After the US Census Bureau released its latest figures on health coverage in September, headlines declared that uninsurance was on the rise in the Trump era.
On the one hand, this is not surprising. Since President Trump’s inauguration, the administration has taken several steps that might shrink coverage, such as cutting spending for Affordable Care Act outreach and navigators and repealing the individual mandate. On the other hand, the economy was improving until the COVID-19 outbreak hit: median incomes reportedly rose in 2019, and unemployment fell, which might well increase coverage rates.
To assess trends in the uninsurance rate, we typically turn to benchmark federal surveys. These surveys have undergone changes in methodology that obscure recent trends in health coverage. However, even when we account for the data limitations, a clear picture of declining coverage emerges — a decline that has cost thousands of lives and worsened health and financial security for many more.
Coverage Losses
Exhibit 1 presents 2016-19 uninsurance estimates, both for the overall population and for non-elderly adults, from three benchmark national surveys: the Current Population Survey (CPS), the National Health Interview Survey (NHIS), and the American Community Survey (ACS). We consider each in turn.
Exhibit 1: Uninsurance estimates from three national surveys, 2016-2019
|
Current Population Survey: ASEC |
National Health Interview Surveyd |
American Community Surveye |
|
|
Overall Population |
|||
|
2016 |
7.9% (25.1 million)a |
9.0% (28.6 million) |
8.6% (27.3 million) |
|
2017 |
7.9% (25.6 million)b |
9.1% (29.3 million) |
8.7% (28.0 million) |
|
2018 |
8.5% (27.5 million)b,c |
9.4% (30.4 million) |
8.9% (28.6 million) |
|
2019 |
8.0% (26.1 million)c |
10.3% (33.2 million) |
9.2% (29.6 million) |
|
Absolute change, 2016 to 2019 |
0.1 percentage point (1.0 million) |
1.3 percentage point (4.6 million) |
0.6 percentage point (2.3 million) |
|
Non-elderly adultsf |
|||
|
2016 |
10.8% (20.8 million)a |
12.4% (24.5 million) |
12.1% (23.3 million) |
|
2017 |
11.0% (21.2 million)b |
12.8% (25.2 million) |
12.3% (23.7 million) |
|
2018 |
11.7% (22.7 million)b,c |
13.3% (26.3 million) |
12.5% (24.1 million) |
|
2019 |
11.1% (21.5 million)c |
14.7% (29.0 million) |
12.9% (24.9 million) |
|
Absolute change, 2016 to 2019 |
0.3 percentage point (0.6 million) |
2.3 percentage point (4.5 million) |
0.8 percentage point (1.6 million) |
Sources:
a Authors’ analysis of the 2017 CPS ASEC Research File: https://www.census.gov/data/datasets/2017/demo/income-poverty/2017-cps-asec-research-file.html.
b 2018 CPS Bridge File: Berchick ER, Barnett JC, Upton RD. Health Insurance Coverage in the United States: 2018. US Census Bureau 2019. https://www.census.gov/content/dam/Census/library/publications/2019/demo/p60-267.pdf.
c 2019 and 2020 CPS ASEC: US Census Bureau. Health Insurance Coverage in the United States: 2019. https://www.census.gov/library/publications/2020/demo/p60-271.html (accessed Sept 15, 2020). Non-elderly adult 2019 figure calculated from percent uninsured ages 19-64 and total number in that age group.
d Health Insurance Coverage: Estimates from the National Health Interview Survey. https://www.cdc.gov/nchs/nhis/healthinsurancecoverage.htm (accessed Sept 16, 2020).
e Census Bureau. American Community Survey Tables for Health Insurance Coverage. The United States Census Bureau. https://www.census.gov/data/tables/time-series/demo/health-insurance/acs-hi.html (accessed Sept 15, 2020).
f Non-elderly adults are defined as individuals aged 19-64 in the Current Population Survey and the American Community Survey and individuals aged 18-64 in the National Health Interview Survey.
The CPS, conducted monthly by the U.S. Census Bureau, is the nation’s primary source of employment data. In February – April of each year, the Bureau conducts an Annual Social and Economic Supplement (ASEC) to the CPS that collects health insurance data. Individuals are classified as uninsured if they report having had no coverage throughout the prior calendar year.
Two issues complicate use of the CPS ASEC to evaluate recent coverage trends. First, to improve the accuracy of health coverage estimates, the Census Bureau undertook a major redesign of CPS health insurance questions in 2014. However, changes in data processing that account for those changes were not fully implemented until the 2019 survey. The upshot, as the Census Bureau has noted, is that calendar year 2013-2017 estimates cannot be directly compared to those of previous years. Meanwhile, calendar year 2018 estimates can be compared to 2016 and 2017 figures only by using specially constructed data files for 2016 and 2017 that incorporate these methodological changes.
Second, the outbreak of the COVID-19 pandemic this Spring upended collection of the 2020 data that serve as the basis for the calendar year 2019 estimates. As a result, the 2019 figures are not comparable to estimates of uninsurance rates in previous years. For safety reasons, in-person interviews were suspended and replaced by telephone interviews. Response rates for the 2020 survey are about 10 percentage points lower than last year, and analysis of administrative data suggested that non-respondents differed from respondents to a greater degree than in previous years. Therefore bias in the 2019 uninsurance estimates is a significant concern.
CPS data show a rise in the total number uninsured from 25.1 million in 2016 to 27.5 million in 2018, and a decline to 26.1 million in 2019 (Exhibit 1). For the reasons mentioned above, however, this 2018 to 2019 change should be interpreted cautiously.
Fortunately, 2019 data from the other two surveys were collected prior to the pandemic’s onset in the US, so it had no effect on their response rates or findings. The NHIS, conducted by the Centers for Disease Control and Prevention (CDC), is a leading health survey. Based on data from the survey, the CDC publishes regular “NHIS Early Release Estimates” that are commonly used to track uninsurance and have proven quite reliable, typically differing from final NHIS estimates by <0.1 percentage point. The standard uninsurance measure reported in the NHIS is lack of coverage at the time of interview.
Like the CPS, the NHIS suggests that uninsurance increased from 2016 to 2018, from 28.6 to 30.4 million. Unlike the CPS, the NHIS found that uninsurance further increased to 33.2 million in 2019.
Unfortunately, the NHIS implemented a major survey redesign in 2019 that complicates interpretation of trends. According to NHIS reports, these design changes may account for “some (but not all) of the increase between 2018 and 2019.” Hence, it seems that the actual change in uninsurance lies somewhere between a 1.8 million increase (the 2016-2018 change) and a 4.6 million increase (the 2016-2019 change).
Finally, the ACS, conducted by the Census Bureau, also collects health insurance data, and has not undergone redesigns that might affect its health coverage estimates. Moreover, as noted above, collection of the 2019 ACS data predated the pandemic. For these reasons, the ACS provides the most reliable data on recent uninsurance trends.
Like both of the other surveys, the ACS found a rise in the number of the uninsured from 2016 to 2018, and, like the NHIS, a further rise from 2018 to 2019. According to the ACS, the number of the uninsured rose from 27.3 million in 2016 to 29.6 million in 2019 — a 2.3 million increase, equivalent to a rise from 8.6% to 9.2% in the overall uninsurance rate.
Findings of increasing uninsurance after 2016 accord with results of other governmental and non-governmental surveys. Most likely, then, more than 2 million individuals lost health coverage during the first three years of the Trump administration. A further, more substantial increase in 2020 seems almost certain to occur. Projections based on unemployment data during the pandemic suggest that 5 to 7 million additional Americans will join the ranks of the uninsured this year.
Consequences For Health And Life
What are the consequences of these coverage losses? Considerable evidence suggests that uninsurance leads to a host of adverse financial, psychological, and health outcomes. In the Oregon Medicaid Experiment, low-income uninsured adults who gained Medicaid coverage experienced a 25% decline in the probability of having unpaid medical bills sent to collection agencies relative to those who remained uninsured. Gaining coverage improved access to mental health care and reduced the probability of screening positive for depression by 30%. It also increased the use of inpatient and outpatient medical care and prescription drugs. Other research, meanwhile, has found that lack of coverage is associated with inferior control of high blood pressure, which is a major cause of stroke, heart attacks, and death.
Perhaps unsurprisingly, findings from one randomized study conducted by the Internal Revenue Service (IRS) and multiple quasi-experimental studies have indicated that lack of insurance costs lives. Based on these data, we estimate the number of excess deaths due to coverage losses during the Trump administration.
Exhibit 2 lists these estimates, which we calculated by dividing Exhibit 1 estimates of coverage losses among non-elderly adults during each year of the Trump administration by estimates of the “number needed to insure” (NNI) in order to save one life annually derived from three quasi-experimental studies. These studies examined the number of deaths averted by each of three coverage changes — pre-ACA state-level Medicaid expansions, ACA Medicaid expansions, and the Massachusetts Health Reform. For the first and third of these studies, researchers defined and reported NNI as the number of individuals who would need to gain insurance coverage to prevent one death over the course of one year; for the second study, we calculated the NNI based on the reported proportion of deaths averted. (The aforementioned IRS study suggests a much larger life-saving effect from insurance coverage than previous studies; to err on the conservative side of estimates, we did not use it in our analysis).
Exhibit 2: Estimated lives lost annually among non-elderly adults due to coverage losses during Trump administration
|
Coverage Losses |
Estimates of Deaths Due to Coverage Losses, Based on Previous Studies of the Number of Deaths Averted by Coverage Expansions |
|||||
|
Year |
Survey |
Increase in non-elderly adult uninsured relative to 2016 baseline |
Study 1 (One Death Annually Per 278 Persons Uninsured)a |
Study 2 (One Death Annually Per 714 Persons Uninsured)b |
Study 3 (One Death Annually Per 830 Persons Uninsured)c |
|
|
2017 |
CPS |
427,000 |
1,536 |
598 |
514 |
|
|
NHIS |
700,000 |
2,518 |
980 |
843 |
||
|
ACS |
426,000 |
1,532 |
597 |
513 |
||
|
2018 |
CPS |
1,869,000 |
6,723 |
2,618 |
2,252 |
|
|
NHIS |
1,800,000 |
6,475 |
2,521 |
2,169 |
||
|
ACS |
826,000 |
2,971 |
1,157 |
995 |
||
|
2019 |
CPS |
632,000 |
2,273 |
885 |
761 |
|
|
NHIS |
4,500,000 |
16,187 |
6,303 |
5,422 |
||
|
ACS |
1,569,000 |
5,644 |
2,197 |
1,890 |
||
|
Total deaths due to coverage losses, 2017-2019 |
CPS |
N/A |
10,532 |
4,101 |
3,528 |
|
|
NHIS |
N/A |
25,180 |
9,804 |
8,434 |
||
|
ACS |
N/A |
10,147 |
3,951 |
3,399 |
||
Sources:
a Study 1 = Analysis of effect of pre-ACA Medicaid expansion. Sommers BD. State Medicaid Expansions and Mortality, Revisited: A Cost-Benefit Analysis. American Journal of Health Economics 2017; 3: 392–421. Figure of 1/278 shown in Exhibit is midpoint of Sommers' estimates.
b Study 2 = Analysis of effect of ACA Medicaid expansions. Miller S, Altekruse S, Johnson N, Wherry LR. Medicaid and Mortality: New Evidence from Linked Survey and Administrative Data. National Bureau of Economic Research, 2019 DOI:10.3386/w26081. Figure of 1/714 shown in Exhibit is calculated from Miller et al's findings for the age 19-64 subgroup.
c Study 3 = Analysis of mortality effect of Massachusetts' coverage expansion. Sommers BD, Long SK, Baicker K. Changes in mortality after Massachusetts health care reform: A quasi-experimental study. Annals of internal medicine 2014; 160: 585–93. Figure of 1/830 shown in Exhibit is Sommers et al's estimate.
Based on the ACS coverage data, we estimate that between 3,399 and 10,147 excess deaths among non-elderly US adults may have occurred over the 2017-2019 time period due to coverage losses during these years. Using the NHIS figures for coverage losses yields a higher estimate (between 8,434 and 25,180 non-elderly adult deaths attributable to coverage losses), while the CPS figures yield an estimate of 3,528-10,532 excess deaths among non-elderly adults. These figures do not completely capture the population effects of coverage loss, as they exclude the excess deaths that would likely result from coverage losses among children. In 2020 and beyond, we can project even more loss of life if, as expected, millions more lose health coverage due to the economic downturn associated with the pandemic.
Declining insurance coverage during the Trump administration has hence come at a heavy cost in physical and mental health, financial security, and loss of life. However, larger policy changes may lie ahead that could have even greater health impacts. In November, the Supreme Court will hear arguments in the case California v. Texas. If, as the Trump administration has urged, it entirely overturns the Affordable Care Act, 19.9 million individuals could lose health coverage. Based on the same approach as outlined above, we estimate that this coverage loss would lead to 22,892 – 68,345 excess deaths among nonelderly adults annually. The life and health ramifications of this case — and of November’s election — are enormous.
Authors’ Note
The authors serve as leaders of Physicians for a National Health Program, a nonprofit organization that favors expanding insurance through a single payer program. None of the authors receive compensation for that work. Some of Gaffney's travel on behalf of the organization has been reimbursed.
"much" - Google News
October 29, 2020 at 07:17PM
https://ift.tt/37Sf79e
How Much Has The Number of Uninsured Risen Since 2016 — And At What Cost To Health And Life? - Health Affairs
"much" - Google News
https://ift.tt/37eLLij
Shoes Man Tutorial
Pos News Update
Meme Update
Korean Entertainment News
Japan News Update
Bagikan Berita Ini
0 Response to "How Much Has The Number of Uninsured Risen Since 2016 — And At What Cost To Health And Life? - Health Affairs"
Post a Comment