A Review of Snacking Patterns among Children and Adolescents: What Are the Implications of Snacking for Weight Status?

Abstract

Given the growing contribution of snacks to dietary intake and the need for effective strategies to reduce obesity, it is important to consider whether snacking behaviors contribute to high BMI in childhood. This review summarizes US research that has addressed trends in snacking behavior and its contribution to dietary intake, as well as research describing snack food availability in settings where youth spend their time. In addition, it comprehensively reviews studies conducted in the United States and internationally that have examined associations of snacking behavior with weight. Research published between January, 2000, and December, 2011, was identified by searching PubMed and MEDLINE databases, and by examining bibliographies of relevant studies. Recent analyses of trends in dietary intake have shown there were significant increases in the contribution of energy-dense, nutrient-poor foods to snacking kilocalories over the past few decades. Although snacks can contribute to intake of key nutrients, frequent snacking has been associated with higher intake of total energy and energy from added and total sugars. Assessments in schools and retail stores have further indicated that energy-dense, nutrient-poor snacks are widely available in settings where youth spend their time. The majority of studies either found no evidence of a relationship between snacking behavior and weight status or found evidence indicating that young people who consumed more snacks were less likely to be obese; however, additional research is needed to address various methodological limitations. Recommendations for future research are provided to address knowledge gaps and inform the development of interventions.

Introduction

The prevalence of obesity dramatically increased among US children and adolescents over the past four decades by a magnitude of two to nearly five times.^1,2 National estimates indicate that nearly one-third of young people ages 2–19 years are overweight or obese.² This epidemic level of obesity is the most challenging nutritional problem facing US children and adolescents and has staggering economic consequences for the nation. Obese children and adolescents are at increased risk for obesity in adulthood and a number of serious health conditions such as hypertension, dyslipidemias, orthopedic problems, and type 2 diabetes.³ On the basis of current trends, the combined medical costs associated with treating obesity-related conditions are expected to increase by $48–$66 billion per year by 2030.⁴ Population-based prevention strategies and interventions are urgently needed to reverse current trends and protect the health of young people.

Several trends in eating patterns have coincided with the rise in obesity, including increases in portion sizes,^5–7 increases in the consumption of food prepared away from home,^8,9 and increases in the contribution of snacks to total energy intake.^10,11 However, the contribution of these trends to obesity in child and adolescent populations is not yet well understood. In particular, there is a lack of consensus in regard to the impact of snacking patterns on weight status to inform the development of recommendations and interventions. National data indicate foods and beverages consumed at snack occasions contribute approximately 25% of total daily energy and are a significant source of healthful nutrients for young people ages 2–19 years.¹² Thus, regular snacking may help children and adolescents to meet dietary recommendations, but may also contribute to excess energy intake in line with evidence for the wide availability of energy-dense snack foods and beverages.^13–18 Given the growing contribution of snacks to dietary intake and the need for effective strategies to reduce and prevent obesity,¹⁰ it is important to consider whether snacking behaviors are associated with high BMI and the development of obesity in childhood or adolescence.

Recent reviews have addressed the potential implications of snacking for weight status in adults. However, to the best of the authors' knowledge, no recent reviews focusing on children and adolescents have been published that comprehensively examine this relationship. The current review was thus conducted with a focus on children (2–11 years) and adolescents (12–19 years) to inform recommendations regarding snacking and the design of future nutrition interventions. To provide context for the review, this article first provides a summary of US research that has addressed trends in snacking behavior and its contribution to dietary intake, as well as research describing the availability of snack foods and beverages in settings where youth spend their time. The summary is followed by a review of studies conducted in the United States and internationally that have examined associations of snacking behavior with weight status. The evidence presented is further discussed in terms of methodological limitations and knowledge gaps are identified to provide direction for future research.

Methods

Peer-reviewed research studies were identified by completing searches in PubMed and MEDLINE using the following key words—snack, dietary intake, BMI, obesity, overweight, children, and adolescents. To be included in this review, articles had to address at least one of the review objectives and be published in English between January, 2000, and December, 2011. The review was limited to literature published during this period in order to focus on recent studies with the greatest potential implications for interventions. Articles describing trends in snacking behaviors or its contribution to dietary intake and the availability of snack foods and beverages were limited to studies conducted in the United States; however, given the small number of scientific publications and their potential for informing future research, relevant studies addressing associations of snacking patterns with weight status were not excluded if completed outside the United States. The bibliographies of identified articles were additionally reviewed to ensure that all relevant studies were included. In addition to peer-reviewed studies, reports prepared for federal US agencies were included to describe the most recent and nationally representative data on snacking patterns. The first author reviewed the retrieved articles to determine their relevance to this review and to evaluate the research methodology; information regarding study design, study population and size, assessment measures, and results was abstracted for each article. The summary of results and discussion that follow was organized to prioritize the findings of observational studies that were longitudinal in design or nationally representative, but no articles were excluded based on study design. A summary of studies addressing associations of snacking patterns with weight status in Table 1 was also organized by study design and provides details on the definition of snacking and categorization of weight status in each study.

Table 1.

Research Assessing Associations between Snacking and Weight Status among Children and Adolescents

Reference	Sample and setting	Snacking measure	Outcome assessment^a	Summary of results
Case–control design
de Novaes and colleagues, 2008⁵¹	n=100 children (n=50 were obese, 50 non-overweight) aged 6–8 years in Vicosa, Brazil	Maternal report of their child's frequency of having a snack in commercial establishments	Measured height and weight	Frequency of having a snack in commercial establishments was directly related to obesity. The majority (79%) of commercial snacks mentioned by mothers were soft drinks and energy-dense foods such as chips, tortillas, and French fries.
Tanasescu and colleagues, 2000⁵²	n=53 Puerto Rican children (n=29 were overweight, 24 controls) aged 7–11 years in Hartford, CT	Parent report of their child's snack food consumption (chips, peanuts, peanut butter, and popcorn) on a food frequency questionnaire	Measured height and weight	A positive correlation was found between snack food consumption and BMI; however, no association with weight status (overweight or not overweight) was observed in multivariate models accounting for demographic and other lifestyle factors.
Cross-sectional design
Kuriyan and colleagues, 2012⁴⁷	n=8444 children aged 3–16 years in Bangalore, India	Parental report of child snacking habits on a mailed questionnaire	Measured height, weight, and waist circumference; weight status was defined using guidelines developed by the Indian Academy of Pediatrics	Snacking between meals was positively related to waist circumference.
Cheah and colleagues, 2011⁴⁵	n=316 adolescents aged 13–17 years in Kuching South City, Malaysia	Adolescent report of consuming foods advertised on television (cereals, soft drinks, fast food, junk food) while viewing	Measured height and weight; weight status was defined using World Health Organization (WHO) reference data	No relationship was found between snacking habits while viewing television and weight status.
Gregori and colleagues, 2011⁴⁶	n=1215 children aged 6–10 years in Italy	Maternal report of whether their child consumed at least three energy-dense products (filled or plain cakes/sweet snacks, chocolate snacks, sugar-sweetened sodas) daily was used to classify them as a ‘snacker’ or ‘nonsnacker’.	Maternal report of height and weight	Classification as a ‘snacker’ was not associated with child overweight status.
Maruapula and colleagues, 2011⁴⁸	Nationally representative sample of n=667 adolescents in Botswana, Africa	Adolescent report of dietary intake for the previous day was used to identify dietary patterns, including a snack-food pattern characterized by more servings of high-fat, high-sugar snack foods and fewer servings of traditional foods	Measured height and weight; weight status was defined using WHO reference data	Adherence to the snack-food pattern was positively associated with adolescent overweight/obesity; however, the association was not found to be statistically significant after accounting for socioeconomic status.
Watanabe and colleagues, 2011⁴⁹	n=1765 children aged 3–6 years in a city within the Tohoku region of Japan	Parental report of eating snacks at fixed versus unfixed times	Measured height and weight; weight status (normal weight or overweight) was defined using the International Obesity Task Force (IOTF) BMI cutoffs	No relationship was found between weight status and whether children had snacks at fixed versus unfixed times.
Food Surveys Research Group, 2010¹⁹	US nationally representative sample of adolescents aged 12–19 years	Frequency of snack consumption assessed by 24-hour recalls with adolescents; snacking occasions were defined by adolescents	Measured height and weight	Daily snacking frequency was not associated with BMI among males or females.
Keast and colleagues, 2010²¹	US nationally representative sample of n=5811 adolescents aged 12–18 years	Frequency of snack consumption and energy intake from snacks assessed by 24-hour recalls with adolescents; self-reported name of eating occasion type was used to identify snacks including food accompanied by a beverage or food with no beverage.	Measured height, weight, and waist circumference	The prevalences of overweight and abdominal obesity were inversely associated with snacking frequency and percentage of energy from snacks.
Isacco and colleagues, 2010⁵⁰	n=278 children aged 6–10 years enrolled at 18 primary schools in Clermont-Ferrand, France	Frequency of snacking between meals was assessed by parental and child report on an interviewer-administered questionnaire.	Measured height, weight, skinfolds, and waist circumference	Frequency of snacking was positively associated with waist circumference, BMI z-score, and the sum of four skinfolds.
Wouters and colleagues, 2010³⁰	n=749 adolescents aged 12–17 years enrolled at five secondary schools in The Netherlands	Adolescent report of consuming five energy-dense snack foods (nuts; chips, cheese, and sausage; pastry, cake, and cookies; candy bars; chocolate) and carbonated soft drinks on a food frequency questionnaire	Measured height and weight	Snack food consumption was higher in adolescents with a lower BMI z-score.
Mercille and colleagues, 2010³¹	n=444 children from 4^th–6^th grade classrooms in a Mohawk community of Quebec, Canada	Frequency of snack consumption and energy intake from snacks was assessed by 24-hour recalls with adolescents; snack occasions were defined by the interviewer based on time of the day and whether food was consumed between two consecutive meals.	Measured height and weight	Energy intake from snacks tended to be higher for overweight children compared with children at a healthy weight and children who were obese. Among boys, an inverse association was observed between energy density of snacks and BMI. Neither the number of snacks nor the number of meals per day was associated with BMI.
Kerr and colleagues, 2009³³	n=434 adolescents aged 13–16 years in mainland Britain and a cohort of n=47 adolescents aged 13–16 years in Northern Ireland	Adolescents completed 7-day weighed dietary records; snacks were defined as eating occasions occurring outside of three specific time frames assigned for breakfast, lunch, and dinner meals.	Measured height and weight; weight status was defined using the IOTF BMI cutoffs	No differences in snack energy intake, percentage of total energy from snacks, or frequency of snack food consumption were observed between normal weight and overweight/obese adolescents.
McDonald and colleagues, 2009³⁵	n=1027 children aged 5–12 years in Bogota, Colombia	Maternal report of child's dietary intake on a food frequency questionnaire and principal components analysis were used to identify food patterns.	Measured height and weight; weight status was defined using the IOTF BMI cutoffs	Adherence to the snacking pattern (high consumption of foods such as candy, ice cream, fried snacks, soda, sugar-sweetened fruit-flavored drinks) was positively associated with child overweight after accounting for child age, sex, and total energy intake; maternal BMI; and number of household assets. The prevalence of overweight was twice as high in the most adherent quartile as in the least adherent quartile.
Sun and colleagues, 2009³²	n=5753 adolescents aged 12–13 years in Toyama, Japan	Adolescents completed questionnaires assessing frequency of snack consumption (rarely, sometimes, almost daily, or daily).	Measured height and weight; weight status was defined using the IOTF BMI cutoffs	Snacking was inversely associated with overweight in girls and not associated with overweight in boys. Nighttime snacking was inversely associated with overweight in boys and girls.
Vader and colleagues, 2009³⁴	n=11,594 children from fourth and eighth grade classrooms in Texas, US	Children completed questionnaires assessing weekday frequency of snack consumption and daily consumption of five foods advertised on television (punch, sports drinks, and other fruit-flavored drinks; sodas and soft drinks; frozen desserts; doughnuts and baked desserts; and chocolate candy).	Measured height and weight	Among students in 8^th grade, daily consuming ≥2 snacks and ≥4 foods advertised on television was associated with reduced odds of overweight regardless of the amount of television watched. Among students in 4^th grade, daily consuming ≥2 snacks was associated with reduced odds of overweight only among those who watched ≤2 hours of television per day.
Aounallah-Skhiri and colleagues, 2008³⁷	Nationally representative sample of n=2872 adolescents aged 15–19 years in Tunisia, Africa	Frequency of snacking assessed by in-person interviews; regular snacking was defined by report of at least one daily eating episode apart from meals.	Measured height and weight; percentiles for determining weight status were based on the WHO/National Center for Health Statistics data	Irregular snacking habits were associated with overweight only among males in urban areas.
Dubois and colleagues, 2008³⁹	n=1549 children aged 4–5 years in Quebec, Canada	Frequency of watching television while eating a snack was assessed by maternal completion of a self-administered questionnaire.	Measured height and weight	Children who consumed snacks while watching television on a daily basis had a higher average BMI than children who did so less frequently.
Lioret and colleagues, 2008³⁸	Nationally representative sample of n=748 children aged 3–11 years in France	The contribution of snacks to total energy intake was determined using 7-day food records completed by children and their caregivers.	Self-reported height and weight; weight status defined using the IOTF BMI cutoffs	The contribution of snacking episodes to total daily energy intake was inversely associated with risk of overweight in children, independent of total energy intake, physical activity, and sedentary behavior.
Maffeis and colleagues, 2008³⁶	n=1837 children aged 8–10 years enrolled at primary schools in Verona, Pisa and Naples, Italy	Type and number of snacks consumed weekly was assessed by parental completion of a physician-administered questionnaire.	Measured height and weight; weight status was defined using the IOTF BMI cutoffs	Obese children consumed more servings of snacks per day compared to overweight and normal weight children, but the difference was not statistically significant. Energy intake with snacks per day was not related to weight status. Mean energy density of snacks was significantly higher for obese and overweight children compared to normal weight children.
Toschke and colleagues, 2005⁴⁰	n=4370 children aged 5–6 years in Germany	Frequency of snacking in front of the television and daily meals was assessed by parental completion of a mailed questionnaire.	Measured height and weight; weight status was defined using the IOTF BMI cutoffs	The prevalence of overweight and obesity decreased by number of daily meals. Overweight and obesity were positively related to regular snacking in front of the television.
Huang and colleagues, 2004⁴¹	US nationally representative sample of n=1995 children aged 3–19 years	Frequency of snack consumption, energy intake from snacks, and energy density of snacks was assessed by two 24-hour recalls with children or a proxy adult household member; self-reported name of eating occasion type was used to identify snacks.	Self-reported height and weight	Frequency of snack consumption, energy intake from snacks, and energy density of snacks were unrelated to BMI percentile.
Nicklas and colleagues, 2003⁴²	n=1562 children aged 10 years in Bogalusa, LA	Frequency of snack consumption, number of total eating episodes, and the amount of foods and beverages consumed at snack occasions were assessed by a single 24-hour recall with children and follow-up phone calls to parents; snack occasions were based on each self-defined snack that was reported without a time-period limitation.	Measured height and weight	The total gram amount of foods and beverages consumed, particularly from snacks, was positively associated with obesity. Frequency of snack consumption, number of total eating episodes, and consumption of salty snacks were not related to obesity.
Sekine and colleagues, 2002⁴³	n=8941 children aged 2–4 years in Toyama, Japan	Frequency and regularity of snack consumption was assessed by parental completion of a self-administered questionnaire.	Measured height and weight; weight status was defined using the IOTF BMI cutoffs	Snacking habits were unrelated to child weight status.
Maffeis and colleagues, 2000⁴⁴	n=530 children aged 5–11 years in northeastern Italy	Percentage of energy consumed at morning snacks, afternoon snacks, and after-dinner snacks was assessed by maternal and child completion of a diet history interview.	Measured height, weight, and triceps and subscapular skinfolds	Simple correlations showed an association between percentage fat mass and percent of energy consumed at the after-dinner snack in the total sample and among boys.
Longitudinal design
Cutler and colleagues, 2012⁵⁷	n=2516 adolescents aged 11–18 years when baseline assessments were completed in Minneapolis/St. Paul, MN	Adolescent report of past-year dietary intake on a food frequency questionnaire and principal components analysis were used to identify eating patterns; a sweet and salty snack food pattern was characterized by high consumption of foods such as chocolate bars, cake, brownies, potato chips, and nachos.	Self-reported height and weight; weight status was defined using reference data from the first National Health and Nutrition Examination Survey (NHANES I)	Adherence to the sweet and salty snack food pattern was associated with lower odds of being overweight/obese (BMI≥85^th percentile) among boys at baseline. No longitudinal association was found between adherence to the sweet and salty snack food pattern and weight status at follow-up after accounting for baseline weight status.
Oellingrath and colleagues, 2011⁵⁹	n=924 children aged 9–10 years when baseline assessments were completed in Telemark County, Norway	Parental report of child's dietary intake on a food frequency questionnaire and principal components analysis were used to identify eating patterns; a snacking pattern was characterized by consumption of energy-dense foods and sugar-sweetened beverages between meals.	Measured height and weight; weight status was defined using the IOTF BMI cutoffs	No cross-sectional association was found between weight status and snacking pattern score. Likewise, snacking pattern score was unrelated to changes in weight status over the 3-year study period.
Gubbels and colleagues, 2011⁵⁸	n=2002 children who completed assessments at ages 5, 7, and 8 years in The Netherlands	Parental report of child's dietary intake on a food frequency questionnaire and principal components analysis were used to identify eating patterns; a sedentary-snacking pattern was characterized by much television watching and high consumption of sweets and sugar, pastry and cookies, savory snacks and sauce.	Parent report of height and weight; weight status was defined using national reference data for The Netherlands	Adherence to the sedentary-snacking pattern at baseline was positively associated with BMI z-score and at both follow-up assessments and the likelihood that children were overweight (BMI z-score≥85^th percentile) at age 7 years.
Field and colleagues, 2004⁵³	n=14,977 US adolescents who completed baseline assessments at ages 9–14 years	Adolescent report of consuming 27 types of snack foods and sugar-sweetened beverages on a food frequency questionnaire was used to assess usual daily servings, energy per day from snacks, and percentage of daily energy contributed.	Self-reported height and weight	No relationship between servings of snack food and subsequent annual changes in BMI z-score was found. Results were similar when sugar-sweetened beverages were included as snack foods and when intake was modeled as energy per day from snacks or percentage of daily energy contributed. Among boys, consumption of reduced-fat snack food was associated with less weight gain.
Phillips and colleagues, 2004⁵⁴	n=196 nonobese girls who completed baseline assessments at ages 8–12 years in Massachusetts, US	Child report of consuming five types of energy-dense snack items (baked goods; ice cream; chips; candy; and sugar-sweetened soda) on a food frequency questionnaire was used to assess usual daily servings and percentage of daily energy contributed.	Measured height, weight, and percent body fat	No longitudinal relationship was observed between total energy-dense snack consumption, expressed as servings per day or percentage of daily calories, and BMI z-score or percent body fat. Percentage of calories from sugar-sweetened soda was significantly related to BMI z-score over the 10-year study period, but it was not related to percent body fat.
Sugimori and colleagues, 2004⁵⁵	8170 children who completed assessments at ages 3 and 6 years in Toyama, Japan	Child report of eating between meals 1–2 times per week or more at age 6 years	Measured height and weight; overweight was defined by a BMI≥90^th age-sex specific percentile within the sample	Among boys, snacking was inversely associated with becoming overweight between ages 3 and 6 years.
Francis and colleagues, 2003⁵⁶	n=173 girls who completed assessments at ages 5, 7, and 9 years in Pennsylvania, US	Snacking while watching television, snacking frequency, and fat intake from energy-dense snack foods were assessed by three 24-hour dietary recalls with girls and their mothers at each time point; self-reported name of eating occasion type was used to identify snacks.	Measured height and weight; weight status was defined using the IOTF BMI cutoffs	Among girls from families in which one or both parents were overweight, increases in BMI from age 5 to 9 were predicted by higher intakes of fat from energy-dense snacks. A similar relationship was found among girls from families in which neither parent was overweight; however, the association was not statistically significant.

Abdominal obesity was defined by a waist circumference ≥90^th percentile for year of age and sex according to US reference data.

Unless otherwise stated, child/adolescent weight status was defined by percentiles based on the 2000 Centers for Disease Control and Prevention Growth Charts; a BMI≥85^th percentile was categorized as overweight and a BMI≥95^th percentile was defined as obese.

Results

Trends in US Snacking and Contributions to Dietary Intake

Recent analyses of nationally representative survey data have indicated that the contribution of snacks to the overall dietary intake of US children and adolescents increased markedly over the past few decades.^10,11,19,20 One analysis used 2 days of 24-hour recall data from four nationally representative surveys to examine trends in snacking behavior.¹⁰ Snacks were largely self-defined by respondents in each survey; however, all foods and beverages reported as snacks that were consumed within a 15-minute period were combined to represent one snack occasion. Among young people aged 2–18 years, the prevalence of snacking as determined by the proportion of respondents having a snack on at least 1 of the 2 recall days increased from 74% in 1977–1978 to 98% in 2003–2006.¹⁰ Additionally, during this same period, the average frequency of snacking increased by approximately one occasion per day and the energy consumed at snack occasions increased by 168 kcal per day.¹⁰ The average frequency of snacking increased to two occasions per day and the contribution of snacking to total daily energy intake increased to 27% by 2003–2006.¹⁰

Nationally representative survey data also indicated significant changes have occurred over the past few decades in the types of foods and beverages consumed at snack occasions.¹⁰ The study described above further observed increases between 1977–1978 and 2003–2006 in the contribution of energy-dense, nutrient-poor foods and beverages (e.g., salty snack foods, candy, fruit-flavored drinks) to the kilocalories consumed at snack occasions. For example, the contribution of high-fat, salty snack foods such as crackers and chips (>5 grams fat per 100 kilocalorie serving) to snacking kilocalories doubled from 7.3% in 1977–1978 to 14.3% in 2003–2006. The contribution of dessert foods to snacking kilocalories decreased over this same period; however, in 2003–2006, dessert foods (e.g., cakes, cookies, ice cream) remained a major source of snacking kilocalories and provided one-fifth of total snacking kilocalories. Sweetened beverages, which provided approximately 14% of snacking kilocalories in 2003–2006, were likewise among the top contributors to snacking kilocalories in both periods.

Although several studies have correspondingly found that more frequent snacking is associated with higher total energy intake and a higher proportion of energy provided by added and total sugars,^19,21–23 national survey data also indicated that snacks can make some positive contributions to intake of key nutrients. In 2009–2010, National Health and Nutrition Examination Survey data for male and female children aged 2–19 years showed that across age and gender groups the foods and beverages consumed at self-defined snack occasions contributed 24% (girls 6–11 years) to 29% (girls 2–5 years) of total daily energy and 34% (girls 6–11 years) to 42% (girls 2–5 years and boys 12–19 years) of total sugar intake based on a single 24-hour recall.¹² Similar to the contribution for total kilocalories, foods and beverages consumed at snack occasions provided 19–32% of calcium intake, 24–31% of magnesium intake, 27–37% of vitamin C intake, 14–29% of vitamin D intake, 26–34% of vitamin E intake, and 20–29% of potassium intake.¹² Moreover, snacks made smaller but meaningful contributions to intakes of protein (14–21%), vitamin A (15–25%), folate (17–20%), iron (18–21%), and zinc (16–22%).¹²

Snack Food and Beverage Availability

Several US studies have examined the availability of foods and beverages that are commonly consumed as snacks by youth in various settings (e.g., retail stores, schools, and neighborhoods surrounding schools) representing major sources of energy consumed away from home.⁸ Although most studies described here did not examine associations between availability and actual consumption or obesity prevalence, the findings have suggested that energy-dense, nutrient-poor snacks were available in retail food stores and also widely available in retail stores that primarily sell other products and services, schools, and locations within walking distance of schools.

The availability and accessibility of snack foods and beverages in retail stores whose primary merchandise was not food (e.g., electronic stores, salons) was examined in 2007 in 19 cities across the United States.¹³ Research teams randomly selected six commercial intersections and attempted to complete observations in 10 retail stores per intersection. Teams only completed observations in street-level areas accessible to customers and assessed the availability of soft drinks and other sweetened beverages, salty snacks (e.g., chips), candy, baked sweets (e.g., cookies), and frozen sweets (e.g., ice cream). Of the 1082 nonfood retail stores where observations were completed, at least one type of snack item was available in 41% of establishments. The types of snack items most commonly available were candy, sweetened beverages, and salty snacks. Depending on the type of item, snacks were located within arm's reach of the cash register in 32% (frozen sweets) to 65% (candy) of establishments where they were available.

Substantial evidence from studies that have conducted on-site observations and self-report surveys indicated that US children and adolescents also have access to a variety of foods and beverages that may be purchased for snacks at school.^16,24–26 According to nationally representative survey data collected in 2010, the average percentage of US secondary schools that allowed students to purchase snack foods or beverages from one or more vending machines at the school or at a school store, canteen, or snack bar was 70%, ranging from 37% to 93% across states.¹⁶ Data from this survey of school principals further indicated that many of the snack items available for purchase at school were energy-dense, nutrient-poor foods or beverages. The average percentage of schools across states that allowed students to purchase sports drinks was greater than 50%, and nearly 30% of schools allowed students to purchase soft drinks or fruit drinks not consisting of 100% fruit juice.¹⁶ Similarly high percentages of schools allowed students to purchase less nutritious snack foods such as salty snacks not low in fat (27%), cookies and other baked goods not low in fat (32%), chocolate candy (20%), and other kinds of candy (25%).¹⁶ The average percentage of schools across states that limited the package or serving size of snack foods or beverages was only 45%.¹⁶

A number of studies additionally indicated that energy-dense, nutrient-poor snack items were often available for purchase from restaurants or retail food stores in neighborhoods surrounding schools. At least three US studies have found evidence that fast-food restaurants tend to be clustered in school neighborhoods.^15,17,27 One of these studies found that a third of secondary schools nationwide have at least one fast-food restaurant or convenience store within easy walking distance (≤0.5 mile).¹⁵ Students at these schools may choose to purchase snacks on the way to or home from school, after sport practices, or at the lunch hour if there is no closed campus policy prohibiting students from leaving the building during the school day. Research in Philadelphia, PA, and Minneapolis/St. Paul, MN, has also provided evidence that convenience stores located near schools may be selling mostly energy-dense, nutrient-poor snack options, and purchases made in these stores contributed significantly to energy intake among urban youth.^14,18,28 Systematic, on-site observations conducted in 17 small convenience stores in Philadelphia showed that no fruit or vegetable items were available and only 20% of snack food inventories could be categorized as healthy according to a criteria of having less than 7 grams of fat, 16 grams of sugar, and 360 mg of sodium per serving.¹⁴ The results of one study in Oakland, CA, further suggested that healthy snack items (e.g., low-fat popcorn, low-fat granola bars) were less often available in food stores located within neighborhoods surrounding lower versus higher income schools.²⁹

Snacking and Weight Status

At least 23 cross-sectional,^{19,21,30–50} two case–control,^51,52 and seven longitudinal,^53–59 studies have examined the relationship between snacking behavior and weight status in children and adolescents (Table 1). Various measures were used by these studies to assess snacking behavior; however, most studies assessed frequency of snack consumption, percentage of energy from snacks, or the consumption of energy-dense snack foods. A small number of these studies (n=9), including two longitudinal studies, found evidence indicating the percentage of energy consumed at snacks or greater consumption of energy-dense snack foods was associated with higher BMI among some groups of children.^{35,36,44,47,50–52,56,58} For example, one longitudinal study found that among US girls from 72 families in which either parent was overweight, frequency of self-defined snacking occasions was associated with higher intake of fat from energy-dense snack foods (i.e., intake of cookies, pastries, crackers, chips, and sweets based on three days of 24-hour recall data), and higher intakes in turn predicted increases in BMI from age 5 to 9 years.⁵⁶ However, this study was conducted in an exclusively white, predominately middle-class, well-educated sample of parents and their daughters and thus generalization of the findings may be limited. Furthermore, the majority of cross-sectional studies (n=12 of 19) and five longitudinal studies either found no evidence of a relationship between snacking behavior and weight status or found evidence indicating that children who consumed food or beverages between meals were less likely to be obese.

Mixed results were also reported by four studies that specifically accounted for the likelihood that obese youth may be more likely than youth at a healthy weight to reduce energy intake for weight loss^21,53 or to underreport intake.^36,41 Two studies found no evidence of a relationship between snacking behavior and weight status,^41,53 one study found a direct relationship between the energy density of snacks consumed and risk for obesity,³⁶ and one study found that more frequent snacking was associated with reduced risk for obesity.²¹ One of these studies made use of nationally representative data to examine the relationship of snacking with weight status and abdominal obesity among 5811 US adolescents aged 12–18 years.²¹ Adolescents completed 24-hour recalls and the self-reported name of eating occasion type was used to identify eating occasions that were snacks. Results showed a lower percentage of energy from snacks was consumed by adolescents who reported they had tried to lose weight in the past year (19%) compared with their peers (21%). Therefore, in addition to accounting for other factors such as age, gender, ethnicity, household income, physical activity, media use, and smoking, all relationships between snacking and indicators of obesity were examined in two samples that included and excluded those adolescents who had tried to lose weight. No cause-and-effect relations could be determined given the cross-sectional design of this study and limitations of the dietary assessment (e.g., energy-dense, nutrient-poor foods, and beverages tend to be underreported); however, the findings showed that mean values of all obesity indicators were inversely associated with frequency of snacking and percentage of energy from snacks regardless of reported efforts to lose weight.

Three cross-sectional studies have investigated whether weight status is associated with consuming snacks while watching television.^39,40,45 The results of two studies indicated that more frequent snacking as reported by parents was associated with an increased risk for obesity in early childhood,^39,40 but no association was found in the one study conducted in adolescent youth.⁴⁵ One of the studies in young children further considered several potential demographic confounders of a relationship, including household income; parental age, education, employment status, and smoking habits; and child sex and child care attendance.³⁹ For this study, anthropometric measurements were completed and data were collected from parents on child television viewing and food consumption habits in a sample of 1549 children aged 4 to 5 years in Quebec, Canada.³⁹ Approximately one-third of the children ate snacks in front of the television less than once per week, 46% ate snacks in front of the television a few times per week, and 17% did so every day. While the results indicated there was no difference in the mean BMI of children according to hours of daily television viewing, children who ate while watching television once daily or more had a higher mean BMI in comparison to children who ate while watching television less than once daily (15.9 versus 15.5 kg/m²). Further, results specific to meals and snacks showed that children who ate snacks or dinner once daily or more in front of the television had a higher mean BMI in comparison to children who did so less than once daily but similar relationships were not found for breakfast or lunch. Despite important strengths of this study, including the large and representative sample, it is noteworthy that the results are based on parental report of children's television viewing at meals and snacks using a limited four-item response scale.

Discussion

The aim of this review was to summarize recent research relating to snacking behaviors of children and adolescents and to evaluate the evidence base regarding the relationship between snacking and weight status. Among US children and adolescents, the contribution of snacks to overall dietary intake has increased markedly and there have been significant increases in the contribution of energy-dense, nutrient-poor foods and beverages to snacking kilocalories over the past few decades.^10,11,19,20 Although snacks can make some positive contributions to intake of key nutrients, several studies have found that more frequent snacking was associated with higher total energy intake and a higher proportion of energy provided by added and total sugars.^19,21–23 Of further concern, there is evidence that energy-dense, nutrient-poor snack foods and beverages are widely available in various settings where young people spend their time.^{14–18,24–28} The majority of studies that investigated whether there is a relationship between snacking behavior and weight status either found no association^{19,33,41–43,45,46,48,49,52–54,59} or found evidence indicating that young people who more often consumed food or beverages between meals were less likely to be obese.^{21,30,32,34,37,38,55,57} However, a smaller number of studies found evidence indicating the frequency of snacking, percentage of energy consumed at snacks, or greater consumption of energy-dense snack foods was associated with increased risk for obesity among some groups,^{31,35,36,44,47,50,51,56,58} and two studies found that more frequent snacking in front of the television was associated with higher BMI.^39,40 Additional research is needed to address various methodological limitations and knowledge gaps so that the relationship between snacking behaviors and weight status may be elucidated and the design of nutrition interventions better informed to promote the consumption of healthful snacks.

One issue deserving of particular attention in the design of future studies is the need to use consistent definitions for snacking occasions and energy-dense snack foods and beverages. Multiple different criteria were used by the studies included in this review to define a snack occasion such as time of day, the types or amounts of food consumed, and subjective assessment of the participant (Table 1, snacking measure). Similarly diverse definitions were used to define energy-dense snacks, ranging from assessments of just four to 27 different types of food.^52,53 A prior review found the association between snacking patterns and weight status may partly depend on how snacking is defined.⁶⁰ Although the previous review considered only two common criteria for defining a snack (type of food or beverage versus time when the food or beverage was consumed) and did not specifically focus on the nature of associations in child and adolescent populations, findings to date suggest that researchers should work toward a consensus on the use of standard definitions and future studies should be sure to fully describe how snacking patterns were defined.^60,61

Other common limitations of studies that have examined whether there is a link between snacking behavior and weight status relate to the cross-sectional, observational nature of most research designs and the challenge of accounting for other dietary and lifestyle factors that may influence associations. Only seven longitudinal studies^53–59 were identified by this review and three of these studies^53,54,56 were conducted in US samples primarily representing non-Hispanic white children or adolescents. Additional longitudinal studies in racially/ethnically and socioeconomically diverse samples are needed to clarify the temporal nature of associations. To allow for comparisons across studies, it will additionally be important for future studies to consistently account for underreporting of dietary intake, efforts to lose weight, and energy expended in physical activity using valid and reliable measures. Few studies have clearly addressed the potential for biased associations to result from the likelihood that overweight youth may reduce their kilocalorie intake for weight loss^21,53 or underreport intake^41,44 more often than youth at a healthy weight.

Gaps in the evidence base reviewed here that need to be addressed by future research include the extent that children and adolescents compensate for the kilocalories consumed at snacks; the potential influence of snacking location (e.g., at home versus a restaurant) and context (e.g., talking with friends versus watching television) on dietary intake and weight status; and determinants of snack food and beverage inventories in settings where young people spend their time. To further clarify the relationship between snacking and weight status, studies are needed to examine what proportion of children and adolescents typically report eating occasions self-defined as snacks in addition to meals versus to replace meals. Frequent snacking may contribute to excess energy intake if snacks are consumed in addition to regular meals, but have little impact on overall daily energy intake or weight status for those who typically consume snacks in place of meals. Also, data are lacking regarding the extent to which children and adolescents compensate for the kilocalories consumed at snack occasions by reducing their intake of kilocalories at future meals. Research of this nature should identify the characteristics of young people associated with different snacking patterns and whether the extent of compensation depends on the context of a snack occasion or the composition of a snack (e.g., beverages versus solid foods) to best guide interventions and messaging. Finally, to better guide the development of interventions, there is also a need for future studies to explore the determinants of snack food and beverage inventories in parks, community centers, convenience stores, and other retail stores where young people may purchase snacks. While there is some evidence that convenience stores in school neighborhoods provide access to snacks that are mostly low in nutrients and high in kilocalories,^14,18 more research is needed to determine what types of foods and beverages are available in various settings and what factors should be addressed to support improvements in nutritional quality.

Conclusions

Preventing obesity among children and adolescents is a public health priority. Existing research does not indicate regular snacking is related to obesity nor does it suggest that children and adolescents should be discouraged from consuming snacks. However, the contribution of snacks to the overall dietary intake of US children and adolescents has increased markedly over the past few decades.^10,11,19,20 The estimated magnitude of the average increase in energy consumed at snack occasions is quite significant at 168 kcal per day relative to estimates of the energy surplus that can contribute to weight gain over time and reductions in energy intake that would be required to achieve national targets for promoting health and reducing obesity among youth.^62,63 In comparison, a recent study indicated that an average net reduction of 64 kcal per day would be required to achieve the Healthy People 2020 targets for reducing the prevalence of obesity among children and adolescents.⁶² Interventions and supportive policy changes are thus needed to ensure the foods and beverages consumed by youth at snack occasions help them to meet dietary recommendations and do not contribute to excess energy intake. There is an urgent need for well-designed studies to evaluate intervention strategies for promoting the consumption of healthful snacks. Furthermore, there is a critical need to address the wide availability of energy-dense, nutrient-poor snack foods and beverages in environments where youth spend their time. Reversing recent trends and significantly improving the nutritional quality of snack foods and beverages will likely require the combined efforts of nutrition professionals working in industry, schools, and other community settings.

Footnotes

Acknowledgments

This manuscript was supported in part by the Robert Wood Johnson Foundation Healthy Eating Research Program.

Author Disclosure Statement

No competing financial interests exist.

References

Ogden

, Flegal

, Carroll

et al. Prevalence and trends in overweight among US children and adolescents, 1999–2000. JAMA, 2002; 288:1728–1732.

Ogden

, Carroll

, Kit

et al. Prevalence of obesity and trends in body mass index among US children and adolescents, 1999–2010. JAMA, 2012; 307:483–490.

Daniels

. Complications of obesity in children and adolescents. Int J Obes, 2009; 33:S60–S65.

Wang

, McPherson

, Marsh

et al. Health and economic burden of the projected obesity trends in the USA and the UK. Lancet, 2011; 378:815–825.

Young

, Nestle

. The contribution of expanding portion sizes to the US obesity epidemic. Am J Public Health, 2002; 92:246–249.

Fisher

. Effects of age on children's intake of large and self-selected food portions. Obesity, 2007; 15:403–412.

Piernas

, Popkin

. Food portion patterns and trends among US children and the relationship to total eating occasion size, 1977–2006. J Nutr, 2011; 141:1159–1164.

Poti

, Popkin

. Trends in energy intake among US children by eating location and food source, 1977–2006. J Am Diet Assoc, 2011; 111:1156–1164.

Guthrie

, Lin

, Frazao

. Role of food prepared away from home in the American diet, 1977–78 versus 1994–96: Changes and consequences. J Nutr Educ Behav, 2002; 34:140–150.

10.

Piernas

, Popkin

. Trends in snacking among US children. Health Aff (Millwood), 2010; 29:398–404.

11.

Jahns

, Siega-Riz

, Popkin

. The increasing prevalence of snacking among US children from 1977 to 1996. J Pediatr, 2001; 138:493–498.

12.

US Department of Agriculture, Agricultural Research Service. Snacks: Percentages of selected nutrients contributed by foods eaten at snack occasions, by gender and age, What We Eat in America, NHANES 2009–2010. 2012. www.ars.usda.gov/ba/bhnrc/fsrg/. 2013 February 9.

13.

Farley

, Baker

, Futrell

et al. The ubiquity of energy-dense snack foods: A national multicity study. Am J Public Health, 2010; 100:306–311.

14.

Lucan

, Karpyn

, Sherman

. Storing empty calories and chronic disease risk: Snack-food products, nutritive content, and manufacturers in Philadelphia corner stores. J Urban Health, 2010; 87:394–408.

15.

Zenk

, Powell

. US secondary schools and food outlets. Health Place, 2008; 14:336–346.

16.

Brener

, Demissie

, Foti

et al. School Health Profiles 2010: Characteristics of health programs among secondary schools. US Department of Health and Human Services, Centers for Disease Control and Prevention, 2011. www.cdc.gov/healthyyouth/profiles/2010/profiles_report.pdf/. 2013 February 9.

17.

Austin

, Melly

, Sanchez

et al. Clustering of fast-food restaurants around schools: A novel application of spatial statistics to the study of food environments. Am J Public Health, 2005; 95:1575–1581.

18.

Gebauer

, Laska

. Convenience stores surrounding urban schools: An assessment of healthy food availability, advertising, and product placement. J Urban Health, 2011; 88:616–622.

19.

US Department of Agriculture, Agricultural Research Service, Beltsville Human Nutrition Research Center. Snacking patterns of US adolescents: What we eat in America, NHANES 2005–2006. Food Surveys Research Group Dietary Data Brief, 2010. http://ars.usda.gov/Services/docs.htm?docid=19476/. 2013 February 9.

20.

Nielsen

, Siega-Riz

, Popkin

. Trends in food locations and sources among adolescents and young adults. Prev Med, 2002; 35:107–113.

21.

Keast

, Nicklas

, O'Neil

. Snacking is associated with reduced risk of overweight and reduced abdominal obesity in adolescents: National Health and Nutrition Examination Survey (NHANES) 1999–2004. Am J Clin Nutr, 2010; 92:428–435.

22.

Sebastian

, Cleveland

, Goldman

. Effect of snacking frequency on adolescents' dietary intakes and meeting national recommendations. J Adolesc Health, 2008; 42:503–511.

23.

Dwyer

, Evans

, Stone

et al. Adolescents' eating patterns influence their nutrient intakes. J Am Diet Assoc, 2001; 101:798–801.

24.

O'Toole

, Anderson

, Miller

et al. Nutrition services and foods and beverages available at school: Results from the School Health Policies and Programs Study 2006. J Sch Health, 2007; 77:500–521.

25.

Finkelstein

, Hill

, Whitaker

. School food environments and policies in US public schools. Pediatrics, 2008; 122:e251–e259.

26.

Johnston

, O'Malley

, Terry-McElrath

et al. School policies and practices to improve health and prevent obesity: National Secondary School Survey results, school years 2006–2007 and 2007–2008. Volume 1 Executive Summary. Bridging the Gap Program, Survey Research Center, Institute for Social Research: Ann Arbor, MI, 2011.

27.

Kipke

, Iverson

, Moore

et al. Food and park environments: Neighborhood-level risks for childhood obesity in East Los Angeles. J Adolesc Health, 2007; 40:325–333.

28.

Borradaile

, Sherman

, Vander Veur

et al. Snacking in children: The role of urban corner stores. Pediatrics, 2009; 124:1293–1298.

29.

Tester

, Yen

, Pallis

et al. Healthy food availability and participation in WIC (Special Supplemental Nutrition Program for Women, Infants, and Children) in food stores around lower- and higher-income elementary schools. Public Health Nutr, 2011; 14:960–964.

30.

Wouters

, Larsen

, Kremers

et al. Peer influence on snacking behavior in adolescence. Appetite, 2010; 55:11–17.

31.

Mercille

, Receveur

, Macaulay

. Are snacking patterns associated with risk of overweight among Kahnawake schoolchildren? Public Health Nutr, 2010; 13:163–171.

32.

Sun

, Sekine

, Kagamimori

. Lifestyle and overweight among Japanese adolescents: The Toyama Birth Cohort Study. J Epidemiol, 2009; 19:303–310.

33.

Kerr

, Rennie

, McCaffrey

et al. Snacking patterns among adolescents: A comparison of type, frequency and portion size between Britain in 1997 and Northern Ireland in 2005. Br J Nutr, 2009; 101:122–131.

34.

Vader

, Walters

, Harris

et al. Television viewing and snacking behaviors of fourth- and eighth-grade schoolchildren in Texas. Prev Chronic Dis, 2009; 6:A89.

35.

McDonald

, Baylin

, Arsenault

et al. Overweight is more prevalent than stunting and is associated with socioeconomic status, maternal obesity, and a snacking dietary pattern in school children from Bogota, Colombia. J Nutr, 2009; 139:370–376.

36.

Maffeis

, Grezzani

, Perrone

et al.

Could the savory taste of snacks be a further risk factor for overweight in children?

J Pediatr Gastroenterol Nutr, 2008; 46:429–437.

37.

Aounallah-Skhiri

, Romdhane

, Traissac

et al. Nutritional status of Tunisian adolescents: Associated gender, environmental and socioeconomic factors. Public Health Nutr, 2008; 11:1306–1317.

38.

Lioret

, Touvier

, Lafay

et al.

Are eating occasions and their energy content related to child overweight and socioeconomic status?

Epidemiol, 2008; 16:2518–2523.

39.

Dubois

, Farmer

, Girard

et al. Social factors and television use during meals and snacks is associated with higher BMI among pre-school children. Public Health Nutr, 2008; 11:1267–1279.

40.

Toschke

, Kuchenhoff

, Koletzko

et al. Meal frequency and childhood obesity. Obes Res, 2005; 13:1932–1938.

41.

Huang

, Howarth

, Lin

et al. Energy intake and meal portions: Associations with BMI percentile in US children. Obes Res, 2004; 12:1875–1885.

42.

Nicklas

, Yang

, Baranowski

et al. Eating patterns and obesity in children: The Bogalusa Heart Study. Am J Prev Med, 2003; 25:9–16.

43.

Sekine

, Yamagami

, Hamanishi

et al. Parental obesity, lifestyle factors and obesity in preschool children: Results of the Toyama Birth Cohort Study. J Epidemiol, 2002; 12:33–39.

44.

Maffeis

, Provera

, Filippi

et al. Distribution of food intake as a risk factor for childhood obesity. Int J Obes Relat Metab Disord, 2000; 24:75–80.

45.

Cheah

, Chang

, Rosalia

et al. The relationship between media use and body mass index among secondary students in Kuching South City, Sarawak, Malaysia. Malaysian J Med Sci, 2011; 18:33–42.

46.

Gregori

, Foltran

, Ghidina

et al. The “snacking child” and its social network: Some insights from an Italian survey. Nutr J, 2011; 10:132.

47.

Kuriyan

, Thomas

, Smithra

et al. Potential factors related to waist circumference in urban South Indian children Indian. Pediatr, 2012; 49:124–128.

48.

Maruapula

, Jackson

, Holsten

et al. Socio-economic status and urbanization are linked to snacks and obesity in adolescents in Botswana. Public Health Nutr, 2011; 14:2260–2267.

49.

Watanabe

, Lee

, Kawakubo

. Associations of maternal employment and three-generation families with pre-school children's overweight and obesity in Japan. Int J Obes, 2011; 35:945–952.

50.

Isacco

, Lazaar

, Ratel

et al. The impact of eating habits on anthropometric characteristics of French primary school children. Child Care Health Dev, 2010; 36:835–842.

51.

de Novaes

, Francheschini Sdo

, Priore

. Mother's overweight, parents' constant limitation on the foods and frequent snack as risk factors for obesity among children in Brazil. Arch Latinoam Nutr, 2008; 58:256–264.

52.

Tanasescu

, Ferris

, Himmelgreen

et al. Biobehavioral factors are associated with obesity in Puerto Rican children. J Nutr, 2000; 130:1734–1742.

53.

Field

, Austin

, Gillman

et al. Snack food intake does not predict weight change among children and adolescents. Int J Obes Relat Metab Disord, 2004; 28:1210–1216.

54.

Phillips

, Bandini

, Naumova

et al. Energy-dense snack food intake in adolescence: Longitudinal relationship to weight and fatness. Obes Res, 2004; 12:461–472.

55.

Sugimori

, Yoshida

, Izuno

et al. Analysis of factors that influence body mass index from ages 3 to 6 years: A study based on the Toyama cohort study. Pediatr Int, 2004; 46:302–310.

56.

Francis

, Lee

, Birch

. Parental weight status and girls' television viewing, snacking, and body mass indexes. Obes Res, 2003; 11:143–151.

57.

Cutler

, Flood

, Hannan

et al. Association between major patterns of dietary intake and weight status in adolescents. Br J Nutr, 2012; 108:349–356.

58.

Gubbels

, Kremers

, Goldbohm

et al. Energy balance-related behavioral patterns in 5-year-old children and the longitudinal association with weight status development in early childhood. Public Health Nutr, 2011; 15:1402–1410.

59.

Oellingrath

, Svendsen

, Brantsaeter

. Tracking of eating patterns and overweight—a follow-up study of Norwegian schoolchildren from middle childhood to early adolescence. Nutr J, 2011; 10:106.

60.

Gregori

, Foltran

, Ghidina

et al. Understanding the influence of the snack definition on the association between snacking and obesity: A review. Int J Food Sci Nutr, 2011; 62:270–275.

61.

Johnson

, Anderson

. Snacking definitions: Impact on interpretation of the literature and dietary recommendations. Crit Rev Food Sci Nutr, 2010; 50:848–871.

62.

Wang

, Orleans

, Gortmaker

. Reaching the Healthy People goals for reducing childhood obesity: Closing the energy gap. Am J Prev Med, 2012; 42:437–444.

63.

Wang

, Gortmaker

, Sobol

et al. Estimating the energy gap among US children: A counterfactual approach. Pediatrics, 2006; 118:e1721–1733.