Article Text

Download PDFPDF

Are dietary interventions effective at increasing fruit and vegetable consumption among overweight children? A systematic review
  1. Michael Bourke1,
  2. Paula J Whittaker2,
  3. Arpana Verma2
  1. 1Manchester Medical School, Manchester Academic Health Sciences Centre, University of Manchester, Manchester, UK
  2. 2Institute of Population Health, Manchester Academic Health Sciences Centre, University of Manchester, Manchester, UK
  1. Correspondecne to Dr Arpana Verma, Director, Manchester Urban Collaboration on Health, Institute of Population Health, Room 2.523, Stopford Building, Manchester Academic Health Sciences Centre, University of Manchester, Manchester M13 9PT, UK; Arpana.verma{at}manchester.ac.uk

Abstract

Introduction Childhood obesity is now a global epidemic and the incidence continues to increase. Dietary interventions and nutritional education are possible options to manage childhood obesity. However, restrictive diets can result in negative outcomes, and therefore it may be more apt to encourage children to consume more fruit and vegetables and thereby develop a healthier positive attitude towards food.

Method A systematic review of literature of interventions to increase fruit and/or vegetable consumption in overweight or obese children and adolescents was conducted, applying a free-text strategy with a set of search terms.

Results A total of five studies describing seven interventions published in international peer-reviewed journals and meeting the review's eligibility criteria were identified. All five studies examined family-focused interventions to increase daily fruit and vegetable consumption measured either by child self-report or parent report. Only one intervention reported a lasting statistically significant increased consumption of fruit and vegetables.

Conclusions This review highlights that in order to tackle obesity narrow interventions focusing on single aspects of behaviour are unlikely to achieve long-term change. Successful public health interventions tackling childhood obesity will need to take a holistic approach and target behaviour change in multiple aspects of children's lifestyles and their surroundings, including nutritional education, parental support and physical activity.

  • OBESITY
  • CHILD HEALTH
  • DIET
  • EXERCISE
  • EDUCATION

Statistics from Altmetric.com

Introduction

Obesity has now become a global epidemic affecting children, adolescents and adults alike.1 In the UK, 31% of boys and 28% of girls aged 2–15 are classed as either overweight or obese.2 Medical conditions related to obesity can start during early life.3 Obese children are at increased risk of becoming obese adults,4 and therefore they are at risk of numerous other medical conditions in later life related to obesity.5–7 Therefore, overweight and obese children are at risk of having a reduced life expectancy.8 Also, the longer medical conditions are present, the greater the risk of complications and associated morbidity, resulting in more days away from work and education.9

The rising prevalence of overweight and obese children is the result of multiple factors, such as increased consumption of energy dense foods and a decrease in physical activity participation.10 Changes in eating habits such as increased consumption of sugar sweetened beverages and high-fat foods accompanied by a decreased consumption of fruit and vegetables are associated with an increasing number of overweight children.11 ,12 Targeting children, particularly overweight children, with nutritional education is important as it helps children form healthier long-term eating habits.13

Obesity management and prevention interventions involve lifestyle modifications such as increasing physical activity participation and dietary changes such as changing unhealthy foods for healthier choices,14 and studies have shown that success can be achieved.15–17 Brown et al18 recommend that dietary changes, such as restricting unhealthy foods and promoting fruit and vegetable consumption, should not be encouraged in isolation but should be combined with physical activity as this approach appears to be effective at preventing children from becoming overweight and helping children already overweight to lose excess weight.

Healthy food promotion interventions could be used as the driving force to enable children and adolescents to develop healthy and balanced attitudes towards food consumption in the long term. This may be more effective and sustainable than using restrictive dieting to achieve short-term weight loss.19

Many dietary interventions used to treat and prevent obesity aim to reduce the intake of high-fat, low-nutrient dense foods, and this may be considered a dietary restriction by people who find these foods reinforcing. This can result in an increased preference for these restricted high-fat foods,20 and therefore people are more likely to revert to previous unhealthy eating habits after structured dietary interventions are removed.21 An alternative approach would be to promote the consumption of healthy high-nutrient dense food, such as fruit and vegetables, within the diet.21 ,22 Even though an increased consumption of fruit and vegetables has not been linked directly with a reduction in obesity, it is assumed that eating more fruit and vegetables could potentially lead to a decreased intake of high-fat and high-sugar energy-dense foods.21 ,23 Fruit and vegetables are an important source of nutrients, and daily intake of fruit and vegetables can prevent chronic diseases, including some cardiovascular diseases and cancers.24

The aim of this study is to review the literature to assess whether dietary interventions can increase the fruit and vegetable consumption of overweight and obese children.

Methods

One author (MB) performed the literature search. PubMed and Ovid MEDLINE electronic search engines were used to search the literature for relevant scientific publications. The search terms were used to search ‘All Fields’. To fully capture the target group, the search terms obese children, overweight children, obese adolescents and overweight adolescents were used. This was combined with the search terms diet intervention and healthy eating intervention to capture interventions involving a dietary component. The search term control was included to capture those studies that included a comparison group. The outcome terms searched for were increase fruit or increase vegetables. The PICO search terms used can be found in table 1.

Table 1

PICO search terms

Inclusion criteria

To be eligible, articles needed to be published in English in a peer-reviewed journal, and report on changes in fruit or vegetable consumption in obese or overweight children or adolescents as an outcome. Studies were only included if they included children who were overweight or obese or where a modifiable risk factor was present, which may be linked to the child's weight (eg, elevated low-density lipoprotein cholesterol, elevated blood pressure) at the start of the trial. Only interventions that included food choices or nutritional components were included, but these did not have to target increased fruit and vegetable consumption as a specific aim. Only studies that measured the children's or adolescent's fruit and vegetable consumption at baseline and at the end of the intervention were included. Only studies that included either a control group or measured the fruit and vegetable consumption of another intervention group for comparison were included.

All the identified study titles from the above search were screened by two authors. Those not deemed relevant were excluded and duplicates removed. The same two authors assessed the abstracts based on the above eligibility criteria, and for included abstracts, the full paper was analysed for eligibility. Disagreements over the inclusion of studies were resolved in the first instance by reaching a consensus through discussion, and, if necessary, remaining disputes were resolved by a final decision being made by the third author (AV). The reference lists of included papers were hand searched for other eligible articles. The quality of the studies included was assessed using the Public Health Critical Appraisal Checklist.25

Results

Summary of included studies

Thirty-five papers were found in the initial search. Four studies satisfied the inclusion criteria for this review (figure 1). One more study was found by cross-checking the references of the included papers to bring the total number of included studies to five.

Figure 1

Flow diagram of the article selection process.

Five papers were included in this study.26–30 The sample sizes were between 26 and 306, with the ages of the children ranging from 4 to 12 years. The studies were either conducted in the USA (n=3)28–30 or Australia (n=2).26 ,27 Four studies repeated their baseline measurements at a 12-month follow-up,26–28 ,30 and one study at a 3-year follow-up.29 Two studies compared an intervention group and a control group.29 ,30 The other three studies compared the outcomes of several different obesity management interventions.26–28 The only groups included were the interventions that involved a dietary component, and therefore seven interventions from five studies have been examined in this review, as two of the included studies incorporated two different dietary interventions.26 ,28 All five studies involved both the parents (or guardians) and the children in the overall intervention; however, in three of the studies,26–28 the nutritional intervention was delivered only to the parents and not directly to the children.

Quality of included studies

The design of the studies was very similar. All five studies were randomised controlled trials. All five studies used questionnaires to measure daily servings of fruit and vegetables. The questionnaires were either completed by the children themselves (n=3)28–30 or by their parents (n=4).26–29 Two studies involved both the parents and the children filling out the questionnaire (see table 2).28 ,29 All the studies achieved similar scores from the Public Health Appraisal Checklist, with the scores ranging from 16/25 to 19/25.

Table 2

Summary of included papers

Outcomes

A total of seven interventions were examined in five studies. All the interventions were aimed at preadolescent children (age range 4–12 years). The majority of the interventions followed children up after 12 months, with one study reporting follow-up after 3 years (see table 2 for details).

Four interventions in three studies produced an increase in the fruit and vegetable consumption after the intervention.27 ,28 ,30 In only one study was the increase statistically significant at final follow-up,30 but despite the failure to reach statistical significance the other two successful studies did record sustained increases in fruit and vegetable consumption at their final follow-up point of 12 months (see table 2).

Two studies, reporting on three interventions, recorded no change in the fruit and vegetable consumption of children,26 ,29 although one of these had a longer follow-up (3 years) than any of the other included studies, and so the lack of a recorded change may be caused by initial changes in food choices reverting after time.

In the three studies involving the youngest sets of children (aged 4–9 years),26–28 nutritional education was only delivered to the parents and not directly to the children themselves. In two of these studies, the children participated in a physical activity intervention.26 ,27

The study reported by Wright et al30 was the only study that targeted changes in more than one setting of the children's lives. The educational component of the study was delivered to both the children and their parents. Wright et al30 reported a statistically significant positive change in the daily servings of fruits and vegetables from baseline. The daily servings of vegetables increased by 1.5 (p=0.03) and the daily serving of fruit increased by 2.0 (p=0.001) compared with the control group that experienced a rise in daily vegetable consumption of 0.4 and a decrease in daily fruit consumption of 0.04. However, 25% of the children included in the Kids Nutrition and Fitness intervention were lost to follow-up at 12 months.

Raynor et al28 and Van Horn et al29 examined the effects of similar family interventions that aimed to change the food choices of both the parents and the children involved. Van Horn et al29 provided both children and parents with nutritional education, while Raynor et al28 only provided educational sessions to parents. The main difference in approach between the two interventions was that Van Horn et al29 promoted a restrictive diet and healthy food promotion in the same subjects at the same time, whereas Raynor et al28 delivered two separate interventions for a restrictive diet and healthy food promotion.

The intervention described by Van Horn et al did not achieve any positive changes in the number of servings of fruit or vegetables eaten per day after 3 years; at baseline an average of 1.1 servings of vegetables and 1.5 servings of fruit were eaten per day, and at 3 years 1.1 servings of vegetables and 1.4 servings of fruit were eaten.29

Raynor et al28 reported on two interventions. Both parents and children were encouraged to make eating habit changes in both the interventions, but only the parents were provided with the educational component. The decrease intervention promoted restrictive dieting to reduce the intake of non-nutrient-dense, energy-dense foods. The increase intervention promoted the consumption of healthy food choices. In the increase intervention arm, a statistically significant rise in the number of fruit and vegetable servings eaten per day was achieved at 6 months (3.3 servings per day) compared with baseline (2.1 servings per day). At 12-month follow–up, the number of servings of fruit and vegetables had fallen from the 6-month peak to 3.1 servings per day, resulting in a non-statistically significant rise in fruit and vegetable consumption at 12 months compared with baseline. In the decrease intervention arm, the same pattern was repeated with a statistically significant rise in fruit and vegetable consumption being achieved from baseline at 6 months and a non-statistically significant rise being recorded from baseline at 12-month follow-up. The number of fruit and vegetable servings being eaten at baseline, 6 months and 12 months was 1.9, 2.3 and 2.4, respectively.

Burrows et al26 and Golley et al27 both reported on the effects of interventions that provided parents with lifestyle education with the aim that the parents should then initiate food choice changes within their families. Burrows et al26 reported on two interventions that included a dietary component. One involved a parent-centred family lifestyle and food choice modification programme (Praise) and the other combined the Praise intervention with a child-centred physical activity programme (Combined). In the Praise intervention arm, the children's daily consumption of fruit and vegetables at baseline was 2.3 and 5.3 servings per day, respectively. At 6 months, the children were eating 3.0 and 5.5 servings of fruit and vegetables per day. At 12-month follow–up, this had fallen to 2.0 and 5.1 servings of fruit and vegetables per day, respectively. In the Combined intervention arm, the children were consuming 2.2 and 4.7 servings of fruit and vegetables per day respectively at baseline. At 6 months, the children consumed 2.2 and 5.0 servings of fruit and vegetables per day, and at 12 months children were eating 2.2 and 4.6 servings of fruit and vegetables, respectively. Even though there was no significant change between baseline and 12-month follow-up, it is noteworthy that the recommended serves per day of fruit and vegetables were already met at baseline and there was no negative change in fruit and vegetable consumption.

Golley et al27 reported that the children in their intervention consumed 1.7 and 2.0 servings of fruits and vegetables per day respectively at baseline. At 6 months, children consumed 2.0 and 2.1 servings of fruits and vegetables per day; and at 12-month follow-up, 2.0 and 2.3 servings of fruits and vegetables were being consumed.

The studies that used self-reporting by the children to measure the servings of fruit and vegetables consumed28–30 recorded greater increases in fruit and vegetable consumption compared with those that relied only on parental reports.26 ,27 This may be explained by a tendency of the children to overestimate and over-report the amount of fruit and vegetables eaten, especially once the children had been told that these were ‘good’ foods to eat.31

Discussion

Overall, the results of the review are disappointing, with only three interventions out of seven reporting statistically significant increases in fruit and vegetable consumptions from baseline.28 ,30 This may reflect the intervention design. The intervention reported by Wright et al30 is the only intervention that targeted changes in the school, home and community settings. This holistic approach, intervening in different aspects of children's lives, may be the reason this intervention had such a positive impact compared with the other interventions. Other interventions using a holistic approach have also been reported as a success in preventing and managing obesity.32 ,33 This may indicate that successful management of childhood obesity requires interventions that not only promote healthy diets, nutritional education and physical activity promotion but also incorporate community support, change of school policies to promote healthy nutrition and physical activity participation and provides parents with nutritional education and support. Children need to not only be provided with the knowledge that they must develop a more healthier lifestyle but they also need to be provided with the opportunities and settings to make these changes, such as increasing the choice of healthy food within school cafeterias and at the same time decreasing the options of high-fat and high-sugar foods.

The most successful intervention30 provided both children and parents with the educational aspect of the intervention, whereas most of the interventions included in this review did not, indicating that this may have been one of the factors contributing to its success at increasing fruit and vegetable consumption. However, the intervention described by Van Horn et al29 also targeted both parents and children with their intervention with disappointing results in vegetable and fruit consumption, suggesting that targeting both parents and children does not guarantee successes in weight management alone, although it has been shown that weight management interventions in adolescents are more successful with the involvement of parents.34 This may be because parents can act as role models to children by exposing them to and encouraging them to consume more healthy food, increasing the likelihood children will consume diets of high nutritional quality.35

In the three studies involving the youngest sets of children (aged 4–9 years),26–28 nutritional education was only delivered to the parents, and not directly to the children themselves. The responsibility with these interventions was firmly placed with the parents to implement and maintain changes in the eating habits of their children. The failure to deliver nutritional education to the children may be one reason for the apparent fall off in increased fruit and vegetable consumption between 6 and 12 months.

Raynor et al28 compared two nutritional education programmes. One involved a restrictive approach to reduce energy-dense foods (decrease) and the other encouraged increased intake of fruit and vegetables (increase). Even though both interventions resulted in a similar reduction in BMI from baseline at 12-month follow-up, the increase intervention group consumed more fruit and vegetables per day although these children also consumed more snack foods and sugar-sweetened beverages per day compared with the decrease intervention group. However, a restrictive diet can result in increased preference for the restricted foods,20 with consumption of the restricted foods becoming more frequent once the intervention has finished.21 There may be signs of this occurring in the increase and increase interventions,28 as the children in these intervention groups consumed more unhealthy food at the 12-month follow-up compared to the 6-month follow-up, suggesting it may be more beneficial to promote healthier eating during childhood to manage obesity than to restrict the diet of children. As both of these interventions were implemented on small sample sizes (n=26), it may be beneficial to repeat this study, comparing restrictive diets with healthy food promotion diets in overweight children.

The age of the children may be another factor that affects the impact of the interventions. Children develop the majority of their food preferences around the ages of 2 and 3 years, and food introduced after the age of four is more likely to be disliked than liked.36 However, the most successful intervention included in this review30 was performed with the oldest set of children. This may reflect older children's ability to understand the nutritional advice better and are able to exercise willpower over their food choices than younger children, for example, choosing to eat foods that they do not necessary enjoy the taste of because they know they are healthy.

Limitations

Limitations of the review and individual studies must be acknowledged. Limitations of the individual studies include using child self-reporting and parental reporting to measure daily servings of fruit and vegetables, small sample sizes, lack of long-term follow-up, a limit of the generalisability to other populations and families that enrolled in these interventions. Some interventions used child self-reporting to measure food consumption and this is associated with a risk of overestimation of consumption.31 In some interventions, food consumption of children was reported by parents only. There may be gaps in the parent's knowledge of what their children have consumed while outside of their care and parents may also report socially desirable responses.27

All but one of the interventions had a 12-month follow–up, and therefore it is unclear whether any positive results will persist beyond this as it can already be seen in some interventions that health behaviours have deteriorated between the 6-month and 12-month follow-up.26 ,28 Some authors indicate that the sample populations investigated in their interventions may not reflect the general population, limiting the generalisability of the intervention.28 ,30

The research identified may be limited in this review by only selecting interventions from published papers in English. The results of this review may also be limited by publication bias. The review may also be limited by not assessing interventions on children with mixed weights that have reported separately how their intervention affected the overweight children.

Conclusion

In order to tackle obesity in childhood and adolescence, narrow interventions focusing on single aspects of behaviour are unlikely to achieve long-term change. Successful public health interventions will need to take a holistic approach and target behaviour changes in multiple aspects of children's lifestyles and their surroundings, including nutritional education, parental support and physical activity.

This review found little evidence focused on increasing fruit and vegetable consumption specifically in overweight and obese children and adolescents. The effects of interventions aiming to increase fruit and vegetable consumption in overweight children should be examined further to determine their capability of producing long-term changes in eating habits. The effects of a restrictive diet compared with a healthy food promotion diet should also be further investigated with an appropriate long-term follow-up of the participants to determine which is more effective at managing childhood obesity.

What is already known on this subject?

  • Childhood obesity is a major public health threat. Increasing consumption of fruits and vegetables is associated with improved long-term health outcomes. It is unclear what interventions are needed to successfully increase the fruit and vegetable consumption of overweight and obese children.

What this study adds?

  • Narrow educational interventions have shown little success in increasing the consumption of fruits and vegetables in overweight and obese children. In order to achieve success, interventions need to address multiple aspects of the children's lifestyles and include parental support.

References

Footnotes

  • Contributors MB and PJW came up with the original idea and conducted the review. MB wrote the first draft of the article. All the authors edited the article. AV is guarantor and advised at all stages of the preparation.

  • Competing interests None.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.