Association of neighbourhood disadvantage and individual socioeconomic position with all-cause mortality: a longitudinal multicohort analysis

Background Few studies have examined the interactions between individual socioeconomic position and neighbourhood deprivation and the findings so far are heterogeneous. Using a large sample of diverse cohorts, we investigated the interaction effect of neighbourhood socioeconomic deprivation and individual socioeconomic position, assessed using education, on mortality. Methods We did a longitudinal multicohort analysis that included six cohort studies participating in the European LIFEPATH consortium: the CoLaus (Lausanne, Switzerland), E3N (France), EPIC-Turin (Turin, Italy), EPIPorto (Porto, Portugal), Melbourne Collaborative Cohort Study (Melbourne, VIC, Australia), and Whitehall II (London, UK) cohorts. All participants with data on mortality, educational attainment, and neighbourhood deprivation were included in the present study. The data sources were the databases of each cohort study. Poisson regression was used to estimate the mortality rates and associations (relative risk, 95% CIs) with neighbourhood deprivation (Q1 being least deprived to Q5 being the most deprived). Baseline educational attainment was used as an indicator of individual socioeconomic position. Estimates were combined using pooled analysis and the relative excess risk due to the interaction was computed to identify additive interactions. Findings The cohorts comprised a total population of 168 801 individuals. The recruitment dates were 2003-06 for CoLaus, 1989-91 for E3N, 1992-98 for EPIC-Turin, 1999-2003 for EPIPorto, 1990-94 for MCCS, and 1991-94 for Whitehall II. We use baseline data only and mortality data obtained using record linkage. Age-adjusted mortality rates were higher among participants residing in more deprived neighbourhoods than those in the least deprived neighbourhoods (Q1 least deprived neighbourhoods, 369.7 per 100 000 person-years [95% CI 356.4-383.2] vs Q5-most deprived neighbourhoods 445.7 per 100 000 person-years [430.2-461.7]), but the magnitude of the association varied according to educational attainment (relative excess risk due to interaction=0.18, 95% CI 0.08-0.28). The relative risk for Q5 versus Q1 was 1.31 (1.23-1.40) among individuals with primary education or less, but less pronounced among those with secondary education (1.12; 1.04-1.21) and tertiary education (1.16; 1.07-1.27). Associations remained after adjustment for individual-level factors, such as smoking, physical activity, and alcohol intake, among others. Interpretation Our study suggests that the detrimental health effect of living in disadvantaged neighbourhoods is more pronounced among individuals with low education attainment, amplifying social inequalities in health. This finding is relevant to policies aimed at reducing health inequalities, suggesting that these issues should be addressed at both the individual level and the community level. Copyright (C) 2022 The Author(s). Published by Elsevier Ltd.

Méthodologie de gestion durable des ressources du sous-sol urbain

Pascal Blunier

The world is becoming more and more urban. During the last fifty years, the population in Swiss cities has grown from 45% to 70% of the total population. This change is characterised by urban sprawl and by an increase in the consumption of resources, particularly those coming from rural areas. This evolution is not compatible with the principles of sustainable development. To solve this problem, new urban policies are being developed. They favour development within the urbanised areas and reduced use of exogenous resources. The urban underground contains resources – space, geomaterials, groundwater and geothermal energy – which can efficiently support such policies. This project analyses how these resources have been exploited until now and how they could be used with sustainability in mind. The use of underground resources was analysed in five cities : Mexico, Paris, Helsinki, Tokyo and Montreal. It appears that, although resources participate in the urban metabolism, there is an important lack of planning and coordination in their use. Developments follow a sectional approach where the resources are considered independently, and only at the scale of a given construction project. As a consequence, conflicts occur between uses and resources are wasted. An understanding of the urban underground as a system was developed with the aim of preventing conflicts and promoting synergies. To this effect, the way in which resources interact was considered attentively. Situations in which interactions occur were described and analysed and compatibility conditions were suggested. A methodology was elaborated to evaluate the resource potential of an urban underground. It aims at considering the resources and the interactions between their uses from the beginning of the process of urban planning. The methodology is based on the use of Geographical Information Systems and 3-D modelling of geological and hydrogeological conditions. It aims at reversing the current paradigm, that proceeds from the needs to the resources, into an approach based on resources to satisfy the needs. Spatial information was exploited to determine indicators and maps of potentiality and of restrictions. The derived results will help support decision making by planners and policy makers who seek to better exploit the resources from the urban underground. The methodology was tested and applied on a case study : Geneva. The developed approach allows more efficiency in the use of urban underground resources, it can help prevent conflicts between uses and develop synergies. When integrated in the urban planning process, it highlights new opportunities and helps to develop projects that take into account the physical conditions of the underground.

EPFL2009

Social Network Analysis for Automatic Role Recognition

Sarah Favre

The computing community has shown a significant interest for the analysis of social interactions in the last decade. Different aspects of social interactions have been studied such as dominance, emotions, conflicts, etc. However, the recognition of roles has been neglected whereas these are a key aspect of social interactions. In fact, sociologists have shown not only that people play roles each time they interact, but also that roles shape behavior and expectations of interacting participants. The aim of this thesis is to fill this gap by investigating the problem of automatic role recognition in a wide range of interaction settings, including production environments, e.g. news and talk-shows, and spontaneous exchanges, e.g. meetings. The proposed role recognition approach includes two main steps. The first step aims at representing the individuals involved in an interaction with feature vectors accounting for their relationships with others. This step includes three main stages, namely segmentation of audio into turns (i.e. time intervals during which only one person talks), conversion of the sequence of turns into a social network, and use of the social network as a tool to extract features for each person. The second step uses machine learning methods to map the feature vectors into roles. The experiments have been carried out over roughly 90 hours of material. This is not only one of the largest databases ever used in literature on role recognition, but also the only one, to the best of our knowledge, including different interaction settings. In the experiments, the accuracy of the percentage of data correctly labeled in terms of roles is roughly 80% in production environments and 70% in spontaneous exchanges (lexical features have been added in the latter case). The importance of roles has been assessed in an application scenario as well. In particular, the thesis shows that roles help to segment talk-shows into stories, i.e. time intervals during which a single topic is discussed, with satisfactory performance. The main contributions of this thesis are as follows: To the best of our knowledge, this is the first work where social network analysis is applied to automatic analysis of conversation recordings. This thesis provides the first quantitative measure of how much roles constrain conversations, and a large corpus of recordings annotated in terms of roles. The results of this work have been published in one journal paper, and in five conference articles.

EPFL2011

A fine scale analysis of socio-economic and landscape variables for the assessment of health status in Geneva

Joël Chételat, Idris Guessous, Stéphane Joost, Abram Pointet

Significant differences in health status between inhabitants of different commune types of the Geneva area have recently been established in a study relying on a large health database managed by the Unit of Population Epidemiology of the Hôpitaux Universitaires de Genève (HUG). For instance, the mean Body Mass Index is lower in richer communes than in other urban areas. Moreover, suburban residents and commuting populations have a higher hypertension prevalence than other communities. Though socio-economic status appears to be an essential driver for quality of life and health, specific knowledge is lacking to define which variables are of main concern. Besides, the relationship between socio-economic conditions and neighbouring living environment is only implicit. The objective of the present study is to detail the spatial differences coarsely observed previously. More particularly, it aims to determine to which extent landscape indicators are related to health risk factors and to identify areas where prevalence of health risk factors is potentially high on a fine scale basis. The study reuses the “Bus santé” database containing health-related information for more than 10'000 participants collected since 1998.Stratified sampling by age and by gender was used for the survey to be representative of the population. Samples were geocoded so that they can be analysed in a Geographic Information System (GIS). The Geneva canton has been divided into some 1600 homogeneous zones called Microregions based on the number of inhabitants, jobs, buildings and dwellings. Demographic, socio-economic and landscape variables such as the proportion of built-in surface and volume or the Shannon land use diversity index inform each of the Microregions. Significant factors characterising them are identified using factor analysis in order to make a spatial typology. A multiple regression is applied to assess the relationship between a set of health variables (including blood pressure and hypertension, glucose level and diabetes) from the sampled individuals contained within the Microregions and the factors describing these. Results of this fine scale analysis show the importance of income and socio-cultural membership as primary variables related to lifestyle and potential health status. The chosen landscape composition and structure variables come as complementary explanatory factors of neighbourhood counting for well-being or on the contrary stress-related illness. As an environmental health research, this study directly contributes to address land planning issues and inform on how landscape design potentially impacts quality of life and population health. Moreover, it serves to identify spatial types where risk factors are high and locate them in the geographical space.

2012