Weighing the importance of neighbourhood: A multilevel exploration of the determinants of overweight and obesity

Abstract

Overweight and obesity have reached epidemic proportions in many countries, including Canada. In addition to becoming critical public health challenges in and of themselves, they represent major risk factors for chronic disease and disability (e.g. cardiovascular disease, diabetes). The various symptoms and co-morbidities associated with these chronic conditions place tremendous stress on the Canadian health care system, generating economic concern. This research takes a population health approach to the study of obesity, examining the complex relationships between individual demographics and behaviours, and aspects of the local social and physical environments. A subset of a nationally representative survey was linked to neighbourhood-level data from the 1991 Canadian Census, and analysed from a multilevel perspective. This study found substantial area-level variation in body mass index and waist circumference, and discovered an important role for neighbourhood-level characteristics independent of individual-level characteristics. These findings provide evidence that the underlying mechanisms driving the increasing prevalence of overweight and obesity may be so called obesogenic environments that encourage physical inactivity and unhealthy eating. An effective policy response must address environmental conditions in order to curb current obesity trends.

Introduction

On a global scale, the increasing prevalence of obesity has reached epidemic proportions, and is a major contributing factor to dramatic increases in the rates of various chronic diseases and disabilities. Worldwide, approximately 1 billion adults are overweight (body mass index [BMI] ≥ 25 kg/m²) and at least 300 million of these individuals can be classified as obese (BMI ≥ 30 kg/m²) as defined by the World Health Organization standards (WHO, 2006). Likewise, in Canada obesity has emerged as a critical public health issue over the last several decades. The 2004 Canadian Community Health Survey found that approximately 36% of Canadian adults were overweight, and almost 23% could be classified as obese (Canadian Institute for Health Information [CIHI], 2004).

Overweight and obesity have been linked to numerous adverse health outcomes such as type II diabetes mellitus (WHO, 2006), various cancers (World Cancer Research Fund/American Institute for Cancer Research, 2007), and adverse psychosocial outcomes such as depression (Kim, Meade, & Haines, 2006). Additionally, it is now clear that overweight and obesity are important risk factors for cardiovascular disease, currently the leading cause of mortality in Canada and much of the western industrialized world (Health Canada, 2006). In 2001, national medical costs attributable to adult overweight and obesity were estimated at $4.3 billion (CAD), or 2.2% of total health care expenditure (Katzmarzyk & Janssen, 2004).

At the physiological level, overweight and obesity are caused by an energy imbalance whereby caloric intake (diet) exceeds expenditure (physical activity) (Black & Macinko, 2008). Despite this relatively simple equation, obesity is a complex, multifactorial disease (Huot, Paradis, & Ledoux, 2004). Evidence has consistently shown variation in obesity rates between individuals to be associated with genetic, psychological, socioeconomic, and behavioural factors, all of which are thought to play a substantial role in onset (Poortinga, 2006). Until recently, obesity research has focused on exploring and identifying these potentially modifiable individual risk factors. While these strategies have had some success in developing intervention approaches for obesity, they have failed to effectively slow or reverse current obesity trends at the population-level, which indicate increasing prevalence across individuals of all ages, genders, socioeconomic groups, educational levels, and geographic regions (McLaren, 2007).

Within more recent obesity research, there is a growing consensus that individual characteristics and behaviours, and in particular dietary and physical activity patterns, are influenced by aspects of the broader socioeconomic, cultural, and environmental contexts in which people live (Moon et al., 2007, Ross et al., 2007). Following arguments developed by Rose (1985), population-level interventions have the power to shift the obesity distribution curve in a direction that would benefit the entire population by removing or modifying these underlying forces. Indeed, major gains in altering other health behaviours, such as reducing smoking-related inequalities in health, have been realized by developing interventions focused on modifying population-level determinants in combination with individual-level interventions (Thomas, Fayter, & Misso, et al., 2008).

A more ecologically informed approach to the obesity problem has the potential to identify, and help us understand the so called obesogenic environments which foster high energy diets and sedentary lifestyles, and develop an appropriate ‘upstream’ preventive response (Black & Macinko, 2008). It is these environmental factors that are explored in this paper. Overweight and obesity are framed as being functions of individual characteristics (e.g. age, sex, education, socioeconomic status, physical activity behaviours, dietary patterns), operating within a local neighbourhood context characterized by various socioeconomic, cultural, and physical characteristics.