Exposome-Wide Association Study of Body Mass Index Using a Novel Meta-Analytical Approach for Random Forest Models.

BACKGROUND: Overweight and obesity impose a considerable individual and social burden, and the urban environments might encompass factors that contribute to obesity. Nevertheless, there is a scarcity of research that takes into account the simultaneous interaction of multiple environmental factors. OBJECTIVES: Our objective was to perform an exposome-wide association study of body mass index (BMI) in a multicohort setting of 15 studies. METHODS: Studies were affiliated with the Dutch Geoscience and Health Cohort Consortium (GECCO), had different population sizes (688-141,825), and covered the entire Netherlands. Ten studies contained general population samples, others focused on specific populations including people with diabetes or impaired hearing. BMI was calculated from self-reported or measured height and weight. Associations with 69 residential neighborhood environmental factors (air pollution, noise, temperature, neighborhood socioeconomic and demographic factors, food environment, drivability, and walkability) were explored. Random forest (RF) regression addressed potential nonlinear and nonadditive associations. In the absence of formal methods for multimodel inference for RF, a rank aggregation-based meta-analytic strategy was used to summarize the results across the studies. RESULTS: Six exposures were associated with BMI: five indicating neighborhood economic or social environments (average home values, percentage of high-income residents, average income, livability score, share of single residents) and one indicating the physical activity environment (walkability in 5-km buffer area). Living in high-income neighborhoods and neighborhoods with higher livability scores was associated with lower BMI. Nonlinear associations were observed with neighborhood home values in all studies. Lower neighborhood home values were associated with higher BMI scores but only for values up to €300,000. The directions of associations were less consistent for walkability and share of single residents. DISCUSSION: Rank aggregation made it possible to flexibly combine the results from various studies, although between-study heterogeneity could not be estimated quantitatively based on RF models. Neighborhood social, economic, and physical environments had the strongest associations with BMI. https://doi.org/10.1289/EHP13393.

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

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.

Ethnic and socioeconomic inequalities in air pollution exposure: a cross-sectional analysis of nationwide individual-level data from the Netherlands.

BACKGROUND: Air pollution contributes to a large disease burden and some populations are disproportionately exposed. We aimed to evaluate ethnic and socioeconomic differences in exposure to air pollution in the Netherlands. METHODS: We did a nationwide, cross-sectional analysis of all residents of the Netherlands on Jan 1, 2019. Sociodemographic information was centralised by Statistics Netherlands and mainly originated from the National Population Register, the tax register, and education registers. Concentrations of NO2, PM2·5, PM10, and elemental carbon, modelled by the National Institute for Public Health and the Environment, were linked to the individual-level demographic data. We assessed differences in air pollution exposures across the 40 largest minority ethnic groups. Evaluation of how ethnicity intersected with socioeconomic position in relation to exposures was done for the ten largest ethnic groups, plus Chinese and Indian groups, in both urban and rural areas using multivariable linear regression analyses. FINDINGS: The total study population consisted of 17 251 511 individuals. Minority ethnic groups were consistently exposed to higher levels of air pollution than the ethnic Dutch population. The magnitude of inequalities varied between the minority ethnic groups, with 3-44% higher exposures to NO2 and 1-9% higher exposures to PM2·5 compared with the ethnic Dutch group. Average exposures were highest for the lowest socioeconomic group. Ethnic inequalities in exposure remained after adjustment for socioeconomic position and were of similar magnitude in urban and rural areas. INTERPRETATION: The variability in air pollution exposure across ethnic and socioeconomic subgroups in the Netherlands indicates environmental injustice at the intersection of social characteristics. The health consequences of the observed inequalities and the underlying processes driving them warrant further investigation. FUNDING: The Gravitation programme of the Dutch Ministry of Education, Culture, and Science, the Netherlands Organization for Scientific Research, the Netherlands Organisation for Health Research and Development, and Amsterdam University Medical Center.

Walking to work: The role of walkability around the workplace in a Dutch adult commuting population.

Current evidence on neighborhood walkability and active commuting focuses on residential rather than workplace environment. This cross-sectional study investigated whether higher workplace walkability (WW) was associated with commute walking, both independently and together with residential walkability, using data from 6769 respondents of the 2017 Dutch national travel survey. In a fully adjusted logistic regression model, 10% increase in WW was associated with 32% higher odds of commute walking (Odds ratio (OR): 1.31, 95% Confidence Interval (CI: 1.27-1.36). The estimates were stronger in rural dwellers than urban residents, (ORrural 1.49, 95%CI: 1.34-1.64 vs ORhighly.urban 1.19, 95%CI: 1.13-1.26). In participants with both high residential walkability and WW, we observed 215% higher odds (OR 3.15, 95% CI: 2.48-3.99) of commute walking compared to those with low walkability in both. Our study indicated the importance and complementary nature of walkable residence and workplace in contribution to physical activity of working individuals through active commuting.

A positive neighborhood walkability is associated with a higher magnitude of leisure walking in adults upon COVID-19 restrictions: a longitudinal cohort study.

BACKGROUND: Previous cross-sectional and longitudinal observational studies revealed positive relationships between contextual built environment components and walking behavior. Due to severe restrictions during COVID-19 pandemic lockdowns, physical activity was primarily performed within the immediate living area. Using this unique opportunity, we evaluated whether built environment components were associated with the magnitude of change in walking activity in adults during COVID-19 restrictions. METHODS: Data on self-reported demographic characteristics and walking behaviour were extracted from the prospective longitudinal Lifelines Cohort Study in the Netherlands of participants ≥ 18 years. For our analyses, we made use of the data acquired between 2014-2017 (n = 100,285). A fifth of the participants completed the questionnaires during COVID-19 restrictive policies in July 2021 (n = 20,806). Seven spatial components were calculated for a 500m and 1650m Euclidean buffer per postal code area in GIS: population density, retail and service destination density, land use mix, street connectivity, green space density, sidewalk density, and public transport stops. Additionally, the walkability index (WI) of these seven components was calculated. Using multivariable linear regression analyses, we analyzed the association between the WI (and separate components) and the change in leisure walking minutes/week. Included demographic variables were age, gender, BMI, education, net income, occupation status, household composition and the season in which the questionnaire was filled in. RESULTS: The average leisure walking time strongly increased by 127 min/week upon COVID-19 restrictions. All seven spatial components of the WI were significantly associated with an increase in leisure walking time; a 10% higher score in the individual spatial component was associated with 5 to 8 more minutes of leisure walking/week. Green space density at the 500m Euclidean buffer and side-walk density at the 1650m Euclidean buffer were associated with the highest increase in leisure walking time/week. Subgroup analysis revealed that the built environment showed its strongest impact on leisure walking time in participants not engaging in leisure walking before the COVID-19 pandemic, compared to participants who already engaged in leisure walking before the COVID-19 pandemic. CONCLUSIONS: These results provide strong evidence that the built environment, corrected for individual-level characteristics, directly links to changes observed in leisure walking time during COVID-19 restrictions. Since this relation was strongest in those who did not engage in leisure walking before the COVID-19 pandemic, our results encourage new perspectives in health promotion and urban planning.

Development of a neighborhood obesogenic built environment characteristics index for the Netherlands.

OBJECTIVE: Environmental factors that drive obesity are often studied individually, whereas obesogenic environments are likely to consist of multiple factors from food and physical activity (PA) environments. This study aimed to compose and describe a comprehensive, theory-based, expert-informed index to quantify obesogenicity for all neighborhoods in the Netherlands. METHODS: The Obesogenic Built Environment CharacterisTics (OBCT) index consists of 17 components. The index was calculated as an average of componential scores across both food and PA environments and was scaled from 0 to 100. The index was visualized and summarized with sensitivity analysis for weighting methods. RESULTS: The OBCT index for all 12,821 neighborhoods was right-skewed, with a median of 44.6 (IQR = 10.1). Obesogenicity was lower in more urbanized neighborhoods except for the extremely urbanized neighborhoods (>2500 addresses/km2 ), where obesogenicity was highest. The overall OBCT index score was moderately correlated with the food environment (Spearman ρ = 0.55, p <0.05) and with the PA environment (ρ = 0.38, p <0.05). Hierarchical weighting increased index correlations with the PA environment but decreased correlations with the food environment. CONCLUSIONS: The novel OBCT index and its comprehensive environmental scores are potentially useful tools to quantify obesogenicity of neighborhoods.

Development of an objectively measured walkability index for the Netherlands.

BACKGROUND: Walkability indices have been developed and linked to behavioural and health outcomes elsewhere in the world, but not comprehensively for Europe. We aimed to 1) develop a theory-based and evidence-informed Dutch walkability index, 2) examine its cross-sectional associations with total and purpose-specific walking behaviours of adults across socioeconomic (SES) and urbanisation strata, 3) explore which walkability components drive these associations. METHODS: Components of the index included: population density, retail and service density, land use mix, street connectivity, green space, sidewalk density and public transport density. Each of the seven components was calculated for three Euclidean buffers: 150 m, 500 m and 1000 m around every 6-digit postal code location and for every administrative neighbourhood in GIS. Componential z-scores were averaged, and final indices normalized between 0 and 100. Data on self-reported demographic characteristics and walking behaviours of 16,055 adult respondents (aged 18-65) were extracted from the Dutch National Travel Survey 2017. Using Tobit regression modelling adjusted for individual- and household-level confounders, we assessed the associations between walkability and minutes walking in total, for non-discretionary and discretionary purposes. By assessing the attenuation in associations between partial indices and walking outcomes, we identified which of the seven components drive these associations. We also tested for effect modification by urbanization degree, SES, age and sex. RESULTS: In fully adjusted models, a 10% increase in walkability was associated with a maximum increase of 8.5 min of total walking per day (95%CI: 7.1-9.9). This association was consistent across buffer sizes and purposes of walking. Public transport density was driving the index's association with walking outcomes. Stratified results showed that associations with minutes of non-discretionary walking were stronger in rural compared to very urban areas, in neighbourhoods with low SES compared to high SES, and in middle-aged (36-49 years) compared to young (18-35 years old) and older adults (50-65 years old). CONCLUSIONS: The walkability index was cross-sectionally associated with Dutch adult's walking behaviours, indicating its validity for further use in research.

Associations of changes in neighbourhood walkability with changes in walking activity in older adults: a fixed effects analysis.

BACKGROUND: Supporting older adults to engage in physically active lifestyles requires supporting environments. Walkable environments may increase walking activity in older adults, but evidence for this subgroup is scarce, and longitudinal studies are lacking. This study therefore examined whether changes in neighbourhood walkability were associated with changes in walking activity in older adults, and whether this association differed by individual-level characteristics and by contextual conditions beyond the built environment. METHODS: Data from 668 participants (57.8-93.4 years at baseline) across three waves (2005/06, 2008/09 and 2011/12) of the Longitudinal Aging Study Amsterdam (LASA) were used. These individuals did not relocate during follow-up. Self-reported outdoor walking activity in minutes per week was assessed using the LASA Physical Activity Questionnaire. Composite exposure measures of neighbourhood walkability (range: 0 (low)-100 (high)) within 500-m Euclidean buffer zones around each participant's residential address were constructed by combining objectively measured high-resolution Geographic Information System data on population density, retail and service destination density, land use mix, street connectivity, green space density, and sidewalk density. Fixed effects linear regression analyses were applied, adjusted for relevant time-varying confounders. RESULTS: Changes in neighbourhood walkability were not statistically significantly associated with changes in walking activity in older adults (ß500m = - 0.99, 95% CI = -6.17-4.20). The association of changes in neighbourhood walkability with changes in walking activity did not differ by any of the individual-level characteristics (i.e., age, sex, educational level, cognitive impairment, mobility disability, and season) and area-level characteristics (i.e., road traffic noise, air pollution, and socioeconomic status). CONCLUSIONS: This study did not show evidence for an association between changes in neighbourhood walkability and changes in walking activity in older adults. If neighbourhood walkability and walking activity are causally linked, then changes in neighbourhood walkability between 2005/06 and 2011/12 might have been not substantial enough to produce meaningful changes in walking activity in older adults.

Deep phenotyping meets big data: the Geoscience and hEalth Cohort COnsortium (GECCO) data to enable exposome studies in The Netherlands.

Environmental exposures are increasingly investigated as possible drivers of health behaviours and disease outcomes. So-called exposome studies that aim to identify and better understand the effects of exposures on behaviours and disease risk across the life course require high-quality environmental exposure data. The Netherlands has a great variety of environmental data available, including high spatial and often temporal resolution information on urban infrastructure, physico-chemical exposures, presence and availability of community services, and others. Until recently, these environmental data were scattered and measured at varying spatial scales, impeding linkage to individual-level (cohort) data as they were not operationalised as personal exposures, that is, the exposure to a certain environmental characteristic specific for a person. Within the Geoscience and hEalth Cohort COnsortium (GECCO) and with support of the Global Geo Health Data Center (GGHDC), a platform has been set up in The Netherlands where environmental variables are centralised, operationalised as personal exposures, and used to enrich 23 cohort studies and provided to researchers upon request. We here present and detail a series of personal exposure data sets that are available within GECCO to date, covering personal exposures of all residents of The Netherlands (currently about 17 M) over the full land surface of the country, and discuss challenges and opportunities for its use now and in the near future.

The association of the neighbourhood built environment with objectively measured physical activity in older adults with and without lower limb osteoarthritis.

BACKGROUND: This study examined the associations of objectively measured neighbourhood built environment characteristics with objectively measured physical activity (PA) in older people with and without lower limb osteoarthritis (LLOA), and assessed whether these relationships differ between both groups. METHODS: Data from the Dutch component of the European Project on OSteoArthritis were used. American College of Rheumatology classification criteria were used to diagnose LLOA (knee and/or hip osteoarthritis). Daily average time spent on total PA and separate PA intensity categories, including light PA, low-light PA, high-light PA, and moderate to vigorous PA, were measured using Actigraph GT3X accelerometers. Geographic Information Systems were used to measure street connectivity (number of street connections per km(2)) and distances (in km) to resources (health care resources, retail resources, meeting places, and public transport) within neighbourhoods. Multiple Linear Regression Analyses were used to examine the associations between measures of the neighbourhood built environment and PA, adjusted for several confounders. RESULTS: Of all 247 participants (66-85 years), 41 (16.6 %) had LLOA. The time spent on any PA did not differ significantly between participants with and without LLOA (LLOA: Mean = 268.3, SD = 83.3 versus non-LLOA: Mean = 275.8, SD = 81.2; p = 0.59). In the full sample, no measures of the neighbourhood built environment were statistically significantly associated with total PA. Larger distances to specific health care resources (general practice and physiotherapist) and retail resources (supermarket) were associated with more time spent on PA in older people with LLOA than in those without LLOA. In particular, the associations of light and high-light PA with distances to these specific resources were stronger in participants with LLOA compared to their counterparts without LLOA. CONCLUSIONS: Specific attributes of the neighbourhood built environment are more strongly associated with PA in older people with LLOA than in those without LLOA. Knowledge on the relationship between objectively measured neighbourhood characteristics and PA in older people with and without LLOA could be used to inform policymakers and city planners about adaptation of neighbourhoods and their infrastructures to appropriately facilitate PA in healthy and functionally impaired older adults.