LTC Bullet: New Data on LTC Incidence, Duration, Cost and Financing Sources
Friday, July 24, 2015
LTC Comment: New numbers, better than the old numbers, but they require further clarification and explanation.
LTC BULLET: NEW DATA ON LTC INCIDENCE, DURATION, COST AND FINANCING SOURCES
Highlights from the report:
Highlights from our analysis:
LTC Comment: The Department of Health and Human Services’ (DHHS) Assistant Secretary for Planning and Evaluation (ASPE) has just published (July 2015) an “Issue Brief” titled “Long-term Services and Supports for Older Americans: Risks and Financing.” Read it here. (Never mind the report’s use of the awkward neologism “LTSS.” What they mean is formal, HIPAA-level “LTC” wherever it is provided. We’ll use the clearer, traditional term.)
This new report is an important contribution to our understanding of the incidence, duration, cost and financing sources for long-term care. But it’s a big change from what we used to think, i.e., that 70 percent of the elderly will require some LTC and 20 percent will need five years or more of care. (For our critique of the study that generated those old estimates, see LTC Bullet: Microsimulate This!, March 28, 2006.) We’re asked now to believe that only 52.3 percent will need any formal LTC and that only 13.9 percent will require five years or more.
Big change. What shall we make of the new utilization numbers, lower risk estimates, and funding source information? That’s what today’s LTC Bullet is about. But, bottom line, these new data give a better picture of the reality of long-term care, because they take into account the cost of housing (not just care) in residential settings and because they focus on higher-acuity, more clearly defined HIPAA-level care for two or more ADLs or incidental to cognitive impairment.
So, this is progress, but that said, let’s go through the report, quote by quote, analyze and comment.
Quote: The issue brief’s “abstract”: “Most Americans underestimate the risk of developing a disability and needing long-term services and supports (LTSS). Using microsimulation modeling, we estimate that about half (52%) of Americans turning 65 today will develop a disability serious enough to require LTSS, although most will need assistance for less than two years. About one in seven adults, however, will have a disability for more than five years. On average, an American turning 65 today will incur $138,000 in future LTSS costs, which could be financed by setting aside $70,000 today. Families will pay about half of the costs themselves out-of-pocket, with the rest covered by public programs and private insurance. While most people with LTSS needs will spend relatively little on their care, about one in six (17%) will spend at least $100,000 out-of-pocket for future LTSS.” (p. 1)
LTC Comment: Those are powerful, but confusing numbers. At age 65, you have roughly a 50/50 chance of needing long-term care that will cost $138,000. But you’d only have to set aside $70,000 to cover that cost. On the other hand, you have a 17% probability of spending $100,000 on LTC out of pocket even though half the cost of long-term care will be paid by public programs or private insurance. You’ll need to read the whole report to unravel this confusion, but we’ll try to clarify the meaning in the following quotes and comments, with a special focus on any ideological bias that has crept into ASPE’s exposition.
Quote: “Most Americans who receive formal LTSS pay out-of-pocket. For those with longer spells, they may pay out-of-pocket until they qualify for Medicaid. Reliance on Medicaid for those that cannot afford the full costs of LTSS may result in increased federal and state spending for LTSS.” (p. 2)
LTC Comment: It’s true that most people pay privately for formal LTC at least for a while. It is also true that they continue paying privately while they receive Medicaid benefits. This report does not explain how such private payment works nor how it impacts the LTC financing and service delivery system. The report simply assumes that people spend down their wealth before qualifying for Medicaid. The truth is much more complicated and critical to understand.
First, Medicaid LTC benefits are easily available to high-income people, because anyone with income below the cost of a nursing home (at least several thousands of dollars per month) qualifies based on income. Second, Medicaid’s LTC asset exemptions are nearly unlimited. Uncounted assets include most home equity and a car, term life insurance, prepaid burial plans, IRAs, and one business with no dollar limits. So for purposes of eligibility, even ignoring legal techniques used to hide or divest assets, neither income nor assets prevent most elderly Americans from qualifying for Medicaid LTC benefits.
Thus, the reality is not that most people spend down their wealth and finally become dependent on Medicaid. The reality is that most people are eligible with little or no spend down. Once on Medicaid, of course, they have to contribute their income to offset Medicaid’s cost for their care. That means that the LTC provider receives Medicaid’s dismally low reimbursement rate which, but Medicaid only has to pay its de minimus rate minus whatever private income, largely Social Security and SSI, that the recipient contributes. The result is downward pressure on quality and misleadingly low Medicaid expenditures. Recipients’ exempt assets are also subject to estate recovery, but loopholes in the federal law and most states’ failure to enforce estate recovery aggressively allow most exempt assets to pass to heirs instead of reimbursing Medicaid. You cannot understand the distribution of payment sources arrayed in this new data without taking these facts into account.
Quote: “A microsimulation model is used to describe the future care needs for Americans. This model can predict what percentage of individuals will develop a disability, have LTSS needs, use paid LTSS, and among those that use paid LTSS, how much they use and for how long. It estimates care costs, and how they would be financed under current policies. Microsimulation modeling provides not only the average likelihood of these outcomes, but also describes the distribution of these needs and costs.” (p. 2)
LTC Comment: All econometric models should be taken with a grain of salt. A key question: if you input data from 30 years ago, does this model accurately predict current conditions in the LTC service delivery and financing system? Unfortunately, we don’t have the necessary data from 30 years ago to answer this question. So the lesson is to challenge all assumptions and watch carefully and critically how the model’s predictions play out over time.
Quote: “As expected, given the aging population, the number with HIPAA-level disability is expected to grow from 6.3 million to almost 15.7 million.” (p. 3)
LTC Comment: Whatever else we can say about LTC services and financing, we’ll have 2.5 times as many people to care for over the next 50 years. Those aging boomers are marching relentlessly toward senescence and need. Absent a plague targeting old people they’re going to need a lot of long-term care. So it behooves us to get these projections right.
Quote: “The typical person who is alive at age 65 can [be] expected to live another 20.9 years. Fifty-two percent can anticipate having at least some needs for LTSS; 19 percent are expected to have needs that last less than a year, and about 14 percent are expected to have needs that extend beyond five years.” (p. 3)
LTC Comment: Instead of being able to say 70 percent of aged Americans will need some long-term care, we can now say that over half will need assistance with two or more activities of daily living and that one in seven will need such help for five years or more. That makes the risk more tangible and realistic, but still insurable. It remains a small risk of a catastrophic loss, which is the necessary and sufficient condition to make private insurance workable.
Quote: “While on average, individuals will need one year of paid LTSS, 48 percent of individuals will not use paid, formal LTSS at all (measured in service days, where one year is 365 days of paid LTSS). Among those who need paid LTSS services, about half will need less than a year, and a little more than 10 percent will need five years or more.” (p. 4)
LTC Comment: Likewise for paid LTC services, a one in ten risk of needing five years or more of paid care is eminently insurable.
Quote: “On average, individuals can expect to spend about $138,000 for LTSS (see Table 3A, or $70,000 in PDV as shown in Table A1). However, among those who ever use paid LTSS, the average cost will be about $266,000 (Table 3B or $134,000 in PDV as shown in Table A2).” (pps. 5-6)
LTC Comment: Big numbers but it’s more important to examine sub-categories and sub-populations as we’ll do below.
For now, consider that the phrase “individuals can expect to spend about $138,000 for LTSS” is a little misleading. The reality is that “various payers, including the individuals themselves, can expect to pay parts of the $138,000 expended on average per individual.”
What bothers me most here, however, is the idea as first stated in the “abstract” above that “$138,000 in future LTSS costs . . . could be financed by setting aside $70,000 today” or that $134,000 set aside today could cover $266,000 in future LTC costs.
What’s being employed to make this assertion is “present discounted value (PDV).” PDV is a legitimate actuarial concept intended to show how much money you would need to have now to be able to meet a future obligation based on certain assumptions regarding investment returns and inflation. For purposes of this paper, the authors computed PDV “using the Social Security Trustees' ultimate real interest rate of 2.9 percent. (Because the Trustees assume long-range price growth to average 2.7 percent, this amounts to a nominal discount rate of about 5.6 percent in the long-run.)” (Footnote 12, p. 12)
Now, here’s the problem with using present discounted value in this context.
Suggesting that people can set aside such small sums to meet the risk of catastrophic LTC costs adds another soporific to the already overwhelming factor anesthetizing the public to LTC risks and costs. To wit, the fact that government pays for most expensive long-term care after the care is needed, which enables the public’s denial by ameliorating the consequences of failing to plan or insure.
Quote: “Out-of-pocket costs average $72,000. Among those who have out-of-pocket costs, these costs average $140,000. About three-fifths of individuals face no out-of-pocket costs.14 Looking at community and institutional expenses together, two predominant payers are Medicaid, comprising 34 percent and out-of-pocket payments, comprising 52 percent of the sum of total LTSS expenditures, respectively. Medicare is the next most important payer, followed by private insurance and other public programs. Payer predominance varies by setting. For example, Medicaid pays for 51 percent of the total for institutional settings. For community expenses, in contrast, out-of-pocket payments by families comprise the majority, about 68 percent.15” (p. 6)
LTC Comment: To read this, you’d get the impression that out-of-pocket LTC expenses are very high compared to Medicaid especially for “community services,” which implies that people are spending down savings to pay for long-term care as was stated without evidence or explanation earlier in this report. The reality is more complicated.
Half of the out-of-pocket expenditures for nursing home care is really just spend-through of Social Security income of people already on Medicaid. This is important because it shows that a very significant portion of out-of-pocket expenditures does not come from asset spend down, but from another fiscally vulnerable federal entitlement program. Sure, it’s money people could otherwise put in their pockets, but think ahead a few years. What happens in 2035 when Social Security can only pay ¾ of what it has promised future beneficiaries? Someone will have to make up the difference. Medicaid? It’s already under water and the age wave bodes ill for tax-funded welfare programs. Medicare? It runs out of money sooner than Social Security (2030). Private payers? That would mean even more cost shifting, further punishing private payers for having behaved more responsibly than others by saving, investing or insuring to pay for their own long-term care.
Do families and individuals pay even more for community care out of pocket (68 percent)? Well, yeah, but that’s just money they would have to spend for room and board anyway. What’s important here is that public financing pays for 28.6 percent of community-based care (Table 3B), which means Medicare and Medicaid are paying for most of the care-cost component whereas individuals and families are paying mostly for room and board expenses they would have had to fund in any case.
Quote: “Expected LTSS costs are higher for women than for men. Women’s costs average $180,000 (Table 4B) compared to $90,000 for men (Table 4A). These could be financed by setting aside about $90,000 for women (Table A5) and about $47,000 for men (Table A3). However, when we focus on those with any LTSS expenditures, this average jumps to $320,000 for women and $194,000 for men (translating to $160,000 and 101,000, respectively, in present value terms as shown in Table A6 and Table A4).” (p. 6)
LTC Comment: OK, if you needed any more proof that long-term care is a “women’s issue,” there you have it. Women have a higher probability than men of needing long-term care; they need it longer on average; and if they need any at all, it’ll cost them nearly one-third of a million dollars.
But here we go again with the present-discount-value painkiller. $320,000 looks like a lot of money at first, but the real cost today is only half that ($160,000). So, not to worry. Analysis of long-term care risk and cost should raise consumers’ awareness and concern, not tamp it down unrealistically.
Quote: “The DYNASIM projections suggest that although Medicaid does reach individuals at all points in the income distribution at age 65, it primarily serves those in the bottom two quintiles. For example, about 36 percent of people in the bottom income quintile at age 65 use Medicaid LTSS, compared to just 5 percent in the top quintile at that age. Those in upper income quintiles who use Medicaid are typically individuals who have survived until their mid- to late 90s, consistent with other research (DeNardi et al., 2013).16” (p. 7)
LTC Comment: Well, hello! Why is it news that Medicaid, a means-tested public welfare program, covers more poor people than rich people? This report displays ideological bias by bending over backwards to minimize the fact that Medicaid LTC benefits accrue to middle class and affluent people as much or more than to the needy.
Let’s cut the numbers from Table 6A a little differently. Two out of five people (40.8 percent) in the top three income quintiles rely on Medicaid. What are the upper limits for all five income quintiles? According to the Census Bureau, as of 2013:
Fifth: $217,032 (This is actually the “lower limit of top 5 percent”)
Hmmm. This looks quite different. Nearly 41 percent of people receiving Medicaid LTC benefits have incomes between $50,521 and $217,032 or more. More than two-thirds (67.4 percent) have incomes between $28,895 and infinity. Not exactly destitute. How does this jibe with the slanted analysis offered in this report? It doesn’t. From now on, every time you read in a newspaper, magazine, or alas, a peer-reviewed academic journal that only “low-income” people qualify for Medicaid LTC benefits and only after they spend down their savings to impoverishment: Think bunk!
Not to put too fine a point on this paper’s bias, but keep an eye out for how its authors round up or down decimal numbers. For example, when they say “about 36 percent of people in the bottom income quintile at age 65 use Medicaid LTSS, compared to just 5 percent in the top quintile at that age,” they’ve bumped up the low-income-quintile number from 35.8 percent and bumped down the top-quintile number from 5.5 percent. That introduces a .7 percent misimpression. Why not just use the actual numbers with the decimals intact? Why indeed? If you like to play “Where’s Waldo,” you’ll love reading this report sleuthing for rounding bias, or searching for typos. Good hunting.
Quote: “Family out-of-pocket expenditures, in contrast, are more concentrated in the higher quintiles. The average out-of-pocket LTSS expense in the top quintile is approximately $97,000 compared to closer to $45,000 in the bottom quintile. But again the mean obscures important distributional information. About 12 percent of people in the top income quintile at age 65 can expect out-of-pocket expenses in excess of a quarter million dollars.” (p. 8)
LTC Comment: The richest people pay only twice as much ($97,000) for LTC as the poorest people ($45,000)? Gee, I wonder if that could have something to do with what we explained immediately above.
It’s not surprising that 11.7 percent of top-income-quintile people have out-of-pocket expenses in excess of $250,000. But Table 6B also says that 5.2 percent of people in the lowest income quintile can expect out-of-pocket expenses to exceed $250,000. Maybe those lowest-income people aren’t quite as broke as we thought they were.
Quote: “Medicaid is an important payer for LTSS, but because it serves only those who meet income and asset criteria, many families pay for LTSS out-of-pocket. Private LTSS insurance has only a modest reach, and it predominantly covers costs for those high in the income distribution. Similarly, other public expenditures (for example, including Veterans Administration care) only help to cover small shares of the population with long-term care needs. The results presented here highlight the need for better planning for LTSS to accommodate both average and catastrophic financial risks associated with chronic disability.” (p. 8)
LTC Comment: Well, true, that’s what these results show. What they do not show without the explanation and clarification offered here is that Medicaid is a major payer for expensive long-term care for all income and asset levels and that as such it has for 50 years crowded out private-payers, impeded the private insurance and reverse mortgage markets as potential long-term care funders, and distorted the service delivery system in favor of the kind of welfare-financed nursing home care that most Americans prefer to avoid.
Bottom line, however, properly interpreted this data on long-term care incidence, duration, cost and financing sources is better than we have ever had before. Use it, but don’t abuse it to suit any political or ideological bias. If you let the facts speak for themselves they’ll shout:
“Give Medicaid back to the poor and everyone else will save, invest or insure for long-term care.”
Do it before it’s too late!