Table 3 Characteristics and findings of garden-based interventions among included reviews
From: Garden-based interventions and early childhood health: an umbrella review
Review | Primary Studiesa | Setting | Countries Included | Relevant Garden-based Intervention Findings | Conclusions of Review |
|---|---|---|---|---|---|
Appleton et al. [49] | 2/77 | Home, school | South Africa, US | Multicomponent interventions reported success at improving vegetable intake and associated outcomes (e.g., selection of vegetables). A multicomponent intervention that included home gardening significantly improved vitamin A status of 2–5y children via production of yellow and dark green leafy vegetables. | There were many barriers to increasing vegetable intake. Although successful interventions have been published, the true value of these, both in cost-efficiency, long-term benefits, and sustainability are yet to be determined. |
Beets et al. [50] | 1/11 | After-school | US | Physical activity: [MA] Positive effect sizes (Hedge’s g: 0.44 [95% CI:0.28–0.60], 6 studies) Herman 2006 effect size: 0.70 (95%CI: 0.05–1.36) | Afterschool programs that include some component of physical activity, can be effective in improving outcomes in children. |
Berezowitz et al. [27] | 1/16 | School | US | Fruit and vegetable consumption: 71% of studies measuring fruit and vegetable intake reported significant improvements. Academic outcomes: Of 4 studies measuring academic outcomes, 3 showed improvements in science achievement and only 1 showed improvements in math scores. | School-based garden interventions improved or maintained both fruit and vegetable consumption and academic performance. Schools should consider school gardens as a hands-on tool to enhance science learning and potentially improve long-term fruit and vegetable consumption. |
Berti et al. [51] | 5/30 | Home | Bangladesh, Vietnam, Guatemala, Thailand, Philippines | Nutrition outcomes improved in 11/13 home garden interventions. Of 17 projects ranked high or mid, 9 were home garden programs that aimed to improve nutritional status. All 9 home garden projects included nutrition education, often with another public health intervention. Of home garden interventions, 16/19 indicators were better in the intervention group. | Home gardening projects were more successful than other intervention types, perhaps because they are easily adoptable and may strengthen human capital. Almost all home gardening projects incorporated gender considerations, which may have partly been responsible for the positive effect on child nutrition outcomes. |
Bhutta et al. [52] | 6/29 | Home | China, Vietnam, Iran, Laos, South Africa, Thailand | Included studies were found to have a positive effect on agricultural production and dietary intake. The 4 studies evaluating impact on nutritional status found a positive effect. | Although some agricultural interventions are potentially promising and culturally relevant, there is not enough evidence to suggest that these dietary diversification strategies are effective in improving nutritional status or micronutrient indicators on a large scale. |
Bird et al. [53] | 2/6 | Home | India, Nepal | Dietary quality: 7 interventions reported improved dietary quality and diversity. Anthropometry: 4 papers reported a lack of convincing evidence | There was not strong evidence that agricultural interventions impacted final measures of nutritional status, the potential of agricultural interventions to improve intermediate outcomes provides support for continued research in this area |
Davis et al. [54] | 2/13 | School | US | Garden programs resulted in improved attitudes towards, willingness to taste, identification of and self-efficacy to prepare and cook fruits and vegetables Vegetable consumption: 6/10 programs examined dietary intake, 6 found increased vegetable intake; 4 showed no effect Vegetable preference: Almost all (7/8) that measured found increased preference | The present analysis showed clear and consistent effects of school garden programs on improved dietary behaviors, with half of the studies showing increases in vegetable intake. Transition from start-up to long term maintenance of garden initiatives is an area of further work to enhance sustainability and thus the duration of health effects. |
Hendrie et al. [55] | 2/22 | Home and community | US | Vegetable consumption: 12/22 studies conducted in home or community settings were effective short-term, with mean short-term change 29% (about ¼ to ½ serving of vegetables); 6/10 were effective long-term (6 + months) | Interventions targeting vegetable intake in home or community settings are generally effective and may increase intake by around 30%. Intervention effectiveness was associated with number of settings targeted and frequency of contact, but not length of intervention. |
Hodder et al. [48] | 2/63 | Child care centers, community | US | Fruit and vegetable consumption: Both trials reported a positive intervention effect. [MA] Multicomponent interventions vs. no intervention had a small effect (SMD 0.34, 95% CI 0.10 to 0.57; 9 trials, 3022 participants; moderate-quality evidence), equivalent to an increase of 0.36 cups of fruit and vegetables per day. Data were insufficient to assess long-term effects. | Child-feeding practice interventions may lead to, and multicomponent interventions probably lead to, small increases in children’s intake of fruit and vegetables in the short term (less than 12 months). However, the quality of evidence is low and effect sizes may be too small for clinical utility. |
Langellotto et al. [56] | 2/20 | School, after-school | US | Vegetable consumption: [MA] Gardening: significant increase [E++ = 0.42, CI = 0.07 to 2.07, df = 3]; Control: no significant effect Nutrition knowledge: [MA] Gardening: No significant effect [E++ = 0.21, CI = −1.19 to 0.43, df = 2]; Control: significant increase [E++ = 0.23, CI = 0.04 to 1.02, df = 2]Fruit consumption: [MA] Gardening: significant increase [E++ = 0.08, CI = 0.02 to 0.12, df = 1]; Control: no significant effect Fruit preference: [MA] Gardening: no significant effect [E++ = −0.02, CI = –0.20 to 0.01, df = 3]; Control: no significant effect Vegetable preference: [MA] Gardening: significant increase [E++ = 0.10, CI = 0.01to 0.19, df = 1]; Control: no significant effect | Gardening interventions had more positive significant effects (including both pre- and post-test comparisons and comparison to control groups) than nutrition education interventions or control conditions. These types of programs should receive more funding for rigorous research, including federal funding. |
Masset et al. [57] | 5/23 | Home, school | South Africa, Lesotho, Thailand, Cambodia | Diet composition: Most studies (19/23) reported a positive effect. With few exceptions, home garden programs increased fruit and vegetable consumption. Hemoglobin concentration: No statistically significant difference. Vitamin A intake: [MA]: Effect of interventions on serum retinol: SMD = 2.42 (95% CI 1.97 to 2.16, 4 studies, fixed-effects model) Child nutrition status: 1 study found a significant effect on stunting prevalence; 3 studies found positive effects on prevalence of underweight; 2 found a positive effect on wasting. | Meta-analysis provides support that home gardens interventions improve vitamin A intake among children <5. Though results provide little support that agricultural interventions reduce undernutrition, this should not be interpreted as absence of an effect. Lack of significance can be the result of absence of effect or of absence of statistical power, and many studies reviewed included small samples of children. |
Mikkelsen et al. [58] | 3/26 | Child care centers, Kindergartens | US, Thailand, Germany | Fruit and vegetable consumption: 6 multicomponent interventions showed a significant increases; 1 found an effect on only fruit consumption after 1 year follow up. Six of the educational only studies showed promising, although non-significant, results. Anthropometric: Educational and multicomponent interventions did not reveal a significant effect on BMI. | Healthy eating interventions in preschools can significantly increase child fruit and vegetable consumption and nutrition-related knowledge if using an educational or multicomponent intervention. Results highlight the scarcity of properly designed interventions with clear indicators and outcomes. Preschool settings could also help decrease child health inequities. |
Nekitsing et al. [59] | 2/30 | Child care centers, Kindergartens | US, Thailand | Vegetable intake: [MA] Effect of intervention vs. comparison on vegetable intake: Hedges g = 0.40 (95% CI: 0.31 to 0.50, random effects model, 30 studies, 4017 participants). Funnel plot test suggest that publication bias is present. Subgroup analysis showed intervention effect sizes varied significantly by study design, outcome measure, recipient, strategy, and type of vegetable. | The most successful strategies included taste exposures and reward. Less effective strategies included food services and nutrition education. Meta-regression revealed the more exposure to a vegetable a child receives, the more likely they are to increase intake of that vegetable. Preschoolers may be more amenable to these interventions than older children, therefore early intervention is key. |
Ohly et al. [31] | 2/18 | Child care centers | US | Fruit and vegetable intake: 2 studies found significant increases Fruit and vegetable preference: 8/13 interventions reported statistically significant effects Food knowledge and attitudes: Most (7/10) studies found positive effects in the intervention groups. Physical activity: Children in gardening groups reported being less sedentary; spent more time engaged in “moderate” activity compared to control group. Diastolic blood pressure was the only significant effect, which decreased more in the intervention group. | There was little objective evidence for changes in eating habits and physical activity. Stronger evidence supported improvements in knowledge, attitudes and preferences towards fruits and vegetables. Quantitative evidence for health and well-being impacts of school gardening are limited by self-report bias, imperfect measures of fruit and vegetable consumptions, heterogeneity of measurement scales and follow-up time. Students who do not excel in classroom activities were thought to particularly benefit from garden-based interventions. |
Savoie-Roskos et al. [30] | 3/14 | Child care centers, school, community | US | Fruit or vegetable consumption: Most articles found improvements after implementation of a gardening intervention. Two studies found that although vegetables consumption at school increased, vegetable consumption at home did not change. Children who received gardening and nutrition education had greater effects when compared to education-only and control groups in 2/3 studies. | Multicomponent garden-based interventions may be effective in increasing fruit and vegetable intake. Garden interventions increased access to fruits and vegetables during the school day. However, children may have limited access at home, resulting in minimal changes over time. |
Sisson et al. [60] | 3/71 | Child care centers, preschools, schools | US, Australia | Most studies were successful in promoting change in obesity or obesogenic behaviors and had the intended effect on the target: obesity 48% (14), physical activity 73% (30), diet 87% (39), and screen time 63% (5). | Environment-level interventions had less impact on child health behavior outcomes than those that specifically included child-level interventions. Child care center environment interventions that included technical support facilitated positive changes. |