The assessment of exposure to the neighbourhood food environment is complex and context-dependent. In studies on exposure to food retailers in relation to diet, measures of geographic accessibility (i.e. proximity) and measures of geographic availability (i.e. density) are most often used. Others use measures that combine proximity and density, or focus on a specific retail category relative to other categories in a certain area. In a study just published in Public Health Nutrition and led by Gabriela Matias de Pinho we tested the associations of absolute and relative measures of exposure to food retailers with dietary patterns, using simpler and more complex measures. We tested the association of different neighbourhood food exposure measures with dietary patterns. We used density and spatial access scores (with the latter accounting for both density and proximity) and also relative measures of exposure to food retailers, namely the mRFEI and a ratio of spatial access scores to healthier and less healthy food retailers.

The correlation analyses showed important differences across the exposure measures used. More complex measures of exposure to food retailers did not produce stronger associations with dietary patterns.

Some people may be better than others in resisting the temptations of the food environment. Are these people - who have a higher level of 'mastery' indeed less prone to such influences and eat healthier as a result? Led by Joreintje Mackenbach, we set out to examine the moderating role of mastery in the association of local fast-food restaurants with diet quality and systolic blood pressure. We used cross-sectional data from 1543 adults participating in wave six of the Netherlands Study of Depression and Anxiety (NESDA). Data were collected between 2013 and 2016. Diet quality was defined by adherence with the dietary approaches to stop hypertension (DASH) diet.The study was recently published in the European Journal of Nutrition and showed that exposure to fast-food restaurants around the home was indeed more strongly related to poorer diet quality in individuals with lower levels of mastery. However, exposure to fastfood restaurant was not found to be associated with diet quality and systolic blood pressure, and we observed little evidence for moderation by level of mastery.

GECCO has formally started. In the next few years we will enrich 20+ Dutch cohorts with geo data and make it more easy for researchers to use geo data in their analyses. We are working on a website. Cannot wait to hear more? Contact us here.

Sitting is sickening - but it matters in what context we sit. Television-related sitting time has shown to be one of the worst sedentary behaviours in terms of risk for cardiovascular disease incidence and mortality. Some subgroups watch more TV than others - but what drives these differences in behaviours? In a recent Amsterdam-Gent-Oslo (AGO) collaborative project led by members of the Upstream Team, we examined the social cognitive, home environmental and health-related factors contributing to socioeconomic differences in television-related sitting. Two Belgian data sets were used to evaluate the differences in an older and younger population. The results were recently published in the European Journal of Public Health, and show that occupation does not seem to matter. Social cognitive variables such as attitude and modelling of the partner explained a large part of the educational inequalities in television-related sitting time, and home-environmental variables to a lesser extent.

The Geoscience and health cohort consortium (GECCO) was founded by the late EMGO Institute - now Amsterdam Public Health research institute. Six large cohort studies were enriched with a variety of Geographic Information Systems (GIS) data, to provide information on the environmental context of cohort participants in addition to the extensive individual-level information on behaviour and health. We previously reported on this initiative and published it here. 

Now, with help of The Netherlands Organisation for Scientific Research (NWO) and the Netherlands Organisation for Health Research and Development (ZonMw), GECCO will extent to all major and longstanding cohorts in the Netherlands. In a stepwise approach we will: i) Gather, combine and complement existing GIS data in a repository; ii) Enrich 20 longstanding and renowned Dutch cohorts with these GIS data by secure and local linkages; and iii) Develop an easily accessible web-portal that facilitates researchers to use the data by informing them about the cohort and GIS meta-data, and provide guidance through all data access procedures.

GECCO is chaired by Upstream Team member Jeroen Lakerveld. Until the website is launched you can obtain more info here.

The SUPREME NUDGE project is underway and its rationale and design was recently published in BMC Public Health. SUPREME NUDGE is a 5-year research project financed by the Dutch Heart Foundation and ZonMw. With SUPREME NUDGE we expand a previous successful Dutch supermarket pricing intervention, and incorporate promising elements such as nudging and ICT applications to provide real-time and context-specific physical activity feedback. More information can also be found on the dedicated website.

During a week-long symposium organised by a small group of multi-disciplinary experts in both spatial science and public health gathered to explore a variety of open questions and pressing issues in the field. Upstream Team was present to provide perspectives on some of the challenges that we face, including:

1. What, primarily, hinders the development of applications of spatial data and approaches for analysis in chronic disease research?
2. How could we link remote sensing to chronic disease epidemiology in theory?
3. How can we break the isolation of multiple disciplines caused by different principles and methods?
4. How can we measure life course and life-space environmental exposures of populations?

The symposium was well prepared and chaired by Peng Jia of the University of Twente (Netherlands) and Peter James from the Harvard Medical School. We look forward to the second edition, and in the meantime participants will use the input for scientific output!

In the Netherlands, a great variety of objectively measured geo-data is available, but these GIS data are scattered and measured at varying spatial and temporal scales. The centralisation of these data and the linkage to individual-level data from longitudinal cohort studies enable large-scale epidemiological research on the impact of the environment on health behaviours and health outcomes. In the Geoscience and Health Cohort Consortium (GECCO), six large-scale and ongoing cohort studies from the Amsterdam Public Health research institute (APH) have been enriched with a variety of existing geo-data. This pilot project is described in a cohort profile paper, led by dr. Erik Timmermans, has just been published in BMJ Open. This initial GECCO project already proved to be of value for many researchers interested in doing analyses on upstream determinants, and its methods provide a great research potential which increases once more cohorts join. More soon!

The availability of outdoor recreational facilities is associated with increased leisure-time physical activity (PA). In a study recently published in the International Journal of Behavioral Nutrition and Physical Activity, we investigated how much of this association is attributable to selection effects, and explored whether usage of recreational facilities was an explanatory mechanism.

In this study first authored by Joreintje Mackenbach we analysed data from 5199 participants across Europe. Adults completed a survey and the availability of outdoor recreational facilities in the residential neighbourhood was objectively measured. We found that 

Subjective – but not objective – availability of outdoor recreational facilities was associated with higher levels of total leisure-time PA. After adjustment for self-selection (which attenuated the association by 25%), we found a 25% difference in weekly minutes of total leisure-time PA between individuals with and without self-reported availability of outdoor recreational facilities. For our study population, this translates to about 28 min per week. Participants who reported outdoor recreational facilities to be present but indicated not to use them (RR = 1.19, 95% CI = 1.03;1.22), and those reporting outdoor recreational facilities to be present and to use them (RR = 1.33, 95% CI = 1.22, 1.45) had higher levels of total leisure-time PA than those who reported outdoor recreational facilities not to be present. Read the details here.

From June 3-6,  the annual conference of the International Society for Behavioral Nutrition and Physical Activity (ISBNPA) took place in Hong Kong. The Upstream Team was well represented, with various oral presentations and poster presentations. A symposium was also organised on challenges in upstream research. Suggestions for future studies that were given based on the content and discussions afterwards included: use food purchasing outcomes rather than dietary intake (because then you know whether individuals actually interacted with specific shops); to engage with land use planners and geographers to better understand why shops are where they are (in addition to understanding why individuals go to certain shops); to take into account affordability as an important food environment exposure; to combine qualitative, quantitative observational and experimental studies to triangulate the findings; to balance specificity vs. complexity; etc.

Next year the conference will be organised in Prague - see you there!

The built environment influences behaviour, like physical activity, diet and sleep, which affects the risk of type 2 diabetes mellitus (T2DM). In order to assess and structure the current evidence base on this topic, we systematically reviewed and meta-analysed evidence on the association between built environmental characteristics related to lifestyle behaviour and T2DM risk/prevalence, worldwide. The results were just published in BMC Medicine in a paper led by Nicole den Braver. Overall, the results indicate that living in an urban residence was associated with higher T2DM risk/prevalence (OR = 1.40; 95% CI, 1.2-1.6) compared to living in a rural residence. Higher neighbourhood walkability was associated with lower T2DM risk/prevalence (OR = 0.79; 95% CI, 0.7-0.9) and more green space tended to be associated with lower T2DM risk/prevalence (OR = 0.90; 95% CI, 0.8-1.0). No convincing evidence was found of an association between food environments with T2DM risk/prevalence.

A considerable proportion of European adults report little or no interest in physical activity. Identifying individual-level and environmental-level characteristics of these individuals can help designing effective interventions and policies to promote physical activity. In our latest paper led by Eliana Carraça and published in Preventive Medicine we showed that adults who were not interested in physical activity reported a higher BMI and a lower self-rated health and were less educated. They were also more prone to have less healthy eating habits, and to perceive more barriers towards physical activity. Only minor differences were observed in environmental attributes: the non-interested were slightly more likely to live in neighborhoods objectively characterized as less aesthetic and containing more destinations, and perceived as less functional, safe, and aesthetic.

n our most recently published Upstream Team paper, just published in PLOS One, we aimed to examine the associations of both objectively assessed and perceived physical and social neighborhood characteristics with happiness in European adults. Thousands of adults across Europe Respondents were surveyed and reported their level of happiness on a 5-point Likert scale, and rated their perceptions of physical and social environmental neighbourhood characteristics. Objective physical environmental characteristics were also assessed, using a Google Street View-based neighbourhood audit. Associations of 14 physical and social environmental characteristics with happiness were analysed.

Living in neighbourhoods that look better ('better aesthetics') and more water and green space was associated with being very happy. Individuals who perceived their neighbourhood to be safer, more functional and more aesthetic were more likely to be very happy. The associations of functionality and aesthetics with happiness were strongest in the Ghent region (Belgium), the Randstad (the Netherlands) and Greater London (United Kingdom). Individuals with a larger social network, more social cohesion and who trusted their neighbours were more likely to be very happy. The association between social networks and happiness was somewhat stronger in men than in women. In general, the associations between environmental characteristics and happiness had similar directions and sizes across socio-economic and socio-demographic subgroups.

Cooking at home has been associated with healthier diets. While individual-level factors such as sex and employment status have been consistently associated with home-cooking, little is known about its upstream determinants. In a study led by Maria Gabriela Pinho and recently published by the IJBNPA we explored the independent association between neighbourhood spatial access to grocery stores and to restaurants with frequency of home cooking. We also tested how the interaction between these two type of stores are associated with home-cooking.

We found that greater access to restaurants was associated with lower frequency of home-cooking, while no association was found for access to grocery stores. We found no strong evidence for a joint association of spatial access to grocery stores and spatial access to restaurants with home-cooking.

On January 1st, the journal Environment and Behavior published a study led by Upstream Team member Joreintje Mackenbach. In this study, we used data from the European Spotlight project to examine via which pathways, and for which subgroups, spatial access to fast food outlets was associated with body weight in adults. We found that there was no direct relation between access to fast food outlets and body weight. However, access to fast food outlets was associated with perceptions about and usage of fast food outlets, and this was in turn associated with greater reported fast food consumption and unhealthier weight status. We did not find evidence for effect modification by age, gender, socioeconomic status or country.

The aim of PEN, which will start in May 2018, is to establish a multi-disciplinary research network for the monitoring, benchmarking and evaluation of policies that affect dietary and physical activity as well as sedentary behavior with a standardized approach across Europe.

The Policy Evaluation Network will make use of the achievements of the DEDIPAC KH, especially Thematic Area 3 and combine it with other existing tools and frameworks (e.g. INFORMAS) to develop a consistent and sustainable approach for the evaluation and benchmarking of existing policies. The overal project coordinator is Wolfgang Ahrens, and the Vice-coordinators are Catherine Woods and Jeroen Lakerveld.