Risk of Financial Precarity from Hospitalization: Implications for Targeting Financial Assistance in Medicare

Abstract

Background:

High out-of-pocket costs in Medicare may leave many beneficiaries in financial precarity. Beneficiaries with modest incomes are often ineligible for Medicaid (which covers most out-of-pocket Medicare costs) and may have insufficient resources to pay an unexpected health care bill. This has prompted calls to improve financial protections, but the target population remains uncharacterized.

Objective:

To identify beneficiaries who would face financial precarity if exposed to the Medicare Part A hospital deductible ($1,600).

Design:

Cross-sectional study of the 2018 wave of the Health and Retirement Study.

Setting:

United States.

Participants:

Community-dwelling Medicare beneficiaries with incomes >100% to ≤400% of federal poverty level.

Measurements:

Nationally representative estimates of financial precarity, defined as having insufficient funds to pay the deductible, examined across four scenarios that considered checking/savings account balances, total liquid assets (with a reserve for future living costs), and supplemental insurance.

Results:

Among 4,881 beneficiaries (population weighted N=26,619,823), 45.0% had insufficient funds in checking/savings accounts to pay the Medicare hospital deductible. Financial precarity was more prevalent among Black and Hispanic vs. White beneficiaries (73.5% and 76.2% vs. 36.2%), those with less vs. more than high school education (70.0% vs. 37.1%), and with ≥3 vs. ≤2 chronic conditions (49.2% vs. 39.1%). In defining financial precarity to include beneficiaries with insufficient liquid assets to pay the deductible while maintaining a $5,000 reserve for future living expenses, 50.7% were financially precarious. Building off this definition to assume supplemental insurance covered the deductible, 39.0% remained financially precarious.

Limitations:

Cost-sharing exposure is limited to hospitalization.

Conclusions:

Many Medicare beneficiaries with modest incomes are at risk of financial hardship from costs of a single hospital stay.

Introduction

Medicare beneficiaries face substantial cost-sharing for most inpatient care, outpatient services, and prescription drugs. Nationally, 36% of beneficiaries report difficulty paying medical bills or delaying care due to cost concerns (1), and those with multiple chronic conditions and serious illnesses are at particular risk for high out-of-pocket costs and economic hardship. supplemental insurance through Medicaid or private plans can shield individuals from these costs (2–6). However, eligibility for Medicaid benefits that cover Medicare cost sharing is limited to those with incomes ≤100% of the federal poverty level and few assets (7). Furthermore, the share of beneficiaries with private supplemental insurance has declined in recent years (8). Consequently, some analysts have called for extending financial protections in Medicare to more broadly address the prevalence of financial precarity among older adults (9–12).

There is no consensus on how to define financial precarity or target greater protections to financially vulnerable Medicare beneficiaries. Prior proposals have focused on income- or asset-based thresholds, but these approaches have limitations (11,13). For instance, defining “near-poor” beneficiaries as those with incomes slightly above the Medicaid eligibility threshold (e.g., 101–150% of the federal poverty level (FPL))(11,12,14) may miss individuals with limited ability to pay costs associated with an unexpected health event, such as the Medicare Part A deductible for a hospitalization ($1,600 in 2023) (15). Some beneficiaries whose incomes exceed the range considered to be “near-poor” might be unable to pay this amount even after depleting savings. Asset-based thresholds are similarly limited as they fail to capture the long-term sequelae of financial precarity: beneficiaries may be able to pay for an unexpected health expenditure in the short-term, but doing so could deplete reserves for future living expenses or health care costs (16). These limitations may be especially consequential for populations at risk for adverse health outcomes, such as racial and ethnic minority groups or individuals with multiple chronic conditions, who are more likely to experience household financial volatility over time and have limited assets at baseline (17).

Ensuring a minimum level of financial protection is a key objective of policymakers and insurance design. The goals of this study are threefold: to measure financial precarity among Medicare beneficiaries in scenarios that model the ability to pay out-of-pocket costs for a hospitalization; to examine how this population differs in scenarios that account for different financial resources, preservation of savings for future living expenses, and supplemental insurance; and to examine demographic and health characteristics of this population across the scenarios.

Methods

Data and Sample

We used the 2018 wave of the Health and Retirement Study (HRS) to identify community-dwelling, Medicare-enrolled respondents. The HRS is a nationally representative survey that follows cohorts of adults ages 50 and older biennially. Black and Hispanic older adults are oversampled (18). The survey collects detailed respondent-reported data on demographics, health status, insurance, income, and wealth. This study was exempt from institutional board review at the University of Pennsylvania because it used public HRS data.

We selected a sample of Medicare beneficiaries with incomes >100% to ≤400% of the FPL for three reasons. First, this range is above the cutoff for Medicaid benefits that cover Medicare cost-sharing. Medicare beneficiaries can qualify for either partial or full Medicaid benefits to assist with cost-sharing. Partial Medicaid benefits, which cover Medicare cost-sharing, is limited to individuals with incomes ≤100% of the FPL and modest assets (≤$9,430 for individuals and ≤$14,130 for couples in 2024) (19). Full Medicaid benefits, which additionally cover Medicaid-funded services such as long-term care, is limited to those with incomes below state-specific cutoffs (ranging from ≤75% to ≤100% of the FPL) and few assets (≤$2,000 for individuals and ≤$3,000 for couples) (20). Second, this income range constitutes the “economic middle” of the Medicare population (approximately 51% of beneficiaries). Finally, it aligns with income ranges prioritized for financial assistance in other programs, such as premium subsidies for Affordable Care Act marketplace insurance (21).

Among community-dwelling Medicare-enrolled respondents, 110 were excluded due to missing covariate data (eFigure 1).

Respondent finances and supplemental insurance variables

The HRS assesses respondent-reported finances across several different asset classes (e.g., checking, savings, and individual retirement accounts). Balances in these accounts were reported as exact dollars if provided by respondents, and as ranges if respondents were unable or unwilling to give exact amounts. We analyzed the RAND HRS file, which imputed balances for respondents who did not report exact values and used information from respondent-reported ranges to bound these values (22). In the 2018 survey wave, balances in individual retirement accounts were imputed for 13% of respondents. The RAND HRS has been used extensively to estimate eligibility for Medicaid and other means-tested (i.e., income-based eligibility) programs for older adults (20,23,24).

We examined liquid assets (e.g., balances in checking, savings, and retirement accounts), rather than illiquid assets (e.g., housing and business assets), for two reasons. First, liquid assets can be drawn down immediately to pay for expenses, while illiquid assets may not be readily converted into cash. Second, our approach is consistent with Medicaid’s asset eligibility test, which does not count the value of an applicant’s home towards total assets (20). We analyzed two measures of liquid assets: one including balances in checking and savings accounts, and another encompassing all liquid assets (i.e., retirement accounts, Keogh accounts, certificates of deposit, stocks, and bonds, in addition to checking and savings accounts).

We assessed whether a respondent reported having private supplemental insurance (excluding long-term care insurance), such as Medigap or an employer-sponsored or retiree plan. Although the HRS does not report benefit design features for individual supplement plans, most individuals who purchase Medigap have plans with the most comprehensive cost-sharing coverage, including full coverage of the Part A deductible (25). Therefore, we assumed individuals with private supplemental insurance would be fully shielded from the Medicare hospital deductible.

Demographic variables

Respondent-reported demographic measures included respondent age, education (categorized as less than high school, high school or equivalent, college or higher), household composition (single or couple), and race and ethnicity (White non-Hispanic, Black non-Hispanic, Hispanic, other). Race and ethnicity were included due to known disparities in household wealth and supplemental insurance among Black and Hispanic older adults (1,26). Health status was assessed from respondent-reported chronic conditions (including arthritis, cancer, diabetes, heart disease, hypertension, lung disease, or stroke) and difficulty performing activities of daily living (ADLs, walking, dressing, bathing, eating, getting in and out of bed, using the toilet).

Analysis

The primary exposure was a financial shock based on the $1,600 Part A Medicare deductible for a hospital stay as representative of an unexpected, costly medical event. The primary outcome was the assessment of respondent-level financial precarity, which we examined across four scenarios (Table 1).

Table 1. Description of four financial shock scenarios.

All Scenarios: Individual must pay a $1,600 hospital bill.

Scenario	Payment source	Financial Precarity Definition
Scenario 1	Checking & Savings accounts	Insufficient balance
Scenario 2	Supplemental insurance if enrolled; Checking & Savings accounts otherwise	No supplemental insurance and insufficient balance
Scenario 3	Liquid assets (IRA/Keough accounts, stocks, bonds, CDs, Checking & Savings accounts)	Remaining balance below either $5,000 or the minimum balance required to produce income stream above the FPL for the person’s remaining life
Scenario 4	Supplemental insurance if enrolled; Liquid assets otherwise	No supplemental insurance AND remaining balance below either $5,000 or the minimum balance required to produce income stream above the FPL for the person’s remaining life.

Notes: $1600 is the Medicare Part A deductible for a hospitalization. A person is considered to have supplemental insurance if they have any private insurance; Medicaid typically does not cover cost sharing for people with incomes greater than 100% FPL, so it is not considered supplemental insurance. Liquid assets is a sum of IRA and Keough accounts, stocks, bonds, CDs, and Checking & Savings accounts. The minimum balance required to produce income stream above the FPL for a person’s remaining life is calculated based on the person’s life expectancy, fixed income, and an assumed annuitization of 3% over all liquid assets.

The first scenario defined financial precarity as having insufficient resources in checking and savings accounts to pay the $1,600 Part A deductible. We assessed individuals’ ability to pay the deductible from balances in checking and savings accounts, rather than from income, because many older adults live on fixed incomes and have limited flexibility to pay a large expense after covering essential living costs (4). For this “paycheck-to-paycheck” population, dipping into savings is often necessary to pay for a large, unexpected bill.

The second scenario expanded on the first but modeled individuals with private supplemental insurance, which we assumed fully covered the hospital deductible.

The third scenario accounted for all liquid assets, including balances in individual retirement and Keogh accounts, stocks, bonds, and certificates of deposit, in addition to checking and savings accounts. This scenario defined financial precarity as having insufficient liquid assets to cover the hospital deductible without going below a minimum financial buffer to cover future living expenses. We defined this buffer as the greater of $5,000 or the minimum amount of savings necessary to protect against future impoverishment. The threshold of $5,000 was chosen based on existing asset limits in social safety-net programs, such as Medicaid, the Supplemental Security Income program, and the Supplemental Nutrition Assistance Program, which tend to be very low and often less than $5,000 for an individual. We used an individual’s life expectancy (given their current age), sources of fixed income (e.g., Social Security benefits), and an annuitization rate of 3% on all liquid assets to calculate the minimum savings necessary to remain above the federal poverty level for the remainder of their expected lifetime (see eFigures 2a/2b for details). We examined this scenario because older adults increasingly depend on savings in retirement investment accounts, rather than pensions, for future income (27). Draining these reserves to cover a hospital bill increases the risk that individuals face future impoverishment.

The fourth scenario expanded on the third but, again, considered whether beneficiaries had supplemental insurance to pay the hospital deductible. Although the third and fourth scenarios account for additional resources, the threshold used to define financial precarity is more stringent. Therefore, the prevalence of financial precarity may be higher under these latter two scenarios.

We described the demographic and health characteristics of respondents identified as financially precarious across the four scenarios. All estimates were weighted using HRS survey weights to produce nationally representative estimates. P-values were not calculated due to the exploratory nature of the analysis. Analyses were performed using SAS 9.4.

Modeling assumptions and sensitivity analyses

Individuals in traditional Medicare must pay the Part A deductible out-of-pocket if they do not have supplemental insurance. Enrollees in Medicare Advantage plans, the private alternative to traditional Medicare, were included in our study and considered to not have supplemental insurance unless they reported having a comprehensive insurance plan outside of their Medicare Advantage plan (e.g., insurance from a former employer).

Medicare Advantage plans typically charge a per-diem copayment for a hospital stay rather than a deductible, with copayment amounts varying across plans (28). To account for this, we conducted a sensitivity analysis that considered the out-of-pocket cost for a hospital stay (without supplemental insurance) to be $900 in Medicare Advantage and $1,600 in traditional Medicare. This $900 amount reflects a median per-diem inpatient copayment of $300 in Medicare Advantage plans and a 3-night hospital stay (see eFigure 3 for details).

We also conducted a sensitivity analysis to account for debt when measuring financial resources. Given our focus on liquid assets, we analyzed unsecured debts (e.g., credit card and medical debts) that are not secured by collateral such as a home, which we subtracted from liquid assets.

Finally, to characterize the persistence of financial precarity over time, we used data from the 2014, 2016, and 2018 biennial HRS waves and evaluated the extent to which the same beneficiary was financially precarious across survey waves.

Role of the Funding Source

This research was supported by the National Institute on Aging of the National Institutes of Health (K23AG073512; R01AG076437). The funder had no role in the design, conduct, or analysis of this manuscript.

Results

Sample Characteristics

The sample included 4,881 community dwelling HRS respondents enrolled in Medicare in 2018 with incomes >100% to ≤400% of the FPL (eFigure 1). Based on survey weighting, this sample represented an estimated 26.6 million people in the community-dwelling population of Medicare beneficiaries ages 50 and older. In the survey-weighted sample, the mean (standard deviation, SD) age was 73.3 (9.0) years, 58.2% of individuals were female, 11.1% identified as Black, and 8.9% identified as Hispanic. Approximately 15.7% had less than a high school education, 54.2% were married or partnered, 58.0% had ≤3 chronic conditions, 6.8% had ≤3 ADL difficulties, and 34.1% were enrolled in a private supplemental insurance plan. Compared to the overall community-dwelling Medicare population, our sample with incomes >100% to ≤400% of FPL included a smaller share of older adults with a college education, lower rates of marriage and partnership, and a larger share of beneficiaries with ≤3 chronic conditions or ≤3 ADL limitations (Table 2).

Table 2.

Sample Characteristics

	Sample Population Income >100% to <=400% of the FPL		General Population Any Income
Unweighted N	4,881		9,144
Weighted N	26,619,823		53,461,452
	Mean (SD)		Mean (SD)
Age	73.3 (9.0)		72.5 (8.5)
	Weighted N	%	Weighted N	%
Gender
Male	11,124,413	41.8%	24,069,523	45.0%
Female	15,495,410	58.2%	29,391,929	55.0%
Race/Ethnicity
White non-Hispanic	20,336,574	76.4%	41,425,269	77.5%
Black non-Hispanic	2,965,334	11.1%	5,762,097	10.8%
Hispanic (any Race)	2,361,398	8.9%	4,434,895	8.3%
Other Race/Ethnicity	909,851	3.4%	1,760,093	3.3%
Unknown	46,666	0.2%	79,098	0.2%
Educational attainment
Less than high school	4,178,610	15.7%	7,081,670	13.3%
High school	10,681,775	40.1%	17,348,723	32.5%
College or greater	11,751,287	44.1%	29,022,908	54.3%
Unknown	8,151	0.0%	8,151	0.0%
Household composition
Single-headed household	12,191,526	45.8%	21,333,543	39.9%
Couple-headed household	14,428,297	54.2%	32,127,909	60.1%
Activities of Daily Living
3+ ADL limitations	1,807,065	6.8%	3,095,198	5.8%
<3 ADL limitations	24,812,758	93.2%	50,366,254	94.2%
Chronic conditions
3+ chronic conditions	15,445,794	58.0%	28,137,399	52.6%
<3 chronic conditions	11,174,029	42.0%	25,324,053	47.4%
Supplemental insurance
Has	9,069,326	34.1%	32,766,949	38.7%
Doesn’t have	17,550,497	65.9%	20,694,503	61.3%
Liquid asset quartiles	Range	Median	Range	Median
Q1	$0–$144	$0	$0-$500	$0
Q2	$144–$9,000	$2,000	$500–$35,000	$5,300
Q3	$9,000–$116,000	$48,000	$35,000–$294,000	$100,500
Q4	$116,000+	$340,000	$294,000+	$700,000

Note: Ns reflect estimated population sizes based on survey weights. Supplemental insurance is considered to be any private insurance; Medicaid typically does not cover cost sharing for people with incomes greater than 100% FPL, so it is not considered supplemental insurance. Liquid assets is a sum of IRA and Keough accounts, stocks, bonds, CDs, and Checking & Savings accounts. HRS is a survey of adults aged 50+.

Source: HRS 2018

Estimates of Financial Precarity

Table 3 summarizes estimates of financial precarity across the four scenarios. In Scenario 1, 45.0% of beneficiaries had insufficient balances in checking and savings accounts to pay the $1,600 Part A deductible. In Scenario 2, 34.6% had insufficient balances in checking and savings accounts and did not have supplemental insurance. In Scenario 3, 50.7% did not have sufficient liquid assets to pay the Part A deductible while also maintaining a minimum financial reserve to cover future living expenses. In Scenario 4, 39.0% had insufficient liquid assets while maintaining a reserve for future living costs and lacked supplemental insurance.

Table 3.

Number and percentage of Medicare beneficiaries in financial precarity in four financial shock scenarios

	Weighted denominator	Scenario 1			Scenario 2			Scenario 3			Scenario 4
		N	%	95% CI	N	%	95% CI	N	%	95% CI	N	%	95% CI
Overall population	26,619,823	11,975,222	45.0	43.2, 46.8	9,218,393	34.6	32.9, 36.3	13,506,121	50.7	48.9, 52.5	10,370,123	39.0	37.2, 40.7
Race and ethnicity
White	20,336,574	7,363,307	36.2	34.2, 38.2	5,245,140	25.8	23.9, 27.7	8,390,780	41.3	39.2, 43.4	5,997,963	29.5	27.5, 31.5
Black	2,965,334	2,179,028	73.5	70.0, 77.0	1,812,629	61.1	57.1, 65.1	2,430,198	82.0	78.9, 85.0	2,006,470	67.7	63.9, 71.5
Hispanic	2,361,398	1,798,847	76.2	71.4, 80.9	1,604,026	67.9	62.6, 73.2	1,946,632	82.4	78.2, 86.7	1,722,763	73.0	67.9, 78.0
Other race/ethnicity	909,851	589,763	64.8	55.2, 74.4	529,101	58.2	48.1, 68.2	694,234	76.3	68.1, 84.5	615,430	67.6	58.5, 76.8
Education
Less than high school	4,178,610	2,925,424	70.0	66.4, 73.6	2,529,722	60.5	56.6, 64.5	3,181,941	76.2	72.8, 79.5	2,709,287	64.8	61.0, 68.7
High school or equivalent	10,681,775	4,681,764	43.8	41.1, 46.6	3,534,682	33.1	30.5, 35.7	5,553,881	52.0	49.3, 54.7	4,216,963	39.5	36.8, 42.2
College or higher	11,751,287	4,363,692	37.1	34.4, 39.9	3,149,647	26.8	24.3, 29.3	4,768,632	40.6	37.8, 43.4	3,442,206	29.3	26.7, 31.9
Activities of Daily Living
3+ ADL limitations	1,807,065	1,197,035	66.2	59.9, 72.6	966,098	53.5	46.6, 60.4	1,352,056	74.8	68.9, 80.8	1,059,417	58.6	51.8, 65.5
<3 ADL limitations	24,812,758	10,778,187	43.4	41.6, 45.3	8,252,295	33.3	31.5, 35.0	12,154,065	49.0	47.1, 50.8	9,310,706	37.5	35.7, 39.3
Chronic conditions
3+ chronic conditions	15,445,794	7,605,161	49.2	46.9, 51.6	5,914,628	38.3	36.0, 40.6	8,590,375	55.6	53.3, 57.9	6,708,339	43.4	41.1, 45.8
<3 chronic conditions	11,174,029	4,370,061	39.1	36.4, 41.8	3,303,765	29.6	27.1, 32.1	4,915,746	44.0	41.3, 46.8	3,661,784	32.8	30.2, 35.3
Household composition
Single-headed households	12,191,526	6,192,074	50.8	48.2, 53.4	4,782,255	39.2	36.7, 41.8	7,001,267	57.4	54.9, 60.0	5,428,058	44.5	41.9, 47.1
Couple households	14,428,297	5,783,148	40.1	37.7, 42.5	4,436,138	30.8	28.5, 33.0	6,504,854	45.1	42.6, 47.5	4,942,065	34.3	31.9, 36.6

Notes: See scenario descriptions in Table 1. Population reflects Medicare beneficiaries living in the community with nominal income >100% to <=400% of the FPL. Ns reflect estimated population sizes based on survey weights. Percentages reflect estimated prevalence of financial precarity in the row populations. For example, in Scenario 1, 73.5% of Black beneficiaries and 76.2% of Hispanic beneficiaries are considered financially precarious. Confidence intervals (CI) are 95%. Hispanic individuals of any race are classified as Hispanic; individuals classified as White and Black are non-Hispanic. HRS is a survey of adults aged 50+

Source: HRS 2018

Financial Precarity in Beneficiary Subpopulations

The prevalence of financial precarity varied across subpopulations of Medicare beneficiaries (Table 3). For example, in Scenario 1, the prevalence of financial precarity among Black and Hispanic beneficiaries was double that of White beneficiaries (73.5% vs. 76.2% vs. 36.2%, respectively). In Scenario 4, 67.7% of Black beneficiaries and 73.0% of Hispanic beneficiaries were financially precarious, compared to 29.5% of White beneficiaries.

Even within the same income range, financial precarity was more prevalent among Black and Hispanic beneficiaries than among White beneficiaries (Figure 1). For example, 87.4% of Black and Hispanic beneficiaries with incomes >100% to ≤140% of the FPL were financially precarious in Scenario 4 vs. 54.4% of White beneficiaries in the same income range. Financial precarity was also more common among beneficiaries with less than a high school education than among beneficiaries in the same income range who had at least a high school education (Figure 2).

Figure 1:

Financial precarity in Scenario 4 by income among Black and Hispanic Medicare beneficiaries vs. White Medicare beneficiaries

Notes: Bars reflect the percent of Medicare beneficiaries without supplemental insurance and insufficient savings to cover the $1,600 Medicare Part A deductible without going below a minimum buffer. Solid lines reflect the weighted populations.

Figure 2:

Financial precarity in Scenario 4 by income among Medicare beneficiaries with less than high school completion vs. high school completion or higher

Notes: Bars reflect the percent of Medicare beneficiaries without supplemental insurance and insufficient savings to cover the $1,600 Medicare Part A deductible without going below a minimum buffer. Solid lines reflect the weighted populations.

In Scenario 1, a higher proportion of beneficiaries with at least 3 chronic conditions (49.2% vs. 39.1%) and at least 3 ADL limitations were financially precarious (66.2% and 43.4%) relative to their respective comparison groups. This pattern was consistent across the four scenarios.

Sensitivity Analyses

In our first sensitivity analysis, we considered beneficiaries enrolled in Medicare Advantage plans to have an out-of-pocket hospital stay cost of $900, reflecting the cost of a 3-night hospital stay with a $300 per-diem copayment; we continued to model a $1,600 deductible in traditional Medicare. Estimates of financial precarity in this analysis ranged from 32.2% to 50.4% across the four scenarios and were 0.2 to 2.7 percentage points lower than our primary estimates (eTable 1). Beneficiaries who identified as Black or Hispanic, those with lower educational attainment, and those with multiple chronic conditions and functional limitations, remained more likely to be categorized as financially precarious than other beneficiaries.

In other sensitivity analyses, we found the prevalence of financial precarity was similar when we deducted unsecured debts from liquid resources (eTable 2). Among respondents classified as financially precarious according to our fourth scenario, 80.1% of those with financial precarity in the 2014 survey wave remained financially precarious by the 2018 wave (eTable 3).

Discussion

This nationally representative study found that 34.6–50.7% of Medicare beneficiaries with incomes >100% to ≤400% of the federal poverty level face financial precarity because they do not have sufficient resources or supplemental insurance to cover the out-of-pocket costs associated with a hospital stay. The prevalence of financial precarity varied according to how we measured available resources (e.g., balances in checking and savings accounts vs. all liquid assets), if we considered the need to reserve funds for future living costs, and whether we accounted for supplemental insurance. However, across all scenarios, the proportion of beneficiaries facing financial precarity exceeded 30%. Financially precarious beneficiaries were more likely to be Black or Hispanic vs. White, have less than a high school education, and have multiple chronic conditions and functional limitations. Financial precarity was also persistent over time, with 80% of beneficiaries who were financially precarious in 2014 remaining so by 2018.

These findings have implications for targeting financial assistance to Medicare beneficiaries. Medicaid, which is the predominant source of support for cost-sharing among Medicare beneficiaries, uses both income and asset limits to target assistance (7). Our findings demonstrate that individuals at the same nominal income level may not have similar resources to withstand a financial shock. This pattern was evident, for example, among Black and Hispanic beneficiaries and those with less than high school education levels, for whom the prevalence of financial precarity was considerably higher than other beneficiaries with similar incomes. Due to systemic disparities in wealth and supplemental insurance among older adults, strict income eligibility limits that exclude all individuals above an income cutoff (regardless of assets) may lead to a regressive allocation of financial protections.

Furthermore, whether an individual can afford the cost of a health care bill depends partly on the level of savings needed to cover future living costs. Older adults increasingly rely on savings in individual retirement accounts, particularly as employers have phased out pension benefits. While account balances of several thousand dollars may appear sufficient for paying a health care bill, they may not be enough to cover living expenses for the remainder of a beneficiary’s life, especially as many older adults have limited retirement savings (29). Medicaid asset limits may not adequately account for the need to maintain a financial buffer. This feature is a particular concern for full Medicaid, where asset limits, which in most states are $2,000 for individuals and $3,000 for couples, have been unchanged since 1989 and have not kept pace with rising living and health care costs (20). Consequently, receiving Medicaid requires individuals to either exhaust most of their resources, which increases future impoverishment risk, or transfer their assets (which is relatively rare) (30).

Our findings suggest that a large proportion of Medicare beneficiaries are just one hospitalization away from depleting their financial reserves and potentially entering a poverty trap. Certain subgroups of beneficiaries may be particularly at risk from such financial shocks, including Black and Hispanic beneficiaries, who have seven-to-eight times lower savings compared to White beneficiaries (1,26). Such financial precarity can also exacerbate health vulnerabilities. For example, wealth shocks can be associated with an increased risk of mortality among older adults (31), and sicker beneficiaries are more likely to be financially precarious, highlighting the bidirectional relationship between health and wealth (32–34). This relationship operates through multiple pathways, including but not limited to, household financial strain as well as the ability to maintain employment and earnings. Accrual of medical debt, although somewhat less prevalent in the older adult population (35) may also play a key role in setting health-wealth traps. Furthermore, while some policies may mitigate certain aspects of the health-wealth trap for younger, working-age populations (e.g., expansion of the earned income tax credit), for older adults, fewer policies for financial protection are available outside of Medicaid. Therefore, our findings indicate that the risk of financial hardship from health-related expenditures among Medicare beneficiaries with modest incomes is common, substantial, and worthy of greater policy attention.

To address Medicare beneficiaries’ financial vulnerability from health expenditures, policymakers could consider several reforms. One option is to revise Medicaid eligibility criteria to consider whether individuals have limited income or assets, as opposed to current approaches that restrict eligibility based on income and assets. This change may benefit individuals who depend on savings to cover future living costs. Another option is to apply more generous income and asset cutoffs to cover a broader range of individuals facing financial precarity. For example, policymakers could expand eligibility for partial Medicaid benefits (provided via the Medicare Savings Programs) that cover Medicare premiums and cost-sharing (36). Finally, policymakers could consider capping out-of-pocket costs in Medicare—a feature of the Medicare Advantage program that is not present in traditional Medicare. Although recent policy reforms have expanded some financial protections in Medicare (e.g., capping annual out-of-pocket Part D spending), they do not target the “economic middle” of beneficiaries who are financially vulnerable but currently qualify for little or no assistance.

This study has several limitations. First, cost-sharing for Medicare beneficiaries is present across the care continuum, but our analysis focuses only on hospital costs. We also did not evaluate the costs of long-term care, which is a major driver of medical costs among older adults (37). Second, this analysis focused on older adults with incomes between 100–400% of FPL which may limit generalizability. However, this population accounts for one-half of Medicare beneficiaries and is consistent with income ranges in which policymakers have targeted economic subsidies in other health insurance programs (e.g., Affordable Care Act Marketplaces). Third, the HRS measures assets using a one-year lookback period, although resources may fluctuate over shorter or longer time horizons. Fourth, our calculation of minimum savings to avoid future impoverishment was based on three factors (life expectancy, sources of fixed income, and an assumed annuitization rate) and may be an imperfect assessment.

Conclusion

Between one-third and one-half of Medicare beneficiaries with incomes >100% to ≤400% of FPL lack sufficient financial resources to pay for a single hospital stay. This population of financially precarious Medicare beneficiaries disproportionately includes Black and Hispanic individuals, beneficiaries with lower education, and those in poor health. Broadening financial protections in Medicare to beneficiaries with moderate incomes and limited assets may achieve greater economic parity and equity among older adults by increasing assistance to those with limited savings and gaps in supplemental insurance.