The neighourhood obesogenic built environment characteristics (OBCT) index: Practice versus theory

Thao Minh Lam; Nicolette R. den Braver; Haykanush Ohanyan; Alfred J. Wagtendonk; Ilonca Vaartjes; Joline WJ Beulens; Jeroen Lakerveld

doi:10.1016/j.envres.2024.118625

The neighourhood obesogenic built environment characteristics (OBCT) index: Practice versus theory

Thao Minh Lam^*, Nicolette R. den Braver, Haykanush Ohanyan, Alfred J. Wagtendonk, Ilonca Vaartjes, Joline WJ Beulens, Jeroen Lakerveld

^*Corresponding author for this work

Research output: Contribution to journal › Article › Academic › peer-review

Abstract

Background: Obesity is a key risk factor for major chronic diseases such as type 2 diabetes and cardiovascular diseases. To extensively characterise the obesogenic built environment, we recently developed a novel Obesogenic Built environment CharacterisTics (OBCT) index, consisting of 17 components that capture both food and physical activity (PA) environments. Objectives: We aimed to assess the association between the OBCT index and body mass index (BMI) in a nationwide health monitor. Furthermore, we explored possible ways to improve the index using unsupervised and supervised methods. Methods: The OBCT index was constructed for 12,821 Dutch administrative neighbourhoods and linked to residential addresses of eligible adult participants in the 2016 Public Health Monitor. We split the data randomly into a training (two-thirds; n = 255,187) and a testing subset (one-third; n = 127,428). In the training set, we used non-parametric restricted cubic regression spline to assess index's association with BMI, adjusted for individual demographic characteristics. Effect modification by age, sex, socioeconomic status (SES) and urbanicity was examined. As improvement, we (1) adjusted the food environment for address density, (2) added housing price to the index and (3) adopted three weighting strategies, two methods were supervised by BMI (variable selection and random forest) in the training set. We compared these methods in the testing set by examining their model fit with BMI as outcome. Results: The OBCT index had a significant non-linear association with BMI in a fully-adjusted model (p<0.05), which was modified by age, sex, SES and urbanicity. However, variance in BMI explained by the index was low (<0.05%). Supervised methods increased this explained variance more than non-supervised methods, though overall improvements were limited as highest explained variance remained <0.5%. Discussion: The index, despite its potential to highlight disparity in obesogenic environments, had limited association with BMI. Complex improvements are not necessarily beneficial, and the components should be re-operationalised.

Original language	English
Article number	118625
Number of pages	13
Journal	Environmental Research
Volume	251
DOIs	https://doi.org/10.1016/j.envres.2024.118625
Publication status	Published - 15 Jun 2024

Bibliographical note

Publisher Copyright:
© 2024 The Authors

Funding

The results presented here were based on calculations by Lam et al./AmsterdamUMC using non-public microdata from Statistics Netherlands. Under certain conditions, these microdata are accessible for statistical and scientific research. For further information: [email protected]. Geo-data were collected as part of the Geoscience and Health Cohort Consortium (GECCO), which was financially supported by the Netherlands Organisation for Scientific Research (NWO), the Netherlands Organisation for Health Research and Development (ZonMw), and Amsterdam UMC. More information on GECCO can be found on www.gecco.nl. This work is supported by EXPOSOME-NL and OBCT consortia. The EXPOSOME-NL consortium is funded through the Gravitation program of the Dutch Ministry of Education, Culture, and Science and the Netherlands Organization for Scientific Research (NWO grant number 024.004.017). The OBCT consortium is funded by the European Union's Horizon Europe research and innovation programme under grant agreement No. 101080250. Views and opinions expressed are however those of the author(s) only and do not necessarily reflect those of the European Union or the European Health and Digital Executive Agency (HADEA). Neither the European Union nor the granting authority can be held responsible for them. We would like to express our gratitude to Prof. Derek Karssenberg for his useful comments on the analysis plan.

Funders	Funder number
EXPOSOME-NL
European Health and Digital Executive Agency
ZonMw
Dutch Ministry of Education, Culture, and Science
Amsterdam University Medical Center
Nederlandse Organisatie voor Wetenschappelijk Onderzoek	024.004.017
European Union's Horizon Europe research and innovation programme	101080250

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1016/j.envres.2024.118625Licence: CC BY

1-s2.0-S0013935124005292-mainFinal published version, 2.66 MBLicence: CC BY

Cite this

@article{ea5b917d93f348e58e0dbc492a901ad8,

title = "The neighourhood obesogenic built environment characteristics (OBCT) index: Practice versus theory",

abstract = "Background: Obesity is a key risk factor for major chronic diseases such as type 2 diabetes and cardiovascular diseases. To extensively characterise the obesogenic built environment, we recently developed a novel Obesogenic Built environment CharacterisTics (OBCT) index, consisting of 17 components that capture both food and physical activity (PA) environments. Objectives: We aimed to assess the association between the OBCT index and body mass index (BMI) in a nationwide health monitor. Furthermore, we explored possible ways to improve the index using unsupervised and supervised methods. Methods: The OBCT index was constructed for 12,821 Dutch administrative neighbourhoods and linked to residential addresses of eligible adult participants in the 2016 Public Health Monitor. We split the data randomly into a training (two-thirds; n = 255,187) and a testing subset (one-third; n = 127,428). In the training set, we used non-parametric restricted cubic regression spline to assess index's association with BMI, adjusted for individual demographic characteristics. Effect modification by age, sex, socioeconomic status (SES) and urbanicity was examined. As improvement, we (1) adjusted the food environment for address density, (2) added housing price to the index and (3) adopted three weighting strategies, two methods were supervised by BMI (variable selection and random forest) in the training set. We compared these methods in the testing set by examining their model fit with BMI as outcome. Results: The OBCT index had a significant non-linear association with BMI in a fully-adjusted model (p<0.05), which was modified by age, sex, SES and urbanicity. However, variance in BMI explained by the index was low (<0.05%). Supervised methods increased this explained variance more than non-supervised methods, though overall improvements were limited as highest explained variance remained <0.5%. Discussion: The index, despite its potential to highlight disparity in obesogenic environments, had limited association with BMI. Complex improvements are not necessarily beneficial, and the components should be re-operationalised.",

author = "Lam, {Thao Minh} and {den Braver}, {Nicolette R.} and Haykanush Ohanyan and Wagtendonk, {Alfred J.} and Ilonca Vaartjes and Beulens, {Joline WJ} and Jeroen Lakerveld",

note = "Publisher Copyright: {\textcopyright} 2024 The Authors",

year = "2024",

month = jun,

day = "15",

doi = "10.1016/j.envres.2024.118625",

language = "English",

volume = "251",

journal = "Environmental Research",

issn = "0013-9351",

publisher = "Academic Press Inc.",

}

TY - JOUR

T1 - The neighourhood obesogenic built environment characteristics (OBCT) index

T2 - Practice versus theory

AU - Lam, Thao Minh

AU - den Braver, Nicolette R.

AU - Ohanyan, Haykanush

AU - Wagtendonk, Alfred J.

AU - Vaartjes, Ilonca

AU - Beulens, Joline WJ

AU - Lakerveld, Jeroen

PY - 2024/6/15

Y1 - 2024/6/15

N2 - Background: Obesity is a key risk factor for major chronic diseases such as type 2 diabetes and cardiovascular diseases. To extensively characterise the obesogenic built environment, we recently developed a novel Obesogenic Built environment CharacterisTics (OBCT) index, consisting of 17 components that capture both food and physical activity (PA) environments. Objectives: We aimed to assess the association between the OBCT index and body mass index (BMI) in a nationwide health monitor. Furthermore, we explored possible ways to improve the index using unsupervised and supervised methods. Methods: The OBCT index was constructed for 12,821 Dutch administrative neighbourhoods and linked to residential addresses of eligible adult participants in the 2016 Public Health Monitor. We split the data randomly into a training (two-thirds; n = 255,187) and a testing subset (one-third; n = 127,428). In the training set, we used non-parametric restricted cubic regression spline to assess index's association with BMI, adjusted for individual demographic characteristics. Effect modification by age, sex, socioeconomic status (SES) and urbanicity was examined. As improvement, we (1) adjusted the food environment for address density, (2) added housing price to the index and (3) adopted three weighting strategies, two methods were supervised by BMI (variable selection and random forest) in the training set. We compared these methods in the testing set by examining their model fit with BMI as outcome. Results: The OBCT index had a significant non-linear association with BMI in a fully-adjusted model (p<0.05), which was modified by age, sex, SES and urbanicity. However, variance in BMI explained by the index was low (<0.05%). Supervised methods increased this explained variance more than non-supervised methods, though overall improvements were limited as highest explained variance remained <0.5%. Discussion: The index, despite its potential to highlight disparity in obesogenic environments, had limited association with BMI. Complex improvements are not necessarily beneficial, and the components should be re-operationalised.

AB - Background: Obesity is a key risk factor for major chronic diseases such as type 2 diabetes and cardiovascular diseases. To extensively characterise the obesogenic built environment, we recently developed a novel Obesogenic Built environment CharacterisTics (OBCT) index, consisting of 17 components that capture both food and physical activity (PA) environments. Objectives: We aimed to assess the association between the OBCT index and body mass index (BMI) in a nationwide health monitor. Furthermore, we explored possible ways to improve the index using unsupervised and supervised methods. Methods: The OBCT index was constructed for 12,821 Dutch administrative neighbourhoods and linked to residential addresses of eligible adult participants in the 2016 Public Health Monitor. We split the data randomly into a training (two-thirds; n = 255,187) and a testing subset (one-third; n = 127,428). In the training set, we used non-parametric restricted cubic regression spline to assess index's association with BMI, adjusted for individual demographic characteristics. Effect modification by age, sex, socioeconomic status (SES) and urbanicity was examined. As improvement, we (1) adjusted the food environment for address density, (2) added housing price to the index and (3) adopted three weighting strategies, two methods were supervised by BMI (variable selection and random forest) in the training set. We compared these methods in the testing set by examining their model fit with BMI as outcome. Results: The OBCT index had a significant non-linear association with BMI in a fully-adjusted model (p<0.05), which was modified by age, sex, SES and urbanicity. However, variance in BMI explained by the index was low (<0.05%). Supervised methods increased this explained variance more than non-supervised methods, though overall improvements were limited as highest explained variance remained <0.5%. Discussion: The index, despite its potential to highlight disparity in obesogenic environments, had limited association with BMI. Complex improvements are not necessarily beneficial, and the components should be re-operationalised.

UR - http://www.scopus.com/inward/record.url?scp=85188503416&partnerID=8YFLogxK

U2 - 10.1016/j.envres.2024.118625

DO - 10.1016/j.envres.2024.118625

M3 - Article

C2 - 38467360

AN - SCOPUS:85188503416

SN - 0013-9351

VL - 251

JO - Environmental Research

JF - Environmental Research

M1 - 118625

ER -

The neighourhood obesogenic built environment characteristics (OBCT) index: Practice versus theory

Abstract

Bibliographical note

Funding

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this