Public Benefit Use and Social Needs in Hospitalized Children With Undocumented Parents

We conducted a cross-sectional, anonymous, survey-based study at a large quaternary care children’s hospital in Houston, Texas, from March 2019 to September 2019. We included English- and Spanish-speaking primary caregivers of children ≤18 years old admitted to general inpatient pediatric services. Subject recruitment took place on weekdays when study team members did not have clinical commitments. Caregivers had the option of completing the survey in English or Spanish on an electronic tablet (using Research Electronic Data Capture), on paper, or verbally with assistance from a study team member.^33,34 

The Baylor College of Medicine Institutional Review Board approved this study. Because of sensitive content (ie, immigration status), we obtained a waiver of signed consent. There was no financial compensation.

Factors that contribute to disparities in immigrant health service use include health care need, resources, predisposing factors, and macrostructural and contextual conditions, as described by Yang and Hwang.³⁵  Consistent with this framework, we designed an 89-item survey assessing child and caregiver demographics, health and resource use, barriers to resource use, and immigration-related fear (Supplemental Fig 3). Survey questions not found on validated tools were developed by the study team on the basis of content expertise and adaptations from existing literature.^36–38  The survey was created in English and translated into Spanish. Semistructured interviews with caregivers (17 in English, 10 in Spanish) were completed to assess the validity of the survey (Supplemental Table 6). Test–retest reliability with 7 English-speaking caregivers demonstrated good reliability.

Child demographic variables included age, sex, native country of origin, and immigration status. Caregiver demographic variables included age, relation to the child, race and ethnicity, marital status, education level, employment, household income, number of children in the household, primary language, native country of origin, and immigration status. Immigration status was addressed with 2 questions. The first asked about the participant’s immigration status with answer choices: “US citizen,” “lawful permanent resident,” “work or school visa,” “refugee/asylee,” “seeking asylum (in process),” “temporary protected status,” or “undocumented (without papers).” The second asked whether the participant’s child had 1 or more parents or guardians who are living in the United States and undocumented (without papers). If a participant responded with “seeking asylum” or “undocumented” for the first question or “yes” to the second question, the family was considered to have 1 or more undocumented parent.

Health and resource use variables included perceived child health, number of emergency department visits and admissions, on-time well-child checks, food security, caregiver stress, and current or past Medicaid or CHIP, SNAP, and WIC use. Food security was measured with the US Department of Agriculture Household Food Security Survey Module.³⁹  Caregiver stress was measured by using the Perceived Stress Scale.^40,41  When applicable, caregivers were asked to select reasons why their child had a delayed well-child check or was not enrolled in Medicaid and CHIP, SNAP, or WIC. Several barriers were listed (including perceived ineligibility due to family income, perceived ineligibility due to immigration status, logistic barriers, and immigration concerns), and families could select more than one response.

Questions about immigration-related fear assessed whether families experienced discrimination, had a family member or friend who was deported in the past year, feared deportation or separation from family, or feared that work, school, or child hospitalization would affect immigration status. We also inquired whether families stopped services from WIC, SNAP, Temporary Assistance for Needy Families, or Medicaid and CHIP because of concerns “about having immigration or citizenship problems.”

All statistical analyses were performed by using SAS (version 9.4; SAS Institute, Inc, Cary, North Carolina). In our first analysis, our exposure variables were demographics, health and resource use, barriers to resource use, and immigration-related fear, with parent documentation status (with versus without undocumented parents) as the outcome. We then performed 3 subanalyses comparing enrollment rates for Medicaid and CHIP, SNAP, and WIC across families with and without undocumented parents while controlling for families with “perceived eligibility” for each resource. Because eligibility for public benefit programs could not be determined for participants without compromising the anonymity of the survey, self-reported perceptions of eligibility were used as a proxy measurement. We used Fisher’s exact tests and χ² tests for descriptive data. Continuous variables were analyzed with analysis of variance, Student’s t test, or Wilcoxon–Mann–Whitney test.

In our second analysis, we assessed the association of parent documentation status as an exposure with 2 outcomes: discontinuation of public benefits and fear of deportation. We performed 2 multivariable logistic regression analyses with these outcomes. Controlling for collinearity, survey variables with P values of <.2 from univariate analyses for each of these outcomes were considered for our multivariable regression models. Stepwise model selection method was used to conduct multivariable logistic regression analysis with calculated odds ratios and corresponding 95% confidence intervals (CIs). Variables with P values of <.05 were chosen for our final multivariable logistic regression model.

Of 1075 potential participants during the study period, 527 were approached and 399 enrolled (76% response rate). Of these participants, 105 reported having 1 or more undocumented parent, and 19 did not disclose immigration status (Fig 1). Data for participants who did not disclose immigration status can be found in Supplemental Tables 7–10.

Caregiver age, child age, child sex, caregiver type, and caregiver marital status did not differ between children with and without undocumented parents. A higher proportion of caregivers among families with undocumented parents identified as Latino or Hispanic anddid not speak English. Caregivers among families with undocumented parents had lower education and income levels compared with caregivers from families without undocumented parents (Table 1).

Forty-six percent of families with undocumented parents reported low or very low food security, compared with 21% of families without undocumented parents. A lower proportion of children with undocumented parents had a primary care physician (83% vs 97%). A higher proportion of children with undocumented parents were uninsured (15% vs 5%) and had delays in obtaining medications because of cost (23% vs 9%). There was no difference in perceived child health or perceived caregiver stress across groups (Table 2).

There was no difference between groups in emergency department visits or hospital admissions in the year before enrollment. Medicaid and CHIP (75% vs 55%) and WIC use (43% vs 21%) were significantly higher in families with undocumented parents, whereas 45% of families with undocumented parents reported SNAP use compared with 29% of families without undocumented parents (P = .05) (Table 2).

In a subgroup analysis of participants with perceived benefit eligibility, families with undocumented parents did not have significantly higher rates of Medicaid (85.9% vs 83.1%; P = .55), SNAP (49.4% vs 52.3%; P = .67), or WIC (64% vs 50%; P = .05) use. Of families with perceived eligibility for SNAP, families with undocumented parents reported higher levels of food insecurity compared with families without undocumented parents (47.1% vs 23.7%; P < .001). We found similarly high levels of food insecurity among families with perceived eligibility for WIC (52.9% vs 19.1%; P < .001).

Both groups reported lack of insurance as the top reason for missing well-child checks. Families with undocumented parents reported ineligibility because of citizenship status as the top reason for not being enrolled in Medicaid and CHIP, followed by insurance being cut off or expired. Families without undocumented parents reported ineligibility because of income level, followed by preference for another type of insurance.

Families with undocumented parents reported immigration concerns as the top reason for not being enrolled in SNAP, followed by enrollment and application process difficulties. Families without undocumented parents reported ineligibility because of income level, followed by not needing the resource.

Ineligibility because of age was the top reason why families with and without undocumented parents were not enrolled in WIC. Families with undocumented parents reported immigration-related concerns as the second-highest reason for nonenrollment in WIC, whereas families without undocumented parents reported ineligibility due to income level (Table 3).

Twenty-nine percent of families with undocumented parents reported discontinuation of public benefits because of immigration concerns, >70% of families with undocumented parents worried about whether a family member, friend, or themselves would be deported or separated from family, and 40% reported that their child had similar concerns. Twenty-four percent of families with undocumented parents worried that their child’s hospitalization would affect their immigration status (Fig 2).

Univariate analysis displaying factors associated with discontinuation of public benefits and fear of deportation can be found in Supplemental Tables 11 and 12, respectively. In multivariable logistic regression analysis, having undocumented parents or having a family member or friend who was deported were associated with both discontinuation of public benefits (Table 4) and fear of deportation (Table 5). Food insecurity was associated with fear of deportation but not discontinuation of public benefits.

We identified significant social and health care needs, disparities in public benefit use, and immigration-related fear among families with undocumented parents. Although approximately one-quarter of our families reported at least 1 undocumented parent, the vast majority of children with undocumented parents were US citizens (87%). Despite the high prevalence of families with mixed immigration status, most existing literature on families with undocumented parents is based on estimates given the sensitivity and privacy concerns of obtaining documentation status.³⁶  Our study is therefore unique in that we anonymously quantified health care and benefit use among undocumented immigrants on behalf of their children.

Social and health care challenges among families with undocumented parents included food insecurity, lack of a primary care provider (PCP), missed well-child checks, and uninsured payer status. These findings are consistent with existing literature describing health disparities in the undocumented immigrant population.^9,10  Inconsistent employment, responsibilities of sending remittances to family members in their country of origin, language barriers, lack of education, and immigration-related fear have been described as contributors to food insecurity and health care barriers in the undocumented population.^30,42 

In assessing public benefit use, we hypothesized that families with undocumented parents would have decreased resource use compared with families without undocumented parents. We instead identified equivalent or higher levels of public benefit use among families with undocumented parents. This is similar to previous observations describing higher odds of WIC and Medicaid use and similar odds of SNAP use in households with undocumented immigrants compared with households without undocumented immigrants.⁴³ 

Despite these findings, it is clear that benefit use among families with undocumented parents is much lower than their need. Excluding families who reported perceived ineligibility, children with undocumented parents had similar rates of SNAP use despite having nearly double the rate of food insecurity. These findings may be explained by bureaucratic obstacles (such as registration systems and paperwork) as well as immigration-related fear described in the undocumented population.⁴²  In addition, food insecurity may be exacerbated in families with undocumented parents because they are only eligible to apply for benefits on behalf of documented family members. Thus, although families with undocumented parents use resources for their children at a similar rate as those without undocumented parents, they do not receive the same level of benefits as families who are all eligible.

Additionally, our study unveiled a significant amount of fear among families with undocumented parents. The majority of these families worried about deportation or separation from family and several reported that their children had similar fears. Approximately one-third of families with undocumented parents reported disenrollment from resources because of immigration-related concerns. Having undocumented parents or a deported family member was associated with higher odds of disenrolling from resources. Furthermore, the top reason for not enrolling in SNAP and WIC among families with undocumented parents was immigration-related concern, which has previously been described.⁴²  However, caregivers did not express similar concerns about enrolling in Medicaid, and the majority of families with undocumented parents perceived to be eligible for Medicaid were enrolled. Although data suggest an overall decline in resource use, including Medicaid, among the immigrant population, immigration-related fear may be a bigger barrier to SNAP and WIC use compared with Medicaid use.^{26,27,44–47}  One possible explanation for higher Medicaid enrollment in our undocumented group is that our study was conducted in the inpatient setting. It is possible that our study participants had a higher insured rate than the general population and were more likely to seek care. Additionally, our hospital has mechanisms in place to help families with insurance applications. Although most patients who applied for Medicaid had applications that were still pending at the time of discharge, we did not control for length of stay, and it is possible that some of our patients gained insurance during their hospitalization. Further research is needed to assess why an enrollment discrepancy exists across Medicaid, SNAP, and WIC.

Our study underscores the importance of social needs screening and referral services by pediatric health care providers. It also highlights the impact of fear on seeking public assistance among families with undocumented parents. Providers should take this into consideration and follow American Academy of Pediatrics guidance on counseling families about immigration-related concerns.⁴⁸  Given the fear associated with using public benefits, disparities in food security among families with undocumented parents may be addressed by connecting families with trusted community organizations and local food banks, which are an underused resource potentially associated with less fear. Finally, the inpatient setting may offer an opportunity to identify and address sensitive stressors unique to families with undocumented parents. However, this must be done while maintaining a safe and protected environment, because a quarter of families with undocumented parents expressed fear that their child’s hospitalization would affect their immigration status.

Our study has several limitations. By limiting inclusion to English- and Spanish-speaking participants, we excluded important perspectives from other undocumented immigrant groups. However, >98% of patients admitted during our study period were either English or Spanish-speaking; thus, we included the vast majority of families on the basis of language. Additionally, 24% of families declined participation, and 5% of enrolled participants did not disclose immigration status, which may have introduced selection bias. Furthermore, our study was performed at a single institution in the inpatient setting, limiting generalizability. However, the inpatient setting may offer the opportunity to interact with several CIF, because children of color are at increased risk of admission to the hospital owing to social and medical disparities. We also captured the perspective of many families whose children did not have established PCPs.⁴⁹  Additionally, our study is limited by self-reported data; it is possible that some participants misclassified themselves. By completing the analysis on families who reported perceived eligibility for public benefits, we likely missed families who did not know they were eligible and included some families who were ineligible. Finally, our multivariable analyses of variables associated with discontinuation of public benefits and fear of deportation have wide CIs, which likely reflect a small sample size. Larger studies are needed to assess the replicability of our results.

Public benefit use among children with undocumented parents does not match their high levels of need. Immigration-related fear is prevalent among families with undocumented parents, is a barrier to benefit use, and is a common reason for public benefit disenrollment. We highlight the importance of provider education on immigration policies that affect the health of children from families of mixed immigration status, assessing barriers to public benefit enrollment, and appropriately counseling or referring eligible families to community resources. Further research is needed to confirm our findings and to explore effective interventions that mitigate disparities in public benefit use and address fear in this population.

We thank our patient families for taking the time to participate in our study. We also thank the Texas Children’s Hospital social workers, the Houston Immigration Legal Services Collaborative, the Tahirih Justice Center, and the Houston Food Bank for supporting our participants and their families. Finally, we thank Dr Ricardo Quiñonez and the section of Pediatric Hospital Medicine for the opportunity to gain perspectives from their patient families.

CHIP

Children’s Health Insurance Program

CI

confidence interval

CIF

children in immigrant families

PCP

primary care provider

SNAP

Supplemental Nutrition Assistance Program

WIC

Special Supplemental Nutrition Program for Women, Infants, and Children