Census Bureau: 'Small Area Health Insurance Estimates – 2019'

WASHINGTON, June 25 -- The U.S. Census Bureau issued the following report (No. P30-09) by Sara Robinson and Katherine Ann Willyard entitled "Small Area Health Insurance Estimates: 2019":

* * *

INTRODUCTION

This report provides a summary of the 2019 release of the U.S. Census Bureau's Small Area Health Insurance Estimates (SAHIE)./1

SAHIE are the only source of data for single-year estimates of health insurance coverage status for all counties in the United States by selected economic and demographic characteristics (refer to the "Small Area Health Insurance Estimates (SAHIE)" text box)./2

The 1-year American Community Survey (ACS) provides detailed estimates of health insurance coverage for counties with populations of 65,000 or more./3

As a data enhancement to the ACS, the SAHIE model-based estimates are a vital source of information for measuring year-to-year change in health insurance coverage at the county level. The data in this report show changes in health insurance coverage between 2018 and 2019, as well as changes in health insurance coverage between 2013 and 2019. In addition, the report provides results on the differences in coverage among selected demographic groups.

* * *

1 The Census Bureau's Disclosure Review Board and Disclosure Avoidance officers reviewed this data product for unauthorized disclosure of confidential information and approved the disclosure avoidance practices applied to this release. CBDRB-FY21-077.

2 There are 3,142 counties in the United States. The SAHIE program does not include Kalawao County, Hawaii, due to insufficient data.

3 Approximately 73.6 percent, or 2,313 of U.S. counties, do not have detailed 1-year estimates of health insurance coverage. However, the ACS 1-year county-level estimates cover about 85.0 percent of the total U.S. population. The ACS also releases 1-year supplemental tables of health insurance coverage estimates for geographic areas with populations greater than 20,000, but these tables do not provide the same economic and demographic detail as SAHIE.

* * *

OVERVIEW OF SAHIE

Each year, the SAHIE program releases timely, reliable estimates of health insurance coverage for the population under the age of 65 by state and county./4

Federal agencies and programs use SAHIE data to determine eligibility for public health services. The SAHIE program is partially funded by the Centers for Disease Control and Prevention's Division of Cancer Prevention and Control (DCPC). The DCPC's National Breast and Cervical Cancer Early Detection Program and its stakeholders use SAHIE to determine the number of low-income, uninsured women who may be eligible for their program at the state and county levels (refer to the "Why Are Small Area Health Insurance Estimates (SAHIE) Important?" text box).

* * *

4 Please refer to the definition of insured at <www.census.gov /programs-surveys/sahie/about/faq.html>.

* * *

HIGHLIGHTS

* From 2018 to 2019, for the population under the age of 65, 90.7 percent of (or 2,850) counties did not have a statistically significant change in their uninsured rate. Among counties that experienced changes in their uninsured rates, more saw an increase (237 counties) than a decrease (54 counties).

* Among the population under the age of 65, the estimated county uninsured rate in 2019 ranged from 2.4 percent to 35.8 percent. The median county uninsured rate was 11.0 percent.

* In 2019, 33.6 percent of (or 1,054) counties had an estimated uninsured rate below 10.0 percent for the population under the age of 65.

* * *

The SAHIE program produces estimates on health insurance coverage at the state and county levels for the full cross-combination of five income-to-poverty ratio (IPR) categories, all incomes, selected age groups, race/ethnicity (state level only), and sex. These IPR categories are defined as the ratio of family income to the federal poverty threshold (refer to the "How Is Poverty Status Measured?" text box). SAHIE data are used to analyze health insurance status by selected characteristics that reflect the federal poverty thresholds and meet the needs of local, state, and federal assistance programs. For instance, the IPR category 0-138 percent of poverty represents the population that may be eligible for Medicaid coverage if they reside in one of the states that expanded Medicaid eligibility under the Patient Protection and Affordable Care Act (ACA).

County-level SAHIE data also allow data users to take a closer look at the distribution and concentration of the uninsured population within states, regions, and metropolitan areas./5

Since the SAHIE program produces single-year estimates for all U.S. counties, SAHIE data are used to analyze changes over time in health insurance coverage, as well as geographic variation. The purpose of this report is to highlight several key findings of such analyses./6

* * *

5 Reference maps on regions and metropolitan/micropolitan area status are available at <https://www2.census.gov /programs-surveys/sahie/reference-maps /2019/ref1-mp-2019.pdf>.

6 All data shown are estimates containing uncertainty. Sources of uncertainty include model error, sampling error, and non-sampling error. Unless specifically noted in the text, apparent differences among the estimates may not be statistically significant. All direct comparisons cited in the text have been statistically tested at the 90 percent confidence level. More information is available at <www.census.gov/programs -surveys/sahie/technical-documentation /source-and-accuracy.html>.

* * *

HEALTH INSURANCE COVERAGE IN U.S. COUNTIES

In 2019, estimated county uninsured rates for the population under the age of 65 ranged from 2.4 percent to 35.8 percent. The median county uninsured rate was 11.0 percent./7

Figure 1 shows how uninsured rates varied among counties throughout the country. The lightest shade in the map represents counties with the lowest uninsured rates (10.0 percent and below). In 2019, 33.6 percent of (or 1,054) counties had an uninsured rate less than 10.0 percent. The Northeast and Midwest had the highest proportion of counties with low uninsured rates./8

In 2019, only 18.1 percent of (or 567) counties had uninsured rates greater than 15.0 percent. These counties were primarily located in the South./9

* * *

7 The median estimated county uninsured rate differs from the ACS's estimated national uninsured rate, which is 9.2 percent (+- 0.1) of the U.S. population under the age of 65 in 2019. The SAHIE program does not produce a national uninsured rate for the United States. SAHIE data are produced using survey estimates from the ACS.

8 The number of counties with uninsured rates below 10.0 percent by region: Northeast--188 out of 217 counties (86.6 percent); Midwest--554 out of 1,055 counties (52.5 percent); South--213 out of 1,422 counties (15.0 percent); West--99 out of 447 counties (22.2 percent).

9 Among the 567 counties with uninsured rates above 15.0 percent, 82.5 percent (468 counties) are in the South. The remaining are in the Midwest (57 counties) and West (42 counties). No counties in the Northeast fell into this category.

* * *

SMALL AREA HEALTH INSURANCE ESTIMATES (SAHIE)

SAHIE are model-based enhancements of the American Community Survey (ACS) estimates, created by integrating additional information from administrative records, postcensal population estimates, and decennial census data. SAHIE methodology employs statistical modeling techniques to combine this supplemental information with survey data to produce estimates that are more reliable. SAHIE are broadly consistent with the direct ACS survey estimates, but with help from other data sources, SAHIE program estimates are more precise than the ACS 1-year and 5-year survey estimates for most counties. Detailed ACS 1-year estimates are not available for most of these smaller geographic areas. A 2019 ACS map of unpublished counties is available at <https://www2.census.gov/programs-surveys/sahie/reference-maps/2019/ref2-mp-2019.pdf>.

Information on the various input data sources used in producing SAHIE is available at <www.census.gov/programs-surveys/sahie/technical-documentation/model-input-data.html>.

SAHIE are subject to several types of uncertainty. Details on this and the SAHIE methodology are available at <www.census.gov/programs-surveys/sahie/technical-documentation/methodology.html>.

* * *

ANNUAL CHANGE IN COUNTY UNINSURED RATES

Between 2018 and 2019, for the population under the age of 65, estimated county uninsured rates significantly decreased in 1.7 percent of (or 54) U.S. counties. More counties experienced a significant increase: 7.6 percent (237 counties). The remaining 2,850 counties did not have a statistically significant change in their uninsured rates.

In 2014, many provisions of the ACA went into effect. From 2013 to 2019, the SAHIE program estimates that 92.7 percent of (or 2,909) counties experienced a significant decrease in their uninsured rates for the population under the age of 65./10

However, the year-to-year changes in county uninsured rates varied.

Figure 2 displays the number of counties where uninsured rates changed from 2013 to 2019. For the periods 2013 to 2014, as well as 2014 to 2015, over 70.0 percent of counties had a significant decrease in their uninsured rates./11

Between 2015 and 2016, that amount dropped to 20.0 percent of (or 629) counties.

* * *

10 When analyzing changes between 2013 and later years, four counties are not included. Bedford County, Virginia, and three counties in Alaska experienced changes in geographic boundaries in 2014. The data for these counties are not comparable to 2013. When analyzing changes between 2017 and later years, three other counties were excluded because of data collection errors in the ACS. For more information, the ACS errata notes are available at <www.census.gov/programs-surveys/acs /technical-documentation/errata.2019 .html>.

11 Between 2013 and 2014, estimated uninsured rates for the population under the age of 65 decreased in 74.1 percent of (or 2,325) counties. Only one county had an increase. From 2014 to 2015, 71.3 percent of (or 2,239) counties experienced a rate decrease. In four counties, the uninsured rate increased. For both periods, the remaining counties had no statistically significant change.

* * *

IMPROVEMENTS TO MEDICAID DATA FOR SAHIE

The SAHIE model utilizes Medicaid enrollment data, among other auxiliary data sources. Major policy changes affected Medicaid in 2014 under the Patient Protection and Affordable Care Act (ACA). Such provisions gave states the option to expand their Medicaid eligibility criteria. To capture changes in the Medicaid enrollment data during this period, the SAHIE program incorporates more up-to-date Medicaid data, starting with the updated 2013 release.

In prior data releases, SAHIE used 2-year lagged Medicaid data from the Medicaid Statistical Information System (MSIS) provided by the Centers for Medicare and Medicaid Services (CMS). For example, the 2013 SAHIE model used 2011 Medicaid data. This 2-year lag is reflected in the 2013 SAHIE data, released in March 2015. In prior years, research supported the 2-year lag because Medicaid enrollment was relatively stable. However, with the implementation of the new ACA provisions in 2014, Medicaid enrollment changed substantially across states. As of December 31, 2019, 33 states and the District of Columbia had expanded their Medicaid enrollment criteria.

The current SAHIE process reduces the 2-year lag of the Medicaid data in the SAHIE model by using more timely sources. SAHIE's updated Medicaid data methods combine MSIS data with two additional Medicaid sources: the CMS Performance Indicator Project Medicaid and Children's Health Insurance Program (CHIP) data, and Kaiser Family Foundation's Medicaid and CHIP data. SAHIE's updated data methods also utilize the most recent Internal Revenue Service 1040 tax data and the American Community Survey estimates to approximate the latest county-level and demographic detail within the state-level Medicaid and CHIP totals. More information on recent changes to SAHIE's use of Medicaid data is available at <www.census.gov/programs-surveys/sahie/technical-documentation /model-input-data/medicaid.html>.

UPDATED 2013 SAHIE DATA AVAILABLE FOR COMPARISON

Methodological improvements, which were applied to 2014 SAHIE and subsequent years, were also used to update 2013 SAHIE for comparability purposes. The original 2013 SAHIE data released in March 2015 (as mentioned above), and the updated 2013 SAHIE released in May 2016, are not comparable due to the changes in SAHIE's use of Medicaid data. The updated 2013 SAHIE was released simultaneously with the 2014 SAHIE data in May 2016. Both data sets are available to download from the SAHIE Web site. For more information, refer to the links in the "Why Are Small Area Health Insurance Estimates (SAHIE) Important?" text box.

* * *

[See link at end of text for Figure 1. Estimated Uninsured Rates for the Population Under the Age of 65: 2019]

Note: The data provided are indirect estimates produced by statistical model-based methods using sample survey, decennial census, and administrative data sources. The estimates contain error stemming from model error, sampling error, and non-sampling error.

Source: U.S. Census Bureau, 2019 Small Area Health Insurance Estimates (SAHIE) program.

* * *

From 2016 to 2017, over 91.0 percent of (or 2,879) counties did not have a statistically significant change in their uninsured rates; however, unlike during the previous 3 years, more counties (183) saw a significant increase than a decrease (79 counties) in their uninsured rates. Similarly, from 2017 to 2018, more counties (81) saw a significant increase than a decrease (70 counties) in their uninsured rates. This increasing trend continued from 2018 to 2019, as even more counties (237) saw a significant increase than a decrease (54 counties) in their uninsured rates.

Given these trends, estimated uninsured rates have fallen below 10.0 percent in many counties. In 2013, only 130 counties, or 4.1 percent of all counties, had an uninsured rate less than or equal to 10.0 percent. In 2018, the number of counties with uninsured rates less than or equal to 10.0 percent increased to 1,184 counties, or 37.7 percent of all U.S. counties. In 2019, the number of counties with uninsured rates less than or equal to 10.0 percent decreased slightly to 1,054 counties, or 33.6 percent of all U.S. counties.

UNINSURED RATES FOR LOW-INCOME, WORKING-AGE ADULTS

The ACA gave states the option to expand Medicaid eligibility to low-income, working-age adults, aged 18 to 64, living at or below 138.0 percent of poverty. Figure 3 contains two maps. The top map displays state Medicaid expansion status as of December 31, 2019. In 2019, two states expanded Medicaid eligibility--Maine and Virginia--making a total of 33 expansion states and the District of Columbia./12

* * *

12 Maine expanded their Medicaid program's eligibility on January 10, 2019. Virginia expanded on January 1, 2019.

* * *

[See link at end of text for Figure 2. Number of Counties With a Change in Their Estimated Uninsured Rates: 2013 to 2019 (Population under age 65)]

* There are 3,142 counties in the United States. The SAHIE program does not include Kalawao County, Hawaii, due to insuficient data. When analyzing changes between 2013 and later years, four counties are not included. Bedford County, Virginia, and three counties in Alaska experienced changes in geographic boundaries in 2014. The data for these counties are not comparable to 2013.

Note: The data provided are indirect estimates produced by statistical model-based methods using sample survey, decennial census, and administrative data sources. The estimates contain error stemming from model error, sampling error, and nonsampling error.

Source: U.S. Census Bureau, 2019 Small Area Health Insurance Estimates (SAHIE) program.

* * *

[See link at end of text for Figure 3. Medicaid Expansion Status by State and Estimated Uninsured Rates for Low-Income, Working Age-Adults by County: 2019]

Source: Centers for Medicare and Medicaid Services (CMS), 2019.

Note: The data provided are indirect estimates produced by statistical model-based methods using sample survey, decennial census, and administrative data sources. The estimates contain error stemming from model error, sampling error, and non-sampling error.

Source: U.S. Census Bureau, 2019 Small Area Health Insurance Estimates (SAHIE) Program.

* * *

The bottom map displays estimated county uninsured rates for low-income, working-age adults who may be eligible for Medicaid. In 2019, county uninsured rates for this population ranged from 6.1 percent to 61.9 percent. The median county uninsured rate among low-income, working-age adults was 22.1 percent. In states that expanded Medicaid eligibility, 9.2 percent of counties (151 out of 1,647 counties) had an estimated uninsured rate above 20.0 percent, compared to 84.8 percent of counties (1,267 out of 1,494 counties) in states that did not expand it.

CHILDREN HAVE LOWER UNINSURED RATES THAN WORKING-AGE ADULTS

At the state level, SAHIE data show that in 2019, children under the age of 19 had a lower estimated uninsured rate than working-age adults, aged 18 to 64, in all 50 states and the District of Columbia. The difference between the two age groups is even found among U.S. counties, where children had significantly lower uninsured rates than working-age adults in 94.5 percent of all counties (Figure 4). There was only one county where the child population had a significantly higher uninsured rate: Dunn County, North Dakota.

* * *

HOW IS POVERTY STATUS MEASURED?

Poverty status is determined by comparing total annual family before-tax income to federal poverty thresholds that vary by family size, number of related children, and age of householder. If a family's income is less than the dollar value of the appropriate threshold, then that family and every individual in it are considered to be in poverty. For people not living in families, poverty status is determined by comparing the individual's total income to their threshold. General information on poverty is available at <www.census.gov/topics/income-poverty/poverty.html>.

The table of federal poverty thresholds is updated annually by the U.S. Census Bureau to allow for changes in the cost of living using the Consumer Price Index (CPI-U). The thresholds do not vary geographically.

SAHIE's primary data input is the estimates of poverty from the American Community Survey (ACS), a continuous survey with people responding throughout the year. Since income is reported for the previous 12 months, the appropriate poverty threshold for each family is determined by multiplying the base-year poverty threshold by the average of the monthly CPI values for the 12 months preceding the survey. More information is available in "How the Census Bureau Measures Poverty" at <www.census.gov/topics /income-poverty/poverty/guidance/poverty-measures.html>.

To determine a family's or an individual's income-to-poverty ratio (IPR), divide a family's or individual's before-tax income by the appropriate federal poverty threshold. Then multiply by 100 to determine how far the family or individual earner is below or above poverty (a family with an IPR of 100 percent is living at the federal poverty threshold).

For example, imagine a family of four, two parents and two children, with a total annual income of $50,000. In 2019, a family of this size had a federal poverty threshold of $25,926. Their income-to-poverty ratio would be:

Total annual income / Federal poverty threshold = $50,000 / $25,926 = 1.929 = 192.9 percent of poverty

The family of four is living just below 200 percent of poverty. This means their income is just below twice the determined federal poverty threshold.

SAHIE Income-to-Poverty Ratio Categories: 0-138 percent, 0-200 percent, 0-250 percent, 0-400 percent, 138-400 percent of poverty, and all incomes.

* * *

[See link at end of text for Figure 4. Estimated Uninsured Rates for Working-Age Adults Aged 18 to 64 and Children Under the Age of 19: 2019]

Note: The data provided are indirect estimates produced by statistical model-based methods using sample survey, decennial census, and administrative data sources. The estimates contain error stemming from model error, sampling error, and non-sampling error.

Source: U.S. Census Bureau, 2019 Small Area Health Insurance Estimates (SAHIE) program.

* * *

WORKING-AGE MEN HAVE HIGHER UNINSURED RATES

In every state and the District of Columbia, the 2019 estimated uninsured rate for working-age men, aged 18 to 64, was higher than for working-age women. Working-age men had a significantly higher uninsured rate than women in 1,661 counties (52.9 percent). There were no statistically significant differences in the remaining counties (Figure 5).

STATE UNINSURED RATES VARY BY RACE AND ETHNICITY

The SAHIE program provides health insurance coverage estimates at the state level by race and ethnicity. In 2019, for the population under the age of 65, non-Hispanic Whites had a lower estimated uninsured rate than Hispanics and non-Hispanic Blacks in every state and the District of Columbia. Non-Hispanic Blacks under the age of 65 also had a lower estimated uninsured rate than Hispanics in every state and the District of Columbia (Figure 6 and Appendix Table 1).

Figure 6 also displays how estimated uninsured rates changed from 2018 to 2019 by race and ethnicity across states. Each line represents the magnitude of change for each group. Longer lines indicate a larger change in the uninsured rate. From 2018 to 2019, for the population under the age of 65, uninsured rates for non-Hispanic Blacks saw significant increases in five states and a significant decrease in one state; the remaining states and the District of Columbia did not have a statistically significant change. Uninsured rates for non-Hispanic Whites significantly increased in 20 states and decreased in three states. Twenty-eight states and the District of Columbia did not have a statistically significant change in their uninsured rates. For the Hispanic population, more states had an increase than a decrease in their uninsured rate. One state had a decrease and 12 states had an increase, while the remaining states did not have a statistically different uninsured rate (Appendix Table 1 for statistically significant changes).

* * *

WHY ARE SMALL AREA HEALTH INSURANCE ESTIMATES (SAHIE) IMPORTANT?

The SAHIE program is partially funded by the Centers for Disease Control and Prevention's Division of Cancer Prevention and Control. It has a congressional mandate to provide screening services for breast and cervical cancer to low-income, uninsured, and underserved women through the National Breast and Cervical Cancer Early Detection Program (NBCCEDP). SAHIE data are used as an important consideration when planning and evaluating public policy on health insurance programs, the impact of common illnesses, or serious health conditions for states and the 3,141 counties in the United States. More information about NBCCEDP is available at <www.cdc.gov/cancer/nbccedp/>.

Additional information is available by data release year from 2000 to 2019. For example, you can download annual reports (for 2010-2019 data release years only), data sets, maps, and interactive data tables at <www.census.gov/programs-surveys/sahie.html>.

The online SAHIE Interactive Data Tool provides detailed customized data tables of the insured and uninsured populations by selected year(s) from 2006-2019, geography (state and county), income-to-poverty ratio (IPR) categories, selected age groups (under the age of 65, ages 18-64, ages 21-64, ages 40-64, ages 50-64, and under the age of 19), sex, and race/ethnicity (state level only). These custom tables can be downloaded to a PDF or CSV file. To access the interactive data online, visit <www.census.gov/data /data-tools/sahie-interactive.html>.

Starting in 2008, SAHIE began utilizing the American Community Survey data. For years prior to 2008, the SAHIE program estimates utilized the Annual Social and Economic Supplement to the Current Population Survey. More information is available at <www.census.gov/programs-surveys/sahie/technical -documentation/methodology/methodology-2008-2019.html>.

* * *

[See link at end of text for Figure 5. Estimated Uninsured Rates for Working-Age Adults Aged 18 to 64 by Sex: 2019]

Note: The data provided are indirect estimates produced by statistical model-based methods using sample survey, decennial census, and administrative data sources. The estimates contain error stemming from model error, sampling error, and non-sampling error.

Source: U.S. Census Bureau, 2019 Small Area Health Insurance Estimates (SAHIE) program.

* * *

[See link at end of text for Figure 6. Change in Estimated Uninsured Rate for the Population Under the Age of 65 by Race and Ethnicity: 2018 to 2019]

* Expanded Medicaid eligibility as of December 31, 2019.

Source: U.S. Census Bureau, 2019 Small Area Health Insurance Estimates (SAHIE) program.

* * *

ACKNOWLEDGMENTS

The Small Area Estimates Branch prepared this document with significant contributions from the Small Area Modeling and Development Branch and the Health and Disability Statistics Branch.

CONTACT

For questions related to the contents of this document, including estimates and methodology of the Small Area Health Insurance Estimates (SAHIE) program, contact the Small Area Estimates Branch at (301) 763-3193 or <[email protected]>.

For questions related to health insurance, income and poverty definitions, the American Community Survey, or other Census Bureau surveys, contact the U.S. Census Bureau Call Center at 1-800-923-8282 (toll free), or visit <https://ask.census.gov>.

SUGGESTED CITATION

Robinson, S. and K. A. Willyard, "Small Area Health Insurance Estimates: 2019," Current Population Reports, P30-09, U.S. Census Bureau, Washington, DC, 2021.

* * *

View figures and report at https://www.census.gov/content/dam/Census/library/publications/2021/demo/p30-09.pdf