Modeling Financial Eligibility for Medicaid Long-term Services and Supports

Abstract

Medicaid plays a significant role in financing long-term services and supports (LTSS) for low-income elderly (65+) in the United States. We modeled the impact of changing income, home equity, and asset limitations on Medicaid eligibility across states. We found that one in five elderly adults (10 million individuals) meet all three tests and would be financially eligible for Medicaid LTSS. Imposing additional restrictions on income allowances and eligibility thresholds had greatest impact on financial eligibility for Medicaid LTSS. Few states have opted to restrict financial eligibility and are instead looking for ways to keep people living independently in the community.

Introduction

Medicaid plays a significant role in financing long-term services and supports (LTSS) for low-income elderly (65+) in the United States. LTSS include not only those localized services provided in nursing facilities but also routine services in a person’s own home including assistance with eating, bathing, dressing, and other household activities such as preparing meals, managing medication, and doing laundry (Kaiser Family Foundation, 2016). These services are expensive, with average costs in 2016 reported at 82,000 USD-$92,000 for a one-year stay in a nursing facility and 46,000 USD per year for home health services (Colello, 2017).

Given the high cost of LTSS, individuals often exhaust their personal resources in paying for services and must rely on Medicaid to finance ongoing care. Medicaid is the fastest-growing component of state budgets and, in FY2016, Medicaid was the largest single component of state spending, accounting for 28.7 percent of all state expenditures (National Association of State Budget Officers (NASBO), 2017). States pay between 50–74.6 percent of all expenditures on LTSS and continue to look for ways to constrain spending. In addition, while users of LTSS represented just 5.9 percent of enrollees in 2013, they represented more than 40 percent ($168 billion) of all spending (Office of Disability, Aging and Long-Term Care Policy, 2018).

States looking to constrain spending for LTSS can look to its eligibility rules and limiting the number of potential enrollees. There are three key aspects of eligibility determination that can be adjusted: household income levels, asset determination, and the assessment of need for LTSS. In general, states are required to provide Medicaid coverage to individuals age 65 or older who receive Supplemental Security Income (SSI) benefits and can choose to extend eligibility to elderly individuals with incomes up to 300 percent of the SSI benefit level, approximately 2,250 USD per month in 2018 for an individual (Office of Disability, Aging and Long-Term Care Policy, 2018).¹ Elderly individuals are also typically required to pass an asset test to qualify for Medicaid, most commonly 2,000 USD for an individual and 3,000 USD for a couple. Finally, in order to qualify for Medicaid payment of LTSS, individuals must demonstrate a level of need for such care. The level of need required varies by state and is based on a person’s need for help with self-care or household activities, the existence of particular medical conditions, or a certain level of cognitive impairment (Kaiser Family Foundation, 2016).

In this paper we model the financial components of Medicaid eligibility rules (income and asset limits) across states and estimate changes to eligibility based on adjustments to these rules. Modeling in this way provides information to better understand the role of different components in determining financial eligibility and the potential impact on enrollment if a state were to restrict eligibility rules for budget purposes. This eligibility model was developed for use in the Minnesota LTSS Projection Model (MN-LPM), which projected growth in Minnesota’s aging population and Medicaid spending on LTSS for years 2020 and 2030. This project was funded by the Minnesota Department of Human Services in a grant to the State Health Access Data Assistance Center (SHADAC).

Resources requirements to determine medicaid eligibility for LTSS

Our work is focused on eligibility for Medicaid LTSS for the population age 65 and older. We use the detailed eligibility framework from Minnesota based on the MN Medicaid Eligibility Policy Manual and in consultation with MN Medicaid eligibility staff. We use the thresholds and allowances most commonly used across states for our modeling scenarios, drawn from the 2015 Kaiser Family Foundation survey of Medicaid eligibility (Watts et al., 2016). Minnesota’s Medicaid eligibility process follows the Special Income Rule used by 44 states (per the Kaiser survey). We present our findings at the national level, aggregating eligibility and projected enrollment across states.

Figure 1 provides the workflow our model uses to determine income eligibility for Medicaid LTSS based on Minnesota’s Medicaid eligibility process. This process demonstrates the complexity and the level of detail required to determine financial eligibility for Medicaid LTSS. There are three key components to determining financial eligibility for Medicaid LTSS: income, assets, and home equity. Each will be described in turn.

Figure 1.

Modeled income eligibility workflow (Minnesota).

Source: [Authors’ analysis of Minnesota Health Care Programs Eligibility Policy Manual.]

Modeled income requirements for LTSS eligibility

Base income

To qualify for Medicaid payment of LTSS, most individuals must spend nearly all of their income on their care. The amount of income individuals must spend on their care is determined using two different income-counting methodologies, the long-term care (LTC) Income Calculation and the Community Income Calculation. The LTC Income Calculation is used for three types of individuals: individuals with incomes at or below 300% of the federal benefit rate (FBR), individuals who have a non-institutionalized spouse (“community spouse”), and nursing facility residents with expected stays of longer than 30 days (“long-term nursing facility residents”). The Community Income Calculation is typically used only for individuals whose income exceeds 300% FBR, do not have a community spouse, and are not long-term nursing facility residents. The current FBR is 750 USD per month for individuals. For purposes of Medicaid LTSS eligibility, all applicants are considered individuals, regardless of marital status or family composition.

To qualify, after deducting the allowances and medical spenddown described below, applicants using the LTC Income Calculation should have no remaining income, and applicants using the Community Income Calculation typically must have incomes at or below 100% of the federal poverty guidelines (FPG) without a spenddown, or at or below 75% FPG with a spenddown.

Allowances

Under the Community Income Calculation, individuals are typically allowed to disregard the first 60 USD of monthly earned income and then disregard half of the remaining earned income. There are three primary monthly allowances under the LTC Income Calculation: an allowance for individuals seeking services in the community (most common: 2,199, USD used in 19 states); a small, personal needs allowance for individuals seeking services in a nursing facility (most common: 50, USD used in 12 states); and an allowance that permits individuals to transfer income to a community spouse (most common: up to 2,981, USD used in 38 states). In addition, individuals can take allowances for expenses such as Medicare premiums, health insurance premiums, and long-term care insurance premiums as well as health insurance deductibles, co-insurance, and co-pays.

Medical spenddown

In Minnesota, applicants are allowed to deduct qualifying medical expenses from their income(s) to spenddown to the relevant income threshold. These expenses include any medically necessary expenditures for medical services, supplies, or devices not subject to payment by a third party. This spenddown can be calculated on a one-month or six-month basis. While many states provide a spenddown option, there are currently 24 states that are considered “income cap” states, which provide set, strict income thresholds with no allowance for spenddown (ElderLaw Answers, 2018b).² Instead, these states allow individuals to put a certain amount of excess income into a “Miller trust,” from which it can be spent on their cost of care. States can recover remaining funds in the Miller trust after the individual’s death to be reimbursed for the cost of their care (NASBO, 2017).

Asset test

To qualify for Medicaid, the value of an individual’s countable assets must be no more than the specified asset limit after allowances. Countable assets typically include any assets that can easily be converted into cash, such as bank savings or investments. Assets such as personal effects, one vehicle used for transportation, and other assets that cannot easily be converted into cash are excluded assets. There is a Community Spouse Asset Allowance that allows up to a certain amount of household assets to be protected for a community spouse (most common: 119,220, USD used in 47 states). Under the Deficit Reduction Act of 2005, the applicant’s homestead is an excluded asset if the individual lives in the residence, is expected to return to the residence, or a community spouse or dependent relative lives in the residence (ElderLaw Answers, 2018a). The most common asset limit is 2,000 USD (used in 38 states), the highest is 4,000 USD (MS and D.C.), and the lowest is 1,500 USD (OH) (NASBO, 2017).

Home equity test

The net value of the individual’s primary residence cannot exceed a limit set by the state. The net value is typically the estimated fair market value (FMV) minus any outstanding mortgages or other loans or liens. The home equity limit does not apply, however, if a dependent relative lives in the home (Minnesota Department of Human Services, 2016). In some states, the home is not considered when determining Medicaid eligibility if the nursing home resident plans to return to the home; in other states, the resident must prove that they are likely to return home (U.S. Department of Health and Human Services, 2005). The most common home equity threshold is also the lowest, set at 552,000 USD and used in 40 states. The highest equity limit is 828,000 USD and is used in 9 states (Watts et al., 2016).

Methods

We developed a model based on key characteristics including income, assets, and expenditures of a nationally representative sample of people age 65 and over, and estimated changes in financial eligibility for Medicaid payment of LTSS based on adjustments to the allowances and thresholds using the most restrictive income and asset rules used by individual states.

Our data come from the 2014 Health and Retirement Study (HRS), a longitudinal household survey of Americans age 50 or older, which has been conducted by the University of Michigan and sponsored by the National Institute on Aging and the Social Security Administration since 1992 (RAND Center for the Study of Aging, 2018). The survey includes rich information on income, assets, functional limitations, and health care expenditures. The nationally representative sample includes 10,388 respondents age 65 or older. The sample includes individuals living in nursing homes and is representative of that population (RAND Center for the Study of Aging, 2019; Sonnega et al., 2014).

Using the HRS data, we model three scenarios: (1) using the most common allowances and limits across states (status quo), (2) using the most restrictive income rules across the states, and (3) using the most restrictive asset rules (Table 2). The thresholds used in our status quo scenario are included in Table 1. These scenarios allow us to better understand the relative impact of the income and asset components of Medicaid eligibility rules on the number of individuals who could be potentially Medicaid eligible.

Table 2.

Status quo, restricted income, and restricted asset scenarios and states where implemented to assess eligibility for medicaid LTSS.

	Allowance/ Limit	Status Quo Scenario	Restricted Scenario	States Where Implemented
Income (Monthly)	Personal Needs Allowance	$50	$30	AL, IL, NC, SC
	HCBS Allowance	$2,199	$600	CA
	Community Spouse Income Allowance	$2,981	$1,991	AL, FL, ME, NJ, NM, NC, OR, SD, VT
Assets	Community Spouse Asset Allowance	$119,220	$66,480	SC
	Asset Limit	$2,000	$1,500	OH
Home Equity	Home Equity Limit	$552,000	No change

Source: Watts et al. (2016).

Table 1.

Most common resource thresholds used across states to determine eligibility for medicaid LTSS.

	Allowance/ Limit	Status Quo Amount	# States That Have Implemented
Income (Monthly)	Personal Needs Allowance	$50	12
	HCBS Allowance	$2,199	19
	Community Spouse Income Allowance	$2,981	38
Assets	Community Spouse Asset Allowance	$119,220	47
	Asset Limit	$2,000	38
Home Equity	Home Equity Limit	$552,000	40

Source: Watts et al. (2016).

For changes to the income threshold, we compare the most restrictive allowances to those most commonly used. In this scenario, we change the Personal Needs Allowance to 30 USD from 50 USD (a 40 percent decrease); we change the HCBS allowance to 600 USD from 2,199 USD (a 73 percent decrease); and we change the Community Spouse Allowance to 1,991 USD from 2,981 USD (a 33 percent decrease). The asset and home equity rules are held constant in the restricted income eligibility scenario.

For changes to the restricted asset eligibility scenario, we hold the income and home equity thresholds constant and vary the assets thresholds. We use the most restrictive Community Spouse Asset allowance of 66,480 USD used in South Carolina compared to the most commonly used allowance of 119,220 USD (a 44 percent difference). For the asset threshold we use the most restrictive asset threshold used in Ohio of 1,500 USD compared to the most commonly used asset threshold of 2,000 USD (a 25 percent difference).

Results

The population of elderly adults studied has an average age of 74.6 and is 56.3 percent female. Among the population, the median household income is 40,912, USD the median value of household net assets is 79,400 USD (excluding housing assets), and the median net primary residence value is 100,000. USD

Figure 2 provides our baseline scenario using the most common thresholds and allowances used across the states for income, assets, and home equity for eligibility for Medicaid LTSS. If the most common state thresholds were applied across all states, we estimate that nearly the entire elderly population would meet the home equity threshold of 552,000. USD Just more than half (54 percent) would meet the home equity and income test, and only 22 percent, or 10 million adults age 65 and older, would meet all three tests – home equity, income, and assets – and be financially eligible for Medicaid LTSS.

Figure 2.

Elderly adults financially eligible for medicaid LTSS under status quo scenario.

Source: [Authors’ analysis of the RAND Health and Retirement Study (HRS) Longitudinal File 2014 (V2) [Public use dataset].]

Figure 3 provides the results of the restricted asset and eligibility scenarios compared to the status quo. We find that applying the most restrictive income allowances across the states would result in an estimated 6.8 million individuals potentially losing financial eligibility for Medicaid LTSS. This reduction in eligibility is statistically significant (p < .001). We find less impact when we focus on the results of applying the most restrictive asset thresholds and allowances while keeping the income allowances at status quo. Under this scenario, we find no statistically significant reduction in the number of elderly individuals eligible for Medicaid LTSS compared to the status quo scenario.

Figure 3.

Elderly adults financially eligible for medicaid LTSS under status quo, restricted asset, and restricted income scenarios.

Notes: [* p < .05, ** p < .01, *** p < .001]

Source: [Authors’ analysis of the RAND Health and Retirement Study (HRS) Longitudinal File 2014 (V2) [Public use dataset].]

Table 3 shows the characteristics of the population age 65 and older by their financial eligibility for Medicaid LTSS under the status quo scenario (columns B and C) and restricted income scenario (columns D and E). All individuals in columns D and E are eligible under the status quo scenario. Relative to the overall population age 65 and older (column A) and the population not financially eligible under each scenario (columns B and D), individuals financially eligible under the status quo (column C) and restricted income (column E) scenarios generally have fewer financial resources and have poorer health (as measured by self-reported health status), limitations to activities of daily living (ADL), low cognitive function, and stays in the hospital or nursing home in the previous two years.

Table 3.

Characteristics of elderly adults by financial eligibility for medicaid LTSS under status quo and restricted income scenarios.

	(A)		(B)		(C)		(D)		(E)
	Elderly Adults (Wgt. N = 46,314,684)		Not Eligible (SQ) (Wgt. N = 36,310,386)		Eligible (SQ) (Wgt. N = 10,004,298)		Not Eligible (RI) (Wgt. N = 6,841,236)		Eligible (RI)(Wgt. N = 3,163,062)
	Estimate	[95% CI]	Estimate	[95% CI]	Estimate	[95% CI]	Estimate	[95% CI]	Estimate	[95% CI]
Mean Age (years)	74.6	[74.3–74.9]	74.5	[74.1–74.8]	75.1	[74.7–75.5]	74.8	[74.4–75.2]	75.7	[74.8–76.6]
Female (%)	56.3	[55.3–57.2]	54.4	[53.4–55.4]	63.0	[61.0–65.0]	62.7	[60.0–65.4]	63.7	[60.3–67.0]
Married, Spouse Present (%)	55.1	[53.8–56.4]	59.3	[57.8–60.8]	39.9	[37.2–42.8]	30.3	[27.5–33.2]	60.8	[56.9–64.5]
White, Non-Hispanic (%)	80.9	[78.2–83.3]	87.6	[85.8–89.1]	56.6	[51.7–61.3]	59.8	[55.1–64.3]	49.6	[43.2–56.0]
Median Household Income ($)	40,912	[38,789– 43,035]	50,400	[48,323–52,477]	16,907	[16,260–17,555]	16,872	[16,135–17,609]	17,040	[15,711–18,369]
Median Household Assets[Non-Housing] ($)	79,400	[66,660– 92,140	165,000	[146,918–183,082]	600	[380–820]	500	[206–794]	900	[312–1,488]
Self-Rated Health Fair/Poor (%)	28.7	[27.3–30.2]	22.5	[21.3–23.7]	51.4	[48.6–54.2]	48.3	[45.1–51.5]	58.1	[53.2–62.8]
2+ ADL Limitations (%)	11.2	[10.3–12.1]	7.9	[7.3–8.6]	23.0	[20.9–25.3]	17.3	[14.8–20.1]	35.5	[31.6–39.5]
Low Cognitive Function (%)	16.9	[15.8–18.0]	11.5	[10.6–12.4]	36.4	[34.3–38.5]	33.3	[30.6–36.1]	43.0	[39.1–47.0]
Hospital Stay in Prv. 2 Yrs. (%)	29.3	[28.1–30.6]	26.7	[25.4–28.1]	38.8	[36.3–41.3]	36.2	[33.5–39.1]	44.2	[39.5–49.1]
Nursing Home Stay in Prv. 2 Yrs. (%)	7.3	[6.8–7.8]	6.7	[6.2–7.1]	9.3	[8.1–10.8]	6.2	[5.2–7.3]	16.2	[12.7–20.5]
Current Nursing Home Resident (%)	2.8	[2.5–3.1]	3.0	[2.7–3.3]	2.1	[1.5–2.8]	0.4	[0.2–0.9]	5.7	[4.1–7.9]
Unweighted N	10,388		7,763		2,625		1,731		894

Under status quo scenario, Medicaid LTSS financial eligibility is modeled using the most common allowances and limits across the states. Under restricted income scenario, Medicaid LTSS financial eligibility is modeled using the most restrictive income rules across the states. See Table 2 for detailed listing of allowances/limits modeled in each scenario.

SQ = status quo scenario; RI = restricted income scenario; CI = confidence interval; ADL = activities of daily living.

Authors’ analysis of the RAND Health and Retirement Study (HRS) Longitudinal File 2014 (V2) [Public use dataset].

To assess if the restricted income scenario improves how well Medicaid LTSS is targeted, it is most relevant to compare the care needs, financial resources, and LTSS utilization of the population excluded from eligibility by the restricted income scenario (column D) to those characteristics of the population not eligible under the status quo scenario (column B). Table 3 shows that the population not eligible under the restricted income scenario (column D) on average has fewer financial resources, is less healthy, likely has a higher level of need for LTSS, but is less likely to be receiving formal LTSS compared to the population not eligible under the status quo scenario (column B). Relative to the population not eligible under the status quo scenario (column B), the population that would become not eligible under the restricted income scenario (column D) is more than twice as likely to report having fair or poor health (48.3 percent versus 22.5 percent), is more than twice as likely to report having at least some difficulty with two or more ADLs (17.3 percent versus 7.9 percent), and is nearly three times as likely to have a low level of cognitive function (33.3 percent versus 11.5 percent), but is no more likely to have had a nursing home stay in the previous two years (6.2 percent versus 6.7 percent) and is less than one fifth as likely to be currently living in a nursing home (0.4 percent versus 3.0 percent). This indicates that although the group made ineligible for Medicaid LTSS by the restricted income scenario (column D) has a high need for services, some individuals in this group may already be facing challenges to accessing formal long-term care services that those with better resources (column B) do not face.

Table 3 also shows that a smaller percent of the population eligible under the status quo scenario (column C) is married compared to the population eligible under the restricted income scenario (column E) (39.9 percent versus 60.8 percent). A greater share of the eligible population under the restricted income scenario is married because the Community Spouse Income Allowance is a larger portion of the total value of the available income allowances compared to the status quo scenario, which requires single individuals to be relatively poorer to be eligible.

Discussion

We modeled income, asset, and home equity eligibility tests used to determine financial eligibility for Medicaid LTSS using a nationally representative sample of individuals age 65 and older to assess the contribution of these three tests to Medicaid eligibility. We used the most common threshold employed by states in the financial eligibility process in order to model the impact of changes to a baseline status quo. Based on current financial eligibility levels, we found that an estimated one out of five adults aged 65 or older – or approximately 10 million individuals – meet all three eligibility tests (home equity, income, and asset), and would be financially eligible for Medicaid LTSS services. We note, however, that not all those who are financially eligible meet the functional eligibility criteria for Medicaid payment of LTSS, making this impact on the number of eligible individuals somewhat theoretical. In addition, not all who are eligible enroll in the Medicaid program, and in 2015 the average monthly Medicaid enrollment for those 65 and older was six million (Office of Disability, Aging and Long-Term Care Policy, 2018).

We also found that restrictions to the income allowances and eligibility thresholds had the largest effect on the number of individuals financially eligible for Medicaid LTSS. There were two main reasons for this result. First, an estimated 82 percent of the target population was eligible to receive one or more income allowances – either the Personal Needs Allowance, the HCBS Allowance, or the Community Spouse Income Allowance – making adjustments to the income allowances a large potential target. In contrast, only 9 percent of the target population received an allowance for the asset component of eligibility (the Community Spouse Asset Allowance). Second, we found that the income rules in general are less restrictive than the asset rules, allowing more room here to restrict eligibility.

We found that the population that would be made financially ineligible for Medicaid LTSS by restricting income allowances and thresholds likely has a greater need for services, is less likely to have a spouse who could potentially provide informal care, has fewer financial resources to pay for formal care, and is less likely to be currently using formal LTSS compared to the population ineligible for Medicaid LTSS under the most common income allowances and thresholds. This indicates that Medicaid LTSS eligibility is already narrowly targeted under the most common allowances and thresholds.

Across the board, states are less restrictive in their application of home equity tests, with almost the entire target population meeting the status quo home equity threshold of 552,000. USD On the surface, this may seem a generous threshold; however, most states have estate recovery laws that allow a state to seek retroactive payments for LTSS from an enrollee’s estate upon their death (ElderLaw Answers, 2017).

While we did not model functional status, we want to note that in addition to financial criteria, enrollees must also meet functional thresholds based on physical and cognitive abilities to determine the level of care required. The use of functional assessment and the application of criteria also varies across the states. The Medicaid and CHIP Payment and Access Commission (MACPAC) documented 124 assessment tools used by states to determine eligibility and care planning, and any restructuring of eligibility must include functional determinations (Medicaid and CHIP Payment and Access Commission (MACPAC), 2016).

Limitations

Our main limitation is that we use one state’s Medicaid LTSS eligibility rules for those ages 65 and older and apply this process to all states. Though Minnesota’s Medicaid LTSS eligibility process is similar to the approach used in most states, it may not be fully representative of all states’ approaches. We also do not include variation in eligibility allowed with Medicaid Home and Community Based Care (HCBC) waivers that allow states to waive Medicaid income and resource limits for those demonstrating a need for nursing facility care. Our goal was to highlight the role income and assets play in Medicaid eligibility and not to model the variation in use of HCBC waivers. While there are clearly differences in the eligibility for Medicaid and for LTSS across states, the complexity inherent in each system makes it difficult to model each state independently. We believe Minnesota’s eligibility model represents the key components of financial eligibility used by all states and provides a reasonable approximation of the impact of changes to these components on eligibility levels. In our modeling, we do take into account the range of eligibility levels of income and assets across states to reflect state variation in these eligibility components.

We also want to note that our primary data source, the Health and Retirement Study, though now said to be representative of the institutionalized population, was not originally designed to be representative of that population and some concerns remain about the sample’s representativeness of the nursing home population, especially when used longitudinally (RAND Center for the Study of Aging, 2019; Sonnega et al., 2014).

Conclusion

Our study highlights the already strict eligibility levels that limit access to Medicaid LTSS. States will feel pressure to tighten eligibility even further in order to address the impending spending increases due to an aging population and increasing health care costs. While reducing the income allowances may make a difference in eligibility, further restrictions will come at a price. The Medicaid program for LTSS among the 65+ population is already well targeted and restricting eligibility would likely exclude individuals in need of services. Few states have opted to further restrict access to needed services and are instead opting to find more ways to keep people living independently in the community. Reduced or delayed access to needed LTSS will have a negative impact on both health status and utilization patterns, and any upfront cost savings are likely to be offset on the acute care side in terms of increased preventable hospital and emergency department use, and on the long-term care side in terms of increased long-term nursing facility stays (DePalma et al., 2012; Hass et al., 2015; Sands et al., 2006; Stewart & Irvin, 2017; Xu et al., 2012).

We suggest that policymakers think more broadly and look at federal as well as state solutions for the future financing of LTSS. In 2013, several strategies were recommended to address this funding need, including the possibility of a broader finance solution that spreads out the cost risk via a social insurance program and would “gradually increase the federal share of Medicaid financing for long-term services and supports, thereby reducing burdens on the states” (Long-Term Care Commission). State policymakers face considerable challenges in paying for LTSS for a growing elderly population and the significant costs of LTSS. We have pointed to key areas of eligibility determination in constraining costs and highlight the already tight restrictions placed on individuals to meet financial eligibility criteria. We caution policymakers who feel pressure to constrain eligibility for Medicaid LTSS as a cost-savings measure against taking this action.