Exploring the Relationship between the Fast Food Environment and Obesity Rates in the US vs. Abroad: A Systematic Review

Obesity remains one of the greatest public health concerns in our modern-day society, necessitating an understanding of the underlying contributing factors. This systematic review examines the existing literature to better understand the relationship between the fast food environment and obesity rates within the United States (US) compared to other countries throughout the world. To be included in our review, studies must have been peer-reviewed, published in English, and had to include some measure of analysis of the fast food environment and rates of obesity. Based on the results of our review (n=46 studies), the findings were largely similar between US and non-US studies; in both cases, there were inconsistent associations between the fast food environment and rates of obesity/overweight. However, in terms of socioeconomic status (SES), our findings were consistent across both US and non-US studies; lower SES was associated with unfavourable fast food environments, higher concentrations of fast food restaurants, higher consumption of junk food, and higher obesity rates. Based on the results of our review, we conclude that more longitudinal research must be performed with consistent methodology in order to more clearly understand the role of the fast food environment in the development of obesity. *Corresponding author: Antonio Gangemi, Division of General, Minimally Invasive & Robotic Surgery, University of Illinois at Chicago, Illinois, USA, Tel: +1.312.355.1493, Fax: +1.312.355.1987; E-mail: agangemii@gmail.com Received January 18 2018; Accepted January 24, 2018; Published February


Introduction
In modern day society, the prevalence of childhood and adult obesity remains one of the greatest public health concerns, especially in the United States (US). Obesity is officially defined by the World Health Organization as a Body Mass Index (BMI) of 30 and above [1]. An estimated 36.5% of US adults and 17.0% of US youth aged 2-19 are affected by obesity. Both youth and adult obesity rates have increased in the US in recent years [2]. Similar concerns regarding increasing rates of obesity exist in other countries throughout the world, particularly those considered developed [3]. Obesity can lead to the development of a myriad of adverse health conditions in affected individuals; for example, obesity has been linked to serious, complex medical conditions such as diabetes, sleep apnea, and cancer [4]. One of the particular concerns is the link between obesity and the prevalence of abnormal levels of cardio metabolic variables, which may eventually lead to the development of cardiovascular disorders [5]. With the wide range of severe health concerns associated with obesity, an understanding of the factors contributing to the development of obesity is essential.
In recent years, there has been a greater focus on environmental factors potentially contributing to the development of obesity. There has been growing consensus that environmental factors play a prevalent role in determining body weight, leading to a shift in focus from individual to population level issues. The presence of fast food restaurants is one particular environmental factor of interest. An increase in the consumption of fast food has been associated with poorer diet quality and higher energy intake at the individual level [6][7][8]. However, previously conducted reviews examining the associations between the fast food environment and obesity have largely found inconsistent associations at the population level [9,10]. In addition, no previous study has examined associations between fast food restaurants and rates of obesity across various settings, such as across countries. Building on previously conducted reviews, this review sought to explicitly examine the methodology used in studies and to compare the associations between the fast food environment, socioeconomic factors, and obesity in the US and other countries throughout the world where literature was present to better understand the cross-cultural impact of the fast food environment. To our knowledge, no such review directly comparing study methodology and the associations between the fast food environments and rates of obesity within and outside the US has previously been conducted. This topic of interest has a variety of potential policy and land-use implications and is particularly relevant as increasing attention has been brought to food environments as a potential contributor to the development of obesity [11][12][13][14][15][16][17][18].

Materials and Methods
A review of the existing literature was performed to identify peer-reviewed works studying the relationship between the fast food environment and obesity. Searches for this review were performed in the online databases Pubmed (Medline) and Scopus for studies published between January 1 st , 2005 and December 31 st , 2015. Based on an examination of previously conducted review articles, the key words included in the online search were "fast food, " "fast food outlet, " "fast food restaurant, " "fast food environment, " "quick service restaurant, " "concentration of fast food restaurants, " "obesity, " "overweight, " "BMI, " "rate of obesity, " and "global obesity. " Several (Put exact number) duplicate studies from Pubmed (Medline) and Scopus were found in the initial search. To be included in our review, studies must have been published in English. Case studies and expert opinion published works were excluded. Studies that met our inclusion criteria were then further filtered based on their abstract; the abstract must have discussed a measure of the surrounding food environment involving fast food restaurants and a measure of rates of overweight/obesity [19]. Finally, the full papers of studies that passed the abstract criteria were reviewed one more time before inclusion to confirm the presence of the desired inclusion criteria. Screening was performed in accordance with the PRISMA statement.

Study designs and statistical analyses utilized by included studies
The vast majority (n=41, 89%) of the included studies were crosssectional in design; only five [14,16,21,30] of the included studies were longitudinal. Every included study performed some sort of regression or correlation analysis to study the relationship between the fast food environment and rates of obesity in a certain study population; a large portion (n=25, 54%) of included studies also utilized a multi-level/ multivariable/weighted/hierarchical regression analysis.

Populations studied and outcomes measured in included studies
Included studies differed in terms of the populations they examined. Some studies (n=17, 37%) explicitly stated that they only studied children (under the age of 18). Other studies (n=18, 39%) explicitly stated that they only studied adults (over the age of 18); the adults included in these studies ranged from young adults to pregnant mothers to adults over the age of 50. Finally, the remainder of studies either did not mention the specific ages of their study populations or studied some combination of both children and adults. The included studies that examined children often studied the effects of fast food environment around schools (elementary school and/or high school). A total of five studies in the US studied the food environment specifically around schools, a total of three studies outside the US studied the food environment around schools, and one study [30][31][32][33][34][35] compared the effects of the food environment around schools in US, Canada, and Scotland. Included studies were relatively similar in terms of the outcomes they measured; almost all the included studies obtained height and weight measures to calculate the BMI of the study population and used the BMIs to find the rate of obesity. Only one study [36][37][38][39][40][41] did not include a measurement for obesity/overweight rates; this study examined pregnant women and their infants, and examined the probability of "excessive weight gain" during pregnancy, defined in the study as weight gain of over 40 pounds in the expecting mother. Several studies studied outcomes in addition to BMI, such as dietary behaviour and psychosocial profiles. Although nearly all the included  -Accounted for age and gender of students.
-BMI was calculated from the height and weight of each child -Accounted for SES through a deprivation score.
-Significant association between higher density of fast food outlet and the child being obese.
-No significant association between proximity to fast food outlet and being overweight/obese. -Children living in areas of higher deprivation have higher densities of fast food outlets in their SOA.  -After adjusting for gender, ethnicity, and IDACI, no significant associations between number of food outlets and sBMI in the child household and commute environments (with the exception of supermarkets, which had negative association with sBMI) -Some significant associations between exposure to outlets and obesity, but were borderline/ negative.
-No evidence of a significant association between distance to the nearest food outlet and sBMI.
-Significant positive relationship between number of food outlets and SES at school and household level. -Number of convenient stores, grocery stores, supermarkets, pizza parlors, and sit-down restaurants.

Location of Food
-Self-reported data on sex, weight, height, and date of birth were collected from surveys. -BMI calculated from selfreported data.
-Accounted for levels of poverty and household median income.
-Neither the proximity of fast food stores near schools nor the number of food stores significantly predicted weight outcomes.
-No correlation between unhealthful food choices and risk of overweight/obese. -Accounted for SES through percentage of students receiving free/ reduced priced meals.
-Nearby fast food restaurants and supermarkets were not associated with overweight rates.
-Presence of a convenience store within the 800 m network buffer, however, was associated with a higher rate of overweight students than schools without a nearby convenient store. -Accounted for race, income, and education level as covariates.
-Found no statistically significant association between the proportion of obese 3-4-year-old children and the relative availability of certain outlets based on the RFEI. -However, there is a statistically significant positive association between number of convenience stores and obesity. -Calculated restaurant density as number of fast food restaurants per 10,000 individuals -Measured density of full-service restaurants.
-Self-reported heights and weights used to calculate the BMI of individuals.
-Accounted for a variety of SES factors, such as race, household income, percentage of adults with a high school diploma.
-Fast food restaurant density and a higher ratio of fast food to full-service restaurants is associated with higher individual weight status, while a higher density of full-service restaurants is associated with lower weight status.
-Area-level restaurant can potentially be associated with risks of obesity. -Measured presence of supermarkets, convenience stores, full-service restaurants.

Neighborhood
-Self-Reported heights and weights used to calculate BMI.
-Incorporated Race/Ethnicity, Education, and Employment status.
-After adjustment for individual-level effects, higher prevalence of obesity was observed in areas with more than one franchised fast food restaurant.
-higher prevalence of obesity in areas of convenience stores and at least one independent grocery store. -Did not study effect of SES -Higher ratio of fast food restaurants to full-service restaurants is associated with obesity most strongly in areas with modern urbanicity. -Significant association found between built environment variables and physical inactivity/obesity when controlling for individual and neighborhood characteristics/ attributes. studies used BMI to measure the outcomes of the population, the included studies were split in terms of how they obtained height and weight measurements. A total of 29 studies (63%) obtained height and weight data of participants in a standardized manner (for example, height and weight measurements were taken during a clinic visit by trained professionals). A total of 17 studies (37%) obtained height and weight data from self-reported measurements.

Fast food access studies categorized by geographic location and scale
A majority of studies included in this review were performed in urban locations or were performed in both urban and rural locations. Only one study was performed solely in a rural location; this study was conducted in rural China [21]. The scale of the included studies varied significantly, ranging from studies performed on entire countries [20] to studies examining Metropolitan Statistical Areas (MSA) to studies examining individual cities and neighbourhoods [42][43][44][45][46][47]. The most commonly used scale in the included studies were the county level (or a similar measure, such as MSA) and the city level.

Characterization of the fast food environment
The included studies also differed significantly in the way the fast food environment was characterized. A majority (n=21, 62%) of the studies performed in the US utilized a similar definition for fast food based on Standard Industrial Classification (SIC) codes, the North American Industrial Classification System (NAICS), or the US Census Bureau's County Business Patterns (CBP) [48][49][50][51]. Two included studies did not explicitly mention a definition for fast food restaurants [20,43,[52][53][54]. The remainder of studies in the US (n=13, 38%) utilized their own definition of fast food restaurants, often loosely based on standards defined by local public health departments. These studies categorized restaurants as fast food restaurants based on characteristics such as the type of utensils offered, whether a customer receives their food before or after they pay, and the type of foods served. In addition, several included studies used popular franchised fast food restaurants, such as McDonald's, Burger King, Subway, KFC, and Taco Bell, as a proxy for the fast food environment. Studies performed outside the US displayed an even greater heterogeneity in the definitions of fast food utilized. A significant portion (n=4, 40%) of studies performed outside the US utilized their own definitions of fast food; similar to the studies performed in the US, these definitions were based on the manner of service, the type of food served, and the use of franchised Western fast food restaurants as proxies to represent the fast food environment. The remainder (n=6, 60%) of studies performed outside the US utilized a definition of fast food based on local public health departments or public data sources, such as Larsen 2014, which defined fast food based on Toronto Healthy Environments Inspection System (THEIS). -Studied effects of persistent poverty present vs. no persistent poverty present -No Significant association between prevalence of fast food restaurants per capita and preschool childhood obesity rates in rural communities.
-In More urbanized communities, relationship between prevalence of fast food restaurants and childhood obesity rate is significant.
-Relationship between availability of fast food restaurants is significant when persistent poverty and high urbanization are present.   Table 2: Summary of associations between the fast food environment and rates of obesity in us studies.

Studies with at least one significant association between the fast food environment and higher rates of obesity/overweight
Studies reporting no significant associations or only indirect associations between the fast food environment and higher rates of obesity/overweight N=4 N=6 To describe the fast food environment, several different methodologies were used; the most commonly utilized methods included densities, proximities, ratios, and indexes used to characterize the food environment. The majority of studies (n=31, 67%) utilized a measure of the density of fast food restaurants. There was some variability, however, within the density measurements; some studies calculated the number of fast food restaurants within a circular buffer of fixed distances around a reference point (such as a residential location or school), while other studies calculated the relative prevalence of fast food restaurants by finding the number of fast food restaurants per fixed unit of population. The next most common measure (n=6, 13%) of describing the food environment utilized was a proximity measure; studies utilizing a proximity measure often measured the distance from a reference point (such as a residential location or a school) to the nearest fast food restaurants. Some studies (n=4, 9%) used indices to characterize the food environment; a commonly utilized measure was some form of the Retail Food Environment Index (RFEI). Finally, only two studies utilized a ratio measure to characterize the food environment; these studies characterized the food environment based on ratios of fast food restaurants to other types of restaurants. A total of three studies [25,26,50] characterized the fast food environment based on both density and proximity measures.
Several studies further described the food environment by incorporating measures of food outlets other than fast food restaurants. A majority of studies (n=33, 72%) included other food outlets in addition to fast food restaurants in their studies [51][52][53]. The most commonly included food outlets were full-service restaurants, convenience stores, grocery stores, supermarkets, outdoor food stalls, and cafes. Several studies [16,49,50,52,54,55] also examined the built environment beyond solely food, studying factors like street connectivity, number of bicycle paths, and the number of subway stations.

Associations found between the fast food environment and obesity: US Studies
One of the 34 included studies that were performed within the US, 20 studies (59%) found at least one significant association between the fast food environment and higher rates of obesity/ overweight or excessive weight gain [41]. Generally, these significant associations were strongest in metropolitan/urban areas with higher rates of deprivation. The associations found were varied in nature. Xu et al. [56] found strongest associations between the fast food restaurant and high rates of obesity in North-eastern US. Xu et al. [55] found the strongest associations in areas with higher urban city measures. Newman et al. [57] found associations between the fast food environment and higher rates of obesity only in urban areas and areas with persistent poverty, but not in rural areas. Ahern et al. [12] found that the significant relationship between a more prevalent fast food restaurant and higher rates of obesity only existed in metropolitan areas. Inagami et al. [35] found the strongest associations for residents who don't have access to cars. Oreskovic et al. [50] found the strongest associations for individuals of the Lowest Income Quartile (LIQ). The remaining 14 studies (41%) found no significant association or only indirect associations between the fast food environment and rates of obesity/overweight. Once again, the conclusions presented by the studies that found no significant association were varied in nature; for example, Hickson et al. [31] found no association between the availability of fast food and BMI, but the study found that greater availability of fast food was significantly associated with higher energy intake in certain study populations. Finally, although some of the studies found no significant relationship between fast food availability and BMI, they found significant relationships between the availability of other food outlets and BMI; for example, Koleilat et al. and Langelier [36,38] both found a significant association between the presence of convenience/corner stores and higher BMIs, but found no significant relationship with regards to the availability of fast food restaurants. In terms of study design, four (12%) longitudinal studies were performed in the US, while 30 (88%) cross-sectional studies were performed in the US. A total of three (75%) longitudinal studies performed in the US found no significant association or only indirect associations between the fast food environment and rates of obesity/overweight. Only one (25%) longitudinal study performed in the US found at least one significant association between the fast food environment and higher rates of obesity/overweight. Amongst the 30 cross-sectional studies performed in the US, 11 (37%) studies found no significant association or only indirect associations between the fast food environment and rates of obesity/overweight; 19 (63%) studies found at least one significant association between the fast food environment and higher rates of obesity/overweight. Finally, of the 5 studies performed studying US schools, three (60%) studies found no significant association or only indirect associations between the fast food environment and rates of obesity/overweight, while two (40%) found at least one significant association between the fast food environment and higher rates of obesity/overweight.

Associations found between the fast food environment and obesity: Studies performed outside the US
Amongst the 10 included studies that were performed in countries outside the US, a total of four (40%) studies found at least one significant association between the fast food environment and higher rates of obesity/overweight. Similar to the studies performed within the US, the strongest associations were found in urban/metropolitan areas. Also, similar to the studies within the US, the associations found were varied in nature. For example, Fraser and Edwards [25], found that within the UK, a significant positive association between fast food availability and obesity rates only when using density measures (no significant association found while using proximity measures). Chiang et al. [18], a study performed in Taiwan, only found a significant positive association existed in boys, but not girls. A total of six (60%) studies reported no significant association or only indirect associations between the fast food environment and rates of obesity/overweight. Interestingly in the final study performed in rural China [21], a significant negative relationship was found between the fast food environment and BMI. No such relationship was found elsewhere. Similar to the US studies, although some of the studies found no significant relationship between fast food availability and BMI, they found significant relationships between the availability of other food outlets and BMI. For example, Larsen et al. [39], performed in Canada, found no association between the fast food environment, but found a significant negative association between "healthy food outlets" and obesity.
In terms of study design, only one (10%) study was longitudinal, while the remaining nine (90%) studies were cross-sectional. Du et al [21], a longitudinal study performed in rural China, found no significant association between the fast food environment and BMI for men, and found a significant negative association between the fast food environment and BMI for women (not found elsewhere). Amongst the remaining nine cross-sectional studies, four (44%) studies found at least one significant association between the fast food environment and higher rates of obesity/overweight, while five (56%) studies found no significant association or only indirect associations between the fast food environment and rates of obesity/overweight. Finally, of the three studies performed studying the fast food environment surrounding schools, two (67%) studies found no significant association between the fast food environment and rates of obesity/overweight, while one (33%) study found at least one significant association between the fast food environment and higher rates of obesity/overweight.

Associations found between the fast food environment and SES: US studies
A majority of included studies conducted in the US incorporated various measures of socioeconomic status (SES). In many of the studies, SES measures served as controls to accurately study the fast food environment and its relationship to obesity/overweight rates without confounding variables. However, 21 (62%) studies further examined the relationship between SES measures and the fast food environment. All of these studies demonstrated similar findings: lower SES was associated with a higher density of fast food restaurants and convenience stores, an unfavourable retail environment, poorer food environments, higher consumption of junk foods, and higher obesity rates. Higher SES was associated with a "healthier" food environment and lower obesity rates. In addition, when a significant positive association was found between the fast food environment and rates of obesity/overweight, the association was often strongest in areas with low SES (persistent poverty, higher levels of deprivation, etc.). Examples of such studies include (24,54,57). Interestingly, the SES findings were consistent across US studies that found no significant relationship between the fast food environment and rates of obesity/overweight and studies that found at least one significant association between the fast food environment and higher rates of obesity/overweight. For example, although Laxy et al. [40] found no direct associations between the fast food environment and obesity, this study found that neighbourhood economic hardship was associated with an unfavourable retail environment.

Associations found between the fast food environment and SES: Studies performed outside the US.
Similarly, a majority of included studies conducted outside the US incorporated various measures of SES, primarily as a controlled variable. However, five (50%) studies further examined the relationship between SES measures and the fast food environment. The findings were similar to the findings of the studies performed in the US: lower SES was associated with a larger number of food outlets, higher rates of obesity, higher consumption of junk food, higher fat intakes, and higher densities of fast food outlets. Higher SES was associated with lower obesity rates and lower concentrations of fast food restaurants. Similar to US studies, these findings were present across all the studies that included SES, even those that found no direct associations between the fast food environment and obesity; for example, Fraser et al. [26] found no consistent associations between the fast food environment and rates of obesity, but found that higher levels of deprivation were associated with greater numbers of food outlets.

Discussion
Many factors contribute to the development of obesity, including lifestyle factors, genetics, consumption of "specific foods and beverages, physical activity, alcohol use, television watching, and smoking habits" [58]. This review sought to specifically compare the relationship between the fast food environment and obesity in the US and other countries throughout the world. Associations found between the fast food environment and obesity: US vs. abroad.

Limitations
Amongst the studies conducted in the US included in our review, 59% found at least one significant association between the fast food environment and higher rates of obesity/overweight, or excessive weight gain, while 41% found no significant association or only indirect associations between the fast food environment and rates of obesity/overweight. Of the included studies conducted outside the US, 40% found at least one significant association between the fast food environment and higher rates of obesity/overweight, while 60% found no significant association or only indirect associations between the fast food environment and rates of obesity/overweight. The studies performed in the US and countries outside the US both demonstrated largely mixed results in terms of the relationship between the fast food environment and obesity. In both the US and non-US studies, when associations were present between the fast food environment and rates of obesity/overweight, the associations were strongest in urban areas and populations with lower SES. Thus, the fast food environment may be associated with higher rates of obesity in areas with lower SES and higher concentrations of ethnic minorities in both the US and countries other than the US. However, as a whole, similar to a review performed by Fleischhacker et al. [59], we find no consistent associations between the fast food environment and measures of obesity/overweight. There are several limitations to our approach. First, the vast majority (89%) of included studies, both US and non-US, were cross-sectional in design. The cross-sectional nature of the included studies in both the US and non-US categories greatly limit our ability to draw definitive conclusions about the relationship between the fast-food environment and rates of obesity/overweight. In addition, there was a significant amount of heterogeneity in various essential components of the included studies, such as the study population, scale of the study, methods of measuring the fast food environment (density, proximity), assessment of other food outlets (Examples: convenience stores, supermarkets), and assessment of BMIs in the study population (selfreported vs. standardized exams). The majority of included studies were performed in metropolitan areas; more studies must be performed in rural areas, in both the US and outside the US, to fully understand the role that the fast food environment. Also, a majority of included studies examined various types of food outlets in addition to fast food, such as convenience stores, sit-down restaurants, and supermarkets. As Fleishhacker et al. [59] points out, it may be necessary to strike a balance between solely examining the associations between the fast food environment and obesity along with examining fast food outlets as a part of the total food environment. Finally, in the non-US category, there were only two non -"Western" countries included: Taiwan and China. More studies must be performed in non -"Western" countries to fully understand the role that the fast food environment plays as a potential risk factor for the development of obesity across cultures. There is a wide range of countries included in the non-US category, complicating the interpretation and comparison of associations found between the fast food environment and rates of obesity in US vs. non-US category. Overall, the vast differences present in methodology across included studies performed in both the US and non-US categories strongly limit our ability to draw conclusions about the association between the fast food environment and rates of overweight/obesity.

Associations found between SES and Fast Food Environment: US vs. Abroad
In the included studies in both the US and non-US groups, there were strong associations between socioeconomic status (SES) and the fast food environment. In nearly all of the included US and non- US studies that examined SES, it was found that areas with lower SES were associated with unfavourable fast food environments, with higher concentrations of fast food restaurants, higher consumption of junk food, and higher obesity rates when compared to areas with higher SES. While there were inconsistent results in the association between the fast food environment and rates of obesity/overweight, lower SES was strongly associated with higher fast food densities and closer proximity to fast food restaurants, especially in urban areas. These findings were consistent across both US and non-US studies. Our findings are in agreement with the review performed by Fleischhacker et al. [59] and several recent studies [60,61] that have affirmed the presence of an unfavourable food environment and higher obesity risks in areas with lower incomes and SES. Several explanations have been posed for this phenomenon, including disproportionate marketing towards children in majority black communities/rural areas/middle-income communities along with other economic factors and social processes [59,62]. Based on the results of our study, the association between low SES and an unfavourable fast food environment is consistently present in both the US and non-US studies. Once again, there were several limitations to our approach. Included studies used a wide range of conditions to define SES, including education, race, personal income, levels of poverty, and median neighbourhood income. As discussed previously, there were also great variations in the measures used to study the fast food environment. Future studies examining SES and the fast food environment must use more comprehensive, uniform methods to define SES and the fast food environment.

Policy Implications in the US
The food environment has come under increasing scrutiny as a potential factor for the development of obesity, and there have been several proposed policy initiatives specifically targeting the fast food environment in the US. Based on the mixed nature of the associations found between the fast food environment and rates of obesity overall in addition to the primarily cross-sectional design of the studies, we find it difficult to justify policy initiatives targeting the fast food environment as a corrective measure to reduce rates of obesity/overweight based on the current literature. More longitudinal studies must be performed with uniform definitions of the fast food environment, study populations, measures of access, and the inclusion of other food outlets in order to more accurately devise policy initiatives that may or may not target the fast food environment. Our conclusions are in agreement with the review performed by Fleischhacker et al. [59].
Relevant to our discussion of policy is the "Los Angeles Fast-Food Ban" introduced in 2007. The ordinance passed by the city of Los Angeles (LA), US is a zoning regulation intending to reduce the prevalence of fast food restaurants in south LA by preventing the expansion or opening of stand-alone fast food restaurants [11]. A study by Sturm and Hattori examined the effects of the so-called fast food ban, and found that there was no significant change in the composition of the food environment, with similar rates of new fast food restaurants in south LA compared to the rest of LA [11]. Overall, the study concluded that the Los Angeles fast-food ban has been ineffective so far, due in large part to the limited focus of the ordinance and the possible limited relationship between the fast food environment and obesity/diet [11].
With the inconsistent associations found between the fast food environment and rates of obesity/overweight in our review along with the limited positive outcomes of initiatives such as the LA fast-food ban, we propose alternative policy measures. While the prevalence of obesity remains high in youth and adult populations in the US, recent research has shown that in the youth populations across the US, the prevalence of obesity among adolescents age 12-17 of high socioeconomic-status has actually begun to decrease, while the prevalence of obesity among adolescents of lower socioeconomic status has continued to increase [63]. Frederick et al. suggests that interventions targeted towards lower SES youth will be particularly effective in helping to alleviate the burden of obesity in the US [63]. Targeted initiatives promoting increased physical activity, increased educational attainment, and healthier lifestyles in lower SES populations will be particularly important in decreasing the obesity epidemic in the US [60,63].

Conclusions
Our review sought to compare the relationship between the fast food environment and rates of obesity/overweight in the US and countries outside the US. A total of 46 studies were included in our final review. Based on the results of our review, the findings were largely similar between US and non-US studies; in both groups, there were inconsistent associations between the fast food environment and rates of obesity/overweight. However, in terms of SES, our findings were consistent across both US and non-US studies; lower SES was consistently associated with unfavourable fast food environments, with higher concentrations of fast food restaurants, higher consumption of junk food, and higher obesity rates. The large variance in methodology and study design across both US and non-US studies potentially limits the conclusions of our study. More longitudinal research must be performed with consistent methodology in order to better understand the role of the fast food environment in the development of obesity. Until then, no definite conclusions regarding the role of the fast food environment in the development of obesity can be drawn.