Elsevier

Social Science & Medicine

Volume 68, Issue 4, February 2009, Pages 601-609
Social Science & Medicine

The link between local environment and obesity: A multilevel analysis in the Lisbon Metropolitan Area, Portugal

https://doi.org/10.1016/j.socscimed.2008.11.033Get rights and content

Abstract

Although individual factors have been shown to predict weight gain, contextual determinants have also attracted attention, with some authors stressing the role played by deprivation, urban sprawl, social capital and safety. Recent evidence has implicated environmental factors that facilitate the consumption of excess calories and/or make it more difficult to expend them in routine physical activity. The interrelationships found in some places between physical and social environments (key mediators) and body mass index (BMI), as well as the potential that exists for the development of healthier places, mean that more research is required into the contextual determinants of health.

In Portugal, particularly in the Lisbon Metropolitan Area (LMA), the effects of physical and social environments on physical activity and BMI have not previously been explored in any detail. This study aims to highlight the associations between residential (physical and social) environment and the risk of weight gain and obesity, over and above individual attributes, assessing which indicators are the best predictors of excess weight in the LMA.

The study involved data from 7669 individuals aged 18 and over from 143 neighbourhoods. Self-reported body height and weight were used to define overweight body mass index (BMI  25). BMI and individual (socio-demographic and behavioural) characteristics were linked to contextual data and analysed in a multilevel framework.

Our findings show that different environmental factors are significantly associated with excess weight and obesity, either directly or indirectly (e.g. health-related behaviours such as eating patterns and physical activity, which are key mediators), after adjustment for individual characteristics. The results suggest that a deeper understanding of these mechanisms is critical if we want to tackle the obesity epidemic, and that policies aimed at weight control and obesity reduction must address people and places in order to bear fruit.

Introduction

Obesity has reached epidemic proportions in almost all developed countries (Backett-Milburn, Wills, Gregory, & Lawton, 2006) and in many urban areas in developing countries. Recent estimates suggest that one billion adults in the world are overweight (Kjellstrom, 2007).

Portugal has a significant obesity problem, which is rapidly worsening. In 1996, 10.3% of males and 11.4% of females had a BMI of over 30 kg/m2 (OECD, 2006); by 2006, this had increased to 16% for males and 16.9% for females (ACS, 2008), representing an additional 55.3% and 43.9% for males and females respectively over a period of ten years. Thus, by 2006, 1 in 2 adults were overweight or obese (ACS, 2008). Targets have been set by the National Health Plan (ACS, 2008) to reduce obesity and excess weight in the population by the year 2010. Obesity is a national public health problem: it has damaging effects on the social, economic and health status of individuals, exacerbating their risks of disease and premature death and increasing health care costs.

Several authors have claimed that excess body weight results from multifactorial causes, including polygenic, metabolic, psychosocial and environmental influences, and behavioural aspects (Borders et al., 2006, Ellaway et al., 1997, Poortinga, 2006a, Poston and Foreyt, 1999). Genes play a role, but the gene pool in industrialised countries has not changed significantly in the few decades in which obesity has rocketed. Therefore, environmental (obesogenic) factors must also contribute significantly to weight gain. Some authors (Cohen et al., 2006, Ellaway et al., 1997, Kim et al., 2006, Poortinga, 2006a, Poston and Foreyt, 1999, Stafford et al., 2007, Van Lenthe and Mackenbach, 2002) have indicated the potential obesogenicity of certain environmental factors (i.e. the extent to which they could promote caloric intake and/or discourage the expenditure of energy in routine physical activity).

However, the precise cause of the epidemic remains unknown, and no possible solutions are yet in sight. It is important, therefore, to identify exactly what is a “toxic environment” (Poston & Foreyt, 1999) in order to focus upon changing it. Swinburn, Egger, & Raza (1999: 564) argue that “there is an urgent need for a conceptual and practical framework to dissect the rather nebulous concept of the environment into concrete elements which are amenable to measurement and interventions”.

Increasing obesity amongst the Portuguese population is associated with trends in: excessive energy intake (3747 calories on average per person, in 2003, representing an increase of 32.6% in the last thirty years, that is 1.6% per year) (OECD, 2006); and low rates of leisure-time physical activity (overall age-adjusted prevalence for a sedentary lifestyle was 62.8% for males and 75.4% for females in 1999) (Demarest et al., 2007). By 2004, 73% of Portuguese people reported that they seldom or never did physical exercise for recreation or as a leisure-time activity, and 54% reported identical levels of inactivity at work (Eurobaromater, 2004). These behavioural risks for obesity, in isolation or together with genetic susceptibilities or vulnerabilities, may vary from individual to individual (in association with socioeconomic status) and according to place of residence (opportunities for healthy behaviours: physical recreation and eating) (Ellaway et al., 1997, Stafford et al., 2007).

The purpose of this paper is to examine the contribution of the local environment and personal attributes to the risk of weight gain in LMA neighbourhoods and evaluate their importance for the understanding and treatment of obesity. This information will be relevant for the design and planning of health promotion interventions, in so far as it will facilitate the identification of areas requiring analysis, major issues and priorities for tackling the obesity epidemic in the Lisbon Metropolitan Area (LMA).

To achieve this, the authors intend to: (1) develop indicators to measure social environment and availability of local resources; (2) link individuals to their neighbourhoods through the characterization of each individual personally and according to the specific features of their place of residence; (3) establish associations between sustainable communities, their citizens' body mass index (BMI) and risk factors for obesity (i.e. behaviours such as physical activity and healthy eating habits).

The next section describes the conceptual framework for obesity, including data and methods used, followed by the main achieved results. Finally, in the last part, possible explanations are suggested, some conclusions are drawn and key proposals are made for changes that could lead to alternative health outcomes.

Section snippets

A conceptual framework for obesity, and hypotheses

How might area characteristics be associated with the risk of increased BMI? Does place act or interact with individual characteristics? How does this happen?

Multilevel models

We present in Table 3 the models that showed the better fit according to DIC, for the four models developed. The values of variance between neighbourhoods and DIC for the null models are presented to highlight the reduction of variance between neighbourhoods after taking into account the included variables and of DIC.

Age, gender, occupation, education and income, marital status, lifestyles and overall health /initial health status (self-rated) were included, although there were some differences

Discussion

This paper considers the features of the local social and physical environment that may affect body mass index (BMI) by encouraging behaviours such as physical activity and eating patterns.

Our results from Lisbon are in concordance with previous empirical work elsewhere (Poortinga, 2006a, Kim et al., 2006) and generally with theoretical reviews on weight gain and its pathways – diet and physical activity – (Stafford et al., 2007). The literature suggests that levels of physical activity

Acknowledgements

The authors would like to thank Sally Macintyre, Anne Ellaway and Min Hua Jen for their useful insights and advice, Cláudia Costa for helping to collate and manage geographical data, Karen Bennett for the language review, and two anonymous referees for their very helpful comments. This research was supported by grant POCTI/GEO/45730/2002 from the Portuguese Foundation for Science and Technology (FCT) “Healthy Urban Planning”.

References (52)

  • S. Macintyre et al.

    Place effects on health: how can we conceptualise, operationalise and measure them?

    Social Science & Medicine

    (2002)
  • L. McNeill et al.

    Social environment and physical activity: a review of concepts and evidence

    Social Science & Medicine

    (2006)
  • A. Parkes et al.

    The multi-dimensional neighbourhood and health: a cross-sectional analysis of the Scottish Housing Survey, 2001

    Health & Place

    (2006)
  • R. Pendola et al.

    BMI, auto use, and the urban environment in San Francisco

    Health & Place

    (2007)
  • W. Poortinga

    Perceptions of the environment, physical activity, and obesity

    Social Science & Medicine

    (2006)
  • W. Poortinga

    Social relations or social capital? Individual and community health effects of bonding social capital

    Social Science & Medicine

    (2006)
  • W.S. Poston et al.

    Obesity is an environmental issue

    Atherosclerosis

    (1999)
  • S. Robert et al.

    A multilevel analysis of race, community disadvantage, and body mass index among adults in the US

    Social Science & Medicine

    (2004)
  • P. Santana

    Poverty, social exclusion and health in Portugal

    Social Science & Medicine

    (2002)
  • M. Stafford et al.

    Pathways to obesity: identifying local, modifiable determinants of physical activity and diet

    Social Science & Medicine

    (2007)
  • K. Sundquist et al.

    Linking social capital and self-rated health: a multi-level analysis of 11,175 men and women in Sweden

    Health and Place

    (2007)
  • F.J. Van Lenthe et al.

    Neighbourhood inequalities in physical inactivity: the role of neighbourhood attractiveness, proximity to local facilities and safety in the Netherlands

    Social Science & Medicine

    (2005)
  • ACS

    Plano Nacional de Saúde: Indicadores e Metas do PNS

    (2008)
  • T.F. Borders et al.

    Gender-specific disparities in obesity

    Journal of Community Health

    (2006)
  • P. Calthorpe et al.

    The regional city. Planning for the end of sprawl

    (2001)
  • V. Carstairs et al.

    Deprivation and health in Scotland

    (1991)
  • Cited by (77)

    • Is walkability equitably distributed across socio-economic groups? – A spatial analysis for Lisbon metropolitan area

      2023, Journal of Transport Geography
      Citation Excerpt :

      It is composed of high-density central neighbourhoods along with a diversity of both low- and high-density suburbs. Residential land use predominates in most suburbs, and public transport supply is poor (Santana et al., 2009; Santos, 2017). In the LMA, high-density areas are found alongside low-density zones and the urban space can be defined using the National Institute for Statistics' (INE) classification of census blocks.

    • Global aging and health determinants in a changing world

      2022, Aging: From Fundamental Biology to Societal Impact
    View all citing articles on Scopus
    View full text