Association Between Material Hardship in Families With Young Children and Federal Relief Program Participation by Race and Ethnicity and Maternal Nativity

Key Points Question What was the association between participation in federal COVID-19 relief programs and mitigation of food and housing hardships and disparities among families with young children? Findings In this cohort study using interviews before and during the COVID-19 crisis of 1396 caregivers with children 48 months or younger, receipt of benefits from the Supplemental Nutrition Assistance Program and/or Economic Impact Payments was associated with reduced prevalence of household food insecurity and being behind on rent. However, there were disparities in receipt of benefits as well as food insecurity and being behind on rent by maternal nativity and caregiver race and ethnicity. Meaning The findings of this study suggest that equity-driven policy design is needed to ensure that benefits reach all families, especially among historically marginalized groups.


Introduction
Economic hardships-including inability to afford food or rent-are strongly associated with adverse health outcomes across the life span. [1][2][3] Food insecurity, defined as inadequate access to enough food for all household members to live active, healthy lives, is associated with poor physical and mental health among children and adults, developmental delays in children, increased hospitalizations, and avoidable health care costs. 1,[4][5][6] Inability to pay rent on time places families at risk of eviction and is associated with poor physical health among children and adults, maternal depression, and child hospitalizations. 3 Young children are particularly vulnerable to adverse outcomes associated with hardship. 3,7,8 The rapidly growing brain of an infant or toddler requires consistent high-quality nutrition and stable housing for healthy development. 9 Even brief periods of hardship and stress during early childhood may have lifelong consequences. 8,10 The onset of the COVID-19 pandemic initiated a period of widespread job loss and disruption due to business, school, and child care closures necessary to slow the rapid spread of the virus.
Research conducted early in the pandemic documented that families with young children reported struggling to meet basic needs like food and rent due to pandemic-related closures and economic distress. 11,12 National surveys found that families with children, Black and Latino adults, and immigrants consistently reported higher rates of economic hardship before the pandemic 13 and disproportionately experienced hardships during the pandemic compared with the population overall. These inequities reflect persistent structural racism and xenophobia. 14,15 To mitigate the effects of the COVID-19 crisis, the US Congress passed a series of relief packages between March 11, 2020, andMarch 25, 2021, allocating trillions of dollars in aid to bolster the US economy and alleviate economic hardships for families and individuals. Federal relief efforts, including investment in nutrition assistance and direct cash transfers, buffered US households from hardships during the crisis. 16 Among the largest investments to facilitate households with low and moderate incomes meeting basic needs were streamlined access to and boosted benefit levels of the Supplemental Nutrition Assistance Program (SNAP), the largest antihunger program in the US, and rapid deployment of Economic Impact Payments (EIPs), also known as stimulus checks. During the early pandemic, SNAP enrollment increased from 36 million people in 2019 to 44 million in 2020 as the program expanded, as designed, to meet the needs of people experiencing food insecurity. 17 The Internal Revenue Service disbursed 3 EIPs to over 92 percent of households over the course of 1 year in amounts ranging from $600 to $1400 per adult and $500 to $1400 per child. 18,19 While there were no restrictions on the use of payments, families, especially those with lower incomes, used payments to meet basic needs, potentially alleviating economic hardships. 20 Multiple relief policies, including SNAP and distribution of EIPs, excluded or limited access to immigrants, imposed administrative barriers to participation for families with low incomes, and were inadequate for families with the greatest needs. These practices disproportionately affected families with Black, Latino, or immigrant members. 21 Anti-immigrant rhetoric, exclusionary immigration policies, and changing eligibility rules for immigrants in relief programs resulted in millions of families with immigrant members not receiving public assistance that could have mitigated hardships. [22][23][24] Few existing studies specifically focus on experiences of families with young children by race and ethnicity and maternal nativity. 25 The current study uses prospective data to examine the association of federal relief programs with mitigation of food insecurity and being behind on rent and illuminates disparities in program participation. We hypothesized (1) that there would be disparities in SNAP and EIP participation with lower participation among families with caregivers who were Black or Latino or that had an immigrant mother, (2) that SNAP and EIP participation would be associated with less food insecurity and being behind on rent compared with nonparticipation, and (3) that, in exploratory analysis, disparities in SNAP and EIP participation would partially account for disparities in food insecurity and being behind on rent during the COVID-19 crisis.

Study Population
Data come from the ongoing repeat cross-sectional Children's HealthWatch (CHW) study, conducted since June 23, 1998, 26 and the CHW Follow-up Study (CHW-COVID), a longitudinal sentinel cohort study of families surveyed in English or Spanish before and during the pandemic. Baseline questionnaires were administered face to face between January 1, 2018, and March 20, 2020 (baseline) in an emergency department or primary care clinic in 5 cities: Boston, Massachusetts; Baltimore, Maryland (with 2 sites, 1 of each type); Minneapolis, Minnesota; Philadelphia, Pennsylvania; and Little Rock, Arkansas. 27 Each research site received institutional review board approval annually. Eligibility criteria at all sites included state residency, child age of 48 months or younger, English-or Spanish-speaking primary caregiver, caregiver residence in the child's household, and consent to be interviewed (written or verbal, depending on the site).
Questionnaires 26,28 were administered by trained interviewers and included sociodemographic characteristics, household material hardships, and public benefit program participation. The telephone CHW-COVID follow-up questionnaires (administered between September 1, 2020, and June 30, 2021) included questions about material hardships and program participation, including SNAP and EIP, since the start of the crisis. 28 There were 2 versions of the follow-up questionnaire. If the same caregiver in the household was interviewed, the first version and the baseline sociodemographic data were used. If a different caregiver in the household (eg, the father vs the mother) was interviewed, the second version, which included that individual's sociodemographic information (eg, race and ethnicity), was used. The study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cohort studies.

Measures
In both baseline and follow-up questionnaires, caregivers reported on household food insecurity, 29 using the validated US Household Food Security Survey Module: Six-Item Short Form, 30 and on whether they were behind on rent or mortgage (hereinafter behind on rent), a risk factor for housing instability, including eviction. For household food insecurity, 0 or 1 question affirmed was coded as food secure and 2 or more questions affirmed was coded as food insecure. For both food insecurity and being behind on rent, the reference period at baseline was within the past 12 months and at follow-up was since the start of the COVID-19 pandemic. Both hardship outcomes were scored dichotomously and analyzed separately.
based on their answers and categorized as Black, non-Latino; Latino; White, non-Latino; or other or multiple non-Latino race or ethnicity (hereinafter other race or ethnicity), which included American Indian or Native American and Asian or Pacific Islander participants, although the numbers of participants in this category were too small to analyze independently. Data on race and ethnicity were collected because it is important to understand how the COVID-19 crisis and related public policy and associations with family health and hardship varied by race and ethnicity. Maternal nativity was categorized as US born (including US territories) or immigrant. Covariates, chosen a priori, were caregiver education, number of children in the household, child age, children's health insurance type (as a proxy for household income), and whether anyone in the household was employed.

Statistical Analysis
We tested our first hypothesis, that there would be disparities in SNAP and EIP participation, with covariate-adjusted multinomial logistic regression analysis. We used the category neither SNAP nor EIP as the reference outcome group and included either race and ethnicity or maternal nativity as the predictors of interest. Results are presented as adjusted odds ratios (aORs) with 95% CIs. The independence of irrelevant alternatives assumption of multinomial logistic regression requires that there be no other alternative outcome choices that, if present, would change the odds of the outcome choices examined. For our analysis of SNAP and EIP participation, there were no competing benefit programs that would replace the value of SNAP or EIP.
We tested our second hypothesis, that SNAP and EIP participation would be associated with less food insecurity and being behind on rent, using adjusted log-binomial regression models with household food insecurity and being behind on rent as outcomes and program participation as the predictor of interest. Results are presented as adjusted prevalence ratios (aPRs) with 95% CIs. Finally, we tested our third hypothesis, that disparities in hardships might be partially attributable to disparities in SNAP and EIP participation, using log-binomial regression models with household food insecurity and being behind on rent as outcomes. Results are presented as prevalence ratios (PRs) with 95% CIs. We used an exploratory mediation analysis approach in which we compared race and ethnicity-and nativity-based disparities in the outcomes before and after adjustment for program participation; attenuation of the disparities after adjustment for program participation would support our hypothesis. We assessed statistical significance using 2-tailed tests with α = .05; 2-sided P values represent results from comparing each characteristic by level of program participation. Each test was performed using χ 2 analysis for categorical characteristics and analysis of variance for independent samples. Data were analyzed using SAS, version 9.4 (SAS Institute Inc).

Results
Out of 6875 eligible caregivers interviewed in person before the pandemic and recontacted, 1396 participated in the telephone follow-up (20.3% response rate). The mean (SD) time between baseline and follow-up questionnaire was 24.0 (7.8) months. The follow-up cohort did not differ from the prepandemic baseline cohort in household income, marital status, SNAP participation at baseline, or caregiver or household employment. The proportion of Latino and immigrant caregivers in the follow-up cohort was higher than at baseline. Among the 1396 caregiver-child dyads, race and ethnicity data were available for 1357 caregivers; 514 (37.9%) caregivers were Black, non-Latino; 558 (41.1%) were Latino; 230 (16.9%) were White, non-Latino; and 55 (4.1%) were of other race or ethnicity. Among 1390 responses with nonmissing data, 417 (30.0%) children had an immigrant mother ( Table 1) Over one-third of caregivers (467; 33.5%) reported household food insecurity during the COVID-19 crisis compared with 292 (20.9%) at baseline (Table 1). Being behind on rent also increased dramatically during the crisis, more than doubling, from 241 (17.9%) at baseline to 567 (40.8%).

Discussion
This cohort study yields timely, policy-relevant information for advancing health equity among families with young children, an understudied population. We found that families with immigrant mothers had lower odds of COVID-19-related relief program participation than families with US-born mothers, although families with immigrant mothers, as well as those with Black or Latino caregivers (compared with White, non-Latino mothers), were more likely to experience household food insecurity or being behind on rent, hardships relief programs were designed to address. In this sample, differential participation in the relief programs examined did not fully account for the disparities in household hardships.
In the US, one-quarter of children live with at least 1 immigrant parent and more than 40% of otherwise eligible lawfully present immigrants who had arrived in the US within the previous 5 years remained ineligible for the program. [40][41][42] In addition to exclusions and barriers in relief policies, the disparities observed in this study among families with immigrant mothers may also have been associated with well-documented, 40 This study may prove instructive for ongoing policy conversations related to another cash transfer program administered through the US tax system, the advance Child Tax Credit (CTC). The advance CTC was more widely available to families with children than EIPs and was associated with positive outcomes. [46][47][48] Legislation to expand the advance the CTC and remove immigration exclusions and reduce barriers to access may provide a transformative opportunity to reduce disparities and advance equity among families with young children.
Further research is necessary to understand how subsequent changes in policy design, including increases in SNAP benefits and the implementation of a third and more inclusively distributed EIP, acting together with other COVID-19-era policies like the disbursement of advance CTC payments, may impact disparities in household hardships among families with young children.
The rapidly changing policy landscape and the sunsetting of many relief policies may deepen disparities in economic hardships by race, ethnicity, and nativity over time.

Strengths and Limitations
This study has several strengths, including its multisite, prospective design with wide diversity of participant race, ethnicity, and nativity. Moreover, the study provides data on an underrepresented group of families with young children that is both difficult to reach through online questionnaires, given generally unreliable internet access, and highly vulnerable to economic hardship. 49 This study also has limitations. The sample is not nationally representative, although it is drawn from a sentinel study design intended to monitor a population at highest risk of showing early signs of either benefit or harm 9 from a given exposure. The sample is similar in family demographic characteristics and child health outcomes to low-income families in the National Survey of Children's Health. 50 All hardships, program participation, and other data were self-reported and thus subject to recall bias, although the food security measure is validated 30 and the measure of being behind on rent has strong criterion validity and has been used in several studies. 3,51 The other race or ethnicity category had a small sample size and should be interpreted with caution. We were not able to further disaggregate racial and ethnic groups because of small sample sizes, which also resulted in wide CIs for some outcomes.

Conclusions
This cohort study found significant race and ethnicity-and nativity-based disparities in food and housing-related hardships among families with young children during the COVID-19 crisis, coupled with significant differences in participation in relief programs to address these hardships. Inclusive, equity-driven policy design is as important as rapid deployment of direct cash transfer payments to ensure that benefits reach all families, especially historically marginalized groups. Specifically, legislators and other leaders must enact and equitably implement robust policies that reduce disparities by removing immigration-related exclusions and eliminating access barriers for families with very low incomes and Black and Latino families. The pandemic brought great economic and societal change, exacerbating preexisting inequities. Thus, the times require equally great policy change to not just reduce but eradicate inequities in the US to ensure that all families with children can afford the food and housing they need to be healthy.