It is estimated that over one-third of all U.S. adults are obese. However, substantial disparities exist among racial/ethnic minorities [1
]. For example, approximately 56.9% of adult African American women are obese compared to 45.7% of Hispanic and 35.5% of Caucasian women [2
]. Obesity and weight gain and their accompanying metabolic abnormalities have been linked to several chronic diseases including cardiovascular disease, type 2 diabetes and cancer at several sites [3
]. Energy-containing beverages could promote weight gain because they are substantial sources of energy in the U.S. diet [4
]. In nationally representative data, Storey and colleagues [5
] and more recently, Han and Powell [6
] reported that beverage consumption by adults varied by race/ethnicity, socioeconomic status, age and sex. Overall, for women, beverage energy intake peaked at ages 20–39 years contributing approximately 19% of total energy intake, or 380 kcal/day. Han and Powell [6
] observed that mean beverage energy intake ranged from 362 kcal/day among Mexican American women and was highest among African American women at 405 kcal/day. Racial minorities were more likely to report consuming more sugar sweetened beverages overall, and among children, adolescents and adults [6
]. Added sugar is one factor which could contribute to obesity and related chronic diseases, thus, sugar-sweetened beverages, the largest source of added sugar in the diet of the US population, have been a major focus of obesity prevention efforts [7
The objective of the present study was to describe the contribution of total beverages and classes of beverages, particularly sugar sweetened beverages, to overall energy intake and to evaluate the associations between beverage consumption and body mass index (BMI) in a population of overweight and obese African American women.
presents total and grouped beverage consumption by demographic and lifestyle characteristics of the population. Overall, in this population of overweight and obese African American women mean energy intake from beverages was approximately 273 ± 192 kcal/day or 18.3% of total energy intake. The most commonly reported beverage was plain water, consumed by 88.2% of participants, followed closely by sweetened beverages (soft drinks, fruit drinks, sweetened teas, sweetened coffees and sweetened/flavored waters) consumed by 78.9% of participants. Unsweetened beverages (tea, coffee, not including water) and artificially sweetened beverages (diet soft drinks, diet fruit drinks, diet tea and diet coffee beverages) were reported by 46.8% and 29.3% of participants, respectively. Milk consumption (without additions) was reported by 68.2% of participants, overall. Study participants who were younger, lived with children, and had more education tended to report greater energy intakes from beverages either for absolute intake (kcal/day) or relative to total energy intake (i.e., percentage of energy intake). Morbidly obese participants reported more absolute energy contributed by beverages (~300 kcal/day) than either obese or overweight (249 and 278 kcal/day, respectively) participants; however, percentage of energy intake from beverages was similar for overweight (19.9%) and morbidly obese (19.3%) participants. Obese participants reported 17.1% of energy from beverages. There were no significant differences in the percentage of participants reporting any (v. no) consumption of sweetened beverages by strata of demographic characteristics. Table 2
presents mean beverage consumption for categories of beverages among consumers with results stratified by BMI (overweight, obese, morbidly obese). Among participants who reported consuming sugar sweetened beverages, those who were morbidly obese reported the highest daily energy intakes from beverages (187 kcal) compared to overweight (161 kcal) and obese participants (150 kcal). Morbidly obese participants also reported a lower percentage of their total beverage intake as water (48.8% total volume (mL)) compared to either overweight (53.4%) or obese (51.7%) participants. In multiple regression analyses total energy from beverages and percent energy from beverages were not significantly associated with BMI modeled as a continuous outcome (Table 3
). Similarly, no clear patterns emerged for energy intake from beverages in multinomial regression models where BMI was considered as a categorical outcome variable [10
]. Finally, in multivariate regression models, none of the a priori selected grouped beverage variables were associated with BMI as a continuous outcome (Table 3
) or as a categorical outcome [10
In this population of low-income overweight and obese African American women mean beverage energy intake was 272 (±195) kcal/day or approximately 18% of total daily energy intake. The most commonly consumed beverages were water (88.2%), sugar sweetened beverages (78.9%) and milk without additions (68.2%).
Daily consumption of sugar sweetened beverages was common across all demographic characteristics considered in our analyses. Approximately 79% of our participants reported consuming sugar sweetened beverages, similar to the 82% observed by Bleich [11
] among US Non-Hispanic Black adults. Although consumption was more common among overweight than obese participants in our study, among consumers, mean energy intake (kcal/day) from sugar sweetened beverages was highest among the morbidly obese (187 ± 135 kcal/day). This value exceeds the American Heart Association’s recommendation for no more than 100–150 kcal/day from all added sugar [12
]. For those who were morbidly obese and reported consuming sugar sweetened beverages, 38% of their total fluid intake was from sugar sweetened beverages.
In multivariate regression analyses we did not observe any associations between intakes of energy-contributing beverages (total), sugar sweetened beverages, or other categories of beverages with current weight. However, based on our data, we cannot reject the possibility that energy-containing beverages could be a contributing factor to future weight gain. In a randomized controlled crossover design including 44 women, DellaValle et al. [13
] observed that energy-containing beverages consumed with a meal added to energy intake without significantly affecting satiety. Similarly, a recent systematic review concluded that energy consumed in liquid form is more difficult to offset in subsequent meals and could promote positive energy balance [7
]. Energy intakes from sugar sweetened beverages of the magnitude reported in this study (e.g., 1000–1300 kcal/week), which appear relatively modest, could lead to weight gain on the order of one or more pounds per month in the context of overall energy intakes above the requirements for weight maintenance. In addition, a growing body of research suggests that regular consumption of sugar sweetened beverages may be associated with incidence of cardiometabolic disease, independent of obesity [14
Previous research of the relationship between sugar sweetened beverage consumption and weight among adults has been summarized in a number of recent comprehensive reviews. A systematic review of sugar-sweetened soft beverages and obesity by Gibson [15
] concluded that the effect of sugar sweetened beverages on obesity among adults was likely small except in susceptible individuals or at very high intakes. Gibson [15
] described results from six previous reviews of sugar sweetened beverages and obesity; two citing the evidence as strong, one probable, and three inconclusive or negligible. Notably, in prospective studies the strongest associations between sugar sweetened beverage consumption and increases in weight have been observed among participants who increased their consumption over time compared to those who maintained either a high or a low consumption. Another recent review focusing on the evidence of common beliefs in obesity research stated that although increased consumption of energy-contributing beverages tends to be coupled with increased total energy intake, associations with BMI have rarely been observed [16
]. In contrast, Hu [17
] concluded that when all the evidence is considered, the relationship between consumption of sugar sweetened beverages and weight is modest but compelling and important from a public health perspective.
Beverage intakes are of interest beyond the contribution to energy intake. Milk is a source of key nutrients, including calcium and vitamin D, and in some emerging research has been linked with cardiovascular health benefits [18
]. In contrast to our results where more than 2/3 of participant reported consuming milk, in nationally representative data for Non-Hispanic Black adults for 1999–2004, Bleich and colleagues [11
] reported that only 31% of those ages 20–44 consumed milk on any given day (data were not stratified by gender). Vitamin C, a water-soluble antioxidant, is an essential nutrient with well-established roles in a number of biological processes including immune function, collagen synthesis, and protein metabolism. There is at least some evidence that citrus juices may have potential benefits for chronic disease prevention [17
]. 100% fruit and vegetable juices, important contributors to vitamin C and phytochemical intake, were consumed by 40% of respondents in our study. In the study by Bleich et al. [11
] 24% of Non-Hispanic Black adults 20–44 years old reported consuming 100% fruit and vegetable juices.
This research had several strengths. We report results for consumption of total beverage and categories of beverages among a unique population of low-income overweight and obese African American women in South Georgia. Demographic data and two days of beverage intake data were collected and processed by trained staff using standardized methods. Despite these strengths, there are several limitations with our study. This population was recruited as part of an intervention study focusing on weight management; therefore, all of the participants were overweight or obese, and the results may not be generalizable to similar adults of healthy weight. This analysis is cross-sectional, and includes self-reported data for both weight and dietary intake data, and thus could be subject to confounding or bias. Although these data were collected at baseline and prior to any intervention, if obese and morbidly obese persons, had already decreased their consumption of energy-containing beverages or replaced consumption of caloric beverages with non-caloric alternatives to facilitate weight loss, then we would underestimate the relationship between beverage energy intake and weight. We are unable to completely explore this possibility with our data. Compared to overweight women (20.6%) more morbidly obese women (30.1%) reported any consumption of artificially sweetened beverage; yet, the percent of total beverage volume contributed by artificially sweetened beverages among the morbidly obese women (27.3%) was not strikingly different than overweight women (25.2%). Under-reporting of energy intake by overweight and obese persons has been consistently reported [20
]. If under-reporting of energy-containing beverages was more substantial among women with higher BMIs, as a result, we could underestimate the relationship between energy-containing beverages and BMI. We cannot completely rule out this possibility; however, mean sugar sweetened beverage volume was greatest among morbidly obese participants in this population.