Association and disparities of food insecurity and exposure to violence: analysis of the National Survey of Children ’ s Health

Context: Lack of access to food is a signi ﬁ cant concern for child well-being, and it creates many health disparities and adverse social outcomes. Food insecurity and its many associated risk factors increase parental stress, which are strongly correlated with an increased risk of child abuse and maltreatment. Research now identi ﬁ es being witness to domestic abuse as a form of child maltreatment, and exposure to violence in the community has been shown to result in similar long-term impacts. Objectives: Given the potential for lifelong adverse e ﬀ ects from experiencing adverse childhood events involving violence and food insecurity, our primary objective was to assess the relationship between the two and disparities among demographic factors. Methods: We conducted an observational study utilizing data from the National Survey of Children ’ s Health (NSCH) 2016 – 2021. The NSCH is a United States nationally representative survey completed by primary caregivers of one child per home aged 0 – 17 years. We determined population estimates (n=216,799; n=83,424,126) and rates of children experiencing food insecurity and parent-reported exposure to violence. We then constructed logistic regression models to assess associations, through odds ratios (ORs), between food security and exposure to violence including demographic factors. Results: Among the sample, 5.42 % of children experienced low food security and 7.4 % were exposed to violence. The odds of exposure to violence are 5.19 times greater for children with low food security compared to food-secure children (95 % con ﬁ dence interval [CI]: 4.48 – 6.02

to food, in itself, is a significant concern for child well-being, creating many health disparities and adverse social outcomesincluding increased risk for iron-deficiency anemia, tooth decay, frequent headaches, asthma, and poor mental and metabolic health outcomes [2,3].Further, lack of access to nutritious food may also result in the consumption of calorie-dense, nutrient-poor foods that are associated with an increased risk for poor physical health outcomes, including diabetes and obesity [4].Food insecurity is associated with other household factors, including poverty, single-parent households, race/ethnicity, and urbanicity [1].For example, in 2021, 32.1 % of households with income below the federal poverty level (FPL) were food insecure [1].Despite the overlap in these household factors, studies have found poor health outcomes to persist with food insecurity even when controlling for poverty, parental education, and more [3].These factors, specifically food insecurity, are known to be associated with high rates of stress, with an estimated 70.2 % of food-insecure individuals reporting increased stress levels [5].This high level of stress results in poor mental health for parents, the use of harsh discipline strategies, and greater frequency and negativity of conflict between parents [6].Furthermore, increased household stress has been identified as a strong predictor of the risk for child abuse [7].
In 2018, the American Public Health Association deemed child abuse a public health crisis in the United States [8].The Centers for Disease Control and Prevention (CDC) estimated that at least one in seven children experienced child abuse or neglect in 2020 [9].Current research has also identified being a witness to domestic violence as a form of child maltreatment [10].With one in three women and one in four men having experienced some form of domestic violence, one in 15 children have been exposed and 90 % of those children are eyewitnesses to domestic violence [11,12].Furthermore, one study found that in the United States from August 2013 to April 2014, 67.5 % of children had at least one exposure to violence either directly, indirectly, or by witness [13].Additionally, one study estimated that over half of all children worldwide (thus approximately one billion children ages 2-17 years old) experienced past-year violence in 2015 [14].Exposure to violence has been identified as one of the most common and severe sources of human stress [15].In addition to maltreatment, early exposures to violence within the communities have been shown to have similar long-term impacts as child abuse [16].Stress-biology research has identified stress-sensitive biomarkers in young individuals exposed to early stressbeing mistreated or abused by an adult, victimized by bullies, witness to serious domestic abuse, or being criminally assaultedthat are known to be associated with increased risk for heart, metabolic, or immune diseases as well as stroke and dementia when found in adults [15,17].
Given the potential for lifelong adverse effects from experiencing adverse childhood events involving violence and food insecurity, investigating the relationship between the two, in addition to disparities among age, race/ethnicity, and urbanicity of the child, may aid in developing mitigation strategies.Many studies show clear racial disparities in food insecurity.Even when controlling for social and economic factors, studies have found a higher risk of food insecurity among ethnoracial minorities [18].Given these risk factors, our primary objective was to assess the relationship between parent-reported exposure to violence and food insecurity among children utilizing data from the National Survey of Children's Health (NSCH).Our secondary objective was to identify the associations among varying demographic factors, because they may disproportionately affect certain groups.

Methods
To assess the relationship between exposure to violence and food insecurity, we conducted an observational study utilizing data from the NSCH.The NSCH is a nationally representative survey funded and directed by the Health Resources and Services Administration's Maternal and Child Bureau (HRSA MCHB) and fielded by the United States Census Bureau [19].The survey is conducted by selecting random addresses across the 50 states and Washington D.C. that are mailed eligibility screeners to be filled out online or on paper.Eligible householdsthose that reported one or more children ages 0-17 living in the householdwere then directed to fill out a screener questionnaire.From the screener, one child from each household was randomly selected to be the subject of the main topical questionnaire.The primary caregiver of the child then completed and submitted the age-appropriate questionnaireeither age 0-5, 6-11, or 12-17 years old.Because the NSCH is conducted annually, we combined data from the 2016 to 2021 cycles to increase our sample size, allowing for more reliable and robust findings.

Food insecurity
To identify children who experienced food insecurity, we utilized the survey question, "Which of these statements best describes your household's ability to afford the food you need during the past 12 months?"[15] Respondents' answers included: "We could always afford to eat good nutritious meals"; "We could always afford enough to eat but not always the kinds of food we should eat"; "Sometimes we could not afford enough to eat"; and "Often we could not afford enough to eat." [19].

Experiences of violence: abuse and domestic or community violence
The NSCH collected adverse childhood experiences with a broad spectrum of factors.To identify children who experienced parent-reported exposure to violence, we extracted data from two potential responses to the prompt, "To the best of your knowledge, has this child EVER experienced any of the following?,"which were: "was a victim of violence or witnessed violence in their neighborhood," and "saw or heard parents or adults slap, hit, kick, or punch one another in the home."[19].

Demographic factors
For age, we re-coded and combined the age groups into the following: 0-4, 5-10, 11-14, and 15-17 years.To identify the race/ethnicity of each child, we utilized the survey question, "What is this child's race/ethnicity?"The self-identified responses included: "Hispanic; White, non-Hispanic; Black, non-Hispanic; Asian, non-Hispanic; American Indian/Alaska Native, non-Hispanic; Native Hawaiian/Other Pacific Islander, non-Hispanic; Two or More Races, non-Hispanic."[19] NSCH data for White, Black, Asian, and Hispanic races are nationally representative.However, American Indian or Alaska Native (AIAN), Native Hawaiian and Other Pacific Islander (NHPI), and "Two or More Races" are not controlled independently, and thus may not constitute national estimates.To assess socioeconomic status, the percent of the federal poverty guidelines (% FPG) that the household income was categorized as was utilized.Urbanicity was defined as households in metro vs. nonmetro areas, according to the National Center for Health Statistics urban-rural classifications.

Statistical analysis
For this study, survey weights provided by the NSCH were employed and adjusted to account for multiple cycles of data [19].First, we reported the sample size and demographics, and the population estimates.Next, we determined the rate of children with parent-reported exposure to violence overall and food insecurity status.We then constructed logistic regression models to assess associations, through odds ratios (ORs), between food security groups and whether the child experienced exposure to violence.
To assess disparities with respect to children experiencing food insecurity and exposure to violence, we combined the six annual cycles of data from 2016 to 2021.Weights were adjusted according to the NSCH guidebook [19].We then constructed logistic regression models through ORs to assess food security and violence exposure by demographic factors.The reported demographic factors included age and race/ ethnicity, in addition to household variables, including food security, federal poverty level (FPL), and urbanicity.
Analyses were conducted utilizing Stata 16.1 MP (StataCorp, LLC, College Station, TX), with alpha set at 0.05 and confidence intervals (CIs) of 95 %.This study did not meet the requirements for human subjects research according to the regulatory definition of human subject research as defined in 45 CFR 46.102(d) and (f) of the Department of Health and Human Services' Code of Federal Regulations (https://www.hhs.gov/ohrp/regulations-and-policy/regulations/45-cfr-46/), and therefore was not submitted for ethics review.Results were reported according to Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.

Results
After combining annual data from the six cycles (2016-2021), the sample size (n) of our analysis was 216,799representing a population estimate of 83,424,126 children.Among these children, 69.1 % experienced high food security, 25.47 % experienced marginal food security, and 5.42 % experienced low food security.Table 1 shows the demographics of children in the United States based on food security.Among the 1,885 children who identified as Indigenous, 10.76 % experienced low food security.Additionally, 9.92 % of Black children (n=13,705), 7.19 % of multiracial children (n=15,599), and 6.88 % of Hispanic children (n=27,042) also experienced low food security.Among the children ages 0-4 (n=55,959), 5.05 % were food insecure.As age increased, the percentage of children experiencing food insecurity also increased, with 6.32 % of children ages 15-17 years (n=47,542) experiencing food insecurity.Among the children from households below the FPL (0-99 % FPL) (n=20,650), 13.13 % were food insecure.Among the children living in metropolitan areas (n=134,748), 5.38 % experienced food insecurity, compared to 6.26 % of children living in nonmetropolitan areas (n=27,830).
Among the sample (n=216,799), 7.4 % were reported by their primary caregiver to have experienced or witnessed violence.Table 2 shows the percentage of groups that has reported exposure to violence and associations with child food insecurity.The correlation between food insecurity and exposure to violence was statistically significant, with the odds of a child with low food security being significantly more likely to have experienced or witnessed violence compared to food-secure children (adjusted odds ratio [AOR]: 5.19, 95 % CI: 4.48-6.02).Indigenous, Multiracial, and Black children were significantly more likely to be exposed to violence when compared to White children (AOR: 1.65, 95 % CI: 1.22-2.21;AOR: 1.61, 95 % CI: 1.39-1.85;and AOR: 1.50, 95 % CI: 1.33-1.69,respectively).As age increased, there was a statistically significant increase in the odds of exposure to violence, with children ages 15 to 17 being the highest (AOR: 4.34, 95 % CI: 3.73-5.04).Compared to the metropolitan area, children in nonmetropolitan areas were more likely to be exposed to violence (AOR: 1.27, 95 % CI: 1.15-1.40).
Table 3 shows the interaction terms for race and food security on exposure to violence controlling for food security, age, percent FPL guidelines, and urbanicity.Compared to foodsecure White children, the odds of exposure to violence were significantly greater for all groups except Asianincluding White childrenwith all AORs exceeding 5. 40

Discussion
Food insecurity was associated with increased odds of children experiencing and/or witnessing violence, with those in households with low food security being 5.19 times more likely to experience violence compared to food-secure children.Additionally, children experiencing marginal food security were 2.30 times more likely to experience violence compared to those who were food secure.Due to the associated stigma and fear of child welfare involvement, parents will often underreport the extent of food insecurity that they experience [20].For this reason, it is common for parents to attempt to shield their child from experiencing food insecurity, either by eating less so the child can eat more or by pretending not to be hungry [20].These actions may increase the stress on the parent, which in turn increases the risk of exposure to violence, including child abuse [21].
Children with specific demographic factors were disproportionately affected by the interaction between experiences of violence and food insecurity.Race/Ethnicity was the most notable contributing demographic factor, with significant disparities affecting historically marginalized populations, including Indigenous, Black, and Multiracial children.Multiracial and Indigenous children with low food security had the highest odds of exposure to violence compared to the reference group.Among those with marginal food security, ethnoracial minority populations, including Black, Indigenous, and Multiracial, were three to four times more likely to be exposed to violence, compared to the reference groups.Finally, Indigenous, Multiracial, and Black children were more likely to be exposed to violence, even when they were food secure, compared to food-secure White children.
Disparities experienced by Indigenous, Black, and Multiracial children likely result from structural and systemic racism.For Indigenous people, forced assimilation and diminished tribal and food sovereignty created generational trauma that produced additional stressors for caregivers, likely contributing to both higher rates of food insecurity and child abuse [22].In order to honor treaty obligations with the federal government, many tribal communities abandoned traditional methods of acquiring food and instead accepted food rations that lacked both cultural relevance and nutritional value [23].Largely due to these government policies, Indigenous peoples have consistently been found to be food insecure [23].In addition to food insecurity, Indigenous peoples experience multiple child abuse risk factors, including higher rates of caregivers who experienced abuse, intimate partner violence, and poverty [24,25].Historical residential segregation (redlining) and reduced government and commercial investments have created significant food-security risk factors for neighborhoods with predominantly Black residents [26].Previous research shows that at equal levels of poverty, Black communities have the lowest presence of food storesresulting in the lowest access of quality food among any racial group [27].Studies have also found that ethnoracial minorities in communities that experienced historical redlining also have limited access to educational and employment opportunities, creating cyclical disparities that likely maintain or increase rates of child abuse [18].Finally, and as anticipated, families with lower socioeconomic status were more likely to experience food insecurity and exposure to violence.This is explained by a lack of financial resources creating a significant barrier to food accessibility, therefore creating an increase in household stress and exposure to violence.Interestingly, as age increased, so did the odds of experiencing food insecurity and exposure to violence; however, within the data, food insecurity was assessed as occurring over the past 12 months, while exposure to violence was assessed as whether the child had ever experienced such events.Thus, as the child's age increases, there is a greater amount of time for the child to have experienced exposure to violence.

Recommendations
There are many hunger-relief organizations with a shared mission to decrease food insecurity, including Feeding America, the nation's largest [28].Despite the altruistic mission to reduce hunger, these organizations are often criticized for providing nutritionally empty, energy-dense foods, and a continuously insufficient supply to meet the demand of families experiencing food insecurity [29].Further, delivery of food is typically made by whatever   means of transportation agencies can provide, which has forced agencies to utilize personal vehicles that often lack refrigerationposing a great challenge for agencies in rural areas [29].The shortcomings displayed by these organizations provide significant evidence for the need of greater funding, resources, and policies addressing food insecurity.Because osteopathic physicians are more likely to work in these rural communities, upholding osteopathic values through providing patient-centered care that considers both the community and environmental impact is essential [30].This includes increased screenings for food insecurity, providing nutritional education to patients as well as patient advocacy and outreach for greater nutritional support for rural areas.
State-funded and regulated farm-to-food bank programs offer a potential strategy to address insufficient supplies and a lack of quality nutrients in the foods typically donated.In 2018, the Growing Together Illinois program was implemented across Illinois, which utilized fresh produce donations for 17 participating food pantries [31].The program received not only positive feedback regarding the high quantity and quality of food provided but also an overall increase in the nutrition assessment score at participating pantries [31].The Agriculture Improvement Act of 2018 was recently amended to allocate funding for Farm to Food Bank Projects for State Emergency Food Assistance Program agencies for the 2023 fiscal year [32].This will not only alleviate the lack of nutritious foods but also provide funding for the harvesting, processing, packaging, and transportation of foods, as well as building relationships between agricultural producers and processors with local food banks [32].
In addition to organizations that directly distribute food, food assistance programs, including the Supplemental Nutrition Assistance Program (SNAP), are aimed at reducing food-related hardships [33].Unfortunately, the current program requirements for SNAP exclude many individuals who would benefit from the program.While the exact qualifications for SNAP vary by state-specific regulations, the general criteria are based on income and work requirements [33].Although poverty and food insecurity are highly correlated, they are not synonymous, and many household variables may contribute to food insecurity despite adequate income.Additionally, numerous barriers prevent many eligible households from applying or recertifying for SNAP benefits [33,34].Limited transportation or lack of the necessary technology to apply disproportionately affects disadvantaged individuals living in rural areas, further from SNAP offices [34].An analysis funded by the Robert Wood Johnson Foundation found that with just two policy adjustments, one in 11 SNAP households would lose eligibility, and the ramifications would disproportionately affect groups of children in rural communities and children of color [35].This prompted the recommendations to utilize more objective data and economic realities to determine eligibility, expansion of SNAP-Ed, and financial incentive programs, as well as to ensure that any changes made to SNAP policy do not disproportionately affect certain groups [35].In addition to these recommendations, we urge policymakers and program administrators to expand enrollment for SNAP participation through the adaptation of more lenient program rules [33].
Culture plays a role in the evaluation of food security in many ways: the classification of nutritious food sources, and the traditionally produced crops of that region, among others [36].Implementing culturally competent initiatives is essential to the success of food-insecurity prevention programs, which may secondarily lower child abuse rates.For example, the Culturally Responsive Food Initiative (CRFI), implemented in August 2020 by the Food Bank of the Rockies, ran a pilot study aiming to overcome barriers experienced by food bank clients of different cultural backgrounds [37].The program identified cultural barriers and developed customized food lists offering culturally responsive foods for their clients [37].By implementing strategies that similarly address cultural variations, it is possible to address disparities created by structural and systemic racism contributing to food availability and utilization, and as an indirect result, child abuse [38].
Finally, additional research is needed to create a more robust analysis of the barriers to food security among historically disempowered groups.Previous studies have identified low native food securitythe accessibility to the desired native foodsas a predictor to food insecurity [39].Therefore, we urge policymakers and state and federal stakeholders to expand funding and develop collaborative research partnerships with tribal leaders and scholars that integrate native food security measures.Additional research and policy is also necessary for understanding and addressing the racialized patterning of food insecurity in the United States.Black communities experience a significant food store shortage, directly contributing to food accessibility.We urge policymakers, community leaders, and business owners to develop food-related policies and practices directed toward addressing the disparities in minority communities [40].Partnerships between local government and supermarket leaders should be developed to increase access to food within neighborhoods that have been historically overlooked by food retailers [40].

Limitations
A significant limitation is that all data are based on parental or primary caregiver responses that have not been independently verified.Additionally, it is important to note the nature of reporting food insecurity and bias.Studies have found food insecurity to be a recurrent state but not a constant [41].Another limitation is that the survey includes noninstitutionalized children only and excludes families experiencing homelessness or children residing in group homes.Therefore, food insecurity and exposure to violence rates are likely underestimated.It is probable that certain ethnic populations, including Indigenous populations, are underrepresented due to being identified as Multiracial in the NSCH survey.Therefore, it is possible that the impact of exposure to violence and food insecurity in these populations is greater than what is reported within the dataset.Additionally, other studies have shown that reports of child abuse substantiation differ between races [42].

Conclusions
Childhood exposure to violence and food insecurity are both complex issues resulting in long-term detrimental effects for those who experience them.The association between the two is significant, with the odds of exposure to violence being 5.19 times greater in food-insecure children compared to food-secure children.The interaction between exposure to violence and food insecurity also disproportionately impacts children with specific demographic factors, notably race/ethnicity including Multiracial, Indigenous, and Black children.By developing and adapting strategies to improve food security, it is possible to indirectly reduce the rates of childhood exposure to violence.
Research ethics: This study did not meet the requirements for human subjects research according to the regulatory definition of human subject research as defined in 45 CFR 46.102(d) and (f) of the Department of Health and Human Services' Code of Federal Regulations (https://www.hhs.gov/ohrp/regulations-and-policy/regulations/45-cfr-46/), and therefore was not submitted for ethics review.Results were reported according to Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines.Informed consent: Not applicable.Author contributions: All authors provided substantial contributions to conception and design, acquisition of data, or analysis and interpretation of data; all authors drafted the article or revised it critically for important intellectual content; all authors gave final approval of the version of the article to be published; and all authors agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.
Competing interests: Dr. Hartwell receives research funding from the National Institute of Child Health and Human Development (U54HD113173; Shreffler), Health Resources and Services Administration (U4AMC44250-01-02, PI: Audra Haney; R41MC45951 PI: Hartwell), and previously from the National Institute of Justice (2020-R2-CX-0014 PI: Beaman)-all unrelated to the current presentation.Research funding: This research was supported in part by the Health Resources and Services Administration (U4AMC44250-01-02, Audra Haney, PI).The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health or the Health Resources Services Administration.

Table  :
Demographics of children in the United States based on food security.

Table  :
The percentage of groups that has reported exposure to violence and associations with child food insecurity (n=,; n=,,).

Table  :
Interaction terms for race and food security on exposure to violence controlling for food security, age, percent FPL guidelines, and urbanicity.
AOR, adjusted odds ratio; CI, confidence interval; FPL, federal poverty level.Bloom et al.: Association and disparities of food insecurity and exposure to violence