Abstract

Objective

The SUECO study examines the relationship between urban obesogenic environments and health outcomes among school-age children in the city of Madrid, Spain. We will study how features of the urban environment (related to the food- and the physical activity environment) associate with children's anthropometrics, eating habits, and physical activity levels.

Method

We describe the study protocol of this multilevel study in a representative sample of school-age children in the city of Madrid (2017; n=5,961 children ages 3-12). Main outcome variables include anthropometrics (body mass index, waist circumference, and body fat), healthy and unhealthy consumption measures, and physical activity measures. The primary explanatory variables are grouped into food environment (e.g., unhealthy food retailers' density) and physical activity environment (e.g., walkability, physical activity opportunities) variable categories. Multilevel models will be used to calculate the associations between each indicator and obesity and physical inactivity.

Keywords:
Obesity; Body mass index; Children; Schools; Neighborhoods; Spain

Resumen

Objetivo

El estudio SUECO evaluará la relación entre el entorno urbano obesogénico y los resultados en salud en la población infantil escolarizada en Madrid. El estudio explorará cómo influyen las características del entorno urbano (relacionadas con la alimentación y con la actividad física) en las variables antropométricas, los hábitos alimentarios y la actividad física.

Método

Se presenta el protocolo de este estudio multinivel, realizado con una muestra representativa de población infantil escolarizada en la ciudad de Madrid (2017; n=5961 escolares de 3 a 12 años). Las principales variables de resultado incluyen antropometría (índice de masa corporal, circunferencia de cintura y porcentaje de grasa corporal), hábitos alimentarios (saludables y no saludables) y actividad física. Las principales variables independientes son contextuales (del entorno alimentario y del entorno de actividad física). Se utilizarán modelos de regresión multinivel para evaluar las asociaciones entre cada indicador contextual, la obesidad y la inactividad física.

Palabras clave:
Obesidad; Índice de masa corporal; Población infantil; Centros educativos; Barrios; España

Introduction

Malnutrition (undernutrition, overweight/obesity, and poor diets) is a major health and social problem worldwide.¹1. Lobstein T, Baur L, Uauy R. Obesity in children and young people: a crisis in public health. Obes Rev. 2004;5:4-85. Moreover, malnutrition during childhood and adolescence is associated with adverse health consequences. The estimated prevalence of childhood overweight and obesity is about 25% in Europe; however, the burden is not homogeneously distributed across the region.²2. Wijhoven TM, van Raaij JM, Spinelli A, et al. WHO European Childhood Obesity Surveillance Initiative 2008: weight, height and body mass index in 6-9-year-oldchildren. Pediatr Obes. 2013;8:79-97. Indeed, estimated prevalences range from about 11% in Belgium to about 40% in Greece.²2. Wijhoven TM, van Raaij JM, Spinelli A, et al. WHO European Childhood Obesity Surveillance Initiative 2008: weight, height and body mass index in 6-9-year-oldchildren. Pediatr Obes. 2013;8:79-97. According to the most recent national health survey in Spain, about 25% of children (aged 2 to 17 years) presented overweight or obesity in Madrid in 2017.³3. Ministerio de Sanidad, Consumo y Bienestar Social. Encuesta Nacional de Salud España 2017. [Accesed 16 november 2021]. Available at: https://www.mscbs.gob.es/estadEstudios/estadisticas/encuestaNacional/encuestaNac2017/encuestaResDetall2017.htm.
https://www.mscbs.gob.es/estadEstudios/e... These prevalence rates are partially attributed to a gradual shift in these countries, from a Mediterranean Diet to a more Westernized diet. Thus, at the forefront of malnutrition and poor children's health is the obesogenic environment.⁴4. Swinburn B, Egger G, Raza F. Dissecting obesogenic environments: the development and application of a framework for identifying and prioritizing environmental interventions for obesity. Prev Med. 1999;29:563-70.

Swinburn et al.⁴4. Swinburn B, Egger G, Raza F. Dissecting obesogenic environments: the development and application of a framework for identifying and prioritizing environmental interventions for obesity. Prev Med. 1999;29:563-70. defined the obesogenic environment as “the sum of the influences that the surroundings, opportunities, or conditions of life have on promoting obesity”. This obesogenic environment contributes to overweight/obesity by encouraging unhealthy diets and sedentarism.⁵5. Swinburn B, Egger G. Preventive strategies against weight gain and obesity. Obes Rev. 2002;4:289-301. For children, this corresponds to an increased access to unhealthy food and sugar-sweetened beverages.⁶6. Diez J, Cebrecos A, Rapela A, et al. Socioeconomic inequalities in the retail food environment around schools in a Southern European context. Nutrients. 2019;11:15. Furthermore, accessibility to recreational facilities (e.g., playgrounds), walkability, sports facilities, and active school transportation opportunities are associated with increased physical activity.⁷7. Ikeda E, Stewart T, Garrett N, et al. Built environment associates of active school travel in New Zealand children and youth: a systematic meta-analysis using individual participant data. J Transp Health. 2018;9:117-31.^,88. McGrath LJ, Hinckson EA, Hopkins WG, et al. Associations between the neighborhood environment and moderate-to-vigorous walking in New Zealand children: findings from the URBAN Study. Sport Med. 2016;46:1003-7.

Yet, scarce research has simultaneously examined more than one dimension of the obesogenic environment concerning children's obesity.⁹9. Nau C, Ellis H, Huang H, et al. Exploring the forest instead of the trees: an innovative method for defining obesogenic and obesoprotective environments. Health Place. 2015;35:136-46.^,1010. Haddad J, Ullah S, Bell L, et al. The influence of home and school environments on children's diet and physical activity, and body mass index: a structural equation modelling approach. Matern Child Health J. 2018;22:364-75. Also, most supporting evidence comes from the US, Australia, and the UK; however, the structure of the built and the food environment differ across geographic contexts.¹¹11. Buehler R. Determinants of transport mode choice: a comparison of Germany and the USA. J Transp Geogr. 2011;9:644-57.

12. Diez J, Bilal U, Cebrecos A, et al. Understanding differences in the local food environment across countries: a case study in Madrid (Spain) and Baltimore (USA). Prev Med. 2016;89:237-44.^-1313. Gullon P, Bilal U, Sanchez P, et al. A comparative case study of walking environment in Madrid and Philadelphia using multiple sampling methods and street virtual audits. Cities Health. 2020;4:336-44. Despite the recognition that childhood obesity in Europe is highest within Mediterranean countries, no study has been conducted in this context.¹⁴14. Garrido-Miguel M, Cavero-Redondo I, Alvarez-Bueno C, et al. Prevalence and trends of overweight and obesity in European children from 1999 to 2016. JAMA Pediatr. 2019;173, e192430.^,1515. NCD Risk Factor Collaboration (NCD-RisC). Worldwide trends in body-mass index, underweight, overweight, and obesity from 1975 to 2016: a pooled analysis of 2416 population-based measurement studies in 128.9 million. Lancet. 2017;390:2627-42. This lack of evidence is hindering effective translation into policy.

The SUECO study aims to assess the relationships between residential- and school-level urban environment features and individual-level anthropometrics, eating behaviors, physical activity among young children in Madrid, Spain.

Method

Study design

This is a multilevel study conducted in Madrid, Spain. We will use de-identified individual data from the 2017 dataset of the Encuesta del estado nutricional de la población infantil de la ciudad de Madrid, a city-wide survey commissioned by the city council to assess the nutritional status of Madrid school-age children (3 to 12 years).¹⁶16. Diaz-Olalla JM, Junco I, Rodriguez-Perez M. Estudio de la situacion nutricional de la poblacion infantil en la ciudad de Madrid. [Accessed 16 november 2021]. Available at: http://www.madridsalud.es/pdfs/INFORME_ESTUDIO_NUTRICIONINFANTIL_v2.pdf.
http://www.madridsalud.es/pdfs/INFORME_E... Contextual measures will be developed based on earlier work (Fig. 1). ⁶6. Diez J, Cebrecos A, Rapela A, et al. Socioeconomic inequalities in the retail food environment around schools in a Southern European context. Nutrients. 2019;11:15.^,1717. Cereijo L, Gullon P, Cebrecos A, et al. Access to and availability of exercise facilities in Madrid: an equity perspective. Int J Health Geogr. 2019;18:15.^,1818. Gullon P, Bilal U, Cebrecos A, et al. Intersection of neighborhood dynamics and socioeconomic status in small-area walkability: the Heart Healthy Hoods project. Int J Health Geogr. 2017;16:21.

Participants and sample

Participants were selected using a stratified cluster sampling design (see fig. I in online ^AppendixAppendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi: 10.1016/j.gaceta.2021.04.005.). First, schools were stratified according to area-level factors, and 84 were randomly selected. Out of these, 60 agreed to participate (response rate: 71.4%). Second, school classes were randomly selected in each school. The study population comprised 7,740 children from which 6,157 provided parental consent to participate in the study (recruitment rate: 79.5%).

The actual number of children for each analysis will vary because individuals could consent to single components of the study (e.g., 5298 for anthropometrics) while abstaining from others (e.g., 5201 completed the survey).

Study variables

Table 1 shows an overview of the study variables. The main outcome variables encompass anthropometrics, healthy and unhealthy consumption measures, and physical activity measures.

Main outcome variables

1) Anthropometrics

Objectively measured heights and weights were used to calculate body mass index (BMI, weight in kg / height in m²) for each child. To allow for international comparisons, this study will use the WHO BMI growth reference which is an age-sex specific BMI z-score to classify overweight and obesity.¹⁹19. De Onis M, Onyango AW, Borghi E, et al. Development of a WHO growth reference for school-aged children and adolescents. Bull World Health Organ. 2007;85:660-7. We will also use as adiposity measures waist circumference and body fat percentage.

Measurements were made by trained staff using standard procedures and regularly calibrated instruments: height was measured with a stadiometer (MZ 10042) and weight with a digital weight scale (TANITA UM-076), both without shoes and with light clothing. Waist circumference was measured to the nearest 0.1cm in a standing position with the use of a measuring tape.

2) Healthy and unhealthy consumption measures

We will calculate fruit and vegetable intake based on the sum of two questions from the questionnaire. Consumption of sugar-sweetened beverages will include carbonated sugar-sweetened drinks and non-carbonated sugar-sweetened drinks. Fast-food consumption will include hamburgers and pizza. Possible answer categories were: never, 1-3 times a week, 4-6 times per week, 7 times a week. Consumption frequencies will be then summed to calculate consumption measures as frequencies per week. Eating at fast-food outlets will be calculated based on the specific question “In a normal week, how many times does your visit a fast-food outlet?”. Possible answer categories were: never, less than one time a week, and one or more times per week.

3) Physical activity

Moderate-to-vigorous physical activity levels will be estimated based on a measure of weekly moderate-to-vigorous physical activity (total amount in minutes) that includes all types of activity out of school hours. We will measure active travel to school based on the following question; “On a typical day is the main part of your child's journey to school made by walking/bicycle, bus/train/subway, car/motorcycle or other means?” Response “walking/bicycle” was coded 1 signifying active travel, and all other responses coded 0.

Main exposure variables

We will construct a geodatabase using Geographic Information Systems, which will allow us to geocode children's home and school addresses, and estimate exposure measurements (e.g., food environment, walkability, and exercise opportunities) around both home and school environment. Data will be spatially joined to census tracts to merge databases obtained at different geographic unit levels.

1) Food environment

Assessment of the retail food environment will be based on earlier work6. We will measure the density of food retailers offering unhealthy food and sugar-sweetened beverages within a 400 m shortest network path buffer around each children's residence and school address. Buffers of 800m will be considered in sensitivity analyses. Retailers' addresses will be collected (as of 2017) using a publicly available administrative dataset (Censo de locales y actividades) which was previously validated.²⁰20. Diez J, Cebrecos A, Galan I, et al. Assessing the Retail Food Environment in Madrid: an evaluation of administrative data against ground truthing. Int J Environ Res Public Health. 2019;16:3538.

2) Walkability

We will measure area-level walkability, at the census tract level, using a previously validated walkability index.18 This index comprises four indicators: residential density, population density, retail destinations, and street connectivity. Data will be obtained from different secondary sources.

3) Surrounding physical activity opportunities

We will use three measures: 1) density of surrounding playgrounds; 2) density of green spaces,²¹21. Plans E, Gullon P, Cebrecos A, et al. Density of green spaces and cardiovascular risk factors in the city of Madrid: the Heart Healthy Hoods Study. Int J Environ Res Public Health. 2019;16:4918. and 3) density of exercise facilities (including both free outdoor sports courts and public sports centers). We obtained green spaces data combining all green land use cover categories from the City Council of Madrid for the year 2016 and all green spaces under mainteinance by the Madrid City Council in 2020.

Potential confounders

We will identify potential confounders for each of the environmental exposures and include individual sociodemographic variables and area-level characteristics: age (years), sex (male or female), country of origin, household composition, Family Affluence Score (FAS)²²22. Currie C, Molcho M, Boyce W, et al. Researching health inequalities in adolescents: the development of the Health Behaviour in School-Aged Children (HBSC) Family Affluence Scale. Soc Sci Med. 2008;66:1429-36. as a measure of the family's socioeconomic status, area-level socioeconomic status,18 and population density.

Statistical analysis

The study has a multilevel structure, where children are nested within schools (Fig. 1). Therefore, multilevel regression models will be used to examine the cross-sectional associations between each outcome measure and each areal level factor. All regression models will be adjusted for a common set of potential confounders and checked for effect modification (e.g., by sex to obtain sex-specific estimates). Effect modification will be tested by including interaction terms in the models. We will design sensitivity analyses to test the robustness of our findings to different model specifications.

Ethical considerations

The study will be conducted according to the guidelines laid down by the Declaration of Helsinki and ethical approval was granted by the Ethics Committee of the Universidad de Alcalá (CEI/HU/2019/35). Participants were assured of anonymity and confidentiality and written consent was obtained from the parents or legal guardians of all children. All data exchanges will adhere to the most up-to-date EU and national data protection regulations.

Discussion

To the best of our knowledge, no research projects on this topic have been conducted or are in progress in Spain. Thus, it will provide significant evidence. First, its multi-level study design will allow for assessing possible area-level effects over and above individual-level effects. Second, it focuses on school-age children which is a current priority given the magnitude of excess body weight, physical inactivity, and sedentarism among this population group. Third, it will measure the obesogenic environment across several domains and locations (both residence and school location).¹⁰10. Haddad J, Ullah S, Bell L, et al. The influence of home and school environments on children's diet and physical activity, and body mass index: a structural equation modelling approach. Matern Child Health J. 2018;22:364-75. Fourth, it will assess the impact of individual- and area-level SES, which are potentially important covariates of eating habits and physical activity and potential effect modifiers of the associations between environmental determinants and eating habits or physical activity.

This study presents several methodological limitations. First, eating and physical activity data are based on self-report, which is prone to bias and measurement error (e.g., parents may be reluctant to disclose the frequency at which their children consume sugar-sweetened beverages). Second, it may be the case that fast-food retailers preferentially locate in areas with greater demand resulting in a causal pathway that is in the opposite direction to that hypothesized.

Conclusions

The study offers a unique opportunity to evaluate the health consequences of obesogenic environments for school-age children in a large Southern European city like Madrid. This study will also provide relevant evidence to influence urban policies to promote children's well-being while addressing social inequities.

Acknowledgments

We thank the schools, families, and children for their participation in the Encuesta del estado nutricional de la población infantil de la ciudad de Madrid. Also, to Madrid Salud for allowing Alcalá University use of the city council data.

References

Publication Dates

History