Abstract
Objective:
Infants with high birthweight have increased risk for adverse outcomes at birth and across childhood. Prenatal risks to healthy food access may increase odds of high birthweight. We tested whether having a poor neighborhood food environment and/or food insecurity had associations with high birthweight.
Methods:
We analyzed cross-sectional baseline data in Greenlight Plus, an obesity prevention trial across 6 US cities (n=787), which included newborns with a gestational age greater than 34 weeks and a birthweight greater than 2500 grams. We assessed neighborhood food environment using the Place-Based Survey and food insecurity using the US Household Food Security Module. We performed logistic regression analyses to assess the individual and additive effects of risk factors on high birthweight. We adjusted for potential confounders: infant sex, race, ethnicity, gestational age, birthing parent age, education, income, and study site.
Results:
Thirty-four percent of birthing parents reported poor neighborhood food environment and/or food insecurity. Compared to those without food insecurity, food insecure families had greater odds of delivering an infant with high birthweight (aOR 1.96, 95% CI: 1.01, 3.82) after adjusting for poor neighborhood food environment, which was not associated with high birthweight (aOR 1.35, 95% CI: 0.78, 2.34). Each additional risk to healthy food access was associated with a 56% (95% CI: 4%−132%) increase in high birthweight odds.
Conclusion:
Prenatal risks to healthy food access may increase high infant birthweight odds. Future studies designed to measure neighborhood factors should examine infant birthweight outcomes in the context of prenatal social determinants of health.
Keywords: birthweight, macrosomia, neighborhood, food insecurity, food environment
INTRODUCTION
Infants born with high birthweight, defined as either macrosomia (4000 grams or greater) or large for gestational age (90th percentile or greater), are at higher risk for adverse neonatal outcomes as well as obesity across the life course.1,2 Prenatal risk factors for high birthweight, such as pre-pregnancy obesity, excessive gestational weight gain and diabetes, are highly influenced by suboptimal maternal dietary patterns.3,4 Risks to healthy food access, including a poor neighborhood food environment and food insecurity, may block access to nutritious food during pregnancy. Studies of the neighborhood food environment have found that high fast food restaurant density is associated with gestational diabetes, and that high densities of convenience stores are associated with greater neighborhood rates of high birthweight.5,6 Food insecurity, the inability to afford adequate food, is also associated with increased gestational weight gain in birthing parents with overweight status as well as lower prenatal dietary quality.7,8 Assessing prenatal risks to healthy food access would increase our understanding of mechanisms that drive high birthweight outcomes.
Early obesity prevention programs in pediatric primary care leverage frequent and widely attended visits in infancy to target feeding practices at a parent level. Yet, risk factors for high birthweight (e.g., gestational weight gain) originate prenatally within a family and neighborhood context. To assess potential challenges faced by an early obesity prevention program, we examined newborn data from Greenlight Plus, a cohort from six US cities. We hypothesized that prenatal risks to healthy food access would be associated with high birthweight.
METHODS
Data Source and Study Sample
We analyzed baseline newborn data from the Greenlight Plus Study, a comparative effectiveness trial of a primary care-based obesity prevention program with and without health information technology (e.g., text messages supporting healthy eating) between well-child visits.9 Between January 2019 and August 2021, bilingual (English and Spanish) study staff recruited birthing parents ( “parents” ) of newborn infants affiliated with 6 academic medical centers across the country. Eligibility criteria for the Greenlight Plus Study included infants between 1–21 days old, born at 34 weeks of gestation or later, weighing at least 2500 grams or greater than 3rd percentile by World Health Organization curves,10 and without a health condition affecting growth. Parent-child dyads were excluded if the parent was not English or Spanish speaking or had uncorrected visual acuity problems. Our analysis included the baseline assessment conducted in the first 21 days of the infant’s life, excluding participants who were not the birthing parent or had missing baseline data (12.5%). Study participant characteristics (N=787) reflect the broader Greenlight Plus Study, with a majority self-identifying as Hispanic (~40%, n=327) or Black (~15%, n=131). Vanderbilt University Medical Center served as the single IRB for this trial: review committees at each participating site also approved this study (NYU, University of North Carolina at Chapel Hill, Duke University, University of Miami, and Stanford University).
Measures
High birthweight.
We extracted birthweight data in grams from the medical record. To incorporate high birthweight definitions associated with adverse outcomes as well as current clinical practice,11 we defined high birthweight using both of the following criteria: 1) macrosomia (4000 grams or more), or 2) large for gestational age (weight 90th percentile or higher by gestational age).12,13
Risks to Healthy Food Access
Poor Neighborhood Food Environment.
We assessed the neighborhood food environment using the “Healthy Food Access” scale from the Neighborhood Characteristics Place-Based Survey.14 Researchers developed this survey to address limitations of Census-tract level neighborhood measures, which do not account for individual experiences of the neighborhood environment.14 Parents assessed the following statements about their neighborhood, defined as “the area about a 20 minute walk around your home,” 1) “It is easy to purchase fresh fruits and vegetables;” 2) “There is a large selection of fresh fruits and vegetables available;” 3) “The fresh produce is of high quality;” 4) “It is easy to purchase low-fat products (such as low-fat milk or lean meats);” 5) “There is a large selection of low-fat products available;” 6) “The low-fat products are of high quality.” Answer choices were provided in a Likert scale of 1–5, with 1 indicating “Strongly agree” and 5 indicating “Strongly disagree.” Scale values were summed across the 6 statements to provide a total score ranging from 5–30 points. Since there is no validated cutpoint for this scale, we created a binary “poor food environment” variable (yes/no) using a cutpoint of ≥15 for analyses, which represented both the top quartile of our sample and also a level of disagreement with at least 2 of the statements.
Food Insecurity.
Food insecurity was assessed using the US Department of Agriculture short form of the Household Food Security Survey Module.15 We asked parents to assess the following over the past 12 months, 1) “The food that we bought just didn’t last, and we didn’t have money to get more;” 2) “We couldn’t afford to eat balanced meals;” 3) “Did you or other adults in your household ever cut the size of your meals or skip meals because there wasn’t enough money for food?” 4) “Did you ever eat less than you felt you should because there wasn’t enough money for food?” 5) “Were you ever hungry but didn’t eat because there wasn’t enough money for food?” Parents could respond: 1) Often true, 2) Sometimes true, 3) Never true, 4) Don’t know/refused. For analyses, we created a binary food insecurity (yes/no) variable at the validated cutpoint of an affirmative answer to at least two items.15
Statistical Analysis.
We summarized our sample with descriptive statistics. We fit separate logistic regressions of high birthweight (yes/no) on poor neighborhood food environment (Model 1a) and food insecurity (Model 1b), then both poor neighborhood food environment and food insecurity together in one model (Model 2). To examine whether an increased number of risks to healthy food access was associated with increased odds of high birthweight, Model 3 assessed risks to healthy food access as a linear variable by modeling the number of risks to healthy food access (0, 1, 2). For all models, we included the following covariates: child sex, race and ethnicity, birthing parent age, education, income, and Greenlight Plus study site. Gestational weight gain and diabetes were not included as covariates given their position on the mediation pathway between healthy food access and high birthweight.4,16 We report results with adjusted odds ratios (aOR) with 95% confidence intervals (CI), defining significance at a 2-tailed 0.05 level. Statistical analyses were performed using R version 4.0.5 (2021–03-31).
RESULTS
Study Sample (Table 1)
Table 1:
Study Sample Characteristics (N=787)
| Study Sample Characteristics, N (%) | |
|---|---|
| Child | |
| Assigned Female Sex | 411 (52.2%) |
| Mean Gestational Age (weeks) (median, interdecile range) | 39.1 (39.1, 37.3–40.6) |
| Race and Ethnicity | |
| Hispanic | 327 (41.6%) |
| White | 178 (22.6%) |
| Black | 131 (16.6%) |
| Asian/Multi-Race/Other | 151 (19.2%) |
| Birthing Parent | |
| Mean Birthing Parent Age (years) (median, interdecile range) | 30.1 (30.2, 22.1–37.7) |
| Mean Birthing Parent Pre-pregnancy BMI (median, interdecile range) | 27.6 (25.8, 20.5, 37.4) |
| Birthing Parent Completed High School | 656 (83.4%) |
| Married or Living as Married | 546 (69.4%) |
| Born outside of United States | 444 (56.4%) |
| Born in Mexico | 118 (15.0%) |
| Born in Honduras | 31 (3.9%) |
| Born in Ecuador | 39 (5.0%) |
| Household | |
| Study Site | |
| Nashville, Tennessee | 322 (40.9%) |
| New York, New York | 177 (22.5%) |
| Durham, North Carolina | 113 (14.4%) |
| Palo Alto, California | 71 (9.0%) |
| Chapel Hill, North Carolina | 60 (7.6%) |
| Miami, Florida | 44 (5.6%) |
| Household Income | |
| <$20,000 | 190 (24.1%) |
| $20,000-$49,999 | 205 (26.0%) |
| $50,000-$99,999 | 96 (12.2%) |
| $100,00 or more | 119 (15.1%) |
| Don’t know / Not sure | 177 (22.5%) |
| Key Study Variables, N (%) | |
| High Birthweight | 81 (10.3%) |
| Large for Gestational Age (>90th percentile for GA) | 75 (9.5%) |
| Macrosomia (>4000g) | 61 (7.8%) |
| Both | 55 (7.0%) |
| Risks to Healthy Food Access | |
| Poor Neighborhood Food Environment | 199 (25.3%) |
| Food Insecurity | 111 (14.1%) |
| Both | 43 (5.5%) |
Approaching half (41.6%) of families self-identified as Hispanic and 16.6% as Black. More than half of birthing parents (56.4%) were born outside of the US. The mean gestational age of the infants was at term (39.1 weeks). More infants met high birthweight criteria by being large for gestational age (9.5%) than for macrosomia (7.8%), 7.0% met both criteria. About a quarter (25.3%) of parents reported a poor neighborhood food environment, 14.1% reported food insecurity, and 5.5% reported both poor neighborhood food environment and food insecurity.
Risks to Healthy Food Access
Figure 1 shows the proportion of infants with high birthweight by the number of risks to healthy food access. Ten percent (10.3%) of overall infants had high birthweight. In the group which had both risks to healthy food access, 20.9% of infants had high birthweight. Table 2 displays that food insecurity was individually associated with greater odds of high birthweight (aOR 2.07 [95% CI: 1.07, 3.98]), even when adjusting for poor neighborhood food environment (Model 2: aOR 1.96 [95% CI: 1.01, 3.82]). Poor neighborhood food environment was not significantly associated with high birthweight (Model 1b: aOR 1.46 [95% CI: 0.85, 2.49]; Model 2: aOR 1.35 [95% CI: 0.78, 2.34]). Each additional risk to healthy food access was associated with a 56% increase in high birthweight odds (Model 3: aOR 1.56 [95% CI: 1.04, 2.32]).
The proportion of infants with high birthweight by the number of risks to healthy food access.
Table 2:
Associations of High Birthweight with Risks to Healthy Food Access
| Odds of High Birthweight; aOR (95% CI) | Model 1a | Model 1b | Model 2 | Model 3 |
|---|---|---|---|---|
| Poor neighborhood food environment (yes/no) | 1.46 0.85, 2.49 p=0.17 |
1.35 0.78, 2.34 p=0.28 |
||
| Food insecurity (yes/no) | 2.07* 1.07, 3.98 p=0.03 |
1.96* 1.01, 3.82 p=0.047 |
||
| Additive Risks to Healthy Food Access (exposure to 0, 1, 2 risks) | 1.56* 1.04, 2.32 p=0.02 |
Model 1 examined each risk to healthy food access individually. Model 2 examined both risks to healthy food access in one model. Model 3 examined risks to healthy food access as a linear variable (exposure to 0,1, and 2 risks). Each of the models were fit with logistic regression models adjusted for child sex, race, ethnicity, and birthing parent age, education, income, and Greenlight Plus study site.
p<0.05
DISCUSSION
In a sample of parent-newborn dyads from six US cities in a primary care-based childhood obesity prevention program, we found evidence that prenatal risks to healthy food access may increase high birthweight odds. Our study sample was well-positioned to examine this important health outcome as over half of our study sample was born outside the US and/or from racial and ethnic groups that experience higher burdens of complications related to obesity.17,18 Pregnancy is a vulnerable time period representing not only what scientists have established to be the origins of obesity risk,19 but also when physiological cravings for sweets and starchy foods increase intake of saturated fat and carbohydrates.20 Prenatal risks to healthy food access may further degrade diet quality, increasing high birthweight odds. These findings inform strategies to connect families during vulnerable time periods to interventions like food prescription programs, which directly link clinics with food suppliers to leverage trusted physician-patient relationships to improve healthy food access.21
We found that food insecurity, a household-level risk to healthy food access, was associated with high birthweight. This finding extends prior work by showing that food insecurity is associated not only with conditions that increase high birthweight risk (e.g., excessive gestational weight gain2,22) but also subsequent high birthweight as well. We did not detect individual associations with poor neighborhood food environment, diverging from prior work showing that poorer food environments were associated with gestational diabetes and neighborhood rates of high birthweight.5,6 Future studies should explore how proximal household-level risks like food insecurity may have stronger health effects than more distal neighborhood-level risks like neighborhood food environment.
We detected additive associations between high birthweight and risks to healthy food access, which contributes to mounting evidence about the cumulative effects of risk factors on adverse early childhood health outcomes.23,24 An explanation for this has been alluded to in prior literature: in the setting of one risk factor, families can reallocate healthy food to pregnant parents25,26 but when families have food insecurity with a poor neighborhood food environment, compensation strategies may be limited. Theories hypothesizing how healthy food access may affect dietary patterns27 describe concepts often missing in geospatial measures, but captured by our neighborhood food environment variable: 1) Availability or selection of healthy food (e.g., our questions assess for a “large selection”), 2) Accessibility or the ease of getting to healthy food (e.g., “easy to purchase”), 3) Acceptability or whether food meets personal standards (e.g., “high quality”). Our results show that our neighborhood food environment variable, which encompasses these domains, may have additive effects with food insecurity, which captures affordability (e.g., “couldn’t afford”). Future studies should apply qualitative methods to gain an in-depth understanding of how these domains may have interrelationships in shaping prenatal dietary patterns.
Limitations include study design, residual confounding and parent-reported measures. Study design is a limitation as this study question was asked within the context of assessing potential challenges to an early obesity prevention program, and thus was not powered for our specific study question, raising the risk for a type II error. Larger studies in the future may also consider designs that account for neighborhood self-selection bias,28 or explore more objective geospatial measures alongside parent-reported measures to comprehensively characterize the prenatal neighborhood food environment. Our sample’s birthweight was skewed by design, as neither preterm nor low birthweight infants were included. Pronounced effects of poor healthy food access may include low birthweight outcomes not captured in our sample. Our study had a cross-sectional design, though we are still able to infer directionality given that high birthweight would not cause risks to healthy food access as defined in this study. We included variables that would likely confound the relationships of interest; however residual confounding may still exist.
CONCLUSION
In this sample of parent-newborn dyads from an obesity prevention program across six US cities, we found evidence that risks to healthy food access, in particular food insecurity, was associated with high birthweight. In other words, the greater a parent’s prenatal risk for poor access to nutritious food – the greater their newborn’s risk of having high birthweight, which has widely recognized associations with adverse child health outcomes. These findings inform future research aiming to address the impact of prenatal social determinants of health on birthweight outcomes.
WHAT’S NEW.
To assess potential challenges to an obesity prevention program across six US cities, we examined prenatal risks to healthy food access like food environment and food insecurity. We found risks to healthy food access may increase odds of high birthweight.
Acknowledgments:
This work was supported by the Patient Centered Outcomes Research Institute (PCORI) [contract number AD-2018C1-11238], the National Center for Advancing Translational Sciences, National Institutes of Health under grant numbers UL1 TR000445 and 2KL2TR001446-06 (Dr. Duh-Leong), as well as the Life Course Intervention Research Network (Health Resources and Services Administration) UA6MC32492 (Dr. Duh-Leong). Study data were collected and managed using REDCap electronic data capture tools hosted at Vanderbilt University Medical Center. Funders/sponsors did not participate in study design, collection, analysis, interpretation of the data, writing of the report, or the decision to submit the paper for publication.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Conflict of Interest Disclosures: The authors have no conflicts of interest nor financial relationships relevant to this article to disclose.
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
REFERENCES
- 1.Cnattingius S, Villamor E, Lagerros YT, Wikström AK, Granath F. High birth weight and obesity—a vicious circle across generations. International Journal of Obesity. 2012;36(10):1320–1324. doi: 10.1038/ijo.2011.248 [DOI] [PubMed] [Google Scholar]
- 2.Khambalia AZ, Algert CS, Bowen JR, Collie RJ, Roberts CL. Long-term outcomes for large for gestational age infants born at term. Journal of Paediatrics and Child Health. 2017;53(9):876–881. doi: 10.1111/jpc.13593 [DOI] [PubMed] [Google Scholar]
- 3.Jaipaul JV, Newburn-Cook CV, O’Brien B, Demianczuk N. Modifiable Risk Factors for Term Large for Gestational Age Births. Health Care for Women International. 2009;30(9):802–823. doi: 10.1080/07399330903066160 [DOI] [PubMed] [Google Scholar]
- 4.Ørskou J, Henriksen TB, Kesmodel U, Secher NJ. Maternal characteristics and lifestyle factors and the risk of delivering high birth weight infants. Obstetrics & Gynecology. 2003;102(1):115–120. doi: 10.1016/S0029-7844(03)00402-2 [DOI] [PubMed] [Google Scholar]
- 5.Kahr MK, Suter MA, Ballas J, et al. Geospatial analysis of food environment demonstrates associations with gestational diabetes. American Journal of Obstetrics and Gynecology. 2016;214(1):110.e1–110.e9. doi: 10.1016/j.ajog.2015.08.048 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Duh-Leong C, Shonna Yin H, Gross RS, et al. The Prenatal Neighborhood Environment and Geographic Hotspots of Infants with At-risk Birthweights in New York City. J Urban Health. Published online May 31, 2022. doi: 10.1007/s11524-022-00662-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Pan L, Sherry B, Njai R, Blanck HM. Food Insecurity Is Associated with Obesity among US Adults in 12 States. J Acad Nutr Diet. 2012;112(9):1403–1409. doi: 10.1016/j.jand.2012.06.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Laraia BA, Siega-Riz AM, Gundersen C. Household Food Insecurity Is Associated with Self-Reported Pregravid Weight Status, Gestational Weight Gain, and Pregnancy Complications. Journal of the American Dietetic Association. 2010;110(5):692–701. doi: 10.1016/j.jada.2010.02.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Heerman WJ, Perrin EM, Yin HS, et al. The Greenlight Plus Trial: Comparative effectiveness of a health information technology intervention vs. health communication intervention in primary care offices to prevent childhood obesity. Contemp Clin Trials. 2022;123:106987. doi: 10.1016/j.cct.2022.106987 [DOI] [PubMed] [Google Scholar]
- 10.World Health Organization (WHO). WHO Anthro (Version 3.2.2, January 2011) and Macros; 2011. http://www.who.int/childgrowth/software/en/.
- 11.Modzelewski J, Pokropek A, Jakubiak-Proć M, et al. Large-for-gestational-age or macrosomia as a classifier for risk of adverse perinatal outcome: a retrospective cross-sectional study. J Matern Fetal Neonatal Med. 2022;35(25):5564–5571. doi: 10.1080/14767058.2021.1887127 [DOI] [PubMed] [Google Scholar]
- 12.Duryea EL, Hawkins JS, McIntire DD, Casey BM, Leveno KJ. A revised birth weight reference for the United States. Obstet Gynecol. 2014;124(1):16–22. doi: 10.1097/AOG.0000000000000345 [DOI] [PubMed] [Google Scholar]
- 13.Aris IM, Kleinman KP, Belfort MB, Kaimal A, Oken E. A 2017 US Reference for Singleton Birth Weight Percentiles Using Obstetric Estimates of Gestation. Pediatrics. 2019;144(1):e20190076. doi: 10.1542/peds.2019-0076 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Echeverria SE, Diez-Roux AV, Link BG. Reliability of self-reported neighborhood characteristics. J Urban Health. 2004;81(4):682–701. doi: 10.1093/jurban/jth151 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Blumberg SJ, Bialostosky K, Hamilton WL, Briefel RR. The effectiveness of a short form of the Household Food Security Scale. Am J Public Health. 1999;89(8):1231–1234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Schisterman EF, Cole SR, Platt RW. Overadjustment Bias and Unnecessary Adjustment in Epidemiologic Studies. Epidemiology. 2009;20(4):488. doi: 10.1097/EDE.0b013e3181a819a1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Ogden CL, Fryar CD, Martin CB, et al. Trends in Obesity Prevalence by Race and Hispanic Origin—1999–2000 to 2017–2018. JAMA. 2020;324(12):1208–1210. doi: 10.1001/jama.2020.14590 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Andrasfay T, Goldman N. Intergenerational Change in Birthweight. Epidemiology. 2020;31(5):649–658. doi: 10.1097/EDE.0000000000001217 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Edwards M. The Barker hypothesis. In: Handbook of Famine, Starvation, and Nutrient Deprivation: From Biology to Policy. ; 2017:191–211. doi: 10.1007/978-3-319-55387-0_71 [DOI] [Google Scholar]
- 20.Farland LV, Rifas-Shiman SL, Gillman MW. Early Pregnancy Cravings, Dietary Intake, and Development of Abnormal Glucose Tolerance. Journal of the Academy of Nutrition and Dietetics. 2015;115(12):1958–1964.e1. doi: 10.1016/j.jand.2015.04.018 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Little M, Rosa E, Heasley C, Asif A, Dodd W, Richter A. Promoting Healthy Food Access and Nutrition in Primary Care: A Systematic Scoping Review of Food Prescription Programs. Am J Health Promot. 2022;36(3):518–536. doi: 10.1177/08901171211056584 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Arzhang P, Ramezan M, Borazjani M, et al. The association between food insecurity and gestational weight gain: A systematic review and meta-analysis. Appetite. 2022;176:106124. doi: 10.1016/j.appet.2022.106124 [DOI] [PubMed] [Google Scholar]
- 23.Frank DA, Casey PH, Black MM, et al. Cumulative hardship and wellness of low-income, young children: multisite surveillance study. Pediatrics. 2010;125(5):e1115–1123. doi: 10.1542/peds.2009-1078 [DOI] [PubMed] [Google Scholar]
- 24.Wells NM, Evans GW, Beavis A, Ong AD. Early Childhood Poverty, Cumulative Risk Exposure, and Body Mass Index Trajectories Through Young Adulthood. Am J Public Health. 2010;100(12):2507–2512. doi: 10.2105/AJPH.2009.184291 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Duh-Leong C, Yin HS, Yi SS, et al. Material Hardship and Stress from COVID-19 in Immigrant Chinese American Families with Infants. J Immigr Minor Health. Published online September 7, 2021. doi: 10.1007/s10903-021-01267-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Hanson KL, Connor LM. Food insecurity and dietary quality in US adults and children: a systematic review. Am J Clin Nutr. 2014;100(2):684–692. doi: 10.3945/ajcn.114.084525 [DOI] [PubMed] [Google Scholar]
- 27.Caspi CE, Sorensen G, Subramanian SV, Kawachi I. The local food environment and diet: a systematic review. Health Place. 2012;18(5):1172–1187. doi: 10.1016/j.healthplace.2012.05.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Lamb KE, Thornton LE, King TL, et al. Methods for accounting for neighbourhood self-selection in physical activity and dietary behaviour research: a systematic review. International Journal of Behavioral Nutrition and Physical Activity. 2020;17(1):45. doi: 10.1186/s12966-020-00947-2 [DOI] [PMC free article] [PubMed] [Google Scholar]