The2030 problem

The 2030 Problem: Caring for AgingBaby BoomersJames R. Knickman and Emily K. Snell

Objective. To assess the coming challenges of caring for laiê numbers of frail elderlyas the Baby B(M>m generation ages.Study Setting. A review of economic and demographic data as well as sitntUatiori.s ofprojected socioeconomic and demographic patterns in the year 2030 form the basis of areview of the challenges related to caring for seniors that need to he faced by society.Study Design. \ series of analyses are used to consider the challenges relaled tocaring for eUU-r-s in the year 2030: (I) meiisures of maciocconomic btirdcn are devel-oped and analj'zed. (2) the literatures on uencis in disability, paîiient approaches forlong-term care, healtliy aging, and cultural views of aging are analyzed and syjithesized,and (3) simulations of Riture int ome and assets patterns of tlie Baby Boom generationare df\elo|M'd.Principal Findings. Thf economic burden of aging in 2030 should Ix- no jfieatertJuin tlic economic burden iissociated wiili raising laige numl>ers of baby boom diildrenin the 1960s. The real challenges of caring for tlie elderly in 2030 will involve: (1) makingsure sCKiety develops payment and insurance systems for long-term care that work In-tierthan exi.sting ones. (2) taking advantage of advances in medicine and liehavionil healdito keep tlie elderly as healthy and active as possible, (3) changing the way societyoi-ganizes community services so that care is more accessible, and (4) altering the cul-tural view of aging to make sure all ages aie integrated into the fabric of community life.Conclusions. To meet tJie long-tenn care needs ol Baby ticKiniers. social and publicpolity changes must begin soon. Meeting the financial and sociiil sei-vice burdens ofgrowing luun bet's of elders will not be a daunting task if necessary changes are made nowrather than when Elaby B<Kimers actually need long-term care.Key Words. Long-term care, financing. Baby Boomei-s, community-based deliverysystem

A major pitblic polio' concem in the long-term care field is the potentialburden an aging society will place on tlie caie-giving system and ptthiicfinances. The "2030 problem" involves the challenge of assuring that sttfFicientresotirces and an effective service system are available in thirty years, when theelderly population is twice what it is today. Mttch of this growtli will be

850 HSR- Health Services Research 37:4 (August 2002)

prompted by the aging of the Baby Boomers, who in 2030 will be aged 66 to84—the "young old"—and will number 61 million people, hi addition to theBaby Boomers, those bom prior to 194ti—the "oldest old"—^will nttmber9 tTiillion people in 2030.

This paper assesses the economic dimensions of the 2030 problem. Thefirst half of the paper reviews the literature and logic that suggest that aging ingeneral, and long-term care senices in particular, will represent an overwhehii-ing economic burden on society by 2030. Then, a new analysis of burden ispresented lo sures t that aggregate resources should not be a major issue forthe midcentttry economy. Finally, the paper presents foiu" key challenges tliatrepresent the real ecotiotnic burden of long-term care in tlie twenty-fii"stcentury. These challenges are significant but different from macro cost issues.

WTiat type of economic btirdcn might be considered ovei"whelming?Existing literattire never explicitly dehnes tliis bui the sense is tliat the burdenmight be considered overwhelming if: (a) tax rates need to be raiseddramatically, (b) economic growth is retarded due to high senice costs tliatpreclude other social investments, or (c) the general well-being of futtiregenerations of workers is worse tlian that of current workers due to semce costsand income transfers.

Tlie discussion has significant implications for public policy and forprivate actors foctised on developing an effective care sj'stem for the mid-twenty-first centttry. Public policy goals related to an aging society must balance theneed to provide adeqtiate services and transfers with an interest in maintainingthe economic and social well-being of tlie nonelderly. The economicchallenges disctissed are such that ptiblic and private progiess Uiat begins inthe near future will make the future burden substitntially easier to handle.

DEFINITIONS AND BACKGROUND

Various aspects of economic burden are associated with an aging population:social security payments will increase, medical care insurance costs will giow.

Address correspondence to James R. Knickmaii, Ph.D., Vice President for Research and Evaltia-lion,The Robert WtKid Johnson Fntiiidation, Box 2316. Princeton. NJ 08543. Emily Snell, B.A., isa Senior Researth ;\s.sistant at The Roben Wood Johnson Foundation. The paper was writtenwhile Dr. Knickman was Regents' I.eclurtT al the L'liiversity of Ciilifoniia (IIC). Berkeley. Thesupport and inpiil ot colleagues botli al UC, Berkeley and The Roben Wood Johnson Foundationare acknowledged. The opinions and conclusions are the authors aiid aic not meant to reflectiliose of the sponsoring insmutions.

Caring for Aging Baby Boomers 851

the burden associated with uncovered medical expenses such as phat-maceu-ticals will become quite serious, atid long-tenn care costs will grow. Mitch of tJielogic of the paper applies to each of these financial tesource challenges.However, we focas principally on the implications of lotig-temi care services,which along with prescription drugs, have had tlie fastest growing co.sLs in iliclist cited.

Every elder has to ptepare for four key "aging shocks": tmcoveted costsof prescription dmgs, tlie costs of tnedical care tliat are not paid by Medicare orprivate insurance, the actual costs of private instirance that partially fills in thegaps left by Medicare, and the tmcoveted costs of long-tetm care.

If the lifetime costs of each of these "aging sliocks" are calcttlated, tlielong-term care burden Is the worst by far. The typical 65year-old faces presetitvahte lifetime costs for uticovered long-term care of $44,(KX). By cotitrast, thepresent value of lifetitne oitt-of-pockct pre.scriptioti dnigs cosLs averagesSrj.OOO, uncovered medical care cotnes to $16,000, atid uticovered privateinsurance premiums come to $18,000* (Table 1). It sliotild be noted tbatbecause of the United States' apptoacb to ftnancing tJiese senices, agitigsbocks represent burdens borne by individuals mote than society. In mostother countries, tbese items tend to be financed socially.

Lotig-tetm care includes a broad contiiututn of services that address theneeds of people who are frail or disabled and reqtiite help witli tbe basicactivities of evervday livitig. The senices can v"ary ftotn itifonnal care deliveredby family and friends to the formal services of bome care, assisted living, ornursitig homes (see Table 2).

Long-term cate professionals generally distingtiish two t)'}jes of stipport-ive care needs for the frail: assistance witb instruviental activities of daily li\ing(IADLs), sttch as shopping or cleaning, and assistance witb /j/ivvicn/activities ofdaily livitig (ADLs), such as eating, bathing, or moving aroutid.^ Amotig the31 uiillion uonitistiuitionalized elderly, 1.8 million have IADLs and 3.3 ttiillionhave ADLs (Table 3). Of tbe 3.3 million, 1.5 tnillion need help with tliree ormore ADLs, indicating a very bigb level of need that requires extensive hotne

Table 1: Expected Lifetime Costs of Significant "Aging Shocks" for a 65-Year-Old Today• Uiicovt'ied Pifscriptiiin Drugs• Uncovered Medical tiare• Uncovered hisurancc Premiums• Uncovered Lonp-irmi Care

$12,000$16,0(H)S! 8.000$44,000

Estimates latt-uluted //y aiil/wrs. See footniitf 1 for iLssumptiom used.

852 HSR: Health Services Research 37:4 (August 2002)

Table 2: Cont inuum of Long-term Care Services

INCRE,ASING SEVERITV Informal Care by Family and FriendsVolunteer ServicesIn-home Supports (e.g., meals)Personal CareAdult Day CareAssisted LivingHome Healtli CareNursing Home Care

Hospice Care-

Table 3: Population Needing Long-term CareThere are 31.2 million community-dwelling elderly

with only IADLs* 1.8 millionwith 1-2 .ADLs** 1.8 millionwith . + ADU** 1.5 million

Total w/LTC needs 5.1 million in the community

There are 1.4 million elderly residing in nursing homes

Sourri': (kwgplmun University histitulefor Health Care Research and Policy preliminary analysis of datafrom Ihe. National Health Interview Survey and the National Health Interview Survey on Disability,Pha.se U. 1994-5'•'lADLs include problems with light housework, laundry, meal preparation, transportation,grocei")' shopping, telephoning, and medical and money management.**ADLs include problems willi eating, bathing, dressing, toileting, transfers, and condnence.care or institutional care. Another 1.4 million elderly are cared for in nursingboines ;md most of these elderly have 3 or more ADLs (Georgetown UtiiversityInstitute for Healtli Care Research and Policy 1994-5).

These statistics measure the number of elderly vifho are disabled at anygiven point in time. Most elderly who live beyond 75 or S.") years of age becomefrail at some point and need some fotm of assistance, making lifetime risksmuch higher. In feet, 42 percent of people who live to the age of seventy willspend time in a nureing home before they die (Murtaitgh et al. 1997).

The current "long-tenn care system" is built around private providers ofservices—some nonprofit and some for-piofit. When resources expand, newservices develop quickly, and when resources contract, capacity can also shrinkquickly. In home health care, for example, annual expenditure giowth rateswent ftom more than 10 percent in the 1980s and early 1990s to minus 3percent between 1998 and 1999 (Levit et al. 2000). Recent policy revisionsincluding the Home Health Prospective Payment System and tbe BalancedBttdget Refinement Act, along with projected strong growth in out-of-pocket

Caring for Aging liahy Boomers 853

spending by indi\idtials and families, catise many analysts to believe tliat homebealth care .spenditig will rise again in the coming decade (Heffleretal. 2001).Of course, expansion and contraction of luirsing bome beds respotid moteslowly to market forces because of the dtirable capital aspect of nursing bomecare.

The Congressional Budget Office (CBO) estimates that expenditures onlong-tenn care totaled more than $120 billion in 2(K)0, with 59 percent of allexpenses covered by the public sector (Congressional Budget Office 1999).Oul-f)f-pocket expenses account for almost all of tbe balance, with privateiustu-cince covenng jiust 1 i>ercent of long-temi caie costs (See Figtue 1).Conservative CBO estimates suggest total long-term care expendittires willIncrease at a rate of 2.6 [X'rcent per year above inflation over tlie next thirtyyears, to $ 154 billion in 2010, $ 195 billion iti 2020, and a staggering $270 billionin 203(). The numbers are slightly different if one assumes that lotig-tcmi careinsurance does not become more common, but the stark upward trendremains.

The $120 billion in cun-ent expeases underestimates the economicresources devoted to long-tenn care, however, because most care is deliveredinformally by family and friends and is not included in economic st;itistics.Among the elderly who reqtiire assistance with daily activities, 65 percent relyexclusively on families and friends and anotlier 30 percent rely, at least in part,

Figure 1: Projections of National Long-term Care Expenditures for the Elderly

375 B Private Long'Term CareInsuranca

y Out of

Pockst

• Medicaid

I Medicare

I Total

2000 2020 2030* 2040*Source: "Projections of expenditures for long-term care sen/ices for the elderfy." CBO 1999Dollar totals in 2030 and 2040 are not disaggregated by payer due to uncertainty about projections.


on informal care. It has been estimated that tbe economic value of suchinformal care-giving in the LInited States reaches $200 billion a year—one anda half times the amount spent on formal care giving (Amo, Levine, andMemmott 1999).

Ecanomic Burden of Long-term Care: The Dire Case

The at^tment that caring for an aging society cottld disable tbe Atnericaneconomy has beeti tnade by various commentators, perhaps most forcefully byPeter Peterson and otbers in the Concord Coalition (Peterson 1996). Tbeirconcern focuses on the Iai"ge growth in the ntitnl>er of elderly over the comingyears, from 3.5 million in 2000 to more than 80 million in 2050 (Figitrc 2). Tliisgiowtb in elderly could lead to a precipitous drop in the number of workers perelderly if current working and retirement patterns do not change (Figure 3).

Tliese trends cottld be partictilarly troublesome for the long-term caresystem because the lai gcst gtowtb in the over-65 popitlatioti will be among the"oldest-old," who are disabled at tbe highest rates. Tbus, the ratio of workers tofi-ail elderly could decrease even more dramatically tban the ratio of workers toall elderly.

A second part of tbe argument that long-term care could be a largeburden focuses on the rapid inflation in expetiditures for long- term care inrecent years. Medicare and Medicaid expendimres on mtreing home care were$9 billion in 1980, tnore than doubling to $25 billion by 1990, atid doubling

Eigure 2: Population of Americans Aged 65 and over, in Millions

90e 80oS 70§ 60.2 50ao

lati

sao

403020100

0 Aged 85+• Aged 75-84~ Aged 65-74

1950 1970 2000 2015 2030 2050

Source: (NP-T4) Projections of the Total Resident Population by 5 Year Age Groups, Race,and Hispanic Origin wilh Special Age Categories: Middle Series, 1999 to 2100

Caring for Aging Baby Buottiers 855

Figure .S: Will the Old Overwhelm Us?

Ratio of all 20-64year olds to elderly

1950 1960 1970 1999 2015 2030 2050

Source: Population estimates from U.S. Census Bureau

again to $54 billion by 1999. Mkewise, Medicare and Mrdicaid expcndimics onhome health care increased from less than $1 billion in 1980 lo $5 billion in1990 and to $16.1 billion in 1999, down from a high of $17 bUlkm in 1996(Health Gire Financing Administration 2000; Hefller et al. 2001). Theseincreases bave not been offset by lower out-ot-pocket payments, altbotighgiowlii rates for tbe latter bave been lower and more consistent, with averageincreases of 7 percent for home health care and 3.3 percent for nursing homecare each year over the last decade (U'vit et al. 2000). Finally, manyprofessionals believe that even at cuiTent expenditure levels, tliere is asignificant amount of unmet need for long-term care among the frail, and noforeseeable end in upward pressure on per diem service costs (Allen and Mor1998).

A final part of ihe argument for concern about long-term care costs comesfrom unsettling survey results reported by Curran, McLanahan, and Knab (inreview) .suggesting that children who experience divorce may be less willing orable to care for tbeir aging parents. Their data indicate tliat die probability of anelderly person perceiving an availability of emotional support from bis or herchildren is reduced from 71 percent for those who marry once and remain


married to 56 percent for those wbo marry and divorce. Remanying afterdivorce provides some additional support, but does not completely offset tlieresult of the first divorce. Witli the legacies of the divorce boom getting older,the percent of the elderly who are divorced or separated is projected to doublefor men and triple for women between now and 2030. It may be the case thatinfonnal care resources will shrink and thus restilt in even more pressure onpublic and private tesources tliat support tlie formal care system.

A Reassessment of Future Economic Burden

It is possible to eonstnict a counter-case that is more optimistic about the macroburden of iong-tenn care in the twenty-firet century. Most importantly, ihetypical analysis of dependency ratios overstates tbe impacts of aging on burden.While standard calculations include only tlie 20- to 64-year-old population a.sproducers, and places al! of tlie 65-and-oIder population in the dependencygroup (because of eligibility for Social Security and Medicare), this approach isnot appropriate in tbe case of long-term care.

At the very least, tbe denominator sbottld incltide only people 75 andolder since tbe 65- to 74r-year-oid age group does not tise large aiiiotmts of long-term care. The percentage of elderly older than 85 years wbo are ADLimpaired or instittidonalized is more than six times the rate of 65- to 74-year-olds (Manton, Corder. and Stallard 1997). In fact, if the Baby Boom generationis healthier than past generations (as argited later in the paper), it very wellcould be that the young elderly migbt work longer and tbus be consideredproducers. In addition, in considering macroeconomic burden, the otliergroup of dependents in society—cliildren—sbotild be included in thedenominator with the elderly, as both groups are dependent on the adultpopulation.

Recalculating dependencv- ratios using these new principles (see Table 4)indicates improvements rather than deteriorations in the adult to dependencyratios, especially when the year 2030 is compared to 1960 wben children werepresent in tlie poptilation in peak numbers. When 20- to 64-year-olds arecompared to children and individtials 75 or older, the dependency ratioactually improves between 1960 and 2030. If the 65- to 74-year-old cohort isconsidered "productive" and enters the numerator, the dependency ratiobecomes stibstantially more favorable in 2030 than it was in 1960.

All of these ratios change in unfavorable directions between 2000 and2030, but the changes are not substantial.^ Tbe key point is that the UnitedStates prospered through the 1960s witb dependency ratios less favorable tban


Table 4: Calculating Dependency Ratios

Ratios 1960 2030 % Change

20-64/65+ 5.66 2.67 -53%(i.e.. The number of 20-64 year olds for every 75+ year old)20-64/0-19 & 65+ 1.10 1.15 +4.5%20-64/0-19 & 75+ 1,26 1.50 +19%20-74/0-19 & 75+ 1.40 1.80 +29%Source: Population estimates from U.S. Cetisvs Burmu

will be experienced in 2030 and liie burden did not overwhelm die economy.Richard Leone, among others, has argued this point (Friedland and Summer191*9; Leone U)97; Singer and Manton 1998). Many odier countries are yeareahead of the United States in population aging; some will reach the "2030burden level" as early as 2010 (Congiessional Budget Office 1994). The UnitedStates will have more time to prepare than most industrialized nadons and willbe able to learn from these nations" experiences.

A second factor that might make Uie burden of long-tenn care lessstriking than expected in 2030 is improvement in the health status of theelderly. Recent data from the National Long-Term Care Survey reported byManton and Gu (2001) indicates that the disability rate for all elderlydropped hom 26.2 percent in 1982 to 19.7 percent in 1999. This meant thatdiere were more than 100,000 fewer disabled elderly in 1999 than in 1982,despite a one-third increase in the population of the elderly. In prior work.Singer and Manton (1998) estimated thai a relative rate of disability declineof 1.5 percent a year over the next few decades would maintain tiie currentlevel of burden each disabled elderly places on economically activeAmericans. This decline is acaially significantly less dian the 2.6 percentrelative rate of decline experienced between 1994 and 1999. Declines in thenursing home population, particularly among the oldest old, have accom-panied these disability trends (Bishop 1999). A more mixed, althoughguardedly optimistic, picture of disability Uends bas been offered by otherdemogi-aphei-s (Crimmins 1997; Reynolds et al. 1998; Schoeni, Freedman,and Wallace 2001). Schoeni, Freedman, and Wallace's analysis of NHISdisability' data reports a 1.1 percent average annual decline in disabilityIx'tween 1982 and 1996. However, Uiey caution tliat this decline was notpersistent or consistent through this pteriod, with most of the declineoccurring in the 1980s.

/Uthough trends will need to be watched closely over die coming decades,declines in disability probably will continue.' This is becau.se the two most

858 HSK- Health Services Research 37:4 (August 2002)

salient factors influencing the recent declines in disability—^better-educatedelderly and better science—will continue to exist in the coming decades(Freedman and Martin 1999; 2000).

The elderly of 2030 will be much better educated, witli a collegegraduation rate twice (and high school drop out rate one-third) that of thecuiTent generation of elderiy (U.S. Depailment of Education 1998). Tliisbodes well for the future phracal healdi of aging Baby Boomei-s, as there is astrong conelation between education level and disability; college graduateshave a disabilit)- rate about half that of high school dropouts.

Expected advances in medicine, through prevention, phamiaceuticals,and surgical tieatinents, should also reduce the need for long-term care. Moreand more older patients are benefiting from restoradve treatments, such asknee leplaceiiienLs and coronaiy angioplasty, once unavailable due to age(LubitJi; et al. 2001). Although such advances in medicine tend to increase acutehealth care costs, these treatments can delay entrance into nursing homes andother long-term care needs. Better pharmaceutlcals for preventing and treatingdisabling conditions such as osteoporosis, arthritis, and rheumatism willcontinue to decrease the number of elderly who need assistance with L\DL.sand ADLs (Neer et al. 2001). Because about 60 to 70 percent of nursing homeresidents have dementia-related symptoms (Ro\Tierand Kat7 1993). progress intreating or preventing AJzIieiraer's and other dementias associated with laterlife could lead to large reductions in the number of elderly with the mostintensive long-tenii care needs. This is already happening: Freedman, Aykan,and Martin (2001) found that the proportion of noninstitutionalizedelderly—paiticularly the very elderly—widi severe coguitive impaiimentbetween 1993 and 1998 fell from 6.1 percent to 3.8 percent.

If as a resuh of further medical advances and social shifts, the di.sahilit)'rates continue to decline in the coming years at the same pace tliat Mantonreported for 1994 to 1999, or an average 0.56 percent a year, the number ofdisabled elderly in the year 2030 would be a remarkably low 1.6 million people,or less than three percent of the elderly population. This seems extremelyunlikely. A more conservative estimate for declines in disability rates would bean average annual decline of 0.13 percent between 1994 and 2030. Althoughthis decline would be half the 0.26 percent average annual rate of declineexperienced between 1982 and 1989, about one-third of tlie 0.38 percentaverage annual decline experienced between 1989 and 1994, and significantlyless than the 0.56 percent average annual decline experienced between 1994and 1999, it would still result in a manageable 11 million disaliled elderly in2030, 40 percent less than die estimate in 1982 of 18 million disabled elderly.

Caring for Aging Baby Boomeis 859

Even a moderate decline in disability would have dramatic impacts on theeconomic bttrden of long-term care.

Tlie optimistic forecasts could be affected by demographic changes thatare difficult to forecast. Most relevant perhaps are trends in immigration. It ispossible ihai unexpected giowih in ininiigration cotild increase tlie numberof elderly in the year 2030, making btudens woi"se. Or, iminigratioti uendsmight also bring laiêr than expected numbers of working-age adults toAmerica, thus decreasing dependency ratios. In addition. Wolf (2001) makesthe case that decteases in disahility rates that are due to higher educationalattiiinments atiiong the Baby Boom generation will not continue past the year2050.

The final and most itnponant factor that will affect the macro btirden ofagitig is the ftitttre of tlic American economy. When the pes.simistic literatuteoti the but den of an aging society was published, the economy was growitig atanemic rates and inflation rates were relatively high. If the economy grows at asteady, healthy pace, the overall btirden of long-tenn care will be moderate.Social secttrity acttiaries issue three diffctent econotnic sccuari<js for thecoming century. In their "bad" economic growth scenario, with avei-iê real(iross Domestic Prodtict (GDP) growth a little tnorc than 1 percctit over eachof the tiext 30 years, long-temi cat e expenses for Uie elderly wottld l>e almost 40percent higher as a percent;ige of GDP than in their "great" economicscenario, with real GDP growth averaging 2.1 percent. During the 1990s, adecade when yearly change in real GDP shifted ftoiti negiUivc numbfi"s toabove 4 percent, GDP actually averaged alôtit 3 percent real aiiritial growth peryear. If GDP growth were to continue at 3 percent over the next three decades,a rate which few I>elieve is possible, LTC expenditures as a percentage of (iDPwould acuially decline between 20(M) and 2030. WTiether I'eal GDP growthaverages 1, 2, or 3 percent will make an enomiotLS difference over a thiity-yeartimeframe.

Taketi togetlier, these three factors—newly calculated dei>endencyratios, declines in disability rates, and healthy economic giowih—could worktogether to make the macro burden of long-term care no worse than it is thebeginning of the twenty-first centtiry. While the future remains tincertaiti,tliere seems to be little reason for dire concenis about tbe ftitttre btitden oflong-tenn care. Our estimates arc tiieant to be broad outlines i-atht-r thanspecific forecasting. Others have eloqtiently discussed the power of andcaveats to making demographic and economic projections (Lee and Skinner1996).''

860 HSR- Health Services Research 37:4 (August 2002)

THE REAL ECONOMIC CHALLENGES EORLONG-TERM CARE IN 2030Despite the preceding positive analysis of the macroeconomics of aging, thereremain some stibstantial challenges lo getting ready to meet the long-term careneeds of Baby Boomers. In fact, fotir types of challenges need to be addressed:

1. Creating a finance .s)'stem for long-term care that works2. Building a viable and affordable community-based delivery system3. Investing in healthy aging in order to achieve lower disability rates, and4. Rechai^ng the concept of family and the value of seniors in American

culture.

Each of these challenges is considered in the remaining sections of this paper.

A Workable Long-term Care Financing System

Current Sources of Financing. Four sources of payments currently financelong-term care services for tbe elderl)': Medicare, Medicaid, private insurance,and out-of-pocket payments (see Figure 4).

The federal Medicare program pays for approximately 24 percent of alllong-term care costs (Congressional Budget Office 1999). Medicare's long-termcoverage, however, focuses mosll)' on home care thai is related to medical

Figure 4: Expenditures on Long-term Care for the Elderly

PrivateInsurance

4% Medicare24%

Medicaid36%

Total Expenditures:$123 Billion

Source: "Projections of expenditures for long-term cace services for the etderly," CBO. 1999


problems, such as broken hips. Services generally are restricted lo peoplereceiving rehabilitation for some medical condition. In principle. Medicaredoes not cover custodial long-term care, bui in practice it is an ongoingcliallenge for Medicare and providei"s to distingtiish ctistodial caic andrehabilitative care.

Tlie federal/state Medicaid program is perhaps the most importantplayer in the long-tetin care financitig system. Medicaid acts as a backstop,paying for long-tenn care services for the frailest elders when they are poor. Inmost states, the Medicaid progi-am pays for care for the poor and for elders whobecome poor when long-term care expenses impoverish them. In 1995, 64percent of elderly nursing home residents used Medicaid to finance at leastsome of their care (Dey 1997). In many states, a laiê share of all Medicaid long-term care dollars supports frail elders who bad been middle class beforebecoming frail. In New York State, for example, the Medicaid program pays for80 percent of all tiursing home costs; clearly, 80 petcent of New York elders arenot poor before they become frail.

Most states diiect the lion's share of Medicaid dollars to ntifsing homes asopposed to botiie care, ln 1995, almost 85 percent of elderl) Medicaid long-term care expenditures were for institutional care and only 10 percent were forhome care semces (Wiener and Stevenson 1998). This is a restilt of Medicaid'snttempt.s to foctis on tbe most frail, who tend to be in ntirsing bomes. Manyexperts feel tbat tliis emphasis on nursitig liotnes means that not enoughresources are devoted to preventing eldere wbo have some disabilities fromIncoming more and mote frail (Kane. Kane, iind l.add 1998). Since ciders tendto avoid nui"siiig homes as long ius possible, tlie Medicaid empbitsis on nureingbomes means that many elders go without commtinity-based services ihat reallycould belp them live better lives.

Private insurance accotmts for jtist 4 percent of long-term care costs.Despite aggressive attenipis by tbe instirance industry to develop a privatemarket for long-term care, tbe growtli of this market has proceeded slowly. Pailof the reason is tlie nature of the contract between an insuterand an elder. Tlieinsurer needs to guarantee a senice tliat often will occur twenty or moie yearsiif ter the contract is set Tbe uncertainty leads insurers to keep prices high andmakes elders nervous about pttrchasing a private insurance policy. In addition,tbe "door-to-door" sales apptoacb by individual agents adds to the costs oflong-tenn care instu"ance. And, tlie relticuince of people to think aboutpurcbasing such insurance at youn^r ages makes the payments on aniusumnce policy beyond tlie reach of many elders. Finally, tbe av-ailabilily ofMedicaid as a substitute for private instirance leads many elders to forego

862 HSR: HeaUh Services Research 37:4 (August 2002)

insurance premiums and take their chances on remaining healthy (McCalletal. 1998).

Out-of-pocket costs finance about 36 percent of long-term care, but tlieburden of these payments is very unevenly distributed. The 42 percent of elderswho spend some time in a ntu'sing home—one-half of them for two years ormore—pay most out-of-pocket costs for long-tenii care.

Barriers to Refonning tJie Financing Systêm. Why is it so hard to devise afinancing system to replace the current patchwork payment approach? Onesubtle but CRicial factor is what Washington officials call the "woodwork"effect, which is the fear that many elders will demand long-term care services if amore hospitable financing system helps them receive care without having touse personal savings. It is very diffictilt to judge who really needs formal long-term care services, and there may be large amounts of pent up demandcurrendy taken care of by families and friends. Tlie woodwork effect predictsthat total expenditures cotild grow substantially if public and or ptivateinsurance expanded (Weissett, Cready, and Pawelak 1988; Kane, Kane, andLadd 1998). Imbedded within the woodwork issue is the real social challenge ofdetermining how best to allocate limited resources.

A second banier to reftHTn is Americans' aversion to taxes and .saving. Anyexpanded public insuiance system would reqtiire new taxes. And privateinsurance would be paid for from private savings, which are in short supply formost middle-income elders. One interpretation of the indifference of working-age Americans to either save privately or approve taxes to cover future long-term care services is that Americans are not adequately aware of theimplications of these nonactions. Perhaps Americans, more so than othersocieties, are less mincUul of the needs of aging due to a relatively age-sU'atifiedsociety. The other interpretation, however, is that Americans do not value long-tenn care services. This interpretation is lx)lstered by the fact that manymoderate-income elderly who could benefit from long-tenn care and couldafford lo pay for some services choose to make do on their own.

In addition, financing reform has had to compete with various othersocial priorities. In recent years, lawmakers have directed more attentiontoward imcovered pharmaceutical costs. Medical care costs in general are highfor elders—even with Medicare, elders and their families pay more than a thirdof their health care costs out of their own pockets. The laiê number ofuninsured among the noneiderly population continues to be a problem thatdeinaiuls attention. Tlie needs of eldei"s also compete for resources witli otherproblems facing other age groups. Many have concluded tliat elders have donerather well with social policy compared with other needy subgroups of our

Caring for Afpng Ha/ty Boomers 863

population. Child poverty, for example, is higher today than it was threedecades ago, while poverty among the elderly has decreased significantly.Finally, tiix ctiLs and a .sluggish economy could completely eliminate the fundsneeded for any new social programs.

The re.sult of the patchwork financing system is the "aging shock"presented iti Table 1: Uncoveted long-temi care averages $44,(XH) in lifctiniccosts for the typical elder. Tlie $44,000 figure also tmderestimates the burdenof long-term care in tliat it represents the expected value for a typical elder. Anyspecific person who fx-comes frail may face higher burdens because theexpected value averages high costs for those who become frail atid zero costs foriliose elders who avoid the need for long-term care services. Conservativeestimates suggest that an elder should be prepared for more tlian $150,000 incosts to cover tlie small l)ut real probability that he or she will spend two oithree years in a ntirsing home. And, home care costs can be jtLst as high for afniil elder wanting to live at home.

Long-term Care Financial \âhility. In oiher re.search, Knickman. Snell. andHunt (2000) estimate which eldei s can ;tflbrd a $I. >O,()(H) long-temi care"shock" and how long-term care financial viability will change over the comingyears as the Baby Booniei's readi old age.' The best way to estimate tfie elderly'sfinancial viability is to consider diree subsets of people (Eigure 5): MedicaidBound, EinanciiUIy Independent, and Tweeners.'

Figure 5; Continuum of Abiiitv' to Absorb Long-term Care Costs andDistribution of Population in 2000 and 2030

2000 2030Medicaid BoundVery limited means to pay 45.2 % 28.9 %forLTC

TweenersLifetime income atid wealth is 27.7 % 32.9 %adequate but cannot handle a LTCshockFinancially Independent '^^tcr IQ^C^-r, *, r * 27.1% 3o.2voThree years of long-term care costsare adequately handled with incomeand wealthSource: Authors'projections using Lewin-VHILong-Term Care Financing Model.

864 HSR Health Seruices Research 37:4 (August 2002)

The Medicaid Bound: Tbese are individuals who have few financial resourcesavailable for long-term care and who have no choice but to rely on Medicaid. Aperson is Medicaid Bound if he or she does not bave at least $50,000 in liquidassets or citrrent income available for long-term care in the year 2000, or$70,000 in the year 2030.**The Rnancially Independent: These are individuals who have $ 150,000 or morein liquid assets or cuiTent income available for long-term care and wbo can takecare of tliemselves financially widi or without private insurance, and surelywithout Medicaid. In 2030, $210,000 is the minimum amotint nece.ssary forindependence.The Tweeners: TTiis is a group whose lifetime income and wealth are adequatebut who cannot handle a long-term care shock. Individttals with liquidity be-tween $50,000 and $150,000 and $70,000 atid $210,000 comprise the Tweenersin 2(K)0 and 2030, respectively.

Estimates calculated ttsing the Lewin Long-Term Care Financing Modelindicate some good news about the fiiture financial viability of elders.^ (See tbefootnote for ati explanatioti of model.) Itt tbe year 2000, an estimated 45percent of eldet s are classified as Medicaid Boutid but this estimatedpercentage drops to 29 percent in the year 2030. By contrast, the share oftbe elderly who are financially independent increases fk>m 27 percent in 2000to 38 percent in 2030.

As with all simttlations and forecasts, tlie estimates into the future dependon key economic assumptions. The assumptions used follow the principles ofthe middle estimates of the social security forecasting model: economic growthaveraging 1.3 to 2.0 percent per year and real wage growtli averaging 0.9percent per year. Perhaps the most "bullisb" assumption in the sitnulation isthat the costs of long-term care will increase just 1 percent per year in realterms. To meet this forecast, long-term care cost itiflation would need to bebrought under control. If long-term care costs iticrease at 2.3 percetit aboveinflation, tbe distribution of the population not able, barely able, and able topay for long-term care wottid retnain almost identical to the distribtition today.Thits, a relatively modest increase in long-term care inflation fates couldeliminate the rosy simulation estimates of changes in the Medicaid Bound asreported above.^"

The other interesting forecast that emerges from the simulation exerciseis that tlie percentage of people iti tlie Tweeners categoty will not shrink, butwill actually increase fi'om 28 percent to 33 percent. This implies tbat there willcontinue to be a latê number of middle class elderly who will spend down to

Caring for A^ng Baby Boomers 865

Medicaid coverage tmless new financing arrangements help make theTweenei^s more self-reliant.

How do assumptions about future disability rates affect tlie simulationresults? In fact, disahility rates only affect the estimates indirectly in that higherdisahility rates lead to lower income and iisset estimates—particularly for thenonelderly—and tiiis increases the number of Medicaid Bound. Disability ratesdo not directly affect the simulation estimates becatise the calculations a.ssesswho is able to afford long-term care at a point in time whether a person isdisabled or not.

Emerging Options to Improve the Financing of Long-term Care

Despite the size of the economic shock associated with long-term care needs,very little polic)' attention is being given to designing new approaches to pay forlong-term care. The Kaiser P'amily Foundation issued a side-by-«ide comparisonof key health policy positions advanced by the Gore campaign and the BtishcatTipaign in early October 2()00. Under the category "long-term care" thepolicybrief reported that the Core campaign had no published proposals whiletlie Bush cainpaign supported tax dedtictibilit) of private long-tt'tm careinsurance. The quiet long-teim care issue for the elderly was completelyeclipsed by attention to prescription drug coverage and the fiiture of socialsecurity citmng tlie presidential campaign.

When attention does focus on financing options for long-term care, threeserious types of options need to be considered: tax deductions for privateinsurance, public pro\'ision of long-tenn care insurance, and mandatoiy savingsstarting at younger ages foi" private insurance.

Tax deductibility for private insurance premiums clearly would expandtbe number of people who ptirchase long-term care iasurance. Dedtictibilitywould lowei- the after tax cost of insurance by 15 percent to 40 percent (therange of the current mai^ginal tax rates). Unfortunately, tlie largest after-taxprice breaks would go to the most wealthy people who do not need insurancebecause they can afford to pay for long-temi care from existing resources.The vast majority of working age, middle-class people—^wlio comprise theTweeners—wottld experience between a 15 and 25 percent reduction in thecosts of insurance premiums after tax deductihility. W'hile tliis would be awelcome incentive, past experience with lowering the niai^nal cost ofinsurance for middle class families suggests tliat most will not Iiegiu topurchase long-term care insurance unless a major portion of the premium ispaid for. (Bilheimer and Colby 2001)

866 HSR: Health Seruices Research 37:4 (Augu.st 2002)

The most likely option for a public program for insuring long-termcare would involve a voltmtaiy-type piogi-am based on out-of-pocketpayments for premiums similar to Part B of Medicare. This type of programcould offer graduated suhsidies to make long-term care insurance moreaffordable for moderate-income people. In order to make the insuranceiiffordable for most people, bowever, the subsidies would probably need tobe large. While this type of program should lead to substantial reductions inMedicaid payments for Tweeners, the net public sector costs would likely besubstantial.

Public offerings of insurance would avoid many of the marketingproblems associated with private long-tenn care insurance and would createsome healthy competition between existing private insurance policies and thenew public offering. A public long-term insurance program with targetedsubsidies would likely cause a much bigger expansion of insurance for middle-class families than would occur in a comparahly scaled, tax deducdbilityprogram.

The third type of financing improvement would follow the logic ofadvocates for privatizing social security: Mandatcny savings in priv-ate invesUnentaccounts could be required for all individuals starting at an age that \vouldmake annual savings afTordahle.^ Then, the annual savings would grow overdme in the private invesUnent accounts, and when a person ttinied 65 orperhaps 75, a private insurance opdon could be selected. Deborah Lucaspresents a detailed plan for a mandatory savings approach to private financing.Her analysis suggests that saving $40 to $80 a month during working yearswould cover approximately 80 percent of long-tenn care expenses. However,the estimated savings rate for prefunding starting at age 55 is almost four timesthat of preftinding stardng at age 35 and would also be quite sensidve to interestrates (Lucas 1996).

More anal)«is of these opdons—and others that might emerge—isneeded to encourage some consensus about how to improve the financing oflong-tenn care for the Medicaid Bound and Tweeners. A few importantprinciples that should guide reform debates emerge from die analysispresented here:

• Long-term care is an expensive item that most middle-class families arenot prepared to pay

• If some type of insurance plan—^whether voluntary or mandatory,whether public or private—does not begin to catch bold, the publicsector is going to see its expenditures grow faster and faster

• ' Caring for Aging Baby Boomers 867

• The sooner alternatives to the citrrent Medicaid-based financing systememerge, the less painfitl die costs of the new systetn will be.

Perhaps coti.sensus about which option is the best refotiii apptoach forlong-temi care financing can emerge otily if tliere is consetisus abottt thecriteria for judgitig options. Unfortunately, the criteria tend to conflict witli oneanother, forcing stark trade-offs. For example, one desirable critetion forasse.ssing options is the extent to which itidividttals have free choice to select amethod of ptepatitig for their po.s,sible long-tenii care needs. Options strongon this criterion tend to be weak on another desirable criterion: tbe assurancetbat a ftnaticitig approach does not restilt in people becoming a ptiblic btitdenif they fail to ptepate for long-temi care needs. A third criteri^ju itlso mayconflict with the other two desirable featttres: the ability to maintain incentivesto allocate scarce long-term care resources efficiently. At the very least, debatesabout options need to make clear what criteria related to effectiveness areiitidei" cotisideration.

BUILDING A VIABLE AND AFEORDABLECOMMUNITY-BASED DELIVERY SYSTEM

c the Baby Boomers were growing up. the needs of yotmg families werea high priotit)' in comtntttiitv' developtneiit, with particular concern forfamily-friendly bousing, parks, and schools. In 2011, these children will startutniing 65 in large nutnbers. Many predict that if comtntmities want to besticcessful iti caring for their aging poptilatioti, ibey will have to makesignificant, yet certainly feasible, changes in housing, health care, and hutnanservices.

In prepating for tbe needs of latê ntimbers of elderly, it is crticial tothink ol tbe challenge as a commtinit)' issite. If the care of the elderly beginsand ends with entry into a formalized system that takes over when a person isalmost ttnable to function day to day, society will face lat ge service costs and willmiss opportttnities to help tbe eldetly function as pioductive, indepetidentcitizens for latter portiotis of tbeir elderly yeare. A cotiimunity's social andeconomic systems need to become attuned to arranging services to meet theneeds of an aging society iti natttral, infotmal vvav's.

Most Bafiy Boomers would like to stay in tbeir own homes, or at least intheir own commttnities, as tbey age. Nearly threes]uarters of all respondents ina recent AARP survey felt strongly that they want to stay in their current

868 HSR: Health Services Research 37:4 {August 2002)

residence as long as possible (Bayer and Harper 2000). The image that mosteiders will move to a retirement village away from their communities is theexception rather than the nile. Most people will not have the resources or theinclination to move to Florida or its equivalents; commtinities cannot rely on"exportitig" to meet the needs of an aging population.

hi thinking ab<jut community capacit), three stages orcommtinit)- agingcan guide planning: the healthy-active phase, the slowing-down phase wheretlie risk of becoming frail or socially isolated increases, and the service-needyphase when an elder can no longer continue to live in the community withontsome active ser\ice in and arotind the home.

Perhaps the most important challenge of the healthy active phase ofaging is for a commtmity to leam how to tap the human resources that eldersrepresent in the community. This is a phase where elders can be key volunteersto impt ove Uie life of many segments of a commtmity. Healthy eldei"s can beconsidered a potendal component of the paid workforce if jobs can bestnictured to meet their changing preferences and capabilities.

The second phiise of aging, when elders begin to slow down and may facesome challenges in doing tlie every day actixities required of community living,represents a subUe challenge for communities. Elders in this phase often needassistance witli transportation to remain independent, and commimiUes needto take the lead to develop affordable transportation s)^t,cms {U. S. Departmentof Ttanspot tation 1997). Safe and affordable housing options also are a priorityfor community capacity efforts. At this phase of £^ng, many elders want tomove into smaller housing units that are more aging-friendly but still aretiffordable and integrated in the community. It is important to begindeveloping such options on a larê scale in the coming 10 to 20 years. In acommunity with five thousand projected elders, for example, a project with 30units will not meaningftiUy attack the problem.

Voluntarism is an imptirtant community need for elders who are mostlyindependent but slowing down (Butler 1997). Volunteers can provide servicesin a manner that makes elders contdntie to feel connected to a community andnot dependent on a formal care system. And, volunteers often can act aspreventive medicine, keeping away the effects of social isolation and keepingelders as active and engaged as possible. Volunteer capacity does not emergewithout effort, however. Commttnides need to recruit, train, and supportvolunteers.

Tlie most prevalent form of "voluntarism," of course. Is the careprovided infonnally by families and friends. Tliese caregivers also need supportthrough training programs and respite progi"ains. Many believe that additional

Caring for A^7ig liahy Boomeis 869

financial assistance for ianiily caregivers is needed as well (Stone and Keigher1994). Such eHons to support family care^ving al.so represent an importantaspect of community capacity lo support elders.

Wbile communities need to make day-to-day asjsects of communitylife more aging friendly and while volunteers and family caregiversrepresent cnicial "capacit)'" to meet the needs of eldere, a well organized,affordable fonnal long-term care system still is essential for everycommunity. It is unclear whether such local care systems can emergenaturally through niarket forces or whether market failures will emerge toblock the evolution of care systems that reflect the wants atid needs ofelders. Clearly, the large financing roles of Medicare and Medicaid givethe public sector an interest in ensuring that adequate systems of eareemerge.

How many people will require formal services in 2030? As disciLssedearlier in the paper, this is an unanswerable question in the year 2002. If effoiisat healthy aging are successful and if infbiTiial caregivei"s and volunteers canhelp to meet tlie needs of elders, the total number of frail who need fomial.services in a community in 2030 could be quite similar to the number in 2000,even though the number of elders will more than dotible. Keeping the ntimberof fmil constant at 2000 levels mtist be the goal of every community to keepcosts affordable.

However, even if tlie aggregate number of frail elders stays the same orgrows slowly, fonrial care capacity must be belter stnictured at the commtinitylevel. Importantly, most communities rely too much on nursing homes as thesource of formal care, at least for Tweeners and tlie Medicaid Boundpopulations. Sixty-seven cents of every public dollar supporting long-term carefor the elderly is spent on institutional care (Congi'essional Budget Ollfice1999), despite the clear preferences of frail eldei s for semces in thecommunity.

Wliy does this mismatch of dollars verstis preferences happen? In part,nursing homes are seen as tlie long-term care "safety net" and most publicdollars are invested using a safety-net mentality: only pay for services when itwould be socially tinconscionable not to do so. We have not developed socialconsensus alxnil when and for whom commiinity-ba.sed sei"vices should besupported widi public dollars; therefbie few public dollai"s are allocated tocommunity^based services.

This oveireliance on nursing homes—what some people call an"unbalanced" long-term care system (Kane, Kane, and Ladd 1998)—maybe changing witli help from the federal cotirt system. Recent court niUngs

870 HSR: Health Sennces Research 37:4 (August 2002)

support tbe idea tbat the disabled have a right to receive services incotntiitinity settings (Pear 2000). Sucb rtilings are putting pressure on pttblicprograms to retliitik tlie balance between nursitig home services andcommunity-based services.

Tbe cballetige over the next 10 to 30 years is to develop newapproaches to delivering community-based care. Home care, using a range oftinskilled to highly skilled wotkers, teptesenLs Uie dotiiiiianl t)pe ofcommitnity-based care. But, this service type, relying on a one-on-one model,is expetisive atid creates challenges for providers to assure quality. Newmodels, such as adult day services and housing-based services that can useone cat egiver to assist more than otie elder at a titne, need to become tnoreprevalent (Feldman 1990). In addition, emergitig technologies mightincrease the ability of one caregiver to meet tbe needs of two or threeelders tliroitgh enhanced ability to communicate and tnonitor a person'sneeds (Gottleib and Caio 1999).

One other key challenge in assuring community capacity is to recruit tberequired numbers of caregivers working in fonnal settings. Wnh chatigitigdemographics and a sttong overall labor market, it is becotning increasinglydifficult for home care agencies and otber providers to find and retainqualified caregivers. New iticcntives atid orgatii/ational strticttires will lierequired to maititain a stable workforce in long-tenn cate settings.

Finally, every community needs to tliink about what types ofinstitutional long-term care shottld be available. Even if community-basedservices expand, tlie most fiTiil among the elderly will sometimes requirethe high level of care that traditionally has been provided by nursinghomes. It is possible, however, to think about restructuring nursing homesto make their living environments and caritig style more attractive to eldersand their families (Alleti and Mor 1998). Assisted living is einerging as a.significant option for many elderly—both disabled and nondisabled. Theidea of institutional care sbould not be considered as a static model thatcannot evolve, improve, and become more responsive to the preferences ofelders.

Expansions in community capacity to care for elders need to be paid forin some way. In the case of formal services, the financing options discussedpreviously are the source of expanded resources. In the case of community-teased (hanges beyond formal services, the give and take of the politicalprocess will shape how high a priority health-promoting community programsbecome atnong the range of local priorities. And, the willingtiess of seniorsand tlieir families to allocate private resources to long term care atid related


services will determine the scofje of "caring features" in twenty-first centurycommunities.

INVESTING IN HEALTHY AGING IN ORDER TOACHIEVE LOWER DISABILITY RATES

. 1 '

Perhaps the most important challenge related to aging poptilations is thechallenge of liealtliy aging. Healthy aging (or successful or prodtictive aging) isthe coucept oi keeping senioi>i disability-free and thtis avoiding some of theneed for long-term care (Rowe and Kahn 1998). Keeping seniors healthy andfunctioning could have sigtiificant economic itnpacLs (Posner 1997). hiaddition to reducing long-tenn care costs, healthier elderly are more likely \ube profhictive memters of society. In contiust to the scarce attention being paidto improving financing for long-term care, the healthy ^ i n g challenge hasgenerated significant ititerest.

Both national and cross-national studies indicate that the rate ofdisability in a population can be extremely vaiiable. Studies of elderlyAmericans with high, average, and low levels of phy'sical activit)' have shownranges in the onset of disability of tip to ten years, with mtich lower lifetimedisability' among exercisers compared to .sedentary people. Right nowAmericans spend 72 percent of their post-65 years free of disability. Ourgoal should be to match the Japanese, who spend 91 percent of their timepast tlie age of 65 disability-tree. For example, Japanese females at age 65have an average life expectancy only 4 months longer than American femalesat age 65, but Japanese elderly women spend jiLst 1.8 years disabled whileAmerican elderly women spend almost 5.5 years disabled {Waidman andMaiiton 1998).''' Wliilc researchci-s caution that some of these difiercncescould be due to varying cultural perceptions of disability and the reporting ofdisability lo survey researchers, most obsetvers of japane.se society believe thattliere ate stibstantial dilferences in disability and tliat tliey relate to liiestylechoices.

Although disability and disease were once thought to be commensuratewith old age, the examples above, along with many others, have made itincreasingly clear that lor all but the most genetically programmed diseases,lifestyle choices, social factors, and the environment playjust as laiê or laiêrroles that! genetics in influencing healtli in later life. Less Uian a tliiid of thebiological process of aging is attributed to genetics, and the potency ol genesthat aifect aging declines even furtlier after age 65 (Finch and Tanzy 1997).


Thus, society has the ability to promote successful aging and reduce andprevent disability among the elderly.

Ffforts to Avoid Disease and Disability Related to Medicine and Medicare

Advances in genomics and medicine may represent the most straightforwardsti-ategy (at least compared to changing behaviors and lifestyles) to reducedisease and disability. Tlie budget for die National Institutes of Health morethan doubled between 1988 and 2000, from $6.6 billion to $18 billion, andappi-opriations are projected to reach $27 billion by 2003. More than $2.5billion has been spent on the Human Cênome Project since 1988 (HtimanGenome Project hiformation 2000). These invesunents should lead to advancesin earlier detection of disease orgenetic predisposition to disease, more rationaldnig design, and possibly even gene therapy. Futtire medical interventionsmight transform the initial stages of chronic disea.se—such as the onset of.Alzheimer's or arthritis—into acute disease events that can be remedied (oreven prevented through vaccination) after one or two visits to the primary carephysician (Singer and Manton 1998). Consider Alzheimer's disease alone; anestimated 14 million people in the United States could suffer from Alzheimer'sin 2040 if today's prevalence rates remain constant In recent years, however,undei-standing about the neurobiology of the disease has increased as genesand proteins that increase siLsceptibilit)' to Alzheimer's have been identified andsttidied (Selkoe 1999). This new knowledge is leading lo earlier diagnosis, thedevelopment of better drugs that treat symptoms, and .some hope that vaccinesand otlier methods for a! least slowing the onset of Alzheimer's will emerge.

Better management by the medical care system of a broad range ofchronic diseases could also reduce the incidence of disability. Society'sunderstanding of what the health system needs to do to encourage preventionand clinical care management of chronic di.seases has improved tremendouslyin recent years. Despite this, the right fomitila has not emerged for settingincentives that will lead to widespread adoption of good clinical caremanagement principles among tlie ntimerotis medical providers who carefor the elderly (Wagner et al. 1999). hicrea.singly, however, public and privatepayers are beginning to demand better clinical caie management approachesfor the chronically ill.

Although Medicare has clearly improved the health status of the elderlythrough access to acute medical care (Lubitz et al. 2001), additional interven-tions are necessary to improve health care services for the elderly. Mtich moremust be done to facilitate better clinical care tnanagement of chronic diseases

• V Caring fm- Aging Baby liootmrs 873

(Wagner etal. 1996) and encourage healthy aging. This cottld include better useof clinical preventive services to redtice tlie costs of Medicare {Rtis.seII 1998),thotottgh itnplemetitation of chronic disease matiagetnent practices, anditicetitives to increase the use of bebavioral interventions tbat could helppatients quit smoking, better monitor dialx-tes, and promote pbysic;\l activity.Ahbotigh Medicare acute care costs are positively affected by prevention effotts,thetc is liitle payoff to tnedical care providei s who invest iti preventive efforts.This lack of connection between Medicare prevention efforts and savings is acurrent barrier to better integration of ptevention efforts into Medicare.

Efforts Related to Socioeconomic and Lifestyle Factors

Altbougb medical advances generate the most excitement, basic social andlifestyle factors migbl have the largest long-tenn impact on di.sitbility rates.Vatioiis stitdies indicate that havitig income alxne tbe lowest qitintile, having ahigh school education or greater, not smoking, atid exercising are among themost impoitanl determinants of healthy ^ n g (Strawbridge et al. 1996; Stitcketal. 1999).

On the positive side, one factor that could lead to healthier aging amongBaby Boomers compared to the current elderly is the changing life circum-stances of childhood and adulthood. It appears that netuobiological circum-suinces in old age may Ix' shaped in part by experiences during early, criticalperiods of brain development, and that many changes in function during agingshow variabilit)' related to these early life experiences. That is, cbildbooddiseases, titttritional deficiency, poverty, and lack of educatioti might becontribtiting to what is now viewed as normal aging. Chikiiiood healthproblems have been linked to a variety of morbidities later in life, includingarthritis, cardiovascular conditions, and cancet% eveti when socioeconomicstattts is cotitioUed for (Blackwell 2001). According to the Batker hypothesis,prenatal conditions can have a large effect on health in later life as well, withpoor ntitrition in u/m?related to increased cardiov-ascular disease, diabetes, andother ailtnents in later life (Barker 1995). Matiy elderly alive today mattireddtuing the two world wars and tlie long depressioti petiod, wheti tiialnuuitiotiand vitamin deficiencies were still commoti. The Baby Btx:>m cohort, bycontrast, grew tip among tiiitcb better bealth. economic, ntttrilional, andeducational conditions. Because of antibiotics and immttnizatiotis they willhave been largely utitoucbed by tbe ravages of childhood di.sease.

Education is also strongly correlated with psychological fitnction, healthbehaviors, and biological conditions (Kubzsansky ei al. 1998). Pei-sons wbo are

874 HSR: Health Sendees Research 37:4 (August 2002)

not high school graduates are at almost twice the risk for experiencing declinesin functional abilides in older adtilthood. It is encouraging Oiat nadonal trendsin educational attainmetit amotig tlie elderly are so posidve, witli future cohortsha\ing completed many more years of sch<:>oling than the current elderly.However, even older adults without much formal education can benefit fromprograms and acdvides that keep tlieir mind.s .supple and active.

Trends in healthy behavions are not as encouraging as the socioeconomicstadstics. Nadonal trends iti healthy behavior have been mixed, with stagnadonin exercise, increa.ses in obesity, and decreases in smoking, hi 1997, only one-half of all 65- to 74-yearolds and one third of all people aged 75 and olderengaged in any leisure dme physical acdvity each week. Twenty-four percent ofpeople aged 60 and older are obese and current obesity trends among youngercohorts indicate that this number will only increase (U.S. Department ofHealdi and Human Senices 2000). Obe.sit>' is a risk factor in tlie elderly forarthrids, lung dysftmcdon, hypertension, diabetes, cardiovascular disease, andcertain forms of cancer (Kotz, Billington, and Levine 1999).

Altliotigb many .siLspect diat rising obesity rates will increase ovei'allmedical costs, it is unknown wbat effect tliis will have on long-tenn careexpenditures. Analysis of mortality stadsdcs indicates that obesity has a muchlarger effect on life expectancy at younger than older ages, but futuregeneradons of elderly are likely to have a much higher rate of obesity thancurrent generations. Also, the health implicadons are tmknown for oveiiveightelderly who have been overweight for much of their adult lives (Kotz,Billington, and Le\ine 1999).

Wliai can l>e done to change these tretids? A sophisticated socialmarkedng campaign might increase awareness and change atdtudes abouteadng and exercising. In the past, successful such efforts made progress inincreasing awareness about caixiiova-scular health risks atid the importance ofcholesterol monitoring and control of hypertension (Shea and Basch 1990;Dustan, Roccella, and Garrison 1996). Better pharmaceudcal agents that helpcotiti'ol obesity also are likely to be developed in the coming yeai"s.

Communides need to pro\ide more and better cjpportunities for healthpromodon for older adults. In 1997, only 12 percent of adtilts aged 65 years andolder participated in one or more organized healdi promotion acdvides (U.S.Department of Healtli and Human Senices 2000). Many commtmides also donot offer acdxity-friendly environments Uiat encotirage seniors to walk orengage in other physical acdvity.

On a positive note, many researchers have shown that it is possible toreach old age in a healdiy condidon; those with healtliy habits have a very good

Caring for Aging Bal)y Boomers 875

chance of reaching old age without disability (Vita, Terry, Hubert, and Fries1998) and without accniitig lai-ge health care costs (Schatifller, D'Agostino,and Kannel 1993; Daviglus, et al. 1998).

Efforts to Nurture Strong Interpersonal Relationships

,'\ltliotigli research has shown that the lack of social relationships is a major riskfactor for poor health, as significant as smoking or inactivity, few directinterventions to work on this issue have emer;ged. Sttidies suggest that mortalityrates rise sharply at low levels of social connection, with death more than twiceas likely compared to people with adeqtiate social relationships (Berkman andSnne 1979). Good social connections also affect mental hcitlth and cognitionas well. One study found that persons who had no social ties were twee as likelyto experience cognitive decline compared to those peisons with five or sixsocial ties (Bassuk, Glass, and Berkman 1999). The 1986 General Social Surveyindicated that twenty percent of the elderly (6.4 million people) are so sociallyisolated their health is at risk. Two million of them have no social network at alt.Clearly, these elderly are at high risk for tinhe;ilthy aging.

Opportunities for healthy aging ctmently are highly related to social andeconomic status. The inequalities that exist in society translate into healthinequalities. Recei\ing heiilth insurance for the first time at age 65 will noteliminate the impact ofyears without insurance. Poor dental caie as a child leadsto lifelong increased susceptibility to many types of infection. Recent researchh;is shown that lifetime differences in social and economic circtimstances aifectdifferences in mortality up to age 89 years (Kubz.sansky et al. 1998). Altbotighresearch indicates that its a whole the elderly will be- iiuidi ixMtt-r off in 20.S0 than(hey are today, closer examination reveals a significant minority of elderly—dis-proportionately women and minorities—in 2030 who cotild be left behind.

REGHARGING THE CONGEPT OF FAMILY ANDTHE VALUE OF SENIORS IN AMERICAN CULTURE

The fotirth challenge related to meeting tlie long-term care needs of an agingpopulation is qttito intangible and is dependent on culture riither than publicpolic). The idea of eldei^ as an economic burden or as Trail and weak is atwentieth-century construct. An interesting book by Thomas Cole traces thehistory of society's views on aging (Cx)le 1992). hi ages when death struckrandomly and evenly at all ages, people did not ftjctts so tnuch on a birth to death,

876 HSR- Health Smnces Research 37:4 (August 2002) '

linear view of life. And, agrarian economies where the young, the middle-aged,and the old all play productive roles enhanced tlie sense of the value of all ages.

So, in past eras life was viewed more as a circle—the Lion King image.But, since the Victorian Age and especially during the twentieth century, asmore people have lived to old age, the linear interpretation of the life cyclehas become dominant. Tlie past centuiy's improvements in medical andeconomic conditions for older people have been accompanied by culturalisolation and a change in the conception of old age. Old age has beenremoved fi om its once spiritual location in the journey of life to beingredefined as a medical problem.

Perhaps it is time to rethink the value of aging and the positive aspects ofaging and to adopt a cultural view capttired hy the imagery of the "Long LateAfternoon of Life." WTiile it is difficult to change "cuUiire" per se and the wayelders are viewed in society, there are practical steps communities, employers,and individuals can begin to take to prepare for a society with greater numbersof healthy elders.

First, it is worth reassessing the responsibilities and assets of elders. Allages need roles in life. Accoixiing to Erik Erikson, die hallmark of successfullate-life deveUipment is the capacity to be generative and to pass on to futuregenerations what one has learned from life. Marc Freedman has called theelderly "America's one growing resource" and views the aging of thepopulation as an opportunity to be seized (Freedman 1999). More than halfof all elderly volunteer their time. In the past few decades with the creation ofNational Senior Service Corps, the Foster Gixuidparent program, and the Faithin Action initiative, more opportunities than ever are available to elderly whowunt to contribute to their communities.

Half a million people age fifty and older have gone back to college (Riley1998). Firms arc integrating workforces through programs of "unredrcment"or by hiring retirees as temps, consultants, and pait-time workers. Surveyssures t that the 60-year trend of a decreasing number of elderly working hasreversed itself as Baby Boomers reconsider dieir financial needs for retirementas well as how they want to spend more dian a third of their adult life, hi 2000,the percentage of elderly who worked, nearly 13 percent, was higher than it hadbeen in 20 years (Walsh 2001). The young elderly (people in their 60s) havereported increased ability to work, with a 24 percent drop in the inability towork at tliis age; tlie percentage of elderly tinable to work at age 65 in 1982 washigher than the percentage utiable to work at age 67 in 1993 (Crimmins,Reynolds, and Saito 1999). Most forecasters project this trend to continue asmore elderl)' work longer for economic, social, aiid personal reasons.


etnployers become more flexible and aware of the needs and benefits of olderworkers, and the labor market remains tight, with a stnaller number of availableyounger workei's.

Since the sheer size and energy of the Baby Boom generation has led toother dratnatic social shifts, sotne expetis see hope that a new itnagery foragitig is possible. A growing interest in "age integration"—a ptocess tliattakes advantage of the broadened range of accutnufated "life course"experiences in society—has occttrred over the last few decades. In an age-integrated society, changes made to bting older people into the maitistteatncottld simttltaneously enlarge personal opportunities aud relieve many otlierpeople who ate in their middle yeat^ of the wotk-family "crtttich"(Uhlenberg 2000; Riley 1998).

Actual ph)'sical integratioti between the generations can take place too.Although some towns have seen a trend towatd ".senior-only" housing, othersare exploring options in integrated apartment buildings. Surveys have shownthat most older people prefer a tnixed-age tieighborhtxid over one restricted topeople their own age. Some cotnmutiity centers are integrating setiior centerswith child<are centers, facilitating cross-age interaction and at the same timeconserving space and resources.

(Ailtttral change also is possible, in tenns of one-t<>one relationsbips.Baby liootners have made an art of enjo)ing and (iiking ptide in everytliitigabottt caring for cbildren; some even go so far as managing to altnost "enjoy"paying $1^0,000 atmtially for college tuition. Tbe needed ctiltttr;\] shift is forchildren and comtnunities to find more enjoyment atid pride iti providing forthe care of parents and neighbors.

The sitnple message—and tbe intangible goal—is to recognize tbe giveand take of all patts of society. Atiyone who has spent time caring for an elderlyfrietid or telative recognizes diat in tlie end, categivers teceive far more tbantliey give in the relationship. Everyone benefits when tlie elderly can beintegrated fully into a caring society.

NOTES

1. Based on average life expectancy of 17 additional years at age 65, out-of-pocketprescription drug costs of approxitnately $400 a year (with 5 percent annualiticreases over inflation), oitt-of-pocket medical care costs of $900 a year, anduncovered insurance premiums a>sts of $ I,(MX) a year, and average life long-termcaie costs.

878 HSR: Health Services Research 37:4 {Augu^st 2002)

2. The ADLs include eating, hathing, dreeing, toileting, transfers, and continence.Tlie IADLs include light housework, hmndiy, meal preparation, ti-an-spoiTation,grocer)' shopping, telephoning, atid medical and money management.

3. Acccndiiig to demographic pR>jecd<)iis, the ratio.s continue to moderately worsenbetween 2030 and 2050 as Baby Ikx)niers reach very old ages.

4. The Lewin-LTC model used in this paper uses relatively conservative assumptionsahont declitie.s in disability over the next thitt)' years and asstitncs that tlie averageainountof time spent in disahility before death will remain the same even its the ageof death rises. Thus, disahility will decrease at an averse annual rate, depending onage, of between 0..5 and 0.6 percent—miiToring tlie social security tnisteesassumptions for declines in moiTality—and keeping the percentage of elderly in thecommtitiit)' who are disahled in 2030 at approximately the same level as in 2000:around 5 percent.

5. In this paper. Lee and Skinner outline the major caveats in making projections ofhealth care needs, mortality, and disability, the lack of consenstis in ititerpretation ofdata, the need for stronger modeling strategies, and what would be ncces.s;iry formore rohast, long-term predictive power.

6. All comparisons were done u.sing 2000 dollars, with projected inflation adjustments.(Knickman, Snell, and Hunt).

7. Tim Smeeding coined tlie tetiii "Tweeners" (Smeediiig 1986) and Smeeding andHolden have considered tlieir prohlenis in earlier papei's (Smeeding and Holden1990).

8. The $150,000 amotmt is lused to define a long-tenn shock becatise this amountwould be needed to support a three-j'ear nursing home stay in many parts ofthe country. To calcttlate "citnênt income" we include income over a three-year period since most nursing home stays are less than three years, but manylast more than one year. Thus, including only one year of current incomemight have imderstated resources avaiiahle for long-term care. Income includesearnings, «>cial sectiritj' income, pen.sions, other annuities, and invesunentincomes. If a peiôn is single, it is assumed the long-term care resourcesavailable include income over three years plus liquid assets. The long-term careresources available to a person who is part of a couple is three times income,minus $16,000 a year for tlie spouse plus liquid axseLs. minus the half tlie liquidassets of $120,000, whichever is smaller. Thtis, lotig temi care resources for acouple = .S[M^X (0, incotne-$16.000')] + (liquid assets - [MIN (1/2 liquidassets, $120,000)]).

9. The Long-term Care Financing Model simulates the utilization and financing oflong-term care services for elderly individuals through 2050 tising natifjnal data. Tlietwo piincipal components of the model ;u e the Pension and Retirement IncomeSimulation Model (PRISM) and the Long-tenn Care Fitiancing Model. The PRISMsimulates future demographic characteristics, labor force paiiicipation, income andassets of the elderly. The l^ng-term Care Financing Model sitnulates disability,admission to and u.se of instittitional and home and community-based care, andmethofLs of financing long-term c:ire services. Tlie mode! tises a Monte Ciirlosimulation methodology. The current version of the tnodel is the second major

Caring for Aging Baby lioomeis 879

revision of the niddel that was developed jointly by Lewin-ICF and the BrookingsInstitution in 1986.

In almost all cases, PRISM tises the Intermediate assumptions from the1999 Social St'ctnity Tmstccs' Rejiort. Tiie PRISM mtHlcls iiuonic lioiii socialsecurity, private and public employee ri-tiremeiit plans, individual RetirementAccounts, Keogh accounts, earnings, assets, and the Supplemental Securityhicoine progi~am. Agpegaie changes in wage levels are iissunicd to increase atdie raic assumed in ihc Intcmiediatc category' of the 1999 Tmstfcs' Report, lngeneral, average wages are assumed to gi-ow b)' (). )-1.0 perct-nuigc poiiiLs inexcess of the inflation rate in each yeai' after 1998. The PRISM simulatesmortality, disability, childbearing, and changes in marital stattis. Mortalit)' ratesvan' hy age, gender, fiisability stattis, years since fjecoming disabled, anrl race(black versus iiuiiblack).

In the Long-term Citre Finaticing Model, disabled individtiais age fj.5 andolder are defined as those who are tmable to condtict at least one instnimentalactivity of daily living or unable to conduct at ie:ist any one of five activities ofdaily living. Tlie di.sabilit} prevuleuce rates tised in the model were calculatedusing data ftom tlie 1**94 NationtU l,ong-tenn (ûe Stuvey (NLTCS) whileassuming that the overall period during which an individual is likely to have adisiibilitv will remain suible. Taking an intenuediate view of long-renn disabilityprevaleiue rates, it itssumcs ihat disability will decrease at an average annualrate of between 0.5 and 0.6 percent—iiiiiToriiig tfie social sectirity tiiLstees'assumptions for declines in mortality—dejKiiding on age. Tiiis rate is approxi-mately one-half the 1.2 perceni per year declines in disability estimated byMantcjn based on the 1982 and 1994 NLTCS (/Mecxib, L. B..J. a)rea, and R.Foreman, 20()0. "Lotig-tcmi C^re Fin:uicing Model: Model Assumptions[Draft]." Tbe I.ewin (iroup.)

10. Il is very difficult to know how tasl long-ienn care costs will inflate over a tliirty-yeai"period. On the otlier hand, if lafxjr becomes more productive in the generaleconomy, service costs could inflate at faster rates tliau average l>ecatise productivity'gains in the .service sector often lag average gains in the economy. However, over thenext thirty yeai's it is possible that new lecbnologies and new service sUâtegies couldimprove the efficiency of tlie long-term care secttjr. The assumption oi 1 percentiuflaiion aliove average inflaliou seems like a nuxierately—but iioi unreasonably—optimistic guess.

If. Ihe gi(junds for mandatory savings would Ix- based on tbe folfowing type of"thinking: (a) people miscalctilate the importance of saving fbr long-term care, and(b) if individuals fail to save, tfie costs oflong-temi care become a .social biiitien. Thislogic motivates a wide range of regtilaloiy betiaviov ranging from social security tomandatory automobile instirance.

12. Some have fe:ued Uiat as fife expectancy iticreases, tlie time spent iu poor fieiiltli willcorrespondingly lise. Recent data does not seem to support these fears. Average lifeexpectancy at age 8. was .six years in 1980, and. by 1997. had incre;tsed by less thanfbui" months (U.S. (Census Bureau 20(K)). fletlines in disability, however, appear tof)e quite significant


REFERENCES •

,\llen, S., atid V. Mor, cds. 1998. Living in the Community with Disability: Service Needs,Use, and Systems. New York: Springer.

Anio, P. S., C. Levine, and M. M. Memmott. 1999. "The Economic Value ofInformal Caregiving." Health Affairs 18 (2): 182-8.

Banas/ak-Holl, J.. S. Allen, V. Mor, and T. Schott. 1998. "Organizadonal Charac-teristics Associated with Agency Position in Community Care Networks."Journal of HeaUh and Social Behavior 39 (4): 368-85.

Barker, D.J. 1995. "Fetal Origins of Coronary Heart Disease." British Medicaljoumal311: 1 7 1 ^ .

Bassuk, S. S., T. A. Glass, and L. F. Berkman. 1999. "Social Disengagement andIncident Cognitive Decline in Commnnit\'-Dwelling Elderly Persons." Annalsof Internal Medicine 131: 165-73.

Bayer, A., and L. Harper. 2000. "Fixing to Stay; A Nadonal Survey on Housingand Home Modification Issues." American Associadon of RetiredPersons.

Berkman, L. F., and S. L. Syme. 1979. "Social Networks, Host Resistance, andMortality: A Nine-Year Follow-up Study of Alameda Count)' Residents."Ammcan Journal of Epidemiology 109 (2): 186-204.

Bilheimer, L. T., and D. C Colby. 2001. "Expanding Coverage: Reflecdons onRecent Efforts." Health Affairs 20 (1): 8.3-95.

Bishop, C. E. 1999. "Where Are the Missing Elders? The Decline in Nursing HomeUse, 1985 and 1995." Health Affairs 18: 146-55.

Blackwell, D. L., M. D. Hayward, and E. M. Crimmins, "Doe.s Childhood HealthAffect Chronic Morbidity in Later Life?" Social Science and Medicine 52:1269-84.

Butler, R. N. 1997. "Living Longer, Contributing Longer." Jcmmal of the AmericanMedical Association 28\ (16): 1372-3.

Cole, T. R. 1992. The Journey of Life: A Cultural History of Aging in America. New York:Cambridge University Press.

Congressional Budget Office. 1999. CBO Memorandum: Projections of Expenditures forLong-Term Care Services for the Elderly. Washington, DC:

Congressional Budget Office ba.sed on data from the Social Security Administrationand from Ediiard Bos. 1994. World Population Projections, 1994-95 Edition.Washington, DC: World Bank.

Crimmins, E. M., S. L. Reynolds, and Y. Saito. 1999. "Trends in Health and Ability toWork among the Older Working-Age Population." youma/ of Geron.tology 54B

Crimmins, E. M., Y. Saito, and S. L. Reynolds. 1997. "Further Evidence on RecentTrends in the Prev'alence and Incidence of Disability among Older Americansfrom Two Sources: The LSOA and the NHIS." Journal of Gerontology 52B:S59-71.

Cuellar, A. E., and j . M. Wiener. 2000. "Can Social Insurance for Long-Term CareWork? The Experience of Germany." HeaUh Affairs 19 (3): 8-25.

Caring for Aging Balry Bootmrs 881

Cnrran, S., S. McLanahan. and ]. Knah. In review. "Does Remarriage ExpandPercepUotis of Kitiship Stippori amotig the Elderly?" Social Eorce.

Daviglus, M. L., K. Liu, P. Greenland, A. R. Dyer, D. B. Garsidc, L. Manheim,L. P. Lowe. M. Rodin, and [. Stamler. 1908. "Benefit of a FavonibleCardiovuscular Risk-Facior Ptofile iti Middle Age witli Respect to MedicareCosts." Neil' England Journal of Medicine ^'i9 (10): 1122-9.

Dey, A. N. 1997. "Characteristics of Elderly Nttrsing Home Residents: Data frotnthe 1995 National Nursing Home Survey." Advance data from Vital andHealth Statistics, no. 2S0. Hyattsville, MD: National Center for HealthStatistics.

Dustan, H. P.. E. J. Roccella, and H. H. Garrison. 1996. "<k>ntrolIingHy[)ertension:A Re.search Success Slory." Archives of Internal Medicine 15(i (17): 192(>-3.'J.

Feldman, P. 1990. W}to Cares for Them? Wtyrkeis in the Home Care Industry. Westport,CT: Greenwood Press.

FiTKh, C. E., and R. E. Tanzi. 1997. "Genetics of Agin^."Science^^^7: 407-11.'Frecdmatt, M. 1999. Prime Time: How Rahy-Boomers Will Revolutionize Retirement and

Transform America. New York: Public AiTairs.Freedman, V. A., and L. G. Mat tin. 1999. "The Role of Education in Explaining and

Forecasting Trends in Functional Limitations among Older Americans."Demography 36 (4): 461-73,

Freedman, V. A., H. Aykan, and L. G. Martin. 2001. "Aggregate Changes in SevereC'ognitive Impairment among Older Arneticans: 1993 and 1998." founial ofCerontology: Social Sciences .'i6B: SI00-11.

Freedtnan, V. A., and L. G. Martin. 2000. "Contribution of Chronic Contlitions tt»Aggregate Changes in Old-Age Futicuoning." American Journal oj Fuhlic Health90 (November): 1755-60.

Friedland, R. B., and L. Summer. 1909. Demography is Not Destiny. Washington, DC:National Academy on ati Agitig Society.

Georgetown Uriivei"sity Institute for Health Care Research and Policy. 1999.Preliminary analysis of dtita from the National Health Intervieiv Suivey and theNational Health Interview Suroey on Disability, Phase II, 1994-5. Wa.shingtoti, DC:Georgetown University.

Gottlieb, A. S.. atid F. Caro. 1999. "Extenditig the Effectiveness of Home Carethrough Low-Cost Adaptive Equipment." Puhlic Policy and Aging Repati 10 (1):13-5.

Health Care Financing Administradon. 2000. "National Health Expenditures byType of Service and Source (jf Funds: Caletidar Years 1960-98" [accessedDecember 2000]. Available at lutp://www.hcfa.gov/stats/nhe-oact/.

Meffler, S.. K. Levit, S. Smith, C. Smith, C. Cowan, H. Lazenby, and M. Freeland."Health Spending Growth Up in 1999; Faster Growth Expected in theFuture." Health Affairs'ii) (2): 193-203.

Htjlden, K. C., and T. M. Smecditig. 1990. "The Poor, the Rich, and ihe InsecureElderly Caught in Between." Milbank Qiiarterly fi8 (2): 191-219.

Hiunan Genome Project Budget. Human Genome Project Information webpage.Available at http://www.ornl.gov/hgniis/proieet/budgot.himt.

882 HSR: Health Sennces Research 37:4 (August 2002) "-.

Kane,R.A.,R. L.Kane,andR. C. Ladd. 1998. The Heart of Lang-term Care. New York:Oxford University Press.

Knickman, J. R., E. K. Snell, and K. A. Hunt. "How Many Elderly Will Be Able loAfford Uncovered Health and Long-Teini Care Senices: Projections to theYear 2030." Working paper. The Robert Wood Johnson Foundation,Princeton, New Jersey.

Kotz, C. M., C. J. Billington, and A. S. Levine. 1999. "Obesity and Aging." Clinics inGeriatric Medicine. 13 (2): 391-412.

Ktibzsansky, L. D., L. F. Berkman, T. A. Glass, and T. E. Seeman. 1998. "IsEdticational Attaintnent Associated with Shared Determinants of Health inthe Elderly? From the Macarthtir Studies of Siiccessftil Aging." PsychosamalicMedicine 60: 578-5858.

Lee, R. D., andJ. Skinner. 1996. "Assessing Forecasts of Mortality, Health Status,and Health Costs During Baby Boomers' Retirement." In Assessing Knowledgeof Ri^tirement Behavior, edited by E. Hanushekand N. Maritato. pp. 195-243.Washington, DC: National Academy Press.

Leone, R. C. 1997. "Why Boomers Don't Spell Bust." The American Prospect 30(January-February).

Levit, K., C. Cowan, H. Lazenby, A. Sensenig, P. McDonnell, J. Stiller, and A. Martin.2000. "HeaUh Spending in 1998: Signals of Change." HeaUh Affairs 19 (1):124-32.

Lubitz, J., L G. Greenberg, Y. Gorina, L. Wartzman. and D. Gibson. 2001. "ThreeDecades of Heakh Care Use by the Elderly, 1965-1998." Health Affairs(March/April): 19-^6.

Lncas, D. 1996, "Managed Competition with Prefunding: The Solution for Long-Tenn Care?" Milbmik Quarterly IA (4): 571-97.

Manton, K. G., and X. Gu.2001. "Changes in ihe Prevalence of Chronic Disability inthe United States Black and Non-black Population above Age 65 from 1982 to1999." Proceedings of ihe Natioiial Acadêmy of Science USA May 8.

Manlon, K. G., L. Corder, and E. Stallard. 1997. "Chronic Disability Trends inElderly United States Populations: 1982-1994." Proceedings of the NationalAcademy of Science USA 94 (6): 2593-8.

McCall, N., S. Mangle. E. Bauer, and J. Knickman. 1998. "Factors Important in thePurchase of Partnership Long-Term Care Insurance." Health Sennces Research33 (2, part 1): 187-203.

Miller, E. A., and W. G. Weissert. 2000. "Predicting Elderly People's Risk for NursingHome Placement, Hospitalization, Functional Impairment, and Mortality: ASynthesis." Medical Care Research and Reiiiew 57(3): 259-97.

Mokdad. A. H.. M. K. Serdula, W. H. Dietz. et al. 1999. "U.S. Obesity Trends inAdults from 1991-1998." Journal of the American Medical Association 282 (16):1519-22.

Murtaugh, C. M., P. Kemper, B. C. Spillman, and B. L. Carlson. 1997. "The Amount,Distribution andTimingof Lifetime Nursing Home Use." Medical Care35 (3):204-18.

Caring for Aging BaJjy Rooixns H83

Neer, R. M., C. D. Amaud, j . R. Zanchetta, R. Prince, G. A. Gaich, J. Reginstcr.A. B. Hodsman, E. F. Eriksen, S. Ish-Shalom. H. K. Genant, O. Wang, andB. If. Mitlak. 2001. "Effect of Parathyroid Hormone on Fractures and BoneMint'iaf Density,' in Postmenopausaf Women vn\h Osteoporosis." NeioEnglandjournal of Medicine M-\ (fiJ): 1-34.

Peterson, P. G. 1996. "Will America Grow Up Before It Grows Old?" Atlantic Monthly277 (.5): 55-86.

Pear, R. 2000. "U.S. Seek.s More Care for Disabled Outside Institutions." New YorkTimes, i;i Februarv. section 1. p. 24.

Posner, R. 1997. A^ng and Old A^. Chicago: University of Chicago Press.Reynolds. S. L., E. M. Crimmins. and Y. Saito. 1998. "Cohott Differences in

Disability and Disease Presence." Cferontolngist $H {-S)-. .37S-90.Riley. M. W. 1998. "The Hidden Age Revolntion: Emergent Integration of All

Ages." Syracuse University, policy brief no. 12.Rowe,J.,and R. L. Kahn. 1998. Successful Aj^ng. New York: Fantheou.Rovner. B. A., and 1. R. Kat/. 1993. "Psychiatric Disorders in the Ntirsing Home: A

Selective Review of Studies Related to Clinical Care." Internationalfoumal of(Wiatric P.\ychiatij 8 {.special i.sstie): 75—87.

Russell, L. B. 1998. "Prevention and Medicare Costs." Nav England Journal ofMedicine3m (16):

Schauffler. H. H., R. B. D'Agostino, and W. B. Kannel. 1993. "Risk for Cardiovas-cular Disease in the Elderly and Associated Medicare Costs: The FramiugliamStudy." American founial of heventive Medicine^ (3): 146-54.

Schoeni, R. F., V. A. Freedman, and R. B. Wallace. 200L "Persistent, Consistent.Widespread, and Robtist? Another Look at Recent Trends In Old-AgeDisability." /ounud.s of Cieunitology: Social Sciences 56B: SI-13.

Selkoe. D. J. 1999. "Translating Cell Biology into Therapeutic Advances inAlzheimer's Disease." Mature 399 (Itni 24); A23-3I.

Shea, S., and C. E. Eascb. 1990. "A Review ot" Five Major Community-BasedCardiovascular Disease Prevention Programs. Part II: Intervention Strategies,Evaluation Methods, and Kv^\\\\^." American Jounial of Health Promotion 4 (4):

Singer, B., and K. Manton. 1998. "Tbe Effects of Health Changes on Pr<)jections ofHealth Service Needs for the Elderly Population of the United States."hoceedings of the National Academy of Scimre'^Tr. l.")(iI8-22.

Stneeding. T. M. 1986. "Nonmoney Income and the Elderly: Tbe Ca.se of ihrTweencr^." foumal of Policy Analysis and Managepifntb (4): 707-24.

Stone, R. I., and S. M. Kcigbcr. 1994. "Toward an Equitable, Universal CarcgiverPolicy: The Potential of Financial Supports for Family Cmv^verfi."fomîal ofAging and Social Policy 6 (1-2): 57-7!).

Strawbridge, W. ).. R. D. Coben, S. j . Shema, and G. A. Kaplan. 1996. "SuccessfulAging: Predictors and Associated Activities." American foumal of Epidemiology144 (2): 1.35-41.

884 HSR: HeaUh Services Research 37:4 (August 2002)

Stuck, A.,J. Waiihert, T. Nikolaus, C. Bula, C. Hohmann, andJ. C. Beck. 1999. "RiskFactoi-s for Functiomil Status Decline in Community-Living Elderly People:A Systematic Literature Review." Social Science and Medicine 48: 445-69.

Uhlenberg, P. 2000. "Integration of Old and Young."f;(Ton/o/o^.*MO (3): 276-9.U.S. Census Bureau. 2000. Statistical Abstract of the United States. Tabk 118.

Washington, DC: U.S. Census Bureau.U.S. Department of Education. 1998. Anabasis of U.S. Department of Commerce and

Cun-ent Population Survey. Washington, DC: U.S. Department of Education.U.S. Department of Health and Human Services. 2000. Healthy People 2010

(Conference Edition). Washington, DC: U.S. Department of Health andHuman Services.

U.S. Department of Transportation. 1997. Improving Transportation foi a MaturingSociety. Officeof the Assistant Secretary for Transportation Policy, Januaiy, no.DOT-PI 0-97-01. Washington, DC: U.S. Department of Transportation.

Vita, A.J., R. B. Terrv, H. B. Hubert, andJ. F. Fries. 1998. "Aging, Health Risks, andCûmulative Disability." Neio England Journal of Medicine ?i?i?>: 1035-41.

Wagner, E. H., B. T. Austin, and M. Von Korff. 1996. "Improving Outcomes inChronic Illness." Managed Care Quarterly 4 (2): 12-25.

Wagner, E. H., C. Davis,J. Schaefer, M. Von Korff, and B. Austin. 1999. "A Survey ofLeading Chronic Disease ManageiTient Program.s: Are They Consistent withthe Literature?" Managed Care Qiiarterly 7 (3): 56-66.

Waidmann, T. A., and K. G. Manton. 1998. "International Evidence on DisabilityTrends among the Elderly." Office of Disability', Aging, and Long-Term CarePolicy, ASPE, Department of Health and Human Service.s. Available at h t tp : / /aspe.os.dhhs.gov/daltcp/reports/trends.htm

Walsh, M. W. 2001. "Reversing Decades-Long Trend, Americans Retiring Later inLife." New York Times, 26 February 26, section A, p. I.

Weissert, W. G., C. Cready, andJ. Pawelak. 1988. "The Past and Future of Home andCIommunity-Based Long-Term Care." Millbank Quarterly 66 (2): 309-88.

Wiener, J., and D. Stevenson. 1998. Long-Term Care for the States. Washington, DC:The Urban Institute.

Wiener, J., L. Hixon Illston, and R. J. Hanley. 1994. "Strategies for Public andPrivate Long-Term Care Insurance—Sbaring the Burden." Washitigton, DC:The Brookings Institudon.

The2030 problem

Education

longterm care senices

mtreing home care

tlie elderly

term care inrecent years

longterm caresystem

aging baby buottiers

ntitnler of elderly

thf economic burden