Associations of Neighborhood Food Retail Environments with Weight Status in a Regional Pediatric Health System

Abstract

Background:

There have been mixed findings on the relationships between childhood obesity and macroscale retail food environments. The current study investigates associations of the neighborhood retail food environment with changes in children’s weight status over 6 years in the Kansas City Metropolitan area.

Methods:

Anthropometrics and home addresses were collected during routine well-child visits in a large pediatric hospital (n = 4493; >75% were Black or Latinx children). Children had measures collected during two time periods ([Time 1] 2012–2014, [Time 2] 2017–2019). Establishment-level food environment data were used to determine the number of four types of food outlets within a 0.5-mile buffer from the children’s residence: supermarkets/large grocery stores, convenience stores/small grocery stores, limited-service restaurants, and full-service restaurants. Children who moved residences between periods were “movers” (n = 1052). Associations of baseline and changes in food environment status with Time 2 weight status were assessed using mixed-effects models.

Results:

Movers who experienced no change in the number of convenience stores or small grocery stores within a 0.5-mile of their home had increased likelihoods of having overweight/obesity and less favorable BMIz changes, compared with movers who experienced a decrease in convenience stores/small grocery stores within a 0.5-mile distance. No associations were observed among nonmovers.

Conclusion:

Findings suggest that moving to an area with fewer unhealthy retail food outlets (e.g., convenience stores) is associated with a lower risk of obesity in children. Future research is needed to determine whether larger-scale changes to the retail food environment within a neighborhood can support children’s healthy weight.

Introduction

Childhood obesity is a significant health issue in the United States, adversely impacting both the physical and mental well-being of children.^1,2 Recent statistics indicate that ∼21% of children aged 6–19 had obesity during 2017–2020.³ The food retail environment is increasingly recognized for its role in shaping dietary behaviors and health outcomes.^4–6 The accessibility, variety, affordability, and quality of food options in neighborhood food outlets, along with the role of programs such as the Supplemental Nutrition Assistance Program (SNAP) and school meal program, can influence dietary choices and nutrition, thereby impacting the prevalence of childhood obesity.^4,5,7

A growing number of studies have focused on how the food retail environment impacts children’s diet and weight status.⁸ There have been mixed findings on the relationships between childhood obesity and macroscale retail food environments (i.e., proximity and access to different types of food outlets).^4–6,8–14 The mixed results may be attributed to differences in study methodologies, including variations in study design, measurement tools, and study sample. For example, supermarkets and large grocery stores, known for providing high-quality healthy foods (e.g., fresh produce, milk, cheese, whole grains, and lean proteins) at relatively lower prices, have been extensively researched for their associations with obesity and other health outcomes in most cross-sectional studies.^4–6,8,9,15 A review by Larson et al. highlighted that greater accessibility to larger grocery stores and supermarkets is associated with lower obesity rates in adults and children.⁶ However, some studies have failed to show significant associations between supermarket availability and children’s weight status.¹⁶ Studies examining the associations between childhood obesity and access to convenience stores and fast-food outlets that mainly provided unhealthy food (e.g., high-calorie, low nutrient density chips, candies, and processed foods) reported both positive^10–12 and no associations.¹³ Most research shows no apparent association between access to full-service restaurants and obesity risk among children.¹⁴ Longitudinal studies on this issue are limited and show conflicting results. For example, research conducted in Los Angeles County observed a nonlinear association of weight status at age 3 with density of healthy food outlets.¹⁷ A nationally representative study among school-age children in the United States found that the proximity to supermarkets had a protective effect against developing obesity, whereas the proximity of small-sized grocery stores and limited-service restaurants increased obesity risk among girls.¹⁸ Additionally, another study identified a correlation between a greater presence of convenience stores in neighborhoods and a significant increase in body mass index (BMI) for age over 3 years in 7-year-old girls.¹⁹ Given inconsistencies in prior evidence, more large-scale prospective studies are needed to provide stronger evidence for informing interventions and policy.

Some studies deliberately select areas undergoing significant neighborhood environmental changes to capture natural experiments on neighborhood environments and health.²⁰ Outside of these types of studies, environmental changes in neighborhoods over a period of a few years are often minimal. Examining children who move residences offers a unique opportunity to explore the health impacts of experienced environmental changes. Research has shown that moving can significantly influence children’s weight status by altering their exposure to various environmental factors.²¹ For example, changes in neighborhood characteristics, such as better physical activity resources and safety, have been linked to changes in healthier weight outcomes in children.²¹ These children may also experience various changes in their new neighborhoods, ranging from reduced, similar, or increased access to healthy or unhealthy food locations in the retail food environment.

The Kansas City regional area presents a complex food environment. While some areas have ample grocery stores, others rely more on convenience stores with limited healthy options, highlighting the need for targeted public health efforts to ensure equitable food access across the region.²² The purpose of the current study was to investigate associations of the neighborhood retail food environments with changes in children’s weight status among movers and nonmovers over 6 years in the Kansas City Metro area. The study aims to test the hypothesis that baseline exposure and changes in exposure to neighborhood retail food environments are associated with changes in children’s weight status, particularly among movers due to the potential to experience larger environmental changes.

Materials and Methods

Participants

The study involved 4493 children who visited a Children’s Mercy pediatric primary care clinic for a Well Child visit during two 3-year time periods: 2012–2014 (Time 1) and 2017–2019 (Time 2). The 2-year gap between the time periods allowed for the investigation of larger changes in weight status that can occur over longer periods of time. The Children’s Mercy pediatric primary care clinics are a safety net health provider. In 2017–2019, over 79.9% of the patients served had public health insurance, and 7.5% were uninsured. To be included in the analysis, children had to be between 6 and 15 years of age at Time 1, reside within the six primary counties of the Kansas City metropolitan area (Cass County, MO; Clay County, MO; Jackson County, MO; Platte County, MO; Johnson County, KS; and Wyandotte County, KS), and have no more than two different addresses during the study periods. The participants were categorized as “movers” or “nonmovers.” Nonmovers maintained the same address throughout the study period, whereas movers had a different address during each time period and no more than two addresses across the study period. Movers with only 1 clinic visit while at their new residence were excluded from the analysis because the length of time at their new address was unknown and could have been very brief (i.e., a few weeks at the new address were not posited to have measurable associations with weight status). See Appendix 1 for the detailed information for sample selection. The groups of movers and nonmovers consisted of 1052 and 3441 children, respectively, spread across 256 and 379 census tracts. The total combined number of unique census tracts included in the study was 405. Notably, 36.9% of movers stayed within the same census tract over the study period. A sensitivity analysis was conducted to assess the impact of the subgroup of children who moved to the different census tracts and found the significant findings remain consistent. The study was approved by the Children’s Mercy Institutional Review Board.

Measures

Sociodemographic characteristics and addresses. Sociodemographic information (i.e., age, sex, race/ethnicity (categorized as non-Hispanic White, non-Hispanic Black, Hispanic, other race/ethnicity, and unknown), and health insurance type (private, government/public [e.g., Medicaid], or none)) was extracted from the electronic health record (EHR) at Time 1. The home addresses of the children, also sourced from the EHR, were geocoded and matched to census tracts utilizing the Python GeoPandas library (0.12.2) for spatial data processing and analysis.

Weight status. Children’s height and weight were measured objectively by clinic staff and recorded in the EHR. Using these measurements, BMI percentiles and z-scores (BMIz) were calculated, taking into account each child’s age and sex.^23,24 The classification of overweight was defined as a BMI at or above the 85th percentile but below the 95th percentile. Obesity was classified as a BMI at or above the 95th percentile. In cases where children had multiple clinic visits or records within a time period, data from the initial visit during Time 1 and the final visit during Time 2 were used for analysis to maximize the ability to capture potential changes in weight status over time.

Neighborhood retail food environment. Following previous research,^25–28 we categorized food retail establishment data from the National Establishment Time Series (NETS) database²⁹ into four categories using the following North American Industry Classification System codes: supermarket/large grocery stores (445110), convenience stores/small grocery stores (445120), full-service restaurants (722511), and limited-service restaurants (722513). A detailed description of retail formats is provided in Appendix 2. The NETS data used in this study align closely with our study timeline of 2012–2014 and 2017–2019. Home addresses, road networks, and establishment-level food environment data during each time were used to determine the number of four types of food outlets within a 0.5-mile road-network buffer from the children’s residences. Food outlets within a 0.5-mile buffer were categorized based on the presence (Yes) or absence (No) of a certain type of outlet. We created a three-category change variable to capture the changes in the number of food stores within the 0.5-mile buffer at two time ranges. The 0.5-mile distance is commonly used in urban food environment studies as it captures the typical area within which families and children are likely to access food outlets on foot.^30,31 The categories were defined based on the comparison between the number of food outlets at two time ranges: decrease (T2 < T1), no change (T2 = T1), and increase (T2 > T1).

Neighborhood sociodemographic characteristics. Time 1 census tract sociodemographic information was obtained from the 2010–2014 American Community Survey 5-year estimates and included population density; total land area; median annual household income; and percentage of non-Hispanic White, non-Hispanic Black, and Hispanic/Latinx residents.

Analysis

Descriptive statistics were computed for all study variables. Mixed-effects regression models, accounting for the nesting of children within census tracts, were conducted to model the association of Time 1 values and change values for the neighborhood environment variable with children’s weight status at Time 2. All four types of food outlets were included concurrently in the same model. We tested the interaction between the baseline and Time 2 values, but there was no significant interaction effect. All models adjusted for Time 1 BMIz, child sociodemographics, and neighborhood sociodemographics. BMIz was tested in linear models and grouped as overweight or obese (vs. healthy weight) in logistic models. Mover status was first explored as a moderator of the aforementioned associations using multiplicative interaction. Due to consistent evidence of moderation across most models, all models were tested separately among movers and nonmovers.

Results

The characteristics of the study participants and neighborhood census tracts are detailed in Table 1. Approximately half of the children were female. In terms of racial/ethnic composition, 34.4% of nonmovers were non-Hispanic Black, compared with 54.8% among movers. Hispanic children constituted 48.5% of the nonmovers and 30.4% of the movers.

Table 1.

Sample Characteristics at Time 1

	Nonmovers (n = 3441)	Movers (n = 1052)
Child characteristics
Age ranges, years	6–15	6–15
Mean age (SD)	8.70 (2.29)	8.74 (2.21)
% female	48.7%	54.9%
Race/ethnicity
% Non-Hispanic White	11.3%	8.8%
% Non-Hispanic Black	34.4%	54.8%
% Hispanic/Latinx	48.5%	30.4%
% Other race/ethnicity	5.8%	6.0%
Insurance type
% With commercial health insurance	14.9%	9.5%
% With government/public health insurance	79.7%	84.5%
% With no health insurance/unknown	5.4%	6.0%
Census tract characteristics
Mean (SD) number of residents	2806.1 (1813.8)	3091.1 (1792.4)
Mean (SD) area, square miles?	4.6 (13.4)	2.6 (6.5)
Mean (SD) % Non-Hispanic White	61.7 (28.9)%	53.3 (29.5)%
Mean (SD) % Non-Hispanic Black	21.3 (25.3)%	27.1 (28.6)%
Mean (SD) % Hispanic	11.6 (13.4)%	13.7 (15.2)%
Mean (SD) median household income	$55,373.7 ($29,659.9)	$46,412.8 ($21,805.3)
% Rural	7.96%	5.49%

SD, standard deviation.

Table 2 provides information on children’s weight and the characteristics of their neighborhood food environments at both Time 1 and Time 2. About 19% of children in both groups were overweight during each time period, and a slightly higher proportion had obesity. The proportion of children with obesity was higher at Time 2, with a more pronounced increase among the movers. The availability of food outlets within a 0.5-mile buffer, among nonmovers and movers, is shown in Table 2 at both time periods. Table 3 presents the changes in neighborhood food environments.

Table 2.

Child Weight and Neighborhood Food Environment Characteristics at Time 1 and Time 2

	Time 1		Time 2
	Nonmovers (n = 3441)	Movers (n = 1052)	Nonmovers (n = 3441)	Movers (n = 1052)
Child weight
Mean (SD) BMIz	0.72 (1.11)	0.81 (1.05)	0.85 (1.14)	0.93 (1.07)
Mean (SD) BMI %ile	68.95 (28.45)	71.72 (26.44)	71.81 (28.55)	74.3 (26.51)
% Overweight	18.7%	19.0%	18.8%	19.0%
% Obese	22.9%	24.1%	28.4%	29.7%
Neighborhood food environment
% Having food location(s) within 0.5-mile buffer
Supermarket/large grocery stores	35.3%	36.2%	34.1%	37.6%
Convenience store/small grocery stores	47.1%	43.8%	43.8%	40.7%
Full-service restaurant	54.2%	51.9%	58.9%	56.9%
Limited-service restaurant	32.9%	31.8%	32.7%	36.6%

SD, standard deviation.

Table 3.

Changes in Neighborhood Food Environments at 0.5-Mile Buffer

	Nonmovers (n = 3411)	Movers (n = 1052)
	% Within in each category	% Within in each category
Supermarket/large grocery stores
Decrease	13.5	14.2
No change	74.5	71.6
Increase	12.0	14.3
Convenience store/small grocery stores
Decrease	18.5	19.4
No change	72.3	72.0
Increase	9.3	8.7
Full-service restaurant
Decrease	15.4	13.4
No change	56.4	58.4
Increase	28.2	28.3
Limited-service restaurant
Decrease	8.6	7.1
No change	82.1	80.2
Increase	9.3	12.7

As shown in Table 4, there were no significant associations between the neighborhood food environment variables at Time 1 and children’s weight status at Time 2. For children who move residences, significant associations were observed between changes in neighborhood food environment variables and children’s weight status, specifically for convenience stores/small grocery stores. Specifically, movers who experienced no change in the number of convenience stores or small grocery stores between time periods had an increased likelihood of having overweight or obesity (odds ratio = 3.561, 95% confidence interval [CI] = [1.694,7.487], p = 001), along with less favorable BMIz changes (B = 0.220, 95%CI = [0.066,0.375], p < 0.05), in comparison with movers who experienced a decrease in convenience stores/small grocery stores. No significant associations were observed between changes in neighborhood food environments and weight status among nonmovers. Appendix 3 provides a detailed breakdown of children’s weight status at Time 2 across each food location category.

Table 4.

Association Between Neighborhood Food Environments Within 0.5 Mile and Weight Status at Time 2

	BMIz				Overweight or obese vs. healthy weight
	Nonmovers		Movers		Nonmovers		Movers
	B (95% CI)	p	B (95% CI)	p	OR (95% CI)	p	OR (95% CI)	p
Time 1 neighborhood food environment^a
Supermarkets/large grocery stores	0.009 (−0.044, 0.062)	0.734	0.078 (−0.020, 0.176)	0.121	1.020 (0.790, 1.317)	0.881	1.067 (0.681, 1.672)	0.776
Convenience stores/small grocery stores	−0.079 (−0.227, 0.069)	0.297	−0.045 (−0.141, 0.051)	0.360	0.881 (0.688, 1.128)	0.314	0.821 (0.529, 1.276)	0.380
Full-service restaurants	0.011 (−0.041, 0.064)	0.675	−0.026 (−0.124, 0.072)	0.601	1.011 (0.788, 1.297)	0.930	1.028 (0.658, 1.607)	0.903
Limited-service restaurants	0.008 (−0.043, 0.059)	0.755	−0.040 (−0.136, 0.056)	0.410	1.075 (0.844, 1.369)	0.558	1.105 (0.711, 1.716)	0.657
Changes in food environment^b
Supermarkets/large grocery stores
No change	0.013 (−0.069, 0.094)	0.761	−0.024 (−0.185, 0.137)	0.771	1.074 (0.726, 1.588)	0.721	0.681 (0.334, 1.390)	0.291
Increase	−0.019 (−0.138, 0.101)	0.762	0.204 (−0.016, 0.425)	0.069	0.996 (0.561, 1.769)	0.989	1.010 (0.382, 2.668)	0.984
Convenience stores/small grocery stores
No change	−0.075 (−0.155, 0.005)	0.066	0.220^** (0.066, 0.375)	0.005	0.793 (0.539, 1.168)	0.241	3.561^** (1.694, 7.487)	0.001
Increase	−0.088 (−0.224, 0.047)	0.201	0.035 (−0.224, 0.295)	0.790	0.894 (0.467, 1.711)	0.735	1.577 (1.47, 5.245)	0.458
Full-service restaurants
No change	0.016 (−0.087, 0.119)	0.761	0.035 (−0.146, 0.216)	0.705	1.068 (0.651, 1.753)	0.795	0.912 (0.409, 2.034)	0.822
Increase	0.009 (−0.118, 0.136)	0.887	−0.043 (−0.270, 0.183)	0.708	1.057 (0.573, 1.952)	0.858	0.782 (0.282, 2.168)	0.636
Limited-service restaurants
No change	0.033 (−0.080, 0.147)	0.567	−0.057 (−0.301, 0.187)	0.647	1.106 (0.639, 1.911)	0.719	1.236 (0.406, 3.765)	0.710
Increase	−0.083 (−0.235, 0.069)	0.287	−0.024 (−0.308, 0.261)	0.870	0.800 (0.381, 1.678)	0.555	1.399 (0.381, 5.139)	0.613

All models were adjusted for children’s race/ethnicity, age, and insurance type, neighborhood median household income, and neighborhood race/ethnicity.

Reference group is not having certain food store.

Reference group is “Decrease”. Models were adjusted for Time 1 neighborhood food environment.

p < 0.05.

p < 0.01.

OR, odds ratio; CI, confidence interval.

Discussion

In the current study, we examined the prospective relationships between children’s exposure to neighborhood food environments and their weight status over a 6-year period. This analysis examined food outlets within a 0.5-mile walking range and included a predominantly Black and Hispanic cohort of children consisting of both movers and nonmovers. Findings reveal that a reduction in the number of neighborhood convenience stores or small grocery stores within walking distance is potentially beneficial for supporting healthier weight changes in children, though these findings were specific to children who moved residences and had multiple wellness visits during time period 2. Living within 0.5 miles of a supermarket was not associated with children’s weight status. Present findings highlight the need for approaches that address the overabundance of unhealthy food options in many neighborhoods and its influence on childhood obesity.

The present finding that a move resulting in a decrease in the number of convenience stores/small grocery stores may support healthier weight changes is consistent with prior research. For example, a previous longitudinal study found positive associations between the number of small grocery stores and obesity risk among girls.¹⁸ This suggests that close physical proximity to food outlets that often sell a proportionally high amount of unhealthy foods may influence dietary choices and, consequently, weight outcomes. Children experiencing a decrease in physical proximity to convenience stores/small grocery stores might adapt by consuming fewer unhealthy snacks,³² purchasing healthier foods that are more typically available at supermarkets, and/or preparing more meals at home. Furthermore, close proximity of convenience stores at walking distance may be more associated with children’s eating behaviors than with those of adults, as children are more likely to walk or bike to these locations due to their inability to drive.³³ This highlights the need for future exploration into how the presence or absence of nearby food outlets influences the dietary preferences of different age groups, especially in areas lacking a nearby grocery store. However, the current study did not find a significant association between the increase in the number of convenience stores or small grocery stores and weight status. One possibility is that the impact of an increase in the number of these stores may be mitigated by other factors in the new environment that promote healthier behaviors. For example, families may increase physical activity in response to the new neighborhood amenities.²⁰

This study adds to prior research by comparing movers and nonmovers. The beneficial association for a reduction in convenience stores/small grocery stores was only observed among movers, which suggests nonmovers may not benefit from a reduction in these food outlets. However, this difference between movers and nonmovers may be attributed to differences in the scale of changes in the food environment between these groups and the quality of food access within these stores. Additionally, some potential confounders (e.g., neighborhood safety, cultural factors, and access to recreational facilities), not measured in the current study, may contribute to the differences observed between movers and nonmovers. Movers naturally face changes more directly because of their relocation, whereas nonmovers would require substantial modifications within their existing neighborhoods. It is possible that more robust changes to the food retail environment within a neighborhood may produce similar associations with weight status as observed among movers, though such changes are difficult to facilitate and thus uncommon. Improving the availability and affordability of healthy food options within convenience stores/small grocery stores has been shown to be feasible in prior studies.³⁴ Related approaches could involve promoting the availability of healthy foods at convenience stores/small grocery stores or supporting local farmers’ markets to enhance physical proximity to fresh produce for nearby residents.

Prior studies on associations of supermarkets/large grocery stores, limited-service restaurants, and full-service restaurants with children’s weight changes have yielded mixed results.^16,17,35 A previous longitudinal analysis conducted with a national representative sample found that poorer physical access to supermarkets and greater physical access to limited-service restaurants were associated with greater odds of having obesity for girls 3 years later.¹⁸ In the present study, we extended this research by examining both baseline and changes in physical proximity to large grocery stores and supermarkets, limited-service restaurants, and full-service restaurants. Contrary to several studies,^18,36 we did not observe a significant association between either baseline or changes in physical proximity to these food outlets and children’s weight changes. This suggests that the relationship between the availability of different food retail outlets and weight outcomes is intricate and likely influenced by factors beyond close physical access to the food outlet. Research indicates that merely increasing the number of healthy food retailers in these areas does not necessarily lead to improved dietary behaviors or weight outcomes.³⁷ Prior evidence demonstrates that food choices are influenced by a variety of other environmental, socioeconomic, and behavioral factors. For example, the affordability and aggressive marketing of unhealthy options often overshadow healthier alternatives, which may be more expensive and require additional time or knowledge to prepare.³⁷ This complexity is further compounded in studies of the family food environment; for instance, increased frequency of consuming dinner while watching television was linked to higher BMIz scores longitudinally.³⁸ Thus, it is important to consider the microscale environment within food outlets and homes in addition to the macroscale environment based on the locations of food outlets.³⁹

The main strength of this study is its longitudinal design spanning a 6-year period, utilizing individualized measurements (i.e., a 0.5-mile walking distance) and including a predominantly Black and Hispanic cohort of children, consisting of both movers and nonmovers. It is important to highlight the methodological advancements of our study compared with previous longitudinal research, which predominantly relied on aggregate data to assess the neighborhood food retail environment, such as census tract and ZIP code levels.^15,17,18,36 This traditional approach, while useful for broad analyses, masks the nuanced variations with neighborhoods, failing to accurately capture the true accessibility and proximity of food options for individuals. By employing individual measures, our study offers insights that are potentially more actionable for interventions aimed at improving access to healthy food options.

The present study also has limitations. First, children’s weight status can be influenced by various factors beyond the scope of the current study such as the home and school environment, along with other neighborhood environment factors. Our study faced limitations in tracking the duration for which movers resided at their new addresses. Despite excluding participants with only one well-child visit at their new location, the average duration at new addresses was likely between 1.5 and 3 years. Our reliance on medical records has constraints as well. For example, the absence of data on children’s diets restricted our exploration of the behavioral mechanisms to connect neighborhood environmental characteristics with childhood obesity. Further, the selection process in the current study may bias our results toward more stable families who seek wellness visits, potentially affecting generalizability due to socioeconomic factors. The unknown reasons behind the families’ decisions to move introduce a potential selection bias that could have influenced the observed associations. We also acknowledge the limitation that the assumption of a monotonic relationship between food environment exposure and child BMI may not fully capture the variations. Further research is needed to understand the impact of food access on neighborhood selection and its implications for the generalizability of observational studies. Moreover, the secondary data source we used to measure the local food environment did not fully capture the nuances of the associations identified, due in part to an inability to account for the variety of products offered by stores. Therefore, future research should aim to enhance the consistency in data collection, geocoding, editing, and analysis of secondary data sources. Additionally, it is important to note that food retail data providers may implement classification schemes inconsistently.⁴⁰ We did not assess the quality or pricing of the food, which could also impact dietary behaviors and weight outcomes. While this study provides valuable insights into the association between food environment changes and child anthropometric outcomes, several important neighborhood contextual variables (e.g., neighborhood walkability, greenspace, and crime) were not included. These factors could confound the observed associations. Future research should aim to include these variables to provide a more comprehensive understanding of the neighborhood environment’s impact on child obesity risk.

Conclusions

The findings of this study offer some support for the hypothesis that reduced access to convenience stores and small grocery stores, which frequently offer unhealthy food options, is linked to healthier weight changes. This highlights the importance of systemic and structural interventions to reshape the food environment. Given challenges in reducing access to food outlets, public health efforts should focus on increasing access to affordable healthy foods in all types of food outlets, particularly convenience stores and small grocery stores. Continued efforts are also needed to increase the appeal and affordability of healthy foods, such as through farmers markets. Additionally, strategies such as implementing soda taxes, modifying SNAP-eligible items to exclude unhealthy options, and changing the retail environment to discourage unhealthy purchases are important. Community-wide initiatives that address access, appeal, and affordability of healthy foods could be instrumental in reshaping food environments to foster and sustain healthy eating practices among children and their families and combating dietary-related health disparities.

Footnotes

Impact Statement

This study reveals that, among the sample population, moving to neighborhoods with fewer unhealthy retail food outlets, such as convenience stores, is associated with a lower risk of obesity in children. These findings point out the potential of improvement in the retail food environment to support healthier weight changes among children, emphasizing a strategy for mitigating childhood obesity.

Acknowledgements

The authors express their gratitude to the anonymous reviewers and Alice Sindzingre for their valuable comments and suggestions.

Authors’ Contributions

Q.J. led all phases of the study, including conceptualization and article preparation. J.C., S.S., and B.F. provided input on data analyses and reviewed and edited the article. L.F. and C.S. contributed to the data collection and data preparation and reviewed and edited the article. H.H.L., S.H., and A.D. reviewed and edited the article.

Ethics Approval and Consent to Participate

The study was approved by the Children’s Mercy Institutional Review Board.

Availability of Data and Materials

The data supporting the findings of this study are available from the author, Carlson, upon reasonable request.

Author Disclosure Statement

The authors declare that they have no competing interests.

Funding Information

This project was funded by Kemper Foundation.

Appendix 1

Appendix 2

Table A2.

Definitions of Retail Food Environment

Category	NAICS code	Definition
Supermarkets/large grocery stores	445110	The industry comprises establishments generally known as supermarkets and grocery stores primarily engaged in retailing a general line of food, such as canned and frozen foods; fresh fruits and vegetables; and fresh and prepared meats, fish and poultry
Convenience stores/small grocery stores	445120 and those with employees four or less in NAICS code 445110 (27, 41)	This industry comprises establishments known as convenience stores or food marts primarily engaged in retailing a limited line of goods that generally includes milk, bread, soda, and snacks
Full-service restaurants	722511	This industry comprises establishments primarily engaged in providing food services to patrons who order and are served while seated (i.e., waiter/waitress service) and pay after eating.
Limited-service restaurants	722513	This industry comprises establishments primarily engaged in providing food services where patrons generally order or select items and pay before eating. Food and drink may be consumed on premises, taken out or delivered to the customer’s location

Appendix 3

Table A3.

Association Across Each Food Location Category Within a 0.5-Mile Buffer and Weight Status at Time 2

	BMIz				Overweight or obese
	Non movers		Movers		Non movers		Movers
	Mean (SE)	p	Mean (SE)	p	Proportion	p	Proportion	p
Time 1 neighborhood food environment^a
Supermarket		0.734		0.121		0.881		0.776
No	0.842 (0.014)		0.901 (0.026)		47.1		48.3
Yes	0.851 (0.020)		0.979 (0.037)		47.3		49.2
Convenience store		0.297		0.360		0.314		0.380
No	0.864 (0.031)		0.949 (0.029)		48.0		49.8
Yes	0.785 (0.048)		0.904 (0.034)		46.3		47.2
Full-service restaurant		0.675		0.601		0.930		0.903
No	0.839 (0.018)		0.943 (0.032)		47.1		48.5
Yes	0.850 (0.016)		0.917 (0.031)		47.2		48.9
Limited-service restaurant		0.755		0.410		0.558		0.657
No	0.843 (0.013)		0.942 (0.025)		46.9		48.2
Yes	0.851 (0.020)		0.902 (0.038)		47.8		49.6
Changes in food environment^b
Supermarket
Decrease	0.835 (0.040)		0.934 (0.080)		46.4		53.2
No change	0.848 (0.011)	0.761	0.911 (0.021)	0.771	47.3	0.721	47.9	0.291
Increase	0.817 (0.042)	0.762	1.139 (0.072)	0.069	46.3	0.989	53.4	0.984
Convenience store
Decrease	0.915 (0.039)		0.731 (0.076)		50.0		33.5
No change	0.840 (0.011)	0.066	0.951 (0.020)^**	0.005	46.9	0.241	50.2^**	0.001
Increase	0.827 (0.054)	0.201	0.766 (0.104)	0.790	48.5	0.735	39.4	0.458
Full-service restaurant
Decrease	0.830 (0.051)		0.904 (0.090)		46.4		50.1
No change	0.846 (0.011)	0.761	0.939 (0.021)	0.705	47.2	0.795	48.8	0.822
Increase	0.840 (0.036)	0.887	0.861 (0.066)	0.708	47.1	0.858	46.7	0.636
Limited-service restaurant
Decrease	0.818 (0.057)		0.983 (0.123)		46.1		45.8
No change	0.851 (0.010)	0.567	0.926 (0.020)	0.647	47.4	0.719	48.6	0.710
Increase	0.735 (0.050)	0.287	0.959 (0.073)	0.870	43.2	0.555	50.3	0.613

Reference group is not having certain food store.

Reference group is decrease.

p values are referenced from Table 4.

p < 0.01.

References

Ayer

, Charakida

, Deanfield

, et al. Lifetime risk: Childhood obesity and cardiovascular risk. Eur Heart J, 2015; 36(22):1371–1376.

Small

, Aplasca

. Child obesity and mental health: A complex interaction. Child Adolesc Psychiatr Clin N Am, 2016; 25(2):269–282.

Stierman

, Afful

, Carroll

, et al. National Health and Nutrition Examination Survey 2017–March 2020 Prepandemic Data Files Development of Files and Prevalence Estimates for Selected Health Outcomes. 2021.

Morland

, Evenson

. Obesity prevalence and the local food environment. Health Place, 2009; 15(2):491–495.

Anderson Steeves

, Martins

, Gittelsohn

. Changing the food environment for obesity prevention: Key gaps and future directions. Curr Obes Rep, 2014; 3(4):451–458.

Larson

, Story

, Nelson

. Neighborhood environments: Disparities in access to healthy foods in the US. Am J Prev Med, 2009; 36(1):74–81.e10.

Mande

, Flaherty

. Supplemental nutrition assistance program as a health intervention. Curr Opin Pediatr, 2023; 35(1):33–38.

Feng

, Glass

, Curriero

, et al. The built environment and obesity: A systematic review of the epidemiologic evidence. Health Place, 2010; 16(2):175–190.

Morland

, Diez Roux

, Wing

. Supermarkets, other food stores, and obesity: The atherosclerosis risk in communities study. Am J Prev Med, 2006; 30(4):333–339.

10.

Lhila

. Does access to fast food lead to super-sized pregnant women and whopper babies? Econ Hum Biol, 2011; 9(4):364–380.

11.

Dunn

, Sharkey

, Horel

. The effect of fast-food availability on fast-food consumption and obesity among rural residents: An analysis by race/ethnicity. Econ Hum Biol, 2012; 10(1):1–13.

12.

Chiang

, Wahlqvist

, Lee

, et al. Fast-food outlets and walkability in school neighbourhoods predict fatness in boys and height in girls: A Taiwanese population study. Public Health Nutr, 2011; 14(9):1601–1609.

13.

, Robinson

, Carter

, et al. Childhood obesity and community food environments in A labama’s B lack B elt region. Child Care Health Dev, 2015; 41(5):668–676.

14.

Jia

, Yang

, Cao

, et al. Association between access to full-service restaurants and childhood obesity. Obes Rev, 2021; 22 (Suppl 1):e13076.

15.

Jiang

, Forseth

, Fitzpatrick

, et al. Prospective associations of neighborhood healthy food access and walkability with weight status in a regional pediatric health system. Int J Behav Nutr Phys Act, 2023; 20(1):113.

16.

Cobb

, Appel

, Franco

, et al. The relationship of the local food environment with obesity: A systematic review of methods, study quality, and results. Obesity (Silver Spring), 2015; 23(7):1331–1344.

17.

Chaparro

, Whaley

, Crespi

, et al. Influences of the neighbourhood food environment on adiposity of low-income preschool-aged children in Los Angeles County: A longitudinal study. J Epidemiol Community Health, 2014; 68(11):1027–1033.

18.

Chen

, Wang

. Changes in the neighborhood food store environment and children’s body mass index at peripuberty in the United States. J Adolesc Health, 2016; 58(1):111–118.

19.

Leung

, Laraia

, Kelly

, et al. The influence of neighborhood food stores on change in young girls’ body mass index. Am J Prev Med, 2011; 41(1):43–51.

20.

MacMillan

, George

, Feng

, et al. Do natural experiments of changes in neighborhood built environment impact physical activity and diet? A systematic review. Int J Environ Res Public Health, 2018; 15(2).

21.

Jones

. Residential instability and obesity over time: The role of the social and built environment. Health Place, 2015; 32:74–82.

22.

Mabachi

, Kimminau

. Leveraging community-academic partnerships to improve healthy food access in an urban, Kansas City, Kansas, community. Prog Community Health Partnersh, 2012; 6(3):279–288.

23.

Kuczmarski

. CDC growth charts: United States: US department of health and human services. Centers for Disease Control and Prevention; 2000.

24.

Centers for Disease Control and Prevention (CDC). The SAS Program for CDC Growth Charts that Includes the Extended BMI Calculations. Centers for Disease Control and Prevention (CDC); 2023.

25.

Cooksey Stowers

, Jiang

, Atoloye

, et al. Racial differences in perceived food swamp and food desert exposure and disparities in self-reported dietary habits. Int J Environ Res Public Health, 2020; 17(19):7143.

26.

Cooksey-Stowers

, Schwartz

, Brownell

. Food swamps predict obesity rates better than food deserts in the United States. Int J Environ Res Public Health, 2017; 14(11):1366.

27.

Jiang

, Ghosh

, Steinbach

, et al. A longitudinal assessment of racial and ethnic inequities in food environment exposure and retail market concentration. Public Health Nutr, 2023; 26(9):1850–1861.

28.

Centers for Disease Control and Prevention. Census Tract Level State Maps of the Modified Retail Food Environment Index (mRFEI). 2013.

29.

Dun and Bradstreet. National establishment time-series database. 2019. Available from: https://www.dnb.com/

30.

Rogus

, Athens

, Cantor

, et al. Measuring micro-level effects of a new supermarket: Do residents within 0.5 mile have improved dietary behaviors? J Acad Nutr Diet, 2018; 118(6):1037–1046.

31.

Ghosh-Dastidar

, Cohen

, Hunter

, et al. Distance to store, food prices, and obesity in urban food deserts. Am J Prev Med, 2014; 47(5):587–595.

32.

Borradaile

, Sherman

, Vander Veur

, et al. Snacking in children: The role of Urban corner stores. Pediatrics, 2009; 124(5):1293–1298.

33.

Villanueva

, Giles-Corti

, Bulsara

, et al. How far do children travel from their homes? Exploring children’s activity spaces in their neighborhood. Health Place, 2012; 18(2):263–273.

34.

Sadler

. Strengthening the core, improving access: Bringing healthy food downtown via a farmers’ market move. Applied Geography, 2016; 67:119–128.

35.

Zhou

, Zhao

, Zhang

, et al. Neighborhood supermarket access and childhood obesity: A systematic review. Obes Rev, 2021; 22 (Suppl 1):e12937.

36.

Daniels

, Schinasi

, Auchincloss

, et al. The built and social neighborhood environment and child obesity: A systematic review of longitudinal studies. Prev Med, 2021; 153:106790.

37.

Ward

, Verity

, Carter

, et al. Food stress in Adelaide: The relationship between low income and the affordability of healthy food. J Environ Public Health, 2013; 2013:968078.

38.

MacFarlane

, Cleland

, Crawford

, et al. Longitudinal examination of the family food environment and weight status among children. Int J Pediatr Obes, 2009; 4(4):343–352.

39.

Lytle

, Myers

. Measures Registry User Guide: Food environment. National Collaborative on Childhood Obesity Research: Washington, DC, USA; 2017.

40.

Forsyth

, Lytle

, Riper

. Finding food: Issues and challenges in using geographic information systems to measure food access. J Transp Land Use, 2010; 3(1):43–65.