Emergency Allotments in SNAP and Food Hardship Among Households With Children

Key Points Question Were emergency allotments in the Supplemental Nutrition Assistance Program (SNAP) associated with a change in the risk of food hardship among households with children? Findings In this cross-sectional study of 44 753 households with incomes 130% or less of the federal poverty level and with children younger than 18 years, implementing emergency allotments in SNAP was associated with a decreased risk of food hardship among SNAP-participating households compared with households that did not participate in SNAP. This decreased risk was observed overall and for households with Hispanic and White children but not for households with Black children. Meaning These results suggest that emergency allotments in SNAP may have contributed to a decreased risk of food hardship among households with children and indicate that efforts are needed to ensure that all populations benefit from economic policies.


Introduction
Food insecurity, or limited access to adequate food due to lack of money or resources, is a critical public health issue in the US.1][12][13] Racial and ethnic disparities in experiences of food insecurity mean that these potential adverse health outcomes may disproportionately affect individuals belonging to minoritized racial and ethnic groups. 2,3e Supplemental Nutrition Assistance Program (SNAP) is the largest programmatic and policy strategy to alleviate food insecurity in the US.Currently, more than 20 million low-income households participate in SNAP and receive a monthly benefit to assist with the cost of purchasing food. 14[20][21] At the onset of the COVID-19 pandemic, amid concerns that rising unemployment would contribute to an increased need for food purchasing assistance, the federal government implemented emergency allotments in SNAP.Beginning in March 2020 under the Families First Coronavirus Response Act, states could request waivers allowing them to provide all SNAPparticipating households with the maximum monthly benefit possible given household size. 22By May 2020, all states had requested and received approval to implement these emergency allotments. 23The federal government issued additional guidance in April 2021, specifying that all SNAP-participating households were to receive a minimum of $95 in monthly SNAP benefits, regardless of the maximum monthly benefit possible for their household size.Recent research has shown that emergency allotments decreased the risk of food insecurity among adults participating in SNAP in the early months of the pandemic. 24At the federal level, emergency allotments in SNAP ended in March 2023, but some states chose to end them early in 2021 or 2022. 22Initial evidence has indicated that the percentage of adults reporting that their household did not have enough food to eat increased after these states ended emergency allotments early, suggesting that emergency allotments may have protected against increases in difficulties affording food during the pandemic. 25e association of emergency allotments in SNAP with food hardship among households with children, a particularly vulnerable population, has not been examined.The aim of this study was to assess whether implementing temporary emergency allotments in SNAP was associated with a change in the risk of food hardship among households with children.

Data Source and Study Population
This ecologic cross-sectional study used population-representative data from the 2016-2022 National Survey of Children's Health (NSCH), an annual survey designed to produce national and state estimates of child and family health and well-being.This study was considered exempt from approval and the need for informed consent due to the use of publicly available data by the institutional review board at the University of North Carolina.This study follows the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for crosssectional studies.
For the NSCH, the US Census Bureau contacts a random sample of households each year by mail to identify households with at least 1 child younger than 18 years. 26For households with multiple children, 1 child is randomly selected to be the survey subject.An adult member of the household (ie, caregiver) who is familiar with the child's health, usually a parent, completes the survey on paper or online in English or Spanish. 26Surveys are completed from June to January of each year (eg, 2021 surveys are completed June 2021 to January 2022).In this study, we included households with incomes 130% or less of the federal poverty level (FPL), the income limit for SNAP eligibility at the federal level. 27

Exposure and Outcome
The NSCH assesses food hardship using a single question that asks caregivers, "Which of these statements best describes your household's ability to afford the food you needed during the past 12 months?"Response options include "We could always afford enough 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."We considered the latter 3 responses, which capture both the quality (ie, the household not being able to afford the kinds of foods they should) and quantity (ie, the household's ability to afford enough food) of food, 28 to indicate household food hardship.The NSCH also asks caregivers whether at any time during the past 12 months, even for 1 month, anyone in their family received food stamps or SNAP benefits, with response options of yes or no.We compared food hardship for households participating in SNAP and income-eligible households not participating in SNAP from before (2016-2019) to during (2020-2022) implementation of emergency allotments.

Confounders
We created a conceptual diagram to identify factors associated with both SNAP participation and household food hardship (eFigure and eTable 1 in Supplement 1).Prior research has indicated that there are systematic differences between income-eligible households that do and do not participate in SNAP.0][31] To account for factors that are likely associated with both the likelihood that households participate in SNAP and household food hardship, we adjusted analyses for household receipt of cash assistance from a government program in the past 12 months, employment status of adults in the household, and household income relative to the FPL.In addition, given that food hardships and poverty disproportionately affect minoritized racial and ethnic populations, 1,14,32 we adjusted analyses for caregiver-reported child race and ethnicity (Black, Hispanic, White, or other [including American Indian or Alaska Native, Asian Indian, Chinese, Filipino, Guamanian or Chamorro, Japanese, Korean, Native Hawaiian, Other Asian, Pacific Islander, Samoan, and Vietnamese]).
Caregiver race and ethnicity are not captured in the NSCH.We also adjusted analyses for time-varying state-level economic policies that may contribute to household food hardship and thus SNAP participation, including state minimum wage, 33,34 refundable Earned Income Tax Credit rate, 35,36

Statistical Analysis
We used a difference-in-differences approach to compare changes in the risk of food hardship from before to during implementation of emergency allotments in SNAP among income-eligible households that did and did not participate in SNAP.We constructed a log-binomial regression model to calculate risk ratios (RRs) and corresponding 95% CIs.We specified the model as follows: where i indexes households, s indexes states, and t indexes time; Y ist is a binary indicator for household food hardship for household i in state s at time t, and g is the log-link function defined as g(u) = ln(u); SNAP i is a binary indicator for whether household i participated in SNAP in the past 12 months; and EA t is a binary indicator for whether the survey responses corresponded to years when emergency allotments in SNAP were implemented (ie, 2020-2022).The coefficient for the interaction term between SNAP i and EA t , β 3 , is the difference-in-differences estimator and indicates whether the change in the risk of food hardship from before to during implementation of emergency allotments in SNAP differs for income-eligible households that did and did not participate in SNAP.nonparticipating households from before to during implementation of emergency allotments.The variable X i is a vector of household characteristics, and X st is a vector of time-varying state economic policies, as described above, included to adjust for potential confounding.Given persistent racial and ethnic disparities in food hardships, [1][2][3] we constructed models overall and separately for households with Black, Hispanic, and White children.

Negative values of β
A key assumption in the difference-in-differences approach is the parallel trends assumption, or that in the absence of emergency allotments in SNAP, trends in food hardship would have been similar between SNAP-participating and nonparticipating households.Our assessment of the parallel trends assumption indicated that it was met, conditional on potential confounders (eMethods in Supplement 1).
We used to determine eligibility. 39We conducted sensitivity analyses that additionally adjusted for state adoption of these policies.Fifth, we conducted sensitivity analyses among households with incomes 200% or less of the FPL given that some states increased the income limit up to 200% of the FPL under BBCE.Finally, we conducted sensitivity analyses using low or very low food insufficiency, defined as responses of "sometimes we could not afford enough to eat" or "often we could not afford enough to eat" during the past 12 months, 27,40,41 as the outcome.
Adjusting for potential confounders, implementing emergency allotments in SNAP was associated with a decreased risk of food hardship among SNAP-participating households with children compared with income-eligible nonparticipating households with children (RR, 0.88; 95% CI, 0.81-0.96)(Figure 2; Table 2).Results were similar in sensitivity analyses excluding the 17 states

JAMA Network Open | Public Health
Emergency Allotments in SNAP and Food Hardship Among Households With Children

Discussion
The results of this ecologic cross-sectional study show that emergency allotments in SNAP were associated with a decreased risk of food hardship among households with children that participated in SNAP compared with income-eligible households with children that did not participate in SNAP.
5][46][47] Importantly, we observed a decreased risk among households with Hispanic and White children but not among households with Black children, suggesting that the potential benefits of emergency allotments were not equally realized for all populations.
Recent studies have shown decreases in the risk of difficulty affording food among SNAPparticipating households during implementation of emergency allotments. 24,25Our results are consistent with these studies and add a focus on households with children, a high-risk population for food hardship. 1Collectively, these results reinforce the critical role of SNAP in mitigating food hardship, particularly during economic crises. 48In addition to emergency allotments in SNAP, other economic supports were implemented during the COVID-19 public health emergency, including an expanded Child Tax Credit (CTC) in 2021. 49Recent research has shown that the expanded CTC was associated with reductions in the likelihood that households with children did not have enough food to eat. 40,41,50Importantly, the expanded CTC was available to all low-income households with children, including those that did and did not participate in SNAP, and thus likely did not systematically affect our results.
We observed a decreased risk of food hardship among SNAP-participating households with Hispanic children and White children compared with income-eligible nonparticipating households during implementation of emergency allotments.However, we did not observe this decreased risk among households with Black children.Food environments are racially segregated, with fewer grocery stores, limited access to healthy food options, and higher food prices in predominantly Black communities. 51,52In addition, among SNAP-participating households with children, Black households report needing a larger increase in SNAP benefits to afford enough food than White households ($51 vs $42 more per week). 53Segregated food environments, as well as the larger gap between the amount of SNAP benefit received and the amount needed, may have limited the potential for emergency allotments to contribute to reductions in food hardship among households with Black children.Furthermore, in our data, households with Black children were disproportionately represented among those with incomes 0% to 65% of the FPL.In SNAP, monthly benefit amounts are proportional to income such that households with lower incomes receive a higher benefit up to a maximum benefit possible given household size. 54SNAP-participating households that were already receiving the maximum benefit possible did not experience an increase in food purchasing assistance under emergency allotments from March 2020 to April 2021.Only those with monthly benefits less than $95 experienced an increase in food purchasing assistance after April 2021. 55If more households with Black children than households with Hispanic or White children were already receiving the maximum benefit possible, they may not have received the same level of economic support from emergency allotments.Existing research has confirmed that among SNAP-participating households with children, Black households were more likely than White households to already be receiving the maximum benefit possible prior to implementation of emergency allotments. 56This research has also shown that Black households experienced a smaller increase in monthly benefit amount during implementation of emergency allotments, including when the minimum monthly benefit was increased to $95 for all households, than White households. 56

Limitations
This study has several limitations.First, though we adjusted multivariable analyses for a comprehensive set of confounders, there is potential for unmeasured or residual confounding specifically due to unmeasured or poorly measured factors that affect both SNAP participation and food hardship.Because income-eligible households that participate in SNAP are more likely to experience more severe food hardship and have fewer economic resources than those that do not participate in SNAP, this unmeasured or residual confounding would likely bias our results toward the null.Second, the NSCH measures food hardship over the past 12 months.There may have been issues with recall, though we do not expect that recall would have systematically differed for households that did and did not participate in SNAP.In addition, given the single report of food hardship over the past 12 months, we were unable to precisely align reports of food hardship with the month of emergency allotment implementation.Third, the NSCH did not include a validated measure of food hardship, such as the US Department of Agriculture's 18-item measure of food insecurity. 28Fourth, the NSCH only collects information on child race and ethnicity, which may not be the same as the race and ethnicity of the child's caregiver or other members of the household.Finally, given the lack of data at a monthly level, we were not able to examine the potential implications of changes in emergency allotment guidance in spring 2021. 55

Conclusions
The results from this ecologic cross-sectional study suggest that policies that provide economic support to children and their families, such as emergency allotments in SNAP, may benefit population-level health.These results can be used to inform ongoing discussions regarding the potential need to permanently increase SNAP benefits. 57,58Future research is needed to examine the implications of the end of emergency allotments in SNAP on child and family food hardship and to understand strategies to ensure that the potential benefits of such economic support policies are realized for all populations.
3 , corresponding to RRs of less than 1.0, indicate that the risk of food hardship decreased among SNAP-participating households compared with nonparticipating households from before to during implementation of emergency allotments (ie, implementing emergency allotments in SNAP was associated with a decreased risk of food hardship among SNAP-participating households relative to nonparticipating households).Positive values of β 3 , corresponding to RRs greater than 1.0, indicate that the risk of food hardship increased among SNAP-participating households compared with nonparticipating households from before to during implementation of emergency allotments.A β 3 value of 0, corresponding to an RR of 1.0, indicates no difference in the change in the risk of food hardship among SNAP-participating households compared with

Figure 1 .
Figure 1.Prevalence of Food Hardship in the Past 12 Months Among Households With Income of 130% or Less of the Federal Poverty Level, Stratified by Supplemental Nutrition Assistance Program (SNAP) Participation100

JAMA Network Open | Public Health Emergency
Allotments in SNAP and Food Hardship Among Households With Children 37,38Network Open.2024;7(8):e2428680.doi:10.1001/jamanetworkopen.2024.28680(Reprinted)August16, 2024 3/12 Downloaded from jamanetwork.combyguest on 08/30/2024maximum Temporary Assistance for Needy Families benefits for a family of 3, 37 and Medicaid expansion.37,38 conducted multiple sensitivity analyses.First, because 10 states ended emergency experiencing food hardship in 2019, compared to 2018, that was not observed among incomeeligible SNAP-participating households (Figure1), we conducted sensitivity analyses excluding 2019.Fourth, under broad-based categorical eligibility (BBCE), states can expand SNAP eligibility by increasing the income limit from 130% to up to 200% of the FPL and/or eliminating the asset test JAMA Network Open | Public Health Emergency Allotments in SNAP and Food Hardship Among Households With Children JAMA Network Open.2024;7(8):e2428680.doi:10.1001/jamanetworkopen.2024.28680(Reprinted) August 16, 2024 4/12 Downloaded from jamanetwork.comby guest on 08/30/2024 42,43cted analyses using SAS, version 9.4 (SAS Institute Inc) and Stata, version 17.0 (StataCorp LLC) statistical software.We used NSCH sampling weights to account for the probability of household selection and nonresponse and to generate population-representative results.For analyses among households with Black, Hispanic, and White children, we used the subpopulation estimation in Stata to ensure correct calculation of standard errors.To interpret results, we relied on the magnitude of the adjusted RR and the width and location of the corresponding 95% CIs, in alignment with guidance from the American Statistical Association.42,43ResultsOf44 753 households with incomes 130% or less of the FPL, a weighted 23.4% had Black children, 56.7% had White children, and 19.9% had children of other races.More than one-third of households (37.8%) had Hispanic children, and 31.8% had young children aged 0 to 5 years.From 2016 to 2022, the percentage of households with children that experienced food hardship during the past 12 months was consistently higher among those that participated in SNAP compared with those that did not (Figure1; eTable 2 in Supplement 1).From 2016 to 2021, the percentage of households that experienced food hardship decreased from 62.9% to 48.2% among those participating in SNAP and from 44.3% to 38.9% among those not participating in SNAP.The percentage of households that experienced food hardship increased in 2022 to 58.0% among SNAP-participating households and to 47.5% among nonparticipating households.Over the entire study period, a higher percentage of SNAP-participating households compared with nonparticipating households received cash assistance from a government program (18.1% vs 3.1%), did not have an employed adult (30.8% vs 14.0%), had incomes between 0% and 65% of the FPL (53.5% vs 38.5%), and had Black children (30.3% vs 15.1%) (Table

Table 2 )
. Implementing emergency allotments in SNAP was not associated with a change in the risk of food hardship among SNAP-participating households with Black children compared with income-

Table 2
Adjusted for household receipt of cash assistance from a government program in the past 12 months, employment status of adults in the household, household income relative to the FPL, child race and ethnicity, and state minimum wage, refundable Earned Income Tax Credit rate, maximum Temporary Assistance for Needy Families benefit for a family of 3, and Medicaid expansion. a Future research is needed to examine whetherJAMA Network Open | Public HealthEmergency Allotments in SNAP and Food Hardship Among Households With Children Downloaded from jamanetwork.comby guest on 08/30/2024 ensuring a minimum monthly SNAP benefit of at least $95 for all participating households is associated with changes in the risk of food hardship.