Physiology & Behavior xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Physiology & Behavior journal homepage: www.elsevier.com/locate/physbeh
Snacking: A cause for concern Richard D. Mattes
Department of Nutrition Science, Purdue University, West Lafayette, IN 47967, United States
A R T I C L E I N F O
A B S T R A C T
Keywords: Snacking Obesity Appetite Energy balance Food intake Eating frequency
Snacking, like any dietary behavior, can be practiced in a manner that is healthful or not. The case presented in this critical review of the literature is that snacking is problematic, primarily due to its contribution to positive energy balance and promotion of overweight/obesity. There is strong evidence that snacking is associated with greater energy intake. How this translates to body weight is less clear, largely due to limitations of experimental measurement tools and research designs. Correction for these shortcomings reveals evidence implicating snacking in the high prevalence of overweight/obesity supported by multiple plausible mechanisms. Given the popularity of snacking and its potential to positively contribute to diet quality, it is recommended that eﬀorts be made to better understand and harness snacking to a better purpose.
The fast and increasing pace of life is facilitated by the wide availability of convenience foods. Items that are palatable, nutritious, aﬀordable, capable of being consumed quickly and preferably while engaged in other activities, are desired by consumers. Snacks can meet all of these criteria, but often emphasize only a sub-set of characteristics resulting in questions about their role in a healthful diet. Assigning positive or negative attributes to snacking is diﬃcult when there is no agreed upon deﬁnition of the ingestive event. The implications of varying eating frequency, timing of ingestive events and/or dietary properties (e.g., sensory stimulation, nutrient contribution) that may stem from snacking can be antithetical and will diﬀer between individuals. For example, snacking-related improvements in diet quality may be accompanied by increased energy intake and the consequences will diﬀer if one has a diet that is nutrient rich or poor and is already contributing to positive or negative energy balance. Nevertheless, the position of this review is that in local and global environments where overweight and obesity are prevalent, snacking poses a cause for concern. Total energy intake is determined by the balance between eating frequency and portion size. Under a regulated homeostatic model, there is a reciprocal relationship whereby a healthful body weight can be maintained with changes in either component. Increased frequency of energy intake can be oﬀset by reduced portion sizes and vice versa. However, precise compensation is the exception rather than the rule as evidenced by the high global prevalence of overweight/obesity [1,2]. Snacking is not synonymous with increased eating frequency due to meal (also not clearly deﬁned) skipping, but generally leads to increased ingestive events. If three meals per day is normative, there has
been a marked change in dietary pattern. Globally, sales of snack foods are expected to exceed US$630 billion by 2020 . Europe is the largest market. A recent review of 27 centers in 10 European countries revealed all had eating frequencies in the range of 4.9 to 7.0 occasions per day . In the Mediterranean, Nordic and Central European countries, snacks contributed 14%, 29% and 31% of daily energy, respectively. For the Mediterranean countries, snack energy equaled energy contributed by breakfast and in the Nordic and Central European countries, snacking provided more energy than either breakfast or lunch. In 2012, Mexicans consumed 1.6 snacks per day which contributed 343 kcal/d or approximately 17% of total energy . In the United States, analyses of National Health and Nutrition Examination Survey (NHANES) data between 1971–1974 and 2007–2010 revealed energy from main meals increased by 63 kcal/d and 112 kcal/d in males and females, respectively whereas energy from snacks increased by 132 kcal/d and 142 kcal/d in males and females . As a percent of daily energy intake, energy from snacks increased by 3% in males and by 5% in females over this time period. Approximately 9% of the US population derives > 50% of their daily energy from snacks. This percentage is slightly lower than the peak in the 1999–2002 survey, but is still markedly higher than the approximately 5% reported in 1971–1974. Though smaller in absolute terms than western nations, the growth of snack sales is greater in developing countries (e.g., Asia-Paciﬁc, Latin America, Middle East/Africa) . In summary, snacking has increased in frequency, is highly prevalent and contributes 15–30% of daily energy in the US, European countries and elsewhere. If total daily energy intake is rising and is disproportionately due to increased energy intake from snacking, it might be expected that there
Corresponding author at: Department of Nutrition Science, Purdue University, 700 W State Street, Stone Hall 212, West Lafayette, IN 47907-2059, United States. E-mail address: [email protected]
https://doi.org/10.1016/j.physbeh.2018.02.010 Received 21 November 2017; Received in revised form 16 January 2018; Accepted 3 February 2018 0031-9384/ © 2018 Elsevier Inc. All rights reserved.
Please cite this article as: Mattes, R.D., Physiology & Behavior (2018), https://doi.org/10.1016/j.physbeh.2018.02.010
Physiology & Behavior xxx (xxxx) xxx–xxx
exerted little impact on appetite and BMI . However, with respect to appetite, none of the 12 studies summarized had a sample size > 20 and in six of the trials, it was < 10. Additionally, in nine of the 12 studies, the duration of observation was less than one day. Given the high variability in appetitive sensation, it is not clear that studies of such limited power and short duration yield reliable ﬁndings. The ﬁndings related to BMI are equally suspect. In eleven of the eighteen studies summarized, the duration of study was < 4 weeks and for sixteen of the eighteen studies, the duration of study was less than ≤8 weeks. This is a very short time to monitor body weight changes in response to dietary interventions. In the one study of one year's duration , there was an attrition rate of 34% and the three meal plus three snack group only increased snack intake by 0.4 snacks per day. The ﬁnding was no eﬀect of snacking on body weight. In summary, it is argued that due to their lower statistical power and perhaps poor compliance to intervention, published randomized controlled trials provide an inadequate basis for drawing conclusions about the eﬀects of snacking on BMI. To further aid evaluation of the potential contribution of snacking to positive energy balance and weight gain, mechanistic studies should be considered. Why should snacking be especially problematic for weight gain? First, the food industry has been especially responsive to consumer demands for products that are palatable, convenient and reasonably priced. This has been coupled with a shift in culture where eating at non-traditional times and in non-traditional locations has gained social acceptability. The American Time Use Survey reveals men and women engage in primary eating (eating is the main focus of their activity) for just over one hour per day, but secondary eating (eating while engaged in other activities) occurs for another 20 min per day. Moreover, secondary drinking occurs for an additional 57 min per day in males and 69 min in females . Secondary drinking is signiﬁcantly, positively associated with BMI. Additionally, snacking is associated with higher food reward sensitivity impulsivity . Thus, the high palatability and convenience of snacks may predispose selected individuals to snacking and in these individuals, snacking is associated with BMI. This may be especially problematic when snacking is initiated in the absence of hunger [41,42]. Second, snacks exert limited eﬀects on hunger, desire to eat and fullness. Indeed, they may actually enhance these sensations. Given that snacks are generally lower in energy and smaller in volume than meals, it is not surprising that the reduction of hunger before and after each eating occasion is smaller for snacks . However, when holding energy and volume constant and just varying eating events, similar ﬁndings are obtained. In a standard preload design trial, individuals were provided a single meal on one occasion or the same meal divided into 4 equal portions provided over a 180-minute period and were then given access to an ad libitum test meal at 240 min . All appetitive indices were less modiﬁed (e.g., less reduction of hunger or enhancement of fullness) by the higher eating frequency intervention over the initial 180 min. This was reversed at the 240-minute time point, but this had no eﬀect on energy intake at the test meal. In another trial, participants were acclimated to eating either three times per day or eight times per day for 3 weeks prior to testing . The low frequency eaters were provided a single meal and the high frequency eaters were provided the same meal divided into two and were presented 2.5 h apart. The composite appetite score was higher in the high frequency eaters indicating this eating pattern is not superior at moderating the sensations that drive feeding. In yet another trial, providing participants common meals divided into three or six eating occasions resulted in higher 24-hr hunger and desire to eat AUC values compared to the three eating occasion intervention. This was due to lesser reductions of these two sensations following each eating occasion with higher eating frequency and comparable rebound sensation levels . This response pattern has been described previously . Third, snacks tend to elicit weak energy compensation (i.e., adjustments to later intake to oﬀset the contribution of energy by the
would be a robust association between snacking and BMI. There are surprisingly limited data on this issue prompting a call for increased research on the topic by the 2010 Dietary Guidelines Advisory Committee . Nevertheless, there are data to this eﬀect (e.g., [9–16]). Findings from the Adventist Health Study 2 is a notable example . These data are particularly strong because they draw from a large sample of individuals (N = 50,660) with generally healthy lifestyles (e.g., low alcohol intake, higher physical activity, high prevalence of vegetarianism) who were studied longitudinally (7 years). They show a signiﬁcant linear association between number of eating occasions and increase in BMI. However, there are also multiple studies reporting no [17–19] or an equivocal  association, an inverse relationship [21–25], a link only when snacks are of high energy density [26,27] or only in individuals with high BMI [26,28,29]. A recent meta-analysis reported an inverse association between eating frequency and fat mass and percent body fat in children, but noted this was attributable to just a single study . Not uncommonly, null or inverse associations between snacking frequency and BMI are noted in observational studies where there is a positive association between snacking and energy intake (e.g., [27,31]). Given that body weight can be measured quite accurately and reliably, but the same cannot be said for food intake in free-living individuals [32,33], this raises the question as to whether the variability in epidemiological studies stems from under-reporting. There is considerable evidence to indicate this is the case. One of the ﬁrst and most inﬂuential reports of an inverse association between eating frequency and BMI was a cross-sectional assessment of Hungarian men . There was a sharp monotonic decline of BMI with reported eating frequency ranging from < 2X/d to > 7X/d. However, as noted by Bellisle et al. , there was also step-wise under-reporting of energy intake with the error greatest for those claiming to eat < 2X/d. This suggests those claiming to eat least frequently either failed to report eating events or actually skipped eating events as a way to moderate their energy intake. In either case, it would challenge the claim of an inverse association between eating frequency and BMI. Under-reporting is also evident in nationally representative data sets of the US population. Using the full Continuing Survey of Food Intake in Individuals (CSFII) data set (N ~ 6500), no association was observed between reported energy intake or eating frequency and BMI . However, with exclusion of implausible reporters, a signiﬁcant positive association is observed for both energy intake and eating frequency. Even more striking are the ﬁndings using NHANES data. Using the unadjusted sample, there was a null or inverse association between overall eating frequency or snacking frequency (deﬁned by energy or eating occasions per day) and risk for overweight/obesity in males (N = 9397; ≥20 y/o) and females (N = 9568; ≥20 y/o) . However, when excluding under-reporters, the associations were positive. The same phenomenon holds for analyses with children (N = 4346; 6–11 y/o) and adolescents (N = 6338; 12–19 y/o). Every negative association switched to positive or was attenuated and positive associations strengthened when corrected for intake plausibility. Similar results were reported with data from the National Diet and Nutrition Survey involving 1487 British adults (19–64 y/o). The deﬁnition of snacking can also lead to discrepant ﬁndings. The INTRERMAP study involving 2696 men and women 40–59 y/o between 1996 and 1999 reported an inverse association between number of eating occasions per day and BMI, but excluded all ingestive events comprised of beverages alone . Given the high level of energy beverages provide and the fact that they are commonly ingested alone  this undermines the ﬁndings. Thus, when critically assessed, the apparent inconsistency between snacking and BMI is due, to a large degree, to biases introduced by under-reporting energy intake and questionable deﬁnitions of snacks. When corrected, a positive association is observed more consistently. Inconsistencies are also reported in clinical trials and warrant critical assessment. A recent review of clinical trials evaluating the eﬀect of varying eating frequency on body weight concluded eating frequency 2
Physiology & Behavior xxx (xxxx) xxx–xxx
so when the patterns entailed 9 eating events per day. This study was then repeated with women with obesity and the same ﬁndings were obtained . The thermogenic response was lower with the irregular eating pattern and energy intake was signiﬁcantly higher. These ﬁndings raise particular concern for snacking that is unplanned; practiced by about 58% of snackers globally . Where snacking becomes incorporated in a usual eating pattern, it is less problematic . It should be acknowledged that the eﬀects of high eating frequency on thermogenesis are small in magnitude. It is argued that cumulatively they may exert an eﬀect over time intervals, but this has not been adequately tested. Alternatively, this eﬀect might be expected to reduce predicted weight loss. This has not been observed [62,63], but again, sample sizes and study durations have been less than optimal for detecting such eﬀects. Related to this, an inverse association between the thermic eﬀect of foods and appetitive sensations has been reported  which would suggest the lower thermogenic eﬀect of snacking maybe associated with greater motivation to eat. Thus, there are multiple plausible mechanisms to support the hypothesis that snacking is especially problematic for management of appetite, energy intake and BMI. They alone do not make a deﬁnitive case, but in the context of the totality of the data, they strengthen the case. However, it must be acknowledged, that this review has been one dimensional, i.e., focused on snacking and energy balance. This is too simplistic a consideration. There are data indicating that snacking contributes meaningfully to overall diet quality (e.g., [25,65–68]). The latter is important, but also too limited a consideration. The balance between positive and negative eﬀects provides a better assessment of the role of snacking in a healthful diet. The Healthy Aging in Neighborhoods of Diversity Across the Lifespan study provides the evidence base to consider this balance . In a sample of 2177 adults, snacking was associated with higher Healthy Eating Index scores. Among African American and White men, the score increased by approximately 2% and 4%, respectively. Among African American and White women, the increments were about 1% and 7%, respectively. However, these HEI score increases were accompanied by increases of 483 kcal in African men, 583 kcal in White men, 365Kcal in African women and 468 kcal in White women. In populations where overweight and obesity are prevalent, the question is whether the trade-oﬀ of a few percent increase in nutrient quality justiﬁes such a large increase in energy intake. In summary, it is not the present position that snacking is inherently problematic. Snacks can be incorporated in healthful diets. However, to do so requires knowledge and vigilance as spontaneous adjustments to oﬀset the energy provided by snacks is not evident and this is a critical issue when two thirds of the global population is overweight or obese. The data reviewed shows snacking is associated with increased energy intake and BMI. Where this is not reported in epidemiological studies, it is often due to the inaccuracy of dietary records. Randomized clinical trials are equivocal, largely because, to-date, they are underpowered. There are multiple plausible mechanisms to support the view that snacking contributes to positive energy balance. Although snacking can contribute important nutrients, this often comes with an energy cost that negatively outweighs the positive contribution to diet quality. Snacking is a relatively new behavior but one that is likely to persist. Learning how to convert it to a positive ingestive behavior should be a high priority.
snack). While snacks do elicit a reduction of hunger or desire to eat and augment fullness acutely, generally the eﬀect size is limited and transient. Consequently, they may extinguish prior to the next scheduled eating event and thereby exert no inﬂuence . This has been documented in controlled eating trials. For example, in one instance , participants were provided a standard lunch and snack that varied by macronutrient composition. Dinner was provided ad libitum to determine the degree of dietary compensation. No signiﬁcant adjustment for any snack was observed. The ingestion of beverages as snacks may be especially problematic because of their low satiety value . Thus, snacks tend to add, rather than displace energy in the diet. This is a well-recognized and exploited phenomenon. There are numerous conditions where it is desirable to increase the energy intake of individuals. Patients with early satiety, food intolerance, dumping syndrome, cirrhosis and/or are losing weight unintentionally are good examples. The common practice is to recommend multiple small meals to boost intake rather than to suggest these individuals attempt to force-feed fewer larger meals. Early satiety can be mitigated, somewhat, by adopting an eating pattern of multiple small eating occasions. Fourth, it is possible that snacking serves to dampen the eﬀects of signaling systems believed to modulate appetite and feeding. Signaling systems generally respond to changes in the conditions to which they are tuned. For example, periodic eating typically results in excursions of blood sugar which, in turn drives the release of the appetite suppressing hormone, insulin. In a trial where participants either fasted, ate two times or twelve times over 8-hr trials, plasma insulin was low and stable in the fasting state, elevated, but stable in the high eating frequency condition and variable with low eating frequency. This lack of insulin variability may itself disrupt the appetitive signaling leading to disordered eating  or it may do so by aﬀecting other signals. In the trial, the absence of a post-prandial insulin spike was associated with the failure of ghrelin, a reported appetite stimulating hormone, to decline post-prandially. The result may be greater hunger and subsequent intake. Fifth, snacking in the evening may contribute to ﬁndings that eating timing inﬂuences BMI. Though the data are mixed [51–53] some work indicates eating in the evening is associated with higher BMI . It is unlikely this is actually due to the timing of intake as there are cultures, such as Spain, where a large meal is consumed late in the evening and they are not necessarily the cultures with the highest prevalence of overweight/obesity . Rather, the association between eating later in the day is mediated by the increased eating frequency snacking generally provides . Sixth, diets with higher eating frequency may be associated with lower thermogenic responses. While there are trials that fail to note an eﬀect of eating frequency on thermogenesis [56,57], a number indicate there is an inverse association. In and early trial, administration of a liquid meal as a single bolus versus 3-hr nasogastric feed led to a lower thermogenic response with the latter treatment . This was followed by work with solid meals. In one trial, participants consumed the same meal as a single event or in six equal portions at 30 min intervals over a three-hour period . The thermogenic response to feeding was signiﬁcantly lower in the multiple meal condition. In another trial, participants were fed either two meals per day or seven meals per day (both patterns matched on energy) for two-day periods. Thermogenesis was higher during the post-prandial period with the lower eating frequency, but overall daily energy intake was not signiﬁcantly diﬀerent. One factor that may account for the variability in responses is the participant's customary intake pattern. This has not been adequately controlled in prior work. Support for this view draws from trials comparing the thermogenic eﬀects of eating six times per day for 14 days versus having a variable eating pattern with 3–9 eating occasions per day in healthy, lean women . The irregular eating pattern was associated with a lower thermogenic eﬀect of feeding relative to baseline and the post intervention with the regular eating pattern. Energy intake also tended to be higher with the irregular eating pattern, signiﬁcantly
References  V.S. Malik, W.C. Willett, F.B. Hu, Global obesity: trends, risk factors and policy implications, Nat. Rev. Endocrinol. 9 (2013) 13–27.  M. Ng, T. Fleming, M. Robinson, B. Thomson, N. Graetz, C. Margono, E.C. Mullany, S. Biryukov, C. Abbafati, S.F. Abera, et al., Global, regional, and national prevalence of overweight and obesity in children and adults during 1980–2013: a systematic analysis for the Global Burden of Disease Study 2013, Lancet 384 (2014) 766–781.  Foods salted baked frozen snacks market trends.asp, http://www.strategyr.com/ MarketResearch/Snack.  E. Huseinovic, A. Winkvist, N. Slimani, M.K. Park, H. Freisling, H. Boeing, G. Buckland, L. Schwingshackl, E. Weiderpass, A.L. Rostgaard-Hansen, et al., Meal
Physiology & Behavior xxx (xxxx) xxx–xxx
   
  
patterns across ten European countries - results from the European Prospective Investigation into Cancer and Nutrition (EPIC) calibration study, Public Health Nutr. 19 (2016) 2769–2780. K.J. Duﬀey, J.A. Rivera, B.M. Popkin, Snacking is prevalent in Mexico, J. Nutr. 144 (2014) 1843–1849. A.K. Kant, B.I. Graubard, 40-year trends in meal and snack eating behaviors of American adults, J. Acad. Nutr. Diet. 115 (2015) 50–63. http://www.nielsen.com/content/dam/corporate/us/en/reports-downloads/ 2014%20Reports/nielsen-global-snacking-report-september-2014.pdf. E.F. Myers, C.S. Khoo, W. Murphy, A. Steiber, S. Agarwal, A critical assessment of research needs identiﬁed by the dietary guidelines committees from 1980 to 2010, J. Acad. Nutr. Diet. 113 (2013) 957–971 (e951). J.O. Fisher, L.L. Birch, Eating in the absence of hunger and overweight in girls from 5 to 7 y of age, Am. J. Clin. Nutr. 76 (2002) 226–231. T.A. Nicklas, S.J. Yang, T. Baranowski, I. Zakeri, G. Berenson, Eating patterns and obesity in children. The bogalusa heart study, Am. J. Prev. Med. 25 (2003) 9–16. S.M. Phillips, L.G. Bandini, E.N. Naumova, H. Cyr, S. Colclough, W.H. Dietz, A. Must, Energy-dense snack food intake in adolescence: longitudinal relationship to weight and fatness, Obes. Res. 12 (2004) 461–472. A.A. van der Heijden, F.B. Hu, E.B. Rimm, R.M. van Dam, A prospective study of breakfast consumption and weight gain among U.S. men, Obesity (Silver Spring) 15 (2007) 2463–2469. N.C. Howarth, T.T. Huang, S.B. Roberts, B.H. Lin, M.A. McCrory, Eating patterns and dietary composition in relation to BMI in younger and older adults, Int. J. Obes. 31 (2007) 675–684. C.M. McDonald, A. Baylin, J.E. Arsenault, M. Mora-Plazas, E. Villamor, 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. 139 (2009) 370–376. M. Bes-Rastrollo, A. Sanchez-Villegas, F.J. Basterra-Gortari, J.M. Nunez-Cordoba, E. Toledo, M. Serrano-Martinez, Prospective study of self-reported usual snacking and weight gain in a Mediterranean cohort: the SUN project, Clin. Nutr. 29 (2010) 323–330. H. Kahleova, J.I. Lloren, A. Mashchak, M. Hill, G.E. Fraser, Meal frequency and timing are associated with changes in body mass index in adventist health study 2, J. Nutr. 147 (2017) 1722–1728. A.E. Field, S.B. Austin, M.W. Gillman, B. Rosner, H.R. Rockett, G.A. Colditz, Snack food intake does not predict weight change among children and adolescents, Int. J. Obes. Relat. Metab. Disord. 28 (2004) 1210–1216. T.A. Nicklas, M. Morales, A. Linares, S.J. Yang, T. Baranowski, C. De Moor, G. Berenson, Children's meal patterns have changed over a 21-year period: the Bogalusa Heart Study, J. Am. Diet. Assoc. 104 (2004) 753–761. T.L. Barnes, S.A. French, L.J. Harnack, N.R. Mitchell, J. Wolfson, Snacking behaviors, diet quality, and body mass index in a community sample of working adults, J. Acad. Nutr. Diet. 115 (2015) 1117–1123. E.W. Evans, P.F. Jacques, G.E. Dallal, J. Sacheck, A. Must, The role of eating frequency on relative weight in urban school-age children, Pediatr. Obes. 10 (2015) 442–447. P. Fabry, S. Hejda, K. Cerny, K. Osancova, J. Pechar, Eﬀect of meal frequency in schoolchildren. Changes in weight-height proportion and skinfold thickness, Am. J. Clin. Nutr. 18 (1966) 358–361. M.S. Westerterp-Plantenga, E.M. Kovacs, K.J. Melanson, Habitual meal frequency and energy intake regulation in partially temporally isolated men, Int. J. Obes. Relat. Metab. Disord. 26 (2002) 102–110. D.R. Keast, T.A. Nicklas, C.E. 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. 92 (2010) 428–435. B.T. House, G.E. Shearrer, S.J. Miller, K.E. Pasch, M.I. Goran, J.N. Davis, Increased eating frequency linked to decreased obesity and improved metabolic outcomes, Int. J. Obes. 39 (2015) 136–141. G.S. Aljuraiban, Q. Chan, L.M. Oude Griep, I.J. Brown, M.L. Daviglus, J. Stamler, L. Van Horn, P. Elliott, G.S. Frost, Group IR, The impact of eating frequency and time of intake on nutrient quality and body mass index: the INTERMAP Study, a Population-Based Study, J. Acad. Nutr. Diet. 115 (2015) 528–536 (e521). L. O'Connor, S. Brage, S.J. Griﬃn, N.J. Wareham, N.G. Forouhi, The cross-sectional association between snacking behaviour and measures of adiposity: the Fenland Study, UK, Br. J. Nutr. 114 (2015) 1286–1293. N.I. Larson, J.M. Miller, A.W. Watts, M.T. Story, D.R. Neumark-Sztainer, Adolescent snacking behaviors are associated with dietary intake and weight status, J. Nutr. 146 (2016) 1348–1355. K. Keller, S. Rodriguez Lopez, M.M. Carmenate Moreno, Association between meal intake behaviour and abdominal obesity in Spanish adults, Appetite 92 (2015) 1–6. L.S. Taillie, D. Wang, B.M. Popkin, Snacking is longitudinally associated with declines in body mass index z scores for overweight children, but increases for underweight children, J. Nutr. 146 (2016) 1268–1275. B.J. Schoenfeld, A.A. Aragon, J.W. Krieger, Eﬀects of meal frequency on weight loss and body composition: a meta-analysis, Nutr. Rev. 73 (2015) 69–82. H. Berteus Forslund, J.S. Torgerson, L. Sjostrom, A.K. Lindroos, Snacking frequency in relation to energy intake and food choices in obese men and women compared to a reference population, Int. J. Obes. 29 (2005) 711–719. N.V. Dhurandhar, D. Schoeller, A.W. Brown, S.B. Heymsﬁeld, D. Thomas, T.I. Sorensen, J.R. Speakman, M. Jeansonne, D.B. Allison, Energy Balance Measurement Working G: energy balance measurement: when something is not better than nothing, Int. J. Obes. 39 (2015) 1109–1113. E. Archer, G.A. Hand, S.N. Blair, Validity of U.S. nutritional surveillance: National
  
Health and Nutrition Examination Survey caloric energy intake data, 1971–2010, PLoS One 8 (2013) e76632. F. Bellisle, R. McDevitt, A.M. Prentice, Meal frequency and energy balance, Br. J. Nutr. 77 (Suppl. 1) (1997) S57–70. M.A. McCrory, N.C. Howarth, S.B. Roberts, T.T. Huang, Eating frequency and energy regulation in free-living adults consuming self-selected diets, J. Nutr. 141 (2011) 148–153. K. Murakami, M.B. Livingstone, Eating frequency is positively associated with overweight and central obesity in U.S. adults, J. Nutr. 145 (2015) 2715–2724. A.K. Kant, B.I. Graubard, R.D. Mattes, Association of food form with self-reported 24-h energy intake and meal patterns in US adults: NHANES 2003–2008, Am. J. Clin. Nutr. 96 (2012) 1369–1378. A.T. Hutchison, L.K. Heilbronn, Metabolic impacts of altering meal frequency and timing - does when we eat matter? Biochimie 124 (2016) 187–197. H. Berteus Forslund, S. Klingstrom, H. Hagberg, M. Londahl, J.S. Torgerson, A.K. Lindroos, Should snacks be recommended in obesity treatment? A 1-year randomized clinical trial, Eur. J. Clin. Nutr. 62 (2008) 1308–1317. C.D. Zick, R.B. Stevens, W.K. Bryant, Time use choices and healthy body weight: a multivariate analysis of data from the American Time Use Survey, Int. J. Behav. Nutr. Phys. Act. 8 (2011) 84. S.H. Fay, M.J. White, G. Finlayson, N.A. King, Psychological predictors of opportunistic snacking in the absence of hunger, Eat. Behav. 18 (2015) 156–159. C. Marmonier, D. Chapelot, M. Fantino, J. Louis-Sylvestre, Snacks consumed in a nonhungry state have poor satiating eﬃciency: inﬂuence of snack composition on substrate utilization and hunger, Am. J. Clin. Nutr. 76 (2002) 518–528. F. Bellisle, A.M. Dalix, L. Mennen, P. Galan, S. Hercberg, J.M. de Castro, N. Gausseres, Contribution of snacks and meals in the diet of French adults: a dietdiary study, Physiol. Behav. 79 (2003) 183–189. X. Allirot, L. Saulais, K. Seyssel, J. Graeppi-Dulac, H. Roth, A. Charrie, J. Drai, J. Goudable, E. Blond, E. Disse, M. Laville, An isocaloric increase of eating episodes in the morning contributes to decrease energy intake at lunch in lean men, Physiol. Behav. 110-111 (2013) 169–178. M.M. Perrigue, A. Drewnowski, C.Y. Wang, M.L. Neuhouser, Higher eating frequency does not decrease appetite in healthy adults, J. Nutr. 146 (2016) 59–64. K. Ohkawara, M.A. Cornier, W.M. Kohrt, E.L. Melanson, Eﬀects of increased meal frequency on fat oxidation and perceived hunger, Obesity (Silver Spring) 21 (2013) 336–343. R.D. Mattes, Hunger and thirst: issues in measurement and prediction of eating and drinking, Physiol. Behav. 100 (2010) 22–32. M. Porrini, A. Santangelo, R. Crovetti, P. Riso, G. Testolin, J.E. Blundell, Weight, protein, fat, and timing of preloads aﬀect food intake, Physiol. Behav. 62 (1997) 563–570. D. Chapelot, The role of snacking in energy balance: a biobehavioral approach, J. Nutr. 141 (2011) 158–162. M.L. Power, J. Schulkin, Anticipatory physiological regulation in feeding biology: cephalic phase responses, Appetite 50 (2008) 194–206. A.K. Kant, R. Ballard-Barbash, A. Schatzkin, Evening eating and its relation to selfreported body weight and nutrient intake in women, CSFII 1985–86, J. Am. Coll. Nutr. 14 (1995) 358–363. A.K. Kant, A. Schatzkin, R. Ballard-Barbash, Evening eating and subsequent longterm weight change in a national cohort, Int. J. Obes. Relat. Metab. Disord. 21 (1997) 407–412. S. Eng, D.A. Wagstaﬀ, S. Kranz, Eating late in the evening is associated with childhood obesity in some age groups but not in all children: the relationship between time of consumption and body weight status in U.S. children, Int. J. Behav. Nutr. Phys. Act. 6 (2009) 27. S. Almoosawi, S. Vingeliene, L.G. Karagounis, G.K. Pot, Chrono-nutrition: a review of current evidence from observational studies on global trends in time-of-day of energy intake and its association with obesity, Proc. Nutr. Soc. 75 (2016) 487–500. K.J. Reid, K.G. Baron, P.C. Zee, Meal timing inﬂuences daily caloric intake in healthy adults, Nutr. Res. 34 (2014) 930–935. A.J. Smeets, M.S. Westerterp-Plantenga, Acute eﬀects on metabolism and appetite proﬁle of one meal diﬀerence in the lower range of meal frequency, Br. J. Nutr. 99 (2008) 1316–1321. A.Z. Belko, M. Van Loan, T.F. Barbieri, P. Mayclin, Diet, exercise, weight loss, and energy expenditure in moderately overweight women, Int. J. Obes. 11 (1987) 93–104. C.A. Nacht, Y. Schutz, O. Vernet, L. Christin, E. Jequier, Continuous versus single bolus enteral nutrition: comparison of energy metabolism in humans, Am. J. Phys. 251 (1986) E524–529. M.M. Tai, P. Castillo, F.X. Pi-Sunyer, Meal size and frequency: eﬀect on the thermic eﬀect of food, Am. J. Clin. Nutr. 54 (1991) 783–787. H.R. Farshchi, M.A. Taylor, I.A. Macdonald, Decreased thermic eﬀect of food after an irregular compared with a regular meal pattern in healthy lean women, Int. J. Obes. Relat. Metab. Disord. 28 (2004) 653–660. H.R. Farshchi, M.A. Taylor, I.A. Macdonald, Beneﬁcial metabolic eﬀects of regular meal frequency on dietary thermogenesis, insulin sensitivity, and fasting lipid proﬁles in healthy obese women, Am. J. Clin. Nutr. 81 (2005) 16–24. A.K. Kant, Evidence for eﬃcacy and eﬀectiveness of changes in eating frequency for body weight management, Adv. Nutr. 5 (2014) 822–828. J.D. Cameron, M.J. Cyr, E. Doucet, Increased meal frequency does not promote greater weight loss in subjects who were prescribed an 8-week equi-energetic energy-restricted diet, Br. J. Nutr. 103 (2010) 1098–1101. R. Crovetti, M. Porrini, A. Santangelo, G. Testolin, The inﬂuence of thermic eﬀect of food on satiety, Eur. J. Clin. Nutr. 52 (7) (1998) 482–488. C.A. Zizza, D.D. Arsiwalla, K.J. Ellison, Contribution of snacking to older adults'
Physiology & Behavior xxx (xxxx) xxx–xxx
and diet quality in US adults: National Health and Nutrition Examination Survey 2003–2012, J. Acad. Nutr. Diet. 116 (2016) 1101–1113.  M. Fanelli Kuczmarski, N. Cotugna, R.T. Pohlig, M.A. Beydoun, E.L. Adams, M.K. Evans, A.B. Zonderman, Snacking and diet quality are associated with the coping strategies used by a socioeconomically diverse urban cohort of AfricanAmerican and White Adults, J. Acad. Nutr. Diet. 117 (2017) 1355–1365.
vitamin, carotenoid, and mineral intakes, J. Am. Diet. Assoc. 110 (2010) 768–772.  C.A. Zizza, B. Xu, Snacking is associated with overall diet quality among adults, J. Acad. Nutr. Diet. 112 (2012) 291–296.  Z. Wang, F. Zhai, B. Zhang, B.M. Popkin, Trends in Chinese snacking behaviors and patterns and the social-demographic role between 1991 and 2009, Asia Pac. J. Clin. Nutr. 21 (2012) 253–262.  K. Murakami, M.B. Livingstone, Associations between meal and snack frequency