Purpose: Adolescence is a period of increased autonomy over decision making, including food choices, and increased exposure to factors outside the home. Combining these factors means the physical food environment could play an important role in adolescents’ independent food purchasing and dietary behaviours. This review aimed to understand how community and consumer nutrition environment exposures were associated with adolescent food purchasing and dietary behaviours.

Methods: Six databases were searched using MESH and free-text terms. Titles and abstracts were screened by one reviewer. If eligible, two reviewers performed data extraction and assessed each article for risk of bias in relation to the research questions using predefined criteria. As meta-analyses were not possible, Cochrane recommendations were followed to synthesise results using a vote counting technique and effect direction plots which recorded the direction of the effect in relation to the expected relationship with health.

Results/findings: The search yielded 4,826 articles; 32 observational and no intervention studies met the inclusion criteria. Two studies were classified as having a high risk of bias, 23 as moderate risk, and seven as low risk of bias.

In the twelve studies that assessed adolescent exposure to healthy community nutrition environments, results did not show clear associations with dietary outcomes. None of these studies considered food purchasing as an outcome. Twenty-eight studies assessed adolescents’ exposure to unhealthy community nutrition environments with the majority (n=17/28, 61%) reporting results in the expected direction suggesting greater exposure to food outlets classified as unhealthy was associated with poorer food purchases and dietary intakes. Inconsistent results were observed across the five studies investigating associations between the consumer nutrition environment and adolescent dietary behaviours. No studies investigated associations between the consumer nutrition environment and food purchasing.

Conclusions: Evidence suggests increased exposure to unhealthy food outlets is associated with increased purchasing and consumption of less healthy food in adolescents. There is limited research describing associations between exposure to healthy food outlets, the consumer nutrition environment, and independent adolescent food-related behaviours, particularly adolescent food purchasing. Further research in these areas may help to identify interventions and policy strategies to improve the diets of adolescent populations.

Purpose: Adolescence is a time when individuals experience increased independence from household influences, often spending more time outside of the home unsupervised. It is also a period when individuals develop an increasing reliance on, and reduced resistance to, peer influences. Social influences, and physical food environment determinants, of adolescent food choice have been described before, but little is known about the way these factors interrelate. This study aimed to use qualitative methods to understand the interrelations between adolescents’ social environments and physical food environments and the role these play in determining adolescents’ food purchasing and dietary behaviours.

Methods: Exploratory online focus groups were conducted with friendship groups of adolescents aged between 11-18 years attending secondary school or college in England.  Forty-two participants took part in 12 online focus groups. Focus groups were transcribed verbatim and analysed thematically.

Results/findings: Preliminary findings suggest that most adolescents regularly visit food outlets with their friends. Adolescents described their independent food purchasing occasions, particularly when accompanied by friends, as an opportunity to treat themselves and recognised that their food choices in these settings were mostly unhealthy. Many participants described having limited opportunities to make their own food choices, so didn’t want to waste these opportunities buying unenjoyable healthy foods. Going to certain food outlets, such as coffee shops and restaurants, provided a feeling of independence and maturity.

Habits and routines played a major role in determining the outlets visited by friendship groups. Many fast-food outlets were chosen for their convenience, not just in relation to time but also decision making and avoiding embarrassment. Friends often described knowing what their friends would purchase in food outlets. Adolescents identified price as a major influence on their food purchasing decisions, many claiming, however, that food promotions and being with friends resulted in spending more than planned.  

Conclusions: Purchasing unhealthy food is one way adolescents assert their autonomy. Healthy manipulations to food environments may reduce the social desirability of unhealthy foods among adolescents. A greater understanding of how such changes align with other adolescent values is also required to develop effective and acceptable interventions. 

Purpose: Social media platforms are increasingly used to advertise unhealthy foods and beverages (“junk food”) to children. Greater awareness regarding public opinion on restricting such advertising is important toward understanding the message environment and potential interventions. This study describes the extent that U.S. adults do or do not support restricting junk food advertising to children on social media and explores associations with sociodemographic, political, and health-related characteristics.

Methods: We used data from the U.S. National Cancer Institute’s 2020 Health Information National Trends Survey to estimate the prevalence of adult opinions on restricting junk food advertising to children on social media (strongly oppose; oppose; neutral; support; strongly support). Using weighted multivariable logistic regression, we examined associations of self-reported sociodemographic, political viewpoint, and health-related characteristics with “not supporting” such an advertising restriction (dichotomous outcome included neutral, oppose, and strongly oppose categories). 

Results: The analytic sample included n=2852 adults (mean age 46.9 (SE 0.4) years; 50.0% male; 65.8% non-Hispanic White). The largest proportion of adults were neutral about advertising restrictions (40.6%). Support (23.3%) and strong support (22.8%) were more common than opposition (7.3%) and strong opposition (6.1%). The odds of not supporting advertising restrictions (53.9%) were 1.4 to 2.6 times higher among non-Hispanic Black (vs non-Hispanic White) adults, those without a college degree (vs with a college degree), reporting a moderate or conservative political viewpoint (vs liberal), and having a BMI in the overweight category (vs normal weight); each association statistically significant at p<0.05. Adults with strong weight and diet-related cancer beliefs had 53% lower odds of not supporting junk food advertising restrictions (vs weaker beliefs; OR: 0.47 (95% CI: 0.36-0.61)).

Conclusions: Tailored communication strategies that increase public awareness of the links between weight, diet, and cancer, and the links between advertising and children’s diets—particularly for adults with higher BMIs, non-Hispanic Black adults, and those with lower education—may increase support for restricting junk food advertising to children on social media. Such advertising restrictions could improve children’s food environments to prevent diet-related diseases.

Background: The Systems Thinking Approach for Retail Transformation (START) map uses systems mapping to guide planning and evaluation of healthy food retail interventions in community-based health-promoting settings. The START map captures the complexity of retail settings and identifies potential healthy food initiatives. However, it is yet to be applied to non-health promoting settings, such as grocery stores. This study aimed to test the feasibility and utility of the START map in grocery retail settings.

Methods: We analysed sixteen semi-structured interviews with grocery retail employees and health promotion practitioners involved in the implementation of grocery healthy food retail initiatives in regional Australia. Interviews were deductively coded against existing START map factors and inductive coding identified new factors that influence implementation of healthy retail in grocery store settings. The Implementation narratives were identified and the START map was updated to reflect these data.

Results: Multiple existing factors identified in the original START map were replicated in the grocery store interview data. Four new factors (competitiveness of market, complexity of organisational structure, number of food retail environments making healthy changes and total profit) and two new implementation narratives were also identified. New narratives demonstrates that healthy retail interventions in grocery settings need to consider: 1) the complexity of grocery store organisational structures, market competition and commercial concerns; and 2) the wider impact that grocery stores could have on changing community nutrition and influencing community expectations of healthy food retail cultures.

Significance: The START map is applicable informing the design, implementation and evaluation of healthy food retail policies in to grocery store settings. 

Purpose: Low-income communities throughout the U.S. continue to experience limited access to affordable healthy foods. Nutrition incentive programs exist to provide low-income families a monetary resource to make healthy food more affordable and accessible. Many of these programs target participants of federal nutrition assistance programs (e.g. SNAP, WIC) and operate at farmers’ markets. This study aimed to use geo-spatial analysis to evaluate availability of a nutrition incentive program in Chicago, IL (Link Match) to determine if nutritionally at-risk communities have adequate access.

Methods: Link Match is the largest nutrition incentive program in Illinois; it provides SNAP recipients a one-to-one dollar match (up to $25) if they redeem their benefits at a participating retailer. We obtained 2018 spatial data on census tract-level socio-demographic characteristics and Link Match locations in Chicago, IL from a variety of sources including the City of Chicago, U.S. Census Bureau, and the Environmental Protection Agency. We found 57 retailers (e.g., farmers markets, food cooperatives) that offered Link Match across the city’s 801 census tracts. We examined spatial lag and ordinary least squares (OLS) regression models to identify tract-level measures associated with distance (in miles) from the nearest Link Match retailer. Measures of interest included % non-Hispanic Black, % Hispanic, median household income, violent crime rate, per capita grocery store availability, and walk score.

Results: Most of the retailers that offered Link Match were located on Chicago’s south and west sides. OLS regression models indicated that census tracts with a higher walk score or median household income below the city’s median in 2018 were on average closer in distance to a Link Match retailer (both p<0.001). However, census tracts in the highest quartile of violent crime rate were also significantly closer to a Link Match retailer (p<0.001). After accounting for spatial dependency of census tracts, only violent crime rate was significantly associated with distance to the nearest Link Match retailer.

Conclusions: Link Match retailers in Chicago, IL appear to be located in areas of need with large populations of nutritionally at-risk families. However, these areas have high violent crime rates, which may deter program usage.

Purpose: Characteristics of the urban environment may contain upstream drivers of the growing rates of obesity. However, research is lacking that considers the combination of environmental factors simultaneously, and what statistical methods are suitable to cope with the multitude of data at individual-level and environmental-level. We therefore applied various machine learning approaches to identify what environmental exposures of the urban exposome are consistently associated with Body Mass Index (BMI) across these approaches, and to reflect on the feasibility/interpretability of these methods.

Methods: A cross-sectional analysis was carried out using the baseline data of 14,829 participants from Occupational and Environmental Health Cohort (AMIGO) study. Self-reported height and weight (used to calculate the BMI) and lifestyle factors were assessed via questionnaires. Overall, 96 environmental exposures, including air pollution, traffic noise, green-space, and other built environmental neighborhood characteristics were estimated and linked to geocoded home addresses. We identified the most consistent exposure-obesity associations across the following approaches: a) Two dimension-reduction methods: sparse group Partial Least Squares, Weighted Quintile Sum, b) Three variable selection approaches: Bayesian Model Averaging, Minimax Concave Penalty, Generalized Additive Model boost and c) One method from grouping of observations: Random Forest. The models were adjusted for relevant socio-demographic variables. The exposures were ranked according to the variable importance scores attributed by each approach, and an overall ranking across all approaches was calculated.

Results: Four neighborhood characteristics were among the top 5 most consistent associations related to BMI: the density of healthy food outlets in the neighborhood, the average value of the houses, the share of people with the lowest registered personal income and the share of divorced people in the neighborhood. Interestingly, a specific air pollutant (i.e.,oxidative potential (OP) of the particulate matter) was also consistently associated with BMI. The median rank of OP across multiple approaches was 4/93 (93 exposures).

Conclusions: This study provides a comprehensive investigation of obesogenic environment. It strengthens the evidence for an association of characteristics of neighborhoods and air pollution exposure with adult obesity.