Individual and contextual correlates of obesity‐related behaviours and obesity: the SPOTLIGHT project

Obesity has grown to be a major health problem across Europe and worldwide 1, and the burden of ill health it imposes is likely to continue to develop for years to come 2. The obesity pandemic affects both individuals through the development of conditions such as type 2 diabetes and cardiovascular diseases, and societies through rising healthcare costs and lower productivity 3-5. Obesity is adiposity, i.e. storage of excess body fat. Body fat can only be stored if more energy is consumed than expended. Thus, over-eating together with a lack of physical activity are the main proximal behavioural risk factors for people to become overweight and obese. It has been posited that these unhealthy lifestyle behaviours are a normal response to environmental characteristics that may influence an individual's level of physical activity (e.g. through the availability of opportunities to walk, interconnectivity of streets, proximity of parks 6-8) and dietary behaviours (e.g. through availability, accessibility and affordability of foods 9, 10) 11. Certain environments may be more ‘obesogenic’ than others – more likely to promote and facilitate unhealthy obesity-promoting behaviours, leading to weight gain in individuals and across populations 12. Accordingly, environmental factors offer a multitude of opportunities for the reduction of obesity prevalence. A number of models describing the multi-dimensional aspects of obesity have been proposed. Conceptual models on how individual-level factors interact with factors acting at the environmental level to influence obesogenic behaviours and weight gain have developed over time. These models have evolved from approaches such as the Analysis Grid for Environments Linked to Obesity (ANGELO) framework 12 and Rothschild's supposition that behavioural determinants can be divided in three broad categories: motivation, abilities and (environmental) opportunities 13. These insights were then incorporated in social–ecological models of health behaviour that further explicated the importance of social and physical contexts as drivers of health 14. In addition, more complex frameworks have sought to capture the multifaceted interactions and mediating pathways of individual and contextual factors (e.g. the EnRG model 15, the IOTF model 16 and the UK Government Foresight obesity systems model 17). In order to contribute to the current knowledge base on these different levels, the European Commission-funded SPOTLIGHT project (derived from ‘sustainable prevention of obesity through integrated strategies’) was established 18. A systems approach was adopted across several independent but inter-related work packages. This was done by identifying mediators of successful outcomes in individual-level obesity interventions 19, by reviewing the state of evidence on social and physical environmental factors related to obesity 7, 20 and evaluating the reach, effectiveness, adoption, implementation and maintenance (RE-AIM) of multi-level interventions reported on in the scientific literature 21, but also of multilevel initiatives that lacked a theoretical basis and were not conducted as part of scientific research 22. We also conducted qualitative analyses of enhancing and impeding factors relating to these multi-level interventions. Outcomes of work packages were used to inform the design of a cross-European survey and a virtual neighbourhood audit 23, 24, upon which the papers in this Obesity Reviews special issue are based. The aim of the SPOTLIGHT survey and neighbourhood audits was to gain understanding of the ways in which diverse factors influence the weight status of individuals, underlying obesogenic behaviours, and the social and physical environmental contexts in which they take place, within a cross-European perspective. While there is growing evidence regarding contextual factors associated with obesity in North America, relatively few such studies have been conducted in Europe 7, where there is substantial variance in terms of obesogenic environments, behaviours and obesity levels. The countries in SPOTLIGHT span the spectrum of obesity prevalence in Europe, ranging from high levels in Hungary and the UK to somewhat lower levels in the Netherlands, Belgium and France. Even within apparently similar contexts (e.g. Belgium-Flanders and the Netherlands), considerable differences in obesity-related behaviours and obesity prevalence exist 25. The contemporary literature, which primarily originates from North America, consistently identifies only a handful of factors linking the built environment to obesity, with appreciable heterogeneity for a wider range of putative factors 6, 7, 26-29. Two correlates in particular stand out: urban sprawl (positively associated with obesity) and land use mix (negatively associated with obesity). Both these measures can be operationalized using characteristics such as residential density. In addition to these physical obesogenic characteristics, a social gradient in obesity-related behaviours and obesity prevalence is consistently seen in developed economies, with higher obesity rates in more socio-economically disadvantaged neighbourhoods 10, 23, 30, 31. The sampling approach used in the SPOTLIGHT survey and neighbourhood audit included five urban areas across Europe with a wide distribution in terms of geography and obesity prevalence: Ghent and suburbs (Belgium), Paris and inner suburbs (France), Budapest and suburbs (Hungary), the Randstad (a conurbation including the cities Amsterdam, Rotterdam, the Hague and Utrecht in the Netherlands) and Greater London (United Kingdom). Neighbourhoods within these areas were stratified by residential density and socio-economic status and then randomly selected. Data collection was harmonized by following the same protocol across all study sites. For the neighbourhood environmental audit, we used an innovative and validated remote sensing tool to categorize the food and physical activity environments 32, 33. Among more than 6000 adult residents of those areas, we collected a wide range of potentially relevant variables using the survey, in which both sides of the energy balance equation was addressed (i.e. physical activity, sedentary behaviour and dietary behaviour). We also assessed psychological constructs relevant to these behaviours, social and perceived environmental attributes, aspects to determine neighbourhood self-selection, and asked participants to draw what they considered as their neighbourhood limits on a map centred on their residential address. The survey faced a number of challenges. Despite a low overall response rate (10%), we collected information from a broad range of participants that reflected the demographic profiles of the countries concerned. Although it remains likely that selection biases have occurred, our sample consisted of equal proportions of men and women, lower and higher educated individuals and wide age ranges 23. Secondly, the use of an online, self-report survey had advantages but also disadvantages. It facilitated a wide reach and made it feasible to include a large sample. In addition, it enabled us to examine domain-specific sedentary and physical activity behaviours – which proved to be important in some of our analyses. Nevertheless, the use of self-report questionnaires can be regarded as a limitation, as they are known to imperfectly capture both exposure and outcome measures 30. Lastly, the survey and neighbourhood audit were cross-sectional – precluding the possibility of identifying causal relations. It is inherently more difficult to perform cross-national comparative research than research within a single country. For instance, because of legal differences across countries, the collection of individual income data was not possible for Belgium, whereas in France, collection of information on ethnicity was not permitted – both for ethical reasons. But the emerging findings – and maybe more importantly the knowledge gained towards these findings – confirm that there is clear value from identifying and comparing contexts, behaviours and health outcomes such as obesity across different cultures and societies. This Obesity Reviews special issue reports findings that use the full breadth of the data from our project. In the methodological part, Charreire et al. present their pioneering work on assessing self-defined neighbourhood boundaries, providing insights into the neighbourhood limits and size as perceived by study respondents 34. Feuillet et al. describe a novel way to ascribe neighbourhoods to one of four possible types based on the obesogenic characteristics identified by the virtual audit 24. The mismatch between perceived and objectively measured environmental obesogenic features is described in the paper by Roda et al. 35, whereas Mackenbach et al. explore why residents of socio-economically deprived neighbourhoods have less favourable perceptions of their neighbourhood environment than residents of wealthy neighbourhoods 36. We present our findings of environmental correlates of cycling for transport 37 and our explorations on how two low-energy output behaviours (i.e. sleep and sedentary behaviour) relate to each other 38. Interactions of individual barriers and neighbourhood destinations with obesity-related behaviours are presented by Mackenbach et al. 39. The (psycho)social environment is covered by systematically summarizing the definitions, measures and existing evidence on associations with weight status 20 and by assessing different ways of operationalizing social capital on a neighbourhood level – and its link with health outcomes 40. In a concluding paper, Rutter et al. 41 build on learning from the project to identify lessons for policy, practice and further research, highlighting the need to develop novel approaches to our study of complex social problems if we are to overcome some of the methodological barriers we face better to understand and tackle obesity at the European and global level. We would like to thank the peer reviewers whose thoughtful comments greatly improved the papers within this supplement: Seraphim Alvanides (Northumbria University Newcastle), Hannah Badland (McCaughey VicHealth Community Wellbeing Unit, Melbourne School of Population and Global Health), Tracie Barnett (INRS-Institut Armand-Frappier), Josephine Chau (School of Public Health, University of Sidney), Billie Giles-Corti (McCaughey VicHealth Community Wellbeing Unit, University of Melbourne), Angela Donkin (UCL Institute of Health Equity), Louisa Ells (Teesside Centre for Evidence-Based Practice), Genevieve Healy (The University of Queensland, Baker IDI Hearth & Diabetes Research Institute, Curtin University), Ruth Hunter (School of Medicine, Dentistry and Biomedical Sciences, Queens University Belfast), Dave van Kann (CAPHRI School for Public Health and Primary Care, Maastricht University), Karen Lamb (School of Exercise and Nutrition Sciences, Deakin University), Lin Lin (The Global Institute for Urban and Regional Sustainability and Urban Planning), Joceline Pomerlau, Margaret Ralston (Department of Sociology, Mississippi State University), Deborah Salvo (Michael & Susan Dell Center for Healthy Living, The University of Texas Health Science Center at Houston, School of Public Health), Stanley Ulijaszek (Unit for Biocultural Variation and Obesity, School of Anthropology, University of Oxford,), Julie Vallée (UMR Géographie-Cités), Chris Rissel (School of Public Health, The University of Sidney). The SPOTLIGHT project (www.spotlightproject.eu) is a cross-European research project with thirteen partners in eight EU member states. Partners in the research consortium include: VU University Medical Center (the Netherlands) Johannes Brug, Jeroen Lakerveld, Joreintje Mackenbach, Gabriela Matias de Pinho, Giel Nijpels; The University of Lisbon (Portugal), Pedro J. Teixeira, Eliana V. Carraça; London School of Hygiene and Tropical Medicine (United Kingdom) Harry Rutter, Ketevan Glonti, Martin McKee; The European Association for the Study of Obesity Euan Woodward; The University of Bergen (Norway) Frida Mathisen, Oddrun Samdal; The University of Oxford (United Kingdom) Lynn Stockley, Gill Cowburn, Charlie Foster; Ghent University (Belgium) Sofie Compernolle, Greet Cardon, Ilse De Bourdeaudhuij; Metropolitan University College Copenhagen (Denmark) Aileen Robertson, Ann Fenger Benwell; Maastricht University (The Netherlands) Jessica Gubbels, Patricia van Oppen Université Paris 13 (France) Jean-Michel Oppert, Maher Ben-Rebah, Thierry Feuillet, Hélène Charreire, Célina Roda, Anaïs Dubreuil; University of Debrecen (Hungary), Helga Bárdos, Róza Ádány; Blox Group (The Netherlands) Hans de Boer; World Obesity Federation Tim Lobstein. The authors have no conflicts of interest to declare. The SPOTLIGHT project was funded by the Seventh Framework Programme (CORDIS FP7) of the European Commission, HEALTH (FP7-HEALTH-2011-two-stage), Grant agreement no. 278186. The content of this article reflects only the authors' views and the European Commission is not liable for any use that may be made of the information contained therein.