Neighborhood Environments and Changes in Obesity and in Lifestyle Behaviors Among Children Enrolled in Obesity Management Interventions: A Systematic Review

ABSTRACT

Introduction

Neighborhood determinants of health have been documented in several populations, yet less is known about their role in pediatric obesity treatment. A systematic review of longitudinal studies examining associations between neighborhood environment features and changes in obesity and in lifestyle behaviors among children participating in obesity management interventions was conducted.

Methods

Searches were conducted in Medline, Embase, CINAHL, and Web of Science for peer‐reviewed articles published in English from database inception until April 2025. We included studies of children with overweight/obesity at baseline, participating in multicomponent obesity management interventions, and with at least one pre‐ and one post‐intervention measurement of obesity or lifestyle behaviors.

Results

Of the 27,310 records screened, six met inclusion criteria. Studies were conducted in the United States (n = 5) and United Kingdom (n = 1), with participants' age ranging from 6 to 18 years, and a total of 13,364 participants. Studies examined availability of parks (n = 3), supermarkets (n = 2), greenspaces (n = 1), walkability (n = 1), recreational facilities (n = 1), and neighborhood deprivation (n = 1). Residing in neighborhoods with more parks was associated with greater reductions in post‐intervention body mass index in two studies. Inconsistent findings relating availability of supermarkets to changes in fruit and vegetable intake were reported. Residing in neighborhoods with more recreational facilities was associated with increases in objectively measured physical activity but not with self‐reported screen time.

Conclusion

Findings among the few studies that examined neighborhood determinants of obesity management outcomes among children were inconsistent. Neighborhood resources that support physical activity (parks, recreational facilities) may be associated with better outcomes.

Abbreviations

BMI

body mass index

EPHPP

effective public health practice project

GIS

geographic information system

GPS

global positioning systems

MVPA

moderate‐to‐vigorous physical activity

PRISMA

preferred reporting items for systematic reviews and meta‐analyses

PROSPERO

prospective register of systematic reviews

RCT

randomized controlled trials

SES

socioeconomic status

SWiM

synthesis without meta‐analysis

1. Introduction

Obesity is a major global public health challenge affecting all age groups, including children and adolescents [1]. An estimated 107 million children and adolescents live with obesity worldwide [2]. This is concerning because obesity is strongly associated with the development of cardiometabolic risk factors such as dyslipidemia, hypertension, and impaired fasting glucose [3] as well as poor mental health, including depression, low self‐esteem, and reduced health‐related quality of life [4]. Obesity treatment and management in children and adolescents aim to improve lifelong lifestyle behaviors and quality of life, manage and treat related physical and psychosocial complications, and prevent the development of chronic diseases [5, 6].

Pediatric obesity management programs, often delivered by multidisciplinary teams, support children, adolescents, and families to improve lifestyle behaviors linked with obesity, notably increasing physical activity, decreasing screen time, eating a healthy diet, and adopting healthy sleep behaviors [7, 8, 9]. Until recently, obesity management programs have largely focused on individual and family determinants of changes in lifestyle behaviors [10, 11]. Although neighborhood environment determinants of lifestyle behaviors and obesity have been extensively studied [12], less is known about whether neighborhood factors can support improvements in lifestyle behaviors and impact obesity management outcomes specifically among children and adolescents enrolled in obesity management programs.

Attributes of neighborhood environments are conceptualized as pertaining to two domains: The built environment comprises all aspects that are constructed by humans (e.g., land use, parks), while the social environment refers to the social processes among neighborhood residents (e.g., safety and violence, neighborhood deprivation) [13]. Several systematic reviews have been published on the associations between neighborhood built and social environment attributes with obesity and lifestyle behaviors in pediatric populations [14, 15, 16, 17, 18]. For example, four systematic reviews have reported that access to a greater number of sports facilities is associated with decreased adiposity, increased physical activity, and decreased sedentary behavior among children and adolescents [15, 16, 17, 18]. Although findings related to the neighborhood food environment are less consistent, a systematic review reported that a higher number of fast‐food restaurants in residential neighborhoods may be associated with more fast‐food consumption among children and adolescents [19]. Among features of the social environment, neighborhood socioeconomic status (SES) has been the most frequently examined in studies of children and adolescents; however, inconsistent associations with PA [14], sedentary behaviors [17], and sleep outcomes [15] were found.

Children and adolescents living with obesity may be more vulnerable to or encounter unique environmental barriers when they are attempting to modify their lifestyle behaviors [11]. For instance, a scoping review highlighted barriers specifically experienced by adolescents with obesity, such as the lack of access to local gyms and sports and recreational centers, and concerns regarding neighborhood safety [11]. Furthermore, emerging evidence suggests that children living with obesity face disparities when initiating obesity management programs and do not respond uniformly to the programs. These differences may reflect the characteristics of their residential neighborhood environments [20, 21, 22]. For example, children residing in neighborhoods with less supportive characteristics (e.g., lower walkability, lower neighborhood‐level SES, and lower park density) tend to enter obesity management programs with higher BMIs [21] and poorer lifestyle behaviors, and may show less favorable responses to obesity management programs [23]. Therefore, synthesizing the evidence on how neighborhood environments relate to obesity management outcomes is critical to inform the design of interventions that are tailored to children and adolescents' neighborhood environments. This knowledge can also guide broader public health initiatives targeting built environments to ensure that interventions are inclusive of children living with chronic conditions including obesity.

Therefore, this study aims to systematically synthesize research on associations between neighborhood environment characteristics and changes in (1) obesity (e.g., BMI z‐score, waist circumference, and waist‐to‐height ratio) and (2) lifestyle behaviors (i.e., physical activity, sedentary behavior, sleep, and diet) among children and adolescents enrolled in multicomponent pediatric obesity management programs.

2. Materials and Methods

The protocol for this systematic review was registered in the prospective register of systematic reviews (PROSPERO) (CRD 42023479725, 11/2023) [23]. This study is reported according to the preferred reporting items for systematic reviews and meta‐analyses (PRISMA) statement [24].

2.1. Search Strategy

Comprehensive searches were conducted in Ovid Medline, Ovid Embase, Ebsco CINAHL, and Web of Science‐Core Collection. Developed in collaboration with a McGill University Health Sciences Librarian (SC), the search strategy combined the following three concepts: children or adolescents, overweight or obesity, and neighborhood environment. Three groups of relevant keywords and subject headings were developed (see Supporting Information S1). The searches were limited to articles published in English and were conducted without any date restrictions. The searches were first conducted on October 30, 2023, and updated on April 24, 2025. Gray literature was not searched for, given that it is generally not indexed in major databases and can therefore be difficult to search in a systematic way [25].

All retrieved records were imported to Covidence, a web‐based collaboration software platform for knowledge syntheses [26]. Duplicates were removed using the automatic removal feature with confirmation by a review author (YT).

2.2. Study Selection

For each stage of the study selection process, the number of records was recorded automatically by Covidence in the PRISMA flow diagram (Figure 1). Study selection was conducted in two steps. In the first step, two independent reviewers (YT and WLT) screened the titles and abstracts of all identified records using pre‐determined inclusion and exclusion criteria (Table 1). Specifically, studies were included if they involved: (1) children and adolescents (aged 2–18 years) with overweight or obesity; (2) a multicomponent (at least two health‐related disciplines involved) obesity management intervention or program; (3) repeated measurement study design (at least one measure pre‐intervention/program and one measure post‐intervention/program with or without a comparison group); (4) self‐report or objective measures of study outcome: obesity outcomes (e.g., BMI, BMI z‐score, waist circumference, waist‐to‐height ratio, and body fat) and lifestyle behaviors (e.g., diet, physical activity, sedentary behavior including screen time, and sleep); (5) at least one residential neighborhood environment exposure measurement; (6) reported estimate(s) of association(s) between neighborhood environment exposure and change in outcome from baseline to follow‐up; and (7) published in English. In the second step, the full texts of potentially relevant studies were obtained and independently reviewed by the same two reviewers. Any discrepancies between the two reviewers in Step 1 were settled by a third reviewer (AVH) who reviewed 50 discrepancies; there were no discrepancies between reviewers in Step 2.

FIGURE 1.

Preferred reporting items for systematic reviews and meta‐analyses (PRISMA) flow diagram.

TABLE 1.

Eligibility criteria.

Inclusion criteria	Exclusion criteria
(1) Children and adolescents (aged 2–18 years) with overweight or obesity at baseline defined as per country specific reference norms	(1) Primarily focused on school neighborhood environment or in‐home environment.
(2) Children participating in multicomponent (at least two health‐related disciplines involved) obesity management intervention/programs that include lifestyle behavior modification interventions (diet, physical activity, and sedentary behavior), mental health, pharmacological, or bariatric surgery for treatment/management of obesity	(2) Study protocols, commentaries, editorials, books, literature reviews, conference proceedings or abstracts, qualitative research articles, and gray literature.
(3) Repeated measurement study design (at least one measure pre‐intervention/program and one measure post‐intervention/program) with or without a comparison group (i.e., quasi‐experiments, randomized controlled trials, and cohort studies)
(4) Self‐report (e.g., questionnaire) or objective (e.g., measured height and weight, accelerometry) measures of study outcome: obesity outcomes (e.g., BMI, BMI z‐score, waist circumference, waist‐to‐height ratio, and body fat), lifestyle behaviors (e.g., diet, physical activity, and sedentary behavior including screen time and sleep)
(5) At least one residential neighborhood environment exposure measurement (e.g., Geographical Information Systems, street audits, and questionnaires)
(6) Reports estimate(s) of association(s) between neighborhood environment exposure and change in outcome from baseline to follow‐up
(7) written in English

The study selection process was reported with the PRISMA flow diagram and included the number of records screened, the number of studies excluded, and the reasons for exclusion after the review of the full texts.

2.3. Assessment of Methodological Quality

The methodological quality of the included studies was assessed using the effective public health practice project (EPHPP) quality assessment tool for quantitative studies [27]. This tool can be used for the critical appraisal of experimental or observational studies and has been used in recent review studies on associations between neighborhood environment attributes and health outcomes [28, 29]. The tool assesses six methodological dimensions: selection bias, study design, confounding, blinding, data collection methods, and withdrawals and dropouts. Each dimension is rated on a 3‐point scale: strong, moderate, or weak, which contributes to the calculation of a global rating [27]. Two authors (YT and WLT) independently performed a critical appraisal of included studies using this tool and discussed any disagreements until a consensus was reached.

2.4. Data Extraction and Synthesis

Data extraction was performed by YT using a standardized data extraction form. The elements extracted were study design, population, sample size, obesity management outcomes, neighborhood exposures, statistical analysis, covariate adjustment, results, and information about obesity intervention characteristics (and control conditions when applicable), namely, intervention content, intervention providers, number of sessions, duration, frequency, and follow‐up time points.

Meta‐analysis was not performed due to the expected heterogeneity (various measures of exposures and outcomes) across eligible studies; thus, we undertook a narrative synthesis of findings according to synthesis without meta‐analysis (SWiM) guidelines [30].

3. Results

Figure 1 provides the PRISMA flow diagram of articles included and excluded from the review. Of the 38,848 records identified, 27,310 records were screened after duplicate removal, 34 articles were assessed at the full‐text stage, and 6 articles met the inclusion criteria for this review. The main reasons for exclusion were participants not being children or adolescents living with obesity, no residential neighborhood features examined, and study protocol without results.

3.1. Characteristics of Study Participants

As shown in Table 2, participants' ages ranged from 6 to 18 years. Although all studies included participants of both sexes, the proportion of girls generally exceeded that of boys. Most studies (n = 5) reported the distribution of participants' ethnicity/race, of which four studies included primarily white participants [22, 31, 32, 33].

TABLE 2.

Participants characteristics of studies included in review (n = 6).

Study and country	Sample size	Study participants age (mean ± SD)	Overweight or obesity diagnostic criteria	Sex (girls: boys, %)	Ethnicity (white: non‐white, %)	Study design
Epstein 2012, USA	191	8–12 years 10.20 (1.20)	BMI > 85th percentile based on CDC growth charts	60.8: 39.2	Not reported	Secondary analysis of the intervention arm only of RCT
Fagg 2014, UK	9563	7–13 years 10.4 (1.75)	BMI > 91st centile of the UK 1990 BMI reference	55.3: 44.7	77:3: 22.7	Prospective cohort study
Armstrong 2015, USA	93	8–14 years 10.88 (1.67)	BMI > 85th percentile based on CDC growth charts	57.0: 43.0	76.3: 23.7	Secondary analysis of RCT
Fiechtner 2016, USA	498	6–12 years 9.70 (1.90)	BMI ≥ 95th percentile based on CDC growth charts	52.0: 48.0	51.6: 48.4	Secondary analysis of RCT
Hayes 2019, USA	181	7–11 years 9.43 (1.32)	BMI ≥ 85th percentile based on the CDC growth charts	62.7: 37.3	71.8: 28.2	Secondary analysis of the intervention arm only of RCT
Neshteruk 2023, USA	2838	6–18 years 11.10 (3.20)	BMI ≥ 95th percentile based on the CDC growth charts	55.0: 45.0	14.5: 85.5	Prospective cohort study

3.2. Characteristics of Obesity Management Interventions

As shown in Table 3, all studies provided family‐based, multicomponent behavioral interventions focusing on the development of parenting skills (for parents), healthy diets, and physical activity promotion (for parents and children/adolescents). Only one study [22] incorporated individualized family behavior plans aimed at encouraging participants to use available built environment resources to support behavior change, e.g., parks and recreational spaces. Three studies [22, 33, 34] included a diet intervention component based on principles of the Traffic Light Diet. The duration of obesity management interventions ranged from 10 weeks to 12 months, with the most common duration being 4 months (n = 2) [22, 33], and weekly delivery of intervention sessions (n = 2) [33, 34].

TABLE 3.

Obesity management intervention characteristics of included studies (n = 6).

Study	Intervention group	Control group	Key intervention features	Intervention duration	Intervention frequency	Total number of intervention sessions	Outcome assessment timepoints
Epstein 2012	All families had weekly weigh‐in and individual meetings with a therapist, followed by separate parent and child groups	Standard treatment	(1) Comprehensive eating plan to increase nutrient density while it decreases energy intake (The Traffic Light Diet) (2) Physical activity program (3) Information on parenting	6 months	Weekly	16–20	6, 12, and 24 months from baseline
Fagg 2014	An integrated, multicomponent healthy lifestyle program based on the principles of nutritional and sports science	N/A (All families received the intervention)	Diet and physical activity through education, skills training and motivational enhancement.	10 weeks	2 sessions/week	20	6 months from baseline
Armstrong 2015	(1) A family‐based behavioral intervention (2) A parent‐only behavioral intervention	Wait‐list control	(1) Education on healthy dietary habits (modified version of the Traffic Light Diet) (2) Increasing physical activity (3) Behavioral change strategies (e.g., monitoring, goal‐setting, modeling, stimulus control, differential attention, and positive reinforcement) (4) Individualized family behavior plans aimed at using available community resources, e.g., parks and recreational spaces	4 months	Weekly sessions for the first 8 weeks, then 4 biweekly sessions over the next 8 weeks	12	4 and 10 months from baseline
Fiechtner 2016	(1) Computerized clinician decision support (CDS) + a family self‐guided behavior change intervention (2) Computerized clinician decision support plus a health coach intervention	Usual care	(1) Decreases in screen time (2) Decreases in consumption of sugar‐sweetened beverages (3) Increases in moderate and vigorous physical activity (4) Improvement of sleep duration and quality (5) Improving diet quality, including increasing fruit and vegetable intake	12 months	Twice weekly text messages	(1) Participants can access to a study website with additional obesity‐related educational materials (2) A phone call from a study health coach at 1, 3, 6, and 9 months, plus two text messages every week	12 months from baseline
Hayes 2019	Family‐based behavioral weight loss treatment	N/A (All families received standard family‐based behavioral weight loss treatment)	(1) Diet: improve dietary quality and reduce caloric intake (The Traffic Light Plan) (2) Physical activity: a maximum goal of 90 min of at least moderate‐intensity activity per day for children at least 5 days per week (3) Self‐monitoring (food and physical activity logs) to set and evaluate behavior change goals (4) Behavior change techniques for dietary and activity modification (e.g., reinforcement, stimulus control, preplanning, and relapse prevention)	4 months	Weekly	16	4 months from baseline
Neshteruk 2023	Children's Healthy Lifestyles pediatric weight management clinic	N/A (All families received pediatric weight management clinic)	Multidisciplinary clinic care that treats children and adolescents with obesity. Patients with obesity and their families see a medical provider as well as a registered dietitian, physical therapist, and licensed counselor to address diet, activity, and medical needs to improve health and treat comorbidities.	Not reported	Not reported	Not reported	The mean number of observations per patient was 5.16 (±5.29)

3.3. Characteristics of Study Design

Two studies were secondary analyses of randomized controlled trials (RCTs), two were secondary analyses of the intervention arm only of RCTs, and two were prospective cohort studies. Most studies were conducted in the United States (n = 5), and one was conducted in the United Kingdom, with all studies conducted between 2012 and 2023. Sample sizes ranged from 93 to 9563 participants, with a total of 13,364 participants and a median of 345 (Table 2).

3.4. Measures of Neighborhood Environment Exposures

All studies used data from geographic information systems (GIS) to characterize built environment attributes. Neighborhoods were defined in several ways: as buffers centered on participants' residences (mean size 1.6 km), which included circular buffer areas (n = 2) [22, 33], and street‐network buffer areas (n = 1) [34]; as census tracts (n = 2) [31, 35]; and by road network distance to the nearest neighborhood feature of interest (n = 1) [32] (Table 4).

TABLE 4.

Methodological characteristics and quality assessment of studies included in review (n = 6).

Study	Neighborhood buffer type and size	Statistical method	Covariates	Quality assessment a
				Selection bias	Study design	Confounders	Blinding b	Data collection	Withdrawals	Global score
Epstein 2012	0.5 miles along the street network of each child's residence	Hierarchical mixed model analyses of covariance (ANCOVA)	Age, sex, and family‐level socioeconomic status	Moderate	Moderate	Moderate	Moderate	Strong	Strong	Strong
Fagg 2014	Lower Super Output Area (LSOA) codes, representing small areas with a mean population of 1500 across England	Multilevel model	Age, sex, ethnicity, and housing ownership	Moderate	Moderate	Strong	Moderate	Strong	Moderate	Moderate
Armstrong 2015	A circle with a 10‐mile radius around participants' residential address	Multilevel growth model	Age and sex	Moderate	Strong	Weak	Moderate	Strong	Moderate	Moderate
Fiechtner 2016	Distances along the street network from each participant's residence to the closest large supermarket	Generalized linear mixed effects model	Children's age, sex, and race/ethnicity Parent's age, country of birth neighborhood median income, and networked street distance to the nearest fast‐food restaurant	Moderate	Strong	Strong	Strong	Strong	Strong	Strong
Hayes 2019	One‐mile Euclidian buffer around each family's home	Linear regression model	Age, sex, and family‐level socioeconomic status	Moderate	Moderate	Moderate	Moderate	Strong	Strong	Strong
Neshteruk 2023	Census tract across the United States	Three‐level generalized linear mixed models	Age, sex, race/ethnicity, and health insurance status	Moderate	Moderate	Strong	Moderate	Strong	Weak	Moderate

^a

^b Among the two prospective cohort studies [31, 35], all participants received interventions, so the statement “the outcome assessor is not aware of the intervention status of participants” is not applicable in this case. Another aspect to consider regarding blinding is whether “the study participants are aware of the research question.” We made assumptions that participants recruited in intervention studies would not have been aware of the specific research question regarding the association between neighborhood environment characteristics and obesity management outcomes when these were reported as secondary data analyses. We chose to keep this category for consistency with the EPHPP and previous review studies on neighborhood environments and health‐related outcomes that have used this tool [28, 29].

3.5. Measures of Outcomes

All studies reported adiposity outcomes measured by weight and height, including BMI, BMI z‐score, and BMI relative to the 95th percentile (BMIp95) (Table 5). Only two studies additionally reported on lifestyle behaviors. These included fruit and vegetable intake by Food Frequency Questionnaire [32] or 24‐h dietary recalls [33] (n = 2) and sugar‐sweetened beverages intake measured by Food Frequency Questionnaire (n = 1) [32], moderate‐to‐vigorous physical activity (MVPA) by accelerometry (n = 1) [33], and screen time by parent‐reported questionnaire (n = 1) [33] (Table 6).

TABLE 5.

Summary of findings for associations between neighborhood environment features and adiposity (n = 6 studies).

Study	Neighborhood exposure	Outcome	Association
Park (n = 3)
Armstrong 2015	Number of parks	BMI z‐score (calculated by height and weight)	Residing in neighborhoods with more parks was associated with greater BMI z‐score reduction among participants in the obesity intervention group (γ = −0.002), but not in the control group (γ = −0.0006).
Epstein 2012	Number of parks	BMI z‐score (calculated by height and weight)	Residing in neighborhoods with more parks was associated with greater BMI z‐score reduction (F = 3.32, p = 0.020).
Hayes 2019	Number of parks	BMI z‐score (Calculated by height and weight)	No association
Supermarket (n = 2)
Epstein 2012	Number of supermarkets	BMI z‐score (Calculated by height and weight)	Residing in neighborhoods with lower number of supermarkets was associated with greater BMI z‐score decrease (F = 7.92, p < 0.001).
Fiechtner 2016	Distances along the street network from each participant's residence to the closest large supermarket (defined by with more than 50 employees)	BMI z‐score (Calculated by height and weight)	Residing closer to supermarkets was associated with a greater BMI z‐score decrease among children in the intervention arm (p = 0.07).
Convenience store (n = 1)
Epstein 2012	Number of convenience stores	BMI z‐score (Calculated by height and weight)	Residing in neighborhoods with lower number of convenience stores was associated with greater BMI z‐score (F = 3.58, p = 0.014).
Grocery store (n = 1)
Epstein 2012	Number of grocery stores	BMI z‐score (Calculated by height and weight)	No association
Fast‐food store (n = 1)
Fagg 2014	Density of local fast‐food outlets	BMI z‐score (Calculated by height and weight)	No association
Greenspace (n = 1)
Neshteruk 2023	Greenspace availability was defined as the inverse of the percentage of each tract covered in impervious surfaces (e.g., rooftops and parking lots)	BMIp95 (Calculated by height and weight)	Residing in neighborhoods with increased greenspace was associated with less BMIp95 reduction (β = 1.93, 95% CI: 0.19, 3.67)
Walkability (n = 1)
Neshteruk 2023	Walkability was calculated using street intersection density, distance to transit stops, employment types and housing	BMIp95 (calculated by height and weight)	Residing in neighborhoods with higher walkability scores was associated with greater BMIp95 reduction (β = −4.40, 95% CI: −5.98, −2.82)
Composite score (n = 1)
Fagg 2014	Composite built environment was calculated based on the percentage of the census tract made up of roads and greenspace	BMI z‐score (calculated by height and weight)	No association
Neighborhood deprivation (n = 1)
Fagg 2014	Income Deprivation Affecting Children Index (IDACI) measured the proportion of all children aged 0 to 15 years living in income deprived families in the United Kingdom	BMI z‐score (calculated by height and weight)	Residing in less deprived neighborhoods was associated greater BMI z‐score reduction (B = 0.034, p = 0.004)

TABLE 6.

Summary of findings for associations between neighborhood environment features and lifestyle behavior (n = 2 studies).

Study	Environmental measurement	Outcome	Outcome measurement	Association
Supermarket (n = 2)
Fiechtner 2016	Distances along the street network from each participant's residence to the closest large supermarket (defined by with more than 50 employees)	Sugar‐sweetened beverages intake	Food frequency questionnaire	No association
	Distances along the street network from each participant's residence to the closest large supermarket (defined by with more than 50 employees)	Fruit and vegetable intake	Food frequency questionnaire	Residing closer to supermarkets was associated with increased fruit and vegetable intake among children in the intervention arm (p = 0.04)
Hayes 2019	Number of healthy stores including supermarkets	Fruit and vegetable intake	24‐h dietary recalls	No association
Recreation facilities (n = 1)
Hayes 2019	Number of public recreation facilities	Physical activity	MVPA measured by accelerometer	Residing in neighborhood with more public recreation facilities was associated with increases in MVPA at the trend‐level (B = 2.107, p = 0.055).
	Number of public recreation facilities	Screen time	Parent‐reported screen time in hours	No association

Among built environment features examined, the number of parks (n = 3) [22, 33, 34] was the most frequently reported exposure variable, followed by the number of supermarkets (n = 2) [32, 34] (Table 5). Only one study examined the neighborhood social environment using an indicator of neighborhood deprivation [31] (Table 5). No study used field audits by trained observers or resident‐reported measures to characterize neighborhood built and social environments.

3.6. Associations Between Neighborhood Environments and Changes in Obesity

Three studies examined the association between parks and changes in BMI z‐score (Table 5). Epstein et al. [34] reported that after accounting for age, sex, and family‐level SES, children had greater reductions in BMI z‐score at the 2‐year follow‐up if they lived in a neighborhood with more parks (F = 3.32, p = 0.020). Similarly, Armstrong and colleagues [22] found that increased park density was related to decreases in BMI z‐score among children in the obesity intervention group (γ = −0.002) at the 10‐month follow‐up, but not in the control group (γ = −0.0006). In contrast, Hayes and colleagues [33] did not find any association between the number of parks and BMI z‐score reduction at the 4‐month follow‐up.

Results from two studies on associations between supermarket availabilities and changes in BMI z‐score were inconsistent. Fiechtner et al. [32] reported that living closer to a supermarket was associated with a greater reduction in BMI z‐score at the 12‐month follow‐up, while Epstein et al. [34] found that a lower number of supermarkets was associated with a greater BMI z‐score reduction at 24‐months follow‐up controlling for family‐level SES (F = 7.92, p < 0.001).

One study conducted sex‐stratified analyses when examining neighborhood environments and changes in BMI [35]. Residing in neighborhoods with higher walkability scores was associated with greater BMI reductions for both girls (β = −2.92, 95% CI: −4.66, −1.18) and boys (β = −4.67, 95% CI: −6.60, −2.73) [35]. However, residing in neighborhoods with higher availability of greenspaces was associated with a lower BMI reduction at follow‐up only for boys (β = 3.18, 95% CI: 0.61, 5.75) [35].

Lastly, the only study examining neighborhood social environments reported that residing in more deprived neighborhoods was associated with less reduction in BMI z‐score at 6‐month follow‐up (B = 0.034, p = 0.004) following adjustment for a number of covariates including parental employment and home ownership [31].

All six studies adjusted for potential individual‐level confounding factors, namely, age and sex. Three studies also adjusted for family‐level SES [31, 33, 34], and only one adjusted for neighborhood‐level covariates (i.e., neighborhood median income) [32].

3.7. Associations Between Neighborhood Environment and Changes in Lifestyle Behaviors

Inconsistent results were found in two studies, which examined associations between neighborhood built environments and changes in fruit and vegetable intake. Fiechtner et al. [32] reported that living closer to a large supermarket was associated with increased fruit and vegetable intake among participants in the intervention group at 1‐year follow‐up (p = 0.04). However, Hayes and colleagues [33] did not find any associations between the number of healthy food stores (including supermarkets but also grocery stores, produce markets, and supercenters) and changes in fruit and vegetable intake at 4‐month post‐intervention. Both studies did not account for family‐level SES.

Only one study examined associations between neighborhood built environment, physical activity, and sedentary behavior. Hayes et al. [33] reported that residing in neighborhoods with more public recreational facilities was associated with a small increase in objectively measured MVPA independently of family‐level SES (B = 2.107, p = 0.055, n = 181), but had no impact on changes in parent‐reported screen time of participants.

3.8. Quality Assessment

Based on the EPHPP criteria, three studies [32, 33, 34] were rated as strong, and three studies [22, 31, 35] were rated as moderate (see Table 4).

4. Discussion

To our knowledge, this is the first systematic review that synthesized evidence linking residential neighborhood environment characteristics to changes in obesity and lifestyle behavior outcomes among children and adolescents participating in obesity management programs. Neighborhood characteristics linked with physical activity environments such as more parks and the availability of public recreation facilities were found to be associated with improved obesity management outcomes, including reduced BMI [22, 34] and increased physical activity [33]. Additionally, limited evidence from individual studies suggests that residing in neighborhoods that are more walkable [35] and less deprived [31] may be associated with greater reductions in BMI. Mixed findings were reported for the associations between neighborhood food environments and obesity management outcomes [32, 33]. Overall, there were substantial differences among the six studies reviewed in terms of neighborhood buffer types and sizes, measures of neighborhood features, and measures of outcomes.

Two out of three studies investigating park access indicated greater reductions in BMI z‐score following obesity management interventions among participants with better park access. The one study [33] that reported a null association had a much shorter intervention duration (4 months) compared to the other studies (≥ 6 months). The relatively short intervention duration in this study may not have allowed sufficient time to result in BMI changes. A meta‐analysis concluded that behavioral interventions lasting 6 to 12 months were associated with small reductions in BMI among children and adolescents [36].

Current research suggests that parks have the capacity to promote physical activity and cardiovascular health among children and adolescents with obesity [37], possibly through the provision of spaces and resources for physical activity [38]. This is in line with the study by Hayes and colleagues which found in a small study of n = 181 participants that residing in neighborhoods with more public recreation facilities was associated with increases in MVPA [33]. Incorporating individualized counseling for participants to use their neighborhood environments in obesity management programs may further promote regular physical activity [39] and improvements in adiposity [22]. For example, one pediatric obesity intervention program in the United States provided individualized counseling sessions to guide adolescents with overweight or obesity to use their specific surrounding built environment (e.g., parks, playgrounds, and sidewalks) to achieve physical activity goals [39]. Adolescents in the intervention group increased and sustained their daily MVPA by 9.3 min over the control group at the 3.5‐month follow‐up, representing a 29% increase in their baseline daily MVPA [39]. This study [39], however, did not examine associations between neighborhood characteristics and and MVPA or other obesity management outcomes.

The two studies examining associations between supermarket availability and changes in BMI used different indicators to operationalize supermarket exposure (i.e., number of supermarkets vs. distance to the nearest supermarket) [32, 34]. One study reported that living in neighborhoods with a lower number of supermarkets (indicating lower availability) was associated with greater BMI reduction [34], while the other reported that residing in neighborhoods that were closer to supermarkets (indicating higher availability) was associated with a greater decrease in BMI z‐score [32]. These inconsistencies may be attributed to variations in how supermarket availability was measured, which limits comparability across studies. Differences in measures of supermarket availability (e.g., presence, number, density, and distance) have also been noted in previous systematic reviews [40]. Also, the inconsistent associations between the two studies [32, 33] might partly reflect potential confounding. For example, other sources of fresh produce, such as farmers' markets and fresh food markets, were not considered in analytical models. Inconsistencies in associations underscore the complex pathways through which supermarket availability may influence dietary and obesity outcomes among children participating in obesity management interventions. Compared to those living near convenience or fast‐food stores, children who live near supermarkets may have more options for fresh fruit and vegetables, potentially enhancing their fruit and vegetable intake. However, the presence of unhealthy food options in supermarkets complicates this association. Thus, future research should assess not only supermarket availability but also the nutritional quality of available foods within these supermarkets, such as the percentage of floor space for fresh fruits and vegetables, ultra‐processed foods, and sugar‐sweetened beverages. Moreover, decisions about food purchases are determined not only by spatial accessibility but also by factors such as food price [41], dietary preferences, and parenting style [42].

Some pediatric obesity management interventions offer, as part of their program, grocery store visits during which a dietitian [43] or a trained patient navigator [44] guides families in selecting low‐cost healthy food. Increased health literacy on foods is associated with healthier dietary practices among parents [45] and adolescents [46]. While all six studies included a dietary component as part of the intervention, none assessed food‐related health literacy in the relationship between neighborhood food environments and obesity management outcomes. One explanation for the mixed results on associations between neighborhood supermarket accessibility and fruit and vegetable intake observed in the current study may be that neighborhood food environments are less influential than health literacy regarding intake of healthy foods. Future research should consider health literacy as a potential effect modifier in understanding associations between neighborhood food environment and obesity management outcomes.

Limited evidence from a single study found that residing in neighborhoods with increased greenspace was associated with less BMI reduction [35]. This association was opposite to the expected [38]. However, given that greenspace was defined as the inverse of the percentage of each census tract covered in impervious surfaces, such as rooftops and parking lots [35], there is a possibility that this definition inaccurately includes surfaces that are not greenspace. Also, this measure might include greenspaces that cannot or are not used by children and adolescents to engage in physical activity. For example, a study of 11–12‐years old children in the United Kingdom found that about half of their physical activity occurred in non‐green environments, as measured by global positioning systems (GPS), such as on roads, pavements, and concreted surfaces [47].

With regard to the obesity management interventions, two studies were secondary analyses of both intervention and control arms of RCTs [22, 32], and only one of the six included studies [22] incorporated neighborhood resources into the intervention. The latter provided families with individualized behavior modification plans that encouraged participants to engage with local community resources, such as parks and recreational spaces [22]. Notably, it reported that higher park density was associated with greater reductions in BMI z‐score at 10‐month follow‐up among youth in the behavioral family weight management intervention group, but not in the wait‐list control group [22]. While the available evidence is limited, this finding suggests that tailoring interventions to explicitly leverage neighborhood resources may enhance their effectiveness.

All studies included in this review used GIS‐based measures to assess neighborhood exposures. It is important to note that none of the included studies used perceived measures of neighborhood features. Objective and perceived measures can capture different facets of the neighborhood environment [48]. The perceived neighborhood environment is also critical, as the objectively measured walkability or availability of parks within a buffer zone around the home or an administrative area may not accurately reflect the area where participants' daily behaviors actually occur [49].

Regarding neighborhood environment exposures, there was significant variability in the measures of the built environment, and only one study included a measure of social environment features (i.e., neighborhood deprivation) [31]. We also observed variations in the types of neighborhood buffers across studies. Street‐network buffers (based on actual travel routes along the street or pedestrian road network) are thought to more accurately represent the relevant spatial context that influences an individual's behavior compared to administrative boundaries [50]. In contrast, circular buffers may include spatial areas that are not accessible and thus potentially less relevant (e.g., areas separated by bodies of water, railways, or highways) [50]. Notably, only one reviewed study [34] used street‐network buffers.

With regard to the outcome variables, BMI was the most common outcome. Despite improving lifelong lifestyle behaviors being one of the primary objectives according to clinical practice guidelines for managing pediatric obesity [5, 6], this review found that only two studies examined changes in diet, physical activity, and sedentary behavior, which mostly relied on self‐report (or proxy report) tools. These included diet outcomes measured by questionnaires [32, 33], accelerometer‐measured MVPA, and parent‐reported screen time [33]. However, self‐reported lifestyle behavior is sometimes susceptible to recall bias [51]. Also, no study examined sleep outcomes (e.g., sleep duration and sleep quality). As recent meta‐analyses reported shorter sleep duration to be associated with an increased risk of developing overweight/obesity in both children and adolescents [52, 53], sleep plays an important role in obesity management.

As for study quality assessment, all six studies had moderate or strong ratings for the global score and for each methodological dimension. Confounding, as one of the quality components, is noteworthy for discussion. Although a majority of studies (5/6) had a moderate or strong rating on the dimension related to confounding, three studies [31, 33, 34] controlled for family SES, and only one study [32] controlled for neighborhood‐level SES (i.e., neighborhood median income). Not accounting for individual/family‐level [54] and neighborhood‐level SES [55] when examining specific environmental effects on health outcomes may result in biased associations. Also, no study adjusted for residential self‐selection (e.g., directly by adjusting for preferences or attitudinal measures), which is frequently considered in studies among adult populations [56]. Additionally, neighborhood exposures (distal exposures) have smaller effect sizes compared to individual (proximal) exposures [57]; therefore, the generally small sample sizes among included studies may have led to limited power to detect associations.

4.1. Future Research

Neighborhood environments, particularly built environments, have been extensively studied in relation to obesity and health behaviors; however, there remains limited evidence on their influence among individuals living with chronic conditions, notably children and adolescents living with obesity and enrolled in a treatment program. Future research could benefit from incorporating objective measures (e.g., accelerometry) and innovative technologies (e.g., wearable sensors [58], GPS [59]). One aspect of particular interest is sleep behaviors—examining both subjective (e.g., perceived sleep problems and sleep quality) and objective measures of sleep (e.g., sleep duration) warrants investigation.

Considering the limited research and inconsistent findings regarding the associations between supermarket accessibility and pediatric obesity management outcomes, there is a need for more rigorously designed studies to understand the nuanced relationship in this population. Furthermore, emerging research exploring digital food environments, which include digital food marketing, online food and grocery ordering platforms, and food and recipe blogging [60] should be considered in future studies. Online food delivery platforms tend to promote unhealthy foods and beverages more frequently than healthy options [60], and digital food environments can influence food choices, preferences, and consumption [61]. Given the increasing use of online food platforms, future studies should also investigate how digital food environments might impact dietary outcomes among families of youth receiving obesity management care.

RCTs of multicomponent behavioral interventions for obesity management should explore potential interactions between intervention effects and neighborhood characteristics, such as whether an intervention is more or less effective in neighborhoods with differing levels of key contextual factors linked to place of residence. In addition, future research should prioritize the inclusion of more diverse and historically underrepresented populations to enhance the applicability of findings across sociodemographic groups. Also, considering that boys and girls, and children of different ages may use their neighborhood environments differently, future studies should conduct analyses stratified by sex and age group.

Although the current information is limited, our findings suggest that behavioral treatment programs for pediatric obesity could incorporate individualized components that help participants leverage supportive features of their neighborhood environments. For example, this could consist of tailored “walking maps” that guide families through safe and walkable routes while avoiding areas with high densities of food outlets (e.g., convenience stores). Programs might also include information on nearby recreational resources, such as parks or local sports teams, to facilitate sustained engagement in physical activity.

To facilitate the incorporation of neighborhood characteristics into tailored programming, the assessment of the relevant neighborhoods would need to be done at program entry. Risk‐stratification dashboards based on postal codes and leveraging GIS data are already being piloted in some US health systems [62], and similar approaches may enhance obesity care planning by enabling an individualized approach to goal setting and intervention delivery. Future research could explore similar approaches in pediatric obesity management programs.

4.2. Strengths and Limitations

This review has some notable strengths. First, it is the first to specifically examine the neighborhood environment and changes in obesity and lifestyle behaviors among children and adolescents participating in obesity management interventions. Second, this review strictly adhered to systematic review methodologies.

In terms of limitations, a very small number of eligible studies was identified despite a comprehensive search strategy; therefore, insights gained from this review are restricted in scope and findings should be interpreted with caution. Second, inclusion criteria were restricted to publications in English, which may limit the generalizability of the findings of studies published in languages other than English. Third, we identified limited diversity in study populations, with most participants being white. Fourth, given the heterogeneity of neighborhood exposures and outcomes among included studies, and the limited number of eligible studies, we were unable to conduct a meta‐analysis, thus hindering the assessment of publication bias. Instead, we conducted a narrative evidence synthesis. Lastly, we included studies that focused exclusively on participants' residential neighborhoods. However, this approach may lead to measurement error in exposure assessment, as children and adolescents spend a significant amount of time at school, in transit, and in other locations.

5. Conclusion

This systematic review highlights inconsistent findings among the few studies that examined neighborhood determinants of obesity management outcomes among children and adolescents. Results suggest that neighborhood resources that support physical activity (parks, recreational facilities) may be associated with better outcomes, namely, greater reductions in obesity indicators at program completion. Future longitudinal studies, featuring longer follow‐up periods and adequately adjusted for potential confounders (e.g., individual‐, family‐, and neighborhood‐level characteristics), are needed to comprehensively examine neighborhood environments, particularly social environments, and their impact on adiposity and lifestyle behaviors among children and adolescents enrolled in obesity management care.

Author Contributions

Y. Tang and A. Van Hulst conceptualized this review. Y. Tang and S. Calleja developed the search strategy, and Y. Tang conducted searches. Y. Tang, W.L. Tock, and A. Van Hulst carried out article screening. Y. Tang conducted data extraction, and Y. Tang and W.L. Tock completed critical appraisal. Y. Tang drafted the manuscript and integrated the feedback received from all co‐authors. A. Baillot, K.M. Morrison, and A. Van Hulst provided expertise related to obesity management, and M. Riva and A. Van Hulst provided expertise related to neighborhood environments. All authors provided critical feedback and helped shape the research, analysis, and manuscript, and approved the final manuscript.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting information

Supporting Information S1: Supporting information.