Machine learning approaches to characterize the obesogenic urban exposome

Association between trajectories of the neighborhood social exposome and mental health in late adolescence: A FinnTwin12 cohort study

BACKGROUND

Adolescent mental health problems impose a significant burden. Exploring evolving social environments could enhance comprehension of their impact on mental health. We aimed to depict the trajectories of the neighborhood social exposome from middle to late adolescence and assess the intricate relationship between them and late adolescent mental health.

METHODS

Participants (n = 3965) from the FinnTwin12 cohort with completed questionnaires at age 17 were used. Nine mental health measures were assessed. The social exposome comprised 28 neighborhood social indicators. Trajectories of these indicators from ages 12 to 17 were summarized via latent growth curve modeling into growth factors, including baseline intercept. Mixture effects of all growth factors were assessed through quantile-based g-computation. Repeated generalized linear regressions identified significant growth factors. Sex stratification was performed.

RESULTS

The linear-quadratic model was the most optimal trajectory model. No mixture effect was detected. Regression models showed some growth factors saliently linked to the p-factor, internalizing problems, anxiety, hyperactivity, and aggression. The majority of them were baseline intercepts. Quadratic growth factors about mother tongues correlated with anxiety among sex-combined participants and males. The linear growth factor in the proportion of households of couples without children was associated with internalizing problems in females.

LIMITATIONS

We were limited to including only neighborhood-level social exposures, and the multilevel contextual exposome situation interfered with our assessment.

CONCLUSIONS

Trajectories of the social neighborhood exposome modestly influenced late adolescent mental health. Tackling root causes of social inequalities through targeted programs for living conditions could improve adolescent mental health.

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.

Greenness, air pollution, and temperature exposure effects in predicting premature mortality and morbidity: A small-area study using spatial random forest model

BACKGROUND

Although studies have provided negative impacts of air pollution, heat or cold exposure on mortality and morbidity, and positive effects of increased greenness on reducing them, a few studies have focused on exploring combined and synergetic effects of these exposures in predicting these health outcomes, and most had ignored the spatial autocorrelation in analyzing their health effects. This study aims to investigate the health effects of air pollution, greenness, and temperature exposure on premature mortality and morbidity within a spatial machine-learning modeling framework.

METHODS

Years of potential life lost reflecting premature mortality and comparative illness and disability ratio reflecting chronic morbidity from 1673 small areas covering Greater Manchester for the year 2008-2013 obtained. Average annual levels of NO2 concentration, normalized difference vegetation index (NDVI) representing greenness, and annual average air temperature were utilized to assess exposure in each area. These exposures were linked to health outcomes using non-spatial and spatial random forest (RF) models while accounting for spatial autocorrelation.

RESULTS

Spatial-RF models provided the best predictive accuracy when accounted for spatial autocorrelation. Among the exposures considered, air pollution emerged as the most influential in predicting mortality and morbidity, followed by NDVI and temperature exposure. Nonlinear exposure-response relations were observed, and interactions between exposures illustrated specific ranges or sweet and sour spots of exposure thresholds where combined effects either exacerbate or moderate health conditions.

CONCLUSION

Air pollution exposure had a greater negative impact on health compared to greenness and temperature exposure. Combined exposure effects may indicate the highest influence of premature mortality and morbidity burden.

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.

Association of the external environmental exposome and obesity: A comprehensive nationwide study in 2019 among Chinese children and adolescents

OBJECTIVE

Children and adolescents are particularly vulnerable to the effects of various environmental factors, which could disrupt growth processes and potentially lead to obesity. Currently, comprehensive and systematic assessments of these environmental exposures during developmental periods are lacking. Therefore, this study aims to evaluate the association between external environmental exposures and the incidence of obesity in children and adolescents.

METHODS

Data was collected from the 2019 Chinese National Survey on Students' Constitution and Health, including 214,659 Han children aged 7 to 19. Body Mass Index (BMI) and BMI-for-age z-score (zBMI) were the metrics used to assess overweight and obesity prevalence. The study assessed 18 environmental factors, including air pollutants, natural space, land cover, meteorological conditions, built environment, road conditions, and artificial light at night. Exposome-wide association study (ExWAS) to analyze individual exposures' associations with health outcomes, and Weighted Quantile Sum (WQS) to assess cumulative exposure effects.

RESULTS

Among the children and adolescents, there were 24.2 % participants classified as overweight or obesity. Notably, 17 out of 18 environmental factors exhibited significant associations with zBMI and overweight/obesity. Seven air pollutants, road conditions, and built density were positively correlated with higher zBMI and obesity risk, while NDVI, forests, and meteorological factors showed negative correlations. Co-exposure analysis highlighted that SO2, ALAN, PM10, and trunk road density significantly increased zBMI, whereas rainfall, grassland, and forest exposure reduced it. Theoretically reduction in the number and prevalence of cases was calculated, indicating potential reductions in prevalence of up to 4.51 % for positive exposures and 5.09 % for negative exposures. Notably, substantial reductions were observed in regions with high pollution levels.

CONCLUSION

This large-scale investigation, encompassing various environmental exposures in schools, highlights the significant impact of air pollution, road characteristics, rainfall, and forest coverage on childhood obesity.

A prospective exploration of the urban exposome in relation to headache in the Dutch population-based Occupational and environmental health cohort study (AMIGO)

OBJECTIVE

Headache is one of the most prevalent and disabling health conditions globally. We prospectively explored the urban exposome in relation to weekly occurrence of headache episodes using data from the Dutch population-based Occupational and Environmental Health Cohort Study (AMIGO).

MATERIAL AND METHODS

Participants (N = 7,339) completed baseline and follow-up questionnaires in 2011 and 2015, reporting headache frequency. Information on the urban exposome covered 80 exposures across 10 domains, such as air pollution, electromagnetic fields, and lifestyle and socio-demographic characteristics. We first identified all relevant exposures using the Boruta algorithm and then, for each exposure separately, we estimated the average treatment effect (ATE) and related standard error (SE) by training causal forests adjusted for age, depression diagnosis, painkiller use, general health indicator, sleep disturbance index and weekly occurrence of headache episodes at baseline.

RESULTS

Occurrence of weekly headache was 12.5 % at baseline and 11.1 % at follow-up. Boruta selected five air pollutants (NO2, NOX, PM10, silicon in PM10, iron in PM2.5) and one urban temperature measure (heat island effect) as factors contributing to the occurrence of weekly headache episodes at follow-up. The estimated causal effect of each exposure on weekly headache indicated positive associations. NO2 showed the largest effect (ATE = 0.007 per interquartile range (IQR) increase; SE = 0.004), followed by PM10 (ATE = 0.006 per IQR increase; SE = 0.004), heat island effect (ATE = 0.006 per one-degree Celsius increase; SE = 0.007), NOx (ATE = 0.004 per IQR increase; SE = 0.004), iron in PM2.5 (ATE = 0.003 per IQR increase; SE = 0.004), and silicon in PM10 (ATE = 0.003 per IQR increase; SE = 0.004).

CONCLUSION

Our results suggested that exposure to air pollution and heat island effects contributed to the reporting of weekly headache episodes in the study population.

Socioexposomics of COVID-19 across New Jersey: a comparison of geostatistical and machine learning approaches

BACKGROUND

Disparities in adverse COVID-19 health outcomes have been associated with multiple social and environmental stressors. However, research is needed to evaluate the consistency and efficiency of methods for studying these associations at local scales.

OBJECTIVE

To assess socioexposomic associations with COVID-19 outcomes across New Jersey and evaluate consistency of findings from multiple modeling approaches.

METHODS

We retrieved data for COVID-19 cases and deaths for the 565 municipalities of New Jersey up to the end of the first phase of the pandemic, and calculated mortality rates with and without long-term-care (LTC) facility deaths. We considered 84 spatially heterogeneous environmental, demographic and socioeconomic factors from publicly available databases, including air pollution, proximity to industrial sites/facilities, transportation-related noise, occupation and commuting, neighborhood and housing characteristics, age structure, racial/ethnic composition, poverty, etc. Six geostatistical models (Poisson/Negative-Binomial regression, Poison/Negative-Binomial mixed effect model, Poisson/Negative-Binomial Bersag-York-Mollie spatial model) and two Machine Learning (ML) methods (Random Forest, Extreme Gradient Boosting) were implemented to assess association patterns. The Shapley effects plot was established for explainable ML and change of support validation was introduced to compare performances of different approaches.

RESULTS

We found robust positive associations of COVID-19 mortality with historic exposures to NO2, population density, percentage of minority and below high school education, and other social and environmental factors. Exclusion of LTC deaths does not significantly affect correlations for most factors but findings can be substantially influenced by model structures and assumptions. The best performing geostatistical models involved flexible structures representing data variations. ML methods captured association patterns consistent with the best performing geostatistical models, and furthermore detected consistent nonlinear associations not captured by geostatistical models.

SIGNIFICANCE

The findings of this work improve the understanding of how social and environmental disparities impacted COVID-19 outcomes across New Jersey.

Seminar: Scalable Preprocessing Tools for Exposomic Data Analysis

BACKGROUND

The exposome serves as a popular framework in which to study exposures from chemical and nonchemical stressors across the life course and the differing roles that these exposures can play in human health. As a result, data relevant to the exposome have been used as a resource in the quest to untangle complicated health trajectories and help connect the dots from exposures to adverse outcome pathways.

OBJECTIVES

The primary aim of this methods seminar is to clarify and review preprocessing techniques critical for accurate and effective external exposomic data analysis. Scalability is emphasized through an application of highly innovative combinatorial techniques coupled with more traditional statistical strategies. The Public Health Exposome is used as an archetypical model. The novelty and innovation of this seminar's focus stem from its methodical, comprehensive treatment of preprocessing and its demonstration of the positive effects preprocessing can have on downstream analytics.

DISCUSSION

State-of-the-art technologies are described for data harmonization and to mitigate noise, which can stymie downstream interpretation, and to select key exposomic features, without which analytics may lose focus. A main task is the reduction of multicollinearity, a particularly formidable problem that frequently arises from repeated measurements of similar events taken at various times and from multiple sources. Empirical results highlight the effectiveness of a carefully planned preprocessing workflow as demonstrated in the context of more highly concentrated variable lists, improved correlational distributions, and enhanced downstream analytics for latent relationship discovery. The nascent field of exposome science can be characterized by the need to analyze and interpret a complex confluence of highly inhomogeneous spatial and temporal data, which may present formidable challenges to even the most powerful analytical tools. A systematic approach to preprocessing can therefore provide an essential first step in the application of modern computer and data science methods. https://doi.org/10.1289/EHP12901.

EAACI guidelines on environmental science in allergic diseases and asthma - Leveraging artificial intelligence and machine learning to develop a causality model in exposomics

Allergic diseases and asthma are intrinsically linked to the environment we live in and to patterns of exposure. The integrated approach to understanding the effects of exposures on the immune system includes the ongoing collection of large-scale and complex data. This requires sophisticated methods to take full advantage of what this data can offer. Here we discuss the progress and further promise of applying artificial intelligence and machine-learning approaches to help unlock the power of complex environmental data sets toward providing causality models of exposure and intervention. We discuss a range of relevant machine-learning paradigms and models including the way such models are trained and validated together with examples of machine learning applied to allergic disease in the context of specific environmental exposures as well as attempts to tie these environmental data streams to the full representative exposome. We also discuss the promise of artificial intelligence in personalized medicine and the methodological approaches to healthcare with the final AI to improve public health.

Predicting self-perceived general health status using machine learning: an external exposome study

BACKGROUND

Self-perceived general health (SPGH) is a general health indicator commonly used in epidemiological research and is associated with a wide range of exposures from different domains. However, most studies on SPGH only investigated a limited set of exposures and did not take the entire external exposome into account. We aimed to develop predictive models for SPGH based on exposome datasets using machine learning techniques and identify the most important predictors of poor SPGH status.

METHODS

Random forest (RF) was used on two datasets based on personal characteristics from the 2012 and 2016 editions of the Dutch national health survey, enriched with environmental and neighborhood characteristics. Model performance was determined using the area under the curve (AUC) score. The most important predictors were identified using a variable importance procedure and individual effects of exposures using partial dependence and accumulated local effect plots. The final 2012 dataset contained information on 199,840 individuals and 81 variables, whereas the final 2016 dataset had 244,557 individuals with 91 variables.

RESULTS

Our RF models had overall good predictive performance (2012: AUC = 0.864 (CI: 0.852-0.876); 2016: AUC = 0.890 (CI: 0.883-0.896)) and the most important predictors were "Control of own life", "Physical activity", "Loneliness" and "Making ends meet". Subjects who felt insufficiently in control of their own life, scored high on the De Jong-Gierveld loneliness scale or had difficulty in making ends meet were more likely to have poor SPGH status, whereas increased physical activity per week reduced the probability of poor SPGH. We observed associations between some neighborhood and environmental characteristics, but these variables did not contribute to the overall predictive strength of the models.

CONCLUSIONS

This study identified that within an external exposome dataset, the most important predictors for SPGH status are related to mental wellbeing, physical exercise, loneliness, and financial status.

The influence of the co-exposure to polycyclic aromatic hydrocarbons and toxic metals on DNA damage in brazilian lactating women and their infants: A cross-sectional study using machine learning approaches

Polycyclic aromatic hydrocarbons (PAHs) and toxic metals are widely spread pollutants of public health concern. The co-contamination of these chemicals in the environment is frequent, but relatively little is known about their combined toxicities. In this context, this study aimed to evaluate the influence of the co-exposure to PAHs and toxic metals on DNA damage in Brazilian lactating women and their infants using machine learning approaches. Data were collected from an observational, cross-sectional study with 96 lactating women and 96 infants living in two cities. The exposure to these pollutants was estimated by determining urinary levels of seven mono-hydroxylated PAH metabolites and the free form of three toxic metals. 8-Hydroxydeoxyguanosine (8-OHdG) levels in the urine were used as the oxidative stress biomarker and set as the outcome. Individual sociodemographic factors were also collected using questionnaires. Sixteen machine learning algorithms were trained using 10-fold cross-validation to investigate the associations of urinary OH-PAHs and metals with 8-OHdG levels. This approach was also compared with models attained by multiple linear regression. The results showed that the urinary concentration of OH-PAHs was highly correlated between the mothers and their infants. Multiple linear regression did not show a statistically significant association between the contaminants and urinary 8OHdG levels. Machine learning models indicated that all investigated variables did not present predictive performance on 8-OHdG concentrations. In conclusion, PAHs and toxic metals were not associated with 8-OHdG levels in Brazilian lactating women and their infants. These novelty and originality results were achieved even after applying sophisticated statistical models to capture non-linear relationships. However, these findings should be interpreted cautiously because the exposure to the studied contaminants was considerably low, which may not reflect other populations at risk.

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.

Exiting the Anthropocene: Achieving personal and planetary health in the 21st century

Planetary health provides a perspective of ecological interdependence that connects the health and vitality of individuals, communities, and Earth's natural systems. It includes the social, political, and economic ecosystems that influence both individuals and whole societies. In an era of interconnected grand challenges threatening health of all systems at all scales, planetary health provides a framework for cross-sectoral collaboration and unified systems approaches to solutions. The field of allergy is at the forefront of these efforts. Allergic conditions are a sentinel measure of environmental impact on human health in early life-illuminating how ecological changes affect immune development and predispose to a wider range of inflammatory noncommunicable diseases (NCDs). This shows how adverse macroscale ecology in the Anthropocene penetrates to the molecular level of personal and microscale ecology, including the microbial systems at the foundations of all ecosystems. It provides the basis for more integrated efforts to address widespread environmental degradation and adverse effects of maladaptive urbanization, food systems, lifestyle behaviors, and socioeconomic disadvantage. Nature-based solutions and efforts to improve nature-relatedness are crucial for restoring symbiosis, balance, and mutualism in every sense, recognizing that both personal lifestyle choices and collective structural actions are needed in tandem. Ultimately, meaningful ecological approaches will depend on placing greater emphasis on psychological and cultural dimensions such as mindfulness, values, and moral wisdom to ensure a sustainable and resilient future.

Associations between the urban exposome and type 2 diabetes: Results from penalised regression by least absolute shrinkage and selection operator and random forest models

BACKGROUND

Type 2 diabetes (T2D) is thought to be influenced by environmental stressors such as air pollution and noise. Although environmental factors are interrelated, studies considering the exposome are lacking. We simultaneously assessed a variety of exposures in their association with prevalent T2D by applying penalised regression Least Absolute Shrinkage and Selection Operator (LASSO), Random Forest (RF), and Artificial Neural Networks (ANN) approaches. We contrasted the findings with single-exposure models including consistently associated risk factors reported by previous studies.

METHODS

Baseline data (n = 14,829) of the Occupational and Environmental Health Cohort study (AMIGO) were enriched with 85 exposome factors (air pollution, noise, built environment, neighbourhood socio-economic factors etc.) using the home addresses of participants. Questionnaires were used to identify participants with T2D (n = 676(4.6 %)). Models in all applied statistical approaches were adjusted for individual-level socio-demographic variables.

RESULTS

Lower average home values, higher share of non-Western immigrants and higher surface temperatures were related to higher risk of T2D in the multivariable models (LASSO, RF). Selected variables differed between the two multi-variable approaches, especially for weaker predictors. Some established risk factors (air pollutants) appeared in univariate analysis but were not among the most important factors in multivariable analysis. Other established factors (green space) did not appear in univariate, but appeared in multivariable analysis (RF). Average estimates of the prediction error (logLoss) from nested cross-validation showed that the LASSO outperformed both RF and ANN approaches.

CONCLUSIONS

Neighbourhood socio-economic and socio-demographic characteristics and surface temperature were consistently associated with the risk of T2D. For other physical-chemical factors associations differed per analytical approach.

A machine-learning approach for predicting the effect of carnitine supplementation on body weight in patients with polycystic ovary syndrome

The present study aimed to explore the effect of carnitine supplementation on body weight in patients with polycystic ovary syndrome (PCOS) and predict an appropriate dosage schedule using a machine-learning approach. Data were obtained from literature mining and the rates of body weight change from the initial values were selected as the therapeutic index. The maximal effect (E_max) model was built up as the machine-learning model. A total of 242 patients with PCOS were included for analysis. In the machine-learning model, the E_max of carnitine supplementation on body weight was -3.92%, the ET₅₀ was 3.6 weeks, and the treatment times to realize 25%, 50%, 75%, and 80% (plateau) E_max of carnitine supplementation on body weight were 1.2, 3.6, 10.8, and 14.4 weeks, respectively. In addition, no significant relationship of dose-response was found in the dosage range of carnitine supplementation used in the present study, indicating the lower limit of carnitine supplementation dosage, 250 mg/day, could be used as a suitable dosage. The present study first explored the effect of carnitine supplementation on body weight in patients with PCOS, and in order to realize the optimal therapeutic effect, carnitine supplementation needs 250 mg/day for at least 14.4 weeks.

Time to Consider the “Exposome Hypothesis” in the Development of the Obesity Pandemic

The obesity epidemic shows no signs of abatement. Genetics and overnutrition together with a dramatic decline in physical activity are the alleged main causes for this pandemic. While they undoubtedly represent the main contributors to the obesity problem, they are not able to fully explain all cases and current trends. In this context, a body of knowledge related to exposure to as yet underappreciated obesogenic factors, which can be referred to as the “exposome”, merits detailed analysis. Contrarily to the genome, the “exposome” is subject to a great dynamism and variability, which unfolds throughout the individual’s lifetime. The development of precise ways of capturing the full exposure spectrum of a person is extraordinarily demanding. Data derived from epidemiological studies linking excess weight with elevated ambient temperatures, in utero, and intergenerational effects as well as epigenetics, microorganisms, microbiota, sleep curtailment, and endocrine disruptors, among others, suggests the possibility that they may work alone or synergistically as several alternative putative contributors to this global epidemic. This narrative review reports the available evidence on as yet underappreciated drivers of the obesity epidemic. Broadly based interventions are needed to better identify these drivers at the same time as stimulating reflection on the potential relevance of the “exposome” in the development and perpetuation of the obesity epidemic.