Physical inactivity is linked to poorer health [1
]. A number of influential reports advocate a significant role for increasing population levels of physical activity to address the growing burden of non-communicable diseases, linked to lifestyle [4
]. Notwithstanding compelling evidence and these calls for action, population levels of physical inactivity remain relatively high in most nations. For example, approximately two-thirds of men and three-quarters of women in the UK are considered not physically active enough to protect their health [5
]. Moreover, this propensity to be inactive appears to begin early in life and to increase throughout the lifespan [8
]. A better understanding of the determinants of physical activity is important to support the development of public health programmes aimed at increasing population levels of physical activity participation.
Most studies that have addressed this issue to date have focused on individual characteristics or circumstances such as age, gender, education, occupation, socioeconomic status and self-efficacy in order to develop interventions aimed at improving knowledge and the motivation to alter individual
lifestyle choices in a particular setting. A recent systematic review of the effectiveness of public health interventions for increasing physical activity [9
] identified common attributes of successful interventions but concluded that the evidence base for policy recommendations in the UK remains sparse. Virtually all of the interventions considered in this review targeted the individual and were limited to a specific setting. It was conceded by the report’s authors that such intervention, at its very best, will have a limited impact on population physical activity and that research on and evaluation of population-based approaches is needed.
There is growing recognition that a sedentary lifestyle is being driven, at least in part, by environmental factors that affect individuals’ physical activity choices and health behaviours. In other words, the environments in which we live, and with which we interact, have become ones that encourage lifestyle choices that decrease physical activity. A number of recent reviews signal an evolutionary shift away from individually orientated theories to broader, more environmentally based approaches for understanding and altering the determinants of population physical activity [10
Neighbourhood environments may either encourage or discourage physical activity [14
] and a number of theoretical models embracing environmental factors as correlates of physical activity have been proposed [15
]. Environmental factors that have been linked to physical activity include: proximity of facilities and spaces and aesthetics [20
], social capital [22
], perceived safety [24
], neighbourhood design, land use mix, transport and traffic [26
], crime [27
] and weather [28
]. However, empirical evidence of a direct association between environmental characteristics and physical activity remains limited. What is more, most of the studies highlighted above suffer from two further limitations. Firstly, in many cases the geographical scale of the area of interest is too large to capture the detail of the interaction between individuals and their immediate environment with loss of sensitivity to detect association. Secondly, few studies have used multi-level analyses, where effects of area level factors (the higher level) on physical activity may be estimated simultaneously with individual level (the lower level) correlates.
Researchers in the United States [29
] and in Australia [31
] have begun to explore the importance of the ‘small’ neighbourhood scale (down to 400 m buffer around respondents’ homes in some instances). The Neighborhood Quality of Life Study (NQLS) [30
] and the Physical Activity in Localities and Community Environments (PLACE) study [32
] were similar in concept to the research reported here, where the primary interest is in the relationship between the neighbourhood environment and health-related behaviours such as physical activity and eating habits. The focus of these two related studies was on active transport and neighbourhood walkability defined in terms of dwelling density, connectivity, land use mix and net retail area. Our focus here is on all physical activity taken outside of the work environment (which includes active transport, garden and domestic and leisure activity but excludes physical activity in the work setting) and we extend the range of environmental factors considered, in particular, including physical activity spaces and facilities, green space and weather. We are not aware of any such studies that have been carried out in urban settings in the UK and that have included simultaneous consideration of weather (an important consideration when the prevailing weather is wet or overcast) and including all components of the available network (paths, short-cuts, cycle routes, streets and roads).
Two recent studies [33
] have used multilevel approaches to examine the association between neighbourhood environment and physical activity. Both studies demonstrated that neighbourhood characteristics are associated with physical activity, both positively and negatively. In the Netherlands (Eindhoven), van Lenthe et al.
] found that those living in more socio-economically disadvantaged areas were more likely to cycle or walk to the shops or work but were less likely to walk, cycle, garden or participate in sport in leisure time than those living in the least disadvantaged areas. In Australia (Melbourne), Kavanagh et al.
] showed significant area level differences in walking, cycling and swimming. Neither of these studies, however, collected sufficient area level detail to be able to offer plausible explanations of environmental characteristics that contribute to the observed variation in physical activity. Kavanagh et al.
go on to suggest that future research should collect detailed environmental data in order to identify key characteristics that could guide urban design to promote greater population levels of physical activity. Added to this, van Lenthe et al.
highlight the importance of individual psycho-social characteristics such as attitudes, self-efficacy and stages of readiness to change and advocate further research that collects environmental characteristics simultaneously with such individual psycho-social factors.
In this paper, we report a detailed multi-level analysis of the role of small area indices and individual factors as correlates of physical activity across a range of areas with different degrees of socio-economic disadvantage. The research provides a detailed mapping of the urban environment down to small local area level and evaluates the relationship between environmental characteristics, individual characteristics (including socio-demographic, psycho-social and perceptions of neighbourhood information) and physical activity behaviours using multi-level analysis.