While we won’t be able to do our in-depth analysis of children’s health coverage trends this fall due to problems collecting American Community Survey data during the pandemic, the Census Bureau did release a report on national trends in health insurance coverage in 2020 using the Current Population Survey Annual Social and Economic Supplement.

Overall, the report found that 5.6 percent of kids were uninsured throughout 2020 – a total of nearly 4.3 million children that went without coverage. In 2017, as readers of SayAhhh! know, the number of uninsured children started going in the wrong direction – using this data set 5 percent of kids were uninsured in 2017. Figure 1 shows that change over time using the Current Population Survey.

National Child Uninsured Rates, by Age Group

Many may be surprised that the Census Bureau report shows no statistically significant increase in the number of uninsured kids from 2018 to 2020. Given the economic effects of the recession associated with the pandemic, many were expecting a large increase in the number of uninsured families. Indeed the Bureau reported that overall median family income went down for the first time since 2011, and the number of full-time, year-round workers declined from 2019 to 2020. Additionally, the uninsured rate for non-elderly adults working less than full-time did increase significantly, from 14.6 percent in 2018 to 16.4 percent in 2020.

But just because the topline number for kids hasn’t changed that much, one should not assume that things weren’t happening below the surface. A few reflections on looking more closely at the Census Bureau’s findings:

• More kids may have been uninsured throughout the year. The CPS only counts people as uninsured if they did not have any insurance coverage during the entire previous year. However, the survey recently started reporting estimates of how many people were only covered for part of the year. In 2020, an additional 3.6 million children were uninsured for part of the year. So just shy of 8 million kids (or 10.4 percent) were uninsured for part or all of 2020.

• Overall, higher income kids did slightly better and the poorest kids did considerably worse. The uninsured rate for children under the poverty line rose significantly from 7.8 percent in 2018 to 9.3 percent in 2020. This is an increase of 19 percent. Meanwhile the highest income kids saw their health insurance coverage rates improve slightly – from an already high 97.4 percent covered to 97.8 percent covered. This no doubt reflects the disparate effects the pandemic had on upper income families and those families struggling to make ends meet. Unfortunately we can’t get a closer look to look at how kids just over the poverty line fared (i.e. from 100% of FPL to 200% of FPL). These kids traditionally have had the highest uninsured rate but that may have changed in 2020.

• We know that the Medicaid disenrollment freeze offered critical protection to kids and families avoiding any “red tape” losses as states were not allowed to disenroll people for small changes in income or paperwork snafus. The Census CPS data likely understates the role Medicaid played when compared to federal administrative data that shows considerable growth in enrollment.

• Racially disparate effects of the pandemic are likely playing out with Black kids seeing the greatest losses. Coverage rates for Black kids have been a rare bright spot over the years when we disaggregate the data by race/ethnicity -- so this is a very troubling trend. However the data on race/ethnicity may be a little wonky given that it shows an increase in uninsured rates for Hispanic kids from 2018 to 2020 but not a significant one. We would be very surprised if coverage for Hispanic kids had not gotten worse during the pandemic. White kids saw no significant changes in their coverage, but Asian kids cut their uninsured rate by nearly one-third from 4.1 percent in 2018 to 2.8 percent in 2020. It is not clear why that would be.

With many families losing income due to the pandemic (which as the report points out was mitigated on the income side by unemployment and stimulus checks), and more kids falling into poverty, more kids may have become newly eligible for Medicaid but their families likely did not realize that and did not enroll them. Virtually no effort was made in 2020 by the Trump Administration to do outreach to families about the possibility of their child being eligible for Medicaid or CHIP. Now, the Biden Administration is gearing up outreach and enrollment efforts– more prominently associated with the recent special enrollment period for the Marketplace – so this may be changing.

The growing number of uninsured children below the poverty line underscores how critical it is to structure outreach efforts with children and lower income families in mind – the vast majority of whom will wind up in Medicaid/CHIP and not the Marketplace. Because families of color are disproportionately represented in the uninsured group, trusted messengers are essential. Medicaid/CHIP enrollment is open year-round, unlike the Marketplace, and the complex differences in child v. adult eligibility may not be well understood by many families. These data suggest that more work needs to be done to ensure that no child goes uninsured.

Note on the Current Population Survey: The Census Bureau’s report has historically included data from the American Community Survey for state-level and other demographic analyses. However, following the announcement last month that the standard one-year American Community Survey estimates would not be released because of data quality issues associated with the pandemic, this year’s report relied exclusively on data from the Current Population Survey Annual Social and Economic Supplement (CPS ASEC). 2019 CPS ASEC data were collected in March 2020 as the Bureau implemented pandemic-related changes to its data collection procedures, and the survey had a much lower response rate than usual. Although the response rate improved somewhat with the 2020 data, the Census Bureau notes that the people who responded to the survey in 2020 and 2021 (representing the 2019 and 2020 data) tended to have higher incomes and higher levels of educational attainment than those who did not respond. The Census Bureau used 2018-2020 comparisons in its report because the 2018 data were collected before the pandemic and the 2020 data were collected after many of the pandemic-related changes to data collection had been lifted.