Well-being of Parents and Children During the COVID-19 Pandemic: A National Survey

We fielded the Vanderbilt Child Health COVID-19 Poll from June 5 to 10, 2020, using the Ipsos KnowledgePanel, a large online research panel created by using probability-based address sampling of US households.^8–13  Households without Internet at the time of recruitment are provided with an Internet-enabled tablet. Participants in KnowledgePanel receive nominal periodic incentives to participate. This study of unidentified persons was considered exempt from human subjects review by the Vanderbilt University Medical Center Institutional Review Board.

For this survey, we included parents in KnowledgePanel with at least 1 child in the household <18 years old. Eligible participants were randomly selected from the standing panel, sent an e-mail notification, and sent a subsequent reminder 3 days later. This survey had a 50% completion rate,¹⁴  with a total of 1011 responses. Survey weights were designed to provide national estimates of parents with children <18 years of age, accounting for differential nonresponse. Benchmarks for survey weighting were obtained from the 2019 March Supplement of the Current Population Survey¹⁵  for all variables, except for language proficiency, which was obtained from the 2018 American Community Survey.¹⁶  Survey weights were constructed by first ranking geodemographic distributions of the ≥18-years parent population with children ages 0 to 17 years. Once all survey data were collected, design weights were adjusted to account for differential nonresponse. The following demographic data were collected and used in survey weights: respondent self-identified sex (male and female), respondent age (18–34, 35–44, 45+), race and ethnicity (white and non-Hispanic, Black and non-Hispanic, other and non-Hispanic, Hispanic, 2+ races and non-Hispanic), census region (Northeast, Midwest, South, and West), metropolitan status (metro and nonmetro), education (less than high school, high school, some college, bachelor or higher), annual household income (<$25 000, $25 000–$49 999, $50 000–$74 999, $75 000–$99 999, $100 000–$149 999, ≥$150 000), and language (Supplemental Information).

We developed the survey instrument to capture changes in physical, mental (parent), and behavioral (child) health; health insurance status; food security; use of public food assistance resources; child care; and use of health care services. We adapted questions from the National Survey of Children’s Health on food security, enrollment in food assistance programs (eg, the Special Supplemental Nutrition Program for Women, Infants, and Children [WIC] and the Supplemental Nutrition Assistance Program [SNAP]), and health insurance.¹⁷  We constructed additional questions to capture changes in the physical health of parents and children, parents’ mental health, children’s behavioral health, children’s insurance status, and child care. For all questions about changes, we asked respondents to use March 2020 as the reference point for the period before or at the beginning of the pandemic. We also asked questions about delays in medical care among children (Supplemental Information). We categorized food insecurity as mild (“we could always afford enough to eat but not always the kinds of food we should eat”), moderate (“sometimes we could not afford enough to eat”), and severe (“often we could not afford enough to eat”). The survey was available in English and Spanish.

We conducted all analyses using survey weights to provide national estimates. Descriptive statistics were calculated to summarize response frequency. Respondents who refused to answer a question were considered missing and not used in calculating proportions. All questions had <0.5% refusals. We report all summary statistics as the weighted proportion estimate with its 95% confidence interval (CI). We conducted significance testing for unpaired questions using Rao-Scott corrected χ² tests. For paired testing of questions that asked for pre– and post–COVID-19 comparisons, we used global P values from McNemar and Exact Multinomial tests for symmetry. The significance level was set at α = .05, and all tests were 2-sided. All analyses were conducted by using R version 3.6.2 (R Core Team, Vienna, Austria).

Compared to March 2020, 26.9% (95% CI: 23.9% to 29.9%) of parents reported worsening of mental health, and 14.3% (95% CI: 12.0% to 16.7%) reported worsening in their children’s behavioral health (Fig 1). For physical health, 17.7% (15.0% to 20.3%) of parents reported a worsening of their own, and 3.8% (2.5% to 5.1%) reported a worsening of their children’s.

The reported declines in mental health for parents, behavioral health for children, and physical health were similar across respondents from most racial and ethnic, income, and education groups and US Census Regions, but female and unmarried parents reported higher rates of worsening of their own mental health. Furthermore, a higher proportion of families with younger children reported worsening mental and behavioral health when compared with families with older children (Table 1). Approximately 1 in 10 (9.6%, 7.6%–11.7%) parents reported worsening of both their mental health and their children’s behavioral health (Supplemental Tables 4 and 5).

Parents reported statistically significant differences in the source of their children’s health insurance compared to March 2020, with a decrease in employer-sponsored insurance, small increases in other sources of private insurance and public insurance, and no substantial change in the proportion who were uninsured (Table 2; P < .001). Slightly more than one-third (39.9%, 36.6%–43.2%) of families reported cancellations or delays in their children’s health care since March 2020. The most commonly delayed visit types were well-child visits (49.4%, 44.0%–54.7%), visits with subspecialists (13.0%, 9.5%–16.5%), and behavioral health visits (9.4%, 6.3%–12.5%) (Supplemental Table 6).

Parents reported more food insecurity at the time of the survey compared to March 2020, with the proportion reporting any food insecurity increasing from 32.6% (29.4%–35.8%) to 36.0% (32.7%–39.3%) (Table 2 and Supplemental Table 9). Parents did not report significant changes in enrollment in SNAP or WIC. There was a statistically significant difference in parents reporting the use of food banks or pantries before and after March 2020, but the absolute increase was small (0.3 percentage points). An estimated 17.5% (14.8%–20.1%) of parents reported that their children received free or reduced-price lunch at schools before the pandemic, making them eligible for free school-related food programs and the Pandemic Electronic Benefit Transfer (P-EBT) program during COVID-19. At the time of the survey, 15.4% (12.9%–17.8%) of all parents reported receiving free food from schools for their children, and 5.0% (3.5%–6.5%) reported enrollment in the P-EBT program.

Disruptions in child care were common, with nearly one-quarter (24.1%, 21.1%–27.1%) of parents reporting loss of regular child care. Among parents who lost child care, the majority (74.1%, 67.6%–80.7%) reported that their child was watched by a parent (Supplemental Table 7). Disruption of regular child care varied substantially by the age of the children in the home. For homes with children 0 to 5 years old, 38.6% (33.4%–43.7%) experienced disruptions, compared with 7.5% (4.9%–10.1%) of families with adolescents 13 to 17 years old (Supplemental Table 8).

Among the 10% of families in which parents reported worsening of both their own mental health and their children’s behavioral health, 47.6% (36.3%–58.9%) lost regular child care during the pandemic, and 11.1% (3.7%–18.5%) also reported less food security (Table 3).

Parents and children have been substantially affected by the COVID-19 pandemic. More than 1 in 4 parents reported worsening mental health, and 1 in 7 parents reported worsening behavioral health for their children since the pandemic began. Worsening of parental mental health and children’s behavioral health were at times intertwined, with nearly 1 in 10 families reporting worsening of both. Loss of child care, delays in health care visits, and worsened food security were common among families experiencing worse mental and behavioral health.

Disruption in routines can be detrimental for children, especially those already with behavioral health diagnoses.¹⁸  For some children, this is complicated by challenges accessing traditional office-based services and the loss of mental health services that students may receive at school.¹⁹  In a recent poll, it was found that parents were worried about how school closure was affecting their children’s mental and emotional health,²⁰  and similar disruptions are evident in our study. The American Academy of Pediatrics, the American Academy of Child and Adolescent Psychiatry,²¹  and the American Psychiatric Association²¹  recently released guidance on school reopening that addresses physical and mental well-being of children. These recommendations suggest that schools prepare for a range of mental health conditions among children, provide enhanced training for teachers, engage mental health professionals in COVID-19–related messaging to children, address the mental health needs of staff, and focus additional resources for children with special needs. The issues around school openings across the United States are complex, and plans will vary by community. In some communities, school will be remote, and schools should consider working with pediatricians and mental health professionals on how they may address mental health of children, parents and staff even when school is remote. To implement these strategies effectively, Congress could consider enhanced funding to schools to address schools’ budgetary challenges related to implementing these recommendations. To address the mental health needs of children, providing funding to support the availability of telebehavioral health services through schools or other sources in the community and parity of coverage and reimbursement between video and audio-only modalities for children who lack reliable Internet access may be important.

We also observed improvements in behavioral, mental, and physical health for a subset of the population and no changes in these domains for the majority of the population, highlighting the heterogeneity in the effects of the pandemic and its consequences on families. Notably, we find that rates of worsening parental mental health and child behavioral health are similar among many socio-demographic groups (eg, race, income). However, our results suggest that some populations have been disproportionately affected, including single-parent families and those with younger children. Supports for families who lost regular caregiver support for young children may need to be tailored.

Our findings revealed small but potentially important changes to children’s insurance status, with an ∼3 percentage point reduction in employer-sponsored insurance, representing ∼2.5 million children. This decrease within just 1 calendar quarter could be due to the large increase in unemployment⁴  that occurred during the economic recession that followed social distancing measures. We did not find changes in uninsurance status, suggesting that small increases in Medicaid and private plans on health insurance exchanges may offset some short-term losses of employer-sponsored insurance. However, changes in insurance status and other social influences on health for children should continue to be monitored as the pandemic and the resulting economic effects continue. Congress could consider additional supports to state Medicaid programs through enhancing the federal match,^22,23  employer-sponsored insurance by providing subsidies to families to maintain converge (ie, Consolidated Omnibus Budget Reconciliation Act subsidies), and health insurance exchanges through expanded open enrollment periods.

Nearly 1 in 10 families reported moderate or severe food insecurity since March 2020. Despite the increase in food insecurity, there was not an increase in parents reporting receipt of SNAP or WIC. We found only a small increase in food bank use; however, it is possible that the use of food banks may be underestimated because families may have received food support from organizations they did not identify as food banks (eg, schools). The Families First Coronavirus Response Act²⁴  and the Coronavirus Aid, Relief, and Economic Securities Act²⁵  were signed into law in March 2020 to provide relief from the economic impact of COVID-19.²⁶  These laws included provisions authorizing emergency funding for food assistance programs, including additional support for existing SNAP recipients and the establishment of the P-EBT program for families with children who receive free or reduced-price meals at school through the National School Lunch Program. The National School Lunch Program serves 30 million children,²⁷  making schools an essential source of nutrition for many children, many of whom had been out of school for months at the time of our study. In this study, the uptake of P-EBT was limited, although many families appeared eligible on the basis of participation in other food assistance programs. Early reports on P-EBT suggest uneven adoption among states even as need for the program surged.²⁸  Our findings suggest additional efforts may be needed to mitigate observed reductions in family food security, including connecting families eligible for existing programs with support through outreach and reduction of administrative burdens.

The results of our study should be interpreted in the context of certain limitations. First, our data were collected at one time point rather than longitudinally. However, the recall period was relatively short (3 months), and our results demonstrating worsening behavioral and mental health for children and parents are qualitatively similar to a panel study early in the pandemic from a large city.¹⁵  Second, our survey respondents had higher levels of socioeconomic status compared with nonrespondents. The potential for nonresponse bias should be mitigated by the relatively rich data on nonrespondents and our approach to survey weights. Finally, our data rely on self-reports rather than clinical health assessments or administrative data on enrollment in programs. Our findings about adult mental health, however, are consistent with a recent poll using the validated Kessler 6 Psychological Distress Scale with an adult sample in April 2020.² 

The COVID-19 pandemic is having a substantial tandem impact on parents and children in the United States. As policy makers consider additional measures to mitigate the health and economic effects of the pandemic, they should consider the unique needs of families with children, including support for mental and behavioral health and efforts to improve food security.

We thank William Dupont, PhD, and Elizabeth McNeer, MS, for their statistical input on this article.

CI

confidence interval

COVID-19

coronavirus disease 2019

P-EBT

Pandemic Electronic Benefit Transfer

SNAP

Supplemental Nutrition Assistance Program

WIC

Special Supplemental Nutrition Program for Women, Infants, and Children