology on dietary intake-related behaviors in children in grade five and older. Nine studies were randomized controlled trials (RCT) [Baranowski, 2011 (positive quality); DeBar, 2006 (positive quality); DeBar, 2009 (neutral quality); Di Noia, 2008 (positive quality); Haerens, 2006 (positive quality); Haerens, 2007b (positive quality); Haerens, 2007a (neutral quality); Thompson, 2009 (positive quality); Williamson, 2005 (positive quality); and four studies were non-randomized controlled trials [Casazza, 2007 (neutral quality); Frenn, 2003 (neutral quality); Long, 2004 (neutral quality); Winett, 1999 (neutral quality)]. Seven studies received a positive quality rating, and six studies received a neutral quality rating. Ten of the studies were conducted in the US and three in Belgium. Sample size of these studies ranged from 50 to 2,840 (two studies had <100 subjects, eight studies had 100 to 500 subjects, one study had 500 to 1,000 subjects, and two studies had >1,000 subjects). Ten of the studies included both boys and girls, one study included only boys, and two studies included only girls. Subjects in these studies had a mean age ranging from 10 years to 16 years. Eight studies included predominantly white subjects, three studies had a mixed ethnicity subject population, and one study included only African American subjects. Few of the studies provided information regarding the socio-economic status of subjects.

The digital media/technology programs used were either computer- or internet-based. Six studies used a multimedia computer game, and five used an interactive internet website. One study combined the use of an internet website with videos.
 
All twelve studies targeted children’s dietary intake, with a primary focus on behaviors that would improve compliance with the Dietary Guidelines for Americans. While most of the studies aimed to increase fruit and vegetable intake, some also focused on increasing the following: Water, 100% fruit juice, fiber, and regular meal consumption. Several studies also aimed to decrease the following: Fat, calories, soft drinks, high fat snacks, high fat dairy products, and fast food consumption.
 
Overall, 12 of the 13 studies showed some improvement in dietary intake following a nutrition education intervention delivered using digital media/technology. Nine studies found that nutrition education delivered via digital media/technology led to improved dietary intake more effectively than a no-intervention control (DeBar, 2006; DeBar, 2009; Haerens, 2006; Haerens, 2007a; Haerens, 2007b) or a comparison intervention with another type of nutrition education method (Baranowski, 2011; Di Noia, 2008; Williamson, 2005; Winett, 1999). One study found that digital media/technology nutrition education improved dietary intake, but not more than control/ comparison (Casazza, 2007). Two studies found that nutrition education delivered via digital media/technology to girls led to improved dietary intake (lower fat intake), but not in boys (Frenn, 2003; Haerens, 2007a), and another found that nutrition education delivered via digital media/technology was effective in changing dietary intake immediately following the intervention, but that the behavior change did not persist six months post-intervention (Thompson, 2009). One study found no significant differences in dietary intake following nutrition education delivered via digital media/technology (Long, 2004). However, the studies varied widely in design and methodology, and it is possible that a number of factors that varied between the studies could have influenced the impact of nutrition education delivered via digital media/technology. Some of these factors are discussed in further detail below.

Click to see Table 4-A.2. Summary of the results from studies examining the effects of nutrition education delivered to children in grade 6–grade 12 using digital media and/or technology

 

Factors That May Impact the Efficacy of Nutrition Education Delivered Via Digital Media/Technology Across Age Groups

Dose of Nutrition Education

One variable that may impact the efficacy of nutrition education delivered via digital media/technology is the dose of education subjects receive. Dose can be characterized by the duration of the educational intervention, the total number of education sessions, or the frequency with which education sessions are held. The doses tested in these studies tended to be shorter in length, with little follow-up to examine long term effects. Therefore, more research is needed to determine whether longer term or more intensive interventions results in further improved outcomes, and whether the effects of these types of interventions can be sustained over longer periods of time, ultimately impacting health outcomes such as body weight. The doses of nutrition education delivered in the studies reviewed are described in more detail below: 
 

Kindergarten Through Grade Five

Among the studies that included subjects in kindergarten through grade five, the length of the interventions lasted from one day to up to 11 months, and the number of education sessions ranged from one to 10 [three studies were one day long (one session each), one study was 16 days long (16 sessions), one study was five weeks long (five sessions), two studies were five weeks long (10 sessions each), one study was eight weeks long (subject logged-in weekly), one study was three months long (six sessions), one study was 20 weeks long (subjects logged-in weekly), and one study was five months long (eight sessions)].  In terms of frequency, the studies ranged from one session to two sessions per week (three studies had a single session, one study had sessions every other week, two studies had sessions once a week, and five studies had two or more sessions per week). Overall, the studies that had longer intervention periods with a greater number and frequency of education sessions had a greater impact on dietary intake change in children (Baranowski, 2003a; Baranowski, 2003b; Cullen, 2005; Horne, 2004; Thompson, 2008; Turnin, 2001; Moore, 2009), compared to interventions that were shorter, and had fewer education sessions less frequently (Mangunkusumo, 2007). However, additional research is needed to determine whether longer term or more intensive interventions results in further improved outcomes, and whether the effects of these types of interventions can be sustained over longer periods of time, ultimately impacting health outcomes, such as body weight.
 
Log-on rate is another aspect of studies using digital media/technology that may impact not only dose and intensity, but also the effect of the intervention on nutrition-related behaviors. The degree to which free-living subjects log-on to a computer- or internet-based program affects the dose of nutrition education they receive. Only two of the studies included in this review measured log-on rates (Baranowski, 2003b; Thompson, 2008). Baranowski et al. (2003b) found that 48% of intervention girls logged-on weekly, and 25% of control girls logged-on monthly, while Thompson et al. (2008) found that 75% of subjects logged-on weekly. Neither study analyzed whether log-on rate affected outcomes; however, Thompson et al. (2008) showed that subjects significantly improved their dietary intake. While Baranowski et al. (2003b) reported some improvements in dietary intake behaviors, they did not reach statistical significance, likely due to a small sample size and lack of statistical power.
 
Older than Grade Five

In the studies that examined children older than grade five, intervention duration lasted from a single session to two years with varied session frequency and session length [one study was a single day (one session), one study had two sessions (length not reported), two studies were four weeks long (one study had four sessions; one study did not provide information on number of sessions), one study was nine weeks long (subjects logged-in weekly), three studies were 12 weeks long (four sessions, six sessions; one study did not provide information on number of sessions), one study was 16 weeks long (five sessions), two studies were nine months long (neither provided information on number of sessions), and two studies were 24 months long (subjects logged-in weekly)]. Overall, most of the studies in this review had a fairly long duration of intervention, with multiple, frequent sessions. Therefore, it is likely that these studies provided subjects an adequate dose of nutrition education to affect dietary intake, as thirteen of the fourteen studies showed significant improvements. More research is needed to determine whether longer term or more intensive interventions results in further improved outcomes, and whether there are the effects of these types of interventions can be sustained over longer periods of time, ultimately impacting health outcomes, such as body weight. 
 

Study Setting

Another factor that may impact the efficacy of nutrition education delivered via digital media/technology is the setting in which the education is provided. For example, it is possible the nutrition education delivered via digital media/technology may be less effective at home compared to in a classroom setting due to competing forms of technology and other forms of entertainment. The study setting used in the studies reviewed is described in more detail below. In this body of literature, there do not appear to be any emerging patterns related to the setting, indicating that nutrition education delivered via digital media/technology has the potential to be used effectively in a variety of settings. However, this set of studies did not systematically test the effects of setting on outcomes, and therefore additional research is needed to determine whether nutrition education delivered via digital media/technology is more effective when delivered in a particular setting. In addition, parental involvement was not addressed across the board in the body of literature reviewed, and therefore, determining whether parental involvement affects outcomes is also of interest. The study settings in the studies reviewed are described in more detail below:

Kindergarten Through Grade Five

In this body of literature, eight studies took place in a school (Bannon, 2006; Cullen, 2005; Baranowski, 2003a; Haire-Joshu, 2010; Horne, 2004; Mangunkusumo, 2007; Moore, 2009; Turnin, 2001), one study was delivered at a day camp (Baranowski, 2003b), one study was delivered in a research laboratory (Pempek, 2009), and one study was done at home (Thompson, 2008). 
 

Older Than Grade Five

In this body of literature, four studies took place in a school (Casazza, 2007; Frenn, 2003; Haerens, 2006; Haerens, 2007a; Haerens, 2007b; Long, 2004; Winett, 1999), one study was done in a youth service agency (Di Noia, 2008), one was delivered in a health care clinic (DeBar, 2006; DeBar, 2009), one was done during Boy Scout troop meetings and at home (Thompson, 2009), one was delivered in a research laboratory (Baranowski, 2011), and one was done in a research laboratory and at home (Williamson, 2005).

Theoretical Frameworks

Whether an intervention is developed using some theoretical basis may also play a role in how effective the intervention is. Using a theoretical framework when developing nutrition education interventions may improve outcomes, and previous reviews have indicated that the use of theoretical models to develop nutrition education interventions has become more common over time, and that use of theoretical models allow the researchers to better account for variables that may explain how nutrition education interventions impact children’s behavioral outcomes.1 The theoretical frameworks used in the studies reviewed are described in more detail below:
 

Kindergarten Through Grade Five

In this set of studies, eight of the 11 studies reported use of a theoretical framework, with four using the Social-Cognitive Theory (Baranowski, 2003a; Baranowski, 2003b; Cullen, 2005; Thompson, 2008), one used Behavioral-Change Theory (Mangunkusumo, 2007), one used the Self-Care Deficit Nursing Theory (Moore, 2009), one used Decision Making Theory (Bannon, 2006), and one used the Social-Ecological Model (Haire-Joshu, 2010). Three studies did not report using a theoretical framework to develop the nutrition education intervention (Horne, 2004; Pempek, 2009; Turnin, 2001). In particular, the studies that used a theoretical framework when developing the educational intervention, particularly the Social-Cognitive Theory or the Self-Care Deficit Nursing Theory, showed significant improvements in nutrition-related behaviors.
 

Older Than Grade Five

In this review, eight of the 13 studies used a theory when developing the intervention (Baranowski, 2011; Di Noia, 2008; Frenn, 2003; Haerens, 2006; Haerens, 2007a; Haerens, 2007b; Thompson, 2009; Winett, 1999), and five studies did not report any theoretical framework (Casazza, 2007; De Bar, 2006; De Bar, 2009; Long, 2004; Williamson, 2005). Of those studies that used a theory when designing the intervention, four used the Transtheoretical Model (Di Noia, 2008; Haerens, 2006; Haerens, 2007a; Haerens, 2007b), two used the Social-Cognitive Theory (Thompson, 2009; Winett, 1999), one used the Social Cognitive, Self-Determination, and Persuasion Theories (Baranowski, 2011), and one used both the Transtheoretical Model and the Health Promotion Model (Frenn, 2003). Of those studies that used a theory, seven out of eight showed that nutrition education improved children’s dietary intake. Of those studies that did not use a theory, three out of five showed that nutrition education improved children’s dietary intake. Therefore, it appears that in this set of studies, interventions that were explicitly grounded in theory may be more effective for improving children’s dietary intake behaviors.

Other Issues to Consider

There are several additional issues that are important to consider when reviewing and interpreting this body of literature. One study found that when children are presented with messages about less healthy foods, this can have a more negative impact on food choices than no message at all (Pempek, 2009). Therefore, when possible, it is important to consider the impact on children’s behavior of competing unhealthy messages that are widely available in the digital environment. In addition, the speed with which technological advances are occurring in our society makes it difficult to know what the impact of the digital media/technology tested in previously published studies would be in the current technological environment. Another consideration relates to subject characteristics, and whether these types of interventions are more effective in certain groups of children, based on race/ethnicity, socioeconomic status, or prior experience with digital media/technology. Also, none of the studies included in this review assessed whether the digital media/technology program used was engaging to students, and whether the degree to which students liked the program affected study outcomes.
 
Finally, it is important to consider potential unintended consequences that may not have been captured in the literature. Strong and consistent evidence in children shows that screen time is directly associated with increased overweight and obesity.  However, the strongest association is with television screen time, and based on the few studies that examined other types of screen time, there was no apparent association between body weight and video game or computer use (DGAC, 2010). Therefore, the impact of using digital media/technology on children’s overall screen time should be taken into consideration when planning interventions or programs.

Evidence Summary Paragraphs

Kindergarten Through Grade Five

Bannon, 2006 (neutral quality) conducted an RCT to test the effects of nutrition messages framing on snack choice in young children. Three classrooms were randomly assigned to watch one of the following 60-second videos: (a) a gain-framed nutrition message (i.e., stated benefits of doing the behavior), (b) a loss-framed message (i.e., stated benefits that would be missed is behavior is not done), or (c) a control scene. Following the video, children were offered a choice of either animal crackers or an apple for snack. The final sample included 50 children (46 % female, mean age=five years). Results showed that children who saw one of the nutrition message videos, were more likely to choose apples as a snack (56%) than those who saw the control condition (33%; P<0.01). There were no differences in snack choices between children who saw the gain-framed compared to the loss-framed videos. These results suggest that videos containing nutritional messages may have a positive influence on children's short-term food choices.
 
Baranowski, 2003a (positive quality) investigated the effects of a group-randomized intervention to increase fruit, juice, and vegetable (FJV) consumption among elementary school children in the US. The intervention involved the use of “Squire's Quest,” a 10-session, psychoeducational, computer game delivered over five weeks, with each session lasting about 25 minutes. The game was developed based on the Social Cognitive Theory. Children in the control group did not receive the intervention. Four days of dietary intake were assessed before and after the intervention using a multiple pass, 24-hour dietary intake interview directly with the children. The final sample included 1,489 subjects (689 boys, 763 girls; mean age=nine years). The results showed that children participating in the game increased their FJV consumption by 1.0 serving more than the children not receiving the program (P=0.002). The authors report that most of this increase was due to increased fruit and vegetable consumption, rather than increased 100% fruit juice intake. The authors concluded that psychoeducational multimedia games have the potential to substantially change dietary behavior. 
 
Baranowski, 2003b (neutral quality) conducted a group randomized intervention to test the effects of an internet-based intervention on dietary intake and body weight in girls. Girls in the intervention group attended a special four-week summer day camp, followed by a special eight-week home Internet intervention for the girls and their parents. Control group girls attended a different four-week summer day camp, followed by a monthly home Internet intervention, neither of which components included the Internet program enhancements. Dietary intake was measured using two 24-hour recalls taken at baseline and after the 12-week intervention. The final sample included 35 African American girls (mean age=eight years). At the end of the 12-week intervention, those in the intervention group decreased calorie intake (-231kcal) and percent calories from fat (-1.6%); increased consumption of water (1.4 servings per day) 100% juice/vegetables (1.2 servings per day); and lower consumption of sweetened beverages (-0.8 servings per day), though the differences were not significant. The authors concluded that summer day camp appears to offer promise for initiating health behavior change. 
 
Cullen, 2005 (positive quality) investigated the effects of a group-randomized intervention to increase fruit, juice, and vegetable (FJV) consumption among elementary school children in the US. The intervention involved the use of “Squire's Quest,” a 10-session, psychoeducational, computer game delivered over five weeks, with each session lasting about 25 minutes. The game was developed based on the Social Cognitive Theory. Children in the control group did not receive the intervention. Four days of dietary intake were assessed before and after the intervention using a multiple pass, 24-hour dietary intake interview directly with the children. The final sample included 1,578 subjects (736 boys, 803 girls; mean age=nine years). Results showed that significant increases were found for servings of fruit (+0.26 servings; P<0.001) and 100% fruit juice (+0.06 serving; P<0.05) at snacks (at home and in school, and regular vegetables (+0.16 servings; P<0.01) at lunch for intervention school children, compared with children in control condition schools. There were no differences at breakfast and dinner. The authors concluded that psychoeducational multimedia games have the potential to substantially change dietary behavior, particularly at eating occasions where children might have more control over food choices.
 
Haire-Joshu, 2010 (neutral quality) conducted an RCT in the US to test the effects of a multi-component diet and activity intervention on children’s dietary intake. Children in the intervention group received eight computer-tailored storybook sessions, parent action newsletters, and trained mentors. Children in the control group received the usual nutrition education provided in the school setting. A survey was used to assess child-related outcomes, including dietary intake. The final sample included 451 children (mean age 8.5 years; 37% African American; 49% female). For all subjects, fruit and vegetable consumption, total calorie intake, and percent energy from fat increased, with no differences between intervention and control groups. When results were stratified by weight status, overweight/obese children in the intervention group decreased consumption of calories from high fat foods, while normal weight children did not (P=0.059).
 
Horne, 2004 (positive quality) conducted an RCT to evaluate a peer-modeling and rewards-based intervention designed to increase children's fruit and vegetable consumption. Over 16 days, children watched six minute video adventures featuring heroic peers (the Food Dudes) who enjoy eating fruit and vegetables, and received small rewards for eating these foods themselves. After 16 days, there were no videos and the rewards became more intermittent. Fruit and vegetable consumption was measured (i) at lunchtime using a five-point observation scale; (ii) at snacktime using a weighed measure; (iii) at home using parental recall. The final sample included 435 children (ages four to 11 years). Results showed that compared to the control school, lunchtime fruit and vegetable consumption in the experimental school was substantially higher at intervention and follow-up than baseline (P<0.001), while snacktime consumption was higher at intervention than baseline (P<0.001). The lunchtime data showed particularly large increases among those who initially ate very few fruits and vegetables. There were also significant increases in fruit and vegetable consumption at home (P<0.05). The authors concluded that the intervention was effective in bringing about substantial increases in children's consumption of fruit and vegetables.
 
Mangunkusumo, 2007 (positive quality) conducted a cluster-RCT in the Netherlands to determine whether Internet-tailored advice for schoolchildren and Internet-supported dietary counseling impact fruit and vegetable intake. During school hours, all children completed Internet-administered questionnaires on fruit/vegetable intake and related determinants. Children in the intervention group received immediate online tailored nutrition feedback, and a nurse also received information via the Internet to support a five-minute counseling protocol to promote fruit/vegetable intake. The control group did not receive any of the intervention components. Children completed a similar post-test questionnaire three months after the first assessment. The final sample included 486 children (263 in the intervention, 223 in the control; mean age=ten years) from 30 classrooms (16 intervention, 14 control). There were no significant effects of the intervention on intake of fruits and vegetables. The authors concluded that while a tailored, Internet-based nutrition education tool can induce positive changes in nutrition-related knowledge, it did not translate into changes in dietary intake.
 
Moore, 2009 (neutral quality) conducted a quasi-experimental pilot study in the US to determine the effect of a nutrition education program, Color My Pyramid, on children's nutrition knowledge and nutrition status. The intervention program incorporates an online component using www.MyPyramid.gov (the Blast-Off game) and consisted of six classes taught over a three-month period. The content focused on general nutrition concepts, moderation and variety, portion sizes, exercise and activity, introduction to MyPyramid.gov for kids, and experiential learning with the Blast-Off Game. The intervention was then delivered in two different schools, with students in School 1 receiving a more didactic presentation on playing the Blast-Off Game, and students in School 2 using individual computers to evaluate their diets in small groups. Finally, a post-test was administered to re-evaluate children’s nutrition knowledge and behaviors. The final sample included 126 students (64 from School 1, and 62 from School 2; age nine to 11 years; 28% overweight, 43% obese; 93% African American). There was a significant increase in nutrition self-care practices from pretest to post-test among all students (P<0.05), but no differences between the intervention groups. The authors concluded that the computer-based Color My Pyramid program was effective in improving children’s nutrition self-care practices.
 
Pempek, 2009 (neutral quality) conducted an RCT to examine how advergames, which are online computer games developed to market a product, affect consumption of healthier and less healthy snacks by low-income African American children. Children played an advergame in which they were rewarded for having their computer character consume healthier or less healthy foods and beverages. Children were randomly assigned to one of the following three conditions: (1) the healthier advergame condition, (2) the less healthy advergame condition, or (3) the control condition. Children in the treatment conditions played a less healthy or a healthier version of an advergame two times before choosing and eating a snack and completing the experimental measures. Children in the control group chose and ate a snack before playing the game and completing the measures. The final sample included 30 African American children (15 girls, 15 boys; mean age=nine years). Children who played the healthier version of the advergame selected and ate significantly more healthy snacks (1.4±0.24 servings) than did those who played the less healthy version (0.20±0.24) (P=0.007), with the control group falling in between (0.90±0.24). Nine children (90%) in the healthy condition chose at least one healthy snack, whereas six children (60%) in the control group and one child (10%) in the less healthy group chose at least one healthy snack. The healthier and less healthy conditions differed significantly (P=0.001). The authors concluded that advergames affect children’s food intake, but that advergames promoting healthier foods and beverages can be used to encourage selection and consumption of healthier foods.
 
Thompson, 2008 (neutral quality) conducted an RCT in the US to determine the efficacy of an Internet-based program on dietary intake. The intervention was eight weeks long, and girls were randomized to receive immediate (weekly) or delayed (program end) incentives ($5.00). The Internet-based program was designed based on the Social Cognitive Theory, and emphasized fruit, 100% juice, vegetable (FJV), and water intake, as well as physical activity. Participants’ were given weekly goals to increase the target behaviors, and included focus on role modeling, problem solving, and goal setting/review. Dietary intake was assessed using a seven-item FFQ. The final sample included 73 African American girls (37 in the intervention, and 36 in the control group) who were eight to 10 years of age. Statistically significant pre-post improvement was observed in FJV consumption, with girls reporting an increase of one serving per day at the end of the intervention (P=0.002). The authors did not report the contributions of each individual food group measured to the total increase in FJV consumption. The authors concluded that this Internet-based program was feasible, and effective in promoting healthy eating.
 
Turnin, 2001 (positive quality) conducted a randomized trial in France to investigate the impact of nutrition computer games on children’s nutrition knowledge and dietary intake. All 16 schools in the same school district were randomized into two groups: Games group and control group, both receiving conventional nutritional teaching by their teachers. The children in the games group played computer games during the conventional nutritional teaching period (two hours per week for five weeks). At completion of the study, nutrition knowledge and three-day diet records were evaluated in both groups; however, these data were not assessed at the start of the study. The final sample included 1,876 children from 15 schools (mean age=nine years; 53% girls; 24% overweight, 11% obese). After the intervention, dietary intake differed significantly between the children in the games group compared to the control group; the game group consumed more carbohydrate (46.4±0.2% vs. 45.7±0.2%, P<0.05), less fat (37.1±0.1% vs. 37.6±0.2%, P<0.05), less protein (16.5±0.1% vs. 16.7±0.1%, P<0.05), less sugar (11.5±0.1% vs. 12.2±0.2%, P<0.001), more calcium (P<0.001) and more fiber (P<0.05). The authors concluded that children provided nutrition education via computer games compared to conventional teaching had slightly, but significantly better nutritional knowledge and dietary intake.
 
 

Older Than Grade Five

Baranowski, 2011 (positive quality) conducted an RCT in the US evaluating the effects of playing computer games on children’s diet. The intervention group played two computer video games: Diab and Nano. The control group played diet and physical activity knowledge-based games on popular websites. Three 24-hour dietary recalls were used to assess dietary intake immediately after playing each game and two months later. The final sample included 133 children (ages 10 to 12 years; 44% female; 40% white). Results showed the children who played the computer video games increased fruit and vegetable intake by 0.67 servings per day, compared to children who played the games on public websites (P<0.018).
 
Casazza, 2007 (neutral quality) conducted a nonrandomized group trial in the US to determine which health education delivery method (computer-based or traditional education) would elicit a greater behavior change. A total of three schools participated: A control school, a traditional education school (investigator taught classes via lecture and pamphlets) and a computer-based education school (investigator designed a nutrition education CD-ROM and students independently navigated the program). Over a period of 16 weeks, subjects participated in five, 45-minute intervention sessions. Measurements were taken at baseline and 11 weeks after baseline. Dietary intake data was collected using two 24-hour recalls and an FFQ. Students also completed a nutrition knowledge questionnaire. The final sample included 275 students (66% female, mean age=16 years). There was a significant decrease in total energy intake in both intervention groups (P<0.01), with no difference between the groups. Saturated fat intake decreased (P<0.01) and dairy intake increased (P<0.001) in the computer-based group, but differences between the groups were not significant. There were no differences between the groups in any other dietary intake measure. This article showed that nutrition education delivered using a computer, as well as that delivered using traditional education techniques, led to improvements in students’ dietary intake.
 
DeBar, 2006 (positive quality) and DeBar, 2009 (neutral quality) reported on results from an RCT to examine Web site use and behavioral outcomes in a multi-component intervention promoting healthy diet and exercise. The two-year intervention consisted of in-person and web components. Participants were encouraged to log-on to the study Web site at least once a week over the two-year intervention. Dietary intake data was measured using 24-hour recalls. In DeBar, 2006, the final sample included 209 girls (mean age=15 years). Participants in the intervention group reported significantly greater consumption of calcium in both study years (P<0.001), vitamin D in the first year (P<0.02), and fruits and vegetables (0.74 and 0.79 servings, respectively, P<0.01) in both years. There was no effect of the intervention on soda consumption. In DeBar, 2009, the final sample included 82 girls (mean age=15 years) who completed the Web-based intervention. Overall, Web site use was associated with increases in calcium intake (P<0.01). However, use of Web pages related to behavioral feedback and communications was not significantly associated with behavioral outcomes. The authors concluded that a multi-component intervention that involved use of a Web site effectively improved dietary intake in adolescent girls, and that use of a Web site may promote retention and engagement in target behaviors.
 
Di Noia, 2008 (positive quality) conducted a group RCT in the US to examine the efficacy of an after-school program intervention for increasing fruit and vegetable consumption among economically disadvantaged African-American adolescents. Youths were randomized into the intervention group that participated in a computer-based intervention program or to a control group that received regular programs offered at after-school program sites. The program, which was based on the Transtheoretical Model, provided youths with four 30-minute sessions of CD-ROM-mediated intervention content related to fruit and vegetable consumption, with a staging measure which classified users into precontemplation, contemplation/preparation or action/maintenance. Fruit and vegetable intake was measured by asking students how many servings per week they consumed. The final sample included 507 students (61% female, mean age=12 years). After adjustment for covariates, youths in the computer intervention arm had higher fruit and vegetable consumption than those in the control arm (3.25±1.5 servings vs. 2.46±1.39 servings, P<0.001), and more youths in the computer intervention arm progressed to later stages and maintained recommended intake levels (P<0.05). The authors concluded that youths who used the computer-based program increased their intake of fruits and vegetables 38%t (0.9 servings per day) more than youths who did not use the computer-based program.
 
Frenn, 2003 (neutral quality) conducted a non-randomized controlled trial in the US to examine the effects of a school-based internet and video intervention on diet. Classrooms were assigned to either the intervention group or a control group. The intervention consisted of four Internet sessions, five videos, a healthy snack session, and a gym class. Dietary intake was collected pre- and post-intervention using a validated dietary intake questionnaire. The final sample included 130 students (ages 12 to 15 years; 52% female; 45% African American). Results showed that following the intervention, girls in the intervention group reduced their fat intake compared to control (P<0.05). However, there were no significant differences in fat intake between the intervention and control groups.
 
Haerens, 2006 (positive quality) and Haerens, 2007a (neutral quality) reported on an RCT done in Belgium to evaluate the effects of a middle-school healthy eating promotion intervention combining environmental changes and computer-tailored feedback, with and without a parental involvement component. Fifteen schools with pupils in seventh and eighth grades were randomized to an intervention group with parental support, an intervention group without parental support, and a control group. The intervention schools implemented an intervention combining environmental changes with computer-tailored feedback based on the Transtheoretical Model. Target behaviors of the intervention were increasing fruit and water intake and reducing soft drink and fat intake. Children in the parental involvement group were also given a CD of the computer program to use at home. Dietary intake was measured with FFQs. Haerens, 2007a (neutral quality) included a final sample of 2,840 students (37% girls; mean age=13 years) who completed nine months of the intervention. In girls, fat intake and percentage of energy from fat decreased significantly in the intervention group with parental support, compared with the intervention alone group (P<0.05) and the control group (P<0.001), while in boys, there were no significant differences in fat intake or percentage of energy from fat as a result of the intervention. In addition, no effects of the intervention were found in boys or girls related to fruit, soft drinks and water consumption. Haerens, 2006 included a final sample of 2,287 students (38% girls; mean age=13 years) who completed the two-year intervention. In boys, there were no significant differences in fat intake or percentage of energy from fat as a result of the intervention. In girls, fat intake and percentage of energy from fat decreased significantly in the intervention groups compared to the control group (P<0.05), with no differences between the interventions groups with and without parental involvement. In addition, no effects of the intervention were found in boys or girls related to fruit, soft drinks and water consumption. The authors concluded that combining physical and social environmental changes with computer-tailored feedback in girls and their parents can result in lower fat intake.
 
Haerens, 2007b (positive quality) conducted an RCT to evaluate the acceptability, feasibility and effectiveness of a computer-tailored dietary fat intake education program for adolescents. A random sample of 10 schools, five with general and five with technical-vocational education programs, were selected to participate. In each of the 10 schools, two classes of 7th graders were randomly assigned to the intervention or control condition. Students were exposed once in class to a 50-minute theory-based computer-tailored dietary fat intake intervention. Questionnaires were completed one week before (FFQ for dietary fat intake + psychosocial determinants) and three months after (process evaluation + FFQ for dietary fat intake + psychosocial determinants) the intervention. The final sample included 304 subjects (70% female; mean age=13 years). Results showed that girls enrolled in technical-vocational schools significantly decreased fat intake (P<0.05), as did boys and girls undertaking general education who reported to have read the intervention messages (P<0.05).
 
Long, 2004 (neutral quality) conducted a group nonrandomized trial in the US to test the effects of a classroom and Web-based educational intervention on self-efficacy for healthy eating. Two schools were assigned to the intervention (five hours of Web-based instruction and 10 hours of classroom curriculum), or to the control (nutrition education embedded in the standard school curriculum during a one-month period, with exposure ranging from zero to three hours). The final sample included 121 adolescents (52% girls, mean age=13 years). Results showed that there were no differences between the schools in consumption of fat, fruits, or vegetables. The authors concluded that the Web-based program significantly improved adolescent’s self-efficacy for fruit and vegetable intake, but that this did not translate into a difference in actual consumption of fruits and vegetables. 
 
Thompson, 2009 (positive quality) conducted an RCT in the US to test the effects of a nutrition education intervention on fruit juice and low-fat vegetable intake in boys. The study population was derived from a group of Boy Scout troops. Forty-two Boy Scout troops were randomly assigned to one of two conditions. The intervention group participated in a nine-week program that included approximately 30 minutes of weekly troop time, plus approximately 25 minutes of weekly Internet programming. The control group participated in a mirror image intervention to increase physical activity. The intervention was designed using the Social Cognitive Theory, and scouts were encouraged to log-on to the internet site twice weekly to participate in a behavior change program and set goals, as well as to report goal attainment. The main outcomes were fruit juice and low-fat vegetable intake, which were assessed at baseline immediately following the intervention, and six months post-intervention using a modified, validated FFQ. The final sample included 473 boys (ages 10 to 14 years). Immediately following the intervention, subjects in the intervention significantly increased consumption of fruit and fruit juice (0.94 servings per day) compared to control subjects (0.56 servings per day; P<0.003). However, this difference was not maintained six months later. Also, at six-month post-intervention, subjects in the intervention group increased low-fat vegetable intake compared to the control group (~1 serving per day, P<0.05). 
 
Williamson, 2005 (positive quality) conducted an RCT in the US to assess the impact of an Internet-based intervention on weight loss in adolescent girls. Family pairs were randomized to either a behavioral group, which incorporated behavior modification techniques and email communication with a specialized case manager trained in weight management, or a control group, receiving basic information about nutrition and physical activity via different Web site. Dietary intake data was collected via a 24-hour recall and FFQ. The final sample included in 50 African American adolescent girls (mean age=13 years; mean BMI=36kg/m2). Girls in the behavioral intervention also reduced total energy intake (P<0.001), protein (P<0.05), and fat intake (P<0.05), while those in the control group only reduced fat intake (P<0.05). 
 
Winett, 1999 (neutral quality) conducted a non-randomized controlled trial in the US to investigate the effects of the Eat4Life Internet-based health behavior program on the nutrition and activity practices of adolescents girls. Eight 9th and 10th grade classes were randomized into two groups: The intervention group used the Eat4Life computer program in their health education classes, while the control group did not use the modules. The Eat4Life group used the computer program during class, and also received a sample of a food recommended in the module. The Eat4Life modules are based on the Social Cognitive Theory, and provide a brief assessment of the subject’s nutritional practices, followed by personalized feedback. Food intake was measured pre- and post-intervention using a 24-hour recall and FFQ. The final sample included 180 girls (103 in the intervention group, 77 in the control group; mean age=15 to 16 years). Across four cohorts of classes, girls using the Eat4Life modules reported that they made relatively consistent changes in all areas of nutrition except for reducing the consumption of high-fat dairy products. The Eat4Life program was effective in increasing consumption of regular meals, fruits and vegetables, and breads/cereals (all P<0.001), and decreasing consumption of regular soft drinks and fast foods (P<0.05). The authors concluded that a computer-based nutrition education program was effective in improving nutrition-related behaviors of adolescent girls.

1. Contento I. The Effectiveness of Nutrition Education and Implications for Nutrition Education Policy, Programs, and Research: A Review of Research. J. Nutr Ed Behav. 1995 Dec;27(6):277-420.



View table in new window
Author, Year,
Study Design,
Class,
Rating
Participants Methods Outcomes Strengths and Limitations
Bannon K, Schwartz MB, 2006 

Study Design: Randomized Crossover Trial

Class: A 

Rating: Neutral

N=50 (46% female).

Age: Five years.

Ethnicity/Race: White.

Attrition rate: 0%.

Location: United States.

 

Methods: Three classrooms randomly assigned to watch one of the following 60-second videos: 1) A gain-framed nutrition message, 2) a loss-framed message, or 3) a control scene.

Outcome measures: Following the video, children were offered a choice of either animal crackers or an apple for snack.

Duration/Intensity: One session.

Intervention Groups: Two of three classrooms randomly assigned to a 60's video with either 1) a gain-framed nutrition message or 2) a loss-framed message.

Comparison Group: One classroom assigned to watch one of the 60's videos with a control scene.

 

Children who saw one of the nutrition message videos were more likely to choose apples as a snack (56%), than those who saw the control condition (33%; P<0.01). No differences in snack choices between children who saw the gain-framed compared to the loss-framed videos.

 

Small sample size.

Short length of intervention makes it difficult to determine long-term effect.

Moderating variables were not accounted for.

Validity of the videos was not pre-tested.

 
Baranowski et al 2003 

Study Design: Group Randomized Controlled Trial

Class: A 

Rating: Positive

N=1,489 (749 intervention, 740 control; 689 boys, 763 girls).

Age: Nine years (eight to 12 years).

Ethnicity/Race: 17% African American, 44% Euro-American, 30% Hispanic, 7% other.

Attrition rate: 6%.

Location: United States.

 

Methods: Intervention tested effects of a multimedia game on diet.

Outcome Measures: Food intake assessed using multiple-pass, 24-hour recalls over four days before and after intervention.

Theory: Social Cognitive Theory.

Duration/Intensity: Ten 25-minute sessions over five weeks.

Intervention group: Used “Squire's Quest,”  a psycho-educational, multimedia game. Game attempts to ↑ intake of fruit, 100% juice and vegetables by associating fun with their consumption.

Comparison group: Children in control group did not receive intervention.

 

Children participating in Squire's Quest ↑ their fruit, 100% juice and vegetable consumption by 1.0 servings more than the children not receiving the program (P=0.002). Authors reported that most of the ↑ was due to ↑ fruit and vegetable intake, rather than 100% fruit juice.

 

Limited detail provided regarding the comparison group.

Self-reported dietary intake. 

Unknown how long the behavior Δ would be sustained after the end of the program.

 
Baranowski et al 2003 PMID 12713209 

Study Design: Randomized Controlled Trial

Class: A 

Rating: Neutral

N=35 African-American girls (19 intervention, 16 control).

Age: Eight years. 

SES: ~55% had household income of >$40,000.

Attrition rate: 0%.

Location: United States.

 

Methods: Two-arm parallel group randomized controlled pilot study, the Girls' Health Enrichment Multisite Studies (GEMS) Fun, Food, and Fitness Project, conducted to ↑ fruit, 100% juice, vegetable and water intake, ↑ physical activity, and to prevent obesity in girls.

Theory: Social Cognitive Theory.

Duration/Intensity: 12 weeks (four-week summer camp and eight-week home intervention). Control girls asked to log on once a month and intervention girls asked to log on weekly (48% of intervention girls logged on weekly, 25% of control girls logged on monthly).

Outcome measures: Diet assessed using two 24-hour recalls taken at baseline and after the 12-week intervention.

Intervention group: Girls in intervention group attended a four-week summer day camp, followed by an eight-week home Internet intervention for the girls and their parents.

Comparison group: Control group girls attended a different four-week summer day camp, followed by a monthly home Internet intervention, neither of which components included the GEMS enhancements.

 

At the end of the 12-week intervention, those in intervention group ↓ in calorie intake (-231kcal), % calories from fat (-1.6%), ↑ consumption of water (1.4 servings per day), fruit/100% juice/vegetables intake (1.2 servings per day) and ↓ consumption of sweetened beverages (-0.8 servings per day), though the differences were NS.

 

Included a very small sample of only girls may have inadequately powered results.

Web site log-on rates were ↓, especially in the control groups.

Self-reported dietary intake.

Significant differences in BMI existed between intervention and control groups at baseline.

Strengths included use of a theoretical framework, randomization, use of a control group and extensive process evaluation.

 
Baranowski T, Baranowski J et al, 2011 

Study Design: Randomized Controlled Trial

Class: A 

Rating: Positive

N=133 (44% female).

Age: 10 to 12 years.

Ethnicity/Race: 40% white, 24% African American, 28% Hispanic, 8% other.

Location: United States.

 

Methods: Intervention tested effects of computer games on diet.

Outcome measures: Food intake assessed using three 24-hour diet recalls.

Theory: Social Cognitive, Self-Determination, Persuasion theories.

Duration/Intensity: Two sessions.

Intervention group: Played two computer games (Diab, Nano).

Comparison Group: Played diet and physical activity knowledge-based games on popular Web sites.

 

Children in the intervention significantly ↑ servings of fruit and vegetables compared to control (0.67 servings per day; P<0.018).

 

Self-reported dietary intake.

Unknown how long the behavior Δ would be sustained after end of program.

Small sample size.

Baseline differences between groups may have obscured results.

Unclear whether the games would be as effective without incentives.

Mediators and moderators not assessed.

 
Casazza and Ciccazzo 2007; PMID 17174854 

Study Design: Nonrandomized Group Trial

Class: C 

Rating: Neutral

N=275 (66% female).

Ethnicity/Race: 52% non-Hispanic black. 

Age: 16 years.

SES: 51% traditional, 65% computer and 71% control were eligible for free school lunch.

Attrition rate: 12%.

Location: United States.

 

Methods: Non-randomized group trial to determine which health education delivery method (computer-based or traditional) would elicit a greater behavior Δ.

Duration/Intensity: Over a period of 16 weeks, subjects had five 45-minute intervention sessions.

Outcome measures: Measurements taken at baseline and 11 weeks after baseline.

Dietary intake data were collected using two 24-hour recalls and an FFQ (both validated) taken one week apart, at pre- (weeks one to three) and post- (weeks 13 to 16) study.

Intervention groups: A total of three schools participated:

1) Control: No intervention

2) Traditional education: Received education via lecture and pamphlets

3) Computer-based education: Received education via CD-ROM that each student independently navigated.

 

Computer and traditional groups had significant ↓ in total energy intake (P<0.01), with no difference between the groups.

NS differences between the groups in any other dietary intake measure (fiber, fruits and vegetables).

 

Subjects were a convenience sample, which limits generalizability.

Study duration was limited due to the school schedule.

 
Cullen KW, Watson K et al, 2005 

Study Design: Randomized Controlled Trial

Class: A 

Rating: Positive

N=1,578 (785 intervention, 793 control; 736 boys, 803 girls).

Age: Nine years (eight to 12 years).

Ethnicity/Race: 17% African American, 44% Euro-American, 30% Hispanic, 7% other.

Location: United States.

 

Methods: Intervention tested effects of a multimedia game on diet.

Outcome measures: Food intake assessed using multiple-pass, 24-hour recalls over four days before and after intervention.

Theory: Social Cognitive Theory.

Duration/Intensity: Ten 25-minute sessions over five weeks.

Intervention group: Used “Squire's Quest,”  a psycho-educational, multimedia game. The game attempts to ↑ intake of fruit, 100% juice and vegetables by associating fun with their consumption.

Comparison group: Children in control group did not receive intervention.

 

Children in the intervention significantly ↑ servings of fruit (+0.26 servings; P<0.001) and 100% fruit juice (+0.06 servings; P<0.05) at snacks (at home and in school), and regular vegetables (+0.16 servings; P<0.01) at lunch for intervention school children, compared with children in control condition schools.

 

Limited detail provided regarding the comparison group.

Self-reported dietary intake.

Unknown how long the behavior Δ would be sustained after end of the program.

 
DeBar L, Dickerson J et al, 2009 

Study Design: Randomized Controlled Trial

Class: A 

Rating: Neutral

N=209.

Age: 15 years.

Ethnicity/Race: White.

Attrition rate: 9%.

Location: United States.

 

Methods: Two-year intervention consisted of in-person and Web components.

Outcome measures: Dietary intake data measured using 24-hour recalls.

Duration/Intensity: Two years.

Intervention group: Two-year intervention consisted of in-person and Web components. Participants encouraged to log on to the study Web site at least once a week over the two-year intervention.

Comparison group: No intervention control.

 

Participants in the intervention group reported significantly ↑ consumption of calcium in both study years (P<0.001), vitamin D in the first year (P<0.02) and fruits and vegetables (0.74 and 0.79 servings respectively, P<0.01) in both years.

No effect of the intervention on soda consumption.

 

Demographics limit the generalizability of the study.

Intervention may not be easily replication in other health care settings.

 
DeBar LL, Ritenbaugh C et al, 2006 

Study Design: Randomized Controlled Trial

Class: A 

Rating: Positive

N=82.

Age: 15 years.

Ethnicity/Race: 80.5% white.

Location: United States.

 

Methods: Two-year intervention consisted of in-person and Web components.

Outcome measures: Dietary intake data measured using 24- hour recalls.

Duration/Intensity: Two years.

Intervention group: A two-year intervention consisting of in-person and Web components. Participants encouraged to log on to the study Web site at least once a week over the two-year intervention.

Comparison group: No intervention control.

 

Overall Web site use was associated with ↑ in calcium intake (P<0.01). However, use of Web pages related to behavioral feedback and communications was NS associated with behavioral outcomes.

 

Study not powered to examine the Web components independently.

Demographics limit the generalizability of the study.

 
Di Noia, Contento and Prochaska 2008; PMID 18517094 

Study Design: Randomized Controlled Trial

Class: A 

Rating: Positive

N=507 (61% female).

Ethnicity/Race: 15% Hispanic.

Age: 12 years.

Attrition rate: 8%.

Location: United States.

 

Methods: Group RCT to examine the efficacy of an after-school program intervention based on increasing fruit and vegetable consumption among economically disadvantaged African-American adolescents.

Theory: Transtheoretical model.

Duration/Intensity: Four 30-minute sessions.

Outcome measures: Fruit and vegetable intake measured by asking students how many servings per week they consumed, using a validated tool from the 5-a-Day Initiative.

Intervention group: Youths participated in a computer-based, CD-ROM-mediated intervention with content related to fruit and vegetable consumption that included a staging measure to classified users into stages of Δ.

Comparison group: Received regular programs offered at after-school program sites.

 

After adjustment for covariates, youths in the computer intervention arm had ↑ fruit and vegetable consumption than those in the control arm (3.25±1.5 servings vs. 2.46±1.39 servings, P<0.001) and more youths in the computer intervention arm progressed to later stages and maintained recommended intake levels (P<0.05).

 

Use of a self-selected sample.

Quasi-experimental design limits internal validity.

More long-term studies are needed.

 
Frenn M, Malin S et al, 2003 

Study Design: Non-Randomized Controlled Trial

Class: C 

Rating: Neutral

N=130 (52% girls).

Age: 12 to 15 years.

SES: 48% were of “lower SES.”

Location: United States.

 

Methods: A quasi-experimental study conducted to evaluate the effects of a school-based Internet/video intervention on diet.

Theory: Transtheoretical and Health Promotion Models.

Duration/Intensity: One semester, six sessions.

Outcome measures: Dietary intake measured with a validated questionnaire

Intervention group: Four Internet sessions, five videos, one healthy snack session, one gym class

Comparison Group: No intervention control.

 

Girls in the intervention group ↓ their fat intake compared to control (P<0.05).

No differences between the intervention and control groups in boys.

 

Limited number of subjects.

Low test-retest of the study instruments used to measure diet.

 
Haerens L, De Bourdeaudhuij I et al, 2007 

Study Design: Randomized Controlled Trial

Class: A 

Rating: Neutral

N = 2,840 (37% girls).

Age: 13 years.

SES: 68% were of “lower SES.”

Location: Belgium.

 

Methods: Group RCT to evaluate the effects of a middle-school healthy eating promotion intervention combining environmental Δs and computer-tailored feedback, with/without a parental involvement component.

Theory: Transtheoretical Model.

Duration/Intensity: Nine months.

Outcome Measures: Dietary intake measured with a validated FFQ.

Intervention Groups: Schools randomized to one of three groups: 1) Intervention group with parental support, 2) intervention group without parental support, 3) control group.

Intervention schools implemented a nine-month intervention combining environmental Δs with computer-tailored feedback.Target behaviors of the intervention were ↑ fruit and water intake and ↓ soft drink and fat intake. Children in parental involvement group were also given a CD of the computer program to use at home.

 

In girls, fat intake and % of energy from fat ↓ significantly in the intervention group with parental support, compared with the intervention alone group (P<0.05) and the control group (P<0.001). While in boys, there were NS differences in fat intake or % of energy from fat as a result of the intervention.

No effects of the intervention were found related to fruit, soft drinks and water consumption.

 

Self-reported measure of diet.

Unable to distinguish results of each component of the intervention; in particular, it is not possible to draw a conclusion on the efficacy of digital media.

 
Haerens L, Deforche B et al, 2006 

Study Design: Randomized Controlled Trial

Class: A 

Rating: Positive

N=2,287 (38% girls).

Age: 13 years.

SES: 68% were of “lower SES.”

Attrition rate: 25% at two years.

Location: Belgium.

 

Methods: Group RCT conducted to evaluate the effects of a middle-school healthy eating promotion intervention combining environmental Δs and computer-tailored feedback, with and without a parental involvement component.

Theory: Transtheoretical Model.

Duration/Intensity: Nine months.

Outcome measures: Dietary intake measured with a validated FFQ.

Intervention Groups: Schools randomized to one of three groups: 1) Intervention group with parental support, 2) intervention group without parental support, 3) control group.

Intervention schools implemented a nine-month intervention combining environmental Δs with computer-tailored feedback. Target behaviors of the intervention were ↑ fruit and water intake and ↓ soft drink and fat intake. Children in the parental involvement group were also given a CD of the computer program to use at home.

 

In boys, NS differences in fat intake or % of energy from fat as a result of intervention.

In girls, fat intake and % of energy from fat ↓ significantly in the intervention groups, compared to the control group (P<0.05), with no differences between the interventions groups with and without parental involvement.

No effects of the intervention were found in boys or girls related to fruit, soft drinks and water consumption.

 

Self-reported measure of diet.

Unable to distinguish results of each component of the intervention; in particular, it is not possible to draw a conclusion on the efficacy of digital media.

 
Haerens L, Deforche B et al, 2007 

Study Design: Randomized Controlled Trial

Class: A 

Rating: Positive

N=304 (70% female).

Age: 13 years.

Ethnicity/Race: White.

Attrition rate: 10%.

Location: Belgium.

 

Methods: RCT to evaluate the acceptability, feasibility and effectiveness of a computer-tailored dietary fat intake education program for adolescents.

Outcome measures: Dietary intake assessed using an FFQ.

Theory: Transtheoretical Model.

Duration/Intensity: Single 50-minute session.

Intervention group: Students were exposed once in class to a 50-minute theory-based computer-tailored dietary fat intake intervention.

Comparison group: No intervention control.

 

Girls enrolled in technical-vocational schools significantly ↓ fat intake (P<0.05), as did boys and girls undertaking general education who reported to have read the intervention messages (P<0.05).

 

Study was short in duration, and long-term effects are not known.

Sample demographics may not be generalizable.

 
Haire-Joshu D, Nanney MS et al, 2010 

Study Design: Randomized Controlled Trial

Class: A 

Rating: Positive

N=451.

Age: 8.5 years.

Ethnicity/Race: 37% African American.

Location: United States.

 

Methods: RCT to test the effects of a multi-component diet and activity intervention on children’s dietary intake.

Outcome measures: Survey used to assess child-related outcomes, including dietary intake.

Theory: Social Ecological Model.

Duration/Intensity: Eight sessions over five months (2.3 to 10.8 months).

Intervention group: Children in intervention group received eight computer-tailored storybook sessions, parent action newsletters, and trained mentors.

Comparison Group: Usual school-based nutrition education.

 

For all subjects, fruit and vegetable consumption, total calorie intake and % energy from fat ↑, with no differences between intervention and control groups.

When results were stratified by weight status, overweight and obese children in the intervention group ↓ consumption of calories from high-fat foods, while normal-weight children did not (P=0.059).

 

Self-reported data.

Limited generalizability of study population.

 
Horne PJ, Tapper K et al, 2004 

Study Design: Cluster Randomized Trial

Class: A 

Rating: Positive

N=435.

Age: Four to 11 years.

Location: United Kingdom.

 

Methods: Over 16 days children watched six-minute video adventures featuring heroic peers (the Food Dudes) who enjoy eating fruit and vegetables, and received small rewards for eating these foods themselves.

Outcome measures: Fruit and vegetable consumption measured at lunchtime using a five-point observation scale; at snack time using a weighed measure; and at home using parental recall.

Duration/Intensity: 16 days; six-minute videos.

Intervention group: Over 16 days children watched six-minute video adventures featuring heroic peers (the Food Dudes) who enjoy eating fruit and vegetables.

 

Compared to the control school, lunchtime fruit and vegetable consumption in the experimental school was substantially ↑ at intervention and follow-up than baseline (P<0.001), while snack time consumption was ↑ at intervention than baseline (P<0.001). Lunchtime data showed particularly large ↑ among those who initially ate very little.

Also significant ↑ in fruit and vegetable consumption at home (P<0.05).

 

Not applicable.

 
Long JD and Stevens KR, 2004 

Study Design: Non-Randomized Controlled Trial

Class: C 

Rating: Neutral

N=121 (52% girls).

Age: 13 years.

Attrition rate: 0%.

Location: United States.

 

Methods: Hroup non-randomized trial conducted to test the effects of a classroom and Web educational intervention on self-efficacy for healthy eating.

Outcome measures: Validated FFQ (YAQ) used to measure dietary intake.

Intervention Group: Nutrition education given using five hours of Web-based instruction and ten hours of classroom curriculum.

Comparison group: Nutrition education embedded in the standard school curriculum during a one-month period, with exposure ranging from zero to three hours.

 

No differences between the schools in consumption of fat, fruits, or vegetables.

 

Sample limited to two schools, and is not generalizable.

Survey used to assess nutrition knowledge had low reliability.

Intervention limited to individuals and did not include Δs to the home or school environment.

 
Mangunkusumo et al 2007 

Study Design: Randomized Controlled Trial

Class: A 

Rating: Positive

N=486 (263 in the intervention, 223 in the control; from 30 classrooms (16 intervention, 14 control).

Age: 10 years.

Attrition rate: 3.5%.

Location: The Netherlands.

 

Methods: Cluster-RCT conducted to determine whether Internet-tailored advice for schoolchildren and Internet-supported dietary counseling impact fruit and vegetable intake. During school hours, all children completed Internet-administered questionnaires on fruit and vegetable intake and related determinants.

Theory: Behavioral Change Theory.

Duration/Intensity: Intervention delivered once; additional details regarding intensity not provided.

Outcome measures: Children completed questionnaire at baseline and at three months that included a validated FFQ component.

Intervention group: Children in the intervention group received immediate online tailored nutrition feedback, and a nurse also received information via the Internet to support a five-minute counseling protocol to promote fruit and vegetable intake.

Comparison group: Control group did not receive any of the intervention components.

 

NS differences between groups in fruit and vegetable intake following the intervention.

Awareness of inadequate fruit intake (OR 3.0; 95% CI: 1.8 to 5.3) and knowledge of recommended vegetable intake levels (OR 2.7; 95% CI: 1.8 to 4.1) were significantly more likely at post-test in the intervention group than in the control group.

 

Dietary intake self-reported by the children.

Study sample homogenous with regards to sociodemographic characteristics.

Only one outcome measure timepoint applied (three months).

 
Moore et al 2009; PMID 19363107 

Study Design: Group Nonrandomized Trial

Class: C 

Rating: Neutral

N=126 (64 from School One and 62 from School Two).

Ethnicity/Race: 93% African American.

Age: Nine to 11 years.

Attrition rate: 35% participation rate; attrition not reported.

Location: United States.

 

Methods: A quasi-experimental study was conducted to determine the effect of Color My Pyramid on children's nutrition knowledge and nutrition status. The intervention program incorporates an online component www.MyPyramid.gov (the Blast-Off game).

Theory: Self-Care Deficit Nursing Theory.

Duration/Intensity: Six classes taught over a three-month period.

Outcome measures: Pre- and post-test administered to re-evaluate children’s nutrition knowledge and self-care practices (content validity and reliability assessed).

Intervention group: School Two used individual computers to evaluate their diets in small groups. 

Comparison group: School One, the control school, received a more didactic presentation on playing the Blast-Off game.

 

Significant improvement in nutrition self-care practices in both groups (P<0.05), but no difference between the groups.

 

Study length was three months, so it is unclear what the longer-term impacts would have been.

Difficult to distinguish difference in the intervention between School One and School Two.

 
Pempek and Calvert 2009 

Study Design: Randomized Controlled Trial

Class: A 

Rating: Neutral

N=30 African American children (15 girls, 15 boys).

Age: Nine years.

Attrition rate: 0%.

Location: United States.

 

Methods: An RCT conducted to examine how advergames affect consumption of healthier and less healthy snacks by low-income African American children. Children played an advergame in which they were rewarded for having their computer character consume healthier or less healthy foods and beverages.

Duration/Intensity: Intervention tested on one occasion.

Conditions: Children randomly assigned to one of the following conditions: 1) The healthier advergame condition or 2) the less healthy advergame.

Children in treatment conditions played a less healthy or a healthier version of an advergame two times before choosing and eating a snack and completing the experimental measures.

Children in control group chose and ate a snack before playing the game and completing the measures.

 

Children who played the healthier version of the advergame selected and ate significantly more healthy snacks (1.4±24 servings) than did those who played the less healthy version (0.20±0.24) (P=0.007), with the control group falling in between (0.90±0.24).

Nine children (90%) in the healthy condition chose at least one healthy snack, whereas six children (60%) in the control group and one child (10%) in the less healthy chose at least one healthy snack. The healthier and less healthy conditions differed significantly (P=0.001).

 

Long-term effects of advergames on diet not assessed.

Study sample was limited to only African American children, and should be expanded to other ethnicities and age groups.

 
Thompson D et al., 2009 PMID: 19765608 

Study Design: Group randomized controlled trial

Class: A 

Rating: Positive

N=473 boys.

Age: 10 to 14 years.

Attrition rate: Not described; participation rate was 75%.

Location: United States.

 

Methods: RCT to test the effects of a nutrition education intervention on fruit juice and low-fat vegetable intake on boys.

Theory: Social Cognitive Theory.

Outcome measures: Fruit juice and low-fat vegetable intake assessed at baseline, immediately following the intervention, and six months post-intervention using a modified, validated FFQ. 

Intervention group: Intervention group participated in a nine-week program that included ~30 minutes of weekly troop time, plus ~25 minutes of weekly Internet programming. Scouts were encouraged to log on to the Internet site twice weekly to participate in a behavior Δ program and set goals, as well as to report goal attainment.

Comparison group: Control group participated in a mirror image intervention to ↑ physical activity.

 

Immediately following the intervention, subjects in the intervention significantly ↑ consumption of fruit juice (0.94 servings per day), compared to control subjects (0.56 servings per day; P<0.003).

However, this difference was not maintained six months later.

At six months post-intervention, intervention subjects ↑ low-fat vegetable intake compared to control (one serving, P<0.05).

 

Log-on rates were acceptable, but the approach was labor intensive and costly.

Limited generalizability of the sample.

Use of an FFQ to assess diet.

 
Thompson D, Baranowski T et al, 2008 

Study Design: Randomized Controlled Trial

Class: A 

Rating: Positive

N=73 African American girls (37 in the intervention, and 36 in the control group).

Age: Eight to 10 years.

Attrition rate: 9%.

Location: United States,

 

Methods: RCT conducted to determine the efficacy of an Internet-based program on dietary intake and physical activity.

Girls randomized to receive immediate (weekly) or delayed (program end) incentives ($5). Participants’ given weekly goals to ↑ target behaviors and included focus on role modeling, problem solving and goal setting and review.

Theory: Social Cognitive Theory.

Duration/Intensity: Eight weeks. Subjects asked to log on weekly, and weekly log on rate averaged 74.5%.

Outcome measures: Dietary intake assessed using a seven-item FFQ (validity not reported).

Intervention Group: Internet-based program emphasized fruit, 100% juice, vegetable (FJV) and water intake, as well as physical activity.

Comparison Group: A delayed group was used as comparison group.

 

Statistically significant pre-post improvement observed in FJV consumption, with girls reporting an ↑ of one serving per day at the end of the intervention (P=0.002).

A significant ↑ in FJV self-efficacy was also observed (P=0.003).

 

Study sample limited to only African American girls, and should be expanded to other ethnicities, genders and age groups.

Dietary intake self-reported by the children.

Did not report separate results for fruit, 100% juice and vegetables.

 
Turnin et al 2001 

Study Design: Group Randomized Controlled Trial

Class: A 

Rating: Neutral

N=1,876 children from 15 schools (53% girls).

Age: Nine years.

Attrition rate: 16%.

Location: France.

 

Methods: 15 schools randomized into two groups (games group and control group), both receiving conventional nutritional teaching by their teachers.

Duration/Intensity: Intervention delivered for one hour, twice a week, for five weeks.

Outcome measures: At completion of study, nutrition knowledge and three-day diet records evaluated in both groups; however, this data not assessed at start of the study (validity not described).

Intervention group: The games group played computer games during the conventional nutritional teaching period.

Comparison group: Control group received conventionally taught nutrition education.

 

The games group had significantly better dietary intake than the control group: ↑ CHOs (46.4%±0.2% vs. 45.7%±0.2%, P<0.05), ↓ fat (37.1%±0.1% vs. 37.6%±0.2%, P<0.05), ↓ protein (16.5%±0.1% vs. 16.7%±0.1%, P<0.05), ↓ sugar (11.5%±0.1% vs. 12.2%±0.2%, P<0.001), ↑ calcium (P<0.001) and ↑ fiber (P<0.05).

Nutrition knowledge test results were better in the games group (P<0.001), compared to the control group.

 

No pre-test, only a post-test, limiting interpretation of the study findings.

Long-term affects of advergames on diet not assessed.

 
Williamson et al 2005; PMID 16277142 

Study Design: Randomized Controlled Trial

Class: A 

Rating: Positive

N = 50 African American girls.

Age: 13 years.

Attrition rate: 12%.

Location: United States.

 

Methods: RCT to compare the efficacy of an interactive Internet-based behavioral weight management program to a passive Internet-based health education program in overweight adolescent girls over a two-year intervention period.

Duration/Intensity: Four face-to-face sessions over a 12-week period, and subjects were followed for six months.

Outcome measures: Dietary intake measured using a multiple-pass 24-hour recall and the Block FFQ.

Intervention group: Internet-based intervention that provided nutrition education plus a behavior modification program (including Internet counseling) that targeted adolescents and their parents.

Comparison group: Internet-based education on healthy nutrition and exercise, but without prescribed behavior Δs.

 

Girls in the behavioral intervention also ↓ total energy intake (P<0.001), protein (P<0.05) and fat intake (P<0.05), while those in the control group only ↓ fat intake (P<0.05).

 

Self-reported measures of dietary intake.

Limited generalizability of the study sample.

 
Winett et al 1999 

Study Design: Nonrandomized Controlled Trial

Class: C 

Rating: Neutral

N = 180 girls (103 in intervention group, 77 in control group).

Age: 15 to 16 years. 

Attrition rate: 0%.

Location: United States.

 

Methods: RCT to investigate the effects of the Eat4Life Internet-based health behavior program on the nutrition and activity practices of adolescent girls.

The Eat4Life modules provide a brief assessment of the subject’s nutritional practices, followed by personalized feedback.

Theory: Social Cognitive Theory.

Duration/Intensity: Intervention was a semester long; additional details regarding intensity were not provided.

Outcome measures: Food intake measured using an adaptation of a 24-hour recall and FFQ (the Block).

Intervention group: Used the Eat4Life computer program in their health education classes and received a sample of a food recommended in the module.

 

Girls using the Eat4Life modules reported that they made relatively consistent Δs in all areas of nutrition except for ↓ the consumption of high-fat dairy products.

Eat4Life program was effective in ↑ consumption of regular meals, fruits and vegetables, and breads and cereals (all P<0.001), and ↓ consumption of regular soft drinks and fast foods (P<0.05).

 

Use of non-randomized group assignments.

Different teachers taught different study groups.

Study was short-term in nature, and long-term effects are unknown.

Use of self-reported measures.

Limited generalizability of the study sample.

 

Research Design and Implementation
For a summary of the Research Design and Implementation results, click here.
Worksheets
Bannon K, Schwartz MB. Impact of nutrition messages on children's food choice: Pilot study. Appetite. 2006 Mar; 46(2): 124-129. Epub 2006 Jan 26.

Baranowski T, Baranowski J, Cullen KW, Marsh T, Islam N, Zakeri I, Honess-Morreale L, deMoor C. Squire’s Quest! Dietary outcome evaluation of a multimedia game. Am J Prev Med. 2003;24(1):52-61.

Baranowski T, Baranowski JC, Cullen KW, Thompson DI, Nicklas T, Zakeri IE, Rochon J. The Fun, Food, and Fitness Project (FFFP): the Baylor GEMS pilot study. Ethn Dis. 2003 Winter;13(1 Suppl 1):S30-9. 

Baranowski T, Baranowski J, Thompson D, Buday R, Jago R, Griffith MJ, Islam N, Nguyen N, Watson KB. Video Game Play, Child Diet, and Physical Activity Behavior Change: A Randomized Clinical Trial. Am J Prev Med 2011 Jan; 40 (1): 33-38.

Casazza K, Ciccazzo M. The method of delivery of nutrition and physical activity information may play a role in eliciting behavior changes in adolescents. Eat Behav. 2007 Jan;8(1):73-82.

Cullen KW, Watson K, Baranowski T, Baranowski JH, Zakeri I. Squire's Quest: Intervention changes occurred at lunch and snack meals. Appetite. 2005; 45: 148-151.

DeBar L, Dickerson J, Clarke G, Stevens V, Ritenbaugh C, Aickin M. Using a website to build community and enhance outcomes in a group, multi-component intervention promoting healthy diet and exercise in adolescents. Journal of Pediatric Psychology 2009; 34 (5): 539-550.

DeBar LL, Ritenbaugh C, Aickin M, DeBar LL, Rittenbaugh C, Aickin , Orwoll E, Elliot D, Dickerson J, Vuckovic N, Stevens VJ, Moe E, Irving LM. Youth: A health plan-based lifestyle intervention increases bone mineral density in adolescent girls. Arch Pediatr Adolesc Med. 2006 Dec; 160(12): 1,269-1,276.

Di Noia J, Contento IR, Prochaska JO. Computer-mediated intervention tailored on transtheoretical model stages and processes of change increases fruit and vegetable consumption among urban African-American adolescents. Am J Health Promot. 2008 May-Jun;22(5):336-41.

Frenn M, Malin S, Bansal N, Delgado M, Greer Y, Havice M, Ho M, Schweizer H. Addressing health disparities in middle school students' nutrition and exercise. J Community Health Nurs. 2003 Spring; 20 (1): 1-14.

Haerens L, De Bourdeaudhuij I, Maes L, Vereecken C, Brug J, Deforche B. The effects of a middle-school healthy eating intervention on adolescents' fat and fruit intake and soft drinks consumption. Public Health Nutr. 2007 May; 10 (5): 443-449. PMID: 17411463.

Haerens L, Deforche B, Maes L, Cardon G, Stevens V, De Bourdeaudhuij I. Evaluation of a 2-year physical activity and healthy eating intervention in middle school children. Health Educ Res. 2006; 21(6): 911-921.

Haerens L, Deforche B, Maes L, Brug J, Vandelanotte C, De Bourdeaudhuij I. A computer-tailored dietary fat intake intervention for adolescents: Results of a randomized controlled trial. Ann Behav Med. 2007 Nov-Dec; 34(3): 253-262.

Haire-Joshu D, Nanney MS, Elliott M, Davey C, Caito N, Loman D, Brownson RC, Kreuter MW. The use of mentoring programs to improve energy balance behaviors in high-risk children. Obesity. 2010; 18(Suppl 1): S75-S83.

Horne PJ, Tapper K, Lowe CF, Hardman CA, Jackson MC, Woolner J. Increasing children's fruit and vegetable consumption: A peer-modelling and rewards-based intervention. Eur J Clin Nutr. 2004 Dec; 58(12): 1,649-1,660.

Long JD, Stevens KR. Using technology to promote self-efficacy for healthy eating in adolescents. J Nurs Scholarsh. 2004; 36: 134-139.

Mangunkusumo RT, Brug J, de Koning HJ, van der Lei J, Raat H. School-based internet-tailored fruit and vegetable education combined with brief counselling increases children's awareness of intake levels. Public Health Nutr. 2007 Mar;10(3):273-9.

Moore JB, Pawloski LR, Goldberg P, Kyeung MO, Stoehr A, Baghi H. Childhood obesity study: a pilot study of the effect of the nutrition education program Color My Pyramid. J Sch Nurs. 2009 Jun;25(3):230-9. Epub 2009 Apr 10.

Pempek TA, Calvert SL. Tipping the balance: use of advergames to promote consumption of nutritious foods and beverages by low-income African American children. Arch Pediatr Adolesc Med. 2009 Jul;163(7):633-7.

Thompson D, Baranowski T, Baranowski J, Cullen K, Jago R, Watson K, Liu Y.  Boy Scout 5-a-Day badge: outcome results of a troop and internet intervention.  Preventive Medicine 2009; 49:518-526.

Thompson D, Baranowski T, Cullen K, Watson K, Liu Y, Canada A, Bhatt R, Zakeri I. Food, fun, and fitness internet program for girls: Pilot evaluation of an e-Health youth obesity prevention program examining predictors of obesity. Prev Med. 2008 Nov; 47(5): 494-497. Epub 2008 Jul 30.

Turnin MC, Tauber MT, Couvaras O, Jouret B, Bolzonella C, Bourgeois O, Buisson JC, Fabre D, Cance-Rouzaud A, Tauber JP, Hanaire-Broutin H. Evaluation of microcomputer nutritional teaching games in 1,876 children at school. Diabetes Metab. 2001 Sep;27(4 Pt 1):459-64.

Williamson DA, Martin PD, White MA, Newton R, Walden H, York-Crowe E, Alfonso A, Gordon S, Ryan D. Efficacy of an internet-based behavioral weight loss program for overweight adolescent African-American girls. Eat Weight Disord. 2005 Sep;10(3):193-203.

Winett RA, Roodman AA, Winett SG, Bajzek W, Rovniak LS, Whiteley JA. The effects of the Eat4Life Internet-based health behavior program on the nutrition and activity practices of high school girls. Journal of Gender, Culture, & Health. 1999;4(3):239-254.