Potential reductions in ultra-processed food consumption substantially improve population cardiometabolic-related dietary nutrient profiles in eight countries.

BACKGROUND AND AIMS: The negative effect on dietary nutrient profiles is the most obvious mechanism explaining the higher risk of cardiometabolic diseases associated with increased dietary share of UPF observed in large cohort studies. We estimate the proportion of diets with excessive energy density, excessive free sugars or saturated fat contents and insufficient fiber that could be avoided, if UPF consumption was reduced to levels among lowest consumers across eight countries, as well as the proportion of diets with multiple inadequacies. METHODS AND RESULTS: Using nationally-representative cross-sectional surveys from Brazil (2008-09), Chile (2010), Colombia (2005), Mexico (2012), Australia (2011-12), the UK (2008-16), Canada (2015), and the US (2015-16), inadequate energy density (≥2.25 kcal/g) or contents of free sugars (>10% of total energy intake), saturated fats (>10% of total energy intake) and fiber (<25 g/2000 kcal) population attributable fractions were quantified. Substantial reductions in nutrient inadequacies would be observed ranging from 50.4% in Chile to 76.8% in US for dietary energy density, from 15.5% in Colombia to 68.4% in Australia for free sugars, from 9.5% in Canada to 35.0% in Mexico for saturated fats, and from 10.3% in UK to 37.9% in Mexico for fiber. Higher reductions would be observed for diets with multiple nutrient inadequacies: from 27.3% in UK to 77.7% in Australia for ≥3 and from 69.4% in Canada to 92.1% in US, for 4 inadequacies. CONCLUSIONS: Lowering dietary contribution of UPF to levels among country-specific lowest consumers is a way to improve population cardiometabolic-related dietary nutrient profiles.

Ultra-Processed Food Consumption Among Chilean Preschoolers Is Associated With Diets Promoting Non-communicable Diseases.

Introduction: In adults, intake of ultra-processed foods (UPF) has been linked with poor diets and adverse health outcomes. In young children, evidence is scarcer but suggests a higher dietary share of UPF. Objective: To quantify the intake of UPF and its association with the nutrient composition of the diet in a sample of preschoolers in Santiago, Chile. Methods: Cross-sectional analysis of dietary data (24-h recall survey) from 960 preschoolers. Foods were categorized according to the extent and purpose of processing (NOVA classification) and participants were classified in quintiles of UPF intake. We explored the associations between UPF intake (% of the total energy) and intake of nutrients of concern for non-communicable disease development (carbohydrates, total sugars, fats, and sodium), and nutrients for promotion (proteins, polyunsaturated fats, iron, calcium, zinc, vitamins A, D, C, and B12, folate, and fiber) using multivariate regression after controlling for covariates. Results: UPF constituted 49% of the total energy intake. Preschoolers with higher intake consumed more energy, saturated and monounsaturated fats, carbohydrates, total sugars, and vitamin D, compared to preschoolers in the lowest quintile of UPF intake. In contrast, UPF intake was negatively associated with the consumption of proteins, polyunsaturated fats, fiber, zinc, vitamin A, and sodium (p < 0.05). Conclusion: In Chilean preschoolers, UPF was the primary source of energy intake. The dietary share of UPF was associated with the nutrient composition of the diet. Improving children's diet should consider not only promoting healthy food consumption but also limiting UPF consumption.

Precocious pubertal events in Chilean children: ethnic disparities.

PURPOSE: Puberty onset exhibits remarkable inter-individual and ethnic differences. 5% of Chileans are indigenous but puberty ethnic disparities have not been studied. We aim for evaluating precocious puberty prevalence in children with Mapuche ancestry vs non-indigenous Chilean children (according to their surnames). METHODS: Longitudinal cohort study: 1003 children (50.2% girls) participating in the Growth and Obesity Chilean Cohort Study (GOCS) were studied. Annual anthropometry was measured since 4-7 years. Subsequently, Tanner staging and anthropometry were measured every 6 months. In girls, Tanner stage was assessed by breast palpation and in boys by testicular volume measurements. The cohort was stratified in three groups depending on Mapuche surname numbers as follows: (A) no indigenous surnames (n = 811), (B) one to two indigenous surnames (n = 147), and (C) three or more indigenous surnames (n = 45). We evaluated the prevalence of precocious thelarche, pubarche, menarche and gonadarche (testicular volume ≥ 4 ml-G2), using a cutoff age of 8 years in girls and 9 years in boys while controlling for socioeconomic status, body mass index, waist circumference, IGF-1 and DHEAS at 7 years. RESULTS: In girls, no significant differences were observed. On the contrary, in boys, precocious gonadarche prevalence was higher in group C (29.2%) vs group A (6.0%) and vs group B (10.5%) (p =0.001, p = 0.004, respectively). Increased precocious gonadarche and pubarche risks in group C were observed even after adjustment [OR 7.31; 95% IC (2.32-23.51); p = 0.001] and [OR 6.17, 95% CI (1.62-23.49); p = 0.008], respectively. CONCLUSION: Indigenous origin in Chile is an independent risk factor for precocious gonadarche and pubarche in boys but not in girls.

Prevention of childhood obesity and food policies in Latin America: from research to practice.

BACKGROUND: Addressing childhood obesity in Latin America requires a package of multisectoral, evidence-based policies that enable environments conducive to healthy lifestyles. OBJECTIVE: Identify and examine key elements to translating research into effective obesity policies in Latin America. METHODS: We examined obesity prevention policies through case studies developed with an expert in the specific policy. Policies were selected based on their level of implementation, visibility and potential impact to reduce childhood obesity. They include: (i) excise taxes on sugar sweetened beverages and energy-dense foods; (ii) front-of-package food label legislation; (iii) trans fatty acids removal from processed foods; and (iv) Ciclovías recreativas or 'open streets'. Case studies were coded to identify components that explained successful implementation and sustainability using the Complex Adaptive Health Systems framework. RESULTS: The analysis identified key elements for effective and sustainable policy, including evidence justifying policy; evidence-based advocacy by civil society; political will; and legislation and skillful negotiations across government, academia, the private sector and civil society. Scientific evidence and evaluation played an important role in achieving tipping points for policies' launch and sustain effective implementation. CONCLUSIONS: Well-coordinated, intersectoral partnerships are needed to successfully implement evidence-based anti-obesity policies. Prospective policy research may be useful for advancing knowledge translation.

Nutrition status of children in Latin America.

The prevalence of overweight and obesity is rapidly increasing among Latin American children, posing challenges for current healthcare systems and increasing the risk for a wide range of diseases. To understand the factors contributing to childhood obesity in Latin America, this paper reviews the current nutrition status and physical activity situation, the disparities between and within countries and the potential challenges for ensuring adequate nutrition and physical activity. Across the region, children face a dual burden of undernutrition and excess weight. While efforts to address undernutrition have made marked improvements, childhood obesity is on the rise as a result of diets that favour energy-dense, nutrient-poor foods and the adoption of a sedentary lifestyle. Over the last decade, changes in socioeconomic conditions, urbanization, retail foods and public transportation have all contributed to childhood obesity in the region. Additional research and research capacity are needed to address this growing epidemic, particularly with respect to designing, implementing and evaluating the impact of evidence-based obesity prevention interventions.

Diseases due to unhealthy environments: an updated estimate of the global burden of disease attributable to environmental determinants of health.

Background: The update of the global burden of disease attributable to the environment is presented. The study focuses on modifiable risks to show the potential health impact from environmental interventions. Methods: Systematic literature reviews on 133 diseases and injuries were performed. Comparative risk assessments were complemented by more limited epidemiological estimates, expert opinion and information on disease transmission pathways. Population attributable fractions were used to calculate global deaths and global disease burden from environmental risks. Results: Twenty-three percent (95% CI: 13-34%) of global deaths and 22% (95% CI: 13-32%) of global disability adjusted life years (DALYs) were attributable to environmental risks in 2012. Sixty-eight percent of deaths and 56% of DALYs could be estimated with comparative risk assessment methods. The global disease burden attributable to the environment is now dominated by noncommunicable diseases. Susceptible ages are children under five and adults between 50 and 75 years. Country level data are presented. Conclusions: Nearly a quarter of global disease burden could be prevented by reducing environmental risks. This analysis confirms that eliminating hazards and reducing environmental risks will greatly benefit our health, will contribute to attaining the recently agreed Sustainable Development Goals and will systematically require intersectoral collaboration to be successful.

Anthropometric indicators as predictors of total body fat and cardiometabolic risk factors in Chilean children at 4, 7 and 10 years of age.

BACKGROUND/OBJECTIVE: To compare the association between anthropometric indicators of global and central obesity as predictors of total body fat (TBF) and cardiometabolic risk factors in children. SUBJECTS/METHODS: A total of 1044 children were evaluated at 4 years (n=320), 7 years (n=1044) and 10 years (n=483). The following anthropometric indices were determined: body mass index (BMI) for age (BAZ, WHO), waist circumference (WC), waist-to-hip ratio (WHR) and waist-to-height ratio (WHtR). To estimate TBF we used validated predictive equations. We measured blood sample concentrations of glucose, insulin, triglycerides, total cholesterol, Low-density lipoprotein (LDL) and High-density lipoprotein (HDL), adiponectin, C-reactive protein (CRP) and Insulin-like growth factor-1 (IGF-1). RESULTS: Adiposity and cardiometabolic markers, particularly those related to glucose metabolism increased from 4 years to 10 years. BAZ and WC were highly correlated to body fat at all ages (all r>0.8) but at 10 years WC was more strongly correlated than BAZ (r=0.94 WC vs r=0.88 BAZ, P<0.05); conversely, WHtR was significantly associated with body fat from 7 years (r=0.85) and 10 years (r=0.88). WHR was unrelated all over the period studied at all ages. Anthropometrical adiposity indicators became associated to cardiometabolic markers only from 7 years on with associations being slightly higher at 10 years, particularly for adiponectin and lipid markers. At all ages, BAZ, WC and WHtR performed similarly as cardiometabolic markers (P<0.05) while WHR was a slightly weaker marker. CONCLUSIONS: Relationship between anthropometrical indicators of adiposity and cardiometabolic markers becomes stronger from 7 years onwards; BAZ, WC and WHtR perform similarly as markers of cardiometabolic risk at least until 10 years of age.

GOCS cohort: children's eating behavior scores and BMI.

BACKGROUND/OBJECTIVE: In Chile, approximately one in three children under 6 years of age reported overweight/obese, while one in four children in elementary school suffer from obesity. There is a paucity of population-based information on the influence of childhood eating behavior on anthropometric measures related to obesity. To assess the association between eating behavior scores and Body Mass Index (BMI) z-scores in 7-10-year-old Chilean children. SUBJECTS/METHODS: We conducted a cross-sectional study in 1058 children aged 7-10 (51% girls) from the 'Growth and Obesity Chilean Cohort Study' (GOCS). Direct measures of weight and height were used to compute BMI z-scores according to World Health Organization (WHO) curves. Children were classified as normal weight (-1<1 s.d.), overweight (1<2 s.d.) and obese (⩾2 s.d.). Eating behavior scores were measured through the Child Eating Behavior Questionnaire (CEBQ), previously adapted and validated for Chilean children. Multiple linear regressions were carried out using BMI z-score as the outcome and eating behavior scores as explanatory variables. All models were adjusted by age and gender. RESULTS: BMI z-scores were positively associated with pro-intake scores in the subscales 'enjoyment of food', 'emotional overeating' and 'food responsiveness' (P<0.0001). Contrary to other studies, 'desire for drinks' scores were also associated with BMI z-scores (P<0.0001). In contrast, food-avoidant 'satiety responsiveness', 'slowness in eating' and 'food-fussiness' scores were negatively associated with BMI z-scores (P<0.0001). CONCLUSION: We found a significant relationship between eating behavior scores and BMI z-scores in Chilean children, showing that BMI in 7-10-year-old Chilean children is positively associated with pro-intake eating behavior scores and negatively associated with anti-intake eating behavior scores. The identification of specific eating behaviors patterns related to obesity will provide important information for the implementation of prevention programs for this disease.

Serum 25-Hydroxyvitamin D associated with indicators of body fat and insulin resistance in prepubertal chilean children.

BACKGROUND: Consistent data on the relation between vitamin D, body fat and insulin resistance (IR) in children are lacking. OBJECTIVES: (1) To evaluate the association between serum 25-Hydroxyvitamin D [25(OH)D] and key indicators of: adiposity (total and central), IR, and (2) to estimate serum 25(OH)D cut-offs that best reflect IR and total and central adiposity in children. SUBJECTS/METHODS: Prepubertal children (n=435, ~53% girls; ~age 7 years) from the Growth and Obesity Chilean Cohort Study were evaluated for potential associations between serum 25(OH)D and indicators of: (1) total adiposity (body mass index by age (BAZ), body fat (including three-component model)), central adiposity (waist circumference and trunk fatness); (2) IR (homeostasis model assessment of IR) and insulin sensitive (quantitative insulin sensitivity check index) using standardized multiple regression models with standardized coefficients and receiver operating characteristic curves. RESULTS: Overall, mean serum 25(OH)D was 32.1±9.2 ng ml(-1), while 19.4% of children were obese (BAZ⩾2 s.d.). Serum 25(OH)D was inversely associated with indicators of total and central adiposity and with IR indicators. Effect sizes were moderate in girls (~0.3 for adiposity and IR indicators), while, weaker values were found in boys. Serum 25(OH)D estimated cut-offs that best predicted total, central adiposity and IR were~30 ng ml(-1). Children with suboptimal serum 25(OH)D (<30 ng ml(-1)) had a higher risk (two to three times) of being obese (high BAZ, body fat percent and/or central adiposity); and three to four times greater risk for IR. CONCLUSIONS: Serum 25(OH)D was inversely associated with adiposity (total and central) and IR indicators in prepubertal Chilean children. The conventional cut-off of vitamin D sufficiency (⩾30 ng ml(-1)) was adequate to assess obesity and IR risk in this age group.

Evaluation of simple body composition methods: assessment of validity in prepubertal Chilean children.

BACKGROUND/OBJECTIVE: The aim of this study was to assess the validity of body fatness estimations based on skinfolds and bioelectrical iImpedance analyses (BIA) measurements compared to a three-component model (3C model) in prepubertal Chilean children, considering potential differences by sex and nutritional status. SUBJECTS/METHODS: Four hundred and twenty four Chilean children (198 females and 226 males) were assessed for body composition. Body fat percentage (BF%) was evaluated by Skinfold equations (Slaughter, Ramirez and Huang) and Bioelectrical impedance (BIA: Tanita BC-418MA) using both the equipment and the Ramirez equation. Measurements based on a 3C model constructed from total body water estimates by isotope dilution and from body volume estimates by air displacement plethysmography were used as gold standard. RESULTS: Coefficient of determination (R2) values were higher in overweight and in the whole group of both gender. All slopes were differed significantly from 1, and most intercepts were significantly different from 0. Skinfold Equations: an underestimation of BF% was found for all equations, being higher with the Slaughter equation. BIA: Tanita underestimated BF% in all groups, whereas Ramirez equation shows an overestimation. CONCLUSIONS: Skinfolds and bio-impedance equations serve well to rank children according to their BF%. However, these methods are not accurate for describing body composition in prepubertal Chilean children.

Early adiposity rebound is associated with metabolic risk in 7-year-old children.

BACKGROUND: Early adiposity rebound (AR <5 years) has been consistently associated with increased obesity risk, but its relationship with metabolic markers is less clear; in addition, the biologic mechanisms involved in these associations have not been established. OBJECTIVE: The objective of this study was to assess the association between timing of AR and metabolic status at age 7 years, evaluating the potential role of adiposity, adipose functionality and skeletal maturation in this association. DESIGN: We estimated the age of AR from the body mass index (BMI) trajectories from 0 to 7 years in 910 children from the Growth and Obesity Chilean Cohort Study (GOCS). At 7 years, we measured waist circumference (WC) and blood glucose, insulin, triglycerides and high-density lipoprotein-cholesterol levels and constructed a metabolic risk score. We also measured percent fat mass (adiposity), plasma concentrations of leptin and adiponectin (adipose functionality) and bone age using wrist ultrasound (skeletal maturation). RESULTS: We found that 44% of the children had an AR <5 years. Earlier AR was associated with larger WC (ß: 5.10 (95% confidence interval (CI): 4.29-5.91)), higher glucose (ß: 1.02 (1.00-1.03)), insulin resistance (ß Homeostatic Model Assessment: 1.06 (1.03-1.09)), triglycerides (ß: 10.37 (4.01-6.73)) and adverse metabolic score (ß: 0.30 (0.02-0.37)). Associations decreased significantly if adiposity was added to the models (i.e. ß WC: 0.85 (0.33-1.38)) and, to a lesser extent, when adipokines (i.e. ß WC: 0.73 (0.14-1.32)) and skeletal maturation (i.e. ß WC: 0.65 (0.10-1.20)) were added. CONCLUSION: In GOCS children, AR at a younger age predicts higher metabolic risk at 7 years; these associations are mostly explained by increased adiposity, but adipose dysfunction and accelerated skeletal maturation also have a role.

Structural responses to the obesity and non-communicable diseases epidemic: the Chilean Law of Food Labeling and Advertising.

In 12 July 2012, the Chilean Senate approved the Law of Food Labeling and Advertising, resulting from the joint efforts of a group of health professionals, researchers and legislators who proposed a regulatory framework in support of healthy diets and active living. Its goal was to curb the ongoing epidemic increase of obesity and non-communicable diseases. Two actions included: (i) improving point of food purchase consumer information by incorporating easy-to-understand front-of-packages labeling and specific messages addressing critical nutrients, and (ii) decreasing children's exposure to unhealthy foods by restricting marketing, advertising and sales. We summarize the work related to the law's release and discuss the conclusions reached by the various expert committees that were convened by the Ministry of Health to guide the development of the regulatory norms. Throughout the process, the food industry has overtly expressed its disagreement with the regulatory effort. The final content of the regulatory norms is still pending; however there are suggestions that its implementation will be delayed and might be modified based on the industry lobbying actions. These lessons should contribute to show the need of anticipating and addressing potential barriers to obesity-prevention policy implementation, particularly with respect to the role of the private sector.

Ambientes Saludables y Prevención de Enfermedades. Hacia una estimación de la carga de morbilidad atribuible al medio ambiente

¿EN QUÉ MEDIDA UN AMBIENTE MÁS SANO PUEDE CONTRIBUIR A EVITAR ENFERMEDADES? Esta es la pregunta en torno a la cual se sitúan los esfuerzos que venimos realizando a nivel mundial para combatir las causas básicas de los problemas de salud mediante estrategias de prevención más eficaces, en las que utilizamos todas las políticas, intervenciones y tecnologías de nuestro arsenal de conocimientos.

Conceptual basis for prescriptive growth standards from conception to early childhood: present and future.

BACKGROUND: Healthy growth in utero and after birth is fundamental for lifelong health and wellbeing. The World Health Organization (WHO) recently published standards for healthy growth from birth to 6 years of age; analogous standards for healthy fetal growth are not currently available. Current fetal growth charts in use are not true standards, since they are based on cross-sectional measurements of attained size under conditions that do not accurately reflect normal growth. In most cases, the pregnant populations and environments studied are far from ideal; thus the data are unlikely to reflect optimal fetal growth. A true standard should reflect how fetuses and newborns 'should' grow under ideal environmental conditions. OBJECTIVE: The development of prescriptive intrauterine and newborn growth standards derived from the INTERGROWTH-21(st) Project provides the data that will allow us for the first time to establish what is 'normal' fetal growth. METHODS: The INTERGROWTH-21(st) study centres provide the data set obtained under pre-established standardised criteria, and details of the methods used are also published. DESIGN: Multicentre study with sites in all major geographical regions of the world using a standard evaluation protocol. RESULTS: These standards will assess risk of abnormal size at birth and serve to evaluate potentially effective interventions to promote optimal growth beyond securing survival. DISCUSSION: The new normative standards have the potential to impact perinatal and neonatal survival and beyond, particularly in developing countries where fetal growth restriction is most prevalent. They will help us identify intrauterine growth restriction at earlier stages of development, when preventive or corrective strategies might be more effective than at present. CONCLUSION: These growth standards will take us one step closer to effective action in preventing and potentially reversing abnormal intrauterine growth. Achieving 'optimal' fetal growth requires that we act not only during pregnancy but that we optimize the maternal uterine environment from the time before conception, through embryonic development until fetal growth is complete. The remaining challenge is how 'early' will we be able to act, now that we can better monitor fetal growth.

Addressing malnutrition while avoiding obesity: minding the balance.

Latin America has experienced rapid demographic, epidemiological and nutrition changes that have successfully contributed to decreasing undernutrition, but concomitantly have resulted in an increase of obesity and associated conditions; in this paper we propose that policies to address undernutrition have not adapted at a sufficiently rapid pace to address the emerging challenges. Taking Chile as an example we show that health promotion policies, implemented only when the obesity epidemic was well advanced, have succeeded in establishing effective regulatory frameworks and in implementing national large-scale programs for treatment; however, they have been insufficient in preventing obesity. The main lessons learned are: (1) Failure to monitor existing malnutrition programs for changing needs and true effectiveness can have unintended consequences. (2) Institutions and professionals working in nutrition and health need to assess changing scenarios and redefine their priorities for action accordingly. (3) There is a need to provide updated information to decision makers, program planners and to the population at large on how to promote and achieve healthy food consumption and active living considering local context. Timely policies and interventions to address these issues may contribute to the prevention of the obesity epidemic in transitional countries, particularly among low socioeconomic groups.

Inactivation kinetics of inoculated Escherichia coli O157:H7, Listeria monocytogenes and Salmonella Poona on whole cantaloupe by chlorine dioxide gas.

The objectives of this study were to examine inactivation kinetics of inoculated Escherichia coli O157:H7, Listeria monocytogenes and Salmonella Poona inoculated onto whole cantaloupe and treated with ClO(2) gas at different concentrations (0.5, 1.0, 1.5, 3.0 and 5.0 mg l(-1)) for different times (0, 2.0, 4.0, 6.0, 8.0 and 10.0 min). The effect of ClO(2) gas on the quality and shelf life of whole cantaloupe was also evaluated during storage at 22 degrees C for 12 days. A 100 microl inoculation of each targeted organism was spotted onto the surface (5 cm(2)) of cantaloupe rind (approximately 8-9 log CFU 5 cm(-2)) separately, air dried (60 min), and then treated with ClO(2) gas at 22 degrees C and 90-95% relative humidity for 10 min. Surviving bacterial populations on cantaloupe surfaces were determined using a membrane transferring method with a non-selective medium followed by a selective medium. The inactivation kinetics of E. coli O157:H7, L. monocytogenes and S. Poona were determined using nonlinear kinetics (Weibull model). A 3 log CFU reduction of E. coli O157:H7, L. monocytogenes and S. Poona were achieved with 5.0 mg l(-1) ClO(2) gas for 5.5, 4.2 and 1.5 min, respectively. A 5l og CFU reduction of S. Poona was achieved with 5.0 and 3.0 mg l(-1) ClO(2) gas for 6 and 8 min, respectively. A 4.6 and 4.3 log reduction was achieved after treatment with 5.0 mg l(-1) ClO(2) gas at 10 min for E. coli O157:H7 and L. monocytogenes, respectively. Treatment with 5.0 mg l(-1) ClO(2) gas significantly (p<0.05) reduced the initial microflora (mesophilic bacteria, psychrotrophic bacteria, and yeasts and molds) on cantaloupe by more than 2 log CFU cm(-2) and kept them significantly (p<0.05) lower than the untreated control during storage at 22 degrees C for 12 days. Treatment with ClO(2) gas did not significantly (p>0.05) affect the color of whole cantaloupe and extended the shelf life to 9 days compared to 3 days for the untreated control, when stored at ambient temperature (22 degrees C).

Size at birth, infant, early and later childhood growth and adult body composition: a prospective study in a stunted population.

BACKGROUND: Pre-natal and post-natal growth are associated with adult body composition, but the relative importance of growth in different periods of childhood is still unclear, particularly in stunted populations. METHODS: We studied 358 women and 352 men measured as children in 1969-77 in four villages in Guatemala, and re-measured as adults in 2002-04 (mean age 32.7 years). We determined the associations of body mass index (BMI) and length at birth, and changes in BMI and length during infancy (0-1.0 year) and early (1.0-3.0 years) and later (3.0-7.0 years) childhood, with adult BMI ((a)BMI), percentage of body fat ((a)PBF), abdominal circumference ((a)AC) and fat-free mass ((a)FFM). RESULTS: Prevalence of stunting was high (64% at 3 years; HAZ < -2SD). Obesity (WHZ > 2SD) prevalence in childhood was <2%, while overweight prevalence in adulthood was 52%. BMI at birth was positively associated with (a)BMI and (a)FFM while length at birth was positively associated with (a)AC and (a)FFM. Increased BMI in infancy and later childhood were positively associated with all four adult body composition measures; associations in later childhood with fatness and abdominal fatness were stronger than those with (a)FFM. Change in length during infancy and early childhood was positively associated with all four adult body composition outcomes; the associations with (a)FFM were stronger than those with fat mass. CONCLUSIONS: Increases in BMI between 3.0 and 7.0 years had stronger associations with adult fat mass and abdominal fat than with (a)FFM; increases in length prior to age 3.0 years were most strongly associated with increases in (a)FFM.