Table 2.
Overview of included studies reporting on associations between aspects of the food environment and diet across different socioeconomic groups—by type of food environmental factor studied.
| Measures | ||||||||
|---|---|---|---|---|---|---|---|---|
| Author (Year) Country | N | Age Group | Study Design | Study Focus | Food Environment | Dietary Outcome | Indicator of SEP | Summary of Findings |
| Economic aspects of the food environment | ||||||||
| Beydoun et al. (2008) USA [44] | 7331 | Adults aged 20–65 yrs | CS | Association: Food prices and dietary intake | Price of FF and of F & V based on geocode data of commonly consumed foods | aMED score, HEI score, F & V intake, FF consumption based on 24 h recalls | Poverty-Income-Ratio | Higher FF prices were associated with higher diet quality in all groups. Higher F & V prices were also associated with higher diet quality, and this associated was mostly present among the poor subgroup. |
| Colchero et al. (2015) Mexico [31] | 2006: 19,512; 2008: 27,994; 2010: 25,805 households | Households | Repeated CS | Own and cross price elasticities for soft drinks and SSBs | Price elasticity of SSBs (prices derived from household expenditures using methodology to produce national consumer price index) | Proportion of household expenditures on SSB with respect to total expenditures on food and drinks | Income | SSB and soft drink demand in low income groups was reduced more by higher prices than in high income groups. |
| Meyer et al. (2014) USA [33] | 5115 | Adults aged 18–30 yrs | LG: Prospective follow-up 5x over 20 yrs | Association: Price changes and FF consumption | Food prices based on Consumer price data from the Council for Community and Economic Research | FF consumption based on a limited number of questions | Education and income | Larger decrease in FF consumption per unit change in price for those with relatively less education or with lower income. |
| Nakamura et al. (2015) UK [45] | 26,986 households | Adults (48.60 ± 15.84 yrs) | CS | Association: price promotions and food purchases | Promotional pricing of food categories (healthy and less-healthy) from 11 supermarket chains | Sales of healthier and less-healthy versions of foods based on transaction records from a household panel | Household SEP | Higher SEP groups were more responsive than lower SEP groups to promotions for foods, most notably for promotions on healthier foods. |
| Powell et al. (2009) USA [46] | 3739 | Adults aged 18–23 yrs | CS | Association: Food prices and F & V intake | Local area food price data. Price index for F & V; meat, dairy & bread; food at home (F & V, meat, dairy, bread); FF Based on American Chamber of commerce price data | 2 questions on F & V intake | Educational level, parental education, income | Lower income, lower educated young adults and those with lower educated mothers and lower income parents were more likely to eat fewer F & V when food prices were higher. |
| Powell et al. (2011) USA [61] | 1134 | Adolescents aged 12–18 yrs | CS | Association: Food prices and availability of food stores and food consumption patterns | Two food-related price indices: a food at home grocery index and away-from-home FF index. Availability food stores and restaurants. Based on American Chamber of commerce price data | Number of days in the last week that F & V, fruit juice, meat, nonmeat protein, dairy, grains, and sweets were consumed, based on an audio computer-assisted self-interview | Maternal education, working status, family income | Among low income adolescents: higher FF prices were associated with a higher number of days nonmeat protein consumption. Increased supermarket availability was associated with higher frequency vegetable intake. FF restaurant availability was not significantly associated with any of the food consumption patterns. |
| Objectively measured access, proximity, quality of food environment | ||||||||
| Burgoine et al. (2016) UK [26] | 5958 | Adults aged 29–62 yrs | CS | Association: Fast food outlet exposure and fast food consumption: moderation by educational attainment | Counts of fast food outlets within 1-mile Euclidean buffers around the home and work location | Consumption of energy-dense foods typically obtained from fast food outlets (e.g., pizza, burgers, chips, fried fish, fried chicken) based on a semi-quantitative food frequency questionnaire | Educational attainment | Greater fast food outlet exposure was associated with greater fast food consumption. The difference in fast food consumption between those with lowest and highest education level was strongest in those most exposed to fast food outlets. |
| Chrisinger et al. (2018) USA [47] | 4962 | Adults aged 18+ yrs | CS | Association: Trip, store and shopper characteristics with trip HEI scores. | Types of food shops used (conventional supermarket, discount/limited assortment store, natural/gourmet store), distance from shopper’s home to full-service supermarket | HEI-2010 score and consumption of multiple food groups based on reported food purchases at household level | Income/SNAP eligibility, educational level | Shopping in conventional supermarket or natural/gourmet store was associated with higher HEI scores. Spending less money was associated with lower HEI scores. Distance travelled from home was not associated with HEI. Non-SNAP eligible households had higher HEI scores when shopping at convenient supermarkets and discount/limited assortment stores than SNAP households and households that were SNAP eligible but not receiving. |
| Duran et al. (2014) Brazil [48] | 1842 | Adults (36.5 ± 11 yrs) | CS | Association: Local retail food environment and consumption of F & V and SSB | Proximity and density of supermarkets and fresh produce markets within 1.6 km buffer from participants’ homes | Consumption of F & V and SSB based on a short number of questions | Income | In neighborhoods with a low density of supermarkets and fresh produce markets, low income individuals had a significantly lower F & V intake than high income individuals. This association disappeared in neighborhoods with a larger number of supermarkets and fresh produce. |
| Gustafson et al. (2017) USA [69] | 2936 households (primary food shoppers) | Households | CS | Association: Neighborhood food store availability and primary food store choice; and primary food store choice and types of food purchases | Availability of food venues within 1 mile of the home: (1) Supermarkets (sells primarily foods); (2) supercenters (food + significant amount other items); (3) convenience stores; (4) combination grocery stores (food + prepared food items + other); (5) medium and large grocery stores | Food purchases of (1) SSB and (2) low-calorie beverages and water based on scanned barcodes on food products; saved store receipts; and information written in a food book | SNAP households or SNAP-eligible households (185% of poverty threshold) | Having supermarkets and supercenters nearby was associated with shopping in supermarkets and supercenters, respectively, but only in SNAP households. Only in non-SNAP households, having grocery stores nearby was associated with shopping there. Shopping at supercenters or convenience stores was associated with higher odds of purchasing SSB. Shopping at supercenters was associated with higher odds of purchasing water/low calorie beverages in both SNAP and non-SNAP households. Shopping at grocery stores was only associated with higher odds of purchasing SSB in SNAP households. Shopping at supermarkets was only associated with higher odds of purchasing water/low calorie beverages in SNAP households. |
| Jack et al. (2013) USA [49] | 15,634 | Adults aged 18+ yrs | CS | Association: Density of food outlets and F & V consumption | Density of healthy food outlets and access to healthy food outlets based on zip codes | F & V intake based on short number of questions | Low and high poverty zip-codes | The density of healthy food outlets did not predict consumption of fruits or vegetables in the total sample, the low poverty sample and the high poverty sample. |
| Macdonald et al. (2011) Scotland [70] | 1149 | Adolescents and adults aged 16+ yrs | CS | Association: Proximity to food retail stores and dietary patterns | Proximity to general stores, F & V stores and supermarkets using GIS data in 500 m and 1 km buffers | Intake of F & V and high fat snacks based on limited number of items in survey | Car ownership and employment | Few significant associations between proximity to food outlets and F & V intake or high fat snacks intake were observed. The borderline significant association between living near a supermarket and not eating F & V regularly was not different between employed and unemployed adults, but did differ between those with and without a car. That is, those with a car had borderline significant higher odds of consuming F & V regularly when a supermarket was present within 1 km. |
| McInerney et al. (2016) Canada [50] | 446 | Adults aged 21+ yrs | CS | Association: Neighborhood food environment and diet quality | Objective measures of food destination presence, density and diversity within walkshed of 400 m from participants’ homes | Canadian adapted Healthy Eating Index (C-HEI) based on FFQ data | Education and income | A higher the number of food destinations within 400 m of home, regardless of type, was associated with higher C-HEI scores. No statistically significant interactions between walkshed food environment variables and socioeconomic status in relation to the C-HEI. |
| Pearce et al. (2008) New Zealand [71] | 12,529 | Adolescents and adults aged 15+ yrs | CS | Association: Neighborhood accessibility to supermarkets and convenience stores and F & V consumption | Access to supermarkets and convenience stores based on travel time along the road network using GIS | Eating recommended F & V levels based on limited number of items in survey | Education, social class, employment and income | No association was observed between neighborhood access to supermarkets or convenience stores and the consumption of F & V. Better access to convenience stores was associated with lower vegetable consumption. None of the interaction effects between access to convenience stores and any of the socioeconomic variables were significant. |
| Rummo et al. (2015) USA [34] | 3299 | Adults (25.0 ± 3.6 yrs) | LG | Association: Neighborhood convenience stores and diet quality | Convenience store relative to total food outlets based on a 3 km buffer around participants’ homes | A priori diet quality score; beneficial foods (whole grains, F & V); adverse foods (SSB, ASB, salty snacks, processed meats, desserts) based on an FFQ | Individual-level income | A higher proportion of convenience stores relative to total food stores/restaurants was associated with lower diet quality scores and this association was stronger among low income participants. For specific food groups; only whole grain consumption was negatively associated with the % neighborhood convenience stores relative to total food stores/restaurants, and this association was also stronger among low income participants. |
| Vogel et al. (2016) UK [25] | 829 | Adults (31.78 ± 6 yrs) | CS | Association: In-store supermarket environment and maternal dietary quality | Composite score representing the healthfulness of the in-store supermarket environment | Prudent dietary pattern score based on a 20-item FFQ | Educational attainment | A strong positive relationship between dietary quality and store healthfulness was observed among low educated mothers, but no significant association among mid educated mothers; and poorer store healthfulness was associated with better dietary quality among high educated mothers. |
| Vogel et al. (2017) UK [51] | 838 | Adults (31.78 ± 6 yrs) | CS | Association: Overall food environment and maternal dietary quality | The balance between healthy and unhealthy food stores in 1000 m buffers using GIS data | Prudent dietary pattern score based on a 20-item FFQ | Educational attainment | Poorer food environments were associated with higher diet quality scores among high educated mothers and (non-significant) lower diet quality scores among low educated mothers. |
| Zenk et al. (2009) USA [52] | 919 | Adults (46.28 ± 0.84 yrs) | CS | Association: Residential neighborhood retail food environment and F & V intake | Observed F & V availability, variety, quality, affordability within ½ mile Euclidean distance from the center of a residential block | F & V intake using an FFQ | Education, income, employment | There was no evidence that individual sociodemographic characteristics moderated the relationship between the neighborhood food environment and F & V intake. |
| School food environment | ||||||||
| Azeredo et al. (2016) Brazil [62] | 109,104 | Adolescents aged 11+ yrs | CS | Association: Food environment in public and private schools and in the immediate surroundings and the consumption of unhealthy food | Availability of healthy/unhealthy foods in school cafeteria or nearby school, reported by school principal. Provision of Brazilian school food program in public schools | Consumption of soft drinks, deep fried salty snacks, bagged salty snacks and sweets based on a validated questionnaire | Public vs. private schools | The presence of cafeteria selling fruit was negatively associated with the consumption of salty snacks in private schools only. Other differences were not statistically significant. Eating foods from the school food programme was associated with lower purchasing of unhealthy foods, but only in public schools. |
| Longacre et al. (2014) USA [63] | 1542 | Adolescents (14.4 ± 1.04 yrs) | CS | Association: F & V intake while school was in session (exposed to school food) and when school was not in session (not exposed to school food) | Exposure to school food based on timing of survey (summer months vs. school year) | F & V intake based on a 2-item measure from the Youth Risk Behavior Surveillance System | Household income | Among adolescents unexposed to school food, household income and F & V intake was positively associated. Among adolescents exposed to school food, F & V intake was similar across income categories. Interaction analysis indicated that adolescents in the lowest income category had higher F & V intake if they obtained school food, and adolescents in the higher income category had lower F & V intake if they obtained school food. The results indicate that exposure to school food mitigates income-related disparities in adolescent F & V intake, and that this mitigation is beneficial for low-income students. |
| Vericker et al. (2013) USA [64] | 5530 | Adolescents | CS | Association: Competitive food and beverage availability in school and F & V and SSB intake | Foods and beverages offered at school that compete with the National School Lunch Program | F & V and SSB intake based on food frequency questionnaires | Family poverty status | Competitive food access was not associated with F & V intake and SSB intake. Only adolescents from families with incomes below the poverty line had lower F & V consumption if they lost access to competitive foods. |
| Virtanen et al. (2015) Finland [65] | 23,182 | Adolescents (15.4 ± 0.63 yrs) | CS | Association: Proximity to FF outlets and grocery stores to school and eating habits | Distance to a food outlet 100, 100–500 and >500 m from school entrance | Skipping free school lunch, obtaining snacks outside of school based on an unknown number of survey items | Parental education | A FF outlet or grocery store close to school was associated with irregular eating habits, but with an accumulation of irregular eating behavior in low-SEP adolescents only. Proximity to a food outlet was associated with higher odds of skipping school lunch in high SEP adolescents. |
| Perceived food environment | ||||||||
| Ho et al. (2009) Hong Kong [66] | 34,369 | Adolescents (14.5 ± 0.11 yrs) | CS | Association: Perceived availability of food stores and intake of F & V, SSB and junk foods | Perceived availability of FF shops, restaurants and convenience stores within 5 min walking distance from home | Intake of F & V, high fat foods and junk food/SSB based on four questions on frequency of consumption | Perceived family affluence | Perceived availability of FF shops, restaurants, and convenience stores were associated with unhealthy dietary intakes. This was stronger in boys from less affluent families. |
CS = cross-sectional. FF = fast food. FFQ = food frequency questionnaire. F & V = fruit and vegetables. Hr = hour. Km = kilometer. LG = longitudinal. SSB = sugar sweetened beverage. Yrs = years.