You are here

Depression and the neighbourhood environment in individuals with and without chronic conditions

TitleDepression and the neighbourhood environment in individuals with and without chronic conditions
Year of Publication2014
AuthorsGariepy, G.
UniversityMcGill University
CityMontréal, QC

{Background: Depression is a serious public-health problem and the leading cause of disability worldwide. Studies have shown neighbourhood characteristics to be associated with depression, but it is not clear which neighbourhood features matter most for depression, for whom this effect is most relevant and what pathways explain this association. Individuals with a chronic disease, such as diabetes, rely on their local area for resources and support, and might be particularly vulnerable in this context. Objectives: To assess the association of neighbourhood characteristics on risk of depression in adults with and without a chronic condition (objective 1); to examine potential moderators and mediators of this association (objective 2); and to explore the relationship between neighbourhood characteristics and changes in depression over time (objective 3). Design and methods: Secondary data analyses were conducted using two cohort studies combined with census, geospatial, satellite imagery and survey data. In a first step, I adopted a wide perspective to examine neighbourhood effects in the general population with and without a chronic condition, using 10 years of data from the National Population Health Survey (NPHS; 2000/01-2010/11; n=17,276 at baseline). In a second step, I focused specifically on people with type 2 diabetes, using 5 years of data from the Diabetes Health Study (DHS; 2008-2013; n=2003 at baseline). The two surveys were analyzed separately. Depression was measured using screening tools. For objective 1, I conducted survival analyses to examine the associations of neighbourhood factors and incident depression, in those with and without a chronic condition. For objective 2, I included interaction terms in the models and stratified analyses to investigate potential moderators. I performed a novel mediation analysis using the additive hazard model to examine potential mediators. For objective 3, I used latent class growth modelling to examine the associations of time-varying neighbourhood factors on trajectories of depression over time. Project 1 - Neighbourhoods and risk of depression in people with and without a chronic condition (NPHS data). Neighbourhood characteristics were not significantly associated with risk of depression in the general sample and in subsamples with a chronic condition. However, moderator analysis revealed that living in proximity to a park was associated with lower risk of depression for people living in crowded households. Project 2 - Neighbourhoods and risk of depression in people with diabetes (DHS data). A greater number of physical activity facilities, cultural services and level of greenness in the neighbourhood were associated with lower risk of depression. Material deprivation was associated with increased risk of depression, particularly in adults with diabetes who were older or retired. Reduction in diabetes complications and disability were significant mediators in the pathway between neighbourhood fitness facilities and depression . Project 3 - Neighbourhoods and trajectories of depression over time (NPHS data). Latent class growth modelling uncovered three distinct trajectories of major depression prevalence: low; moderate decreasing; and high persistent. The presence of neighbourhood parks and cultural services was associated with a significant shift in the trajectory of high persistent depression towards lower probability of major depression. Conclusions: Aspects of the neighbourhood environment were significantly related to risk of depression, particularly in vulnerable subgroups, such as those with diabetes, those living in crowded households and those with persistent major depression symptoms. Future intervention research is needed for health policy recommendations. /// Contexte: La dépression est un problème de santé publique majeur. Des études montrent des associations entre certaines caractéristiques du voisinage et la dépression. Par contre, nous ne savons toujours pas quels aspects du voisinage ont le plus grand impact sur la dépression, quelles personnes sont les plus affectées et par quel mécanisme existe cette association. Les personnes ayant une maladie chronique dépendent du soutien et des ressources de leur voisinage, et donc pourraient être particulièrement vulnérables par leurs contextes environnementaux. Objectifs: Évaluer l'association entre les caractéristiques du voisinage et le risque de dépression chez des adultes avec et sans maladie chronique (objectif 1); examiner les modérateurs et les médiateurs qui contribuent à cette association (objectif 2); et étudier la relation entre les caractéristiques de voisinage et les changements de la dépression à travers le temps (objectif 3). Conception et méthodes: Cette thèse est une l'analyse de données secondaires tirées de deux enquêtes canadiennes, combinées avec des données provenant de recensements, de bases de données géospatiales, d'imagerie satellite, et de sondages. En premier lieu, j'ai adopté une perspective globale des effets de voisinage dans la population générale en me servant de 10 ans de données de l'Enquête nationale sur la santé de la population (ENSP 2000/01-2010/11

Document URL