OUP user menu

Economic incentives and nutritional behavior of children in the school setting: a systematic review

Jørgen Dejgård Jensen, Helene Hartmann, Anika de Mul, Albertine Schuit, Johannes Brug,
DOI: http://dx.doi.org/10.1111/j.1753-4887.2011.00422.x 660-674 First published online: 1 November 2011

Abstract

The aim of the present review was to examine the existing literature on the effectiveness of economic incentives for producing sound nutritional behavior in schools. Studies published in the English-language literature that included baseline and/or outcome data regarding food and beverage intake of schoolchildren were eligible for inclusion. A systematic search of the literature was conducted to identify relevant primary studies and relevant systematic reviews of primary studies. Altogether, 3,472 research publications were identified in the systematic search, of which 50 papers were retrieved. Of these, 30 publications representing 28 studies fulfilled the criteria for inclusion. The studies addressing price incentives suggest that such incentives are effective for altering consumption in the school setting. Other types of economic incentives have been included in combined intervention schemes, but the inclusion of other intervention elements makes it difficult to draw conclusions about the effectiveness of the economic incentive instruments per se in these studies.

  • child behavior
  • economic incentives
  • effectiveness
  • feasibility
  • nutrition
  • school setting

INTRODUCTION

Childhood obesity and, thus, overnutrition are arguably major concerns in developed countries and, increasingly, in the developing world as well. Overnutrition in children and the related obesity epidemic have been attributed to what has been labeled the “obesogenic environment”,1,2 i.e., an environment in which palatable high-energy foods are almost always readily available, accessible, and affordable. A large number of studies on environmental interventions to prevent obesity and promote healthy eating and physical activity have been conducted and reviewed in several meta-analytical studies addressing the population in general36 as well as children and adolescents in particular.715 It has been argued that changes in the economic environment should be considered to promote healthier eating and physical activity as ways to help curb the obesity epidemic.16,17

Economic motives, in terms of relative prices of energy-dense versus less-energy-dense foods, as well as the relation between food prices and incomes, affect the demand for foods and beverages.18 In the school setting, economic motives may also influence the availability of foods and beverages offered to schoolchildren.1923

In a literature review related to preventive health behavior, Kane et al.5 found that economic incentives work in the majority of studies, but the success rate depends on the goal of the incentive, with relatively simple incentives being the most effective. Marteau et al.6 found evidence for the effectiveness of financial disincentives (higher prices on unhealthy commodities) to be clearer than that for positive incentives (lower prices on healthy commodities). Several of these reviews found stronger evidence for short-term effects of economic incentives than for long-term effects.

Despite the substantial number of literature reviews on intervention strategies to promote healthy nutrition and physical activity, the literature on economic incentive instruments in this area is relatively scarce, even though the issue of economic incentives has attracted a fair amount of interest in recent years.1625 These reviews suggest that the school setting may be suitable for promoting health strategies targeted at children and adolescents. They also provide some evidence of the effectiveness of economic incentive strategies for promoting healthy nutrition. The use of economic incentives for nutrition-improving interventions in the school setting, however, has generated only scattered evidence and – to the authors' knowledge – has not been the subject of any previous systematic reviews.

As several studies have demonstrated the potential effectiveness of economic incentives in other areas of health promotion, and as only a limited number of studies have investigated the use of economic incentives to promote healthy dietary behavior in children, the objectives of the present study were as follows: 1) to review the existing literature on economic incentives of importance to children's dietary behavior in schools; 2) to evaluate the potential of such incentive interventions; and 3) to derive experiences that may be useful for the design of new economic incentive instruments. A systematic review of the scientific literature reporting nutrition-oriented intervention studies that involve economic incentives in schools was conducted. The instruments included incentives for the children and their families to exhibit healthy nutritional behavior as well as incentives for the schools, the authorities, and the schools' surroundings to establish an environment supportive of sound nutritional behavior in children. A corresponding review of economic incentives to affect the physical activity behavior of children is provided elsewhere.

ANALYTICAL FRAMEWORK

Economic incentives are defined as any factor (financial or semifinancial) that changes (monetary and nonmonetary) costs and thereby enables or motivates one particular course of action relative to another. Applied to the dietary behavior of children, i.e., the intake of different healthy and unhealthy foods and beverages, this means that economic incentives are measures that decrease the costs of exhibiting healthy dietary behavior relative to the costs of an unhealthy diet.

Economic incentives may take many forms, either in terms of reinforcing behavior (rewarding desired behavior) or enabling behavior (removing barriers to desired behavior). Commodity prices constitute one well-known set of economic incentives. For example, a lower price of, for example, healthy versus unhealthy foods, snacks, or drinks provides the individual with an economic incentive to eat more healthy products and fewer unhealthy products. Economic incentives may also take the form of rewarding desired behavior (e.g., a gift or bonus received upon reaching specified behavioral or dietary goals) or penalizing undesired behavior.

Economic incentives can affect the nutritional environment in several ways.26 In decisions involving cognitive efforts, price incentives, together with other factors, influence the individual's decision-making process directly. Economic incentives may also have implications for other environmental factors that children and adolescents encounter at school. Economic incentives can play a role in automatic decision processes, affecting, for example, whether individuals react to special offers in an automatic manner, an issue being addressed increasingly in the literature on behavioral economics.27 The physical influence (e.g., availability of healthy and unhealthy foods at or close to schools) may, to a large extent, be influenced by economic incentives for the school or for surrounding businesses that sell food (physical environment). The set of rules applied at the school can also be influenced by, for example, funding that is conditional on the existence of a nutrition/physical activity policy (political environment). Underlying mediators (such as attitudes, norms, and intentions) in this decision-making may also be influenced by economic incentives (e.g., high prices for healthy foods may have a negative, or a positive, impact on attitudes related to healthy eating).

Finally, economic incentives may influence decision processes if, for example, such incentives, which represent extrinsic motivational factors (positive or negative), affect intrinsic motivational factors. For instance, an individual's awareness, involvement, or habits may be influenced by economic incentives. These issues are addressed, as far as possible, in the present study.

REVIEW METHODOLOGY

This review focuses on interventions that aimed to improve the diets of children using economic incentives, which could be directed toward the child, the parents, the school, the local government, private companies, or nongovernmental organizations. Interventions that aimed to improve the diet separately and interventions that had a combined focus on diet and physical activity are included, provided economic incentives were expected to play a role. Studies are included if they evaluated dietary behavior – measured as the intake of relevant foods, beverages, and snacks – or the availability of healthy foods and beverages in schools. Primary studies since 1990 (any type of study) that focused on nonobese children aged 10–12 years have mainly been used (other age groups of children/adolescents are also included due a low number of studies fulfilling all these criteria). However, since children, in general, depend on their parents to pay their expenses, parents are also included in the study population, with a focus on economic incentives directed at parents to encourage healthy diets in their children. Because schools play an important role in the nutritional behavior of children, studies on economic incentives that motivate schools to undertake nutrition-promoting activities or to provide a healthy school food environment have also been included.

The following publication databases were used for the literature search: ScienceDirect, Cochrane Database System, CSA Illumina, Google Scholar, PubMed (Medline), and Social Science Research Network. These databases were searched for combinations of the following search terms: [“cost”or“econom*”]and[“food”or“lunch”or“snack”or“fruit”or“vegetable”or“diet”or“nutrition*”]and[“school”or“children”or“adolesc*”]. Each group of terms represented the following three dimensions of the subject, respectively: economic aspects, nutritional aspects, and the school setting.

A preliminary assessment of the relevance of publications identified in these searches was based on a review of the publication titles, which was undertaken by one researcher and verified by another. In the second stage, the selection of publications for in-depth review was based on review of the published abstracts. This was performed independently by two researchers. The publications selected were then retrieved and reviewed in depth to determine whether they fulfilled the above-mentioned inclusion criteria and to extract the relevant results. The design and quality of the studies were assessed, as were the feasibility and effectiveness of the interventions implemented. Specifically, the quality of the studies for the purposes of this review was determined on the basis of the following criteria. Studies in which conclusions could be justified from the analysis conducted were assigned grade I. Studies which, in addition to this, also contained clear definitions of intervention and correctly performed analyses were assigned grade II. Studies that furthermore contained accurate and standard definitions of appropriate outcome variables were assigned grade III. Finally, studies that additionally contained comparable intervention and control groups and an appropriate follow-up were assigned the highest grade (IV).

RESULTS

The search resulted in more than 4,000 titles (including multiple hits of the same paper); from these, 104 were selected for abstract review. Based on the abstract review, as well as backward searches and other references, 50 publications were reviewed in depth. Of these, 28 studies, reported in 30 publications, fulfilled the criteria sufficiently for inclusion in the systematic part of the review (Figure 1).

Figure 1

Diagram showing the process of the literature search and selection.

The findings from the literature review are summarized in Table 1.

View this table:
Table 1

Overview of the studies reviewed (studies are presented in alphabetical order and separately for direct and indirect comparison of the effects of economic incentives.

ReferenceType of studyGradeCountry, period of studyParticipantsIntervention/explanatory variablesOutcome measuresConclusion
Effects of direct economic incentives
Bere et al.28Baseline (2001) + intervention + follow-up studyIVNorway, 2001–2002795 7th graders (13–14 years) from 38 schoolsFree fruit or vegetables every school day versus fee-based school fruit for 8 monthsSelf-reported intake of fruits and vegetablesFree fruit or vegetables an effective strategy to increase schoolchildren's fruit and vegetable intake. Free fruit increased mean intake by almost one portion per day
Bere et al.29Baseline (2001) + intervention + follow-up (2002, 2003) studyIVNorway, 2001–2003517 6th graders from 19 schoolsFree fruit or vegetables every school day, followed by user-paid fruit/vegetable program versus no fruit/vegetable programSelf-reported intake of fruits and vegetablesHigher intake of fruits and vegetables (approx. 0.5 portion per day) due to intervention sustained after 1 year
Bere et al.30Baseline (2001) + intervention + follow-up (2002, 2005) studyIVNorway, 2001–20051,950 pupils (6–7th grade)Free fruit or vegetables every school day versus fee-based or no school fruit/vegetable programSelf-reported intake of fruits and vegetablesHigher intake of fruits and vegetables due to intervention sustained to some extent after 3 years (approx. 0.4 portion per day)
Epstein et al.31Pilot study, price simulation experimentIVNew York, United States10 children aged 10–12 years and their mothersPrice manipulations of fruits/vegetables and energy-dense snack foodsPrice elasticities, relationship between mothers' and children's purchasesPrice elasticity between −0.74 and −0.5 for both children and mothers. Parental food choice and purchasing behavior may play a role in the development of children's purchasing of both healthy and unhealthy foods
Epstein et al.32Price simulation experimentIV32 children aged 10–12 years + 20 children aged 10–14 yearsPrice manipulations of fruits/vegetables and energy-dense snack foods, and manipulation of total budget availablePurchase of fruits/vegetables and energy-dense snack foods. Price elasticitiesPrice elasticity between −2.0 and −1.0. Price elasticitiy decreases (in absolute terms) with available budget
Epstein et al.33Price simulation experimentsIVNew York, United States47 mothers aged 25–50 yearsPrice manipulations of LED and HED foodsMothers' food purchasesPrice changes may be a way to influence food purchases of LED compared with HED foods. Price elasticity −1.6 for HED foods and −0.6 for LED foods
French et al.34Cafeteria sales data, baseline + intervention + follow-upIIITwo high schoolsLow-price interventions (approx. 50% reduction) on fruits, baby carrots, and salad servings; 3-week baseline period, 3-week intervention, and 3-week return-to- baseline periodSales of fruits, baby carrots, and salad servings from cafeteriaFruit sales increased by approx. fourfold and carrot sales by about twofold during the low-price period and remained higher in the 3-week post-intervention period. No significant changes in total dollar sales for a-la-carte purchases
French et al.35Effects of sales data, pricing, and point-of-purchase promotion on sales of low-fat and regular vending snacksIIIMinnesota, United States12 secondary schools and 12 worksitesFour levels of pricing and three levels of promotionSales of low-fat snacks from vending machinesLowering the prices of low-fat vending snacks had a strong effect on sales of low-fat snacks from vending machines at diverse worksites and secondary schools, corresponding to a price elasticity of between −1.9 and −1.0
Hannan et al.36Sales and revenue data from a high-school cafeteriaIIIPrices of three high-fat foods were raised and prices of four low-fat foods were lowered in a high-school cafeteria over 1 school yearData collected on food sales and revenues regarding targeted foodsResults supported the feasibility of a pricing strategy that offered low-fat foods at lower prices and high-fat foods at higher prices
Gentile et al.,37 Eisenmann et al.38Four-stage RCTIIMinnesota & Iowa, United States, 2004–20061,323 children (3rd–5th grade)Creation of a supportive environment for healthy behaviors (physical activity, sedentary behavior, nutrition) by providing knowledge, goal-setting advice, and tips on physical activity. Goal and activity points were introduced in a lotterySelf- and parent-reported consumption of fruits and vegetablesThe experimental group showed a significant increase in parent-reported fruit and vegetable consumption (1.2 extra servings per week) after 6 months and a marginally significant increase in child-reported fruit and vegetable consumption (0.1 serving/day). Effect influenced by sex of the child and family involvement
Horne et al.39RCT and 12-month follow-upIVIreland435 children aged 4–11 yearsFood Dudes heroes as role models. Reward to children for tasting/eating both fruits and vegetables during a 16-day intervention. Follow-up after 12 monthsFood intake during morning and lunch break measured by researchersAfter 1 year, children in experiment group consumed more fruit and vegetables (55 g/day) than control group
Horowitz et al.,40 Shilts et al.41Pilot studyICalifornia, United States34 middle school students aged 11–15 yearsEatFit has three components: workbook, web-based assessment, and classroom curriculum. Raffle tickets and prizes for goal attainment used as incentives to motivate studentsSelf-reported dietary and physical activity recordsPositive changes in dietary behavior and self-efficacy (29–56% of participants) and positive change in physical activity behavior and self-efficacy
Lowe et al.42RCTIIIEngland & Wales, United Kingdom402 children aged 4–11 yearsFood Dudes heroes as role models. Reward to children for tasting/eating both fruits and vegetables during intervention. Follow-up after 12 monthsVisually estimated food intake during morning and lunch break, parent-reported consumption at homeConsumption of fruits and vegetables – and children's liking of these foods – increased during intervention: 131–153 g/day on weekdays, no clear effect on weekends
Sallis et al.43RCTIIICalifornia, United States, 1997–199924 middle schools (6th–8th grades), mean enrollment 1,109Combined interventions addressing physical activity and nutrition, including healthy lunch contest, in which students observed to have low-fat lunch were provided raffle tickets. Schools received financial support for physical education equipment, kitchen equipment, and physical activity programsAmount of physical activity, quality of lunch, 24-hour diet recall, 7-day physical activity recallIntervention had significant effect on physical activity for boys, but not for girls, and no significant effect on nutrition
Smith & Epstein44Laboratory experiments, with time representative of user costIV6 obese children aged 10 ± 1.5 yearsFive different settings for scores necessary for conversion to low-calorie or high-calorie foodsAllocation of time for obtaining scores to convert to low-calorie versus high-calorie foods, respectivelyFood choice is the result of a combination of hedonic ratings and costs
Effect of indirect economic incentives via other environmental factors
Ard et al.45Cross-sectionalIIIAlabama, United StatesHomes of 1,355 childrenPrice of fruits, vegetables, income, race, genderPresence of up to 27 fruit or vegetable itemsHome-availability of fruits and vegetables was negatively associated with price (estimated price elasticity −0.2)
Briley et al.46Cross-sectionalITexas, United States9 childcare centersIdentification of factors that influence the menuQuality of childcare menusFactors with a strong influence on menus include costs
Brown & Tammineni47Sales dataIIIMississippi, United States15 schools, 2005–2006Passive marketing, 10–25% price discounts, changing types and proportions of beverages offeredSales of carbonated drinks, sports drinks, fruit juice, and water from vending machinesIncreased profits reported in 12 of the intervention schools. No conflict between intervention and economic incentives
Cullen et al.48Quasi-experimentalIIITexas, United States, 2006–20072,000 (intervention) + 1,600 (control) high-school studentsUSDA free fruit and vegetable program, one free fruit per student per day. Selection dependent on availability and costSurvey data on exposure and likingsAccess to free fruit and vegetable program did not improve fruit and vegetable exposure or preferences reported by high-school students
Harris et al.49IIIKansas, United States136 4th gradersCombined intervention to increase physical activity, modify school lunch, and provide nutritional education, the last one including economic incentives for teachersKnowledge survey, fitness skills, attitudesMore reduced-fat dishes on school lunch menu (from a mean of 38–40% of energy intake from fat to 30%), better knowledge and skills among students, improved physical fitness
Honisett et al.50Pilot studyIVictoria, Australia62 primary schools and 84 early childhood servicesAward program in which schools received an award upon reaching specified goals regarding food environment and physical activitySchool participation in program26% of schools had joined the program after 4 weeks
Mitchell et al.51Baseline and two follow-ups (3 and 12 months)IIIHypercholesterolemic children aged 4–10 yearsLow-fat diet versus no-intervention dietCalculated food costs based on 24-hour dietary recallsAmong children adhering to a low-fat diet, there was no increase in food costs
Montgomery et al.52ITexas, 1991–199424 school cafeteriasEat Smart School Nutrition Program, designed to lower the fat, saturated fat, and sodium in school mealsAverage school lunch and breakfast meal costsDecreasing the fat in school meals does not significantly affect the long-term food costs of either lunch or breakfast menus in schools
Nicklas et al.53Cross-sectionalIIIUnited States597 9th graders aged 14–15 yearsThree breakfasts: fast-food, ready-to-eat cereals, other foodsNutrient intake: total, per kcal, per $Ready-to-eat cereals does not conflict with economic incentives, but provides more energy, proteins, and fiber than fast-food breakfasts
Probart et al.54Cross sectionalIPennsylvania, United States228 school food-service directors, randomly selected from high schoolsFactors that predict sales of competitive foods and participation in school lunch programA la carte sales, vending machines per student, daily participation in school lunchPercentage of students eligible for free or reduced-price meals is an important determinant for a-la-carte sales and average daily participation in school lunch and, hence, for economic outcome
Rainville55Cross-sectionalI6 US Midwestern statesHigh schoolsComparison of school lunch with lunches from home and fast-food lunch mealsCost per mealSchool lunches were significantly less expensive than fast food and lunches from home
Raynor et al.56Baseline and two follow-ups (6 and 12 months)IIIUnited States31 families with obese children aged 8–12 years20-week behavior modification interventionEnergy intake, energy composition, servings, daily food costSignificant decrease in energy intake (around 500 kcal/day), increase in nutrient density, and decrease in daily food cost (around $1.7/day) after 1 year. No conflict between nutritional goals and economic incentives
Wagner et al.57Cross-sectionalIIIMinnesota, United States, 1999–2004330 school districtsImproved nutritional quality of school mealsMeal sales, costsSales do not necessarily decrease and costs do not necessarily increase when meals are healthy
  • Abbrevations: HED, high energy density; LED, low energy density; RCT, randomized controlled trial.

Relevant outcome measures

The outcome measurements in the empirical literature on children's (and their parents') decision-making regarding the relationship between children's nutritional behavior and economic incentives in schools fall into different categories: food choice observations from controlled experiments (representing intake), self-reported intake with changed economic incentives, observed sales data (also a measurement of intake in schools), and intake data measured directly by researchers. The outcome measurement of the effect of the physical and political environment on the nutritional behavior of schoolchildren is less clearly defined, resulting in more scattered evidence that is indicative rather than conclusive.

In a few studies,39,42 children's food intake was measured by researchers, either by visual inspection or by weighing the individual food components. Such a procedure enables relatively precise measurements to be taken, but it also demands considerable resources. Some studies were based on participants' self-reported intake of commodities that were offered within school-based incentive-oriented interventions.2843 Although self-reporting involves a risk of erroneous (or even biased) reporting, it has the advantage of assessing interventions in the school setting while still maintaining the possibility of linking individual response data to relevant background information about the participants.

In a number of simulation experiments,3144 participants were given a budget and told to allocate the amount between different goods, constrained by varying price and budget conditions. Outcomes were measured by observing the participants' choices. These kinds of experiments attempt to estimate parameters for the participants' preferences and budget allocation behavior (such as sensitivity of demand to price changes), and it is possible to associate these parameters with, for example, sociodemographic variables. It should be noted that results from such studies are based on (relatively few) observations from an experimental setting, using a budget that does not necessarily reflect the participants' normal budget for such purchases and without the influence of peers, etc. Hence, the estimates of effectiveness obtained from such simulation experiments may represent a biased estimation of the effectiveness in a real life setting. Another approach is to monitor sales data from points of food sale, such as school cafeterias or vending machines, to examine different price configurations as an approximation of the average intake at school.3436 This approach reflects children's actual behavior at school and requires little extra effort. However, it is normally not possible to link these purchase data to, for example, the sociodemographic characteristics of the children. One drawback of this approach is that the results from such studies regarding 10–12-year-olds is very limited, probably because children in this age group are less likely to bring money to school than adolescents.

By nature, the measurement of indirect impacts of economic incentives via other environmental factors is more diversified in the literature. Regarding the influence of economic incentives on the physical school food environment, the range of products offered from school food suppliers or vending machines may constitute a basis for outcome measurement, and so may, for example, the number of soft drink vending machines at a school.55 Suppliers' economic incentives to supply nutritious foods and beverages may constitute another, more indirect, measure of the economic framework that determines the school food environment, represented, for example, by differences in costs4651 or profits between healthy and less healthy foods and beverages. For intervention programs that address schools as a whole, one outcome measure could be the schools' participation rates,50 which represent the schools' policies rather than the actual school food environment. In general, these measures only provide indirect indications of the school food environment and should thus be considered as indicative, rather than conclusive.

Direct effects of economic incentives relevant to children in schools

In general, the studies reviewed support the hypothesis that the choice of foods, snacks, and beverages by schoolchildren can be influenced by economic incentives. In an attempt to investigate the direct effect of prices on individuals' trade-offs, Epstein et al.31 found significant price responsiveness in children's choice between fruit/vegetables and energy-dense snack foods. Significant price responsiveness was found in choices made by the children's mothers, and the children's choices between healthy and unhealthy foods were significantly related to their mothers' choices. In another price-simulation study, Epstein et al.33 found that prices had a significant influence on mothers' purchases of high- and low-energy-density foods, with an asymmetry in cross-price effects, so that a higher price of highly energy-dense foods significantly increased the purchase of foods with low energy density. However, a lower price of low-density foods was not associated with a significantly reduced purchase of energy-dense foods. Epstein et al.32 found that price sensitivity decreases with a larger budget, and Smith and Epstein44 showed that both subjective (liking) and environmental factors (availability) contribute to children's choice of foods. These simulation-based studies support standard economic theory and suggest that price incentives may be effective in children as well.

A number of studies have investigated the effects of price incentives in combination with changes in the physical availability of certain foods. Bere et al.28 investigated the interaction of the economic and physical environment by studying the effects of two different fruit/vegetable programs at the school level. By the end of the intervention period, they found that students at schools with a free fruit/vegetable program had a higher intake of fruit and vegetables than students at schools with a paid scheme; the latter students, in turn, had a higher intake than students at schools with no school fruit/vegetable program. Children also reported a lower intake of soda/candy/chips at schools with (free or paid) school fruit/vegetable programs. This suggests that both availability and cost may have an influence on intake. The effects of free school fruit on students' intake of fruit and vegetables were found to be sustained for 1 year29 and 3 years30 after the intervention.

French et al.34 found that increasing the diversity of offerings and halving the price of fruits, carrots, and salad servings led to a fourfold increase in the sale of fruit, a twofold increase in the sale of carrots, but no significant change in the sale of salads in high-school cafeterias. If sales are considered a measurement of intake, this combined intervention had a significant effect on the intake of fruit and carrots. At the same time, the number of meal-pattern customers, as well as total dollar sales for a la carte purchases, did not change significantly. With a similar research design, French et al.58 investigated the influence of price on sales of low-fat snacks from university vending machines. They found that sales of low-fat snacks increased by approximately 80% during the low-price intervention but returned to baseline levels during the post-intervention period. As a refinement to this study, French et al.35 conducted an experiment to address the effect of price changes and promotional activities on the sale of low-fat snacks from vending machines at secondary schools and workplaces. They found that price reductions had a significant effect on the sales of these snacks, but the study also revealed a threshold effect in that only major price reductions (more than 25%) had a significant effect. Hannan et al.36 examined a combination of increased prices for high-fat foods and lower prices for low-fat foods in high-school cafeterias and also found support for a similar effect of such price changes. Other studies related to an adult population at university cafeterias23,59 and in a restaurant15 also generated comparable results.

In existing empirical studies, direct economic incentives aimed at schoolchildren mainly take the form of enabling incentives in terms of price differences between nutritious and non-nutritious foods and the subsequent demand responses to changes in prices. However, other types of economic incentives exist, and some of them have been used as reinforcing or supporting mechanisms in multifaceted behavioral intervention studies. Such incentive schemes include rewards (e.g., pens, balls, stickers, or participation in a raffle) upon desired behavior. For example, in the SWITCH study,37,38 a set of educational and awareness-enhancing instruments aimed at increasing children's physical activity and intake of fruit and vegetables and reducing their sedentary behavior was implemented. Upon achievement of stated weekly goals in the three areas, participants earned enough “SWITCH Points” to participate in a lottery. The intervention had a beneficial effect on the intake of fruit and vegetables and on sedentary behavior but did not have a significant effect on physical activity. The EatFit intervention41,42 guided students through an assessment of their nutritional behavior, their physical activity, and their personal goals. The project used economic incentives, in the form of raffle tickets and prices, to motivate students to attain their goals. In the studies by Lowe et al.42 and Horne et al.,39 British and Irish children watched video adventures featuring the heroic “Food Dudes” and received small rewards (e.g., erasers, pens, and pencil cases) for tasting fruit and vegetables during the intervention, which also comprised a number of other elements. Relative to the baseline, consumption of food provided by the school increased during the intervention, whereas it declined significantly in the control school. These intervention studies combine extensive educational/informational activities with relatively modest reinforcing economic incentives, but since these are closely integrated, it is not possible to distinguish the contributions from the separate elements and, hence, to assess the partial effectiveness of the economic incentives applied.

Indirect impact of economic incentives via other environmental factors and moderators

Economic incentives may indirectly influence nutritional behavior via the physical environment at schools in terms of incentives to provide a nutritious food environment and via the political environment in terms of incentives to undertake (and emphasize) nutritional education and encouragement activities at schools.

With regard to economic incentives to modify the school food environment, two elements of the food environment are considered important: the schools' own supplies of nutritionally approved meals, and the supply of competitive foods. For foods or beverages supplied at schools, it is relevant to investigate the vendor's economic incentive to maximize the nutritional quality of the foods/drinks supplied. Briley et al.46 investigated factors that influence menus at childcare centers and found that, in addition to production costs, the most influential factors included food program requirements, staff perceptions of children's food preferences, and history of the food program at the center. Conversely, a study by Mitchell et al.,51 which investigated the additional costs of supplying a low-fat diet relative to a control group diet, found that the costs of a low-fat diet were not significantly higher than the dietary costs of the control group, a conclusion also reached in other studies.5257 A study by Cullen and Watson60 showed that a stricter linkage between nutritional requirements in the Texas Public School Nutrition Policy and reimbursements from the US National School Lunch Program improved the foods served or sold to students at school. This suggests that the economic incentive embedded in the reimbursement was significant for the foods served, compared to the additional costs of improving the nutritional quality of the foods. Regarding the availability of competitive foods, Probart et al.54 found that the number of vending machines per student in Pennsylvanian schools was significantly and positively associated with student enrollment, number of less nutritious food items, and the existence of soft drink machines owned by soft drink companies (for which the school received a percentage of sales). This indirectly suggests that economic considerations influenced the number of vending machines. In an intervention study that investigated increased availability and lower prices of healthy soft drinks in vending machines, half of the participating schools experienced increased beverage profits, whereas only one-quarter of the participating schools reported lower beverage profits.47

With regard to economic incentives for nutrition-supportive school policies, Honisett et al.50 conducted a pilot study to consider the establishment of a school award program in order to stimulate schools to promote healthy eating and physical activity in schoolchildren. Based on the fulfillment of a list of criteria in these areas, schools received an award, which was supposed to document the school's serious commitment to these aspects. In their experiment, 26% of the invited schools had joined the program within 4 weeks, but the study did not report outcomes of the intervention in terms of final goal achievement. In the Kansas LEAN School Health Project,49 a combined intervention involving economic incentives for teachers to implement the nutritional education curriculum (e.g., grants to buy classroom materials), economic incentives for students to engage in physical activity (reward of class parties for using the fitness stations), and economic incentives for the modification of school lunches facilitated changes in the school to reduce the risk of chronic disease by improving the nutritional quality of school lunches and enhancing the students' nutritional knowledge and physical fitness. However, the study did not allow for the identification of the specific effects of the individual incentive schemes.

As mentioned above, economic incentives may influence decision processes and moderators. However, very limited empirical or experimental research in these areas was identified. One approach to measuring such effects might be to measure whether changed economic incentives influence the preferences of adolescents. Cullen et al.48 investigated the influence of free fruit and vegetables on high-school students' preferences for various fruits and vegetables but found that the intervention had no significant effect on preferences. Ard et al.45 investigated the in-home availability of 27 fruit and vegetable items as a function of price, income, and socioeconomic factors. They found that availability was negatively associated with price and positively associated with income, which is in accordance with economic theory and a wide range of empirical studies of consumer demand behavior.

Compared to evidence from the reviewed literature on direct economic incentives for children, the evidence of indirect influences within the school food environment is somewhat weaker. The reviewed studies, which involve economic incentives for the teachers and school administration, are based on interventions that combine economic incentives with other elements – where economic incentives mainly serve as a catalyst for the other elements of the study – thus making it difficult to separate the effect of the incentive. The studies related to the costs of more nutritious supplies tend to suffer from weaknesses in study design, for example, lack of a control group, while the results may also be difficult to generalize due to dependency on specific cost structures, etc.

DISCUSSION

The existing research on the effectiveness and feasibility of economic incentive instruments to improve children's nutritional behavior in schools is relatively scarce, with limited opportunities to assess the explicit, singled-out effects. Two basic types of economic incentive strategies directly related to economic environmental factors have been investigated: price incentives and incentives to stimulate participation in health-promoting multi-instrumental interventions. Whereas price incentives appear to be effective for promoting fruit and vegetable consumption and reducing the consumption of energy-rich foods and beverages, the efficacy of incentives to stimulate participation is more difficult to assess from the research reviewed due to the intervention and study designs of the studies. The effects of changes in economic incentives on other (physical, social, and political) environmental factors are even more difficult to ascertain from the studies reviewed, but the results seem to indicate that economic incentives may support the supply of healthy meals within school meal programs.

Effectiveness of economic incentive instruments

Some of the main types of economic incentive instruments that could be applicable in schools are summarized in Table 2.

View this table:
Table 2

Overview of economic incentive schemes for nutritional improvements in the school setting.

Type of incentiveIncentives for students and their familiesIncentives for schools and communities
Price incentivesHigher prices for unhealthy foods, snacks, beverages
Lower prices for healthy foods, snacks, beverages
Reward incentivesReward for participation in health-promoting activities, either individual- or group-basedReward upon school achievement of specified nutritional goals
Reward upon individual achievement of specified nutrition goals, either individual- or group-basedCompensation for efforts devoted to promotion of healthy nutrition

Overall, studies that address price incentives in schools suggest that such incentives are effective for altering the consumption in school cafeterias or from vending machines and for increasing fruit/vegetable consumption in schools in the short run and, to some extent, in the long run. One study35 suggests that such economic incentives should have “some magnitude” to have a significant effect, while other studies suggest that such “large” economic incentive changes (e.g., a period with zero-priced fruit and vegetables) may even have fairly long-lasting effects.30 The apparent evidence for the effectiveness of price incentives has, to some extent, been found for schools at which the students are older than 10–12 years. However, the extent to which these findings can be translated to 10–12-year-olds is not clear, although studies of price simulation experiments31 have shown that 10–12-year-olds may be expected to react in a similar way to price changes. There are two crucial determinants for the effectiveness of price instruments: 1) foods or beverages are offered (for sale) at the schools, and 2) 10–12-year-old children bring money to school to buy some of these items. Further research is required to determine to what extent these conditions are met in schools in different countries.

When evaluating the effectiveness of economic incentive instruments, it is important to know whether all groups of children modify their food intake or whether there is, for example, a social bias in the effects (which might possibly be expected with enabling incentives such as price reductions on healthy foods). A social bias in actual dietary behavior has been observed,61 and it is crucial to determine whether the economic incentives increase or decrease these social biases.

This review includes a few studies in which economic incentives were used to facilitate schools' participation and teachers' involvement in health-promoting educational activities. Since these incentives have been combined with several other intervention elements in all the studies considered, it has not been possible to identify the partial motivating effects of these incentives, but it can be assumed (as was probably done by the researchers undertaking the studies) that the impacts of the incentives would support the aim of the intervention. Whereas the effectiveness of price incentives has, to some extent, been documented at schools in the literature, the effectiveness of other types of economic incentive instruments (e.g., reward schemes) and the influence of the targeting and design of the incentives seems to be an underinvestigated topic. However, studies from other fields, such as health promotion in the workplace,6268 suggest that such other types of instruments may have potential use at schools. Furthermore, rewards for schools upon achieving stated goals regarding the school food environment might be an instrument worth considering.

In their literature review, Marteau et al.5 note that economic incentives may affect individuals' motivation, since a meta-analysis of 128 experiments69 found an undermining effect of rewards on intrinsic motivation, as assessed by persistence in a task when rewards were stopped. Such mechanisms should be taken into account when assessing the potential effectiveness of economic incentive instruments.

Feasibility of economic incentive instruments

One important aspect of intervention instruments is the multidimensional issue of feasibility. One dimension is the feasibility related to the specific design of the intervention instrument, including the logistics surrounding it. Such practical issues may be of minor importance when modifying the prices of snacks in a vending machine or when providing all schoolchildren with a free piece of fruit per day but may be more crucial when it comes to implementing, for example, nutrition-based reward schemes according to some measurement of goal achievement.

With regard to economic incentive interventions aimed at teachers and schools, one important feasibility constraint could be the acceptability of the incentive schemes among teachers (and other staff affected by the incentives). Many of the proposed schemes require the support of teachers in the form of time and effort. A few of the studies reviewed have considered enabling economic incentives that compensate teachers for their extra efforts (and they seem to be feasible), but it is conceivable that reinforcing incentives aimed at stimulating some teachers' extra efforts and creativity in, for example, nutrition teaching, might be considered unfair by other teachers, which, in turn, may hamper the feasibility of such measures.

Financial considerations pose other important challenges to the feasibility of economic incentive instruments. For example, a free fruit/vegetable scheme for schoolchildren or lower prices of healthy foods from vending machines at schools requires a source of financing (to the extent that higher demand does not compensate for the reduced gross margins), and some incentive schemes may require additional human resources. One way of financing such budgetary outlays could be to raise the prices of unhealthy products sold at schools. However, as the availability of such unhealthy products at schools ought to be minimized, from a nutritional perspective, this is not likely to be a sustainable way of financing such deficits. Hence, price-related incentive schemes may require some sort of external financing to be economically sustainable. It should, however, be kept in mind that many other types of intervention, such as educational activities, most often require financing of some sort as well.

In addition to such practical and financial considerations of feasibility, there are a number of ethical and political considerations related to the use of economic incentive instruments to promote healthy nutritional behavior in schools. One such consideration relates to the acceptability of public intervention in healthy children's nutritional behavior in general and the specific interventions in particular, which tends to interfere with the normal conception that the intake of meals belongs to the private domain of children and their parents. Likewise, it might be questioned whether public tax money should be spent on paying schoolchildren to eat and drink healthfully. Another issue is whether the use of certain economic incentive schemes will tend to be socially biased, either by imposing relatively high costs on low-income households or by providing benefits that are more easily reaped by children and parents in higher-income households. Furthermore, it is worth considering whether some economic incentive instruments are more intrusive than others. For example, it might be expected that individual or group reward schemes would be considered more intrusive than price incentives. Again, it can be debated whether these private-public domain conflicts are more pronounced with regard to economic incentive instruments than other types of intervention instruments.

Perspectives for policy and management

From an economic perspective, it can be argued that individuals' decisions about energy balance are a matter of their own responsibility and that if individuals are rational in their trade-offs between utility from consumption, physical activities, and anticipated health in a free and well-functioning market, government interventions are unnecessary and may distort this decision-making, thereby leading to suboptimal outcomes.70 However, Cawley17 mentions several economic rationales that may justify government intervention in markets to address childhood obesity. First, because free markets generally underprovide information, the government may intervene to provide consumers with necessary nutritional information. Second, because society bears the growing costs of obesity, the government may intervene to lower the costs to taxpayers. Third, because children are probably less able to evaluate information critically and are less likely to assess the future consequences of their actions, the government may step in to help them make better choices. The case for policy intervention is further supported by a contingent valuation study,71 in which respondents show a significantly positive willingness to pay for reduction in childhood obesity, which suggests that citizens consider policy interventions to reduce childhood obesity economically justified.

The results reported in this review suggest that economic incentive mechanisms could potentially have a larger role to play in the formulation of school-based nutritional intervention strategies, although the amount of existing research in this area is still relatively limited. So far, research related to “pure” economic incentive interventions has focused on “free offer” and price interventions. These show some effectiveness in altering children's nutritional behavior, the latter to the extent that children bring money to school and buy the considered food items (which may differ across cultures). However, the literature also suggests other economic incentive tools that might be worthy of consideration, including reward schemes for children and their families, or economic incentives for schools and teachers upon implementation of specified intervention activities or upon the achievement of specified nutritional goals.

Shortcomings and perspectives for further research

Two obvious shortcomings of the present literature review are the limited amount of relevant empirical studies and the problem of separating the effects of economic incentives from those of other intervention elements. Thus, although the literature reviewed suggests a role for economic incentive instruments in schools, it must be recognized that empirical evidence supporting the use of such instruments, including their appropriate design and their strengths and weaknesses, is relatively limited. This calls for additional research in a number of areas.

First, a need exists for deeper understanding of the mechanisms that may be triggered when using economic incentive instruments, how such incentives change children's dietary behavior, body weight, and health condition, and how this cause-effect chain works. For example, there may be cases in which a tax on a high-calorie food decreases the consumption of the food good but leads to an increase in body weight due to various substitution and complementary effects in food consumption.72 Furthermore, Just et al.73 investigated economic incentives in light of behavioral and psychological studies, which indicate that people regularly and predictably behave in ways that contradict some standard assumptions of economic analysis.74

There is a need for research on the effectiveness and feasibility of economic incentive instruments other than prices or free offers in schools. Such novel instruments might be developed using the ideas of various reward schemes from adjacent fields, albeit adapted to the school setting in terms of goal variables, incentive instruments, and incentive sizes. In the assessment of effectiveness, it is also relevant to compare the effectiveness of economic incentive instruments with the effectiveness of other types of instruments, such as purely educational instruments (nutritional education, cooking courses, etc.), availability instruments, etc. Such comparisons might address the maximum potential effects of different types of instruments or the cost-effectiveness of the instruments, i.e., the ratio between the effect and cost of instruments. Such comparative cost-effectiveness analyses of instruments to promote sound nutritional behavior in children appear to be very scarce, if they exist at all, and may be a subject for further research.

There is also a need for research into the potential of and barriers to the use of economic incentives aimed at schools and teachers, and whether such incentives can lead to improved performance in the schools' nutritional education efforts and the creation of a physical environment that supports healthy nutritional behavior. A few of the studies reviewed have included such incentives, but the incentives were fully integrated with other intervention elements, which makes it difficult to distinguish between the effects of the different incentive schemes. Improvements in study design, in order to allow the effects of economic incentives to be separated from the effects of other intervention elements, may add important knowledge to this area.

Another shortcoming in the existing literature is the lack of results within different groups of children. As social bias in dietary behavior is well established in the literature,61 there is a crucial need to know whether the use of economic incentive instruments in health-promoting interventions amplifies or reduces these social differences. Moreover, further research is warranted to determine the extent to which these effects differ for different incentive instruments and the extent to which economic incentive instruments can be designed to take such effects into account. Remaining questions include whether the use of price incentives has more serious economic consequences for some groups of children than for others, whether the use of enabling economic incentives for children from low-income families leads to the stigmatization of these children, and whether reward incentives based on group performance lead to peer support for low-resource children or to stigmatization.

The studies reviewed in this paper have generally dealt with the active use of incentive instruments in the design of interventions to improve dietary behavior among children and adolescents in schools. It should be kept in mind, however, that the economic framework and, hence, the incentives for decisions relevant to this dietary behavior are influenced to a large extent by other economic forces and incentives, such as the general budgetary situation of the schools and local communities, the financial situation and time constraints faced by the children's parents, regulation and marketing efforts related to food shops and food service outlets in the vicinity of schools, commercial advertising, etc. It is likely that economic incentives induced by such (external) forces will interact with the incentives that may be activated by intervention strategies in schools. However, this is an area that has received little research attention, making it an important avenue for future research in the field of economic incentives and child nutrition.

CONCLUSION

This review has summarized the literature on the use of economic incentives to promote healthy eating in schools. The results suggest that responses to price changes are in accordance with standard economic theory. The few studies that address price incentives in schools suggest that such incentives are effective for altering consumption in school cafeterias or from vending machines and for increasing fruit/vegetable consumption at schools. Such incentives, therefore, are also likely to modify the food intake of the children.

In addition to price incentives, other economic incentives may be useful for promoting improved nutrition at schools, including economic incentives to facilitate schools' participation and teachers' involvement in health-promoting educational activities as well as enabling incentives to remove or reduce barriers for participation in such activities. Although the studies reviewed in this article examined such incentives in combination with several other intervention elements and thus do not allow the identification of the specific effect of the individual incentives, it is assumed that the impacts of these incentives would support the aim of the intervention.

As the amount of existing research on the use of economic incentives for improving nutrition in schools is limited, further research in this area is encouraged.

Acknowledgments

Funding

The EuropeaN Energy balance Research to prevent excessive weight Gain among Youth (ENERGY) Project is funded by the Seventh Framework Programme (CORDIS FP7) of the European Commission, HEALTH (FP7-HEALTH-2007-B) and is conducted by a partnership of researchers from 11 countries and two international organizations (WHO and International Obesity Task Force). The overall aim of the ENERGY study is the development of a new theory- and evidence-based multicomponent intervention scheme ready to be implemented across Europe promoting the adoption or continuation of health behaviors that contribute to a healthy energy balance. The content of this article reflects the authors' views only, and the European Community is not liable for any use that may be made of the information it contains.

Declaration of interest

The authors have no relevant interests to declare.

REFERENCES

View Abstract