Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Jan 1.
Published in final edited form as: J Pediatr. 2024 Oct 18;276:114355. doi: 10.1016/j.jpeds.2024.114355

Good Friends and Good Neighbors: Social Capital and Food Insecurity in Families with Newborns

Jennifer O Lambert 1, Melissa R Lutz 2, Colin J Orr 3, Jonathan S Schildcrout 4, Aihua Bian 5, Kori B Flower 6, H Shonna Yin 7, Lee M Sanders 8, William J Heerman 9, Russell L Rothman 10, Alan M Delamater 11, Charles T Wood 12, Michelle J White 13, Eliana M Perrin 14
PMCID: PMC11645200  NIHMSID: NIHMS2033485  PMID: 39428091

Abstract

Objective

To examine the association between social capital and household food insecurity among US families with newborns.

Study design

This cross-sectional analysis used enrollment data from 881 newborn-caregiver dyads at six geographically-diverse US academic sites enrolled in the Greenlight Plus Trial, a comparative effectiveness trial to prevent childhood obesity. Ordinal proportional-odds models were used to characterize the associations of two self-reported measures of social capital: 1) caregiver social support and 2) neighborhood social cohesion, with household food insecurity after controlling for sociodemographic characteristics.

Results

Among 881 newborn-caregiver dyads (49% Hispanic, 23% non-Hispanic white, 17% non-Hispanic Black; 49% with annual household income <$50,000), food security was high for 75%, marginal for 9%, low for 11% and very low for 4%. In covariate-adjusted analyses, caregivers with a low social support score of 18 had five times the odds (aOR=5.03 95%CI=3.28-7.74) of greater food insecurity compared with caregivers with a high social support score of 30. Caregivers with a low neighborhood social cohesion score of 10 had nearly three times the odds (aOR=2.87 95%CI 1.61-5.11) of greater food insecurity compared with caregivers with a high neighborhood social cohesion score of 20. These associations remained robust when both social capital measures were included in one model.

Conclusions

Caregiver social support and neighborhood social cohesion each appear to be inversely associated with food insecurity among US families with newborns. Longitudinal research is needed to determine the directionality of these relationships and whether improving social capital for families with young children reduces household food insecurity.

Household food insecurity, defined as when a household’s “ability to acquire adequate food is limited by a lack of money and other resources,” affected over one in six US households with children in 2022.[1] Food insecurity is associated with numerous adverse health outcomes in children including increased body mass index (BMI),[2] anemia,[3] abnormal early childhood development, [4,5] academic problems,[4,6-8] and mental and behavioral health problems among both children[4,7,9,10] and their parents.[9,11] Although poverty is a well-established risk factor for food insecurity,[12] the other resources that affect food security status have been less thoroughly explored. Social resources may represent some of these other resources and could be novel targets for interventions to prevent or reduce food insecurity and its deleterious impacts on the health of children and families.

Social capital, defined in pediatrics as “the benefits that a child receives from social relationships”[13] is a positive social driver of health. The relationship between social capital measures and food insecurity is underexplored in families with infants. It is important to study social capital in the newborn period, as it represents a time when social networks are especially relied upon, experience significant change,[14] and may be more modifiable than at other life stages. It is a time of frequent contact with the healthcare system, which provides the opportunity for connection to food security resources.[13]

Two fundamental components of social capital are social support and social cohesion.[13] Social support has been defined as “support accessible to an individual through social ties to other individuals, groups, and the larger community”[15] and has at least four domains: emotional support, informational support, social companionship, and instrumental support. Greater social support has been linked with improved health outcomes in adults with chronic conditions.[16-18] Evidence supports the buffering hypothesis that social support can attenuate the potentially deleterious effects of stressors on health,[19] and specific neurobiological pathways that mediate these relationships have been proposed.[20] Among US adults, there are conflicting results regarding the potential association between social support and food insecurity. One analysis found a significant, inverse association between social support and food insecurity,[21] yet another found no significant association.[22] Among families with young children, there are very limited data on the potential relationship between social support and food insecurity. One analysis found that greater instrumental social support defined as “the provision of financial aid, material resources, and needed services”[19] was protective against food insecurity among families with young children,[23] but data are lacking regarding other domains of social support in this population.

Social cohesion measured at multiple levels, including the neighborhood level, is defined as “the strength of a group to which a person belongs,”[13] and has also been linked to health outcomes [24] by both direct and indirect sharing of resources, through the sharing of information.[25] Multiple studies have found associations between higher levels of neighborhood social cohesion and lower food insecurity among families with children. [23,26-29] However, these studies did not include families with children under age two[23,27,28] or did not include data on child ages,[26,29] and most were geographically limited to a single city[27,28] or state.[26]

Therefore, data are lacking regarding the potential associations between social capital and food insecurity for families with very young children. It is important to study this relationship in racially and ethnically diverse populations due to the demonstrated significance of race in social capital constructs, including neighborhood social cohesion,[30] and the well-established racial and ethnic inequities in food insecurity among families with children.[1]

We sought to examine the association between household food insecurity and two measures of social capital, 1) caregiver social support and 2) neighborhood social cohesion, after controlling for income and other sociodemographic factors in a large, diverse cohort of families with newborns. We hypothesized that there would be an inverse relationship between each measure of social capital and food insecurity among families with newborns.

Methods

Participants

This secondary data analysis used baseline data from newborns and their caregivers enrolled in the Greenlight Plus Trial, a previously described comparative effectiveness trial to prevent childhood obesity (registered as NCT04042467).[31] The Vanderbilt University Medical Center Review Board served as the single IRB for this trial. Review committees at each participating site also approved this study, and all participants provided written informed consent to participate. English- and Spanish-speaking caregivers of newborns were recruited from six US academic medical center sites: Duke University, New York University, Stanford University, University of Miami, University of North Carolina, and Vanderbilt University Medical Center. Caregivers completed baseline questionnaires between October 2019 and August 2021. Families were excluded if their infant was greater than 21 days old, had very low birthweight <1500g, gestational age <34 weeks at birth, weight less than the third percentile at enrollment, or had a chronic medical problem that could affect weight gain.[31]

Measures

Caregiver social support was measured using the ENRICHD Social Support Instrument, the six-item form of a validated scale for an individual’s self-reported social support.[32,33] The respondent is asked how frequently someone is available to provide them with various types of social support, such as how often “is there someone available to you to give you good advice about a problem.” Responses range from “none of the time” to “all of the time.” Overall scores range from 6-30; a higher score indicates greater social support. [33]

Neighborhood social cohesion was measured using the social cohesion scale, a five-item form of a validated scale for an individual’s self-reported social cohesion in the area within a twenty-minute walk from their home (ie, their neighborhood).[34,35] The respondent is asked how much they agree or disagree with statements about the neighborhood, such as “people around here are willing to help their neighbors.” Responses range from “strongly disagree” to “strongly agree.” Overall scores range from 5-25, and the scale was designed so that a lower score indicates greater neighborhood social cohesion.[34] For clarity in this analysis, the scale was reverse coded such that a higher score indicates greater neighborhood social cohesion.

Food insecurity was measured using the validated six-item short form of the US Household Food Security Survey.[36] Scores range from 0-6; higher scores indicate greater food insecurity. A score of 0 indicates high food security, 1 indicates marginal food security, 2-4 indicates low food security, and 5-6 indicates very low food security.

Seven variables measured at baseline were included in our analysis to control potential confounding. Household income was assessed via self-report and categorized as follows: <$20,000, $20,000 to $49,999, $50,000 to $99,999, $100,000 or more, and Other (including "Don't know/Not sure" and "Prefer not to answer"). Household WIC status was determined through self-report and categorized as "Yes" (if the mother and/or child was enrolled in or received WIC food/formula) or "No" (if neither the mother nor the child was enrolled in or received WIC food/formula). The number of children in the household was self-reported and categorized as 1, 2, or >3. The number of adults in the household was also self-reported and categorized as 1, 2, or >3. Caregiver race and ethnicity was self-reported and categorized as "Hispanic," "Black, non-Hispanic," "White, non-Hispanic," or "Other, non-Hispanic." Caregiver education was self-reported and categorized as "Less than high school graduate," "High school graduate, but less than college degree," or "College degree or higher." The study site was recorded in the study database as Duke, Miami, NYU, Stanford, UNC, or Vanderbilt.

Statistical Analysis

We summarized the baseline characteristics using the median and interquartile range for continuous variables, and proportions and counts for categorical variables. Cronbach’s alpha was used to assess the internal consistency and reliability of caregiver social support, neighborhood social cohesion, and food insecurity scores. Spearman rank correlations were used to characterize the strength of the bivariate relationships between neighborhood social cohesion, caregiver social support, and food insecurity.

Our outcome variable, the food insecurity score, is an ordinal variable ranging from 0 to 6. To fully assess the covariate-adjusted relationship between neighborhood social cohesion and caregiver social support with food insecurity, we used ordinal proportional-odds logistic regression models. Ordinal proportional-odds models [37,38] are one type of ordinal logistic regression model wherein a proportional odds assumption is imposed on the risk factor associations with the outcome. The proportional odds assumption forces the odds ratio (OR) association measure to remain constant across all levels of the outcome variable. In our case, this means that the OR associated with a change in a risk factor is the same across level of the food insecurity outcome. For example, as the food insecurity score ranges from 0 to 6, the odds ratio remains constant for outcome odds of 1-6 versus 0, 2-6 versus 0-1, 3-6 versus 0-2, etc. Even when the proportional odds assumption is violated, estimated OR most often provide meaningful and parsimonious measures of association.

We fit three models to examine associations with food insecurity: one that included caregiver social support and covariates as independent variables, a second that included neighborhood social cohesion and covariates as independent variables, and a third that included caregiver social support, neighborhood cohesion, and covariates as independent variables. Model covariates were prespecified based on existing literature on social capital and food insecurity, as well as the expertise of the author team, to minimize confounding bias.

The estimated OR characterize the strength of association between risk factors and outcome variables, and we report 95% confidence intervals (CI) to capture uncertainty. Odds ratios greater than one are consistent with higher food insecurity, and OR less than one are consistent with lower food insecurity. Because caregiver social support and neighborhood social cohesion are continuous variables, and to avoid assuming a linear relationship between these independent variables and food insecurity, we decided a priori to use restricted cubic spline functions. A brief description of these functions has been provided in the supplemental material.[39] To display the associations graphically we calculated OR with 95% CI for a series of values of caregiver social support and neighborhood social cohesion in comparison with the mode for each measure.

Out of 900 caregiver-infant dyads enrolled in the Greenlight Plus study, only 19 were excluded from this analysis due to missing food insecurity scores. Missing covariates, including three caregiver social support scores and 30 neighborhood social cohesion scores, were multiply imputed with chained equations using predictive mean matching implemented through the areglmpute function from the R package Hmisc.[40] One hundred distinct imputations were constructed and estimates from the imputation-specific analyses were combined with Rubin’s Rule.[41] Overall p-values, corresponding to the global test for the association between each of the independent variables and the outcome, were calculated using the Wald test in the adjusted models. A two-sided p-value less than 0.05 was considered statistically significant. All analyses were conducted with R version 4.2.3.[42]

Results

A total of 881 caregiver-infant dyads (98% of the Greenlight Plus sample) were included in the analysis. Approximately half of the caregivers, 49%, identified as Hispanic or Latino; 23% identified as non-Hispanic white; 17% identified as non-Hispanic Black. Forty-nine percent of caregivers reported an annual household income less than $50,000, and 21% reported educational attainment less than high school graduate. In 63% of families, there was participation in WIC by the mother, infant, or both. Demographic and baseline characteristics of the participants are shown in Table I.

Table I.

Characteristics of study participants (N=881)

Variables N
Infant age (days) 881 4 [3,6]
Caregiver income 881
 < $20,000 210 (24%)
 $20,000 to $49,999 221 (25%)
 $50,000 to $99,999 99 (11%)
 $100,000 or more 122 (14%)
 Other (don’t know/not sure or prefer not to answer) 229 (26%)
Caregiver education 881
 Less than high school graduate 185 (21%)
 High school graduate, but less than a college degree 379 (43%)
 College degree or higher 317 (36%)
Caregiver race and ethnicity 881
 Black, non-Hispanic 150 (17%)
 Hispanic/Latino 431 (49%)
 White, non-Hispanic 207 (23%)
 Other, non-Hispanic 93 (11%)
Caregiver preferred spoken language 881
 English 581 (66%)
 Spanish 300 (34%)
Number of adults at household 881
 1 95 (11%)
 2 564 (64%)
 ≥3 222 (25%)
Number of children at household 881
 1 337 (38%)
 2 261 (30%)
 ≥3 283 (32%)
Household WIC status 881
 Enrolled 553 (63%)
 Not enrolled 328 (37%)
Food insecurity score 881
 High food security (0) 665 (75%)
 Marginal food security (1) 80 (9%)
 Low food security (2-4) 97 (11%)
 Very low food security (5-6) 39 (4%)
Caregiver social support 878 29 [26, 30]
Neighborhood social cohesion 851 18 [15, 20]
Open in a new tab

Values are n (%) for categorical variables and median [25th to 75th percentiles] for continuous variables. Food insecurity scores range from 0-6; higher scores indicate greater food insecurity.

WIC = Women, Infants & Children Program

The median caregiver social support score was 29 (IQR: 26-30), and the median neighborhood social cohesion score was 18 (IQR: 15-20). Seventy-five percent of families reported high food security, 9% reported marginal food security, 11% reported low food security, and 4% reported very low food security. There was high internal consistency of all three measurement scales with Cronbach’s alpha of 0.85 (95%CI= 0.83-0.86) for caregiver social support, 0.80 (95%CI= 0.77-0.82) for neighborhood social cohesion, and 0.86 (95%CI= 0.84-0.87) for food insecurity. In unadjusted analyses, food insecurity was significantly inversely correlated with caregiver social support (Spearman’s rho= −0.32, p value<0.0001) and neighborhood social cohesion (Spearman’s rho= −0.21, p value <0.0001), and caregiver social support was significantly positively correlated with neighborhood social cohesion (Spearman’s rho= 0.28, p value <0.0001).

Caregiver social support and neighborhood social cohesion were each highly associated with household food insecurity. In covariate adjusted analyses, caregiver social support and neighborhood social cohesion were significantly associated with the food insecurity score at the two-sided 0.05 significance level. Graphical results of OR for a series of values compared with the mode are shown in Figure 1a for caregiver social support and Figure 1b for neighborhood social cohesion and are also reported in supplemental Table I and supplemental Table 2 in the supplemental material.

Figure 1a: The Association Between Caregiver Social Support and Food Insecurity.

Figure 1a:

Open in a new tab

Adjusted OR and 95% CI from the fully adjusted ordinal proportional-odds regression model of food insecurity (US Household Food Security Survey score) on caregiver social support (ENRICHD Social Support Instrument score). Values are compared with the reference mode value of caregiver social support=30. The Y-axis is natural log-transformed. The overall p value was calculated with the Wald test. This p-value corresponds to the global test for the association between the caregiver social support and the outcome. Because we used a restricted cubic spline with three knots to model caregiver social support, the Wald test is a two degrees of freedom test, one for the linear component of the association and one for the non-linear component. Under the null hypothesis, coefficients corresponding to linear and non-linear associations are both zero and under the alternative hypothesis, at least one of them is non-zero.

Figure 1b: The Association Between Neighborhood Social Cohesion and Food Insecurity.

Figure 1b:

Open in a new tab

Adjusted OR and 95% CI from the fully adjusted ordinal proportional-odds regression model of food insecurity (US Household Food Security Survey score) on neighborhood social cohesion (social cohesion scale score). Values are compared with the reference, mode value of neighborhood social cohesion=20. The Y-axis is natural log-transformed. The overall p value was calculated with the Wald test. This p-value corresponds to the global test for the association between the social cohesion and the outcome. Because we used a restricted cubic spline with three knots to model social cohesion, the Wald test is a two degrees of freedom test, one for the linear component of the association and one for the non-linear component. Under the null hypothesis, coefficients corresponding to linear and non-linear associations are both zero and under the alternative hypothesis, at least one of them is non-zero.

As shown in Figure 1a, caregivers with a low caregiver social support score of 18 had approximately five times the odds of greater food insecurity compared with caregivers with a high caregiver social support score of 30 (aOR=5.03 with 95%CI= 3.28-7.74). After further adjustment for neighborhood social cohesion, the relationship remained strong (aOR=4.50 with 95%CI= 2.90-6.98). As shown in Figure 1b, caregivers with a low neighborhood social cohesion score of 10 had nearly three times the odds of greater food insecurity compared with those with a high neighborhood social cohesion score of 20 (aOR=2.87 with 95% CI 1.61-5.11) after controlling for the covariates listed above, and the curve is approximately linear with respect to the natural log scale of the OR. The association remained significant after further adjustment for caregiver social support (aOR=2.15 with 95% CI=1.18-3.93). Additional OR are presented in supplemental Table 1 and supplemental Table 2. Graphical representation of the aOR of all covariates for all three models are included in Supplemental Figure 1, Supplemental Figure 2, and Supplemental figure 3 of the supplemental material.

Discussion

In this diverse sample of nearly 900 families with newborns from six US academic medical centers, we found support for our hypothesis that caregiver social support and neighborhood social cohesion are each inversely associated with food insecurity after controlling for sociodemographic characteristics. This supports the larger understanding that social resources are significantly associated with food security status for families with young children. We found a weak, positive correlation between caregiver social support and neighborhood social cohesion in our unadjusted analysis and found that the relationship between each of these components of social capital and food insecurity remained statistically significant even when including the other component in the fully adjusted model. This supports the understanding that each of these measures is associated with food insecurity and that the two measures should not be inappropriately conflated.

Our results provide new insights into the relationship between social support and food insecurity among families with young children. Although social support has not been consistently associated with food insecurity among US adults in prior literature,[21,22] this analysis demonstrates a significant association between caregiver social support and household food insecurity among US families with young children. One prior study of families with children showed an inverse association between one domain of social support, instrumental social support, and food insecurity.[23] The present analysis adds to the evidence base by finding a significant inverse association between a broader measure of social support and food insecurity. This result suggests that other domains of social support, such as emotional support and informational support, may also be associated with food insecurity for families with young children.

Our study adds to the existing literature on neighborhood social cohesion and food insecurity among families with children by analyzing this relationship in a novel population, families with newborns, in a geographically, racially, and ethnically diverse sample. It is important to study families with newborns because this life period involves significant social network changes[14] and frequent contact with the healthcare system, which may provide additional opportunities for connection to relevant neighborhood resources.[13] These results also highlight the importance of public policies that serve to support social networks and neighborhood social cohesion for families with young children.

Although lower caregiver social support and neighborhood social cohesion were each associated with greater food insecurity in our sample, the adjusted models differed in that the curve for the caregiver social support model was not linear, while the curve for the neighborhood social cohesion model was approximately linear. The fact that the lower odds of greater food insecurity were only observed at the highest levels of caregiver social support in our analysis calls into question whether interventions to enhance caregiver social support could be an effective way to decrease food insecurity. The approximately linear curve demonstrating the relationship between neighborhood social cohesion and the odds of greater food insecurity in our analysis suggests that even small increases in neighborhood social cohesion could have significant impacts on food insecurity or vice versa. These results lead us to hypothesize that social interventions at the neighborhood level may be more effective at reducing food insecurity, compared with interventions at the individual or household level. This hypothesis is consistent with the conclusions of similar research in families with older children,[23] but longitudinal data on both measures of social capital and food insecurity among families with young children are needed to test this hypothesis. It is also possible that the results of this analysis are related to an undefined shared upstream causal path.

Further study is warranted regarding the mechanisms by which social capital may decrease food insecurity. A systematic review of adaptive strategies used by families with food insecurity revealed numerous social strategies including borrowing food from friends and family, bringing children to relatives’ homes for meals, shopping with friends and family for bulk items, receiving cash assistance from friends or family, and living with extended family or friends.[43] The review found limited evidence that these strategies may differ by race and ethnicity, but further research is needed to determine whether it is best to culturally tailor social capital interventions, especially for populations facing disproportionally high rates of food insecurity.

Our study has some important limitations. First, the cross-sectional design leaves us unable to determine causality or directionality of the associations between our variables of interest. With this design, our variables of interest could be subject to a shared upstream causal path that we were not able to identify. The observational study design also raises concern for potential residual confounding, although we diligently sought to control for the relevant confounding factors. Second, the social capital measures used in this study were validated among US adults who were older, more male, and less racially and ethnically diverse than our sample,[32,35] which may limit their validity in our sample. In addition, our sample’s social capital scores were concentrated at higher values. As a result, the generalizability of our findings may be limited to populations with higher social capital. Another limitation of our neighborhood social cohesion scale is the lack of a codified threshold for lower neighborhood social cohesion.

The exclusion of families with language preferences other than English or Spanish and of infants with very low birthweight limits the generalizability of these findings to families with these characteristics. Furthermore, although data support the bidirectional relationship between parental mental health and food insecurity for families with children,[44] we did not include a measure of parental mental health in our model. This was because the available mental health measure in our study, the nine-item patient health questionnaire (PHQ-9), has been shown to have a low positive predictive value, and therefore a high false positive rate, among mothers with newborns less than one month old.[45] Fortunately, the ongoing Greenlight Plus trial[46] will survey caregiver social support, neighborhood social cohesion, food insecurity, and caregiver mental health at multiple timepoints among this sample over the newborns’ first two years of life to allow for longitudinal analyses of these associations.

Despite these limitations, our study has important strengths. In addition to the novel population of families with newborns, the strengths of this study include the large, diverse sample of English- and Spanish- speaking families from six sites across the US in which approximately half of caregivers identified as Hispanic or Latino. These data help to address an important gap in the literature on social drivers associated with food insecurity among Hispanic/Latinx households with children under age three, which is a growing population in the US with a disproportionately high rate of food insecurity compared with other ethnic groups.[47] We also use a validated measure of food insecurity that captures the severity of this phenomenon, rather than dichotomizing food insecurity as present or absent.

Future research should evaluate measures of social capital among families with young children longitudinally, including caregiver social support and neighborhood social cohesion. Data are needed regarding the modifiability of these measures, as interventions that successfully increase social support have only been demonstrated in adults[48] and data are lacking on interventions that increase neighborhood social cohesion. Ultimately, additional research is needed to determine whether enhancing caregiver social support or neighborhood social cohesion might represent novel means of preventing or addressing food insecurity for families with young children, in addition to connecting families with nutritional resources.

In conclusion, two measures of social capital, caregiver social support and neighborhood social cohesion, are each inversely associated with food insecurity among US families with newborns in this study. Longitudinal data are needed to determine the directionality of these associations and to determine the modifiability of social capital measures, as they may represent targets to prevent or address food insecurity among families with young children.

Supplementary Material

1

Sources of Financial Assistance

This secondary analysis utilized data from the Greenlight Plus Trial which was supported by the Patient Centered Outcomes Research Institute (PCORI) [contract number AD-2018C1-11238]. Study data were collected and managed using REDCap electronic data capture tools hosted by Vanderbilt University Medical Center and supported by NCATS/NIH grant number: UL1 TR000445. The funders of the Greenlight Plus Trial, PCORI and NCATS/NIH, had no role in the design and conduct of the study, collection and analysis of the data, preparation of the manuscript, or the decision to submit the manuscript for publication. Dr. Lambert is a postdoctoral fellow supported by NIH-NIDDK T32DK062707 grant.

Abbreviations and Acronyms

BMI

Body Mass Index

WIC

Special Supplemental Nutrition Program for Women, Infants, and Children

OR

Odds Ratio

CI

Confidence Interval

IQR

Interquartile Range

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.

CRediT Author Statement:

Jennifer O. Lambert: Conceptualization, Methodology, Formal analysis, Investigation, Writing- Original Draft, Writing- Review and Editing; Melissa R. Lutz: Conceptualization, Methodology, Writing- Review and Editing; Colin J. Orr: Conceptualization, Writing- Review and Editing; Jonathan S. Schildcrout: Methodology, Data Curation, Investigation, Formal analysis, Writing- Original Draft, Writing- Review and Editing; Aihua Bian: Data curation, Formal analysis, Writing-Review and Editing; Kori B. Flower: Conceptualization, Supervision, Writing- Review and Editing; H. Shonna Yin: Conceptualization, Funding acquisition, Investigation, Writing- Review and Editing; Lee M. Sanders: Methodology, Funding acquisition, Data curation, Supervision, Resources, Writing- Review and Editing; William J. Heerman: Conceptualization, Writing- Review and Editing, Russell L. Rothman: Conceptualization, Funding acquisition, Investigation, Resources, Writing- Review and Editing; Alan M. Delamater: Conceptualization, Investigation, Writing- Review and Editing; Charles T. Wood: Conceptualization, Supervision, Writing- Review and Editing; Michelle J. White: Methodology, Investigation, Writing- Review and Editing; Eliana M. Perrin: Conceptualization, Funding acquisition, Methodology, Data curation, Investigation, Supervision, Resources, Writing- Review and Editing.

Conflicts of Interest and Prior Presentations

Dr. Flower previously reviewed abstracts for Patient Centered Outcomes Research Institute funded studies unrelated to this study. The authors have no other conflicts of interest to disclose. An earlier version of this work was presented as an oral abstract at the 2023 Academic Pediatric Association Region IV conference and the 2023 Pediatric Academic Societies conference.

Contributor Information

Jennifer O. Lambert, Johns Hopkins University School of Medicine, Department of Pediatrics.

Melissa R. Lutz, Johns Hopkins University School of Medicine, Department of Pediatrics.

Colin J. Orr, University of North Carolina at Chapel Hill School of Medicine, Department of Pediatrics.

Jonathan S. Schildcrout, Vanderbilt University Medical Center, Department of Biostatistics.

Aihua Bian, Vanderbilt University Medical Center, Department of Biostatistics.

Kori B. Flower, University of North Carolina at Chapel Hill School of Medicine, Department of Pediatrics.

H. Shonna Yin, New York University School of Medicine, Department of Pediatrics.

Lee M. Sanders, Stanford University School of Medicine, Department of Pediatrics.

William J. Heerman, Vanderbilt University Medical Center, Department of Pediatrics.

Russell L. Rothman, Vanderbilt University Medical Center, Institute for Medicine and Public Health.

Alan M. Delamater, University of Miami Leonard M. Miller School of Medicine, Department of Pediatrics.

Charles T. Wood, Duke University School of Medicine, Department of Pediatrics.

Michelle J. White, Duke University School of Medicine, Department of Pediatrics.

Eliana M. Perrin, Johns Hopkins University Schools of Medicine and Nursing, Department of Pediatrics.

Data Sharing Statement

This secondary analysis utilized data from the ongoing Greenlight Plus Trial which was registered as “Greenlight Plus Study: Approaches to Early Childhood Obesity Prevention” NCT04042467 [https://clinicaltrials.gov/study/NCT04042467] on August 2, 2019. Deidentified individual participant data will not be made available until the Greenlight Plus Trial is complete.

References

  • [1].Rabbitt MP, Hales LJ, Burke MP, Coleman-Jensen A. Household food security in the United States in 2022. 2023. 10.32747/2023.8134351.ers. [DOI] [Google Scholar]
  • [2].Metallinos-Katsaras E, Must A, Gorman K. A Longitudinal Study of Food Insecurity on Obesity in Preschool Children. J Acad Nutr Diet 2012;112:1949–58. 10.1016/j.jand.2012.08.031. [DOI] [PubMed] [Google Scholar]
  • [3].Eicher-Miller HA, Mason AC, Weaver CM, McCabe GP, Boushey CJ. Food insecurity is associated with iron deficiency anemia in US adolescents. Am J Clin Nutr 2009;90:1358–71. 10.3945/ajcn.2009.27886. [DOI] [PubMed] [Google Scholar]
  • [4].Gallegos D, Eivers A, Sondergeld P, Pattinson C. Food Insecurity and Child Development A State-of-the-Art Review. Int J Environ Res Public Health 2021;18:8990. 10.3390/ijerph18178990. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [5].Rose-Jacobs R, Black MM, Casey PH, Cook JT, Cutts DB, Chilton M, et al. Household Food Insecurity: Associations With At-Risk Infant and Toddler Development. Pediatrics 2008;121:65–72. 10.1542/peds.2006-3717. [DOI] [PubMed] [Google Scholar]
  • [6].Alaimo K, Olson CM, Frongillo EA. Food insufficiency and American school-aged children’s cognitive, academic, and psychosocial development. Pediatrics 2001;108:44–53. [PubMed] [Google Scholar]
  • [7].Shankar P, Chung R, Frank DA. Association of Food Insecurity with Children’s Behavioral, Emotional, and Academic Outcomes: A Systematic Review. Journal of Developmental & Behavioral Pediatrics 2017;38:135–50. 10.1097/DBP.0000000000000383. [DOI] [PubMed] [Google Scholar]
  • [8].Winicki J, Jemison K. Food Insecurity and Hunger in the Kindergarten Classroom: Its Effect on Learning and Growth. Contemp Econ Policy 2003;21:145–57. 10.1093/cep/byg001. [DOI] [Google Scholar]
  • [9].Cain KS, Meyer SC, Cummer E, Patel KK, Casacchia NJ, Montez K, et al. Association of Food Insecurity with Mental Health Outcomes in Parents and Children. Acad Pediatr 2022;22:1105–14. 10.1016/j.acap.2022.04.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [10].Slopen N, Fitzmaurice G, Williams DR, Gilman SE. Poverty, Food Insecurity, and the Behavior for Childhood Internalizing and Externalizing Disorders. J Am Acad Child Adolesc Psychiatry 2010;49:444–52. 10.1097/00004583-201005000-00005. [DOI] [PubMed] [Google Scholar]
  • [11].Whitaker RC, Phillips SM, Orzol SM. Food Insecurity and the Risks of Depression and Anxiety in Mothers and Behavior Problems in their Preschool-Aged Children. Pediatrics 2006;118:e859–68. 10.1542/peds.2006-0239. [DOI] [PubMed] [Google Scholar]
  • [12].Wight V, Kaushal N, Waldfogel J, Garfinkel I. Understanding the link between poverty and food insecurity among children: Does the definition of poverty matter? J Child Poverty 2014;20:1–20. 10.1080/10796126.2014.891973. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [13].Duh-Leong C, Dreyer BP, Huang TT-K, Katzow M, Gross RS, Fierman AH, et al. Social Capital as a Positive Social Determinant of Health: A Narrative Review. Acad Pediatr 2021;21:594–9. 10.1016/j.acap.2020.09.013. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [14].Moon RY, Mathews A, Oden R, Carlin R. A Qualitative Analysis of How Mothers’ Social Networks Are Established and Used to Make Infant Care Decisions. Clin Pediatr (Phila) 2019;58:985–92. 10.1177/0009922819845332. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [15].Lin N, Simeone RS, Ensel WM, Kuo W. Social support, stressful life events, and illness: a model and an empirical test. J Health Soc Behav 1979;20:108–19. [PubMed] [Google Scholar]
  • [16].Reblin M, Uchino BN. Social and Emotional Support and its Implication for Health. Curr Opin Psychiatry 2008;21:201–5. 10.1097/YCO.0b013e3282f3ad89. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [17].Cámara RJA, Lukas PS, Begré S, Pittet V, von Känel R. Effects of social support on the clinical course of Crohn’s disease. Inflamm Bowel Dis 2011;17:1277–86. 10.1002/ibd.21481. [DOI] [PubMed] [Google Scholar]
  • [18].Hughes S, Jaremka LM, Alfano CM, Glaser R, Povoski SP, Lipari AM, et al. Social support predicts inflammation, pain, and depressive symptoms: Longitudinal relationships among breast cancer survivors. Psychoneuroendocrinology 2014;42:38–44. 10.1016/j.psyneuen.2013.12.016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [19].Cohen S, Wills TA. Stress, social support, and the buffering hypothesis. Psychol Bull 1985;98:310–57. 10.1037/0033-2909.98.2.310. [DOI] [PubMed] [Google Scholar]
  • [20].Ozbay F, Johnson DC, Dimoulas E, Morgan CA, Charney D, Southwick S. Social Support and Resilience to Stress. Psychiatry (Edgmont) 2007;4:35–40. [PMC free article] [PubMed] [Google Scholar]
  • [21].Ashe KM, Lapane KL. Food Insecurity and Obesity: Exploring the Role of Social Support. J Womens Health 2018;27:651–8. 10.1089/jwh.2017.6454. [DOI] [PubMed] [Google Scholar]
  • [22].De Marco M, Thorbum S. The relationship between income and food insecurity among Oregon residents: does social support matter? Public Health Nutr 2009;12:2104–12. 10.1017/S1368980009990243. [DOI] [PubMed] [Google Scholar]
  • [23].King C. Informal assistance to urban families and the risk of household food insecurity. Soc Sci Med 2017;189:105–13. 10.1016/j.socscimed.2017.07.030. [DOI] [PubMed] [Google Scholar]
  • [24].Miller HN, Thornton CP, Rodney T, Thorpe RJ, Allen J. Social Cohesion in Health. ANS Adv Nurs Sci 2020;43:375–90. 10.1097/ANS.0000000000000327. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [25].Szreter S, Woolcock M. Health by association? Social capital, social theory, and the political economy of public health. Int J Epidemiol 2004;33:650–67. 10.1093/ije/dyh013. [DOI] [PubMed] [Google Scholar]
  • [26].Denney JT, Kimbro RT, Heck K, Cubbin C. Social Cohesion and Food Insecurity: Insights from the Geographic Research on Wellbeing (GROW) Study. Matern Child Health J 2017;21:343–50. 10.1007/s10995-016-2119-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [27].DiFiore G, Hannan C, Fiks AG, Virudachalam S, Glanz K, Mayne SL. Associations Between Food Insecurity and Neighborhood Safety, Social Cohesion, Social Control, and Crime Among Mothers of Preschool-Aged Children. J Health Care Poor Underserved 2022;33:1258–74. 10.1353/hpu.2022.0111. [DOI] [PubMed] [Google Scholar]
  • [28].Carter MA, Dubois L, Tremblay MS, Taljaard M. Local social environmental factors are associated with household food insecurity in a longitudinal study of children. BMC Public Health 2012;12:1038. 10.1186/1471-2458-12-1038. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [29].Brisson D. Neighborhood Social Cohesion and Food Insecurity: A Longitudinal Study. J Soc Social Work Res 2012;3:268–79. 10.5243/jsswr.2012.16. [DOI] [Google Scholar]
  • [30].Hobson-Prater T, Leech TGJ. The Significance of Race for Neighborhood Social Cohesion: Perceived Difficulty of Collective Action in Majority Black Neighborhoods. The Journal of Sociology & Social Welfare 2012;39. 10.15453/0191-5096.3651. [DOI] [Google Scholar]
  • [31].Heerman WJ, Perrin EM, Yin HS, Schildcrout JS, Delamater AM, Flower KB, 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. 10.1016/j.cct.2022.106987. [DOI] [PubMed] [Google Scholar]
  • [32].Vaglio J, Conard M, Poston WS, O’Keefe J, Haddock CK, House J, et al. Testing the performance of the ENRICHD Social Support Instrument in cardiac patients. Health Qual Life Outcomes 2004;2:24. 10.1186/1477-7525-2-24. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [33].Mitchell PH, Powell L, Blumenthal J, Norten J, Ironson G, Pitula CR, et al. A Short Social Support Measure for Patients Recovering From Myocardial Infarction. J Cardiopulm Rehabil 2003;23:398–403. 10.1097/00008483-200311000-00001. [DOI] [PubMed] [Google Scholar]
  • [34].Echeverria SE. Reliability of Self-Reported Neighborhood Characteristics. J Urban Health 2004;81:682–701. 10.1093/jurban/jth151. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [35].Mujahid MS, Diez Roux A V, Morenoff JD, Raghunathan T. Assessing the Measurement Properties of Neighborhood Scales: From Psychometrics to Ecometrics. Am J Epidemiol 2007;165:858–67. 10.1093/aje/kwm040. [DOI] [PubMed] [Google Scholar]
  • [36].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;89:1231–4. 10.2105/AJPH.89.8.1231. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [37].Agresti A. Categorical Data Analysis. Wiley; 2002. 10.1002/0471249688. [DOI] [Google Scholar]
  • [38].Walker SH, Duncan DB. Estimation of the Probability of an Event as a Function of Several Independent Variables. Biometrika 1967;54:167–79. 10.2307/2333860. [DOI] [PubMed] [Google Scholar]
  • [39].Harrell FE. Regression Modeling Strategies: With Applications to Linear Models, Logistic and Ordinal Regression, and Survival Analysis. 2nd ed. Springer International Publishing; 2015. 10.1007/978-3-319-19425-7. [DOI] [Google Scholar]
  • [40].Harrell F Jr. Hmisc: Harrell Miscellaneous. R package version 5.1–1 2023. [Google Scholar]
  • [41].Rubin DB. Multiple Imputation for Nonresponse in Surveys. Wiley; 1987. 10.1002/9780470316696. [DOI] [Google Scholar]
  • [42].R Core Team. R: A Language and Environment for Statistical Computing. 2023. [Google Scholar]
  • [43].Kamdar N, Rozmus CL, Grimes DE, Meininger JC. Ethnic/Racial Comparisons in Strategies Parents Use to Cope with Food Insecurity: A Systematic Review of Published Research. J Immigr Minor Health 2019;21:175–88. 10.1007/s10903-018-0720-y. [DOI] [PubMed] [Google Scholar]
  • [44].Bruening M, Dinour LM, Chavez JBR. Food insecurity and emotional health in the USA: a systematic narrative review of longitudinal research. Public Health Nutr 2017;20:3200–8. 10.1017/S1368980017002221. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [45].Gjerdingen D, Crow S, McGovern P, Miner M, Center B. Postpartum Depression Screening at Well-Child Visits: Validity of a 2-Question Screen and the PHQ-9. The Annals of Family Medicine 2009;7:63–70. 10.1370/afm.933. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [46].Heerman WJ, Perrin EM, Yin HS, Schildcrout JS, Delamater AM, Flower KB, 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. 10.1016/j.cct.2022.106987. [DOI] [PubMed] [Google Scholar]
  • [47].Varela EG, McVay MA, Shelnutt KP, Mobley AR. The Determinants of Food Insecurity Among Hispanic/Latinx Households With Young Children: A Narrative Review. Advances in Nutrition 2023;14:190–210. 10.1016/j.advnut.2022.12.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • [48].Writing Committee for the ENRICHD Investigators. Effects of Treating Depression and Low Perceived Social Support on Clinical Events After Myocardial Infarction. JAMA 2003;289:3106. 10.1001/jama.289.23.3106. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS2033485-supplement-1.docx (476.6KB, docx)

Data Availability Statement

This secondary analysis utilized data from the ongoing Greenlight Plus Trial which was registered as “Greenlight Plus Study: Approaches to Early Childhood Obesity Prevention” NCT04042467 [https://clinicaltrials.gov/study/NCT04042467] on August 2, 2019. Deidentified individual participant data will not be made available until the Greenlight Plus Trial is complete.

RESOURCES