gms | German Medical Science

Introduction: Exposome research to investigate health effects of the entirety of environmental exposures has seen a recent increase. However, many exposome approaches lack a comprehensive consideration of social exposures. The social exposome extends initial exposome concepts to the entirety of social exposures at individual and contextual level [1]. It is assumed to interact with the physical environment and to impact the development and health across the life-course. Within the Equal-Life project we aim to detect clusters of children sharing similar exposomes and to analyse their relevance for mental health.

Methods: We analysed data of 2,850 children aged 11-12 of the Amsterdam born children and their development (ABCD) cohort study [2]. 60 variables on the social and physical environment were used to characterise the exposome. As a preprocessing step uniform manifold approximation and projection [3] was applied to project the 2,850 observations of the 60-dimensional covariate space onto two-dimensional Euclidean space. Clustering was performed by agglomerative nesting on the retrieved embedding. Differences between clusters were analysed by random forests and comparison of exposure and health outcome distributions. Children’s mental health was investigated based on prevalence of problematic behavior and symptoms, measured by the strengths and difficulties questionnaire (SDQ) scores, and personality risk for substance abuse, measured by the substance use risk profile scale (SURPS).

Results: We were able to identify eight exposome clusters. Factors such as degree of urbanisation, area deprivation and neighborhood income among others seem to best discriminate between cluster membership. Two clusters stand out with significantly worse SDQ scores and slightly worse SURPS scores among children. Prevalence of mental health problems as indicated by worse SDQ and SURPS scores was higher in exposome clusters characterised as more deprived areas, indicated by low area socioeconomic position, lower mean income and higher percentage of nonwestern migrants.

Conclusion: The results of this innovative approach to identify exposome clusters suggest that in particular social and physical characteristics at the neighborhood level determine exposome clusters in this study population, but not social characteristics at the individual level of the children and their families. Identification of exposome clusters associated with adverse health outcomes may help to target health promotion and prevention measures.

The authors declare that they have no competing interests.

The authors declare that an ethics committee vote is not required.