Neighborhood Social and Economic Change, Food Environment Change and Diabetes Incidence in Madrid (Spain)
Embargo until
Date
2017-03-17
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Johns Hopkins University
Abstract
Background
The dynamic nature of residential environments is an understudied macro-social determinant of health.
Objectives
The aims of this dissertation were to measure neighborhood social and economic change, and to evaluate its association with changes in the retail food environment and diabetes incidence.
Methods
We collected area-level data from multiple administrative sources in the city of Madrid (Spain) from 2005 to 2016. For Aim 1, we computed measures of change in indicators related to residential mobility, socioeconomic and sociodemographic characteristics, and housing construction and renovations. We used a finite mixture model to measure types of areas according to how they change, revealing four types of neighborhood social and economic change. For Aim 2, we geocoded and categorized retail food stores into 4 categories: a) any food store, b) supermarkets, c) small specialized stores and d) fruit and vegetable stores. We used a multinomial logistic regression model to evaluate the association between neighborhood change type and food environment change. For Aim 3, we used data from the HeartHealthyHoods Retrospective Study that included electronic health records on every individual registered in a health center of four districts of Madrid. We used a Cox proportional hazards model to estimate the association between neighborhood change type and diabetes incidence.
Results
Our discrete measurement model identified four types of neighborhoods (census sections) according to their change: (Type 1) areas with an increasing proportion of foreign-born migrants and a relative worsening in SES markers; (Type 2) areas with high residential mobility and housing constructions, relative reduction in average age and increase in total population; (Type 3) areas with a relative improvement in SES markers and increases in housing renovations; and (Type 4) areas with low residential mobility, and a relative increase in average age, reduction in foreign-born migrants and total population. Type 3 areas were associated with an increase in the number of small specialized stores, a decrease in the number of supermarkets and an increased incidence of diabetes. Type 1 and 4 areas were associated with an increase in the number of supermarkets and decrease in the incidence of diabetes.
Conclusions
Measuring a complex exposure such as neighborhood social and economic change is a challenging endeavor. Further study of this association with food environment changes should include consideration of opening hours. If the finding of an association between neighborhood type and diabetes incidence can be replicated, policy-based diabetes prevention programs may be developed and tested.
Description
Keywords
neighborhoods, residential environments, diabetes, food environment, food stores, urban health, Europe, Spain