|Home | About | Journals | Submit | Contact Us | Français|
Total energy consumption among United States adults has increased in recent decades, and energy-containing beverages are a significant contributor to this increase. Because beverages are less satiating than solid foods, consumption of energy-containing beverages may increase energy intake and lead to weight gain; trends in food and beverage consumption coinciding with increases in overweight and obesity support this possibility. The purpose of this review is to present what is known about the effect of beverage consumption on short-term (i.e., meal) energy intake, as well as longer-term effects on body weight. Specific beverages addressed include water, other energy-free beverages (diet soft drinks, coffee and tea), and energy-containing beverages (soft drinks, juices and juice drinks, milk and soy beverages, alcohol). Existing evidence, albeit limited, suggests that encouraging water consumption, and substituting water and other energy-free beverages (diet soft drinks, coffee and tea) for energy-containing beverages may facilitate weight management. Energy-containing beverages acutely increase energy intake, however long-term effects on body weight are uncertain. While there may be health benefits for some beverage categories, additional energy provided by beverages should be compensated for by reduced consumption of other foods in the diet.
Caloric intake has increased over the past 20 years, with the majority of the increase derived from snacks foods (Nielsen, Siega-Riz, & Popkin, 2002) and beverages (Duffey & Popkin, 2007). Beverage consumption patterns in children and adolescents are an important target for promoting body weight management (Ludwig, Peterson, & Gortmaker, 2001; Sanigorski, Bell, & Swinburn, 2007), yet until recently, less attention has been directed at beverage consumption and adult weight management. The potential negative public health consequences of sweetened beverage consumption have led some to suggest a sugar-sweetened beverage tax (Brownell & Frieden, 2009), which is controversial due in part to a lack of direct evidence linking changes in sweetened beverage intake to weight outcomes (Allison & Mattes, 2009). Thus, the purpose of this article is to review literature addressing beverage consumption and adult weight management.
Though beverages of all types satisfy thirst (Almiron-Roig & Drewnowski, 2003), caloric (i.e., energy-containing) beverages do not effect food intake during acute meal settings (Almiron-Roig & Drewnowski, 2003; DellaValle, Roe, & Rolls, 2005; Harper, James, Flint, & Astrup, 2007; Tsuchiya, Almiron-Roig, Lluch, Guyonnet, & Drewnowski, 2006). However, when compared to water or energy-free, artificially-sweetened beverages, consumption of energy-containing beverages increases total meal energy intake (Almiron-Roig & Drewnowski, 2003; DellaValle et al., 2005). The low satiety value of beverages may be explained by the fact that for most of human evolutionary history, water was the only beverage consumed (Wolf, Bray, & Popkin, 2008); regulation of beverage energy was not of biological importance. Though the exact mechanism of this weak satiety response is unclear, there are several possibilities. First, relative to solid foods, beverages take less time to consume and are rapidly emptied from the stomach (Kissileff, 1985). This is supported by research demonstrating that consumption of an isocaloric liquid compared with solid food given prior to an ad libitum meal increases food intake (Stull, Apolzan, Thalacker-Mercer, Iglay, & Campbell, 2008). Second, food form (i.e. solid vs. liquid) influences appetite-related hormonal response (Tieken et al., 2007). Finally, energy-containing beverages are largely comprised of carbohydrates, which stimulate fewer satiety signals than fat or protein (Wolf et al., 2008).
Because energy from beverages is less satiating than consumption of solid foods (DiMeglio & Mattes, 2000; R. Mattes, 2006; R. D. Mattes, 1996; Stull et al., 2008), consumption of energy-containing beverages may produce weight gain (Ludwig et al., 2001). Trends in beverage consumption coinciding with increases in overweight and obesity support this possibility; energy intake from sweetened beverages (i.e., soft drinks and fruit drinks) has increased by 222 kcal in recent decades (Duffey & Popkin, 2007). Adults with a high energy-containing beverage intake consume less healthy diets (Duffey & Popkin, 2006; Popkin, Barclay, & Nielsen, 2005), while water consumers ingest less total energy (Duffey & Popkin, 2006; Popkin et al., 2005). Soft drinks are the single greatest energy source (i.e., 7% total energy) in the American diet (Block, 2004); average daily consumption of soft drinks is > 12.3 fl oz per person (Wolf et al., 2008). Therefore, beverage consumption patterns may be an important dietary factor influencing adult weight management.
This review was conducted using a PubMed (National Library of Medicine) online literature search to identify publications (within 15 years) investigating beverage consumption and either food/energy intake or body weight management. A small number of less recent publications were included if they were a key reference for later work on this topic. Articles were limited to those published in English, and in adults. Findings were categorized according to beverage type as follows: water, other energy-free beverages (artificially sweetened beverages, coffee and tea), and energy-containing beverages (soft drinks, juice and juice drinks, milk and soy beverages, alcohol). Details of studies reviewed are presented in the Table in alphabetical order by first author’s name within each subsection. Research findings are summarized within the text in the following order: epidemiological research, test meal/short-term feeding studies, and randomized controlled trials (if available). Relevant meta-analyses and reviews are also presented.
There is a common belief that water ingestion facilitates weight management. Epidemiological data indicate that energy intake among water drinkers is ~9% (194 kcal/d) lower than non-water drinkers (Popkin et al., 2005), yet few studies have directly addressed the effect of increased water consumption on weight management. Water consumed before or with a meal reduces sensations of hunger and increases satiety among nonobese adults (Lappalainen, Mennen, van Weert, & Mykkanen, 1993; Van Walleghen, Orr, Gentile, & Davy, 2007). However, young, normal weight men consume the same amount of energy at an ad libitum meal when given water (8 and 16 fl. oz) as compared to no beverage 30 minutes prior to the meal (Rolls et al., 1990). Pre-meal water consumption reduces meal energy intake in nonobese older adults (Van Walleghen et al., 2007), and also among middle-aged and older overweight/obese adults (Davy, Dennis, Dengo, Wilson, & Davy, 2008). Thus, it appears ingesting water (~16 fl oz) prior to a meal reduces meal energy intake among middle-aged and older adults by ~75-90 kcals, but may not impact meal energy intake among younger population segments (Davy, Dennis, Dengo et al., 2008; Van Walleghen et al., 2007).
Longer term studies have investigated changes in self-reported beverage intake to energy intake and weight during a weight loss intervention (see Table) (Stookey, Constant, Gardner, & Popkin, 2007). Over 12 months, drinking ≥1 L water (>33.8 fl oz) per day increased weight loss by ~ 2 kg as compared to drinking less water (Stookey, Constant, Popkin, & Gardner, 2008). However, these investigations relied on self-reported data and to date, no clinical intervention studies have assessed the influence of increased water consumption on weight loss in adults; the aforementioned intervention compared popular weight loss diets and water/beverage consumption was not the primary focus. Preliminary data suggests increasing daily water intake may facilitate long term weight loss among middle-aged and older adults (Davy, Dennis, Akers, Dengo, & Davy, 2008); a trend (p=0.09) was detected for greater weight loss after 6 months among individuals (n=20) following a hypocaloric diet along with three 500-ml bottled water per day (16.9 fl oz) as compared to those in a diet-only control condition. Additional long term studies with larger sample sizes are needed to confirm this possibility.
There may be other benefits of water consumption for weight management, such as increased metabolic rate (Boschmann et al., 2003; Brown, Dulloo, & Montani, 2006). Additional research is needed to confirm these findings, as well as to address issues of optimal volume and timing of water ingestion for managing hunger and reducing meal energy intake.
The Beverage Guidance Panel provided an overview of the nutritional benefits and risks of different beverage categories, and provided recommendations for beverage intake in relation to weight status and health (Popkin et al., 2006). These recommendations were controversial, in that not all were supported by scientific data (Kaplan, 2006; Weaver et al., 2006). Though there is not a Recommended Daily Allowance (RDA) for total water intake due to insufficient evidence and individual variability, the Institute of Medicine’s Food and Nutrition Board suggested a Daily Recommended Intake (DRI) for adult males and nonpregnant, nonlactating females over 19 years of 3.7 L/day and 2.7 L/day, respectively. However, this includes total water from all beverages such as coffee, tea, juices, soda and water, as well as moisture from foods (Panel on Dietary Reference Intakes for Electrolytes and Water, 2004). An Adequate Intake (AI) of 3.0 L/day for men and 2.2 L/day for women (~104 fl oz and 72 fl oz) of total beverages, including water, was also recommended. These guidelines recommend that individuals be attentive to thirst cues and consume beverages with meals to maintain adequate hydration status, but do not make specific recommendations pertaining to beverage consumption and weight management. Thus, epidemiological and acute meal studies suggests a beneficial role for water consumption in reducing energy intake and promoting weight management, yet there is insufficient data from intervention studies to make conclusive evidence-based intake recommendations for water consumption.
Approximately one fifth of US adults consume energy-free, artificially sweetened beverages, such as diet soft drinks and other artificially sweetened beverages (Duffey & Popkin, 2006). These beverages provide a source of water and sweetness in the diet without additional energy (Popkin et al., 2006). Concern about a negative effect of diet soft drink consumption on energy intake arose from animal studies reporting an increased food intake and weight gain following prolonged exposure to saccharin-sweetened yogurt (Swithers & Davidson, 2008). This suggested that artificial sweeteners may “uncouple” a relationship between sweet taste and energy content, which prompted rodents to consume more food and gain weight (Swithers & Davidson, 2008). Yet, in humans there are conflicting data on artificial sweetener intake and weight status and most available data are correlational; few intervention studies have investigated the influence of energy-free artificially-sweetened beverages on energy intake regulation and body weight. Specifically, habitual intake of artificially sweetened beverages is associated with self-reported BMI, and high consumers of artificially sweetened beverages (>28 fl oz/d) were more likely to report body weight concerns such as restrained eating and body dissatisfaction (Appleton & Conner, 2001). A dose-response relationship between artificially sweetened beverage consumption and BMI change over time has also been reported; odds of becoming overweight/obese for individuals consuming >22 artificially sweetened drinks per week was almost twice that of individuals consuming no artificially sweetened beverages (Fowler et al., 2008). However, those consuming artificially sweetened beverages were more likely to be female, dieting, and overweight/obese at baseline.
In contrast, epidemiological investigations have reported that women who increased diet soft drink consumption gained less weight over a four-year period than women who decreased diet soft drink consumption (Schulze et al., 2004). In addition, replacing energy-containing beverages with artificially sweetened beverages was associated with greater weight loss (~1.6 kg) over 12 months, which is slightly less than that reported with water (Stookey et al., 2008).
In ad libitum meal studies, artificially sweetened beverage preloads (i.e. prior to the ingestion of a meal) do not appear to stimulate appetite or affect subsequent food intake (Canty & Chan, 1991; Holt et al., 2000; Rodin, 1990), and short term intervention studies incorporating artificially sweetened beverages into the usual diet have demonstrated weight loss in both overweight (Raben, Vasilaras, Moller, & Astrup, 2002) and normal weight adults (Tordoff & Alleva, 1990). Thus, although epidemiological data in this area are conflicting, existing experimental/intervention trials do not suggest that energy-free artificially sweetened beverages increase food intake or body weight. These beverages could be used as an alternative to water as a substitution for energy-containing beverages, particularly for interventions targeting weight management.
Coffee and tea in their unaltered form contribute fluid to the diet without contributing energy. Increased coffee and tea consumption is associated with less weight gain over time in men and women; this observation is not attributed to beverage caffeine content (Lopez-Garcia et al., 2006). Adults who routinely (i.e., ≥1 time per week) consumed green, oolong or black tea for >10 years have lower body fat percentages and smaller waist circumferences than nonconsumers (Wu et al., 2003), and research has suggested that green tea in particular may play a role in weight loss and weight maintenance. Green tea contains catechins, particularly epigallocatechin gallate (EGCG), which in concentrated supplement form has been shown to increase fat oxidation (Auvichayapat et al., 2008; Dulloo et al., 1999; Venables, Hulston, Cox, & Jeukendrup, 2008), reduce body fat (Nagao, Hase, & Tokimitsu, 2007), and increase weight loss (Auvichayapat et al., 2008). However, several cups of tea must be consumed to achieve the benefits of the catechin levels found in concentrated forms. In contrast, daily energy intake among high consumers of popular energy-containing coffee and tea drinks (i.e. lattes, espressos, café mochas, cappuccinos) is 206 kcal higher than nonconsumers (Shields, Corrales, & Metallinos-Katsaras, 2004). Thus, while there may be some benefits to consuming coffee and tea in their energy-free form, consumption of energy-containing coffee and tea drinks may increase total energy intake.
Regular (sweetened) soft drinks are the greatest contributor of energy in the American diet (Block, 2004). A recent meta-analysis (Vartanian, Schwartz, & Brownell, 2007) reported a clear association between soft drink consumption and increased energy intake, and previous reviews have concluded that a strong association exists between sweetened soft drink consumption and risk of overweight/obesity (Malik, Schulze, & Hu, 2006). However, others suggest that dietary behaviors (i.e., food choices, dietary restraint, energy intake, beverage type and usage) and economics (i.e., poverty, food and beverage costs) contribute to obesity more than sugar-sweetened beverages (Drewnowski & Bellisle, 2007). Increased intake of soft drinks is associated with BMI (Lin, Huang, & French, 2004), weight gain and the risk of type 2 diabetes (Schulze et al., 2004), and risk factors for the metabolic syndrome (Dhingra et al., 2007). Soft drink consumers (>4 servings/wk) have higher total energy intakes and engage in less physical activity than low soft drink consumers (≤3 servings/wk) (Kvaavik, Andersen, & Klepp, 2005; Schulze et al., 2004). Longitudinal studies report that women who decrease their intake of soft drinks over a four-year period gain less weight than those who maintain or increase their soft drink intake (Schulze et al., 2004). Intervention studies indicate that consuming soft drinks for 3-weeks increases usual energy intake and body weight (Tordoff & Alleva, 1990), and although data are conflicting, some acute meal studies suggest an increase in total meal energy intake (i.e., beverage + meal) when sugar-sweetened soft drinks are consumed (see Table). Methodological differences may explain conflicting study outcomes in this area.
High fructose corn syrup (HFCS) is a primary ingredient in soft drinks, mainly because of its sweet taste, low cost and wide availability. Its potential role in the obesity epidemic has been suggested (Bray, Nielsen, & Popkin, 2004), but an expert panel assembled by The Center for Food, Nutrition, and Agricultural Policy conducted a review and concluded that HFCS consumption is not “uniquely” responsible for increasing rates of obesity (Forshee et al., 2007). Individual sweeteners (i.e., sucrose, HFCS) appear to have similar effects on intake-related hormones and energy intake (Melanson et al., 2007). Although controversy remains (Allison & Mattes, 2009; Brownell & Frieden, 2009), most research in this area suggests that energy-containing soft drink intake should be limited, regardless of sweetener type.
Fruit juices (100% juice) contain important nutrients, and are a contributor to total fruit intake (USDA). However, fruit juices contain little or no fiber, and because beverages are less satiating than solid foods (Flood-Obbagy & Rolls, 2009; R. D. Mattes, 1996), whole fruit is preferable for weight management (Popkin et al., 2006; USDA). Increased consumption of fruit juice (juice or juice drinks) compared with decreased consumption over four years is associated with weight gain, as is fruit drink intake (i.e., fruit punch, sweetened fruit-flavored drinks)(Schulze et al., 2004). Due to their minimal nutritional value and energy content, juice drink intake should be minimized (Popkin et al., 2006), and whole fruit should be encouraged rather than fruit juice due to its greater satiating ability.
Fluid milk is a major source of energy and calcium in the US diet (Cotton, Subar, Friday, & Cook, 2004). In recent years, a great deal of attention has focused upon the role of dietary calcium in weight loss. Interest in calcium arose from its role in the regulation of energy metabolism, specifically related to hormones influencing lipolysis and fatty acid synthesis (Zemel & Miller, 2004). Studies have associated a higher milk intake with less weight gain over time (Rosell, Hakansson, & Wolk, 2006), which suggests that milk consumption may be beneficial for weight management. Research has demonstrated that consumption of calcium-containing beverages promotes weight loss (Lukaszuk, Luebbers, & Gordon, 2007; Zemel & Miller, 2004), however conflicting evidence also exists (Barr et al., 2000).
Milk is rich in nutrients, however high milk consumption could increase dietary fat, cholesterol, and energy intake, and contribute to weight gain (Barr et al., 2000; Cotton et al., 2004; Hollis & Mattes, 2007). Satiety and fullness are greater after consumption of a skim milk beverage as compared to a sugar-sweetened beverage, however meal energy intake with the two beverages is not different (Harper et al., 2007; Soenen & Westerterp-Plantenga, 2007). Intervention studies (Barr et al., 2000) have demonstrated that increasing skim or 1% milk consumption by three servings per day without other intentional dietary changes leads to weight gain among adults aged 55-85 yrs, although less than would be expected based upon the additional energy provided in the milk.
Most research in this area has focused on dairy products; however, soy milk may be comparable to skim milk at facilitating weight loss (see Table). Individuals on a hypocaloric diet supplemented either with soy or skim milk lost equal amounts of weight and had similar nutrient intakes (Lukaszuk et al., 2007). Thus, consumption of either dairy or soy milk may be beneficial as part of a hypocaloric diet. Due to their energy and fat content, low-fat or non-fat options may be recommended, and appropriate reductions in other sources of energy should be encouraged to compensate for the energy content of the dairy or soy beverages.
Beer is among the top ten sources of energy in the US diet (Block, 2004). Although possibly controversial, the Beverage Guidance Panel (Popkin et al., 2006) categorized alcohol as an energy-containing beverage with some nutrients. Alcohol drinkers experience health benefits not seen in abstainers including smaller waist circumference (Tolstrup et al., 2008) and a lower prevalence of metabolic syndrome (Freiberg, Cabral, Heeren, Vasan, & Curtis Ellison, 2004; Rosell, De Faire, & Hellenius, 2003). Others have associated a high alcohol intake (>81 g/wk) with abdominal obesity; however, after accounting for alcoholic beverage type (beer, wine, liquor), only liquor consumption was associated with abdominal obesity (Riserus & Ingelsson, 2007). Similarly, a linear relationship between intake of spirits and BMI has been reported (see Table) (Lukasiewicz et al., 2005). Associations have also been found between a heavy alcohol intake (>30 g/day; a glass of wine represented ~10g alcohol), weight gain and obesity, but not light-moderate alcohol intake (Wannamethee & Shaper, 2003). This appears consistent with studies reporting a J-shaped relationship between total alcohol consumption and waist-to-hip ratio and BMI (Lukasiewicz et al., 2005). Conflicting results may be due to confounding factors related to alcohol intake and weight status, such as smoking, physical activity, educational level, and economic status, as well as to alcoholic beverage type (Rosell et al., 2003).
Clinical studies investigating alcohol consumption and meal energy intake have found that ad libitum alcoholic beverage consumption increased total meal energy intake as compared to soft drink consumption (Buemann, Toubro, & Astrup, 2002), and that daily energy intake was greater after consuming a 5% alcohol beer as compared to a nonalcoholic beer or cola (R. D. Mattes, 1996). Overconsumption of alcoholic beverages at a meal may be related to increased thirst due to their diuretic effect (Buemann et al., 2002), although others report that moderate alcohol consumption does not exert a diuretic effect (McDonald & Margen, 1979). A high plasma alcohol concentration may reduce dietary cognitive restraint and promote energy overconsumption (Buemann et al., 2002).
Moderate alcohol consumption may have health benefits (Goldberg, Mosca, Piano, & Fisher, 2001), including less weight gain with time (Wannamethee & Shaper, 2003) and a lower risk of abdominal obesity (Freiberg et al., 2004; Rosell et al., 2003; Tolstrup et al., 2008). Some evidence suggests more benefits of wine and beer consumption (Freiberg et al., 2004) as opposed to consumption of liquor/spirits (Lukasiewicz et al., 2005). However, as with other energy-containing beverages that appear to be additive to meal/daily energy intake, energy intake may acutely increase with alcohol ingestion. Thus, as with other energy-containing beverages that appear to be additive to meal/daily energy intake, portion control and moderation should be encouraged.
In the US, energy intake from beverages has increased dramatically in recent decades (Nielsen & Popkin, 2004). Americans are consuming more energy from soft drinks (Block, 2004) which may be replacing water and nutrient-containing beverages such as milk. Because energy-containing beverages do not promote satiety (R. D. Mattes, 1996), intake of energy-containing beverages may be an important dietary factor contributing to weight gain. In contrast, water consumption may reduce hunger and energy intake (Davy, Dennis, Dengo et al., 2008; Stookey et al., 2007; Stookey et al., 2008; Van Walleghen et al., 2007). These benefits, as well as its low cost and wide availability, suggests that increased water consumption ( >1 L/day; >33.8 fl oz) and a reduced intake of energy-containing beverages (Stookey et al., 2008) may facilitate weight management. However, this has not been directly addressed beyond the acute meal setting.
A limited amount of research suggests that other energy-free beverages (i.e., diet soft drinks, coffee/tea) may facilitate weight management, particularly if substituted for energy-containing beverages (Stookey et al., 2008). While there may be health benefits to consuming milk, 100% fruit juice, and in moderation, alcoholic beverages, additional energy provided by these beverages must be compensated for by reducing food intake. A summary of research investigating beverage consumption (by category), acute energy intake and body weight is provided in the Figure. Due to the lack of long-term intervention studies, further research is warranted to determine if specific patterns of beverage consumption improves health and weight status.
There are a number of limitations to existing research which should be acknowledged. First, a conflict of interest may exist between financial sponsors of beverage research and researchers. Industry-sponsored research may be less likely to report unfavorable results compared to studies with no industry support (Lesser, Ebbeling, Goozner, Wypij, & Ludwig, 2007). Differences in effect sizes between industry-funded and non-funded research studies have also been reported (Vartanian et al., 2007). Therefore, financial disclosures should be considered when evaluating this literature. A second limitation, as mentioned previously, is the lack of randomized controlled trials. Available literature largely consists of epidemiological studies, which provide evidence of associations of beverage intake and weight status but not causal relationships, and from short-term feeding studies. Other limitations include possible confounding lifestyle behaviors by categories of beverage consumers (Duffey & Popkin, 2006; Popkin et al., 2005), and aging, which may independently influence gastric emptying and energy intake regulation (Davy, Dennis, Dengo et al., 2008; Van Walleghen et al., 2007).
Weight management practitioners may consider the recently proposed beverage intake guidelines (Popkin et al., 2006), and existing fluid intake recommendations. To maintain hydration, an AI of 3.0 L/day for men and 2.2 L/day for women (~104 fl oz and 72 fl oz, respectively) of total beverages is recommended, but no specific recommendations for water intake and weight management exist. An increase in the total beverage volume consumed is associated with greater weight loss during a hypocaloric diet (Stookey et al., 2008); therefore, water (> 1 L/day; >33.8 fl oz) and other energy-free beverages should be consumed to satisfy thirst and promote optimal hydration status, without increasing total energy intake. Solid fruits should be emphasized to meet fruit intake recommendations, and low-fat/non-fat dairy or soy products may be recommended to provide micronutrients while minimizing saturated fat and energy intake. Alcoholic beverages should be consumed in moderation, for those who wish to consume alcohol. Other energy-containing beverages (i.e., soft drinks, fruit drinks and whole milk) are not recommended for frequent consumption due to their energy content and low satiating ability, as energy provided by beverages appears additive to total energy intake. As highlighted in a recent commentary (Allison & Mattes, 2009), additional research addressing the role of beverage consumption in adult weight management is clearly needed.
Disclosure: The corresponding author has received research funding for prior work from the Institute for Public Health and Water Research.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.