Health Insurance Affordability Concerns and Health Care Avoidance Among U.S. Adults Approaching Retirement

Abstract

Importance:

As U.S. adults approach Medicare eligibility at age 65, they face important decisions about health care and employment. Recent legislative, regulatory, and legal challenges to the Affordable Care Act may add new uncertainties to this decision-making.

Objective:

To understand adults’ perspectives on health insurance, health care, and employment near retirement.

Design:

In October 2018, a cross-sectional online survey was conducted with completion rate=62%.

Setting:

Ipsos KnowledgePanel®, a nationally representative internet survey panel.

Participants:

Community-dwelling U.S. adults, age 50-64.

Main Outcome(s) and Measure(s):

Confidence in affording health insurance; keeping a job or delaying retirement to have employer-sponsored health insurance; concern about potential changes to health insurance due to changes in federal policies; and avoidance of medical care or medication due to cost.

Results:

Among 1,028 respondents, 52% (95% CI 49%-55%) were female, 40% had high school education or less (95% CI 37%-44%), and 65% (95% CI 62%-68%) were employed while 35% (95% CI 32%-38%) were retired or not working. Approximately one-quarter of respondents age 50-64 (27%, 95% CI 25%-30%) had little to no confidence in being able to afford health insurance over the next year, and nearly half (45%, 95% CI 41%-48%) had little to no confidence in their ability to afford health insurance when they retire; 14% (95% CI 12%-17%) reported keeping a job or delaying or considered delaying retirement (11%, 95% CI 10%-14%) to have employer-sponsored health insurance. Most respondents (68%, 95% CI 65%-71%) were very or somewhat concerned about potential changes to their health insurance due to changes in federal policies. In the past year, 13% (95% CI 11%-16%) did not get medical care and 12% (95% CI 10%-14%) avoided filling a prescription medication because of cost. After adjusting for demographic and health characteristics, individuals with low confidence in health insurance affordability either in the next year or in retirement were significantly more likely to avoid medical care (adjusted odds radio [aOR] 2.89, 95% CI 11.86-4.49) and to avoid filling a prescription medication (aOR 2.87, 95% CI 1.71-4.80) because of cost concerns.

Conclusions and Relevance:

Many adults age 50-64 worry about their ability to afford health insurance in retirement and about potential future changes to their health insurance options due to changes in federal policy. Policy solutions are needed to help patients better understand their coverage options and navigate health care in this critical transition period before Medicare eligibility.

Introduction

As U.S. adults approach Medicare eligibility at age 65, they face complex decisions about health insurance and health care.¹ These include considerations of current or anticipated future health conditions, and related decisions about whether or when retirement should occur. Although Medicare provides near universal coverage for U.S. adults age 65 and older, some may retire earlier for health or other reasons,² while others may continue working well past that age. Adults age 50-64 represent a group for whom such decisions can be complicated and consequential, as they commonly develop chronic health conditions^3,4 and face rising out-of-pocket health care costs,^1,5 but are not yet age-eligible for Medicare coverage. Therefore, as chronic health conditions accumulate and because a lack of health insurance has been associated with unmet health needs and poor health and financial outcomes, maintaining health insurance in this period is especially critical.^6-8

The Affordable Care Act’s (ACA) formation of Marketplace plans and expansions of Medicaid coverage have created a greater number of potential insurance options for 50-64-year-olds, further complicating decisions about health insurance near and in retirement.^9,10 Recent legislative, regulatory, and legal challenges to the ACA may create additional uncertainty in this decision-making process.^11,12 Since January 2017, challenges at the federal level include Congress’s attempted repeal and replace of the ACA, removal of the penalty associated with the individual mandate to have health insurance, and announcements by the Department of Justice stating that the ACA pre-existing condition protections would no longer be enforced.^11-13

Since implementation of the ACA and its subsequent challenges, there has been limited research comprehensively examining 50-64-year-olds’ perceived challenges to navigating health insurance as they approach retirement and how that might influence decisions about health care and employment. Prior studies of this age group have primarily focused on tracking the uninsured rate, which has declined from 15% in 2010 to 6% in 2015 after implementation of the ACA’s major coverage provisions,¹² and likely had continued declines in later years.¹⁴ One 2014 survey conducted among all non-elderly adults found limited numeracy and literacy among many adults across socioeconomic strata, and that nearly half of individuals with these limitations had difficulty navigating the process of selecting a health insurance plan.¹⁵ Prior to the ACA, McWilliams and colleagues found increased health care utilization and costs after acquiring Medicare coverage at age 65 among previously uninsured adults with a history of diabetes, hypertension, heart disease, or stroke, compared with adults with those conditions who were previously insured.¹⁶ However, there has been limited research examining how adults age 50-64 navigate and use health insurance and health care since implementation of the ACA.

In the current setting of major changes to U.S. health insurance under the ACA and more recent federal health policy uncertainty, the objective of this study was to understand the perspectives of U.S. adults age 50-64 with regard to health insurance affordability, health-insurance-related decision-making about employment including retirement, and decision-making about health care. Also assessed was whether potential concerns about health insurance affordability were associated with decisions to delay or avoid health care.

Methods

Study Design

A cross-sectional online survey of 50-64-year-old adults was conducted in October 2018, as part of the University of Michigan National Poll on Healthy Aging (NPHA). This study was reviewed and deemed exempt by the University of Michigan Institutional Review Board, as it was a study of deidentified respondents. Thus, the requirement for informed consent was waived.

Data Source

The NPHA is a recurring nationally representative survey of U.S. households (https://www.healthyagingpoll.org/) conducted using the Ipsos KnowledgePanel®, a large, nationally representative probability-based panel (Ipsos Public Affairs, LLC). The poll is directed by the University of Michigan Institute for Healthcare Policy and Innovation and sponsored by AARP and Michigan Medicine, the academic medical center for the University of Michigan. NPHA surveys are conducted online 2-3 times each year with U.S. adults ages 50-80. This study was conducted with a subset of adults age 50-64 as part of a multi-part survey that included questions about health insurance, medication storage, memory, and pets.

Callegaro and DiSogra’s method for computing response metrics for online panels was used.¹⁷ For the overall survey fielding that included modules for adults age 50-80, a random sample of 3,202 panel members was drawn from Ipsos’s KnowledgePanel®. 2,051 (excluding breakoffs) responded to the invitation and 2,051 qualified for the survey, yielding a final stage completion rate of 64.1% and a qualification rate of 100% percent. The recruitment rate for this study, reported by Ipsos, was 13.2% and the profile rate was 63.5%, for a cumulative response rate of 5.4%.

Measures

Concern about health insurance affordability was assessed with the following survey items: “Please rate your confidence with the following: 1) being able to afford the cost of your health insurance and out of pocket costs over the next year; and 2) being able to afford the cost of your health insurance and out of pocket costs when you retire.” Responses options included not at all confident, slightly confident, moderately confident, and very confident. Respondents were categorized as having low confidence in health insurance affordability if they reported being not at all confident or slightly confident to either or both of these survey items.

Avoidance of health care was determined by a no/yes response to the following items: “In the last year, have you 1) not gotten medical care because of how much it would cost you; and 2) not filled a prescription because of how much it would cost you?”. Respondents were considered to have avoided health care if they reported not getting medical care and/or did not fill a prescription.

Employment decision-making was assessed by a no/yes response to the following survey items: “In the past year, have you 1) kept a job specifically to have health insurance through your employer; and 2) delayed or considered delaying retirement specifically to have health insurance through your employer?”

Concern about federal policy changes to health insurance was assessed with the item, “How concerned are you about potential changes to your health insurance coverage due to changes in federal policies?” Response options included very concerned, somewhat concerned, not very concerned, or not at all concerned. Responses reporting being very or somewhat concerned were combined to indicate those concerned about federal policy changes.

Standard demographics on panel members were obtained from Ipsos, except for employment status, which was obtained from the NPHA survey. For 13 respondents (0.6% of the sample) who did not respond to the survey question, employment status was obtained from Ipsos data.

Statistical analysis

Post-stratification survey weights provided by Ipsos were applied to reflect the U.S. population and estimate nationally representative descriptive demographic and health statistics. Each survey item had less than 1-2% missing responses and over 97% of respondents missed two or fewer survey items. All respondents were included in the analysis but percentages are reported for each survey item only among those who answered the item. Thus, missing responses are dropped from the denominator.

For comparisons of survey responses between respondents who were working and not working, chi-square testing was used to determine statistical significance. To determine the association between health insurance affordability concerns and avoidance of health care, a multivariable logistic regression analysis was conducted, adjusting for age, gender, race/ethnicity, education, income, marital status, employment status, health status, and type of health insurance. Sensitivity analyses with alternative categorizations of variables and models with additional covariates were also conducted. Two-sided p<0.05 were considered statistically significant. All analyses were conducted with Stata version 15 (StataCorp, College Station, TX).

Results

Survey sample

The study sample included 1,028 community-dwelling U.S. adults age 50-64 (completion rate of 62% among panel members contacted to participate), with 52% (95% CI 49%-55%) of respondents female, 68% (95% CI 65%-72%) Non-Hispanic White, 12% (95% CI 10%-14%) Non-Hispanic Black, and 13% (95% CI 11%-16%) Hispanic (Table 1).¹⁷ Forty percent (95% CI 37%-44%) reported high school or less as their highest level of education, and 18% (95% CI 15%-21%) had annual household incomes of $30,000 or less. Most respondents were employed (65%, 95% CI 62%-68%), while 35% (95% CI 32%-38%) were retired or not working. Less than half (43%, 95% CI 40%-46%) rated their health as excellent or very good, 41% (95% CI 37%-44%) rated their health as good, and 17% (95% CI 14%-19%) rated their health as fair or poor. Most respondents had employer-sponsored health insurance (66%, 95% CI 63%-69%); 8% (95% CI 7%-10%) had an individual private insurance plan; 8% (95% CI 6%-10%) had Medicaid; 8% (95% CI 6%, 10%) had Medicare; 3% (95% CI 2%-4%) were dually enrolled in Medicare and Medicaid; and 3% (95% CI 2%-4%) had VA/Tricare coverage. Four percent (95% CI 3%-6%) were uninsured.

Table 1.

National Poll on Healthy Aging Respondent Demographics, U.S. Adults Ages 50-64, October 2018

	N=1,028
	n	Weighted % (95% CI)
Age
50-54	305	33.3 (30.2, 36.5)
55-59	393	34.6 (31.7, 37.6)
60-64	330	32.1 (29.2, 35.2)
Gender
Male	506	48.2 (45.0, 51.4)
Female	522	51.8 (48.6, 55.0)
Race/ethnicity
White, non-Hispanic	762	68.4 (65.1, 71.5)
Black, non-Hispanic	93	11.7 (9.6, 14.1)
Hispanic	101	12.9 (10.7, 15.5)
Other, non-Hispanic	72	7.0 (5.4, 9.1)
Education
≤ High school	342	40.4 (37.2, 43.7)
Some college	340	27.0 (24.5, 29.8)
≥ Bachelor’s degree	346	32.6 (29.7, 35.6)
Total annual household income
≤ $30,000	151	17.9 (15.4, 20.8)
$30,000-$59,999	194	19.7 (17.2, 22.4)
≥ $60,000	683	62.4 (59.2, 65.5)
Marital status
Married/partnered	740	69.8% (66.7, 72.7)
Not married/partnered	288	30.2% (27.3, 33.3)
Employment status
Employed	693	65.2% (62.1, 68.3)
Retired/Not working	335	34.8% (31.7, 37.9)
Health status
Excellent/very good	448	42.9 (39.8, 46.1)
Good	417	40.5 (37.4, 43.6)
Fair/poor	159	16.7 (14.3, 19.3)
Insurance type
Employer-sponsored	699	66.2 (63.0, 69.3)
Individual private	81	8.3 (6.7, 10.3)
Medicare	77	7.8 (6.2, 9.8)
Medicaid	60	7.8 (6.1, 10.0)
Dual Medicare/Medicaid	27	2.9 (2.0, 4.3)
VA/Tricare	33	2.8 (1.9, 3.9)
None	36	4.2 (3.0, 5.8)

Respondents were overall similar to nonrespondents, except that a greater proportion of nonrespondents were female, Black or Hispanic (eTable 1). Survey estimates incorporate weights that account for these differences.

Health insurance affordability concerns

Approximately one-fourth (27%, 95% CI 25%-30%) of respondents reported low confidence in their ability to afford health insurance and out-of-pocket costs over the next year (Table 2). Nearly half (45%, 95% CI 41%-48%) reported low confidence in their ability to afford health insurance when they retire. A total of 47% (95% CI 44%-51%) of respondents had low confidence in health insurance affordability either in the next year or during retirement. Respondents who were retired or not working at the time of the survey were more likely to report low confidence in health insurance affordability in the next year, but not in retirement, compared to those who were working (Table 2).

Table 2.

Perspectives of U.S. Adults Ages 50-64 on Health Insurance, Health Care and Employment

	N=1,028 (n)	All Respondents	Weighted % (95% CI) Respondents who are working	Respondents who are retired or not working
Little to no confidence in health insurance affordability a
Over the next year	268	27.4 (24.6, 30.4)	23.8 (20.6, 27.4)	34.2 (28.9, 39.8)*
When retired	453	44.6 (41.4, 47.8)	46.6 (42.8, 50.5)	40.8 (35.3, 46.5)
Either	475	47.3 (44.1, 50.5)	48.6 (44.8, 52.6)	44.8 (39.2, 50.5)
Avoided health care in past year due to cost
Medical care	141	13.2 (11.2, 15.5)	13.9 (11.4, 16.7)	12.1 (8.9, 16.2)
Prescription medication fill	119	11.9 (9.9, 14.1)	9.6 (7.6, 12.2)	16.1 (12.3, 20.8)*
Either	190	18.2 (15.9, 20.8)	16.8 (14.1, 19.9)	21.0 (16.8, 25.9)
Concerned about potential federal policy changes to health insurance
Yesb	686	67.7 (64.6, 70.6)	65.9 (62.1, 69.5)	71.2 (65.8, 76.0)
Kept job in last year to keep employer health insurance
Yes	145	14.1 (12.0, 16.5)	19.2 (16.2, 22.5)	4.7 (2.8, 7.9)*
Delayed or considered delaying retirement in last year to keep employer health insurance
Yes	123	11.4 (9.6, 13.6)	15.1 (12.5, 18.0)	4.6 (2.7, 7.5)*

^aRespondents who were slightly or not at all confident.

^bRespondents who were very or somewhat concerned.

^*p<0.05 by χ² test for comparison of respondents who are working and respondents who are retired or not working.

Decisions about employment and retirement

With regard to employment decision-making, 14% (95% CI 12%-17%) of respondents reported keeping a job in the last year specifically to maintain employer-sponsored health insurance, and 11% (95% CI 10%-14%) delayed or considered delaying retirement to keep their employer-sponsored health insurance (Table 2). Overall, 19% (95% CI 16%-21%) of respondents either kept a job, considered delaying retirement, or delayed retirement to maintain their employer-sponsored health insurance. Respondents who were working were more likely to report either keeping a job or considering delaying retirement to maintain employer-sponsored health insurance, compared to those who were retired or not working (Table 2).

Concerns about changes to federal health insurance policy

A majority (68%, 95% CI 65%-71%) of respondents were very or somewhat concerned about potential future changes to their health insurance coverage due to changes in federal policies (Table 2). Respondents who were working were similarly likely to those who were retired or not working to report these concerns (Table 2).

Avoidance of medical care and association with health insurance affordability concerns

Approximately 13% (95% CI 11%-16%) of respondents said that they did not get medical care in the past year because of cost (Table 2). In addition, nearly 12% (95% CI 10%-14%) reported not filling a prescription medication because of cost concerns. Overall, 19% (95% CI 16%-21%) either avoided medical care or filling a prescription because of cost concerns. Respondents who were retired or not working at the time of the survey were more likely to report avoiding filling a prescription, but not medical care, compared to those who were working (Table 2).

After adjusting for demographic and health characteristics, individuals with low confidence in health insurance affordability either in the next year or in retirement were significantly more likely to avoid medical care because of cost concerns (adjusted odds radio [aOR] 2.89, 95% CI, 1.86-4.49). Individuals with health insurance affordability concerns were also more likely to avoid filling a prescription medication (aOR 2.87, 95% CI 1.71-4.80). Overall, adults age 50-64 with low confidence in health insurance affordability had three times the odds of either avoiding medical care or avoiding filling a prescription due to cost concerns (aOR 3.00, 95% CI 2.00-4.50).

Sensitivity analyses

In sensitivity analyses, additional models were considered that classified the primary predictor as low confidence in health insurance affordability 1) in the next year only; or 2) in retirement only. Findings were similar to the primary analysis. Models that included as covariates measures of health insurance literacy (defined as confidence in understanding health insurance terms, covered services, and out of pocket costs) were also considered. Findings remained similar.

Discussion

In this nationally representative survey of U.S. adults age 50-64 conducted after implementation of, and subsequent legal challenges to, the ACA, nearly half of adults in this age group had low confidence in their ability to afford health insurance in retirement and a majority were concerned about potential future changes to their health insurance due to changes in federal policies. Individuals reporting concerns about health insurance affordability were significantly more likely to avoid medical care, including prescription medication fills. This is troubling, given the frequent diagnosis of chronic health conditions among adults over age 50,³ the high burden of poor health with 17% of our respondents rating their health as fair or poor, and the overall rising morbidity and mortality for low-socioeconomic status individuals in this age group.^18,19 These results suggest that the health insurance affordability concerns identified could translate to negative long-term health outcomes for U.S. adults approaching Medicare age. In addition, as health insurance may be associated not only with health but also with capacity to work,^20,21 such health insurance concerns may impact whether adults age 50-64 feel flexibility to change jobs or retire, both directly and indirectly.

The current findings about affordability concerns among adults age 50-64 track with other recent national data about U.S. workers’ retirement planning. In 2019, the Employee Benefit Research Institute found that 41% of workers were not confident they would have enough money to take care of their medical expenses during retirement.² This is not surprising, as 50-to-64-year-olds spend 10% of their household budget on health care, and this percentage increases with age.²²

It is important to note that the timing of Medicare eligibility at age 65 may not necessarily correlate with the timing of retirement, as many individuals will retire before or after age 65. In this sample, 14% reported they had already retired. Adults age 50-64 who are healthy may continue to work for many years, while those with multimorbidity due to several chronic conditions may need to retire early for health reasons. Given the variability in timing of retirement,²³ concerns about health insurance affordability near retirement likely have less to do with concerns about Medicare and more to do with costs of current plans or transitions between plans. Rising out-of-pocket costs in recent years^1,24 may negatively affect people’s perceptions of their ability to afford insurance now and in the future. Likewise, adults age 50-64 may worry about transitions between employer-sponsored insurance plans, or from employer-sponsored insurance to Medicaid or Marketplace plans. In fact, prior studies of adults in their 50s and 60s have found that they are more likely to be enrolled in Marketplace plans compared to younger adults,²⁵ and that this proportion has grown between 2013 and 2015.¹

In addition to concerns about affordability of health insurance, 14% of adults age 50-64 reported keeping a job specifically to have employer-sponsored health insurance, suggesting ‘job lock’, while 11% reported delaying or considering delaying retirement to have employer-sponsored health insurance. The ACA's insurance coverage expansion was intended, in part, to reduce job lock and allow individuals to change or leave their job without concern about becoming uninsured.²⁶ However, the current findings suggest that many older adults age 50-64 still worry about maintaining employer-sponsored insurance and keep a job for that reason, perhaps due to concern about affordability of other health insurance options. Although other studies have found no change in labor force participation or early retirement related to ACA Marketplace or Medicaid expansion coverage,¹⁴ these results suggest that the ACA may not have full eliminated job lock for 50-64-year-olds who remain employed.

Limitations

The current study has several potential limitations. First, as with all surveys, there is the potential for recall or response bias. However, the use of a large national panel and response rate mitigate some of the concern for response bias, and survey weights were applied to account for any response differences. The selection of survey items focused on relatively recent behavior and attitudes also help mitigate potential recall bias. Second, it is possible that respondents to an online survey may differ from the general population. However, nearly 90% of U.S. adults age 50-64 used the internet by 2018.²⁷ Several studies have found similar characteristics among older adult respondents to internet vs. non-internet surveys, while other studies have found that internet samples were younger than in-person samples.²⁸ Third, this is a cross-sectional study conducted after legal challenges to the ACA; results should be interpreted in the national policy context at the time of the survey in October 2018. Fourth, the correlation identified between health insurance affordability concerns and avoidance of medical care does not signify causality. As this is a cross-sectional study, it is not possible to determine direction of effects, for example whether worries about health insurance led to avoidance of medical care or vice versa. It is also possible that there are other unobserved mediators of our outcomes, though results were adjusted for standard demographic and health characteristics.

Implications

Despite these limitations, the study findings have important policy implications for multiple stakeholders. For health care providers, the results underscore the importance of discussions with patients about out-of-pocket costs to help inform decisions about timing, choice, and appropriateness of health care services.²⁹ For employers and health insurance navigators, the findings suggest that these groups should help individuals near retirement understand coverage options and navigate health care in this critical transition period. Such education and outreach efforts are needed to help adults navigate the myriad possible health insurance transitions that can occur at age 65. For policymakers, it is clear that health insurance and health care remain top issues for U.S. adults age 50-64 and that this group worries about how policymakers are addressing it.³⁰ Policy solutions are needed to enhance stability of health insurance affordability and availability for adults in this age group.

Currently, several proposals are being considered by Congress to extend health insurance options to adults age 50-64, as well as to other age groups. These proposals include a “Medicare for all” or single-payer health insurance plan that would replace the current employer-sponsored health insurance system; a renewed proposal for a “public option” in the ACA Marketplace; and Medicare buy-in proposals that would allow adults age 50-64 the option to buy into Medicare coverage before age 65.³¹ Future research should continue to investigate the perspectives of adults age 50-64 as changes are considered to federal health insurance policy.

Conclusions

In summary, this study found that many adults age 50-64 worry about their ability to afford health insurance in retirement and a majority worry about potential future changes to their health insurance options due to changes in federal policy. Of concern, such worries about health insurance were associated with avoidance of medical care. While federal legislative efforts to change the ACA and other health insurance reforms largely failed in 2018, these findings suggest that health care policy solutions are needed to help patients near Medicare age understand their coverage options and navigate health care in this critical transition period.

Table 3.

Association of Low Confidence in Affording Health Insurance with Avoidance of Health Care Among U.S. Adults Ages 50-64

	Adjusted odds ratioa (95% CI)
	Avoided medical care	Avoided filling a prescription	Avoided either medical care or filling a prescription
Low confidence in health insurance affordability b	2.89 (1.86, 4.49)*	2.87 (1.71, 4.80)*	3.00 (2.00, 4.50)*
Age
50-54	Ref	Ref	Ref
55-59	1.12 (0.68, 1.84)	0.48 (0.28, 0.84)*	0.81 (0.52, 1.28)
60-64	0.98 (0.57, 1.70)	0.60 (0.34, 1.08)	0.80 (0.49, 1.30)
Gender
Male	Ref	Ref	Ref
Female	1.58 (1.03, 2.41)*	1.88 (1.19, 2.99)*	1.88 (1.27, 2.77)*
Race/ethnicity
White, non-Hispanic	Ref	Ref	Ref
Black, non-Hispanic	0.48 (0.20, 1.12)	0.46 (0.18, 1.18)	0.45 (0.21, 1.00)
Hispanic	0.58 (0.28, 1.21)	1.27 (0.62, 2.60)	0.89 (0.47, 1.68)
Other, non-Hispanic	0.76 (0.28, 2.05)	0.97 (0.41, 2.31)	1.05 (0.45, 2.45)
Household income
≤ $30,000	Ref	Ref	Ref
$30,000-$59,999	0.95 (0.47, 1.91)	0.59 (0.28, 1.23)	0.68 (0.35, 1.30)
≥ $60,000	0.56 (0.28, 1.14)	0.41 (0.19, 0.90)*	0.39 (0.20, 0.77)*
Education
≤ High school	Ref	Ref	Ref
Some college	1.60 (0.95, 2.70)	1.49 (0.85, 2.63)	1.60 (1.00, 2.58)
≥ Bachelor’s degree	1.30 (0.74, 2.31)	2.03 (1.06, 3.88)*	1.78 (1.06, 3.00)*
Marital status
Married/partnered	Ref	Ref	Ref
Not married/partnered	0.75 (0.47, 1.20)	1.17 (0.67, 2.04)	0.93 (0.59, 1.46)
Employment status
Working	Ref	Ref	Ref
Retired/Not working	0.38 (0.21, 0.68)*	0.51 (0.26, 1.00)*	0.54 (0.32, 0.93)*
Health status
Excellent/very good	Ref	Ref	Ref
Good	1.87 (1.15, 3.03)*	1.73 (0.98, 3.03)	1.75 (1.14, 2.71)*
Fair/poor	3.64 (1.94, 6.83)*	4.30 (2.29, 8.04)*	3.99 (2.30, 6.92)*
Health insurance type
Employer-sponsored	Ref	Ref	Ref
Individual private	2.78 (1.40, 5.52)*	2.25 (1.07, 4.75)*	2.66 (1.38, 5.14)*
Medicare	2.59 (1.03, 6.51)*	6.87 (2.76, 17.10)*	3.02 (1.32, 6.89)*
Medicaid	0.55 (0.17, 1.78)	1.96 (0.76, 5.05)	0.87 (0.35, 2.14)
Dual Medicare/Medicaid	1.85 (0.55, 6.22)	4.69 (1.45, 15.18)*	1.67 (0.57, 4.85)
VA/Tricare	0.46 (0.07, 2.92)	c	0.31 (0.05, 1.98)
None	4.2 (1.77, 9.91)*	3.36 (1.26, 8.98)*	3.34 (1.40, 7.96)*

^aAdjusted for age, gender, race/ethnicity, income, education, marital status, employment status, health status, and health insurance type.

^bRespondents with low confidence (slightly or not at all confident) in health insurance affordability either in the next year or in retirement. Reference group is respondents who were moderately or very confident.

^cEmpty cell in multivariable analysis

^*p<0.05

Key Points.

Question:

Do U.S. adults have concerns about health insurance in retirement, and if so, are these concerns associated with avoidance of medical care?

Findings:

In this nationally representative survey of adults age 50-64, 45% had low confidence in affording health insurance in retirement and 68% were concerned about potential changes to health insurance due to changes in federal policies. Individuals with low confidence in health insurance affordability were significantly more likely to avoid medical care and medications.

Meaning:

Policy solutions are needed to help patients better understand coverage options and navigate health care during the critical transition period before Medicare eligibility.