The Journal of Nutrition Nutritional Epidemiology
Usual Dietary Energy Density Distribution Is Positively Associated with Excess Body Weight in Mexican Children1,2 3 Tania C Aburto,3 Alejandra Cantoral,3 Lucia Hernandez-Barrera, Alicia L Carriquiry,4 ´ 3 and Juan A Rivera * 3 Center for Nutrition and Health Research, National Institute of Public Health, Cuernavaca, Morelos, Mexico; and 4Department of Statistics, Iowa State University, Ames, IA
Background: Studies suggest a positive association between dietary energy density (DED) and body weight in adults, but evidence in children is inconclusive. Objective: The objective of this study was to compare usual DED distributions of nonoverweight vs. overweight or obese (OW/O) Mexican children. Methods: The study used 24-h recall (24HR) data from 2367 children aged 5–11 y from the 2012 Mexican National Health and Nutrition Survey (ENSANUT 2012). Repeated 24HR measures were obtained in a random sample (;10%) to estimate usual intake distributions by using the Iowa State University (PC-Side) method. Implausible dietary reports were identified. Multivariate linear regression models were used to evaluate the relation between DED and body mass index status and to compare results with and without PC-Side adjustment and restriction to plausible reporters. Results: A total of 35.1% of the children in the sample were OW/O. The usual DED mean was ;175 kcal/100 g in both the complete sample and the plausible reporters subsample. Regression models adjusted by PC-Side and for potential confounders showed higher DED in OW/O relative to nonoverweight children for both plausible reporters (9.7 kcal/100 g; n = 1452, P < 0.0001) and the complete sample (7.9 kcal/100 g; n = 2367, P < 0.0001). The DED difference in plausible reporters translates into 88 additional kilocalories in daily energy intake of OW/O children. In the absence of PC-side adjustment, the difference was significant for plausible reporters (P < 0.05) but not for the complete sample (P > 0.10). Conclusions: A positive association between usual DED and OW/O was found in Mexican children. The association was stronger when only plausible reporters were considered. This suggests that there is a need for strategies to reduce energy density in the diet of Mexican children.
Keywords:
J Nutr 2015;145:1524–30.
childhood obesity, dietary energy density, nutrition survey, diet, usual distribution, Mexico
Introduction Energy density is the amount of energy in food per unit weight (1), often expressed as kilocalories per 100 g. Studies have shown that dietary energy density (DED)5 may alter normal processes of appetite regulation, because we tend to consume a constant weight of food despite changes in energy density; thus, an increase in energy density leads to a higher energy intake, a concept referred to as ‘‘passive overconsumption’’ (2).
1
Supported by Bloomberg Philanthropies and the National Institute of Public Health, Mexico. 2 ´ Author disclosures: TC Aburto, A Cantoral, L Hernandez-Barrera, AL Carriquiry, and JA Rivera, no conflicts of interest. 5 Abbreviations used: DED, dietary energy density; ISU, Iowa State University; OW/O, overweight or obese; pER, predicted energy requirement; rEI, reported energy intake; WCRF, World Cancer Research Fund; 24HR, 24-h recall. * To whom correspondence should be addressed. E-mail: [email protected].
1524
A positive association between DED, energy intake, and weight gain has been observed in adults (3, 4). Consequently, energy density is considered a ‘‘convincing’’ risk factor for obesity by the WHO (5) and as a ‘‘probable’’ risk factor for obesity by the World Cancer Research Fund (WCRF) (6) expert panel. The WCRF encourages lowering the average DED toward 125 kcal/100 g as a public health goal. Nonetheless, the association between DED, energy intake, and weight status in children is still inconclusive. Some feeding studies in young children suggest that they are able to adjust their subsequent energy intake in response to variations in energy density of a preload (7–9). However, a study in 5- to 6-y-old children, in which portion size and energy density of 1 meal were increased, found that the children had greater energy intake in both experimental conditions (10), suggesting inadequate dietary compensation for increases in DED. Still, it is also positive that dietary compensation occurs over several days. Studies in adults
ã 2015 American Society for Nutrition. Manuscript received November 11, 2014. Initial review completed December 24, 2014. Revision accepted April 6, 2015. First published online April 29, 2015; doi:10.3945/jn.114.206359.
Downloaded from jn.nutrition.org at LIBRARIE DU UNIV DU QUEBEC METRO on July 15, 2015
Abstract
Methods Design and participants Data were obtained from the ENSANUT 2012, which was conducted in Mexico between October 2011 and May 2012. The survey had a probabilistic stratified cluster-sampling design and was representative at the national and state levels for rural and urban areas. Detailed sampling procedures are described elsewhere (19). For the purposes of this analysis, the study population consisted of a subsample of surveyed school-aged children (5–11 y), who had at least one 24-h recall (24HR) with complete anthropometric and socioeconomic information (n = 2399), and who were randomly selected. Children classified as thin were excluded (n = 32). Therefore, the analytic sample consisted of 2367 individuals of whom 228 had a second 24HR. Data collection and variable construction Anthropometric measurements. Weight was measured by using digital scales (model 872; Seca) with 0.1-kg precision, and height was measured by using stadiometers with 0.1-cm precision. Measurements were taken by trained personnel using standard procedures (20, 21). BMI was obtained as kilograms per meters squared (kg/m2), and z scores were calculated according to the WHO Reference 2007 (22). Children with a z score above +1 SD were classified as OW/O. BMI values between 65 SDs were considered valid, whereas values out of this range were excluded. Sociodemographic variables. Sociodemographic variables such as sex, age, and maternal education were obtained by using a household questionnaire. The country was divided into 4 geographic regions: North, Central, Mexico City, and South. Locations with
Usual Dietary Energy Density Distribution Is Positively Associated with Excess Body Weight in Mexican Children1,2 3 Tania C Aburto,3 Alejandra Cantoral,3 Lucia Hernandez-Barrera, Alicia L Carriquiry,4 ´ 3 and Juan A Rivera * 3 Center for Nutrition and Health Research, National Institute of Public Health, Cuernavaca, Morelos, Mexico; and 4Department of Statistics, Iowa State University, Ames, IA
Background: Studies suggest a positive association between dietary energy density (DED) and body weight in adults, but evidence in children is inconclusive. Objective: The objective of this study was to compare usual DED distributions of nonoverweight vs. overweight or obese (OW/O) Mexican children. Methods: The study used 24-h recall (24HR) data from 2367 children aged 5–11 y from the 2012 Mexican National Health and Nutrition Survey (ENSANUT 2012). Repeated 24HR measures were obtained in a random sample (;10%) to estimate usual intake distributions by using the Iowa State University (PC-Side) method. Implausible dietary reports were identified. Multivariate linear regression models were used to evaluate the relation between DED and body mass index status and to compare results with and without PC-Side adjustment and restriction to plausible reporters. Results: A total of 35.1% of the children in the sample were OW/O. The usual DED mean was ;175 kcal/100 g in both the complete sample and the plausible reporters subsample. Regression models adjusted by PC-Side and for potential confounders showed higher DED in OW/O relative to nonoverweight children for both plausible reporters (9.7 kcal/100 g; n = 1452, P < 0.0001) and the complete sample (7.9 kcal/100 g; n = 2367, P < 0.0001). The DED difference in plausible reporters translates into 88 additional kilocalories in daily energy intake of OW/O children. In the absence of PC-side adjustment, the difference was significant for plausible reporters (P < 0.05) but not for the complete sample (P > 0.10). Conclusions: A positive association between usual DED and OW/O was found in Mexican children. The association was stronger when only plausible reporters were considered. This suggests that there is a need for strategies to reduce energy density in the diet of Mexican children.
Keywords:
J Nutr 2015;145:1524–30.
childhood obesity, dietary energy density, nutrition survey, diet, usual distribution, Mexico
Introduction Energy density is the amount of energy in food per unit weight (1), often expressed as kilocalories per 100 g. Studies have shown that dietary energy density (DED)5 may alter normal processes of appetite regulation, because we tend to consume a constant weight of food despite changes in energy density; thus, an increase in energy density leads to a higher energy intake, a concept referred to as ‘‘passive overconsumption’’ (2).
1
Supported by Bloomberg Philanthropies and the National Institute of Public Health, Mexico. 2 ´ Author disclosures: TC Aburto, A Cantoral, L Hernandez-Barrera, AL Carriquiry, and JA Rivera, no conflicts of interest. 5 Abbreviations used: DED, dietary energy density; ISU, Iowa State University; OW/O, overweight or obese; pER, predicted energy requirement; rEI, reported energy intake; WCRF, World Cancer Research Fund; 24HR, 24-h recall. * To whom correspondence should be addressed. E-mail: [email protected].
1524
A positive association between DED, energy intake, and weight gain has been observed in adults (3, 4). Consequently, energy density is considered a ‘‘convincing’’ risk factor for obesity by the WHO (5) and as a ‘‘probable’’ risk factor for obesity by the World Cancer Research Fund (WCRF) (6) expert panel. The WCRF encourages lowering the average DED toward 125 kcal/100 g as a public health goal. Nonetheless, the association between DED, energy intake, and weight status in children is still inconclusive. Some feeding studies in young children suggest that they are able to adjust their subsequent energy intake in response to variations in energy density of a preload (7–9). However, a study in 5- to 6-y-old children, in which portion size and energy density of 1 meal were increased, found that the children had greater energy intake in both experimental conditions (10), suggesting inadequate dietary compensation for increases in DED. Still, it is also positive that dietary compensation occurs over several days. Studies in adults
ã 2015 American Society for Nutrition. Manuscript received November 11, 2014. Initial review completed December 24, 2014. Revision accepted April 6, 2015. First published online April 29, 2015; doi:10.3945/jn.114.206359.
Downloaded from jn.nutrition.org at LIBRARIE DU UNIV DU QUEBEC METRO on July 15, 2015
Abstract
Methods Design and participants Data were obtained from the ENSANUT 2012, which was conducted in Mexico between October 2011 and May 2012. The survey had a probabilistic stratified cluster-sampling design and was representative at the national and state levels for rural and urban areas. Detailed sampling procedures are described elsewhere (19). For the purposes of this analysis, the study population consisted of a subsample of surveyed school-aged children (5–11 y), who had at least one 24-h recall (24HR) with complete anthropometric and socioeconomic information (n = 2399), and who were randomly selected. Children classified as thin were excluded (n = 32). Therefore, the analytic sample consisted of 2367 individuals of whom 228 had a second 24HR. Data collection and variable construction Anthropometric measurements. Weight was measured by using digital scales (model 872; Seca) with 0.1-kg precision, and height was measured by using stadiometers with 0.1-cm precision. Measurements were taken by trained personnel using standard procedures (20, 21). BMI was obtained as kilograms per meters squared (kg/m2), and z scores were calculated according to the WHO Reference 2007 (22). Children with a z score above +1 SD were classified as OW/O. BMI values between 65 SDs were considered valid, whereas values out of this range were excluded. Sociodemographic variables. Sociodemographic variables such as sex, age, and maternal education were obtained by using a household questionnaire. The country was divided into 4 geographic regions: North, Central, Mexico City, and South. Locations with
