Stewart, Andrea
Research Fellow

p 612-624-4850
e astewart@umn.edu

2022 NHIS Early Release: Estimates for First Half of the Year Show Continued Fall of Uninsured Rates

As part of the National Health Interview Survey (NHIS) Early Release Program, the National Center for Health Statistics (NCHS) recently published a second round of health insurance coverage estimates for the first half of 2022, covering the months of January to June.

Uninsurance was again measured at historic lows, with just 8.3% of all ages reporting being uninsured compared to 9.6% in the first half of 2021. This significant decrease represents 3.7 million fewer people without insurance through June 2022 (27.4 million as opposed to 31.1 million in 2021).

Other overarching comparisons between January-June estimates in 2022 and 2021 showed:

• The rate of individuals without health insurance coverage was significantly lower among those under 65, particularly for nonelderly adults (age 18-64), who saw a 1.9 percentage-point (PP) drop in their uninsured rate, from 14.0% in 2021 to 12.1% in 2022.
• Rates of public and private coverage remained statistically unchanged for all age groups.

Much like in our analysis of the NHIS Q1 estimates, no measure of statistical significance in either public or private coverage changes was able to account for the large drop in uninsurance rates. Therefore, we can only surmise that the potential shifts in coverage shown by the data—nearly unchanged among public coverage and slight indications of increasing private coverage—may in time point to confirmations of a post-pandemic recovery.

Coverage Changes for All Ages and Nonelderly Adults (18-64)

*Statistically significant at the 95% confidence level.

Important Findings by Demographic Groups

Analyzing coverage rates by more granular population levels revealed few significant changes between January to June of 2022 and the same time in 2021. However, we did note sizeable drops in uninsured rates for nonelderly adults with an income less than 100% of the federal poverty level (down from 19.5% in 2021 to 14.9% in 2022) and nonelderly adults living in states that have not yet expanded Medicaid (decreasing from 17.1% in 2021 to 14.5% in 2022).

While it is heartening to see persistently low overall rates of uninsurance, to anticipate that any perceived gains in private coverage^ have been potentially driven by employer-sponsored insurance and thus by rising post-pandemic employment levels, and improvements in coverage for those experiencing greater levels of poverty, it is also important to understand where there are still opportunities to improve coverage rates. For instance, Hispanic/Latino individuals were more than twice as likely to be uninsured compared with all nonelderly individuals under 65 (20.5% versus 9.9%) for the third year in a row.

Examining uninsured rates for individuals of other races/ethnicities showed that uninsurance decreased to 6.4% from 7.5% between 2022 and 2021 for nonelderly White individuals, but was stable among Black individuals (10.0%), Asian individuals (6.9%), and persons who identified as being of multiple races and other races (8.6%).

Much like overarching national trends, no racial or ethnic groups experienced any significant differences in rates of public coverage; however, White individuals saw gains in private coverage, rising from 72.7% in 2021 to 75.1% in 2022, while Hispanic individuals, Black individuals, Asian individuals, and individuals who reported being of multiple races and other races remained statistically unchanged between the first half of 2021 and 2022, at 44.7%, 49.7%, 72.4%, and 53.2%, respectively.

Impact of COVID-19 on NHIS 2021 Data

While federal survey data collection has recovered greatly from pandemic-related disruptions in 2020, it’s important to note that the presence of COVID-19 did impact collection operations in 2021 to a lesser degree. From January through April 2021, NHIS interviews continued to be attempted by telephone first, with personal visits used only to follow up on nonresponse, deliver recruitment materials, and conduct interviews when telephone numbers were unknown. Starting in May, field representatives were given the discretion to resume personal visits based on their assessment of risks and conditions in the area.

As such, differences between early estimate reports may contain greater variation than in years past; however, all data presented in this report continues to meet NCHS standards of reliability.¹

For further details on the impacts of the COVID pandemic on federal surveys in 2020 and 2021, see SHADAC briefs: “Comparing Federal Government Surveys That Count the Uninsured: 2022” and “Changes in Federal Surveys Due to and During COVID-19.”

SHADAC Staff

p 612.624.4802
e shadac@umn.edu

Key Resources and 2022 Updates: SHADAC's Health Insurance Unit

Updated April 2023

In the years following the passage of the Patient Protection and Affordable Care Act (ACA), SHADAC developed and introduced its Health Insurance Unit, or “HIU,” a tool to aid in defining family interrelationships in federal population surveys for the purpose of analyzing health insurance coverage.

Since the inception of this tool, SHADAC has regularly produced updated HIU variables, including in fall 2022 following the release of the 2021 American Community Survey (ACS) and 2022 Current Population Survey Annual Social and Economic Supplement (CPS ASEC). These variables were recently made available via IPUMS-CPS (October 19, 2022) and IPUMS-USA (March 10, 2023). These HIU variables use the updated HIU methodology introduced in 2020 and incorporate the latest years of survey data and federal poverty guidelines, as relevant.

What is the SHADAC HIU?
First released in 2012, one of the goals of the SHADAC HIU is to establish a common definition of a “family unit” to help ensure consistent and comparative research across major national surveys when analyzing health insurance coverage.

Most federal surveys define a “family” differently from the way it is defined by most private and public insurance programs. For instance, the definition of "family" or "household" in U.S. Census Bureau surveys includes all related members of a household, no matter the degree of relationship, and does not necessarily align with dependent coverage or public program eligibility.

The SHADAC HIU attempts to more closely align with a definition of family used for private and public health insurance coverage eligibility. The SHADAC HIU uses a narrower definition of family that looks at specific interrelationships between individuals within a household and excludes all non-dependent relatives (grandparents, adult siblings, aunts/uncles, etc.) who may be household members but are unlikely to be considered as part of the “family unit” as defined for the purposes of determining eligibility for health insurance.

The SHADAC HIU, then, is defined as an economic unit that consists of those members of a household who would likely be eligible as a group for family health insurance coverage, or whose resources (i.e., income) would be considered in determining eligibility for public coverage.

Related Resources

• Defining Family for Studies of Health Insurance Coverage (August 2021): A brief that outlines the impacts of using the SHADAC HIU in analysis—specifically, analysis showing how the population distribution of family income changes using three different definitions of family: all members in the same household (Census definition), the definition used by the IPUMS (described below), and the SHADAC HIU (described in detail in a companion brief). Researchers can use this brief to assess whether the SHADAC HIU is suitable for their analysis and what the potential impacts of its use might be.
• Stata Code (January 2021): Technical documentation of statistical code in STATA to help researchers to employ the SHADAC HIU in their own analysis using ACS microdata downloaded from IPUMS USA.

Comparing Federal Government Surveys That Count the Uninsured: 2022

October 2022:

With the release of new insurance coverage estimates from surveys conducted by the U.S. Census Bureau, the Agency for Healthcare Research and Quality (AHRQ), and the Centers for Disease Control and Prevention (CDC), SHADAC has updated our annual “Comparing Federal Government Surveys that Count the Uninsured” brief.

The brief provides an annual update to comparisons of uninsurance estimates from four federal surveys:

• The American Community Survey (ACS)
• The Current Population Survey (CPS)
• The Medical Expenditure Panel Survey - Household Component (MEPS-HC)
• The National Health Interview Survey (NHIS)

In this brief, SHADAC presents current and historical national estimates of uninsurance along with the most recent available state-level estimates from these surveys (where possible). We also discuss the main reasons for variation in the estimates across the different surveys as well as possible reasons for incomparability of estimates across and within the surveys.

Download a PDF of the Comparing Federal Government Surveys Brief.

Last year’s brief with data from the 2020 collection year, and certain 2019 collection-year data, can be accessed here.

Better Health through Data: Improved Collection of Demographic Data in Medicaid

October 2022:

SHADAC Deputy Director Elizabeth Lukanen served as an expert panelist for an Alliance for Health Policy briefing hosted by Arnold Ventures titled "Better Health through Data: Opportunities and Limitations of Health Data" which was held virtually on October 6, 2022. This event reviewed the importance and use of health data in addressing health disparities, the main barriers that persist in limitations to health data, and policy levers that can improve health data collection and utilization at the federal level including changes to data standards and systems, incentives and requirements, education and technical assistance, and regulation.

Elizabeth's presentation specifically focused on Medicaid Administrative Data and how SHADAC has been working to evaluate activities aimed at improving response rates and enhancing the quality of these data at the state level. She also highlights efforts by a few states – Oregon, New York, and Maryland – that have done a lot of work to fill gaps in their demographic data collection of Medicaid enrollees. 

Visit the Alliance for Health Policy event page for further details, resources provided by the panelists, and a recording of this event.

Related resources supported by State Health and Value Strategies (SHVS):

• Issue Brief - Collection of Race, Ethnicity, Language (REL) Data in Medicaid Applications: A 50-state Review of the Current Landscape 
• Issue Brief - Collection of Sexual Orientation and Gender Identity (SOGI) Data: Considerations for Medicaid and Spotlight on Oregon 
• Expert Perspective - Exploring Strategies to Fill Gaps in Medicaid Race, Ethnicity, and Language Data 
• Expert Perspective - New York State of Health Pilot Yields Increased Race and Ethnicity Question Response Rates 
   

SHADAC Staff

p 612.624.4802
e shadac@umn.edu

CPS ASEC: 2021 National Health Insurance Coverage Estimates Show Falling Uninsured and Private Coverage Rates Across Population Demographics (Infographic)

The U.S. Census Bureau has released an initial report detailing the state of health insurance coverage in the United States for 2021. Drawing on data from the Current Population Survey Annual Social and Economic Supplement (CPS ASEC), the report highlights good news overall as the percentage of the population without health insurance coverage at any point in 2021 fell to 8.3% from 8.6% in 2020—a decrease of 0.4 percentage points (PP) or approximately 1.1 million people (27.2 million in 2021 vs. 28.3 million in 2020).

While this year marks a return to a more standard release of health insurance coverage estimates after COVID-related factors impacted the release of 2020 from the CPS and, to a larger degree, the American Community Survey (ACS), the report did not include any state-level estimates from the ACS, as have been provided in past years. Instead, the Census Bureau plans to release all data relating to the ACS on Thursday, September 15, 2022.

The CPS ASEC does provide important breakdowns of available health insurance coverage estimates at the national level, including by age, race and ethnicity, immigration status, marital status, geographic region, employment, poverty status, and Medicaid expansion status.

The following blog and infographic explore changes in coverage rates by demographic for nonelderly adults (age 19-64) and for children (age 0-18), who experienced a greater number of changes in insurance coverage both overall and across categories than their adult counterparts.

Changes in Uninsurance by Age and Characteristics

Among nonelderly adults (age 19-64), the overall rate of uninsurance remained statistically unchanged from the previous year, reported at 11.6% in 2021 compared to 11.9% in 2020.

Comparatively, children (age 0-18) had an uninsured rate of 5.0% in 2021, a decrease from 5.6% the previous year.

Race and ethnicity
While uninsurance rates also remained unchanged across a number of racial and ethnic groups in 2021—7.5% for White adults, 7.7% for Asian adults, and 25.1% for Hispanic adults—the rate of uninsurance fell significantly among Black nonelderly adults, decreasing to 12.7% in 2021 from 14.2% in 2020.

Similarly, while uninsurance rates remained stable for children of several racial and ethnic groups in 2021—3.4% for White children and 8.6% for Hispanic children—the rate of uninsurance dropped significantly among Black children, falling to 4.3% in 2021 from 5.9% in 2020. Asian children, however, saw an increase in rates of uninsurance, rising to 4.6% in 2021 from 2.8% in 2020.

Immigration status
The rate of uninsurance among native-born nonelderly adults fell to 9.1% in 2021 from 9.6% in 2020, while the uninsurance rate among foreign-born adults remained statistically unchanged at 22.8% in 2021 compared to 22.9% in 2020.

Rates of uninsurance also fell among native-born children, decreasing to 4.5% in 2021 from 5.2% in 2020, while again, the uninsurance rate for foreign-born children was statistically unchanged at 18.6% in 2021.

Poverty level
When examining coverage by poverty level, rates of uninsurance among nonelderly adults were broadly unchanged for each measurable category: 24.0% among those with incomes below 100% poverty, 16.7% for those between 100%-399% poverty, and 4.5% for those living at or above 400% poverty.

In 2021, rates of uninsurance were unchanged for children in the two lower poverty level categories—8.3% for those in families with incomes below 100% poverty and 6.4% for those living in families with incomes between 100%-399% poverty. However, uninsured rates fell to 1.7% among children in families with incomes at or above 400% of poverty, a decrease of 0.4 PP from 2020.

Medicaid expansion status
As of January 1, 2021, 36 states and the District of Columbia have chosen to act on the option provided by the Affordable Care Act (ACA) and expand Medicaid eligibility requirements. Comparing rates of uninsurance by expansion and nonexpansion states revealed parallel trends for both nonelderly adults and children.

In 2021, the overall uninsured rate remained statistically unchanged from 2020 (9.0% in both years) for adults living in expansion states. However, uninsurance dropped from 17.7% in 2020 to 16.7% in 2021 for adults living in nonexpansion states.

Similarly, children in expansion states saw no change in uninsured rates for 2021 compared to 2020 (4.0% in both years). And again, uninsurance fell in nonexpansion states, falling to 7.1% in 2021 from 8.6% in 2020.

Other Notable Coverage Changes

Public and private coverage
Other notable overarching trends included a higher percentage of the population with public coverage, increasing 1.2 PP to 35.7% in 2021 from 34.5% in 2020, and a consequent decrease in the percentage of the population with private coverage, measuring at 66.0% in 2021, a decrease of 0.6 PP from 2020 (66.5%). The largest drivers of public and private coverage changes were a 0.9 PP increase in Medicaid coverage (18.9% in 2021, up from 17.9% in 2020) and a 0.2 PP drop in employment-based coverage (54.3% in 2021, down from 54.6% in 2020).

Poverty level
Of particular note in the report was a consistent correlation between poverty level for both adults and children and changes in rates of public and private insurance coverage between 2020 and 2021.

Rates of public coverage rose among adults in all poverty categories, increasing 2.4 PP to 51.8% among those with incomes below 100% of poverty, 0.8 PP to 25.4% for those between 100%-399% of poverty, and 0.5 PP to 5.7% among those living at or above 400% of poverty.

Rates of public coverage also increased by 2.0 PP to 44.2% among children in families with incomes between 100%-399% and by 1.2 PP to 7.2% for children in families at or above 400% of poverty. Rates of public coverage were unchanged among children below 100% of poverty (79.7%).

Looking at private coverage changes, uninsurance rates remained unchanged among adults in all poverty categories at 27.1%, 61.8%, and 91.8% for those with incomes below 100% of poverty, between 100%-399% of poverty, and at or above 400% of poverty, respectively.

Similarly, rates of private coverage were largely unchanged among children in families with incomes below 100% of poverty and those in families at or above 400% of poverty at 15.5% and 93.2%, respectively. Rates of private coverage did decrease in one category: children in families with incomes between 100%-399% of poverty, dropping by 1.5 PP to 53.4% in 2021.

Employment
Overall, rates of uninsurance fell amongst workers, decreasing to 10.7% in 2021 from 11.2% in 2020, driven by a decrease in adults who worked less than full-time year-round (15.1% in 2021 vs. 16.4% in 2020). Among those who worked full-time year-round, uninsurance did increase to 9.1% in 2021 from 8.4% in 2020, and among those who did not work in 2021, the rate of uninsurance remained unchanged at 14.8%.

Access the U.S. Census Bureau's "Health Insurance Coverage in the United States: 2021 - Current Population Reports".

As previously mentioned, the Census Bureau will be releasing the full roster of 2021 estimates from the American Community Survey (ACS) on Thursday, September 15. SHADAC will be monitoring this release and providing updates, along with several blog posts analyzing the estimates, as soon as they are available.

In a follow-up to several blogs and reports released in 2021 that detailed the effects of COVID-19 on the 2020 CPS and ACS health insurance coverage data estimates, a new blog from the Census Bureau looks at the continuing impact of the pandemic on survey data collection and release processes and nonresponse bias.

Though data collection operations were mostly able to resume as normal, lower-than-usual response rates persisted into 2022, measuring at 72% compared to 76% in 2021. Prior to the pandemic (2018 and earlier), the CPS enjoyed some of the highest response rates across a number of federal surveys, consistently recording percentages in the high 80s and low 90s.

Additionally, the blog notes the continued presence of measurable nonresponse bias, particularly in categories of income statistics and poverty rates. While the CPS standard weighting methodology (described fully in the ASEC technical documentation) was able to account for differences in survey estimates and administrative data prior to 2019, figures in the blog show that both standard and even alternative weights employed by the Census Bureau have not been able to fully adjust for differences between respondents and nonrespondents since 2020, highlighting the fact that certain pandemic-era challenges still persist.

Notes
All changes are statistically significant at the 90 percent confidence level.

All estimates come from the 2021 and 2022 Current Population Survey Annual Social and Economic Supplement, which provide data for years 2020 and 2021, respectively.

Both the 2020 and 2021 estimates use 2020 Census-based population controls, meaning that estimates for 2020 listed in the current report may not always match those listed in the report published in September 2021, which used 2010 Census-based population controls.

Uninsured estimates represent individuals with no health insurance coverage for the entire calendar year.

Estimates of health insurance coverage type represent individuals with that type of coverage at any point during the calendar year. Types of health insurance coverage are not mutually exclusive.