INTRODUCTION

We can consider the obesity as a complex and multifactorial chronical disease. Its development is influenced by genetic, environmental and behavioral factors. Among of them, we can highlight several environmental and behavioral factors as feeding habits, the lack of physical activity and sedentary lifestyle¹.

Both the overweight and obesity in children have a great influence in the development of different physical comorbidities such as diabetes mellitus type-II, metabolic syndrome, cardiovascular diseases, osteoarticular disorders or different types of cancer. Moreover, these ill children are also exposed to other psychosocial consequences like low self-esteem, body image disorders, depression, stigmatization, margination or bulling¹.

In the las three decades the childhood obesity has been doubled worldwide. In 1980 the obesity rate in children between 6-11 years old was around 7%, increasing till 18% in 2010¹.

According to a US study which analyzed the tendency of the prevalence of obesity in children and teenager between 2013 and 2014, the 17% of the children between 2 and 19 years old were suffering obesity and the 5.8% extreme obesity. In 1990 the obesity in children between 6 and 11 years old was about 11.3% numbers that raised up to 19.6% in 2008 and keeping constant till 2014².

In Spain, the ALADINO study carried out in 2015 and which analyzed the overweight and obesity among children ion the in the primary school showed that the overweight rate in children between 6 and 9 years old was 23.2% (22.4% in boys and 23.9% in girls) while the obesity reached 18.1% (20.4% in boys and 15.8 in girls)³.

Aranceta Bartrina et al.⁴⁾ proposed that the educational actions for health focused on children in the primary school (6-12 years old) are the most effective way to act against the increase of childhood obesity. It seems that these are educational strategies are useful in the prevention and modification of unhealthy habits which favor the weight overload.

The primary school is one of the most adequate place for carrying out these type of educational actions since the children spend a lot of hours there and, moreover, it is possible to reach almost all the population in short period of time⁵.

On the other hand, the period in the primary school is an ideal frame for these type of actions because the most susceptible age to change in habits and behavior⁵. It also quite clear that the overweight and obesity became permanent between 11 and 12 years old meaning that the risk of suffering obesity and comorbidities as an adult is much higher. this fact justifies the need of educational actions in children below 12 years old in order to avoid a weight overload at young age⁶.

A recent meta-analysis concluded that the actions on the diet and the physical activity performed in primary schools are very useful in the prevention of the overweight⁷. However, other authors claimed that the actions on the feeding and physical exercise during the school period are not effective in the prevention of childhood obesity and, if some positive effects are achieved, these are not kept in time. There is also no agreement about the ideal extension of the education programs to achieve long-lived changes in the children weight⁸.

For all these reasons and due to the high prevalence of the childhood obesity and the associated health risks, the following research question is posed following the PICOS format (Patient/Problem, Intervention, Control/Comparison, Results, Design of studies): Are the educational actions on nutrition and/or physical activity carried out on primary schools (6-12 years old) effective for the prevention and/or decrease the childhood obesity?

- P: children on primary school (6-12 years old)

- I: educational actions in primary schools about nutrition and/or physical activity

- C: no interventions or education about nutrition and/or physical activity is given

- O: effect on the weight, body mass index (BMI), waist circumference or skin folds (obesity measurements)

- S: experimental studios on control groups

On this way, an extensive bibliographic review is carried out to:

- Describe the educational actions on feeding habits and/or physical activity carried out in children on primary school in order to reduce or prevent the childhood obesity

- Analyze the efficiency of this interventions.

METHODOLOGY

Protocol

A systematic bibliography review was carried out following the guidelines established in PRISMA 2010 statement for systematic reviews and meta-analysis.

Eligibility criteria

Studies about children in the primary school period (6-12 years old) were included in this review.

The experimental studies with control groups in which nutritional and/or physical activity interventions have been carried out in the schools against no interventions or only the typical education about feeding and physical activity.The studies carried out with actions out of the school environment (the community, at home or primary attention) or without any control group have been avoided.

Studies whose objectives were the evaluation of the impact of nutrition and/or physical activity intervention programs on the childhood obesity with an analysis. The reports that did not include this information were avoided.

Other eligibility criteria were the date and language. Only studies published in the las 5 years (January 2013 - September 2017) in Spanish or English were considered. The occupation of the authors of the study or the amount of time employed in the study were not taken into consideration.

Information and searching sources

The electronic databases used for the bibliographic search were Web of Science (WOS) y SCOPUS. These databases were employed from July 2017 till September 2017.

The search of articles was carried out using the following combination of terms and Booleansoperators both in WOS and SCOPUS: ("pediatricobesity" OR "obesity" AND ("healtheducation" OR "prevention" OR "intervention") AND ("schools" OR "educationprimary" OR "primaryschool") AND ("nutrition") AND("physicalactivity").

The employed filters in both databases were the date and type of document (2013 - 2017 and article or review, respectively).

Moreover, the references included in these studies were revised and incorporated if the eligibility criteria were fulfilled and they have not been found previously with the systematic search.

Selection of the studies

After the search, the duplicated studies were removed using the tools provided by Mendeley Desktop. Then, a deep analysis of the titles and abstracts was carried out to discard the irrelevant articles. The abstracts that were ambiguous about the eligibility criteria were selected to a more in detail review by a complete reading. These documents were then read and included if they fulfilled the eligibility criteria.

The quality of the methodology employed in the different studies were examined by a detailed read of the documents that satisfied the eligibility criteria. This was made by the use of CASPe, specifically: "11 preguntas para darsentido a un ensayoclínico"⁹.

Finally, a fine screening was carried out considering different evidence levels according to the Scottish Intercollegiate Guidelines Network (SIGN). The studies with evidence level SIGN "1-" y "2-" were removed since this tool does not recommend to use this type of studies due to their bias high potential.

Data Acquisition Process

The acquisition of the data from the articles that fulfilled the eligibility criteria and included in the final review, was carried out preparing twodifferent Tables (Table 1 y 2).In these Tables, the following data is summarized: year and reference, country, design of the study, sample, type of actuation, and main results and conclusions. Also the evidence degree according to SIGN scale was also included.

List of data

The main data extracted from each article for the preparation of this work were: the age of the children, information about nutrition and/or physical activity interventions, related to school, parent's participation, duration of the interventions and duration of the results.

RESULTS

After the systematic bibliographic search in the databases, a total of 571 articles were found. First, 185 articles were removed since they were duplicated. Then, another 355 irrelevant articles were excluded after reading the titles and abstracts. After this first selection, 62 articles were identified as potentially relevant. After analyzing the full content of 61 of these articles, finally only 22 articles were considered for the systematic review (Figure 1).

Figure 1.  Flowchart. 

Among the selected 22 articles, 19 of them were randomized experimental studies^{6)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20)(21)(22)(23)(24)(25)(26)(27}, 1 of them was a non-randomized experimental study with control group²⁸and 2 of them werequasi-experimental studies with control group²⁹⁾⁽³⁰, as it summarized in Table 1 together with the evidence level SIGN assigned to each one.

In all the selected articles, the population under study was 6-12 years old children in the primary school. Two of the studies only evaluated actions on nutrition²⁶⁾⁽³⁰, five of them only on physical activity^{10,17,18,27,29}and fifteen of them evaluated combined actions on nutrition and physical activity^{6)(11)(12)(13)(14)(15)(16)(19)(20)(21)(22)(23)(24)(25)(28}. These data were summarized in Table 2 together with the rest of the extracted information.

DISCUSSION

Nutrition interventions

It has been found that the activity of distributing free fruits in the schools increases its consumption while reduces the consumption of unhealthy snacks and helps to decrease the prevalence of the obesity. Therefore, this activity is a clear strategy to prevent overweight and improve eating habits²⁶.

On the other hand, it was observed that the interventions focused on stimulate a good breakfast help to improve the healthy eating habits but do not show any effect on the BMI of the children³⁰.

Physical activity interventions

The strategy of increase the physical activity in the curriculum seems to have positive effects since the practice of exercise is increased and the physical attitude improved¹⁷⁾⁽¹⁸ while also, the waist circumference is reduced¹⁸. However, one study in which this intervention was used found no improvement in the physical condition of the children nor any effect in the obesity measurements¹⁰.

It was also observed that this type of intervention produces long term results since even 2 years after its completion, it was found a decrease in the body fat and overweight and obesity prevalence,having these children less risk of becoming obese^17,29. However, in another study in which at the end of the intervention it was possible to reduce fat, improve physical attitude and an increase in exercise, years later only the effects on the increase in activity were maintained²⁷.

Combined interventions: nutrition and physical activity

Among the fifteen studies that performed combined interventions^{6)(11)(12)(13)(14)(15)(16)(19)(20)(21)(22)(23)(24)(25)(28}, nine obtained positive results in the decrease or maintenance of BMI^{11)(12)(13)(14)(16)(19)(21)(24)(28}and sixdid not achieve statistically positive changes on children's BMI^{6)(15)(20)(22)(23)(25}.

Programs with educational interventions of physical activity and healthy eating, either included in the curriculum or extracurricular, achieved a decrease in waist circumference and BMI. In addition, physical activity was increased and eating habits were improved, with a decrease in the prevalence of obesity^{11)(12)(19)(21}.

Comparing the combined interventions with the simple ones, it was observed that the first ones were more useful to increase activity, improve diet and decrease body fat²⁸. It was also found that comprehensive programs achieve, years after their completion, a significant reduction in the excess of weight²⁴.

On the other hand, one study found that these types of interventions do not always reduce the prevalence of obesity, being these programs useful to control it but not to prevent it¹⁴. Additionally, it cannot be guaranteed that the reductions in the BMI are maintained in the long term¹⁶.

In contrast with these results, there are other studies that failed to reduce BMI^{6)(15)(20)(22)(23)(25}. Although it is also true that in some cases they did reduce sedentary lifestyle, the consumption of sugary drinks and unhealthy snacks and increase the consumption of fruits and vegeTables and physical exercise^{6)(20)(22)(23}.

Another remarkable about this type of interventions was the significant increase in knowledge about healthy eating and physical activity that children showed^15)(23)(25.

Parental involvement in educational programs

There are many studies that have included the participation of parents/caregivers in the educational interventions^{6)(11)(12)(13)(14)(16)(19)(21)(22)(23)(24)(25)(28)(29)(30}.

Fairclough et al.¹¹explained that the interventions that were most effective were those combined, those included in the curriculum and those with parental involvement. This conclusion was also reached by other authors who proposed that, for obesity interventions to be successful, school support and family participation are important, since parents are essential for the success of these interventions^23)(25)(30.

A study that did not achieve changes in obesity measures, proposes that it is important to invest more in parental involvement to achieve better results¹⁵.

Sex and socioeconomic level

Three experimental studies found that girls had better results in lowering BMI than boys^11,13,14.Only one study found better results in boys than girls¹⁹.

The socioeconomic level also seems to influence the results of the programs. It was found that the higher the socioeconomic and educational level of the parents is, the better the results on the BMI of the children are¹¹⁾⁽¹³. Another study observed that the lower the educational level of the parents is, the greater the increase in BMI is²⁴.

Thus, it seems that both gender and socioeconomic status must be taken into account in the design of educational interventions that attempt to influence the weight of children¹³.

Duration of the interventions

The results in this aspect are quite heterogeneous. The duration of the interventions is an important aspect, although it is not clear what the ideal duration is⁸.

The duration of the experimental studies included in this review varies from 10 weeks to 5 years, being the average duration between 1 and 2 years.

Two studies with a duration of one year that obtained positive results on obesity, did not guarantee that the effects were maintained in the long term¹⁴⁾⁽¹⁶. A 28-month intervention that was also effective at the end of the study, failed to subsequently maintain the effects achieved on BMI¹⁹.

However, studies that lasted 2 years, achieved positive effects with their interventions and / or that the effects were maintained over time^{15)(17)(24)(29}.

Theories on which the studies are based

Among all the studies included in this review only six mentioned the use of some theory^{6,11,15,20,25,28}, although all included interventions with behavioral components.

The most common was the theory of social cognitive behavior, used by four authors^6,11,25,28. Another one used the theory of planned behavior¹⁵and other one the health promotion theory²⁰.

In this review, it was observed that of the four studies using the psychological theory of behavior change, only two achieved positive results¹¹⁾⁽²⁸.Studies based on other theories did not obtain significant results¹⁵⁾⁽²⁰.

Summary of the results

This systematic review has concluded that educational interventions on nutrition and / or physical activity carried out in primary schools seem to have a positive effect on eating habits and physical activity. The consumption of fruits and vegeTables increased, decreasing the consumption of sugary drinks and unhealthy snacks. The practice of physical also increased exercise and the physical attitude of children improved. However, not all studies included in this work have positive effects on BMI.

Although it was observed that the combined interventions on nutrition and physical activity seem to be more promising in the treatment and prevention of childhood obesity^{11)(12)(13)(14)(16)(19)(21)(24)(28}, it was also found that some studies of a single component, either interventions only on nutrition or only on physical activity, have also a positive effect on the treatment or prevention of weight overload^{17)(18)(26)(29}.

Regarding the duration of the interventions, it seems clear that it is an important aspect of their effectiveness and that those that extend longer in time obtain better results¹⁰⁾⁽¹⁷. However, it is not clear what is the ideal duration for a program to be successful, nor what interventions are considered in the long term⁵⁾⁽⁸. In this review it was found that programs with a duration of one year mostly achieve positive results and those with a duration of 2 or more years can mantain these positive effects over time^{14)(15)(17)(19)(24)(29}.

Parental involvement in educational interventions is essential in obtaining good results^{11)(15)(18)(22)(23)(25)(30}. It is also important to take into account both sex and socioeconomic status of participants in educational programs, as it can influence the effectiveness of interventions⁸.

Most studies did not mention the theory on which they based their educational interventions. Despite this, it was found that the most used was the theory of social cognitive behavior, being the only one that achieved significant results on obesity¹¹⁾⁽²⁸.

Summary of the evidence

It was found that the combined interventions on nutrition and physical activity are the most used and effective on BMI. Moreover, they are mostly rated according to the SIGN scale, with high levels of evidence ("1 ++" and "1+"). On the other hand, among the interventions on only nutrition, it was observed that the strategy of distributing free fruit in schools improves eating habits and decrease the prevalence of overweight, being identified with a level of evidence "1+". Conversely, the intervention on increasing the number of breakfast was rated with a level of evidence "2+".

In general, programs on physical activity increase exercise practice, improve physical attitude and help prevent obesity. These studies are identified with levels of evidence similar to the combined interventions.

Parental involvement has been identified as positive, obtaining good results in most of the studies that counted on their participation and identified according to the SIGN scale with high levels of evidence.

It is important to keep in mind that in order to develop this type of interventions the socioeconomic and educational level of the participants as well as the duration are important aspects. Interventions with a duration of 1 to 2 years, get better results, and after 2 years they manage to keep them for longer. These studies were identified with a level of evidence according to SIGN of "1 ++", "2 ++" and "2+".

On the other hand, it was found that the theory of social cognitive behavior is the most used, being classified the studies that use it and obtain good results with the highest levels of evidence.

Limitations

Some limitations were detected during the work on this systematic review. First, the search and selection of articles was carried out by the first author, and supported by the rest of the researchers, so there may be a bias in the selection of studies. The possibility of publication bias also exists, since it is more likely that intervention studies that failed to have a positive effect have not been published.

On the other hand, limitations were also found on the evidence quantification. The studies are heterogeneous in terms of size and characteristics of the sample and characteristics and duration of interventions. "Unclear risk" of bias was also found in most of the included studies when evaluating the quality of the articles, mainly due to lack of information on randomization, blinding and losses during follow-up.

CONCLUSIONS

The initial proposed objectives were fulfilled:

- On the one hand, educational interventions on food and / or physical activity carried out in order to reduce or prevent childhood obesityin Primary School students have been described. It was concluded that the most used educational interventions in these last 5 years and which have achieved better results are the combined interventions on nutrition and physical activity, with activities included in the curriculum and the participation of parents. The optimal duration of interventions is not clear, although it can be concluded that long-term interventions seem to have better results.

- We have analyzed the effectiveness of these interventions and we could conclude that educational interventions on food and / or physical activity help to improve and acquire healthy habits: increasing the consumption of fruit and vegeTables, reducing the consumption of unhealthy snacks and sugary drinks, increasing the time devoted to physical activity and improving the physical attitude of children. There seems to be some evidences that they can also be useful in reducing children's BMI or preventing childhood obesity.