Résilience de la Chaîne Logistique Vaccinale Face à la COVID-19 en Côte d'Ivoire

## **Are City Features Related to Obesity in Preschool Children? Evidence from Latin America**

**Jessica Hanae Zafra-Tanaka1,2, Ariela Braverman3 , Cecilia Anza-Ramirez1 , Ana Ortigoza3 ,**

#### **Mariana Lazo3 , Tamara Doberti4 , Lorena Rodriguez-Osiac4 , Gina S. Lovasi3 , Monica Mazariegos5 , Olga Sarmiento6 , and J. Jaime Miranda1,7**

<sup>1</sup> CRONICAS Center of Excellence in Chronic Diseases, Universidad Peruana Cayetano Heredia, Lima, Perú

<sup>2</sup> Division of Tropical and Humanitarian Medicine, University of Geneva and Geneva University Hospitals, Geneva, Switzerland

3Urban Health Collaborative, Dornsife School of Public Health, Drexel University, Philadelphia, PA, USA

4Escuela de Salud Pública, Universidad de Chile, Santiago de Chile, Chile

5 INCAP Research Center for the Prevention of Chronic Diseases (CIIPEC), Institute of Nutrition of Central America and Panama (INCAP), Guatemala City, Guatemala

<sup>6</sup> School of Medicine, Universidad de los Andes, Bogota, Colombia

<sup>7</sup> School of Medicine, Universidad Peruana Cayetano Heredia, Lima, Perú

Urbanization shapes health through pathways involving cities' built and social environment features. Studies, mainly focused on high-income countries, have found links between these features and children and adolescents' obesity. Latin America (LA) is currently facing rapid urbanization and increases in childhood obesity. However, little is known about the burden of obesity in preschoolers and how cities could shape it. Thus, our study aimed to describe the prevalence of preschool children overweight/obesity across countries and cities in LA and to explore its association with city features. Cross-sectional analysis of 18,933 children aged 1-5 years who lived in 158 large cities in five countries in LA (Chile, Colombia, El Salvador, Mexico, and Peru) as part of the Salud Urbana en America Latina (SALURBAL) Project. We used individual-level anthropometric data from harmonized health surveys.

Overweight/obesity was defined as a z-score for weight-for-length/height>2, based on World Health Organization guidelines. We also characterized features of the social (living conditions, service provision, and educational attainment) and built environment (fragmentation, isolation, presence of mass transit, population density, intersection density, and percent greenness). We estimated the prevalence of childhood overweight/obesity in cities and used multi-level logistic models to explore associations between built and social, environmental features with overweight/obesity at the city and sub-city levels, adjusting for age, sex, and head of household education. The overall prevalence of overweight/obesity among pre-school children was 8% but varied substantially between and within countries. In the adjusted models, we found that lower odds of pre-school overweight/obesity in children were associated with higher distance between the urban patches-isolation- (OR:0.90, 95%CI:0.82-0.99) at the city-level and better educational attainment (OR:0.86, 95%CI:0.76-0.97) at the sub-city level. In contrast, higher odds were associated with better living conditions at the sub-city level (OR:1.15, 95%CI:1.00-1.31). In LA urban areas, we found substantial variability in the prevalence of childhood overweight/obesity between cities and countries. In the context of rapid urbanization in LA and other regions, our results suggest that features of the urban environment should be considered as key targets of policies to reduce childhood obesity.

Jessica Hanae Zafra-Tanaka1,2, Ariela Braverman3, Cecilia Anza-Ramirez1, Ana Ortigoza3, Mariana Lazo3, Tamara Doberti4, Lorena Rodriguez-Osiac4, Gina S. Lovasi3, Monica Mazariegos5, Olga Sarmiento6, and J. Jaime Miranda1,7

<sup>1</sup> CRONICAS Center of Excellence in Chronic Diseases, Universidad Peruana Cayetano Heredia, Lima, Perú

<sup>2</sup> Division of Tropical and Humanitarian Medicine, University of Geneva and Geneva University Hospitals, Geneva, Switzerland

3 Urban Health Collaborative, Dornsife School of Public Health, Drexel University, Philadelphia, PA, USA

<sup>4</sup> Escuela de Salud Pública, Universidad de Chile, Santiago de Chile, Chile

<sup>5</sup> INCAP Research Center for the Prevention of Chronic Diseases (CIIPEC), Institute of Nutrition of Central America and Panama (INCAP), Guatemala City, Guatemala

<sup>6</sup> School of Medicine, Universidad de los Andes, Bogota, Colombia

<sup>7</sup> School of Medicine, Universidad Peruana Cayetano Heredia, Lima, Perú

All authors are part of the SALURBAL project (Salud Urbana en America Latina).

#### Key Messages


#### Results

Prevalence of preschool overweight/obesity varied substantially between and within countries

#### Introduction

Urbanization shapes health through pathways involving cities' built and social environment features.

Latin America (LA) is currently facing rapid urbanization and increases in

childhood obesity.

Little is known about how cities can contribute to the risk of childhood obesity, particularly at early stages of life.

#### Objectives


Colombia 20I0, El Salvador 2008, Peru 20I6, Mexico 20I2, Chile 20I7 *Figure 1. City−level estimates of prevalence and 95%CI. Estmates are standardized to the 2010 SALURBAL population.*

Lower odds of OW/OB were associated with:


Higher odds of OW/OB were associated with better living conditions at the sub-city level.

*Table 1. Association between the built and social environment with preschool OW/ OB in I58 Latin American cities.*


#### Methods

Study design and sample: Cross-sectional analysis of 18,933 children aged 1-5 years who lived in 519 sub-cities of 158 cities with 100,000 or more residents in 5 countries in LA (Chile, Colombia, El Salvador, Mexico, and Peru).

Outcome: Overweight/obesity (a z-score for weight- for-length/height>2)

Exposure: city features of the social (living conditions, service provision and educational attainment) and built environment (fragmentation, isolation, presence of mass transit, population density, intersection density and percent greenness).

Statistical approach: multilevel logistic regression examining associations between built and social environmental features and OW/OB at the city and sub-city levels, adjusting for individual-level characteristics (age, sex, and head of household education).

> *Model adjusted for individual variables: age in months, sex, and head of household education, and country fixed effects.*
