Ethics Statement

Ethical approval was not required as this is a synthesis of previously published research. The authors received no specific funding for this study and declare no conflicts of interest.

Component analysis

This component analysis was methodologically similar to a previous paper that analysed the components of self-monitoring interventions for CVD [19] and was carried out in six stages:

1. A literature search for relevant interventions.
2. Description of the included interventions.
3. Extraction of the intervention components into a pre-defined template.
4. Extraction of further information about the included interventions such as theoretical basis and the results achieved.
5. Authors contacted to provide missing information if any of the previous stages are not able to be completed from the papers.
6. Finally the extracted information was examined for components that were most frequently associated with success in interventions.

We searched the following databases: EMBASE 1974-week 3 November 2014, Ovid Medline 1946 Present, and PsychINFO 1967-Present. The search terms detailed in Fig 1 were truncated with wild card characters where necessary and searched in each of the databases. The PRISMA Checklist for this study can be seen in S1 PRISMA Checklist.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. Search terms.

https://doi.org/10.1371/journal.pone.0135115.g001

Studies included in the component analysis had to report on an intervention carried out in healthy young people under the age of 18 and use a control group so that the likelihood of any results occurring due to the effect of the intervention could be judged. We defined an improvement or healthy change in the variables of interest as a change in the intervention group compared to the control group, that could either be an increase or decrease depending on the variable, which reached a significance level of p≤0.05 as reported by the study. Included interventions had to contain an education component and had to have been carried out in a community, school, or clinical setting. Papers had to report on at least one of the CVD risk parameters searched as an outcome as well as the change in at least one non-biological outcome from knowledge, attitudes or behaviours. The inclusion criteria are summarised below:

• Intervention studies.
• Must have a control group.
• Must be group interventions.
• Interventions must be carried out in a school, community or clinical setting.
• The intervention must have an education component.
• Must measure at least one CVD risk parameters searched.
• Must measure change in at least one of knowledge, attitudes or behaviours.

Studies were excluded if they contained a pharmacological element or if the paper did not report on changes in both the CVD risk parameters and the change in the knowledge, attitudes, or behaviours of the children. Papers were not excluded based on country of origin or date or language of publication. The exclusion criteria are summarised below:

• Use of a pharmacological treatment.
• Participants have a known co-morbid condition.
• Failure to report sufficient outcome data.

Components common to changes in outcomes

In order to ascertain which components were present in studies that brought about a positive change in the outcomes of interest, any papers which failed to improve at least one biological and one non-biological outcome were excluded. This left nine out of the twelve included studies remaining [20–30, 32–36]. Of the three excluded studies, Schofield, Mummery and Schofield (2005) [38] and Johnson et al. (1991) [37] were able to bring about some significant improvements in physical activity and dietary measures, however these were not accompanied by significant improvements in the biological factors of interest as defined above. Park et al. (2007) [31] reported significant improvements in many of the biological factors of interest, but did not report any changes in non-biological factors. These studies reported interventions that were among the shortest of those included in this analysis which could have meant that they were not long enough for significant improvements to be seen. There were those, however, that did achieve improvements in both the biological and non-biological factors that were of similar length, and so we must assume that the different outcomes seen were due to the differences in how the interventions were constructed, delivered, and assessed. For the details of which factors were improved in all 12 studies see S1 Table.

Table 2 shows the components common to the nine studies that had achieved an improvement in both biological and non-biological outcomes. The components were grouped into component themes and each theme is discussed below.

Health behaviours

The subcomponents of this theme focussed on instructing the children in healthy eating [20–30, 32–34, 36] and physical activity [20–26, 32–36]. The healthy eating component included learning about the food pyramid [32], setting targets like reducing the consumption of sugar sweetened drinks, increasing the consumption of fruit and vegetables [27, 33, 34], and increasing the children’s familiarity with different fruit and vegetables [20]. Physical activity included encouraging the children to increase the amount of time they spent being active [32], physical education (PE) lessons that placed greater importance on fitness [20, 32, 35], and activities that could easily be incorporated into the children’s lifestyles [21, 33, 34]. Of the studies that included instruction in healthy eating, all nine studies brought about significant (p-values ≤ 0.05) improvements in the children’s diets. Such improvements included reductions in carbonated and fruit juice drinks, reduction in fats and oils consumed, increases in the number of children eating breakfast and increases in fruit and vegetable servings consumed. Seven of the eight studies targeting physical activity brought about significant (p-values ≤ 0.05) improvements which included reductions in screen time and increases in extra-curricular physical activity.

Self-concept

This theme included three sub-components; self-esteem, healthy body image [20–26, 35], and self-efficacy [20]. Self-esteem and healthy body image was addressed through discussions to change self-concept, role-modelling, education in growth and development, and introducing techniques such as positive thinking and self-talk. Self-esteem and body image were assessed using a pictorial current-self and ideal-self rating scale, the Beck Youth Inventory [39], which includes a sub-scale on self-concept, or the Harter Self-Perception Profile [40]. Significant improvements in self-esteem or body satisfaction were reported in the LOOZIT study (global self-worth; 0.21, 95% CI 0.01 to 0.32, p<0.001 and body shape dissatisfaction; -0.56, 95% CI -0.74 to -0.38, p<0.001) and the MEND study (global self-esteem score increase; 0.3, 95% C.I. 0.0 to 0.5, p = 0.03) [23, 27]. However, even in the studies where such improvements were not found, there were improvements to the children’s attitudes towards their health, which could have been facilitated by attempts to increase self-esteem.

Increasing self-efficacy to lead healthy lives was targeted specifically by the CHILDREN programme [20]. Self-efficacy was targeted through techniques such as modelling, enactment and guided practice in weekly sessions that lasted one to two hours [20]. This study did not, however, measure the change in self-efficacy, instead, change in diet, physical activity and sedentary behaviour were measured.

Practical and cognitive tools

The practical and cognitive tools theme included components such as goal setting, where the children had to make goals for themselves and report back at a later time whether these were achieved or not [20, 22–26]. Where they were not achieved, discussions were held about the reasons why this may be and ways in which problems arising could be dealt with. Children were given practical skills such as how to read nutrition labels on food packaging, how to modify recipes to make them healthier, and how to make healthy choices, for example when eating at a restaurant [27, 32]. Cognitive skills, such as ways in which stress could be managed more effectively through improving the children’s communication skills, were also used by some studies [22–26].

All of the studies that aimed to provide children with practical or cognitive skills achieved improvements in either one or both of their physical activity levels and diet.

Intervention characteristics

Unlike the preceding three themes, the intervention characteristics theme summarised the ways in which the successful interventions were delivered rather than what they delivered. Common characteristics included interventions that were interactive, the children were encouraged to discuss, play games and use role play to facilitate their learning [20], and that both children and those delivering the intervention were provided with materials such as workbooks [20, 22–26, 28–30, 32–34]. The children’s workbooks contained physiological and health information, activities, and space to enter their own thoughts that could be kept after the intervention had finished. Materials given to the researchers or teachers were often provided with the intention of keeping the delivery of the intervention consistent across lessons and were made up of instructions on delivering the intervention, reproducible materials for the children, and guidance on how the intervention should be assessed.

A common component was the involvement of parents, peers or the community so that general knowledge could be increased and support given to the children participating in the intervention [22–26, 28–30, 32, 37]. Finally, some of the studies used professionals, such as dieticians, in the design and delivery of the intervention [28–30, 32].

Strengths and weaknesses

The strength of this component analysis lies in the structured and systematic way in which relevant studies were identified for inclusion and the efforts made to ensure that their design was as comparable as possible. In this way 12 studies were selected and four component themes were identified that have been used to promote heart health to children and were successful at least in the short term. This component analysis is however not without its weaknesses. It is possible that our database search failed to identify some relevant studies that should have been included in the analysis. It is our belief however that as our searches were designed carefully and carried out in multiple databases this is unlikely. A second weakness is that although great care was taken to ensure that the included studies were as comparable and matched our requirements as closely as possible, there was still a level of heterogeneity among them in the ways in which they were delivered, their duration and the characteristics of the children to whom they were delivered. We believe however that as this analysis was descriptive rather than quantitative, this limitation is not serious enough to negate the findings.

Implication for research and practice

The purpose of this study was to identify the components associated with education interventions that were successful in improving the health and health behaviours of children so that they might be incorporated into new interventions to target childhood cardiovascular health. It was found that an emphasis should be placed on how behaviours impact on internal mechanisms so that why healthy living is important might be highlighted to children. Second, the self-esteem of the child should always be central to any intervention and interventions should be designed so that weight bias is not perpetuated, but instead children are encouraged to see themselves as able to influence their own health outcomes and are equipped with the confidence and self-efficacy to do so. Third, children should be given practical and cognitive tools by health lessons that will assist them in their health decision making. Such skills could include stress management, reading nutrition labels, understanding how diet and physical activity interact to either promote or impede health, adapting recipes, and communication skills. Finally, health lessons should be interactive and hands on with materials for both the student and the teachers delivering the lessons so that they may be delivered with fidelity to the intervention design, allowing the effectiveness of the intervention to be assessed appropriately.