![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
,
* Robert Wood Johnson Clinical Scholars Program and Department of Pediatrics, Child Health Institute, Center for Public Health Nutrition and Departments of Medicine and Epidemiology, and ** Department of Health Services, University of Washington, Seattle, WA; and Children's Hospital and Regional Medical Center, Seattle, WA
3 To whom correspondence should be addressed. E-mail: jasonmen{at}u.washington.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT SUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
19 y old and to describe associations between DED and predictors of overweight. We used a subset of data from the 1994–1996, 1998 Continuing Survey of Food Intake for Individuals (CSFII) and a multivariate regression model to determine independent associations between DED and socioeconomic and demographic variables after controlling for covariates. In this cross-sectional data set, DED was positively associated with total energy intakes and varied with both age and gender. DED increased from birth, peaked at 7–8 y of age, and then declined. Boys consumed more energy-dense diets than girls. Among children 4 y old, higher DED was associated in the regression model with lower household incomes and with enrollment in the food stamp program. Among adolescents 12–19 y old, higher DED was associated with being African-American. In contrast, lower DED among children 11 y old was associated with being Asian or Hispanic and with total daily consumption of fluid milk. The quality of the diet for young children, as indexed by high DED, may be adversely affected by limited household economic resources. Although food insecurity and WIC enrollment were not associated with DED in this study sample, milk consumption in children 4 y old was associated with lower DED.
KEY WORDS: • nutrition • dietary energy density • pediatrics • obesity • overweight
The epidemic of childhood overweight is a major public health concern with multiple causes (1–5). The increased consumption of energy-dense, micronutrient-poor foods, i.e., processed foods usually high in starches, added sugars, and added fats, is thought to play an important role in this epidemic (3,6,7). Foods with high energy density are less healthy (3), associated with higher energy intakes (8–11), and less expensive (12). Their availability and intake may explain in part why the epidemic of overweight has disproportionately affected people of lower socioeconomic status (13). Recent recommendations by the U.S. Departments of Health and Human Services and Agriculture emphasized that most Americans should limit their intake of food or beverages with added sugars and decrease their intake of saturated and trans fats (14). Similarly, the WHO recently recommended that an important method for preventing overweight in children and adolescents was to restrict the intake of energy-dense foods (3).
Despite these recommendations, our knowledge of the role that dietary energy density plays in the U.S. overweight epidemic is poorly understood for several reasons. First, we lack population-based estimates of U.S. dietary energy density in children. Second, the association between the intake of energy-dense foods and risk factors for overweight has not been thoroughly explored.
The goals of this study were to describe dietary energy density (DED)4 in a nationally representative, cross-sectional survey of U.S. children and to test the hypothesis that increased DED would be predicted by overweight risk factors, such as race/ethnicity and poverty.
|
SUBJECTS AND METHODS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
This study was reviewed and deemed exempt by the University of Washington Human Subjects Division.
Subjects. For this analysis, we chose all subjects and households with children and adolescents <20 y old who had completed 2 d of dietary intake for a sample size of 11,284 subjects.
Outcome variable. The dependent variable, dietary energy density, was calculated by taking each subject's mean daily total energy in kilocalories (1 kcal = 4.184 kJ) and dividing by the mean daily total amount of food and drink in grams as previously described (17). Intake of water consumed as a beverage was excluded from this calculation because the assessment for water intake was not collected according to the same rigorous, well-validated method as the dietary intake data. Energy (kilocalories) from the consumption of human milk was not included in the calculation of dietary energy density or total fluid milk.
Covariates.
We chose the following socioeconomic and demographic variables as key covariates in constructing each of our multivariate linear regression models as discussed below: 1) gender; 2) age; 3) race/ethnicity categorized as Hispanic, Non-Hispanic white, Non-Hispanic black, Asian/Pacific Islander, or Other; 4) household income reported as a percentage of poverty level and further divided into the following poverty categories for this analysis: <100%, 100–199%, 200–299%, and 
300% of the poverty level; and 5) the highest household education level achieved among the male and female heads of households. These key covariates along with total fluid milk consumption comprise the "base" covariates included in all of the linear regression models. Total daily fluid milk consumption was included as an important covariate because it is a major, nutrient-rich beverage among children. Additionally, milk consumption is encouraged by supplemental nutrition programs such as the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) to improve the nutritional quality of children's diets. By adjusting for milk consumption, which the WIC encourages through both counseling and milk distribution, we are able to distinguish other benefits of the WIC program.
In addition to the base covariates, other predictors of interest include the following: 1) current household participation in the WIC or the Food Stamp Program (FSP) at the time of the survey; 2) region of the country as defined by the U.S. Department of Commerce for the 1990 census; 3) urbanization status based on metropolitan statistical areas (MSA) as defined by the Office of Management and Budget; and 4) household food insecurity assessed by self-report using a validated (18–20) question with 4 possible responses that identifies varying levels of food insecurity. For this analysis, this variable was further collapsed into 2 categories, i.e., no food insecurity and some food insecurity in which the latter category encompassed the 3 levels of food insecurity.
Statistical analyses. We used bivariate analyses to assess the unadjusted relation between DED and our predictors of interest. We used a multivariate linear regression model to adjust for potential confounders and covariates in which DED was the dependent variable, and gender, age, race/ethnicity, poverty category, highest education level achieved among the head of household, and total fluid milk intake were the independent variables. The mean daily total weight of food and drink in grams was included as a covariate in the regression model as recommended by Willett when using a ratio of nutritional variables, e.g., dietary energy density, in regression models (21). The age-squared term (age2) was included to adjust for a nonlinear relation between age and energy density. In separate analyses, we added WIC status, FSP status, food insecurity, region of the country, and urbanization status individually to the regression model.
Planned subanalyses were conducted using the above multivariate regression models according to the following subgroups: children 0–4, 5–11, and 12–19 y old for stratification similar to that used in previous population-based studies on childhood obesity (1,5). We chose 0–4 y of age as a subgroup because generally only children 0–4 y old are eligible to participate in WIC.
Survey estimation commands for complex survey data were used in the analyses taking into account weighted observations and the probability of selection, nonresponse, and poststratification adjustments, to obtain representative estimates of U.S. children < 20 y old. A significance level of 0.05 was used for all analyses. We present means ± SE where indicated. Taylor series linearization was used to estimate SE. Stata version 8.2 for Windows was used for all analyses.
|
RESULTS |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
|
In the unadjusted analysis, DED was positively correlated with total energy intake (r = 0.26 for 0- to 19-y olds; r = 0.42 for 0- to 4-y olds; r = 0.22 for 5- to 11-y olds; and r = 0.14, for 12- to 19-y olds, P < 0.001). DED varied by the sociodemographic variables. DED did not significantly differ among the categories of gender, poverty, food stamp, food insecurity, and urbanization (Table 2). DED differed significantly among race/ethnicity categories such that non-Hispanic blacks and whites had the highest DED (Table 2). Of the federally funded nutrition programs, children in households enrolled in WIC had a significantly lower DED than those that were not (Table 2). The Midwest region had the highest DED (Table 2). Children whose head of household had a 12th grade education had the highest DED, and children whose head of household had less than a 12th grade education had the lowest DED (Table 2).
|
|
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
In the multivariate linear regression model, age was the strongest predictor for increasing DED among the 0- to 4-y-old subjects. Given that energy requirements increase with age for 0- to 4-y-old children, this finding is not surprising because DED is likely a key mediator toward increasing energy intake in addition to increasing the quantity of foods consumed.
For 0- to 11-y-old children, Asian race was the strongest predictor for lower DED, and Hispanic ethnicity was also significantly associated with lower DED. Black race was associated with higher DED in adolescents, a finding that mirrors the disproportionate percentage of black adolescents who are at risk for overweight or were overweight in the 1999–2002 National Health and Nutrition Examination Survey (NHANES) (1,5).
Federally funded nutrition programs such as the WIC or the FSP provide nutrition resources to lower-income families and individuals. These programs may counter economic forces that encourage consumption of inexpensive, energy-dense foods. Household participation in the FSP was associated with a higher DED for 0- to 11-y-old children, independent of poverty level or the education level of the head of household. This association identifies an important opportunity for interventions to decrease DED and improve nutrition quality among the youngest children from socioeconomically disadvantaged households. No association between DED and WIC participation was noted, although children 0–4 y of age trended (P = 0.053) toward lower DED with WIC participation.
Lower household incomes and higher DED were significantly associated among children 0–4 y old, consistent with previous research that scarce economic resources promote the consumption of energy-dense foods (12,22). Moreover, household food stamp enrollment was independently associated with higher DED. These associations are not present in older children likely because they are more developmentally independent in their dietary intake and have more opportunities to exert this independence at school and other venues outside the home (23). The youngest children are more directly influenced by their parents (23); thus, their diets better reflect their parents' economic status. Interventions to improve diets should not only provide education but also address other key components such as economic constraints and cultural influences.
Food insecurity, defined as limited or uncertain availability of nutritionally acceptable or safe foods, is another factor related to overweight and certain low SES groups. Although food insecurity was associated with lower income respondents and with overweight among women in NHANES III (24), no association existed between food insecurity and increased DED in children. One explanation may be that in the setting of food insecurity, adult nutrition declines first because adults attempt to spare their children from poor nutrition first (25).
This study has several important limitations. First, it is a cross-sectional study; therefore, we are limited to demonstrating associations and not the directionality of the associations. Second, DED was calculated without taking into account water consumption; thus, children who consume large amounts of water will have higher estimates of DED. Third, a large number of children were excluded from the analysis because they did not have the requisite 2 d of dietary intake data. Excluded children were more likely to be older, nonwhite, and live in low SES households. Because children from low SES and disadvantaged minorities are at greater risk for obesity, our results may have been attenuated toward the null hypothesis. Moreover, these differences in sociodemographic characteristics limit generalizability. Fourth, the race/ethnicity categories are very broad. Each of the race/ethnicity categories represents a complex combination of ethnicities and cultures, which limits the generalizability of the associations to specific subgroups. Nevertheless, these traditional race/ethnicity categories may provide a rough screening measure for increased DED and poor nutrition. Finally, we were unable to examine the relation between DED and BMI as measured in this dataset. Although the CSFII provides reported height and weight, BMI based on reported height and weight is less valid than BMI from physical measurements. In a recent study in the United Kingdom, only 25% of parents recognized their children as overweight (26). Moreover, for adults, reported height tends to be overestimated and weight underestimated (27,28). Future studies in children should examine the relation between DED and BMI using measured height and weight.
Despite these limitations, this study provides the first nationally representative reference values by age for dietary energy density in U.S. children. The estimates may serve as a useful baseline for future clinical and public health interventions. Moreover, we report significant positive associations between DED, total energy intake, and key sociodemographic variables that previously were associated with overweight including black race among adolescents, and poverty among 0–4 y olds. It should be noted that only some risk factors for higher DED are shared with obesity, suggesting that DED is only one of many potential mechanisms for causing obesity. Further research to better define the associations between DED and BMI are warranted as are culturally sensitive means to successfully encourage the consumption of lower DED.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
2 Supported by the Robert Wood Johnson Clinical Scholars Program, University of Washington. The views expressed in this article are those of the authors and do not necessarily represent the views of the Robert Wood Johnson Foundation or the University of Washington.
4 Abbreviations used: CSFII, Continuing Survey of Food Intake for Individuals; DED, dietary energy density; FSP, Food Stamp Program; NHANES, National Health and Nutrition Examination Survey; WIC, Special Supplemental Nutrition Program for Women, Infants, and Children.
Manuscript received 29 September 2005. Initial review completed 9 November 2005. Revision accepted 22 January 2006.
|
LITERATURE CITED |
|---|
|
TOPABSTRACTSUBJECTS AND METHODSRESULTSDISCUSSIONLITERATURE CITED |
|---|
1. Ogden CL, Flegal KM, Carroll MD, Johnson CL. Prevalence and trends in overweight among US children and adolescents, 1999–2000. JAMA. 2002;288:1728–32.
2. Healthy People 2010. 2nd ed. Washington, DC: U.S. Government Printing Office, November 2000.
3. Diet, nutrition and the prevention of chronic diseases. World Health Organ Tech Rep Ser. 2003;916:i–viii, 1–149, backcover.[Medline]
4. Barlow SE, Dietz WH. Obesity evaluation and treatment: Expert Committee recommendations. The Maternal and Child Health Bureau, Health Resources and Services Administration and the Department of Health and Human Services. Pediatrics. 1998;102:E29.
5. Hedley AA, Ogden CL, Johnson CL, Carroll MD, Curtin LR, Flegal KM. Prevalence of overweight and obesity among US children, adolescents, and adults, 1999–2002. JAMA. 2004;291:2847–50.
6. Bowman SA, Gortmaker SL, Ebbeling CB, Pereira MA, Ludwig DS. Effects of fast-food consumption on energy intake and diet quality among children in a national household survey. Pediatrics. 2004;113:112–8.
7. Prentice AM, Jebb SA. Fast foods, energy density and obesity: a possible mechanistic link. Obes Rev. 2003;4:187–94.[Medline]
8. Rolls BJ, Bell EA. Intake of fat and carbohydrate: role of energy density. Eur J Clin Nutr. 1999;53: Suppl 1:S166–73.
9. Cuco G, Arija V, Marti-Henneberg C, Fernandez-Ballart J. Food and nutritional profile of high energy density consumers in an adult Mediterranean population. Eur J Clin Nutr. 2001;55:192–9.[Medline]
10. Poppitt SD, Prentice AM. Energy density and its role in the control of food intake: evidence from metabolic and community studies. Appetite. 1996;26:153–74.[Medline]
11. Stookey JD. Energy density, energy intake and weight status in a large free-living sample of Chinese adults: exploring the underlying roles of fat, protein, carbohydrate, fiber and water intakes. Eur J Clin Nutr. 2001;55:349–59.[Medline]
12. Darmon N, Briend A, Drewnowski A. Energy-dense diets are associated with lower diet costs: a community study of French adults. Public Health Nutr. 2004;7:21–7.[Medline]
13. Drewnowski A, Specter SE. Poverty and obesity: the role of energy density and energy costs. Am J Clin Nutr. 2004;79:6–16.
14. U.S. Department of Health and Human Services and U.S. Department of Agriculture. Dietary guidelines for Americans, 2005. 6th ed. Washington, DC: U.S. Government Printing Office, January 2005.
15. Rizek RL, Pao EM. Dietary intake methodology I. USDA surveys and supporting research. J Nutr. 1990;120: Suppl 11:1525–9.
16. What we eat in America 1994–1996: interviewer manual for the Continuing Survey of Food Intake by Individuals 1994–1996 and the Diet and Health Knowledge Survey 1994–1996. Rockville, MD: Westat, 1993.
17. Ledikwe JH, Blanck HM, Khan LK, Serdula MK, Seymour JD, Tohill BC, Rolls BJ. Dietary energy density determined by eight calculation methods in a nationally representative United States population. J Nutr. 2005;135:273–8.
18. Cristofar SP, Basiotis PP. Dietary intakes and selected characteristics of women ages 19–50 years and their children ages 1–5 years by perception of food insufficiency. J Nutr Educ. 1992;24:53–58.
19. Sidel VW. The public health impact of hunger. Am J Public Health. 1997;87:1921–2.
20. Rose D, Oliveira V. Validation of self-reported measure of household food insufficiency with nutrient intake data. USDA, Economic Research Service, 1997.
21. Willet W. Nutritional epidemiology. 2nd ed. New York, NY: Oxford University Press, 1998.
22. Darmon N, Ferguson E, Briend A. Do economic constraints encourage the selection of energy dense diets? Appetite. 2003;41:315–22.[Medline]
23. Kleinman R., ed. Pediatric nutrition handbook. 5th ed: American Academy of Pediatrics, 2004.
24. Townsend MS, Peerson J, Love B, Achterberg C, Murphy SP. Food insecurity is positively related to overweight in women. J Nutr. 2001;131:1738–45.
25. Messer E, Ross EM. Talking to patients about food insecurity. Nutr Clin Care. 2002;5:168–81.[Medline]
26. Jeffery AN, Voss LD, Metcalf BS, Alba S, Wilkin TJ. Parents' awareness of overweight in themselves and their children: cross sectional study within a cohort (EarlyBird 21). BMJ. 2005;330:23–4.
27. Palta M, Prineas RJ, Berman R, Hannan P. Comparison of self-reported and measured height and weight. Am J Epidemiol. 1982;115:223–30.
28. Rowland ML. Self-reported weight and height. Am J Clin Nutr. 1990;52:1125–33.
This article has been cited by other articles:
|
|
 |
|
  S. I O'Donnell, S. L Hoerr, J. A Mendoza, and E. Tsuei Goh Nutrient quality of fast food kids meals Am. J. Clinical Nutrition, November 1, 2008; 88(5): 1388 - 1395. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. A. Nicklas, C. E. O'Neil, J. Mendoza, Y. Liu, I. F. Zakeri, and G. S. Berenson Are Energy Dense Diets Also Nutrient Dense? J. Am. Coll. Nutr., October 1, 2008; 27(5): 553 - 560. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. A McCaffrey, K. L Rennie, M. A Kerr, J. M Wallace, M. P Hannon-Fletcher, W A. Coward, S. A Jebb, and M B. E Livingstone Energy density of the diet and change in body fatness from childhood to adolescence; is there a relation? Am. J. Clinical Nutrition, May 1, 2008; 87(5): 1230 - 1237. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. E Schebendach, L. E. Mayer, M. J Devlin, E. Attia, I. R Contento, R. L Wolf, and B T. Walsh Dietary energy density and diet variety as predictors of outcome in anorexia nervosa Am. J. Clinical Nutrition, April 1, 2008; 87(4): 810 - 816. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. O Fisher, Y. Liu, L. L Birch, and B. J Rolls Effects of portion size and energy density on young children's intake at a meal Am. J. Clinical Nutrition, July 1, 2007; 86(1): 174 - 179. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. K. Newby Are Dietary Intakes and Eating Behaviors Related to Childhood Obesity? A Comprehensive Review of the Evidence J. Law Med. Ethics, March 1, 2007; 35(1): 35 - 60. [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
