Consistency and Adequacy of Public and Commercial Health Insurance for US Children, 2016 to 2021

Key Points Question How do the rates and child and family characteristics associated with inadequate and inconsistent health insurance coverage compare for publicly vs commercially insured children in the US? Findings This cross-sectional study of 203 691 children found that publicly insured children experienced higher rates of inconsistent coverage, whereas commercially insured children faced higher rates of inadequate coverage. Public insurance consistency and commercial insurance adequacy improved substantially during the COVID-19 public health emergency. Meaning The findings of this cross-sectional study suggest that policies are needed to address the unique issues faced by each population of insured children to improve the consistency and quality of children’s health coverage in the postpandemic context.


Introduction
Consistent and adequate health insurance is critical to ensure children have access to affordable and high-quality health care.2][3] Among insured children, having adequate insurance-coverage that offers affordable access to needed services and health care professionals-is associated with lower unmet health care needs and higher quality care. 4,5[8] During the past 2 decades, the uninsured rate among children has steadily declined in the US.In 2021, 61.9% of US children were commercially insured, 36.4% were publicly insured, and only 5% were uninsured. 9This progress has been driven by policy reforms including expansions of Medicaid and the Children's Health Insurance Program (CHIP); commercial insurance regulations on cost sharing and coverage for preventive services; the establishment of the Affordable Care Act (ACA) Health Insurance Marketplace and subsidies as well as other efforts to enhance outreach and streamline enrollment. 10However, alongside declines in uninsurance, the proportion of US children who had inconsistent or inadequate insurance increased from 30.6% in 2016 to 34.0% in 2019. 4ior research has focused on documenting rates and trends in insurance consistency and adequacy at a national level for children covered by all insurance types 4,5,11 ; however, the insurancerelated challenges faced by publicly and commercially insured children are likely to differ due to differences in eligibility criteria, application processes, health care networks, cost-sharing requirements, as well as the accessibility, affordability, and quality of available care.The

Methods
This study was deemed exempt by the institutional review board of the University of Michigan, and informed consent was waived because only deidentified data were used.We followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

JAMA Health Forum | Original Investigation
Consistency and Adequacy of Public and Commercial Health Insurance for US Children

Study Sample
We conducted a secondary analysis of the 2016 to 2021 National Survey of Children's Health (NSCH), a nationally representative survey of US children from birth to 17 years old living in noninstitutional settings.Our analysis used the NSCH topical survey, which includes more than 100 survey items focused on 1 child in the sampled household.Nearly all NSCH respondents (93.8%) were biological, foster, and adoptive parents or stepparents of the selected child; 5.8% were other relatives (eg, grandparents) and 0.3% were nonrelatives.Detailed information on the NSCH sampling and data collection procedures are available from the US Census Bureau. 12Exposures The primary exposure was child insurance type at the time of the NSCH survey.Public insurance included coverage from any form of government assistance including Medicaid and CHIP.
Commercial insurance included coverage through a family member's current or former employer or union, the ACA Marketplace, direct purchase from an insurance company, and TRICARE (the health care program of the US Department of Defense Military Health System) or other military health care.
We excluded children who were covered by both public and commercial insurance (3.8% of the total sample) and those uninsured at the time of the survey (4.6%), including those insured only through the US Indian Health Service or through a religious health share.

Outcomes
The 2 primary outcomes were (1) inconsistent insurance, defined as having an insurance gap in the past 12 months and (2) inadequate insurance, defined as coverage failing to meet the following criteria: (i) benefits were usually or always sufficient to meet child's needs; (ii) coverage usually or always allowed child to see needed health care practitioners; and either (iii) no annual out-of-pocket (OOP) payments for child's health care or (iv) OOP costs were usually or always reasonable.
Secondary outcomes were the 4 individual criteria that comprised insurance adequacy.These outcome definitions are the National Performance Measures for the US Department of Health and Human Services Title V Maternal and Child Health Services Block Grant program and have been applied in prior research on health insurance using the NSCH. 4,11,13,14variates Sociodemographic characteristics of the child were reported by the adult NSCH respondent and included age, sex, race and/or ethnicity (non-Hispanic Black, Hispanic, non-Hispanic White, and other [including American Indian, Alaska Native, Asian Indian, Chinese, Filipino, Guamanian, Japanese, Korean, Native Hawaiian, Samoan, Vietnamese, other Pacific Islander, other Asian, some other race, and multiracial]), family structure (2 married parents; 2 unmarried parents, single parent, other), family income (as a percentage of the federal poverty level [FPL]), having at least 1 primary caregiver born in the US, and primary household language (English, Spanish, other).Clinical characteristics included whether the child had none, 1, or 2 or more of a list of 24 chronic physical or mental health conditions or disabilities, or any special health care needs (CSHCN). 12The NSCH CSHCN screener identifies children who require more than average use of health care services, counseling, or medications, or who experience a functional limitation due to a condition for a duration of 12 months or longer. 12

Statistical Analysis
We calculated rates of inconsistent and inadequate insurance overall and by sociodemographic and clinical characteristics stratified by insurance type.Because state Medicaid policies could drive heterogeneity in the outcomes, we also calculated rates of the outcomes for publicly insured children by state.To assess changes in the outcomes over time, we calculated rates by year and for the pooled prepandemic (2016-2019) and COVID-19 PHE (2020-2021) periods.We conducted unadjusted and adjusted logistic regressions to (1)

Results
The

Changes During the COVID-19 PHE
The Figure shows inconsistent and inadequate coverage rates by year and insurance type.Among publicly insured children, inconsistent insurance decreased from 4.8% before the PHE to 2.9% during the PHE (adjusted difference, −2.0 pp; 95% CI, −2.8 to −1.2 pp; 42% decline from baseline).
Among commercially insured children, inadequate insurance decreased from 33.6% to 31.7% during the PHE (adjusted difference, −2.0 pp; 95% CI, −2.9 to −1.0 pp; 5.9% decline from baseline), primarily due to improvements in reasonable OOP costs (eTable 2 in Supplement 1).No PHE-related changes were identified for public insurance adequacy or commercial insurance consistency.

Child Characteristics Associated With Inconsistent and Inadequate Coverage
In adjusted models among publicly insured children, inconsistent coverage was significantly higher among Hispanic children and those with household incomes from 200% to 399% FPL (Table 3).
Although differences by age category were not statistically significant, when comparing inconsistent insurance rates by year of age, we found that publicly insured children had markedly higher rates (7.3%) of inconsistent coverage at 1 year of age (eFigure 1 in Supplement 1).Inadequate coverage increased with child age and was significantly higher among publicly insured children with household incomes of 200% to 299% FPL and those with multiple chronic conditions and disabilities.Inconsistent coverage among publicly insured children varied from 0.6% in Vermont to 7.7% in Georgia, while inadequate coverage varied from 7.6% in Maine to 19.7% in Illinois (eTable 4, eFigures 2 and 3 in Supplement 1).
In adjusted models among commercially insured children, inconsistent coverage was significantly higher for non-Hispanic Black children, children of other race or ethnicity, as well as children in households with unmarried parents, single parents, lower incomes, and those with multiple chronic conditions and disabilities.In contrast to publicly insured children, differences in inadequate coverage were identified for most commercially insured child characteristics including age, sex, race and/or ethnicity, family structure, income, and health needs.Inadequate coverage was particularly high for commercially insured children younger than 1 year (adjusted predicted probability, 40.0%), children with household income less than 100% FPL (38.0%) and those with 2 or more chronic conditions and (42.2%).eTable 3 in Supplement 1 provides the unadjusted differences by child characteristics and insurance type.

Discussion
Using nationally representative data, we found that inconsistent coverage is 3 times higher among publicly insured compared with commercially insured children.However, inadequate insurance is more prevalent overall, affecting nearly 1 in 5 children (16.5 million annually) in the US, with particularly high rates among the commercially insured.We also identified substantial improvements in public insurance consistency and commercial insurance adequacy during the COVID-19 PHE.Furthermore, we found that the child and family characteristics associated with higher rates of inconsistent and inadequate coverage differed by insurance type.
Consistent with prior research showing that insurance gaps are a particular issue for publicly insured children, 3,15,16 we found that 4.2% of publicly insured children had a gap in the past year compared with only 1.4% of commercially insured children.Although some gaps are due to household income changes, a substantial share are for procedural reasons and nearly half of children who lose Medicaid-CHIP re-enroll within 12 months. 15As evidence of this, we found a notable spike in insurance gaps among publicly insured children at 1 year of age (7.3%), which reflects the first point of eligibility determination for most publicly insured children-given that being born to a mother with Medicaid automatically covers the child until their first birthday.
Compared with 2016 to 2019, we found that inconsistent public insurance declined by 42% during the PHE when continuous Medicaid eligibility requirements were in place.The unwinding of these protections in 2023 is projected to leave 5.3 million children without Medicaid-CHIP coverage, potentially resulting in delays and forgone care. 17Among disenrolled children, 74% are projected to be disenrolled despite being eligible for Medicaid. 17The remaining 26% of Medicaid ineligible children will need to enroll in commercial coverage, which our findings indicate offers less adequate coverage, particularly for low-and middle-income families.
States have promising policy options to bridge insurance gaps for publicly insured children in the post-PHE context.Since 1997, the US Centers for Medicare & Medicaid Services (CMS) has allowed states to grant 12-month continuous Medicaid-CHIP eligibility for children, which has been associated with reduced insurance gaps. 18However, as of January 2023, only 23 states have implemented this policy. 19In 2022, CMS approved Oregon's 1115 waiver that allows children to continuously maintain Medicaid-CHIP from birth until age 6 years, with 2-year continuous eligibility from age 6 to 17 years.Although multiyear continuous eligibility rules are the most durable approaches to improve coverage consistency, states can also address procedural disenrollment through automatic renewal, increased funding for consumer assistance, and working with managed care plans to maintain updated beneficiary contact information. 20Recent proposed rulemaking from CMS would make it easier for states to implement these streamlined enrollment and renewal intitiatives. 21r findings also point to a particular need for state Medicaid programs to conduct targeted outreach and linguistically and culturally competent navigation assistance for immigrant families.We found that 1 in 20 publicly insured Hispanic children had inconsistent coverage in the past year, with similar rates among children without a US-born caregiver (4.6%).These findings could partly reflect reticence among immigrant parents to enroll eligible children in Medicaid-CHIP for fear of immigration-related consequences.Although the Biden Administration reversed the Trump characteristics of children and families at greater risk of inconsistent or inadequate insurance may also vary by insurance type.Furthermore, policy responses during the COVID-19 public health emergency (PHE), such as continuous eligibility requirements for Medicaid and bolstered ACA Marketplace subsidies may have substantially affected children's insurance.Although a national study found no change in US children's insurance consistency and adequacy from 2019 to 2020 (the first year of the COVID-19 PHE), 11 aggregate estimates may mask changes between publicly insured compared with commercially insured children, who were differentially affected by COVID-19 PHE policies.Using nationally representative survey data from 2016 to 2021, the objective of this study was to evaluate the consistency and adequacy of health insurance for children insured by public compared with commercial insurance.We also compared changes during the COVID-19 PHE and identified sociodemographic and clinical characteristics associated with inconsistent and inadequate coverage within each insurance type.

Figure .
Figure.Inconsistent and Inadequate Health Insurance Coverage for US Children, by Year and Insurance Type, 2016 to 2021

JAMA Health Forum | Original Investigation Consistency
compare outcome differences by insurance type,(2)estimate and Adequacy of Public and Commercial Health Insurance for US Children the PHE stratified by insurance type, and (3) identify child characteristics associated with the outcomes stratified by insurance type.For outcome differences by insurance type, we were primarily interested in unadjusted differences, which reflect both the unique populations served and the features of public compared with commercial insurance.All analyses applied survey weights to produce nationally representative estimates that account for the NSCH sampling structure and nonresponse.The NSCH public-use files include imputed data for several variables.12Childsexand race and/or ethnicity had low levels of missingness (<1%) and were imputed by the Census Bureau using hot-deck imputation.Household income had a high level of missingness (18%) and was inputted by the Census Bureau using sequential regression imputation.We used the Stata multiple imputation command to appropriately estimate means and variance based on the 6 imputed income values in the public-use data set.For all other variables with missing values not imputed by the Census Bureau, we included missing as a category in the analysis.Statistical tests were 2-tailed and P < .05 was considered statistically significant.Data analyses were performed from March to August 2023.

Table 1 .
Sample Characteristics by Child's Current Health Insurance Type, 2016 to 2021 a Other race or ethnicity included Alaska Native, American Indian, Asian Indian, Chinese, Filipino, Guamanian, Japanese, Korean, Native Hawaiian, Samoan, Vietnamese, other Pacific Islander, other Asian, any other race, and multiracial.Unweighted sample sizes and survey-weighted prevalence estimates.Variables without a missing category had no missing values.

Table 2 .
Prevalence of Health Insurance Consistency and Adequacy Among US Children, by Insurance Type, 2016 to 2021

Table 3 .
Adjusted Predicted Probabilities of Inconsistent and Inadequate Health Coverage for Publicly Compared With Commercially Insured Children, 2016 to 2021 Insurance Consistency and Adequacy by Year and Pre (2016-19) vs. Post-Pandemic (2020-21) Differences for Publicly Insured Children eTable 2. Insurance Consistency and Adequacy by Year and Pre (2016-19) vs. Post-Pandemic (2020-21) Differences for Commercially Insured Children eFigure 1. Inconsistent and Inadequate Insurance by Child Age and Insurance Type, US 2016-2021 eTable 3. Unadjusted Predicted Probabilities of Inconsistent and Inadequate Coverage for Publicly and Commercially Insured Children, US 2016-2021.eTable 4. Inconsistent and Inadequate Insurance for Publicly Insured Children by State, 2016-2021 eFigure 2. Inconsistent Insurance for Publicly Insured Children by State, 2016-2021 eFigure 3. Inadequate Insurance for Publicly Insured Children by State, 2016-2021