AJCN Tufts Nutrition Symposium, Boston & Online Sept 2009
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by McCaffrey, T. A
Right arrow Articles by E Livingstone, M B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by McCaffrey, T. A
Right arrow Articles by E Livingstone, M B.
Agricola
Right arrow Articles by McCaffrey, T. A
Right arrow Articles by E Livingstone, M B.
American Journal of Clinical Nutrition, Vol. 87, No. 5, 1230-1237, May 2008
© 2008 American Society for Nutrition

ORIGINAL RESEARCH COMMUNICATION

Energy density of the diet and change in body fatness from childhood to adolescence; is there a relation? 1,2,3,4

Tracy A McCaffrey, Kirsten L Rennie, Maeve A Kerr, Julie M Wallace, Mary P Hannon-Fletcher, W Andy Coward, Susan A Jebb and M Barbara E Livingstone

1 From the Northern Ireland Centre For Food and Health (NICHE), University of Ulster, Coleraine, United Kingdom (TAMcC, KLR, MAK, JMW, MPH-F, and MBEL), and the Medical Research Council, Human Nutrition Research, Elsie Widdowson Laboratory, Cambridge, United Kingdom (WAC and SAB)

2 We dedicate this article to the memory of Dr Andy Coward, 17 October 1941–3 November 2007.

3 This paper forms part of the dissemination of the Foods Standards Agency, UK-commissioned research project at follow-up. The baseline project was supported by an educational grant from the Sugar Bureau, Kellogg's, Coca Cola, and Masterfoods UK. TAMcC is supported by a PhD award from the Department of Employment and Learning Award, United Kingdom.

4 Reprints not available. Address correspondence to MBE Livingstone, Northern Ireland Centre for Food and Health (NICHE), University of Ulster, Cromore Road, Coleraine BT52 1SA, United Kingdom. E-mail: mbe.livingstone{at}ulster.ac.uk.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: The contribution of energy density (ED) of the total diet to increased risk of obesity from childhood into adolescence is unclear.

Objective: We assessed the relation between the ED of the diet in childhood, calculated in a number of ways, and change in adiposity from childhood to adolescence.

Design: In a prospective study, 48 children (30 boys, 18 girls) were initially studied at age 6–8 y (baseline) and followed up at age 13–17 y. Daily ED, energy intake, and food intake were assessed at baseline by 7-d weighed food records concurrent with estimates of total energy expenditure (TEE) by doubly labeled water. ED was calculated with the use of 5 published methods. Obesity risk was defined with the use of body fat from total body water by isotope dilution. Body fat was normalized for height and expressed as fat mass index (FMI). Change in adiposity was calculated as follow-up FMI minus baseline FMI.

Results: Misreporting of energy intake at the group level at baseline was low relative to the TEE. ED of the total diet at baseline by the 3 methods for calculating ED that excluded all or most beverages was prospectively associated with change in FMI. However, ED of the total diet by any of the methods was not associated with change in the percentage body fat, body mass index, or waist circumference z scores.

Conclusion: The methods used to calculate ED and to assess obesity risk lead to different conclusions about the relation between the ED of the diet in childhood and gain in fat into adolescence.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The causes of obesity are multifactorial, and, in relation to dietary intake, no one factor has been identified as playing a major role in contributing to excess energy intake (EI). In recent years, the role of the energy density (ED; in kJ/g) of the diet in promoting overweight and obesity has received increasing attention. It was postulated, based on the results of experimental studies in adults (1, 2), that reducing the ED of the diet could be used to attenuate weight gain (3, 4). However, in children and adolescents the relation between the ED of the diet and risk of obesity is not so clear with some cross-sectional (5, 6) and prospective (7) studies showing a positive relation between ED and obesity risk factors, whereas other cross-sectional (8) and prospective (7, 9, 10) studies showed no relation. There are a number of possible reasons for the lack of consistency. First, there is no consensus about the most appropriate way of calculating ED (11, 12). One dilemma when calculating ED is whether to include beverages, because they can disproportionately influence the calculation of ED. Second, the way in which ED is calculated also depends on the method used to assess dietary intake because intake of some beverages, particularly water, may not have been recorded. Third, the problems associated with misreporting of dietary intake need to be acknowledged in children, adolescents, and adults (13). Finally, the method used to calculate adiposity in children and adolescents is particularly important. In most studies, body mass index (BMI; in kg/m2) was used as a proxy measure for adiposity. However, for a given BMI contemporary children now have a greater fat mass than did children 20 y ago (14), and use of BMI and percentage of body fat as measures of adiposity may be inappropriate and misleading (15).

Concurrent with the rise in overweight and obesity, there is a popular perception that the frequency of snacking in children and adolescents has also increased, but with no apparent change in the average portion size of snacks in the United States during the past 20 y (16). Snacking has often been associated with more energy-dense foods (16, 17); thus, it is plausible that changes in eating patterns from 3 meals/d to a more energy dense snacking or "grazing" style may be causally related to obesity. However, understandably the contribution of snacking to excess adiposity remains problematic because there is currently no consensus on the definition of meals or snacks. Thus, the relation of snacking or snack foods with obesity status remains equivocal (18-22). Taken together, it is not surprising that at present the literature presents an inconsistent picture about the relation of ED of the diet and obesity in children and adolescents.

The primary aim of this study was to calculate ED with the use of 5 different published methods and to assess the relation between the ED of the diet in childhood and change in fat mass from childhood to adolescence. The secondary aim was to evaluate whether the relation between ED and change in fat mass were independently associated with the ED of snacks compared with meals.


   SUBJECTS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Subjects
At baseline, 115 healthy children were recruited through primary schools in the Coleraine area of Northern Ireland between 1996 and 1998 (23). Parents of eligible children (aged 6-8 y and living with biological parents) were first contacted by letter through local primary schools, after which those who expressed interest in participating in the study were interviewed in their homes to explain the study in detail. At follow-up, when the children were aged 13-17 y, 50 subjects (44%) of the original cohort agreed to participate. As with the baseline study (23), no direct reference to obesity was made by the researchers. Parents and adolescents were informed that the study was concerned with the assessment of healthy growth and development through an examination of changes in food intake and energy expenditure across childhood and adolescence. Following the protocol for the baseline study, all follow-up measurements took place during the school term and were conducted during a 1-y period.

At baseline and follow-up, the parents of each subject gave written informed consent to their child's participation in the study. Ethical approval for the study was obtained from the Research Ethics Committee of the University of Ulster.

Total energy expenditure
At both baseline and follow-up, total energy expenditure (TEE) was measured over 10 d by the doubly labeled water (DLW) method. After collection of a predose urine sample, every child was given an oral dose of 2H2O and H218O. At baseline, this was 0.05g 2H2O/kg body weight and 0.125g H218O/kg body weight, and at follow-up the dose was 0.07g 2H2O/kg body weight and 0.174g H218O/kg body weight. Further samples of urine were collected at a known time, daily for 10 d. A detailed description of the protocol and calculation of TEE was previously described (23).

Body composition
In addition to calculation of TEE, the intercepts of the isotope disappearance curves were used to provide estimates of total body water at baseline and follow-up. Body water was calculated as the mean of the time zero H218O distribution/1.01 and of the time zero 2H2O distribution space/1.04. Fat-free mass (FFM) was calculated from total body water by dividing the water content of fat-free tissue with age- and sex-specific values (24, 25), which was shown as having high accuracy for measuring FFM relative to the 4-compartment model in children (26). Hydration of FFM was assumed to be 76.6% for boys and 74.9% for girls at baseline and 73% at follow-up (24, 25). Fat mass (FM) was calculated as the difference between body weight and FFM (27). To adjust for body size at baseline and follow-up, adjustment for the subject's height was made. From log-log regression analyses, height squared (in m2) removed the effect of height on the measure of FM at both baseline and follow-up (27). Therefore, FM was expressed as fat mass index [FMI; fat (in kg)/height squared (in m2)] (15, 28). The difference between FMI at baseline and follow-up was calculated as follows: follow-up FMI –baseline FMI = FMIdif. Similar to the calculation of FMI, FFM index (in kg/m2) was also calculated. Percentage of body fat was calculated as body fat (in kg)/body weight (in kg) x 100.

Anthropometry and pubertal status
At baseline and follow-up, body weight in a swimming costume was measured to the nearest 0.1 kg, and height was measured to the nearest 0.1 cm. BMI was calculated and converted to age- and sex-specific z scores (29). Weight status was defined with the use of the International Obesity Task Force BMI cutoffs (30). As an indicator of abdominal fat distribution at baseline and follow-up, waist circumference (WC) at the umbilicus was measured to the nearest 0.1 cm and then converted to age- and sex-specific z scores (31). Similar to the method of calculating FMIdif, change in BMI z score and WC z score between baseline and follow-up was also calculated as follow-up minus baseline. At follow-up, pubertal status was also self-assessed (n = 48) based on Tanner line drawings (32).

Dietary intake
At baseline, dietary intake by 7-d weighed dietary record was assessed concurrently with DLW measurements of TEE, as previously described by McGloin et al (23). Briefly, parents were issued with digital weighing scales and instructed how to weigh and record all food and drinks consumed, as well as leftovers, for 7 consecutive days. Parents were asked to record the time of consumption and their perception of whether the eating occasion was a meal or a snack. Researchers gave detailed explanations and showed the cumulative weighing technique and then asked the parents to repeat the procedure in their presence. In addition, written instructions and an example of a complete diary were provided for reference inside the food diary. Families were visited at home on 2 further occasions to ensure the protocol was adhered to and to monitor compliance. Researchers checked the diaries for any obvious omissions or lack of detail by probing the parent. Food eaten outside the home was identified by brand name and packet size or by the empty wrappers. At each home visit, the researcher checked these and added missing detail as appropriate.

An earlier version of the dietary analysis package (v 1.25) did not provide information on the time of eating occasion, type of eating occasion (meal or snack as defined by parents), or food groupings, and subsequently the food diaries were reanalyzed with the use of WISP (version 3.01; Tinuviel Software, United Kingdom). Diaries were excluded from analysis if insufficient information was recorded (eg, if no foods were recorded after midday; n = 2), resulting in complete food diaries for 48 children aged 6–8 y for this analysis. The accuracy of EI reporting relative to TEE was calculated (EI:EE) as described by Black and Cole (33).

Energy density calculations
On the basis of previous analyses by Cox and Mela (11) and Ledikwe et al (12), 5 different methods were used to calculate ED (Table 1Go). The methods included the ED of all food and beverages consumed (EDall), the ED of solid foods only (EDsolid), and ED of food and energy-containing beverages (>21 kJ/100 g; EDenergy). It was possible from the dietary recording method to distinguish between milk "eaten as food" (eg, eaten with breakfast cereals) and milk consumed "as drinks" (consumed separately from food, eg, as a beverage). Therefore, milk as food was included in the calculation of EDfood and EDsoup, because excluding it inappropriately adjusts or calculates ED for foods such as breakfast cereals, which are typically eaten with milk (34).


View this table:
[in this window]
[in a new window]

  		TABLE 1. Methods used to calculate energy density (ED)1

 
Statistical analysis
Results are expressed as means ± SDs, or when data could not be normalized as medians and interquartile ranges. Data were assessed for normality with the use of the Shapiro Wilks test. Differences between sexes were assessed with the use of t tests and the nonparametric Mann-Whitney U test, as appropriate. The variance between ED methods was calculated as the intraindividual (within-person) CVs (CVW = SDW/mean) and the interindividual (between-person) CVs (CVB = SDB/mean). Correlation between the ED methods was assessed with the use of Spearman's correlation coefficient (r). When the distribution of FMIdif was examined, it was found that girls had gained more FM between baseline and follow-up. To adjust for the skewed data, the data were categorized based on sex-specific tertiles and because participants in the first and second tertiles of FMIdif had significantly lower change in FMI than did participants in the third tertile of FMIdif, the data were subsequently compared as the lowest tertiles (first and second) compared with the third tertile. Logistic regression analysis was used to examine associations between the change in FMI and ED of the diet at baseline, with first and second tertiles compared with the third tertile of FMIdif as the dependent variable and the ED methods as separate independent variables, with adjustment for sex, pubertal status, and misreporting (EI:EE) in the most parsimonious model. Other variables that did not contribute to the model included age at baseline, time between measurements (in mo), and tertile of baseline FMI. The most parsimonious models examined were model 1 that included sex + pubertal status + ED method and model 2 that included sex + pubertal status + EI:EE + ED method.

As with FMIdif, girls had a significantly higher change in percentage of body fat and WC z score than did boys (P < 0.01); thus, similar to change in FMIdif, change in percentage of body fat and WC z score were analyzed as sex-specific tertiles (first and second tertiles compared with third tertile). For change in BMI z score, no significant differences were observed between boys and girls (P = 0.856); therefore, the data were analyzed with the use of multiple regression. Values of P < 0.05 were regarded as statistically significant, and all statistical analyses were performed with the use of SPSS (version 11.5; SPSS, Chicago, IL).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Anthropometry
No significant differences were observed between the participants who declined to take part in the follow-up (n = 67) and those who participated in both baseline and follow-up, in terms of age, weight, height, BMI, WC, TEE, EI:EE, parental BMI, and energy and macronutrient intakes (data not shown). At baseline, boys had a significantly higher TEE (P = 0.004), FFM (P = 0.020), and FFM index (P = 0.007) than did girls (Table 2Go). Girls had higher FM (P = 0.035), FMI (P = 0.010), and percentage of body fat (P = 0.001) than did boys. When the International Obesity Task Force BMI cutoffs were applied, 13% of boys (3% obese) and 28% of girls (6% obese) were classified as overweight and obese, although the proportion of boys and girls overweight or obese was not significantly different (P = 0.227).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Physical characteristics of the participants at baseline (aged 6–8 y)1

 
Dietary intake
At baseline mean reported EI was 96% of measured TEE, indicating low levels of misreporting at the group level (Table 3Go). At baseline, girls had lower intakes of energy, carbohydrate, fat, and protein (expressed as g/d) than did boys (P < 0.05), but no differences were observed between boys and girls in terms of the percentage of energy from carbohydrate, fat, and protein; weight of food consumed (in g/d); and EI:EE (Table 3Go). Irrespective of the method used to calculate ED (Table 1Go), no significant differences were observed between girls and boys in the overall ED of their diet (Table 3Go).


View this table:
[in this window]
[in a new window]

  		TABLE 3. Dietary intake data from 7-d weighed food records at baseline in participants aged 6–8 y1

 
Variation between ED methods
The mean within-person variance was highest for EDall (49.8%) compared with {approx}20% for the other methods of calculating ED (Table 4Go). For the most conservative method of calculating ED, inclusion of beverages that provided >21 kJ/100 g (EDenergy), the within-person variance (20.4%) was similar to that for EDfood, EDsoup, and EDsolid. As expected, the correlation coefficients between ED methods were highly significant (Table 5Go).


View this table:
[in this window]
[in a new window]

  		TABLE 4. The between- and within-person CV for the energy density (ED) calculation methods1

 

View this table:
[in this window]
[in a new window]

  		TABLE 5. Correlation coefficients between each of the energy density (ED) calculation methods1

 
Self-defined eating occasions
When total food and drinks consumed were analyzed according to self-defined meal and snack occasions, no differences were observed between the ED of meals and snacks for EDall and EDenergy methods. However, the ED of snacks was significantly higher than was that of meals when ED was calculated by EDfood, EDsoup, and EDsolid methods (P < 0.001 in each case; Figure 1Go).


Figure 1
View larger version (6K):
[in this window]
[in a new window]

  		FIGURE 1.. Energy density (ED) of self-defined meals ({square}) and snacks ({square}) (median and interquartile ranges; n = 48). EDall, ED of all food and beverages; EDfood, ED of solid foods and milk as a food; EDsoup, ED of solid foods, milk as a food, and soup; EDsolid, ED of solid foods only; EDenergy, ED of foods and beverages in which energy was > 21 kJ/100 g. *Self-defined snacks were significantly different from meals, P < 0.001.

 
Association between ED and change in body composition over time
At the group level, the EDall and EDenergy of the total diet at baseline were not significantly related to change in body fatness (FMI) between baseline and follow-up (Table 6Go). However, the EDfood, EDsoup, and EDsolid of the total diet at baseline increased the odds of being in the highest category of FMIdif by 2.105 (95% CI: 1.081, 4.1; P = 0.029), 2.132 (95% CI: 1.098, 4.141; P = 0.025), and 1.922 (95% CI: 1.049, 3.525; P = 0.035), respectively (Table 6Go). Adjustment for misreporting at baseline (EI:EE) did not significantly alter the association with the change in FM, albeit the odds ratio increased slightly (Table 6Go).


View this table:
[in this window]
[in a new window]

  		TABLE 6. Logistic regression analyses of lowest gain compared with highest gain in fat mass index (FMI; in kg/m2) between childhood (6–8 y) and adolescence (13–16 y) according to energy density (ED) models1

 
When baseline EDs of self-defined meals and snacks were entered as 2 independent variables into the logistic regression analyses, they were not associated with change in body fat between baseline and follow-up for any of the ED methods at baseline (data not shown). No relations were found between ED of the total diet or self-defined meals and snacks at baseline by the ED methods with changes in the surrogate measures of body fat (BMI and WC z score change) either before or after adjustment for covariates and potential confounders (data not shown).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
In this cohort, ED of the total diet at baseline for 3 of the 5 methods for calculating ED were prospectively associated with change in body fat normalized for body size, but it was not associated with change in percentage of body fat, BMI, or WC z scores. These data highlight the importance of examining prospective associations between ED of the total diet at baseline and changes in body fatness with the use of different approaches of calculating ED and obesity risk, because the method chosen can greatly affect the interpretation of the findings. EDall or EDenergy were not associated with change in FMI, but when ED was calculated excluding beverages as EDfood, EDsoup, or EDsolid, participants who had the most energy-dense diet in childhood had the highest gain in FMI into adolescence.

Most studies in children and adults have examined the ED of the diet by excluding caloric or noncaloric beverages or both because the method used to measure dietary intake did not assess beverage intake, particularly water consumption, and therefore limited the ways to calculate ED. In studies of adults, the primary reason cited for examining ED of food only is that the within-person variance is lower than for of all food and beverages (35-37), which is also reflected in this study (Table 4Go). Although the most conservative method of calculating ED, inclusion of beverages that provide >21 kJ/100 g (EDenergy), had the lowest within-person variance, no relation was observed between EDenergy and change in FMI. However, when ED was calculated as EDfood, EDsoup, or EDsolid of the total diet, ED at baseline predicted those children who were the fattest by adolescence. The differences in the relation between ED methods and change in FMI are evident despite the high correlation between the ED methods (Table 5Go). The apparent subtle differences in calculating ED by various methods may result in a positive or null association between ED and change in FMI over time.

Although there are no reports of negative associations, the current inconsistencies about the relation, or lack of, between the ED of the diet and risk of overweight or obesity in children and adolescents may also be due to the common use of surrogate measures of body fatness such as BMI and WC. In this study, proxy measures attenuated any potential relation between ED and obesity status. The importance of normalizing body fat for height is also highlighted by the lack of association between change in percentage of body fat and ED. Indeed, most studies in children and adolescents have found no association between ED of the diet and measures of obesity status (including BMI, WC, or body fat) but have with risk factors for obesity (including race, poverty, and parenting practices) (5, 6, 8, 10). Large cross-sectional studies in adults have shown that the ED of the diet is associated with overweight and obesity [BMI (35, 37, 38)]. Similar to this cohort the associations also depend on the method used to calculate ED (39). One study in the United Kingdom that assessed body fat by dual-energy X-ray absorptiometry at age 9 y and ED (excluding all drinks, equivalent to EDsolid) with the use of 3-d unweighed diaries, found no association between ED of the diet at age 5 y and FMI at age 9 y, but a positive association was observed between ED at age 7 y and FMI at age 9 y (7).

In studies of children and adolescents, the role of snacking in promoting obesity remains unclear (18, 19, 21, 22). A unique aspect of this study is that meal and snack occasions were self-defined by respondents at the time of consumption, thus overcoming the issue of defining a snack occasion. In this cohort, the ED of self-defined meals or snacks by any method of calculating ED did not predict change in body fatness, although this could be partly explained by the low percentage of EI from snacks than from meals (24.4% compared with 75.6%, respectively). However, self-defined snacks had higher ED than did meals, but only when ED was calculated as EDfood, EDsoup, or EDsolid, once again highlighting that the method used to calculate ED can profoundly affect the conclusions drawn.

A strength of this study is the low amounts of EI misreported at baseline as assessed with estimates of TEE by DLW. Misreporting of EI affects the numerator (in kJ) and possibly the denominator (weight of food; in g/d) in the calculation of ED. Thus, attenuation of the relation between dietary intake and risk of obesity may occur in studies in which high misreporting is present or adjustment for misreporting is not considered. In the present analysis, adjustment for misreporting was included in the second regression model, but it was not found to be a significant predictor of change in FMI. Thus, EDfood, EDsoup, and EDsolid at baseline remained significant predictors of increased body fatness during a period of 8 y. Clearly, it is important that some evaluation of misreporting is included in studies of relations between potential dietary risk factors for obesity with the use of established techniques (40). Although dietary data by 7-d weighed diaries were collected at 13-17 y of age, unlike the diaries collected at 6-8 y of age, validation against TEE by DLW showed high underreporting of EI (66%). This high underreporting of EI would have severely impeded evaluation of the cross-sectional relations between ED and body composition at follow-up. In any case, the primary focus was to examine the relation between the exposure (ED at 6-8 y of age) in relation to change in obesity status between childhood and adolescence.

The small numbers of subjects who participated at baseline and follow-up (44% of the original cohort) are a limitation; however, there were no significant differences between anthropometric and dietary data for those subjects who participated at baseline only compared with those subjects who participated at baseline and follow-up. The subjects were originally studied at 6-8 y of age and again at 13-17 y of age. This interval removed the possibility of misclassifying rapid body fat gain because of variability in puberty status. By follow-up, the majority of subjects (83%) were beyond the peak puberty period (Tanner stages 4 or 5); thus, the rapid changes in body fatness and lean tissue mass associated with the peak puberty period were avoided. The accuracy of self-assessed puberty status is lower compared with physical examination (32, 41-43). In this 13-17-y-old age group a physical examination was not feasible in a follow-up observational study; nonetheless, it is important to include self-reported pubertal status as a covariate in the regression models. Thus, we can be confident that observed changes in fat mass are not due to some subjects who entered puberty earlier than others, as would be the case if body composition had been reassessed between the ages of 9 and 13 y. Unfortunately, there are no z score data for WC that has been measured at the umbilicus level, as in this study. Therefore, WC z scores were calculated with the use of data collected by McCarthy et al (31) that was measured at the natural waist level. Thus, it is possible that the method of measuring WC in this study may have attenuated the relation between ED of the diet and change in WC, although the magnitude of change may be expected to be similar (HD McCarthy, personal communication, July 2007). Although this is a small cohort, the study shows that the interpretation of the results differs considerably depending on the methods used to calculate ED and the accuracy of the measure of body composition or obesity status. Whether these associations between change in body composition and ED are replicated in a larger population study remains to be observed.

This study shows the dilemmas in the methods of both assessing adiposity status and ED of the diet. In this cohort, the inclusion of beverages in the calculation of ED attenuated the relation between ED at baseline and gain in FMI. Until a consensus can be reached on whether to include or exclude beverages, researchers in this area should err on the side of caution and present ED of the diet in as many of the ED methods as possible, as suggested by Cox and Mela (11) and Ledikwe et al (12), particularly if the results can be interpreted differently depending on the ED method used. This would allow for more comparisons to be made between studies over the relation between ED of the diet and risk of excess body fat gain. It would also allow more informed conclusions to be drawn from studies for the development of public health recommendations to alleviate the burden of obesity.


   ACKNOWLEDGMENTS
 
We thank the subjects and their parents for their willingness to participate in the study. We also thank David McCarthy and Jonathan Wells for their input into the study, Maranna McCloskey for assisting with collection of the data, and Antony Wright for the stable isotope analysis. We also thank Mark Chatfield for his advice on the statistical analysis.

The contributions of the authors were as follows—TAM and KLR: data acquisition and analysis and writing of the manuscript; KLR and MBEL: were responsible for the study concept and design, writing of the manuscript, and obtaining funding; SAJ and WAC: contributed to the study design, contribution to manuscript, and obtaining funding; JMW, MAK, and MPH-F: contributed to the writing of the manuscript. KLR is currently employed by Unilever, but at the time of the study she was employed by the University of Ulster. None of the other authors had a personal or financial conflict of interest.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 SUBJECTS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Ledikwe JH, Rolls BJ, Smiciklas-Wright H, et al. Reductions in dietary energy density are associated with weight loss in overweight and obese participants in the PREMIER trial. Am J Clin Nutr 2007;85:1212–21.[Abstract/Free Full Text]
  2. Ello-Martin JA, Roe LS, Ledikwe JH, Beach AM, Rolls BJ. Dietary energy density in the treatment of obesity: a year-long trial comparing 2 weight-loss diets. Am J Clin Nutr 2007;85:1465–77.[Abstract/Free Full Text]
  3. Rolls BJ, Drewnowski A, Ledikwe JH. Changing the energy density of the diet as a strategy for weight management. J Am Diet Assoc 2005;105(suppl):S98–103.[Medline]
  4. Ello-Martin JA, Ledikwe JH, Rolls BJ. The influence of food portion size and energy density on energy intake: implications for weight management. Am J Clin Nutr 2005;82(suppl):236S–41S.[Abstract/Free Full Text]
  5. Mendoza JA, Drewnowski A, Cheadle A, Christakis DA. Dietary energy density is associated with selected predictors of obesity in U.S. children. J Nutr 2006;136:1318–22.[Abstract/Free Full Text]
  6. Matheson DM, Robinson TN, Varady A, Killen JD. Do Mexican-American mothers' food-related parenting practices influence their children's weight and dietary intake? J Am Diet Assoc 2006;106:1861–5.[Medline]
  7. Johnson L, Mander AP, Jones LR, Emmett PM, Jebb SA. A prospective analysis of dietary energy density at age 5 and 7 years and fatness at 9 years among UK children. Int J Obes (Lond) 2007; October 2 (Epub ahead of print; DOI:10.1038/sj.ijo.0803746).
  8. Gibson SA. Associations between energy density and macronutrient composition in the diets of pre-school children: sugars vs. starch. Int J Obes Relat Metab Disord 2000;24:633–8.[Medline]
  9. Alexy U, Sichert-Hellert W, Kersting M, Schultze-Pawlitschko V. Pattern of long-term fat intake and BMI during childhood and adolescence–results of the DONALD Study. Int J Obes Relat Metab Disord 2004;28:1203–9.[Medline]
  10. Kral TV, Berkowitz RI, Stunkard AJ, Stallings VA, Brown DD, Faith MS. Dietary energy density increases during early childhood irrespective of familial predisposition to obesity: results from a prospective cohort study. Int J Obes (Lond) 2007;31:1061–7.[Medline]
  11. Cox DN, Mela DJ. Determination of energy density of freely selected diets: methodological issues and implications. Int J Obes Relat Metab Disord 2000;24:49–54.[Medline]
  12. Ledikwe JH, Blanck HM, Khan LK, et al. Dietary energy density determined by eight calculation methods in a nationally representative United States population. J Nutr 2005;135:273–8.[Abstract/Free Full Text]
  13. Livingstone MB, Robson PJ, Wallace JM. Issues in dietary intake assessment of children and adolescents. Br J Nutr 2004;92(suppl 2):S213–22.[Medline]
  14. Wells JC, Coward WA, Cole TJ, Davies PS. The contribution of fat and fat-free tissue to body mass index in contemporary children and the reference child. Int J Obes Relat Metab Disord 2002;26:1323–8.[Medline]
  15. Wells JC. A critique of the expression of paediatric body composition data. Arch Dis Child 2001;85:67–72.[Abstract/Free Full Text]
  16. Jahns L, Siega-Riz AM, Popkin BM. The increasing prevalence of snacking among US children from 1977 to 1996. J Pediatr 2001;138:493–8.[Medline]
  17. de Graaf C. Effects of snacks on energy intake: an evolutionary perspective. Appetite 2006;47:18–23.[Medline]
  18. Huang TTK, Howarth NC, Lin B-H, Roberts SB, McCrory MA. Energy intake and meal portions: associations with BMI percentile in U.S. children. Obes Res 2004;12:1875–85.[Medline]
  19. Nicklas TA, Morales M, Linares A, et al. Children's meal patterns have changed over a 21-year period: the Bogalusa Heart Study. J Am Diet Assoc 2004;104:753–61.[Medline]
  20. Phillips SM, Bandini LG, Naumova EN, et al. Energy-dense snack food intake in adolescence: longitudinal relationship to weight and fatness. Obes Res 2004;12:461–72.[Medline]
  21. Field AE, Austin SB, Gillman MW, Rosner B, Rockett HR, Colditz GA. Snack food intake does not predict weight change among children and adolescents. Int J Obes Relat Metab Disord 2004;28:1210–6.[Medline]
  22. Rodriguez G, Moreno LA. Is dietary intake able to explain differences in body fatness in children and adolescents? Nutr Metab Cardiovasc Dis 2006;16:294–301.[Medline]
  23. McGloin AF, Livingstone MB, Greene LC, et al. Energy and fat intake in obese and lean children at varying risk of obesity. Int J Obes Relat Metab Disord 2002;26:200–7.[Medline]
  24. Fomon SJ, Haschke F, Ziegler EE, Nelson SE. Body composition of reference children from birth to age 10 years. Am J Clin Nutr 1982;35:1169–75.[Free Full Text]
  25. Pace N, Rathbun EN. Studies on body composition. III. The body water and chemically combined nitrogen content in relation to fat content. J Biol Chem 1945;158:685–91.[Free Full Text]
  26. Wells JC, Fuller NJ, Dewit O, Fewtrell MS, Elia M, Cole TJ. Four-component model of body composition in children: density and hydration of fat-free mass and comparison with simpler models. Am J Clin Nutr 1999;69:904–12.[Abstract/Free Full Text]
  27. Rennie KL, Livingstone MB, Wells JC, et al. Association of physical activity with body-composition indexes in children aged 6–8 y at varied risk of obesity. Am J Clin Nutr 2005;82:13–20.[Abstract/Free Full Text]
  28. VanItallie TB, Yang MU, Heymsfield SB, Funk RC, Boileau RA. Height-normalized indices of the body's fat-free mass and fat mass: potentially useful indicators of nutritional status. Am J Clin Nutr 1990;52:953–9.[Abstract/Free Full Text]
  29. Cole TJ, Freeman JV, Preece MA. Body mass index reference curves for the UK, 1990. Arch Dis Child 1995;73:25–9.[Abstract/Free Full Text]
  30. Cole TJ, Bellizzi MC, Flegal KM, Dietz WH. Establishing a standard definition for child overweight and obesity worldwide: international survey. Br Med J 2000;320:1240–3.[Abstract/Free Full Text]
  31. McCarthy HD, Jarrett KV, Crawley HF. The development of waist circumference percentiles in British children aged 5.0–16.9 y. Eur J Clin Nutr 2001;55:902–7.[Medline]
  32. Taylor SJ, Whincup PH, Hindmarsh PC, Lampe F, Odoki K, Cook DG. Performance of a new pubertal self-assessment questionnaire: a preliminary study. Paediatr Perinat Epidemiol 2001;15:88–94.[Medline]
  33. Black AE, Cole TJ. Within- and between-subject variation in energy expenditure measured by the doubly-labelled water technique: implications for validating reported dietary energy intake. Eur J Clin Nutr 2000;54:386–94.[Medline]
  34. Drewnowski A, Almiron-Roig E, Marmonier C, Lluch A. Dietary energy density and body weight: is there a relationship? Nutr Rev 2004;62:403–13.[Medline]
  35. Ledikwe JH, Blanck HM, Kettel Khan L, et al. Dietary energy density is associated with energy intake and weight status in US adults. Am J Clin Nutr 2006;83:1362–8.[Abstract/Free Full Text]
  36. Ledikwe JH, Blanck HM, Khan LK, et al. Low-energy-density diets are associated with high diet quality in adults in the United States. J Am Diet Assoc 2006;106:1172–80.[Medline]
  37. Mendoza JA, Drewnowski A, Christakis DA. Dietary energy density is associated with obesity and the metabolic syndrome in U.S. adults. Diabetes Care 2007;30:974–9.[Abstract/Free Full Text]
  38. Howarth NC, Murphy SP, Wilkens LR, Hankin JH, Kolonel LN. Dietary energy density is associated with overweight status among 5 ethnic groups in the multiethnic cohort study. J Nutr 2006;136:2243–8.[Abstract/Free Full Text]
  39. Kant AK, Graubard BI. Energy density of diets reported by American adults: association with food group intake, nutrient intake, and body weight. Int J Obes (Lond) 2005;29:950–6.[Medline]
  40. Rennie KL, Jebb SA, Wright A, Coward WA. Under-reporting of energy intake in the National Diet and Nutrition Survey of young people. Br J Nutr 2005;93:241–7.[Medline]
  41. Desmangles JC, Lappe JM, Lipaczewski G, Haynatzki G. Accuracy of pubertal Tanner staging self-reporting. J Pediatr Endocrinol Metab 2006;19:213–21.[Medline]
  42. Bonat S, Pathomvanich A, Keil MF, Field AE, Yanovski JA. Self-assessment of pubertal stage in overweight children. Pediatrics 2002;110:743–7.[Abstract/Free Full Text]
  43. Wacharasindhu S, Pri-Ngam P, Kongchonrak T. Self-assessment of sexual maturation in Thai children by Tanner photograph. J Med Assoc Thai 2002;85:308–19.[Medline]
Received for publication September 26, 2007. Accepted for publication December 6, 2007.




This article has been cited by other articles:


Home page
 Diabetes CareHome page
J. Wang, R. Luben, K.-T. Khaw, S. Bingham, N. J. Wareham, and N. G. Forouhi
Dietary Energy Density Predicts the Risk of Incident Type 2 Diabetes: The European Prospective Investigation of Cancer (EPIC)-Norfolk Study
Diabetes Care, November 1, 2008; 31(11): 2120 - 2125.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by McCaffrey, T. A
Right arrow Articles by E Livingstone, M B.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by McCaffrey, T. A
Right arrow Articles by E Livingstone, M B.
Agricola
Right arrow Articles by McCaffrey, T. A
Right arrow Articles by E Livingstone, M B.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS