Risk profiles for overweight/obesity among preschoolers

doi:10.1016/j.earlhumdev.2010.07.006

Early Human Development

Volume 86, Issue 9, September 2010, Pages 563-568

https://doi.org/10.1016/j.earlhumdev.2010.07.006 Get rights and content

Abstract

Background

The epidemic of overweight/obesity among U.S. children has led to an alarming increase in health-related consequences, including early-onset diabetes and cardiovascular disease. Recent research has identified the independent contribution of several maternal and child factors to the development of childhood overweight/obesity. Few studies, however, have examined risk profiles of childhood obesity.

Aim

This study used classification and regression tree (CART) analysis to examine the combined effect of maternal and child factors in generating risk profiles for overweight/obesity among preschoolers.

Study design

Data from the Early Childhood Longitudinal Study-Birth Cohort (ECLS-B) study were used. The sample was comprised of preschool children. CART and logistic regression models were built and compared.

Results

Children who were overweight/obese at two years of age had an increased risk of being overweight/obese at four years of age. Children born to overweight/obese mothers were more likely to be overweight/obese by age four, even if their BMI at two years of age was normal. Children with high birth weight (≥ 4000 g.) were also more likely to be overweight/obese at age four years if they were born to mothers with a normal pregravid BMI, but were of a lower socioeconomic status. Among preschoolers whose mothers were black or white and who had a high pregravid BMI, breastfeeding duration and parity played an important role in determining their risk of being overweight/obese.

Conclusions

Classification tree analysis confirms and extends current knowledge of preschool overweight/obesity by providing preliminary risk profiles that are structured within the context of prenatal and postnatal maternal and child characteristics.

Introduction

Escalating childhood overweight and obesity rates represent a growing epidemic in the U.S. Data from the National Health and Nutrition Examination Surveys (NHANES) (1976–1980 and 2003–2006) demonstrated an increase from 5% to 12.4% in the prevalence of obesity among two- to five-year old children [1], [2]. More recently, data from the nationally-representative Early Childhood Longitudinal Study, Birth Cohort (ECLS-B) found an obesity rate of 18.4% among four year old children [3]. When overweight and obesity were considered together, 2007–2008 NHANES surveys found that the prevalence rate for BMI ≥ 85th percentile among preschool children was 21.2%. The health-related consequences of preschool overweight/obesity are pervasive. Psychosocial outcomes include early social discrimination that leads to low self-esteem, poor academic performance, and a delayed emotional development that may persist into adulthood [4]. The physiological sequelae, including cardiovascular disease and early-onset diabetes, have been widely reported [5], [6].

Maternal and child characteristics have been explored for their independent association with childhood overweight/obesity. Several studies have found direct relationships with race and ethnicity, pregravid and gestational obesity, smoking during pregnancy, and reduced breastfeeding duration [7], [8], [9], [10], [11], [12], [13]. Gillman et al. [12] found that children exposed to adverse maternal behaviors during and after pregnancy (i.e., smoking, high gestational weight gain, and reduced breastfeeding duration) were at an increased risk of overweight at three years of age. Other research has examined the combined effect of risk factors on overweight for older children [13].

To our knowledge, no previous studies have explored the combined effect of maternal and early childhood factors in developing pediatric overweight/obesity risk profiles. The purpose of this study was to use classification and regression trees (CART) to generate profiles of preschool children (approximately 4 years old) who are at risk for overweight/obesity based on a set of prenatal and postnatal maternal and child characteristics that have been found in previous research to have an independent association with pediatric obesity.

CART is an innovative approach in the field of perinatal and pediatric epidemiology that offers an alternative to parametric regression techniques [14]. It uncovers complex relationships among a large number of variables and provides information on their combined effect in predicting a health outcome. In this study, the CART methodology allowed for the construction of risk profiles, based on a large number of both non-modifiable and modifiable factors related to preschool overweight/obesity that have the potential to adversely affect long-term health.

Section snippets

Study population

This study used data from the ECLS-B, which is sponsored by the U.S. Department of Education, National Center for Education Statistics (NCES) [15]. This study was also approved by the Human Subject Review Board at the researchers' institution. The ECLS-B is a nationally-representative data set, designed to provide detailed information on child development, health, and early learning experiences. The children participating in the ECLS-B were followed from birth (2001) through kindergarten entry.

Results

The distribution of the maternal and child characteristics that were considered in the construction of risk profiles for preschool overweight/obesity is displayed in Table 1. Significant associations were found between children's BMI status at four years of age and several maternal characteristics. A higher proportion of children born to Hispanic mothers were overweight/obese compared to those born to white or black mothers. Overweight/obese children were more likely to have mothers who were

Discussion

The current study used the nationally-representative ECLS-B study and the CART methodology to construct preliminary risk profiles for preschooler overweight/obesity within the context of a wide array of maternal and child characteristics. The individual variables that emerged in the risk profiles were comparable to those found when traditional logistic regression analysis was applied to this data set. The association between these variables and childhood obesity also has been well-documented in

Conclusions

The prevalence of childhood overweight and obesity has increased rapidly in the past two decades and constitutes a serious public health problem in the U.S. Halting this epidemic requires prevention efforts that begin early in life. This exploratory study uses a novel methodological approach and a current, nationally-representative sample to identify risk profiles for preschool overweight/obesity. The findings confirm current knowledge about the associations that socioeconomic status,

Conflict of interest statement

The authors declare that they have no conflicts of interest.

References (37)

D.S. Freedman et al.
Cardiovascular risk factors and excess adiposity among overweight children and adolescents: the Bogalusa Heart Study
J Pediatr
(2007)
E. Oken et al.
Gestational weight gain and child adiposity at age 3 years
Am J Obstet Gynecol
(2007)
K.R. Welch et al.
Characteristics of obese children aged 1–4 years at a referral clinic
J Natl Med Assoc
(2008)
P. Kitsantas et al.
Assessing adverse birth outcomes via classification trees
Artif Intell Med
(2006)
P.B. Crawford et al.
Counseling Latina mothers of preschool children about weight issues: suggestions for a new framework
J Am Diet Assoc
(2004)
K.M. Sussner et al.
The influence of maternal acculturation on child body mass index at age 24 months
J Am Diet Assoc
(2009)
J. Armstrong et al.
Breastfeeding and lowering the risk of childhood obesity
Lancet
(2002)
P. Kitsantas et al.
Assessing the stability of classification trees
J Stat Plan Inference
(2007)
NHANES data on the prevalence of overweight among children and adolescents: United States, 2003–2006. CDC National...
C.L. Ogden et al.
High body mass index for age among US children and adolescents, 2003–2006
JAMA
(2008)

S.E. Anderson et al.

Prevalence of obesity among US preschool children in different racial and ethnic groups

Arch Pediatr Adolesc Med

(2009)

M.B. Swatz et al.

Childhood obesity: a societal problem to solve

Obesity Rev

(2003)

A. Must et al.

Effects of obesity on morbidity in children and adolescents

Nutr Clin Care

(2003)

A. Chen et al.

Maternal smoking during pregnancy in relation to child overweight: follow-up to age 8 years

Int J Epidemiol

(2006)

T. Harder et al.

Duration of breastfeeding and risk of overweight: a meta-analysis

Am J Epidemiol

(2005)

P.J. Salsberry et al.

Dynamics of early childhood overweight

Peidatrics

(2005)

M.W. Gillman et al.

Developmental origins of childhood overweight: potential public health impact

Obesity

(2008)

A.M. Toschke et al.

Children at high risk for overweight: a classification and regression trees analysis approach

Obes Res

(2005)

Cited by (66)

From childhood obesity risk to healthy growth in the U.S.: A 10-year social work research & policy update
2023, Preventive Medicine Reports
Citation Excerpt :
Children with obesity have poorer physical and mental health, making prevention across critical stages of development crucial. Despite the positive association between economic inequality and obesity risk (Gibbs and Forste, 2014; Kitsantas and Gaffney, 2010; Schuler et al., 2021; Vazquez and Cubbin, 2020; Schuler, 2019), intervention trials among low-income samples have typically yielded null findings, attributed to barriers in economically depressed neighborhoods (Moore et al., 2019; Dietz, 2019; Kumanyika, 2017; Skinner et al., 2016; Robinson et al., 2010; Butte et al., 2017; Woo Baidal et al., 2017; Barkin et al., 2018; French et al., 2018). Interventions must move beyond individual behaviors to address contextual factors driving weight-related health disparities (Davison et al., 2013).
Childhood obesity is a major health issue and a prominent chronic health condition for children in the United States (U.S.), caused by a multitude of factors. Most existing models of childhood obesity prevention have not worked, yielding little to no effect on improving weight status or the proximal health behaviors most attributed to obesity risk: nutritional intake, physical activity, sedentary behaviors, and sleep. There is an urgent need for new approaches to prevent health disparities that are responsive to impacts of economic inequality on healthy child growth in marginalized populations. In this Short Commentary, a social justice update is provided to motivate a new generation of research that promotes equitable and healthy child growth under present-day social, economic, and political circumstances. Social work-specific research and policy recommendations are provided to guide future research that targets underlying social and economic determinants of weight-related health disparities in childhood. Recommendations include research on cross-disciplinary metrics to better capture reductions in health disparities and the development and testing of policy and system interventions that address structural issues and strengthen health resources in marginalized communities. Progress in reducing disparities in childhood obesity will likely remain inhibited until recommendations from social work research are incorporated to strengthen existing medical and public health models and redirect the childhood obesity epidemic toward equitable, healthy child growth.
Identifying overnutrition risk groups in Bangladeshi married women based on sociodemographic factors: A classification and regression tree model
2022, Obesity Medicine
Citation Excerpt :
The route from root node to leaf node is rule-based (e.g., “if-then” rule). Several splitting criteria are used in the partitioning process, and of them, Gini index (the measurement of impurity) is the most popular and extensively used criteria (Kitsantas and Gaffney, 2010), which was also used in this study. The Gini index ranges between 0 (purity of classification) to 1 (randomness of the sample distribution among different classes).
Overweight and obesity among women are major health issues in Bangladesh. To develop preventive strategies to deal with this epidemic, identifying women who are at risk for overweight and obesity is critical. This study aims to identify married women at high risk for overweight and obesity in Bangladesh.
The study has used a sample of 14207 married women aged 15–49 years with a body mass index (BMI) ≥18.5 kg/m² from the 2017-18 Bangladesh Demographic and Health Survey. The overweight and obese status of women based on BMI is the primary outcome variable, and a range of sociodemographic variables are included as predictors. The Classification and Regression Tree (CART) is used to identify important predictors and high-risk groups.
The CART models show that household wealth status is the most crucial predictor of overweight and obesity in married women. Women aged 26 to 49 years who live in affluent households and do not work or breastfeed are at the high-risk group.
This study suggests that the formulation of strategies to prevent overweight and obesity in married women in Bangladesh should be tailored for different social classes and age groups. To reduce the burden, preventive interventions may target the high-risk group.
From Birth to Overweight and Atopic Disease: Multiple and Common Pathways of the Infant Gut Microbiome
2021, Gastroenterology
Citation Excerpt :
We extend these observations by identifying cesarean birth as the prime initiating event and by showing that E/B abundance ratio associations with atopy also involve Veillonellaceae and C difficile in later infancy. Pregnancy or birth order associations with atopic disease are well documented46 and emerging in relation to offspring obesity.47,48 We found first pregnancy to be associated with lowered abundance of bifidobacteria but with a higher E/B ratio and greater colonization of C difficile in 3-month-old infants.
Few studies, even those with cohort designs, test the mediating effects of infant gut microbes and metabolites on the onset of disease. We undertook such a study. Methods: Using structural equation modeling path analysis, we tested directional relationships between first pregnancy, birth mode, prolonged labor and breastfeeding; infant gut microbiota, metabolites, and IgA; and childhood body mass index and atopy in 1667 infants.
After both cesarean birth and prolonged labor with a first pregnancy, a higher Enterobacteriaceae/Bacteroidaceae ratio at 3 months was the dominant path to overweight; higher Enterobacteriaceae/Bacteroidaceae ratios and Clostridioides difficile colonization at 12 months were the main pathway to atopic sensitization. Depletion of Bifidobacterium after prolonged labor was a secondary pathway to overweight. Influenced by C difficile colonization at 3 months, metabolites propionate and formate were secondary pathways to child outcomes, with a key finding that formate was at the intersection of several paths.
Pathways from cesarean section and first pregnancy to child overweight and atopy share many common mediators of the infant gut microbiome, notably C difficile colonization.
Improving birth and breastfeeding outcomes among low resource women in Alabama by including doulas in the interprofessional birth care team
2019, Journal of Interprofessional Education and Practice
Birth outcome disparities are particularly evident in Alabama and evidence has shown that including doulas in an interprofessional birth care team (IBCT) may improve these outcomes.
The purpose of this study is to assess whether including doulas in the birth care team was associated with improved birth outcomes among low resource mothers in Alabama.
Doula supported birth outcomes in 2013–2014 (n = 120) were restrospectively compared to all 2014 Medicaid funded births in Jefferson County, Alabama (n = 3782).
Doula supported births were associate with lower rates of epidural anesthesia and birth by cesarean as compared to the reference population (OR 3.0, 95% CI 2.1–4.4 and OR 1.8, 95% CI 1.1–2.9, respectively, for reference population). Doulas were also associated with a ten-fold increase in breastfeeding initiation (OR 10.5, 95% CI 5.4–23.2).
Doula support is associated with increased breastfeeding initiation and reduced rates of epidurals and cesareans among low resource mothers in Alabama.
Risk Factors for Childhood Obesity in the First 1,000 Days: A Systematic Review
2016, American Journal of Preventive Medicine
Citation Excerpt :
Twenty-eight studies40,42,53,62,68,69,78,79,90,98,163,168–184 examined the association between birth weight and later child overweight. Higher birth weight was consistently associated with later childhood overweight, with 24 studies40,42,53,68,69,78,79,90,98,163,168–175,177–181,184 identifying an association between higher birth weight and later overweight. Studies defined birth weight as a continuous measure, as categorical absolute weight (e.g., macrosomia), or in reference to gestational age (e.g., large for gestational age).
Mounting evidence suggests that the origins of childhood obesity and related disparities can be found as early as the “first 1,000 days”—the period from conception to age 2 years. The main goal of this study is to systematically review existing evidence for modifiable childhood obesity risk factors present from conception to age 2 years.
PubMed, Embase, and Web of Science were searched for studies published between January 1, 1980, and December 12, 2014, of childhood obesity risk factors present during the first 1,000 days. Prospective, original human subject, English-language research with exposure occurrence during the first 1,000 days and with the outcome of childhood overweight or obesity (BMI ≥85th percentile for age and sex) collected between age 6 months and 18 years were analyzed between December 13, 2014, and March 15, 2015.
Of 5,952 identified citations, 282 studies met inclusion criteria. Several risk factors during the first 1,000 days were consistently associated with later childhood obesity. These included higher maternal pre-pregnancy BMI, prenatal tobacco exposure, maternal excess gestational weight gain, high infant birth weight, and accelerated infant weight gain. Fewer studies also supported gestational diabetes, child care attendance, low strength of maternal–infant relationship, low SES, curtailed infant sleep, inappropriate bottle use, introduction of solid food intake before age 4 months, and infant antibiotic exposure as risk factors for childhood obesity.
Modifiable risk factors in the first 1,000 days can inform future research and policy priorities and intervention efforts to prevent childhood obesity.
Prospective BMI changes in preschool children are associated with parental characteristics and body weight perceptions: the ToyBox-study
2022, Public Health Nutrition

View all citing articles on Scopus

¹: Tel.: + 1 703 993 1961; fax: + 1 703 993 1942.

View full text

Published by Elsevier Ireland Ltd.

Risk profiles for overweight/obesity among preschoolers

Abstract

Background

Aim

Study design

Results

Conclusions

Introduction

Section snippets

Study population

Results

Discussion

Conclusions

Conflict of interest statement

J Pediatr

Am J Obstet Gynecol

J Natl Med Assoc

Artif Intell Med

J Am Diet Assoc

J Am Diet Assoc

Lancet

J Stat Plan Inference

High body mass index for age among US children and adolescents, 2003–2006

JAMA

Prevalence of obesity among US preschool children in different racial and ethnic groups

Arch Pediatr Adolesc Med

Childhood obesity: a societal problem to solve

Obesity Rev

Effects of obesity on morbidity in children and adolescents

Nutr Clin Care

Maternal smoking during pregnancy in relation to child overweight: follow-up to age 8 years

Int J Epidemiol

Duration of breastfeeding and risk of overweight: a meta-analysis

Am J Epidemiol

Dynamics of early childhood overweight

Peidatrics

Developmental origins of childhood overweight: potential public health impact

Obesity

Children at high risk for overweight: a classification and regression trees analysis approach

Obes Res