Special Care Units for People With Alzheimer's and Other

Chapter 1

Overview and Policy Implications

ContentsPage

INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3Congressional Requests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . 4Policy Context . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4Organization of the Report . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

NURSING HOMES AND DEMENTIA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9The Clinical Syndrome of Dementia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9The Prevalence of Dementia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . * . . . . . . . . . . . . . . 9Nursing Home Use by Individuals With Dementia . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10Characteristics of Nursing Home Residents With Dementia . . . . . . . . . . . . . . . . . . . . . . . . 10Problems in the Care Provided for Nursing Home Residents With Dementia . . . . . . . . 12Negative Consequences for Nursing Home Residents With Dementia, Their

Families, Nursing Home Staff Members, and Nondemented NursingHome Residents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

SPECIAL CARE UNITS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .15Six Theoretical Concepts of Specialized Dementia Care and Their Implications

for Staff Composition and Training and the Individualization of Care . . . . . . . . . . . . . 16Ideas About Special Care Units From Other Countries . . . . . . . . . . . . . . . . . . . . . . . . . . . . 23Findings From Research on Special Care Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

THE REGULATORY ENVIRONMENT FOR SPECIAL CARE UNITS . . . . . . . . . . . . . . 32The Nursing Home Reform Provisions of OBRA-87 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33Existing State Regulations for Special Care Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34

POLICY IMPLICATIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .35Implications for Consumer Education About Special Care Units . . . . . . . . . . . . . . . . . . . 35Implications for Research on Special Care Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37Implications for Government Regulation of Special Care Units . . . . . . . . . . . . . . . . . . . . . 39Implications for Reimbursement for Special Care Units . . . . . . . . . . . . . . . . . . . . . . . . . . . 42

LEGAL AND ETHICAL ISSUES IN SPECIAL CARE UNITS . . . . . . . . . . . . . . . . . . . . ..43Issues With Respect to Locked Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44Issues With Respect to Admission and Discharge . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44Issues With Respect to Consent for Research Participation . . . . . . . . . . . . . . . . . . . . . . . . . 45

OTHER ISSUES OF IMPORTANCE TO NURSING HOME RESIDENTSWITH DEMENTIA . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 47The Availability of Physicians’ Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .47The Availability of Mental Health Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48The Use of Psychotropic Medications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .49

ALTERNATES TO SPECIAL CARE UNITS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .49Initiatives To Reduce The Use of Physical Restraints for All Nursing Home Residents. 50Dementia Training Programs for Nursing Home Staff Members . . . . . . . . . . . . . . . . . . . . 50Specialized Programs for Residents With Dementia in Nonspecialized Nursing

Home Units . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51Specialized Living Arrangements Outside Nursing Homes . . . . . . . . . . . . . . . . . . . . . . . . . 51

CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

BoxesBOX Pagel-A. A Special Care Unit in Lynden, Washington . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5l-B. The Development of Excess Disability in a Nursing Home Resident With Dementia. 14l-C. The Use of Behavioral Interventions With a Nursing Home Resident With Dementia. 18

Chapter 1

Overview and Policy Implications

INTRODUCTIONAt least half of all nursing home residents in the

United States have dementia. As awareness ofAlzheimer’s disease and other diseases that causedementia has increased in recent years, so havecomplaints and concerns about the quality andappropriateness of the care provided for individualswith dementia by most nursing homes. In responseto these complaints and concerns, some nursinghomes have established a special care unit-that is,a physically separate unit in the nursing home thatprovides, or claims to provide, care that meets thespecial needs of individuals with dementia. Suchunits are referred to generically as special care units,dedicated care units, Alzheimer’s units, or dementiaunits. OTA uses the term special care units in thisreport.

The number of special care units for individualswith dementia has increased rapidly over the pastfew years. No comprehensive data are available onthe number of special care units before 1987, butinformation from several studies indicates that thegreat majority of existing special care units wereestablished after 1983 (181,413,485). The frostcomprehensive data on special care units in thiscountry were collected in 1987, as part of theNational Medical Expenditure Survey. That surveyfound that 1668 nursing homes—8 percent of all

—had a special care unit for individu-nursing homesals with dementia in 1987, and that these special careunits accounted for more than 53,000 nursing homebeds (249). The survey also found that an additional1444 nursing homes planned to establish a specialcare unit by 1991, and 535 of the nursing homes thatalready had a special care unit in 1987 planned toexpand the unit by 1991. If all these plans hadmaterialized, more than 3100 nursing homes—14percent of all nursing homes in the United States—would have had a special care unit in 1991, andalmost 100,000 nursing home beds would have beenin special care units.

When published in 1990, the figures from the1987 National Medical Expenditure Survey sur-

prised researchers and others because they weremuch higher than any previous estimates. Twostudies conducted since then indicate that the truenumber and proportion of nursing homes with aspecial care unit are probably somewhat lower(194,247). On the basis of these studies, OTAestimates that 10 percent of all U.S. nursing homeshad at least one special care unit in 1991.1 Regard-less of the precise figures, however, it is clear that thenumber and proportion of nursing homes with aspecial care unit are growing rapidly.

The proliferation of special care units creates bothproblems and opportunities for individuals withdementia, their families, and many other people andorganizations that have an interest in the quality andappropriateness of nursing home care for individualswith dementia. These other interested parties in-clude: nursing home administrators and staff mem-bers who provide care for individuals with dementiaboth in and out of special care units; physicians,nurses, social workers, hospital discharge planners,community agencies, Alzheimer’s Association chap-ters, and other voluntary organizations that referpeople with dementia and their families to nursinghomes; and nursing home licensing and certificationofficials, nursing home surveyors, and long-termcare ombudsmen who are responsible for regulatingand monitoring the quality of nursing home care.

The problems created by the proliferation ofspecial care units are due primarily to the lack ofagreement about what a special care unit is or shouldbe and the related lack of standards to evaluatespecial care units. Existing special care units varygreatly in every respect, including their guidingphilosophy, physical design, staff composition, staff-to-resident ratio, activity programs, and patient carepractices (64,181,194,199,232,256,275,332,4 13,485,494). Despite this variation, the operators of virtu-ally all special care units express confidence thatthey are providing appropriate care for their resi-dents. According to researchers who studied thedifferences among special care units:

The differences are of such significance that theyappear to place special units in direct opposition to

1 As discussed later in the chapter, tbis number includes nursing homes that place some of their residents with dementia in a physically distinct groupor cluster in a unit that also serves some nondemented residents.

–3–

4 ● Special Care Units for People With Alzheimer’s and Other Dementias

each other. Nevertheless, without exception, theirproponents have hailed the success of the units (332).

Many people have told OTA that some nursinghomes that have a special care unit just use the wordsspecial care as a marketing tool and actually provideno special services for their residents. Most nursinghomes charge more for care in their special care unitthan in other parts of the facility (413,494). In specialcare units that provide no special services, individu-als with dementia and their families may pay morebut receive no better care than they would in anotherunit in that nursing home or a different nursinghome. At worst, they may pay more and receiveinferior care in the special care unit.

Many families of individuals with dementia areextremely concerned about the quality and appropri-ateness of services they may use for these individu-als (166,5 13). As a result, they are likely to respondenthusiastically to claims of “special care. ’ With-out standards by which to evaluate special care units,families and individuals and organizations that referpatients and their families to nursing homes cannotknow with any certainty whether the units areproviding better care than other nursing home units.

Despite these problems, the proliferation of spe-cial care units also creates opportunities for individ-uals with dementia, their families, and others whoare concerned about the quality and appropriatenessof the nursing home care available to these individu-als. Even without standards by which to evaluate theunits, it is obvious to all observers that some specialcare units are providing better care for their residentswith dementia than these individuals would receiveinmost nursing homes. One such unit is described inbox l-A.

The proliferation of special care units means thatfor the first time in the United States there arenumerous nursing homes in which administratorsand staff members are concentrating on developingbetter methods of care for their residents withdementia. This attention to the special needs ofnursing home residents with dementia reverses thelong-standing reality in many nursing homes inwhich the special needs of these residents have notbeen recognized and the residents frequently havenot even been identified as individuals with demen-tia.

This OTA report discusses the complaints andconcerns about the care provided for nursing home

residents with dementia that have led to the develop-ment of special care units, the theoretical conceptsthat underlie their design and operation, and thefindings of studies that describe and evaluate them.The report analyzes the problems and opportunitiescreated by the proliferation of special care units anddiscusses the ways in which government has re-sponded or could respond to these problems andopportunities.

Congressional Requests

This report was requested by Senator DavidPryor, chairman of the Senate Special Committee onAging, and Congresswoman Olympia J. Snowe,ranking minority member of the Subcommittee onHuman Services of the House Select Committee onAging. The congressional letters of request for thereport stress the need for information about specialcare units to inform Federal policy with respect toconsumer education, research, regulation, and reim-bursement for special care units. CongresswomanSnowe noted the lack of information about the costand effectiveness of special care units and stressedthe need for quality standards to help families andothers evaluate the units and assess their options fornursing home care for an individual with dementia.Senator Pryor noted the problem of overuse andmisuse of physical restraints in nursing homes andasked whether restraints are used less often inspecial care units and, if so, what alternatives torestraints are being used.

Policy Context

Nursing home care for individuals with dementiais an important public policy issue for three reasons.One reason is that a large number and proportion ofnursing home residents have dementia. The 1985National Nursing Home Survey, a large-scale surveyof a nationally representative sample of nursinghomes, found that 696,800 nursing home residents—47 percent of all residents-had dementia (469). The1985 survey also found that 922,500 nursing homeresidents--62 percent of all residents—were sodisoriented or memory-impaired that their perform-ance of the activities of daily living was impairednearly every day (467). The 1987 National MedicalExpenditure Survey, which also included a nation-ally representative sample of nursing homes, foundthat 637,600 nursing home residents-42 percent ofall nursing home residents—had dementia (237).These figures are based on judgments by nursing

Chapter l-Overview and Policy Implications .5

Box 1-A—A Special Care Unit in Lynden, Washington

The Christian Rest Home, a 150-bed nursing home in Lynden, WA has had a special care unit since 1988. The15-bed special care unit was established because of staff concerns about the safety and well-being of residents withdementia who wander or have other behavioral symptoms that cannot be handled on the facility’s regular units.

The special care unit consists of resident bedrooms, an activity/dining area, and an enclosed outdoor courtyard.Three physical changes were made to the building to create the unit: 1) a set of doors was installed in an existing

unit to partition off the resident bedrooms and the activity/dining area; 2) a door was made in an exterior wall togive the residents access to the enclosed courtyard; and 3) keypad-operated locks were installed on the exit doors;the doors open when a number code is punched in on the keypad; the doors open automatically if the fire alarm goesoff. These physical changes cost less than $5000.

The special care unit functions as a self-contained entity, but technically it is part of an adjacent unit.Washington State regulations require each nursing home unit to have a separate nurses’ station, a separate shower,a separate bathroom for staff, and a separate utility room. To avoid the cost of these separate facilities, the specialcare unit is considered part of the adjacent unit. Medications, medical treatments, and rehabilitative services for thespecial care unit residents are delivered from the nurses’ station on the adjacent unit.

Some residents of the special care unit have been transferred to the unit from other parts of the nursing home,usually because they wander or have other behavioral symptoms that are more easily handled on the special careunit. Other residents have been admitted directly from home. Although all the special care unit residents havedementia in the opinion of the facility staff, a few have not had a diagnosis of dementia in their medical records,

The objectives of the unit are to assure the residents’ safety, to reduce agitation and behavioral symptoms, tomaintain independent functioning, and to improve the residents’ quality of life. The staff members perceive residentagitation and behavioral symptoms as meaningful expressions of feelings and unmet needs. They attempt tounderstand and respond to those feelings and needs, in the belief that by doing so, they will reduce agitation andbehavioral symptoms and improve the residents’ quality of life.

The unit has a relaxed atmosphere. The residents appear calm and contented. They wander freely around theunit and respond to and sometimes initiate verbal interactions with staff members and visitors, Although many ofthe residents exhibited severe behavioral symptoms before coming to the unit, the unit staff reports that thesesymptoms are relatively easily managed in the special care unit.

The only type of physical restraint that is used on the unit is a geriatric chair with a tray table that keeps aresident from getting up. These ‘geri-chairs’ are used only temporarily and only with a doctor’s order. Psychotropicmedications are used sparingly. They are used in low doses and only after other, behavioral interventions have beentried. On Jan. 13, 1992,7 of the 15 residents were receiving psychotropic medications, including 4 residents whowere receiving antipsychotic medications.

Formal and informal activity programs are conducted on the unit. Each afternoon there is a formal activityprogram, such as a weekly Bible study and music group, a weekly reminiscence group, a weekly “validation”group, and “high tea”-a Monday afternoon event with real china and lace tablecloths. Other activities, such asfood preparation and singing, take place informally on the unit. One resident who likes to fold laundry is encouragedto do SO.

Each morning, there is a half-hour hymn sing for all residents of the nursing home. Most of the special careunit residents are taken to this activity. In the afternoons, a few of the special care unit residents are taken to whateveractivity program is scheduled for the facility as a whole.

Family members are welcome on the unit at any time. The staff knows the residents’ families and involves themindecisions about the residents’ care. The staff reports that family members often thank them for the help they givethe residents and the emotional support they give the family members. Two formal events-a Thanksgiving potlucksupper and a summer barbecue-involve all the unit residents and their families.

During the day, the staff on the special care unit consists of one registered nurse, who functions as the unitcoordinator, and two muse aides. A licensed practical nurse and two other nurse aides take over for the evening shift.Since staff consistency is considered important for the unit, the unit staff members generally are not rotated to otherunits, although staff rotation is the norm in the rest of the facility. The special care unit staff members work as ateam, with little apparent difference in status between the nurses and aides.

(Continued on next page)


Box l-A—A Special Care Unit in Lynden, Washington-(Continued)

Until recently, the unit had no separate staff for the night shift (11:00 p.m. to 7:O0 &m.). Before being admittedto the special care unit, many of the residents had been awake, agitated, and difficult to manage at night. Once theycame onto the unit, these individuals began to sleep through the night, and the facility found it was possible to leavethe unit doors open and have the unit supervised by a staff member on the adjacent unit. Nevertheless, as ofDecember 1991, the facility had decided to assign an aide to the unit for the night shift.

The unit administrator and the facility’s staff development coordinator stress the importance of training for thespecial care unit staff, but they place greater emphasis on staff attitudes. The unit administrator believes there arepeople who cannot be trained to work effectively on the special care unit because their attitudes and personalitiesare not suited to the unit. Both the unit administrator and the staff development coordinator stress the need for aflexible, “trial and error,” approach to dealing with an individual resident’s problems and for staff members whocan implement this approach.

Several individuals besides the unit staff members are involved in the care of the residents. The weekly Biblestudy and reminiscence groups are run by staff of the facility’s Therapeutic Recreation Department. The weeklyvalidation group is run by the director of the facility’s Social Services Department, who is a psychiatric nurse. Shealso works with the geriatric mental health team from the local community mental health center to assess andrespond to residents’ mental health needs. A monthly staff meeting is held to discuss problems and ideas amongthe special care unit staff and other individuals who are involved in the residents’ care.

Special care unit residents are discharged from the unit when the staff considers that the residents can no longerbenefit from the unit. The unit discharge poilcies are explained to family members when a resident is admitted, butmany family members are upset when their relative is moved to a different unit, Several spouses of former specialcare unit residents have created an informal support group that meets almost daily in the facility, presumably toreplace the emotional support they previously received from the unit staff.

Discharges are hard on the unit staff members, since they often become attached to the resident and theresident’s family. The facility believes, however, that it is important to make space available in the unit for otherindividuals who will benefit from it. Priority is given to individuals who are at risk because of wandering.

The Christian Rest Home is a private, nonprofit facility. The specia1 care unit serves both Medicaid and privatepay residents. Until January 1992, there was no additional charge for care in the unit. Starting in January 1992,private pay residents are charged $10 more per day in the special care unit than they would be charged in other unitsin the facility. The special care unit has a waiting list, as does the facility as a whole.

SfXJRC!E: Angie Brouwer, Adtmms‘ - trator, Christian Rest Home, Lynde~ WA, personal communicatio~ Jan. 13, 1992; Linda Jager, RN, StaffDevelopment Coordinator, Christian Rest Home, Lynde& WA, personal communications, Oct. 19, 1990, Dec. 30, 1991, Jam 13,1992; Betty LOU Rau, RN, Day Charge Nurse, Special Care Uni$ Cbristiart Rest Home, Lyndeq WA, personal communications, Oct.19, 1990, Dec. 30, 1991; Jennifer Johnson, RN, Director of Social Services, Christian Rest Home, Lynde~ WA personalcommunications, Oct. 19,1990, Jan 13, 1992.

home staff members about the residents’ mentalstatus. Several small-scale studies based on compre-hensive medical and psychiatric evaluations havefound that an even higher proportion of residents (67to 78 percent) have clinically diagnosable dementia(82,389,390).

The second reason nursing home care for individ-uals with dementia is an important public policyissue is that government expenditures for nursinghome care for individuals with dementia are substan-tial. In 1990, total expenditures for nursing homecare from all sources were $53.1 billion. Federal,

State, and local government expenditures accountedfor slightly more than half (52 percent) of thatamount (250).2 Excluding expenditures for the careof individuals in facilities for the mentally retarded,total government expenditures for nursing homecare were $22.8 billion. Individuals with dementiatend to be among those who stay longest in nursinghomes and so are most likely to become eligible forgovernment reimbursement through Medicaid(229,258,465). As a result, government probablypays for more than half of all nursing home care forindividuals with dementia. Since individuals withdementia constitute at least half of all nursing home

zTotalgov ernment expenditures for nursing home care were $27.7 billion in 1990. This amount included $17.2 billion in Fedeml expemhms ($2.5billion from Medicare, $13.7 billion ffom Medicaid, and $1.0 billion from other sources, e.g., the Department of Veterans Affairs) and $10.5 billion inState and local government expenditures, virtually all of which are Medicaid expenditures (250).

Chapter l-Overview and Policy Implications ● 7

residents, OTA estimates that government expendi-tures for nursing home care for individuals withdementia amounted to more than $11 billion in1990.3

The third reason nursing home care for individu-als with dementia is an important public policy issueis that government is extensively involved in regu-lating nursing homes. The Federal Governmentregulates nursing homes that participate in theMedicare or Medicaid programs. In 1985,75 percentof all nursing homes participated in one or bothprograms, and these participating facilities ac-counted for 89 percent of all nursing home beds(467). All States also regulate nursing homes.

Complaints and concerns about the quality andappropriateness of the nursing home care providedfor individuals with dementia are pervasive. Giventhese complaints and concerns and government’sextensive role in regulating nursing homes andpaying for nursing home care, the claim of specialcare unit operators and others that special care unitsprovide better care for individuals with dementiadeserves the attention of policymakers.

The existence and proliferation of special careunits raise four policy questions. One questionpertains to consumer education. The Alzheimer’sAssociation and several other organizations havedeveloped informational brochures and guidelinesto assist families and others in evaluating specialcare units.4 New Hampshire has also taken thisapproach (325). The policy question is what, if any,additional steps government should take to informconsumers about special care units.

The second policy question pertains to the ade-quacy of government funding for research on specialcare units. Until recently, Federal agencies hadfunded very little research on special care units. Inthe fall 1991, the National Institute on Aging fundednine special care unit studies through its “SpecialCare Units Initiative,’ and a tenth study was fundedthrough the initiative in 1992. When the results ofthese studies are available in a few years, they willgreatly expand knowledge about special care units.In the meantime, it is important to consider whether

additional government-funded research is needed,and if so, on what topics.

The third policy question pertains to regulation ofspecial care units. As of early 1992, six States—Colorado, Iowa, Kansas, Tennessee, Texas, andWashington-had added requirements for specialcare units to their general regulations for all nursinghomes. Five States—Nebraska, North Carolina,New Jersey, Oklahoma, and Oregon-were devel-oping regulations for special care units, and moreStates were considering doing so. The policy ques-tion is whether the Federal Government or otherStates should develop special regulations for specialcare units.

Many special care unit operators and others say itcosts more to operate a special care unit than anonspecialized nursing home unit (12,64,377,477,485).Thus, the fourth policy question is whether govern-ment should pay more for the care of eligibleindividuals in special care units than in other nursinghome units.

Until the publication in 1990 of figures on thenumber of nursing homes that had a special care unitin 1987, most commentators believed there might beseveral hundred special care units in the UnitedStates. It was reasonable then to regard special careunits as a relatively small phenomenon and toconsider government policies for special care unitsin that context. Recent data suggesting that 10percent of all nursing homes had a special care unitin 1991 indicate that special care units are not a smallphenomenon. The rapid proliferation of special careunits means such units are likely to become a muchlarger phenomenon. Government policies for specialcare units should be considered in this new contextand in relation to the long-range possibilities andsocietal objectives for special care units.

Various long-range possibilities for special careunits can be imagined. One possibility would be forall nursing home residents with dementia to be caredfor in special care units (or in whole nursing homesdevoted exclusively to serving individuals withdementia). To OTA’s knowledge, no one advocatesthis alternative, in part because of the huge numberof individuals involved< 37,600 to 922,500 indi-

S Some and per~ps my nmsing home residents with dementia are admitted for reasons other than or in addition to their dementia. OTA’S estimaterefers to the overall cost to government of nursing home care for residents with dementia regardless of the primary reason for their admission.

4 See, for exmple, Mace and @@er> “Selecting a Nursing Home With a Dedicated Dementia Care UniG” Akheimer’s Disease and RelatedDisorders Association (276).


viduals according to national surveys-and the costand other implications of creating a whole separatenursing home industry to serve them.

A second possibility would be for special careunits to serve only certain types of nursing homeresidents with dementia—for example, residentswith behavioral symptoms or residents in a particu-lar stage of their dementing illness. To implementthis alternative would require a rationale for deter-mining which types of residents with dementiashould be in special care units and criteria foridentifying these individuals.

A third possibility would be for special care unitsto serve: 1) individuals with dementia whose fami-lies choose to place them in the unit for any reason,including ability to pay, and 2) individuals thenursing home chooses to place in the unit for anyreason, including ability to pay. In this scenario, thetotal number of special care units and the numberand types of individuals with dementia who arecared for in these units would be determined in thefuture, as they are now, by market demand and thedecisions of individual nursing home administratorsand staff members.

A fourth possibility would be for special careunits to function as research settings to develop andevaluate methods of care for individuals withdementia. Once shown to be effective, the methodsof care developed in special care units could beincorporated into the care practices of all nursinghomes, thus potentially benefiting all residents withdementia.

Government policies adopted now with respect toconsumer education, research, regulation, and reim-bursement for special care units will influence whichof these long-range possibilities becomes the futurereality. Which of the long-range possibilities isdesirable depends on several factors, the mostimportant of which are:

●

●

●

the effectiveness of special care units in generaland for particular types of individuals withdementia;the relative cost of caring for individuals withdementia in special care units vs. nonspecial-ized nursing home units; andthe impact of the different long-range possibili-ties on nondemented nursing home residents.

By definition, special care units segregate individ-uals with dementia from other nursing home resi-

dents. Some commentators believe this segregationbenefits both demented and nondemented nursinghome residents. Other commentators believe thatalthough segregation may benefit nondementedresidents, it will result in poorer care for residentswith dementia who will, in effect, be warehoused’in segregated units. In the view of these commenta-tors, the anticipated negative effects of segregatingnursing home residents with dementia outweigh anypossible positive effects of the units. Some of thelatter commentators are particularly disturbed by thefact that most special care units are either locked or“secured” in some other way so that residents withdementia cannot get out. The reactions of thesecommentators to proposed government policies forspecial care units are likely to reflect their objectionsto locked units rather than to special care units per se.

Finally, in considering government policies forspecial care units, it is important to note that theproliferation of special care units is occurring at thesame time as numerous other government andnongovernment initiatives that are likely to improvethe care of nursing home residents with dementia orprovide them with alternatives to nursing home care.These initiatives include the following:

●

●

●

initiatives intended to improve the care of allnursing home residents, including nursing homeresidents with dementia, e.g., the regulatoryand other changes associated with implementa-tion of the nursing home reform provisions ofthe Omnibus Budget Reconciliation Act of1987 (OBRA-87), and separate but relatedefforts to create ‘restraint-flee’ nursing homes;

initiatives intended to improve the care ofindividuals with dementia in any nursing homeunit, e.g., training programs for nursing homestaff members, special activity and other pro-grams for residents with dementia in nonspe-cialized units, and the development of effectivestrategies for resident assessment, care plan-ning, and treatment of behavioral symptoms;and

initiatives intended to provide appropriate careoutside nursing homes for individuals withdementia, e.g., specialized residential care pro-grams inboard and care facilities, group homes,and assisted living facilities; specialized adultday programs; and specialized in-home serv-ices.


This OTA report focuses on special care units innursing homes. A full evaluation of the initiativeslisted above is beyond the scope of the report,although the implications of OBRA-87 for nursinghome residents with dementia are discussed in thischapter and at greater length in chapter 5, and someof the other initiatives are discussed briefly at theend of this chapter. Ultimately, government policiesfor special care units should be considered in thecontext of these other initiatives which may providealternate or even better ways of accomplishing someof the same objectives as special care units.

Organization of the Report

The remainder of this chapter summarizes OTA’sfindings with respect to the characteristics of nursinghome residents with dementia and problems in thecare they receive in many nursing homes, thecharacteristics of existing special care units, theavailable information about their effectiveness, andthe regulatory environment for special care units.The implications of these findings for governmentpolicies about special care units are discussed. Thechapter also discusses several topics not addressedelsewhere in the report, including the theoreticalconcepts of specialized care for individuals withdementia and legal and ethical issues related tospecial care units.

Chapter 2 discusses the prevalence of dementia innursing homes, the characteristics of nursing homeresidents with dementia, and the most frequentlycited complaints and concerns about the nursinghome care provided for these individuals. Chapters3 and 4 analyze the results of the available descrip-tive and evaluative studies of special care units.Chapter 5 discusses the government regulations thatapply to special care units, including the specialrequirements that are now in effect in six States, andthe guidelines for special care units that have beendeveloped by various public and private organiza-tions. Chapter 6 analyzes the problem of governmentregulations that discourage innovation in the designand operation of special care units.

NURSING HOMESAND DEMENTIA

Because of the aging of the U.S. population, thenumber of individuals with Alzheimer’s disease andother diseases that cause dementia is growingrapidly. The proportion of individuals with dementia

that is in nursing homes now or will ever be innursing homes is not known, but it is likely that mostindividuals with dementia will spend some time ina nursing home in the course of their illness. Theseindividuals constitute the pool of potential users ofspecial care units.

This section provides background informationabout the clinical syndrome of dementia and itscauses, the prevalence of dementia, and the use ofnursing homes by individuals with dementia. Itdescribes the characteristics of nursing home resi-dents with dementia and discusses the problems inthe care they receive in many nursing homes and theimpact of those problems on the residents, theirfamilies, nursing home staff members, and nonde-mented nursing home residents.

The Clinical Syndrome of Dementia

Dementia is a clinical syndrome characterized bythe decline of cognitive abilities in an alert individ-ual. By definition, dementia involves some degree ofmemory loss. Other cognitive abilities that arefrequently diminished or lost in dementia includejudgment, learning capacity, reasoning, comprehen-sion, attention, and orientation to time and place andto oneself. Language functions, including the abilityto express oneself meaningfully and to understandwhat others communicate, are usually also affected.

Dementia can be caused by many diseases andconditions (see app. A). Alzheimer’s disease is themost common cause of dementia, accounting for 50to 80 percent or more of all cases (131,227,448). Thesecond most common cause of dementia is multiplesmall strokes that lead to multi-infarct dementia.

Alzheimer’s disease and most other diseases andconditions that cause dementia are progressive. Overtime, as individuals with these diseases and condi-tions lose cognitive abilities, they become increas-ingly unable to care for themselves independently.Eventually most individuals with dementia require24-hour supervision and assistance with everyaspect of their daily lives.

The Prevalence of Dementia

OTA estimates that there are now about 1.8million people with severe dementia in the UnitedStates and an additional 1 to 5 million people withmild or moderate dementia (458). The results of astudy conducted in East Boston in the early 1980ssuggest that as many as 3.75 million people may


have Alzheimer’s disease at all levels of severity(129), but some researchers and clinicians considerthis estimate high.

The prevalence of dementia increases dramatic-ally with age. OTA estimates that the prevalence ofsevere dementia increases from less than 1 percentof people under age 65, to about 1 percent of thoseage 65 to 74,7 percent of those age 75 to 84, and 25percent of those over age 85 (458). It has beenhypothesized that the incidence of new cases ofdementia may level off in individuals over age 85,but followup data from the East Boston study andother sources indicate that the incidence of dementiacontinues to increase (130,495).

The U.S. population over age 65 is growing fasterthan younger age groups, and the 85+ age group isgrowing faster than other segments of the olderpopulation. As a result, the number and proportionof individuals with dementia in the population aregrowing rapidly.

Nursing Home Use by IndividualsWith Dementia

The proportion of individuals with dementia thatis in a nursing home at any one time is not known.Nor is it known what proportion of individuals withdementia will ever be in a nursing home in the courseof their illness.

On the basis of figures from the 1985 NationalNursing Home Survey--i.e., 696,800 nursing homeresidents who had senile dementia or chronic ororganic brain syndrome and 922,500 nursing homeresidents who were so disoriented or memory-impaired that their performance of the activities ofdaily living was impaired nearly every day—andOTA’s estimates of the prevalence of dementianationwide—i.e., 1.8 million Americans who havesevere dementia, and 1 to 5 million who have mildor moderate dementia--one could estimate thatanywhere from 10 to 33 percent of individuals withdementia of any degree of severity are in a nursinghome now. If one surmises that only individuals withsevere dementia are likely to be in a nursing home,one could estimate that anywhere from 39 to 51percent of individuals with severe dementia are in anursing home now.

A much larger proportion of individuals withdementia are likely to spend some time in a nursinghome in the course of their illness, although some

individuals with dementia will never be in a nursinghome. Recent projections from data on elderlyindividuals who died in 1986 suggest that 43 percentof all Americans who reached age 65 in 1990 willspend some time in a nursing home before they die(230). Individuals with dementia are far more likelythan elderly individuals in general to be admitted toa nursing home, and it may be that almost allindividuals with dementia will spend some time ina nursing home in the course of their illness.

The proportion of individuals with dementia thatis in a nursing home at any given time and theproportion that will be in a nursing home at sometime in the course of their illness could increase ordecrease as a result of several factors. These factorsinclude the availability of appropriate residentialcare in alternate settings, such as board and carefacilities; the availability of appropriate in-home andcommunity services; and Medicaid eligibility, cov-erage, and reimbursement policies that encourage ordiscourage nursing home placement for individualswith dementia.

Characteristics of Nursing Home ResidentsWith Dementia

Available information about the characteristics ofnursing home residents with dementia is presentedin chapter 2. As noted there, nursing home residentswith dementia are older on average than othernursing home residents. The 1985 National NursingHome Survey found that half of the residents withdementia were over age 85, compared with one-thirdof the other residents (469). The survey also foundthat three-quarters of the residents with dementiawere female. Although a preponderance of femaleresidents with dementia is to be expected sincefemale nursing home residents greatly outnumbermale residents, the survey data indicate that femalenursing home residents were somewhat more likelythan male residents to have dementia (48 percent vs.40 percent, respectively) (469).

Nursing home residents with dementia are morelikely than other nursing home residents to needassistance with activities of daily living (i.e., bath-ing, dressing, using the toilet, transferring from bedto chair, remaining continent, and eating). The 1985National Nursing Home Survey found, for example,that 69 percent of residents with dementia neededassistance to remain continent, compared with 37percent of the other residents (469) (see fig. l-l).

Chapter 1--Overview and Policy Implications . 11

Figure I-1—impairments in Activities of Daily Livingin Demented and Nondemented Nursing Home

Residents, United States, 1985

100 9 6 %

80

60

40

20

0

87%’.

74%70% 69%

5 4 %

Bathing Dressing Using Transferring Continence Eatingthe toilet bed to chair

- Demented residents m Nondemented residents

SOURCE: Adapted from U.S. Department of Health and Human Services,“Mental Illness in Nursing Homes: United States, 1985,” PublicHealth Service, National Center for Health Statistics, DHHSPub. No. (PHS) 89-1758, Hyattsville, MD, February 1991.

Psychiatric symptoms are more common amongnursing home residents with dementia than amongother nursing home residents. The 1987 NationalMedical Expenditure Survey found, for example,that 36 percent of residents with dementia hadpsychiatric symptoms, such as delusions and hallu-cinations, compared with 26 percent of other resi-dents (464) (see ch. 2).

Behavioral symptoms are also more commonamong nursing home residents with dementia thanamong other nursing home residents. The 1987National Medical Expenditure Survey found that 59percent of residents with dementia had one or moreof ten behavioral symptoms (wandering, physicallyhurting others, physically hurting oneself, dressinginappropriately, crying for long periods, hoarding,getting upset, not avoiding dangerous things, steal-ing, and inappropriate sexual behavior) (464). Incontrast, 40 percent of other nursing home residentshad one or more of these symptoms (see fig. 1-2).

Although these data show that nursing homeresidents with dementia are more likely than othernursing home residents to have impairments inactivities of daily living and psychiatric and behav-ioral symptoms, not all nursing home residents withdementia have these problems. The survey data

Figure 1-2—Behavioral Symptoms in Demented andNondemented Nursing Home Residents, United

States, 1987

59%60

40- - 37%

20- -

6%2%

0 {

1 or more 1 to 4 5 to 10behavioral behavioral behavioralsymptoms symptoms symptoms

- Demented residents ~] Nondemented residents

SOURCE: Adapted from U.S. Department of Health and Human Services,published and unpublished datafromthe 1987 National MedicalExpenditure Survey, Institutional Population Component, Cur-rent Residents, Agency for Health Care Policy and Research,Rockville, MD, 1991.

indicate that 4 to 46 percent of residents withdementia do not have impairments in activities ofdaily living, depending on the activity, and that morethan 40 percent of residents with dementia do nothave behavioral symptoms.

Nursing home residents with dementia also differin their coexisting medical conditions and physicalimpairments. OTA is not aware of any informationfrom national studies on the proportion of nursinghome residents with dementia who have coexistingmedical conditions or physical impairments. Asdiscussed in chapter 2, data on the characteristics of3427 residents of New York nursing homes showthat residents with dementia vary greatly in thisrespect (283). Some are relatively healthy except fortheir dementia, and others have numerous diseasesand physical impairments in addition to their demen-tia.

The diversity of nursing home residents withdementia has important implications for special careunits. First, it is unlikely any particular type of unitwill be appropriate for all types of nursing homeresidents with dementia. Second, with respect to thelong-range possibilities discussed earlier, it is clear


that if special care units were designated to serveonly individuals with behavioral symptoms, theunits would not serve all individuals with dementiawho need nursing home care, because more than 40percent of nursing home residents with dementia donot have behavioral symptoms.

Problems in the Care Provided for NursingHome Residents With Dementia

Many complaints and concerns have been ex-pressed about the quality and appropriateness of thecare provided for nursing home residents withdementia. These complaints and concerns are theprimary reason for the development and prolifera-tion of special care units. They explain to a greatdegree why there is a market for special care units.They are also the rationale for many of the specificchanges in physical design features, patient carepractices, and staff training that are recommendedfor special care units.

Table 1-1 lists the most frequently cited com-plaints and concerns about the care provided fornursing home residents with dementia. This list isbased on OTA’s review of numerous articles andbooks on nursing home care for individuals withdementia (see ch. 2). The inclusion of items in thelist does not imply that there is evidence to prove theitems are true but rather that the items are aspects ofwhat is believed to be wrong with the care providedfor individuals with dementia in many nursinghomes.

Some of the complaints and concerns listed intable 1-1 apply particularly to residents with demen-tia, and others apply equally to nondemented resi-dents. To differentiate these two types of problems,OTA compared the most frequently cited complaintsand concerns about the care of nursing homeresidents with dementia, as listed in table 1-1, withthe problems identified by the Institute of Medicinein its 1986 report, Improving the Quality of Care inNursing Homes, which dealt with nursing home carefor all types of residents (318). This comparison,which is discussed in greater detail in chapter 2,shows that the complaints and concerns aboutnursing home care for residents with dementia focusmore on the physical aspects of nursing homes thatare perceived to be inappropriate for individualswith dementia (e.g., the lack of cues to help residents

find their way and the lack of appropriate space forresidents to wander) and the lack of staff knowledgeabout how to respond to behavioral symptoms. Incontrast, the Institute of Medicine report focusesmore on the lack of sufficient attention to residents’rights and the lack of choices for residents.

Both the Institute of Medicine’s report and theliterature on nursing home care for individuals withdementia cite the failure of many nursing homes tocreate a home-like environment and their failure toidentify and treat residents’ acute and chronicdiseases and conditions. Both sources also cite thelack of adequately trained staff in many nursinghomes. The Institute of Medicine’s report focuses onthe lack of training in general, whereas the literatureon nursing home care for individuals with dementiafocuses on the lack of training about dementia andthe care of residents with dementia.

Both the Institute of Medicine’s report and theliterature on nursing home care for individuals withdementia cite the overuse and inappropriate use ofpsychotropic medications and physical restraints.Although these two problems affect all nursinghome residents to some degree, they are more likelyto affect residents with dementia.

From 35 to 65 percent of all nursing homeresidents are prescribed and/or receive at least onepsychotropic medication? and 9 to 26 percent ofresidents are prescribed and/or receive more thanone such medication (18,19,52,366,425,429,433,461). Nursing home residents with dementia aremore likely than other nursing home residents toreceive these medications (19,389,425,429). Oftenthe medications are used to control behavioralsymptoms in residents with dementia, even thoughmany of the frequently used medications have notbeen demonstrated to be effective for this purpose(l8,l9,l80,208,277,285,339,381,389,397,406,414,425).Moreover, some of the most frequently used medica-tions are known to cause confusion, disorientation,and oversedation in older people and are likely toworsen the fictional impairments of individualswith dementia.

From 25 to 59 percent of all nursing homeresidents are physically restrained at any one time(133,446,520). Nursing home residents with demen-tia are far more likely than other nursing homeresidents to be physically restrained (133,389,446).

S Psycho@opic m~ications include antipsychotic, antidepressan~ antianxiety, and se&tive/hypnOtk agents.


Table 1-1—Frequently Cited Complaints and Concerns About the Care Provided forNursing Home Residents With Dementia

● Dementia in nursing home residents often is not carefully or accurately diagnosed and sometimes is notdiagnosed at all.

• Acute and chronic illnesses, depression, and sensory impairments that can exacerbate cognitive impairmentin an individual with dementia frequently are not diagnosed or treated.

•There is a pervasive sense of nihilism about nursing home residents with dementia; that is, a general feelingamong nursing home administrators and staff that nothing can be done for these residents.

• Nursing home staff members frequently are not knowledgeable about dementia or effective methods ofcaring for residents with dementia. They generally are not aware of effective methods of responding tobehavioral symptoms in residents with dementia.

• Psychotropic medications are used inappropriately for residents with dementia, particularly to controlbehavioral symptoms.

• Physical restraints are used inappropriately for residents with dementia, particularly to control behavioralsymptoms.

• The basic needs of residents with dementia, e.g., hunger, thirst, and pain relief, sometimes are not metbecause the individuals cannot identify or communicate their needs, and nursing home staff members maynot anticipate the needs.

• The level of stimulation and noise in many nursing homes is confusing for residents with dementia● Nursing homes generally do not provide activities that are appropriate for residents with dementia

• Nursing homes generally do not provide enough exercise and physical movement to meet the needs ofresidents with dementia.

● Nursing homes do not provide enouqh continuity in staff and daily routines to meet the needs of residentswith dementia.

• Nursing home staff members do not have enough time or flexibility to respond to the individual needs ofresidents with dementia.

. Nursing home staff members encourage dependency in residents with dementia by performing personal carefunctions, such as bathing and dressing, for them instead of allowing and assisting the residents to performthese functions themselves.

. The physical environment of most nursing homes is too “institutional” and not “home-like” enough forresidents with dementia.

• Most nursing homes do not provide cues to help residents find their way.

• Most nursing homes do not provide appropriate space for residents to wander.•o Most nursing homes do not make use of design features that could support residents’ independent

functioning.• The needs of families of residents with dementia are not met in many nursing homes.

SOUR=: ~lce of ‘I&bnolOgy Assessment 1992.

A study of restraint use in 12 Connecticut nursing of bone and muscle mass and other physiologicalhomes found, for example, that 51 percent of the effects of immobility; increased agitation; aggra-disoriented residents were newly restrained over the vated behavioral symptoms, such as screaming,l-year course of the study, compared with only 17 hitting, and biting; decreased social behavior; loss ofpercent of the residents who were not disoriented self-esteem; emotional withdrawal; and injuries and(446). The potential negative effects of physical death due to improper use of the restraints andrestraint use for both demented and nondemented residents’ attempts to escape from them (30,133,residents include the following: incontinence; loss 139,182,208,300,305,383,427,446,490,498).

14 ● Special Care Units for People with Alzheimer’s and Other Dementias

Box I-B—The Development of Excess Disability in a Nursing Home Resident With Dementia

One evening an elderly man with dementia who had recently been admitted to a nursing home was pickingup his newspaper at the receptionist’s desk Abruptly, he threatened to hit the receptionist with his cane if she didnot call him a cab, so he could “go to town.’ The receptionist contacted the nurses’ station and kept the man talkinguntil help arrived. Three staff members responded. They attempted to calm the man verbally, but when theseattempts failed, they snatched the cane and forcefully placed him in a “geri-chair.” He was wheeled to his room,yelling and kicking. Several visitors and other residents stood by, wide-eyed, watching this scene.

A negative pattern developed with the new resident. He did well during the day with minimal assistance, butevery evening he became very confused, agitated, and disruptive. The nursing home staff met with his family, andthe family agreed to visit him each evening for a few weeks, until he adjusted to the new environment.

Several weeks passed, the agitation and confusion continued, and the family requested sedation, in part becausethey were embarrassed about his behavior. An antipsychotic medication was prescribed. Different dosages andadministration times were tried to determine a therapeutic level. Several more weeks passed. The resident becameless disruptive, but he also began to walk unsteadily, drool, and slur his words. He became incontinent, and he couldno longer dress himself.

SOURCE: Adapted from M. Bowsher, “A Unique and Successful Approach to Care for Moderate Stage Alzbeimer’s Victims,” Green HillsCenter, West Liberty, OH, unpublished manuscript no date.

Overuse and inappropriate use of psychotropic functional impairments that are caused by his or hermedications and physical restraints are problems in dementing disease or condition and other functionalthemselves. They are also perceived by special careunit advocates and others as manifestations of otherproblems in the nursing home care provided forindividuals with dementia—notably the failure ofmany nursing homes to use more appropriatemethods of responding to the individuals’ physicaland emotional needs and behavioral symptoms.

Reduction in the use of psychotropic medicationsand physical restraints is a major objective of manyspecial care units. Evidence cited later in this chapterand discussed at greater length in chapter 3 indicatesthat in general special care units have been success-ful in reducing the use of physical restraints but thatuse of psychotropic medications is as high or higherin special care units than in nonspecialized units.

Negative Consequences for Nursing HomeResidents With Dementia, Their Families,

Nursing Home Staff Members, andNondemented Nursing Home Residents

Problems in the care provided for nursing homeresidents with dementia have many negative conse-quences for the residents. These negative conse-quences include reduced quality of life, reducedphysical safety, and excess disability. The termexcess disability refers to functional impairment thatis greater than is warranted by an individual’sdisease or condition (47,219). The concept of excessdisability implies that an individual has certain

impairments that are caused by other factors. Thelatter impairments constitute excess disability.

Inappropriate or poor-quality nursing home carecan lead to excess disability in cognitive function-ing, mood, activities of daily living, and behavior.Box 1-B illustrates the development of excessdisability in a nursing home resident with dementia.The immediate cause of excess disability in this casewas a psychotropic medication. Box 1-C later in thischapter describes an alternate set of staff responsesin the same situation that solved the problem andavoided the use of psychotropic medications and theexcess disability.

In practice, it is often difficult to distinguishfictional impairments caused by an individual’sdementing disease or condition and functionalimpairments caused by inappropriate or poor-qualitynursing home care. Many commentators contend,however, that some and perhaps many of thefunctional impairments of nursing home residentswith dementia are due to problems in the care theyreceive rather than to their dementing disease orcondition (107,1 15,125,165,171 241,263,359,385,386).

Problems in the nursing home care provided forindividuals with dementia have negative conse-quences for the residents’ families. Many families ofindividuals with dementia feel intensely guilty,anxious, and sad about having to place the individualin a nursing home. These feelings may be due

Chapter I-Overview and Policy Implications ● 15

primarily to the patient’s condition and other factorsthat have made nursing home placement necessary,but the feelings are intensified if the family believesthe individual is receiving inappropriate or poor-quality care (84,162,263). In addition, the failure ofmany nursing homes to facilitate and supportfamilies’ ongoing involvement in their relative’scare may result in the development of a competitiveor adversarial relationship between the staff and thefamily which further increases the family members’anxiety (45,50,55,167,349,418).

Problems in the care provided for individuals withdementia also have negative consequences for nurs-ing home staff members. Residents with dementiaare often difficult for staff members to care forbecause of their communication deficits, impair-ments in activities of daily living, and behavioralsymptoms (60,107,167,170,181,191,263,352,359,385).The difficulty of caring for residents with dementiais said to cause stress, lowered morale, and burnoutfor staff members (191,263,346,352,398). Thesereactions may in turn lead to increased absenteeismand staff turnover. To the extent that residents’impairments are caused or exacerbated by inappro-priate or poor-quality care, the job of staff membersis unnecessarily difficult, and any resulting stress,absenteeism, and turnover are also attributable to theinappropriate care.

Lastly, nondemented nursing home residents mayexperience negative consequences because of prob-lems in the care provided for residents with demen-tia. Behavioral symptoms of residents with demen-tia, e.g., restlessness, screaming, repetitive verbali-zations, and combativeness, are upsetting for nonde-mented residents (46,220,241,263,268,352,373). Thecognitive and fictional impairments of residentswith dementia may also be upsetting for nonde-mented residents. Experts disagree about the overallimpact on nondemented nursing home residents ofliving in close proximity to residents with dementia,but the two studies OTA is aware of that address thisissue found significant negative effects for thenondemented residents (438,507). In a study of 72nondemented nursing home residents, Teresi et al.found that the nondemented residents who shared aroom or had a room adjacent to a demented residentwere significantly more likely than the other nonde-mented residents to express dissatisfaction with theirlife and their environment and to be perceived asdepressed by staff members (438). They were also

significantly less likely to receive visits or phonecalls from family or friends.

It is unclear whether the negative effects onnondemented nursing home residents of living inclose proximity to residents with dementia are dueprimarily to characteristics of the demented resi-dents that are caused by their dementing illness or tocharacteristics that are caused by inappropriatenursing home care. To the extent that the negativeeffects are due to characteristics caused by inappro-priate care, the inappropriate care is also responsiblefor the reduced quality of life of the nondementedresidents.

Special care units promise to provide betternursing home care than is currently available forindividuals with dementia. By providing better care,they expect to benefit residents, residents’ families,and nursing home staff members. Better care canonly reduce impairments that are not inevitablycaused by the residents’ dementing disease orcondition. Likewise, better care for residents canonly alleviate that portion of family members’feelings of guilt, anxiety, and sadness that is due toinappropriate care, not the portion of those feelingsthat is caused by the residents’ impairments ordeteriorating condition. Similar considerations applyto the potential impact of better care on nursinghome staff members. Research findings with respectto the effect of special care units on residents,families, and nursing home staff members should beconsidered in the context of these inherent limita-tions on potential positive outcomes.

The situation is different for nondemented nurs-ing home residents. Placing demented residents inseparate units eliminates for nondemented residentsthe negative effects of living in close proximity withdemented residents regardless of the factors thatcause the negative effects. Some commentatorsbelieve that placing individuals with dementia inphysically separate units may be justifiable solely onthe grounds that it benefits nondemented residents,assuming the placements do not harm the dementedresidents (221,356).

SPECIAL CARE UNITSThe first special care units in this country were

established in the mid 1960s and early 1970s(413,485,494). In the mid to late 1970s and the firsthalf of the 1980s, interest in specialized nursinghome care for individuals with dementia grew


rapidly because of increasing general awareness ofAlzheimer’s disease and the special needs of nursinghome residents with dementia (273). In this period,some nursing homes established special care units.6

Other nursing homes established special activityprograms for their residents with dementia.7

Reports on these early special care units andprograms reflect each facility’s search for workableapproaches in caring for individuals with dementia(273). The reports are primarily descriptive. Many ofthem include case examples that illustrate thebehavioral and other resident problems the unit wasdesigned to address.

Much of the literature on special care unitsconsists of descriptive reports of this kind. Thesereports generally cite one or more theoretical con-cepts as the rationale for the physical design featuresand patient care practices that have been imple-mented in a particular unit and make that unit specialin the view of the report authors. Many of the reportsalso provide nonquantitative, anecdotal evidence ofthe beneficial outcomes of the unit.

Reports on early special care units do not suggestmarketing interests, but some recent reports doreflect such interests. In the past few years, marketdemand has clearly become an important factor inthe establishment of special care units (273).

This section discusses the theoretical concepts ofspecialized dementia care that are frequently cited inthe special care unit literature. It briefly describesseveral ideas about special care units from othercountries that have influenced the development ofspecial care units in this country. Lastly, it summa-rizes the findings from the available descriptive andevaluative studies of special care units.

Six Theoretical Concepts of SpecializedDementia Care and Their Implications forStaff Composition and Training and the

Individualization of Care

Six interrelated concepts pervade the literature onspecial care units. The six concepts are discussed atsome length in this report because OTA’s review ofthe literature on special care units and discussions

with experts on dementia care indicate that theseconcepts constitute the core of what is or should bespecial about special care units, more so than anyparticular physical design features or other charac-teristics of the units. Although experts disagreeabout particular physical design features and otherspecial care unit characteristics, there appears to beconsiderable agreement about the concepts.

The six theoretical concepts apply to the care ofindividuals with dementia generally and are notlimited to special care units or even to nursing homecare. One or more of the concepts are cited invirtually all articles and books about special careunits, although few sources cite them all. Theconcepts are often used to explain and justify theparticular physical design features and patient carepractices used in a given special care unit orrecommended for special care units generally. Theconcepts also have important implications for staffcomposition and training and the individualizationof care.

1. Something can be done for individuals withdementia.

This concept argues against the pervasive nihil-ism that has characterized the care of individualswith dementia. It posits instead that even thoughmost of the diseases and conditions that causedementia are incurable at present, some aspects ofdementia are treatable, and treatment will improvethe individual’s functioning and quality of life(91,125,165,268,353,364,371,403). The other fiveconcepts discussed in this section can be thought ofas ways of operationalizing the first concept. Acorollary to the first concept that is implicit in muchof the special care unit literature but explicitly statedby only a few commentators is the value judgmentthat individuals with dementia have a right to carethat improves their functioning and quality of lifeeven if the disease or condition that causes theirdementia is irreversible and progressive (33,66,170,399).

2. Many factors cause excess disability inindividuals with dementia. Identifying andchanging these factors will reduce excess

6For ~xmple, of ~peci~ ~me ~~ ~~~bli~hed ~ ~~ ~eri~d, see Berger (27), Bl~en~ Jewish Home (32), Bohg md Bohg (34), Bowsher(38), Bnce (44), Clarke (87), Goodman (158), Grossman et al. (163), Kromm and Kromm (234), Liebowitz et al. (253), Peppard (345), Wallace (478),and Wilson and Patterson (505).

T See, for ~wple, H~c~k ~d Ba~ (173), Johnson and Chapman (21 1), McGrowder-Lin and Bhatt (299), Sawer ~d Mendolovi~ (400)sand Schwab et al. (403).


disability and improve the individuals’functioning and quality of life.

As discussed earlier, excess disability is fuc-tional impairment that is greater than is warranted byan individual’s disease or condition (47,219). Ex-cess disability in individuals with dementia can becaused by untreated acute or chronic illnesses,depression, and sensory impairments; overuse orinappropriate use of psychotropic or other medica-tions or physical restraints; excessive environmentalnoise; lack of stimulation and exercise; inappropri-ate caregiver responses to individuals’ behavioralsymptoms, and other factors. The literature onspecial care units contains numerous examples ofsituations in which changing a factor that wascausing excess disability resulted in dramatic im-provement in an individual’s functioning and qualityof life.

3. Individuals with dementia have residualstrengths. Building on these strengths willimprove their functioning and quality of life.

Although individuals with dementia are usuallydescribed in terms of their impairments, even thosewith severe impairments have residual strengths andabilities (125,328,353,399,519). It has been noted,for example, that some individuals with dementiawho are no longer able to speak coherently can stillsing, and some can remember the words to old songs(295,487,491). By building on this strength, musicprograms and music therapy are intended to improvethese individuals’ quality of life and allow them tointeract on some level with other people.

Another example of the implementation of thisconcept is the use of familiar activities. Manyindividuals with dementia remember how to do tasksthey did earlier in their lives. Activities such ascooking and laundry-folding for women and wood-working for men are intended to build on theseremaining abilities and give the individuals a feelingof competence (108,518).

4. The behavior of individuals with dementiarepresents understandable feelings and needs,even if the individuals are unable to expressthe feelings or needs. Identifying andresponding to those feelings and needs willreduce the incidence of behavioral symptoms.

The behavior of individuals with dementia isfrequently regarded as an inevitable and essentiallymeaningless consequence of their dementing dis-

ease or condition, and little effort is made tounderstand or explain it. In contrast, experts indementia care point out that the behavior of individ-uals with dementia often expresses meaningfulfeelings, intentions, and needs (60,125,273,287,353,361,385,403,408,482,517). They contend that ifnursing home staff members and other caregiverscan figure out the meaning of the individuals’behavior and respond to that meaning, the caregiversmay be able to prevent or resolve behavioralsymptoms without resorting to psychotropic medi-cations or physical restraints. Box 1-C describes thesame elderly man with dementia who is described inbox 1-B and illustrates the way in which interven-tions based on an understanding of the meaning of anindividual’s behavior may prevent the developmentof behavioral symptoms and avoid the use ofpsychotropic medications and physical restraints.The special care unit literature contains manysimilar accounts.

The first efforts to explain specific behavioralsymptoms in individuals with dementia focused onwandering. Beginning in the 1970s, several re-searchers have studied wandering behavior andconcluded that although the behavior often seemsmeaningless on the surface, it actually represents avariety of meaningful intentions and needs fordifferent individuals (e.g., a search for someone orsomething, a search for security, a wish to go home,or a lifelong coping style) (106,306,361,417). Basedon this conclusion, a number of innovative andreportedly effective methods of responding to wan-dering behavior have been developed.

Two books-Care of Alzheimer’s Patients: AManual for Nursing Home Staff (165) and Under-standing Difficult Behaviors (385)-discuss themany possible reasons for behavioral symptoms andsuggest ways of responding to the problems basedon these reasons. Both books recommend andexemplify a flexible, problem-solving approach tobehavioral symptoms. Other commentators havealso noted that responding effectively to the behav-ioral symptoms of individuals with dementia ofteninvolves a flexible, trial and error approach (353,399,516).

Rader refers to wandering and other behaviors ofindividuals with dementia as agenda behavior; thatis, behavior by which a person with dementiaattempts to meet his or her own agenda (359,361).She urges caregivers of individuals with dementia to


Box 1-C—The Use of Behavioral Interventions With a Nursing Home Resident With Dementia

One evening an elderly man with dementia who had recently been admitted to a nursing home was pickingup his newspaper at the receptionist’s desk. Abruptly, he threatened to hit the receptionist with his cane if she didnot call him a cab, so he could “go to town, ’ The receptionist stood up, looked directly at the resident and said ina respectful, matter-of-fact tone, “I see something is bothering you.’ The resident answered in a low, harsh voice,“I should be working, not being lazy.” The receptionist asked him about his work and listened intently as he talkedabout the work he used to do.

A pattern developed with the new resident. He did well during the day with minimal assistance, but everyevening he became very confused and agitated. A nurse aide was assigned to take a walk with him at these times.As they walked together around the facility, they often talked about the past and the resident’s busy professionallife. Sometimes they just walked. When the resident showed sorrow, the nurse aide shared the sorrow with him byactive listening and gently touching him on the arm.

Several weeks passed, The resident became less agitated and more content to wander around the unit,sometimes stopping to take imaginary measurements of a doorway or a piece of furniture. The intervention of thefamiliar nurse aide prevented the development of a behavioral problem that might have led to the use of psychotropicmedications or physical restraints.

SOURCE: Adapted from M. Bowsher, “A Unique and Successful Approach to Care for Moderate Stage Alzheimer’s Victims,” Green HillsCenter, West Liberty, OH, unpublished manuscrip~ no date.

try to understand the agenda that underlies theindividual’s behavior and to allow the individual toplay out that agenda as much as possible, rather thansuperimposing the caregiver’s own agenda.

On the basis of the concept that the behavior ofindividuals with dementia represents understand-able feelings and needs, Feil and others advocate theuse of validation therapy (120,136,407). Validationtherapy involves understanding and validating thepersonal meaning of an individual’s behavior. It isan alternative to reality orientation, a therapymethod which requires the caregiver to consistentlyreorient the confused person to current reality. Manycommentators contend that reality orientation isfrustrating and usually ineffective for individualswith dementia, except perhaps early in the course oftheir dementing disease or condition (120,170,273,359,361, 436,483).

5. Many aspects of the physical and socialenvironment affect the functioning ofindividuals with dementia. Providingappropriate environments will improve theirfunctioning and quality of life.

The relationship between the environment and thefunctioning of older people has been the topic ofempirical research and theory-building in environ-mental psychology for 30 years (183,242). It is nowgenerally accepted that the interaction between anolder person’s environment and the person’s charac-

teristics can affect his or her functioning, eitherpositively or negatively. According to Lawton:

The quality of the outcome of a person-environment transaction is a function of the degreeof environmental demand or press. . and the compe-tence of the person. When the degree of demand ismatched to the person’s competence, a positiveoutcome in terms of affective response or adaptivebehavior is the rule. When press is high in relation tocompetence, psychological disturbance in the formof strain is likely to occur. When press is low inrelation to competence, sensory deprivation andatrophy of skills are likely (243).

In this theory, the terms environmental demandand environmental press refer to the motivating oractivating quality for a particular individual of thephysical and other aspects of that individual’senvironment (242). The term person-environment fitdenotes the degree of congruence between environ-mental demand or environmental press and the needsand characteristics of an individual. The theoryproposes that person-environment fit can be im-proved by changing the environment (218,242).

The theory also proposes that the impact of theenvironment is greater for individuals with lowcompetence, including individuals with dementia,than for other people. According to Lawton:

As individual competence decreases, the environ-ment assumes increasing importance in determiningwell-being. One corollary of this hypothesis is that


the low-competent are increasingly sensitive tonoxious environments. The opposite and morepositive corollary is that a small environmentalimprovement may produce a disproportionate amountof improvement in affect or behavior in the low-competent individual (241).

The concept that appropriate environments willimprove the functioning and quality of life ofindividuals with dementia appears frequently in thespecial care unit literature. In the context of thetheory, the term environment includes all aspects ofa person’s surroundings, but the concept is citedmost often in connection with physical aspects of theunits. Many articles and books that discuss thedesign of special care units identify one or moreimpairments or needs of individuals with dementiaand propose physical design features to compensatefor or respond to the impairments or needs. Twobooks exemplify this approach: Designing for De-mentia: Planning Environments for the Elderly andConfused (67) and Holding Onto Home: DesigningEnvironments for People With Dementia (93).

Physical design features are seen as potentiallycompensating for or responding to the impairmentsand needs of individuals with dementia in thefollowing general ways:

. by assuring safety and security;

. by supporting functional abilities;

. by assisting with way-finding and orientation;

. by prompting memory;

. by establishing links with the familiar, healthypast;

. by conveying expectations and eliciting andreinforcing appropriate behavior;

. by reducing agitation;● by facilitating privacy;. by facilitating social interactions;. by stimulating interest and curiosity;. by supporting independence, autonomy, and

control; and. by facilitating the involvement of families

(62,67,93,184).

Many different physical design features are justi-fied on the basis of the concept that appropriateenvironments will improve the functioning andquality of life of individuals with dementia. Thesedesign features range from the overall shape andfloor plan of the unit (see fig. 1-3) to the use ofenvironmental cues, such as color coding of roomsand corridors to help residents find their way, and

personal markers, such as residents’ pictures placednear their rooms to help them identify the rooms.

Physical design features are often referred in thespecial care unit literature as prosthetic because theyare intended to compensate for, rather than cure,impairments that are believed to be unchangeable.Since the impairments are unchangeable, it isassumed the prosthetic features will be neededpermanently. Physical design features that compen-sate for functional impairments are said to be costeffective because the design features act continu-ously and may substitute for more costly staffinterventions (185,243).

Sometimes very strong claims are made aboutparticular physical design features for special careunits, as if there were proof of the effectiveness orlack of effectiveness of the features. Numerousarticles state with certainty, for example, that floorpatterns with dark areas or dark borders should notbe used in special care units because individualswith dementia will perceive the dark areas as holesand be afraid to walk on or over them. Likewise it isoften said that certain types of art work, wallpaper,and carpet patterns cause delusions and hallucina-tions in nursing home residents with dementia. ToOTA’s knowledge, there is no research-based evi-dence for these claims.

OTA has heard particular physical design featuresjustified on the basis of claims, such as thatindividuals with dementia may mistake a lightreflected from a shiny floor as a blob that is chasingthem, that they feel threatened by the person in themirror who does not respond to their greeting, thatthey sometimes mistake their shadows for pools ofwater and try to jump over, that they try to pick theflowers in floral-print wallpaper, etc. One suspectsthat these claims arise from anecdotes about individ-ual residents or someone’s guess about the responseof individuals with dementia to a particular designfeature and that the anecdotes and guesses are thengeneralized to all residents with dementia.

In reality, very little research has been done to testthe impact of particular physical design features onindividuals with dementia. Moreover, the conclu-sions of several of the existing studies are contradic-tory. Some of these studies are described in chapter4. Unfortunately, some nursing homes incorporatephysical design features for which strong claims aremade and believe they have thereby created anappropriate environment for their residents with

John Douglas French Center, Los Angeles, CA

The building is structured in a “butterfly” shape with 4 units maintainingrooms for “families” of 12-13 residents located around a shared nurses’station. Each family unit includes a mix of private and semi-private rooms.There is direct access to a secure courtyard.

IJ 1

-

d

Weiss Institute, Philadelphia Geriatric Center, Philadelphia, PA

The unit is comprised of a large central space, around which residents’rooms are located. The open plan of the 40-bed unit allows staff easy visualaccess to all residents and provides a continuous path for wanderers. Theunit has a therapeutic kitchen for residents.

Figure 1-3—Alternate Shapes and Floor Plans for Special Care Units

Corinne DoIan Alzheimer’s Center, Heather Hill, Chardon,OH

The building is comprised of 2 triangular units with a shared supportand bathing core. The open plan of each 12-bed unit allows staff easy visualaccess to all residents, and provides a continuous path for wanderers. Eachunit has a fully equipped residential-style kitchen. There is direct access to asecure courtyard, as well as to several paved paths beyond the yard forresidents and visitors.

n

Friendship House, West Bend, IN

The building is comprised of 2 units with 4 “households” each. A nurses’station, elevator and services are located at the center of each unit of4 households. A protected outdoor courtyard is defined by the two units.

SOURCE: M.P. Calkins, Design for Dementia: Planning Environments for the Ekfetfyand Confused, 1988; U. Cohen and G.D. Weisman, Holding On To Home, 1991; U. Cohen and G.D. Weisman,Environments for People Wth Dementia: Case Studies, 1988.


dementia, when, in fact, no evidence exists that thespecifc features are effective. Lawton has notedthat:

There is a strong tendency for intuitive, a priorireasoning about what is “good” for Alzheimerpatients to become accepted as fact. . .The hunger forinformation is so great among practitioners thatalmost any unsupported assertion can be rapidlyaccepted (244).

As noted earlier, the concept that appropriateenvironments will improve the functioning andquality of life of individuals with dementia is citedmost often in connection with physical designfeatures for special care units, but it is sometimesalso cited in connection with other unit characteris-tics, such as activity programs and daily routine.Activity programs and the daily routine on the unitare perceived as potentially compensating for theimpairments of residents with dementia in manyways, e.g., by supporting functional abilities, prompt-ing memory, conveying expectations, eliciting andreinforcing appropriate behavior, facilitating socialinteractions, and stimulating interest and curiosity(358,392,519).

Coons has gone farthest in developing a model ofspecialized dementia care, referred to as a therapeu-tic milieu, in which all aspects of the physical andsocial environment and the daily routine on the unitare designed to be therapeutic (104,105,109).8 Thismodel was demonstrated for several years at WesleyHall, a special care unit in a retirement facility inChelsea, MI.

A different model of care, referred to as a lowstimulus unit, has been developed by Hall and hercolleagues (170,171). This model is based on theconcept that appropriate environments will improvethe functioning and quality of life of individuals withdementia and the perception of these clinicians thatindividuals with dementia have a “progressivelylowered threshold for stress” due to their reducedability to receive and process external stimuli. Halland others believe that in traditional nursing homeunits, residents with dementia are overwhelmed bymultiple environmental stimuli, including noisefrom telephones, televisions, radios, Muzak, andpaging systems; high-glare floors; hurrying staff;visitors; other residents; and large group activities.They believe that in response to these stimuli, the

residents become increasingly agitated, confused,and sometimes combative. To compensate for theresidents’ lowered threshold for stress, Hall and hercolleagues propose units in which environmentalstimuli are reduced: no telephones ring on the unit;television, radio, Muzak, and paging are eliminated;staff and visitor traffic through the unit is reduced;dining and activities take place in small groups; andresting is encouraged by environmental cues, such ascomfortable chairs in the hallways. Many lowstimulus units have been established on the basis ofthis model (169,209,334). While agreeing with someaspects of the low stimulus model, other cliniciansand researchers contend that the main problem is notexcessive stimuli, but insufficient stimuli of appro-priate types. They argue that an increase in selectedstimuli will improve the functioning and quality oflife of individuals with dementia (107,183,243,259,272). The ideal level and type of stimuli areunclear, however (96,185,244,287).

Like the other five concepts discussed in thissection, the concept that appropriate environmentswill improve the functioning and quality of life ofindividuals with dementia is theoretical. It is inter-preted differently by different individuals and isused to justify a great variety of physical designfeatures and other unit characteristics. Disagree-ments among experts about the right characteristicsfor a special care unit make it difficult for nursinghome administrators and others to design a specialcare unit. These disagreements do not, however,invalidate the underlying concept. Instead, theypoint out the need for research to test the effective-ness of the recommended characteristics.

6. Individuals with dementia and their familiesconstitute an integral unit. Addressing theneeds of the families and involving them in theindividuals’ care will benefit both theindividuals and the families.

Families of individuals with dementia are oftensaid to be the second victim of the dementia. Theyare generally perceived by experts in dementia careas part of the client unit. As a result, meeting theirneeds becomes a legitimate objective of specializeddementia care.

Families can also assist in various ways in the careof nursing home residents with dementia. They area source of valuable information about the residents,

g me Con=pt of therapeutic m-lieu was f~st used in the treatment of mentally ill persons in psychiatric hospitis (215).


who often cannot provide accurate informationabout themselves. As Hegeman and Tobin havenoted, families can ‘‘help to preserve the uniqueidentity of residents and help the staff and theresident be aware of that identity” (178). Familiescan also provide physical assistance, emotionalsupport, and advocacy. Their presence helps to makeany setting more home-like and familiar for theresident (174,296,358,418).

Meeting the needs of families of nursing homeresidents with dementia means providing them withinformation, emotional support, and a structure thatfacilitates their involvement in the residents’ care.Families are perceived to benefit from informationabout dementia and ways of communicating with aperson with dementia, as well as from supportgroups, counseling, and other forms of emotionalsupport (55,128,168,296,358,41 8,5 16).

To facilitate the involvement of families in theresidents’ care, it is necessary to provide both awelcoming atmosphere and administrative and care-giving practices that recognize the families’ legiti-mate role in the residents’ care. Families can beinvolved, for example, in care planning conferencesand other situations in which decisions are beingmade about the residents’ care. They may also beencouraged to act as volunteers on the unit (46,55,125,168,174,418).

By providing information, emotional support, anda structure that facilitates the involvement of fami-lies, it may be possible to lessen their feelings ofanxiety and guilt and avoid the development of acompetitive or adversarial relationship between thestaff and the families. Families differ, however, andthe best ways of providing information and supportand involving families also differ (128,168,358).

Implications for Staff Composition and Training

The six concepts discussed above have importantimplications for staff composition and training. Withrespect to staff composition, the concepts indicatethe need for a multidisciplinary approach to care. Toidentify and change the factors that cause excessdisability requires the involvement of health careprofessionals capable of diagnosing and treating thecauses of excess disability, e.g., acute and chronicillnesses, depression, and sensory impairments.Likewise, to provide activity programs that build onresidents’ residual strengths, support functionalabilities, and facilitate social interactions requires

the involvement of individuals who are skilled invarious therapeutic recreation specialties. Althoughthese health care professionals and other therapistsdo not necessarily have to be part of the unitstaff-and to make them part of the staff may beprohibitively expensive-some means of involvingthem in the residents’ ongoing care is essential foreffective implementation of the concepts.

With respect to staff training, the concepts requirea change for all staff members in widely heldnihilistic attitudes about nursing home residentswith dementia. In addition, since the concepts do notprovide precise formulas for care, staff membersmust not only understand the concepts but also beable to interpret and apply them in caregivingsituations. In most special care units, as in nursinghomes generally, nurse aides provide most of thedaily care. These aides must be able to interpret andapply the concepts—sometimes in difficult, emo-tionally-charged situations. To do so requires knowl-edge, problem-solving skills, and judgment. Specialcare units that adopt the concept of therapeuticmilieu often regard housekeepers and other nonprofes-sional staff members as part of the care team. Theseindividuals also must understand the concepts andbe able to apply them.

Implications for the Individualization of Care

Three of the six concepts clearly emphasize theindividualization of care. They require the staffmembers to: 1) identify and change the factors thatcause excess disability in individual residents; 2)identify and build on the residual strengths ofindividual residents; and 3) identify and respond tothe feelings arid needs expressed in the behavior ofindividual residents. As noted earlier, nursing homeresidents with dementia are diverse, and theircharacteristics and needs change over time. Thethree concepts that emphasize the individualizationof care fit well, at least in theory, with this diversity.

The concept that appropriate environments willimprove the functioning and quality of life ofindividuals with dementia may also fit well in theorywith the diverse and changing needs of nursinghome residents with dementia. In practice, however,the concept is probably more difficult to apply, sincespecial care units must be designed and built forgroups of individuals. The objectives in special careunit design include flexibility and the capacity toadapt to resident change (10,67,287,296,358). Ne-vertheless, given the extreme diversity of nursing

Chapter I-Overview and Policy Implications ● 23

home residents with dementia, it would seem thatthe more closely the physical environment of aspecial care unit matches the needs of one individualor one type of individual with dementia, the lesslikely the unit would provide the best environmentfor other types of individuals with dementia. Thesame concern may apply to other features of specialcare units, such as activity programs.

This concern has led a few nursing homes toestablish several special care units that providedifferent levels and types of care intended to matchthe characteristics and needs of residents in differentstages of their illness (34,473). A second alternative,adopted by some nursing homes with only onespecial care unit, is to discharge residents from theunit-usually to a nonspecialized unit in the samefacility-when the level and type of care provided inthe special care unit no longer matches the residents’characteristics and needs. Both these alternativesrequire moving residents, which is likely to increasetheir confusion. Moving residents also may havenegative consequences for the residents’ familieswho are often emotionally attached to the unit staffmembers and for the unit staff members who areoften attached to the residents and their families(40,375,473).

A third alternative is to allow special care unitresidents to age in place, that is, to remain on the unituntil they die. Anecdotal evidence suggests thatsome special care units that adopt this policybecome, in effect, terminal care settings as most ofthe residents progress into the later stages of theirillness (40,419). This creates problems for newresidents who are admitted to a unit in which mostof the other residents are severely cognitively andphysically impaired. OTA is not aware of anyresearch that compares these three alternatives, andthe special care unit literature contains little discus-sion of this important issue.

Ideas About Special Care UnitsFrom Other Countries

Special care units for people with dementia existin many other countries. Information about theseunits reaches the United States primarily throughreports from foreign visitors who are knowledgeableabout the special care units in their own countriesand through reports of Americans who have visited

the units in other countries. There are a fewdescriptive studies on special care units in particularcountries,9 but most of the available information isanecdotal. OTA is not aware of any formal researchcomparing special care units in different countries.

Information about special care units in othercountries influences thinking about special careunits in the United States in several ways. First,special care units in other countries demonstratealternate models of care. For example, a primaryobjective of special care units in some countries is toprovide a comfortable, home-like environment fortheir residents. These units have few rules andmaintain a flexible daily schedule that is responsiveto the habits and preferences of individual residents.In visiting these units, American observers havebeen impressed with their relaxed atmosphere andthe apparent contentment of the residents (273).Reports on special care units of this kind in othercountries create an incentive for the establishment ofsimilar units in this country.

Physical restraints are used less frequently or notat all in special care units in some other countries(273,498). The knowledge that restraints are lessoften used in other countries has been one incentivefor reducing their use in the United States.

Special care units in some other countries aremore able to innovate than special care units in theUnited States (273). Awareness of this differencecalls attention to the factors that encourage orconstrain innovation in different countries. One suchfactor is nursing home regulations. As discussed inchapter 6, nursing home regulations in the UnitedStates sometimes interfere with the implementationof innovative physical design and other features inspecial care units. Nursing homes are less tightlyregulated in most other countries and are thereforemore able to innovate. Public programs in manyother countries also make a less rigid distinction thanpublic programs in the United States between healthcare and social services, and the same publicprograms are more likely to pay for both types ofservices in other countries. As a result, there arefewer artificial barriers to the development of specialcare units that provide a mix of medical and socialservices. Lastly, public funding is more likely to beavailable for nonmedical residential care in othercountries than in the United States. When the same

9 see, for ex~ple, No- Severe Dementia: The Provision of Longstay care (330).


public programs pay for both medical and socialservices and public funding is available for nonmed-ical residential care, there is a strong financialincentive for government agencies to support thedevelopment of nonmedical residential care modelsthat are less costly than nursing homes. Since 1986,for example, the Australian government has pro-vided grants to stimulate the development of specialcare units in hostels as an alternative to nursinghomes for individuals with dementia (101).

Despite these advantages in other countries, nocountry has the answers with respect to special careunits or problems in the care of nursing homeresidents with dementia (273). Questions about theeffectiveness of various models and components ofcare are pervasive. Clinicians and researchers fromother countries frequently come to the United Statesin search of ideas about physical design features andpatient care practices for special care units. Ade-quately trained staff and sufficient funding are inshort supply everywhere.

Findings From Research onSpecial Care Units

Research on special care units is in an early stage,but some descriptive and evaluative studies havebeen conducted in the past few years. OTA’sconclusions from the available descriptive studiesare listed in table 1-2. The findings from thesestudies are discussed in detail in chapter 3, and someof the most important findings for policy purposesare reviewed in this section. The findings from theavailable evaluative studies are discussed in detail inchapter 4 and reviewed briefly in this section.

Number of Nursing Homes That Have a SpecialCare Unit

OTA estimates that in 1991, 10 percent of U.S.nursing homes had a special care unit. This numberincludes nursing homes that group some of theirresidents with dementia in physically distinct clus-ters in units that also serve some nondementedresidents.

As noted earlier, OTA’s estimate is based on thefindings of two recent studies. One of the studies-a1991 survey of all U.S. nursing homes with morethan 30 beds—found that 9 percent of the nursinghomes reported having either a special care unit ora special program for residents with dementia in aphysically distinct part of the facility (246). The

second study-a 1990 survey of all nursing homesin five northeastern States—found that seven per-cent of the nursing homes reported having a specialcare unit, and an additional five percent reported thatalthough they did not have a special care unit, theydid place some of their residents with dementia inphysically distinct groups or clusters in units thatalso served some nondemented residents (194).Thus, a total of 12 percent of the nursing homesreported using some method to physically groupresidents with dementia--either in a special careunit or a cluster unit.

The lack of an accepted definition of the termspecial care unit makes it difficult to developaccurate figures on the number and proportion ofnursing homes that have a special care unit. Thefigures cited above are based on self-report. Thefigures from the 1991 survey generally reflect theopinion of each nursing home administrator or othersurvey respondent about what a special care unit is.According to the researchers who conducted the1990 survey, however, some nursing homes thatplace residents with dementia in a physically sepa-rate unit and provide special services in the unit donot use the term ‘‘special care’ for these arrange-ments and therefore may not respond affirmativelyto a survey question about whether they have aspecial care unit (436). Surprisingly, the researchersalso found that in some nursing homes, the adminis-trator and the director of nursing disagreed aboutwhether the facility had a special care unit (194).

Some people believe the term special care unitshould mean more than just a physically separatespace and the nursing home’s claim that it provides‘‘special care. Depending on the additional criteriathat are used, some and perhaps many of the nursinghomes included in the figures just cited might not becounted as having a special care unit.

To OTA’s knowledge, the 1990 survey of allnursing homes in five northeastern States was thefirst to identify large numbers of nursing homes withcluster units. It is unclear whether cluster unitsshould be counted as special care units. Many of thecluster units identified in the 1990 survey incorpo-rated features that are recommended for special careunits (e.g., physical design features, special stafftraining, and family support groups), althoughcluster units were less likely than special care unitsto incorporate these features (194).


Table 1-2-Conclusions From Descriptive Studies of Special Care Units

Number of Nursing Homes That Have a Special Care Unit

• OTA estimates that in 1991, 10 percent of all nursing homes in the United States had a special careunit. In at least some States, this figure includes nursing homes that place some of their residents withdementia in “clusters” in units that also serve nondemented residents.

• The proportion of nursing homes that have a special care unit varies in different parts of the countryand in different States,

• Many nursing homes that do not have a special care unit are planning to establish one, and somenursing homes that have a special care unit are planning to expand the unit.

Characteristics of Nursing Homes That Have a Special Care Unit● Larger nursing homes are more likely than smaller nursing homes to have a special care unit.• As of late 1987, most nursing homes that had a special care unit were private, for-profit facilities. At

that time, multi-facility nursing home corporations owned about one-third of all the facilities that hada special care unit. There is no evidence, however, that ownership of special care units is dominatedby a small number of multi-facility nursing home corporations.

Characteristics of Special Care Units● Special care units are extremely diverse.• Most special care units have been established since 1983, although a few have been in operation for

20 to 25 years.• The goals of special care units differ. For some units, the primary goal is to maintain residents’ ability

to perform activities of daily living. Other units focus on maintaining residents’ quality of life,eliminating behavioral symptoms, or meeting residents’ physical needs,

. Most existing special care units were not originally constructed as special care units, and at leastone-fifth were neither originally constructed nor remodeled for this purpose.

• The use of specific physical design and other environmental features varies in existing special careunits. Many of the physical design and other environmental features cited as important in the specialcare unit literature are used in only a small proportion of special care units.

• The most extensively used environmental feature in special care units is an alarm or locking system,found in more than three-fourths of existing units.

• On average, special care units probably have fewer residents than nonspecialized nursing home units.

• On average, special care units probably have more staff per resident than nonspecialized nursing homeunits.

• Although the majority of existing special care units provide special training for the unit staff, at leastone-fourth of existing units do not.

• Less than half of existing special care units provide a support group for unit staff members.

● The types of activity programs provided by special care units vary greatly, but existing special careunits are probably no more likely than nonspecialized units to provide activity programs for theirresidents.

* About half of existing special care units provide a support group for residents’ families.

● Special care unit residents areas likely or more likely than other nursing home residents with dementiato receive psychotropic medications.

• Special care unit residents we probably less likely than other nursing home residents with dementiato receive medications of all types.

(Continued on next page)


Table l-2-Conclusions From Descriptive Studies of Special Care Units-(Continued)

● Special care unit residents are less likely than other nursing home residents with dementia to bephysically restrained.

o Special care units vary greatly in their admission and discharge policies and practices. About half ofall special care units admit residents with the intention that the residents will remain on the unit untilthey die.

● The cost of special care units varies depending on the cost of new construction or remodeling, if any,and ongoing operating costs. On average, existing special care units probably cost more to operate thannonspecialized nursing home units, primarily because of the higher average staffing levels on specialcare units.

• Special care units generally have a higher proportion of private-pay residents than nonspecializednursing home units, and the private-pay residents are often charged more for their care in the specialcare unit than they would be in a nonspecialized unit.

Characteristics of Special Care Unit Residents

• Special care unit residents are younger than other nursing home residents, and they are more likelythan other nursing home residents to be male and white.

● Special care unit residents are more likely than other nursing home residents to have a specificdiagnosis for their dementing illness.

• Special care unit residents are probably somewhat more cognitively impaired and somewhat lessphysically and functionally impaired than other nursing home residents with dementia

● Special care unit residents are probably somewhat more likely than other nursing home residents withdementia to participate in activity programs.

* Special care unit residents are more likely than other nursing home residents with dementia to fall.

SOulmr!: CM%ce of ‘lk@nology Assessment, 1992.

In this context, it is interesting to note that the Because of this diversity, no single descriptivespecial care unit described in box 1-A at the statement is true of all special care units.beginning of this chapter is technically not a separateunit, because it does not have a nurses’ station andother features the State requires for a nursing homeunit. That unit is viewed by the facility’s administra-tors as a separate entity. A similar arrangement inanother nursing home might be viewed by itsadministrators as a clustering of residents withdementia in one section of a larger unit that alsoserves nondemented residents, and they might reportit as such on a survey questionnaire.

Characteristics of Special Care Units and SpecialCare Unit Residents

All studies of special care units show that existingunits are extremely diverse. They vary in their goals,physical design features, staff-to-resident ratios,staff training programs, provision of staff and familysupport groups, activity programs, use of psy-chotropic medications and physical restraints, andadmission and discharge policies and practices.

On average, special care units probably havefewer residents and more staff per resident thannonspecialized nursing home units (291). Staff-to-resident ratios vary greatly among units, however.

Most special care units provide special trainingfor their staff, but at least one-fourth of existing unitsdo not provide special training. In response to the1987 National Medical Expenditure Survey, 26percent of the nursing homes with a special care unitreported they did not provide special training for theunit staff (248). Likewise, in response to the 1990survey of all nursing homes in 5 northeastern States,30 percent of the facilities with a special care unitand 47 percent of the facilities with a cluster unitreported they did not provide special training for theunit staff (194). Given the emphasis on staff trainingin the special care unit literature, the finding thatmore than one-fourth of existing units do not providespecial training is surprising. The finding is proba-

Chapter l--Overview and Policy Implications ● 27

bly correct, however, since nursing homes areunlikely to underreport the provision of staff train-ing.

The most widely used physical design feature inspecial care units is an alarm or locking system,found in more than three-fourths of existing units(181,194,247). Although numerous physical designfeatures have been recommended for special careunits, most of the recommended features are used inonly a small proportion of existing units (194,485,494).

Some special care units have formal (written)admission and discharge policies, but most probablydo not (194). In response to the 1990 study of allnursing homes in five northeastern States, three-fourths of the facilities with a special care unitreported using each of three criteria to select theirresidents: 1) the degree of the individual’s dementia;2) the individual’s need for supervision; and 3) theindividual’s behavioral symptoms (194). Most of thefacilities reported that they seek individuals withmore rather than less severe behavioral symptoms,but 15 percent reported that they seek individualswith less severe behavioral symptoms for their unit.One-third reported that the individuals they admitmust be able to ambulate independently.

Reported admission practices may or may notreflect actual admission practices in special careunits. Findings from the Multi-State Nursing HomeCase-Mix and Quality Demonstration-a 5-yearcongressionally mandated study that includes spe-cial care unit residents among the 6800 nursinghome residents in the study sample-suggest thatthe major factor distinguishing special care unitresidents from individuals with dementia in nonspe-cialized nursing home units is the severity of theirphysical impairments (382). Data from a subsampleof 127 special care unit residents and 103 residentswith dementia in nonspecialized units in the samefacilities indicate that individuals with severe physi-cal impairments and physical care needs are lesslikely to be admitted to special care units than tononspecialized units. Once other variables werecontrolled, there was no significant difference inbehavioral symptoms between the special care unitresidents and the residents with dementia in thenonspecialized units.

About half of existing special care units admitresidents with the expectation that the individualswill remain in the unit until they die (194). Otherspecial care units admit residents with the expecta-

328-405 - 92 - 2 QL 3

tion that they will be discharged from the unit atsome time prior to their death. In the latter units, thereported reasons for discharge are: 1) that a residenthas become nonresponsive, physically abusive, orunable to ambulate independently; 2) that theresident needs intensive medical care; and 3) that theresident’s private funds are exhausted (194,485,492).

As noted in table 1-2, special care unit residentsare as likely or more likely than individuals withdementia in nonspecialized units to receive psy-chotropic medications (256,292,413). They are muchless likely to be physically restrained, however(256,292,391,413). A University of North Carolinastudy of 31 randomly selected special care units and32 matched, nonspecialized units in 5 States foundthat only 16 percent of the special care unit residentswere physically restrained at one point in time,compared with 36 percent of the residents withdementia in nonspecialized units (413).

Finally, five studies show that special care unitresidents are significantly more likely to fall thanother nursing home residents with dementia(99,265,292,497,521). In one study, the special careunit residents were not only more likely to fall butalso more likely to be hospitalized for a hip fracture(99). In another study, the increase in falls amongspecial care unit residents did not result in anincrease in injuries due to the falls (54). The greaterincidence of falls among special care unit residentshas received little attention thus far, in part becausethe relevant data from three of the studies have notyet been published. The reasons for the greaterincidence of falls are not known.

Costs, Charges, and Payment Methods

Very little information is available about the costof special care units. The cost of creating a specialcare unit obviously varies, depending on the extentof new construction or remodeling, if any. One studyof 12 nonrandomly selected special care units foundthat the reported costs for new construction andremodeling ranged from $4100 to $150,000 (275).Another unit was created for $1300, which coveredthe cost of an alarm system, color coding, and a fewother physical changes to the unit (70).

Most—but not all-special care units report thattheir operating costs are higher than the operatingcosts of nonspecialized units (70,477,485). Of 13nonrandomly selected special care units in Florida,for example, 7 reported that their operating costs


were higher than the operating costs of nonspecial-ized units in the same facility; 5 reported nodifference in operating costs, and one reported loweroperating costs (64).

The Multi-State Nursing Home Case-Mix andQuality Demonstration found that on average theamount of staff time spent caring for residents withdementia was greater in the special care units than inthe nonspecialized units in the study sample (143).The University of North Carolina study had similarfindings (413). The greater amount of staff timespent caring for special care unit residents undoubt-edly translates into higher average operating costs inthe special care units.

Many-but not all-nursing homes charge morefor care in their special care unit than in theirnonspecialized units (247,256,413,477,494). Mostspecial care units also have a higher proportion ofprivate-pay residents (292,413,477). It is the private-pay residents who are charged more for their care ina special care unit than they would be in anonspecialized unit. To OTA’s knowledge, nopublic program currently pays more for care in aspecial care unit than in a nonspecialized nursinghome unit.

According to preliminary data from the 1991survey of all U.S. nursing homes with more than 30beds, about half the nursing homes with special careunits charged their private-pay residents more in aspecial care unit than the residents would have beencharged in a nonspecialized unit in the same facility(246). The excess charge averaged $9.24 a day andranged from $1 to $83 a day.

Effectiveness of Special Care Units

OTA is aware of 15 studies that evaluate theeffectiveness of special care units for residents anda few additional studies that evaluate the effective-ness of special care units for residents’ families andunit staff members. These studies are discussed indetail in chapter 4.

Nine of the 15 studies did not use a control group(22,24,56,88,160,171,245,297,312). Each of thesestudies found some positive outcomes. The positiveoutcomes vary from one study to another, and someof the studies’ findings are contradictory. Excludingthese contradictory findings, the positive residentoutcomes found in more than one of the nine studiesare decreased nighttime wakefulness, improvedhygiene, and weight gain. A few of the studies found

improvements in the important areas of residents’ability to perform activities of daily living andresidents’ behavioral symptoms, but an equal num-ber of studies did not find such improvements.

All nine studies suffer from one or more methodo-logical problems that could affect the validity oftheir findings. One such problem is small samplesizes: 6 of the 9 studies had fewer than 12 subjects.Another methodological problem is inadequate re-search design and implementation. Some of thestudies are more like descriptive reports than rigor-ous research from which valid conclusions can bedrawn; in these studies, the outcomes are not clearlydefined, and the measurement process is moreimpressionistic than objective or standardized. Onlyfour of the nine studies report the statistical signifi-cance of their findings. Lack of control groups isanother methodological problem, since without acontrol group, the impact of the special care unitcannot be separated from the impact of other factorsthat may affect resident outcomes. Finally, many ofthe studies were conducted by unit staff members orother individuals who were involved in planning oradministering the unit. These individuals have anobvious interest in finding positive outcomes. Thepotentially powerful effect of their expectations,coupled with small sample sizes, lack of a rigorousresearch design, and lack of control groups mean thestudies’ results—both positive and negative-arequestionable.

Six of the 15 studies evaluating the effectivenessof special care units for their residents used a controlgroup. Four of the six studies with a control groupfound no statistically significant positive residentoutcomes that could be attributed to the special careunits (80,99,195,489). The resident outcomes meas-ured in one or more of these four studies werecognitive functioning, ability to perform activities ofdaily living, mood, behavioral symptoms, and rate ofhospitalization.

Two of the six studies with a control group foundpositive resident outcomes. One study found thatover a l-year period, 14 residents of one special careunit declined significantly less than 14 residentswith dementia in nonspecialized units of the samefacility in their ability to perform activities of dailyliving (392). The other study found that 13 residentsof one special care unit exhibited significantly fewercatastrophic reactions than 9 residents with demen-tia in nonspecialized units of the same facility (265).

Chapter 1--Overview and Policy Implications . 29

In the latter study, the special care unit residents alsointeracted significantly more with staff members,but there was no effect of the unit on the residents’ability to perform activities of daily living.

The samples for the six studies that used a controlgroup are larger than the samples for the nine studiesthat did not use a control group. Their researchdesign and implementation are more rigorous, andthe study outcomes are more precisely defined andmeasured. Use of a control group also increases thepresumed validity of their findings. On the otherhand, each of the studies has one or more methodo-logical problems that could affect the validity of itsfindings. Although the study samples are, on aver-age, larger than the study samples in the nine studiesthat did not use a control group, some of the samplesare still quite small. Selection bias is anotherproblem that could affect the validity of the studies’findings. If the special care unit residents and thecontrol group subjects differed in significant ways atthe start of the studies, these differences, rather thanthe impact of the special care unit, could account forthe observed outcomes. Randomization of subjectsto the special care unit or control group would be theideal way to address this problem, but familypreferences, subject attrition, and other factorsinterfered with randomization in one of the twostudies in which it was attempted (265,489). Othermethodological problems that could affect the valid-ity of the studies’ findings are discussed in chapter4.

Four studies evaluate the effect of special careunits on the unit staff over time. Three of thesestudies found no statistically significant effects(81,88,195). The fourth study found a significantreduction in stress among 15 special care unit staffmembers and a significant difference on one of threeindicators of burnout between the 15 special careunit staff members and 49 staff members onnonspecialized nursing home units (265). This studyalso found a statistically significant improvement inthe scores of the special care unit staff members onone of six indicators of job satisfaction. The studyfound no other significant effects of the special careunit on staff stress, burnout, or job satisfaction.

Three studies measured staff knowledge aboutdementia (81,88,265). In each of the studies, thespecial care unit staff members received trainingabout dementia. None of the studies found anystatistically significant effect of the training on the

special care unit staff members’ knowledge aboutdementia (see ch. 4).

Four studies evaluate the effect of special careunits on residents’ families over time. Two of thefour studies found no statistically significant effects(76,265). One of the remaining studies found asignificant increase in family members’ satisfactionwith the care provided for their relative withdementia over the 3-month period after the individ-ual was admitted to a special care unit (88). Theother study found a significant reduction in familymembers’ feelings of anxiety, depression, guilt, andgrief after their relative with dementia was admittedto a special care unit (489). One descriptive studyfound that families of special care unit residentswere significantly more likely than families ofresidents with dementia in nonspecialized nursinghome units to visit their relative regularly (413). It isnot clear whether the latter finding is attributable tothe effect of the special care units or to preexistingdifferences between the two groups of families,however.

A few of the 15 evaluative studies had negativefindings. Maas and Buckwalter report a trend forindividuals with dementia to become more activeafter being admitted to a special care unit (265). Thisincreased activity includes both positive behaviors,such as interacting with staff members, and negativebehaviors, such as noisiness, restlessness, and scream-ing. Bullock et al. found an increase in verbal abuseand resistiveness over time among the special careunit residents they studied (56).

Insummary, only two of the six evaluative studiesthat used a control group found any positive residentoutcomes. Only one of the four studies that evalu-ated the effect of special care units on the unit stafffound any positive outcomes, and only two of thefour studies that evaluated the effect of special careunits on residents’ families found any positiveoutcomes. For most outcomes, the positive findingsof one study are contradicted by the findings of otherstudies. Moreover, some of the statistically signifi-cant positive findings in these studies are relativelytrivial, and a few of the studies had negativefindings.

The limited positive findings in some of theseevaluative studies and the lack of positive findingsin other studies are surprising. After reporting thelack of positive findings in a study of families of


special care unit residents, one researcher com-mented:

Finally, I am left trying to reconcile these results,showing no special care unit superiority, with thepalpable sense of excitement, of mission, and ofrelief that the special care unit families, but not theother families, show (76).

This comment mirrors the response of manyresearchers and others to whom OTA has spoken inthe course of this study: that is, surprise that theevaluative studies conducted thus far generally donot show the positive outcomes they expected to findand thought they had observed informally.

Methodological problems may account in part forthe failure of some of the studies to find positiveoutcomes. Small sample sizes are a particularproblem because studies with very small sampleslack the statistical power to detect small, butclinically significant, positive outcomes (279).

In addition to methodological problems, numer-ous difficult conceptual and methodological issuescomplicate the process of designing and conductingspecial care unit research. Table 1-3 lists many ofthese issues, some of which are discussed in moredetail in appendix B.

Citing these methodological problems and con-ceptual and methodological issues, some commenta-tors discount the findings of the available studies.They imply that no credible research has been doneon special care units or that the studies that had nopositive findings had no findings at all.

In contrast, OTA concludes that at least the sixevaluative studies that used a control group arecredible studies in an area in which good research isdifficult to design and conduct. These studies werecarefully designed and implemented. The specialcare units they studied incorporated the patient carephilosophies, staff training, activity programs, andphysical design features recommended in the specialcare unit literature. Only one of the studies success-fully randomized subjects to the special care unit andthe control group, but the other studies used acceptedstatistical methods to correct for pre-existing differ-ences among the subjects that could affect theoutcomes. Although each of the studies has method-ological problems, it is unlikely the lack of positivefindings is due entirely to these problems. Despite

methodological problems, the studies’ findings aremeaningful and deserve careful consideration bypolicymakers, special care unit advocates, andothers.

It is important to note that none of the availablestudies directly measured the impact of special careunits on residents’ quality of life. Quality of life isdifficult to define operationally and particularlydifficult to measure in individuals with dementia.Several of the clinicians who reviewed this report forOTA pointed out, however, that improvements inresidents’ quality of life maybe the primary positiveoutcome of special care units.

Finally, for policy purposes, it is important to notethat the available evaluative studies provide little orno information about the effectiveness of differenttypes of special care units or particular features inspecial care units. In each of the six evaluativestudies with a control group, the special care unitsdiffered in many ways from the control groupsettings. 10 It is unclear whether the overall milieu ofthe special care units or their particular featuresaccount for the studies’ findings. If particularfeatures account for the findings, it is unclear whichfeatures.

The only evaluative study with a control groupthat found a significant effect of the special care uniton the residents’ ability to perform activities of dailyliving focused on a unit that was created with theaddition of an activity room but no other physicaldesign changes (392). The distinguishing character-istics of the unit, in the view of the researchers, werethe staff’s efforts to accomplish the followingobjectives:

. to identify residents’ specific cognitive impair-ments,

. to treat depression, delusions, and hallucina-tions,

. to identify medication side effects,● to maintain residents’ physical health,. to reduce the use of physical restraints, and. to increase residents’ participation in activities

(392).

The ongoing involvement of a psychiatrist on thestaff also seems to be unique to this study. It isunclear which, if any, of these characteristics aredifferent enough from the characteristics of the

10 Table 4.2 ~ Ch. 4 fists he ch~ges tit were ~de to create he Special Cme tits in each of the Six studies.


Table l-3-Conceptual and Methodological Issues in Designing andConducting Special Care Unit Research

• Special care units are extremely diverse. It is difficult to determine which units should be included ma study sample and which of the many possible unit characteristics are important to study. For purposesof evaluative research, it is difficult to determine whether the intervention to be studied should be theunit’s overall milieu or its particular features and, if particular features, which features.

• Individuals with dementia are extremely diverse. It is difficult to determine which of theircharacteristics are important to study.

• The characteristics of individuals with dementia are interrelated and changeover time. In the contextof an evaluative study, it is difficult to determine whether these changes reflect the progression of theresidents’ dementing disease or the effects of the special care units.

. Residents’ families and special care unit and other nursing home staff members are diverse. It is

●

•

Ž

•*

●

•

●

e

●

•o●

*

●

●

difficult to determine which of their characteristics are important to study.Many of the potentially important characteristics of the units, the residents, their families, and the staffmembers are conceptually vague, difficult to define operationally, and difficult to measure.The available assessment instruments do not include all the potentially important characteristics of theunits, the residents, their families, or the unit staff members. The reliability and validity of some ofthe available instruments has not been demonstrated, and many of the available instruments exhibitceiling or floor effects that obscure the full range of responses.There is insufficient baseline information about many potentially important resident, family, and staffcharacteristics.It is difficult to identify an appropriate control or comparison group.Preexisting differences between special care unit residents and individuals with dementia in othersettings are likely to bias a study’s findings. Because of family preferences and other factors, randomassignment of subjects to a special care unit or a control group setting maybe impractical.Researchers often cannot control the services that subjects in the control group receive.There is disagreement about the outcomes to be studied. This disagreement reflects different valuesin the care of nursing home residents with dementia and different expectations about the areas in whichpositive outcomes may be found.Many potentially important resident outcomes, e.g., quality of life and satisfaction with care, are verydifficult to measure in persons with dementia. The outcomes that are easiest to measure are likely tobe trivial.There are many conceptual and practical difficulties in obtaining consent for research participationfrom individuals with dementia and their families.Because of their cognitive impairments, nursing home residents with dementia are often unable toparticipate in conventional research interviews or to provide accurate information about themselves.Sensory impairments and physical illnesses exacerbate this problem.Proxy-derived information may not be reliable or valid.It is difficult to effectively blind interviewers to the subjects’ treatment status.Sample attrition is very high. Some special care unit studies have lost one-third or more of theirsubjects in a year. Although longer studies may be more likely to find significant effects, attrition isso great that the final sample may be too small to show the effects.The findings of small studies conducted in different special care units often cannot be pooled becauseof differences in the characteristics of the units.It is unclear when measurements should be made. New admissions to a special care unit may exhibittemporary negative effects of the move. Long-time residents may have experienced any positiveeffects of the unit before the beginning of the study.

SOURCE: Offke of Technology Assessment, 1992.


special care units in the other five evaluative studieswith a control group to account for their contradic-tory findings.

THE REGULATORYENVIRONMENT FOR SPECIAL

CARE UNITSBecause of the diversity of special care units, the

fact that existing units frequently do not incorporaterecommended physical design and other features,and pervasive claims that some special care unitsactually provide nothing special for their residents,many Alzheimer’s advocates, State officials, andothers believe there should be special regulations forspecial care units. As of early 1992, special regula-tions were in place or in various stages of develop-ment in many States:

●

●

●

●

●

Six States-Colorado, Iowa, Kansas, Tennes-see, Texas, and Washington-had special regu-lations for special care units.Five States-Nebraska, New Jersey, NorthCarolina, Oklahoma, and Oregon-were in theprocess of drafting or approving special regula-tions for special care units.One additional State-Arkansas-had legisla-tion mandating the development of specialregulations for special care units.Two States—Kentucky and Michigan-hadspecial requirements for special care units orspecial Alzheimer’s nursing homes establishedwith exemptions from the States’ certificate ofneed process.In three additional States—Arizona, Indiana,and Rhode Island, the State-appointed Alz-heimer’s task force or long-term care advisorycouncil had recommended the development ofregulations, and in two of the States—Arizonaand Rhode Island-the State-appointed bodyhad developed draft regulations.

At the State level, interest in regulating specialcare units is growing rapidly. In some States, thisinterest is unopposed. In other States, the issue ofspecial regulations for special care units is highlycontroversial.

State regulations for special care units have beenor will be superimposed on the existing regulatorystructure for nursing homes—a complex, multifac-eted structure with six major components:

1)

2)3)

4)

5)

6)

In

the Federal regulations for Medicare andMedicaid certification of nursing homes,State licensing regulations for nursing homes,State certificate of need regulations for nursinghomes,other State and local government regulationsthat affect nursing homes,the survey and certification procedures associ-ated with each type of regulations, andthe oversight procedures of each State’s Long-Term Care Ombudsman Program.

addition to these six components, Federal,State, and local government regulations for nursinghomes incorporate standards established by privateorganizations, such as the National Fire PreventionAssociation’s Life Safety Code standards. Specialcare units must comply with these standards, as wellas the regulations and survey, certification, andoversight procedures listed above and any specialregulations that may apply.

Special care unit operators and others oftencomplain that the regulations and survey, certifica-tion, and oversight procedures for nursing homesdiscourage innovation in special care units byinterfering with the use of physical design and otherfeatures they believe would be effective for residentswith dementia. OTA has been told about instances inwhich special care units could not get approval forthe use of innovative features of various kinds;instances in which approval was held up for years,thus adding enormously to the cost of establishingthe unit; and instances in which approval was givenby one government agency and later denied byanother government agency, sometimes after thespecial care unit opened. Thus, while there ispressure on the one hand for more regulation ofspecial care units, some people advocate less regula-tion, at least on a selective basis, to allow greaterinnovation.

The regulatory structure for nursing homes iscurrently in flux due to implementation of thenursing home reform provisions of OBRA-87 andrelated legislation. The nursing home reform provi-sions of OBRA-87 changed the Federal regulationsfor Medicare and Medicaid certification of nursinghomes and the survey and certification proceduresassociated with those regulations. Many provisionsof OBRA-87 are relevant to the frequently citedcomplaints about the care provided for nursing homeresidents with dementia. This section summarizes

Chapter l-Overview and Policy Implications . 33

OTA’s findings with respect to the relevant provi-sions of OBRA-87 and the existing State regulationsfor special care units. Both of these topics arediscussed at greater length in chapter 5.

On the basis of the information presented here andin chapter 5, OTA concludes that OBRA-87 pro-vides a better framework for regulating special careunits than any of the existing State special care unitregulations or any special regulations that could bedevised at this time. This conclusion and alternativesto address the concerns that lead some people toadvocate special regulations for special care unitsare discussed in a later section of this chapter, as aremethods to allow greater innovation in special careunits.

The Nursing Home Reform Provisionsof OBRA-87

Through OBRA-87, Congress sought to create acomprehensive regulatory structure that would as-sure high-quality, individualized care for all nursinghome residents. Under OBRA-87, a nursing homemust now meet the following requirements to becertified for Medicare or Medicaid:

●

●

●

●

●

“The facility must care for its residents in amanner and in an environment that promotesmaintenance or enhancement of each resident’squality of life.”‘‘The facility must promote care for residents ina manner and in an environment that maintainsor enhances each resident’s dignity and respectin full recognition of his or her individuality. ’‘‘The facility must conduct initially and period-ically a comprehensive, accurate, standardized,reproducible assessment of each resident’sfunctional capacity.’“The facility must develop a comprehensivecare plan for each resident that includes meas-urable objectives and timetables to meet aresident’s medical, nursing, mental, and psy-chosocial needs that are identified in thecomprehensive assessment. ’“Each resident must receive and the facilitymust provide the necessary care and services toattain or maintain the highest practicable physi-cal, mental, and psychosocial well-being, inaccordance with the comprehensive assessmentand plan of care” (463).

Chapter 5 lists other provisions of OBRA-87 thatare relevant to the frequently cited complaints about

the care provided for nursing home residents withdementia. These other provisions deal with main-taining residents’ functional abilities, providingactivities that meet residents’ needs, providingspecialized rehabilitative services, minimizing theuse of psychotropic medications and physical re-straints, allowing residents to use their own belong-ings, involving residents and their families in careplanning, training for nurse aides, and other issues.

The provisions of OBRA-87 rarely mentiondementia, but the resident assessment system devel-oped to implement OBRA-87 emphasizes the evalu-ation of a resident’s cognitive status and theproblems and care needs that are common amongnursing home residents with dementia (see ch. 5). Asjust noted, the regulations require that residents’needs must be assessed and that once their needs areidentified, appropriate services must be provided tomeet the needs.

If fully implemented, the provisions of OBRA-87would greatly improve the care of nursing homeresidents with dementia. Two factors could limit thebenefits of OBRA-87 for individuals with dementia.One obvious factor is a failure to implement theprovisions, which could occur for a variety ofreasons, including insufficient government fundingfor nursing home care, for inspections, or forsurveyor training. The second factor is lack ofknowledge among many nursing home administra-tors, staff members, and surveyors about whatconstitutes appropriate care for individuals withdementia-e. g., lack of knowledge about whatactivities and rehabilitative services would meet theresidents’ needs.

Existing State Regulations forSpecial Care Units

As noted above, six States—Colorado’ Iowa,Kansas, Tennessee, Texas, and Washington-hadregulations for special care units as of early 1992.Each of the States’ regulations address severalcommon areas, e.g., admission criteria, safety, stafftraining, and physical design, but their requirementsin these areas differ (see ch. 5). Each State requiressome features that are not addressed in the otherStates’ regulations, e.g., Iowa’s requirement that aunit and its outdoor area must have no steps or slopesand Washington’s requirement that the units floors’walls, and ceilings must be of contrasting colors.Some of the requirements are very detailed.


Thus far, State regulations for special care unitshave been developed largely without regard for theprovisions of OBRA-87. Some of the six States’requirements for special care units duplicate OBRArequirements that apply to all nursing homes. Someof the special care unit requirements, e.g., thosedealing with residents’ rights to have visitors, areweaker than the comparable OBRA requirements.

OTA’s analysis of the six States’ regulationsindicates several problems that are likely to arise inany special care unit regulations that could bedevised at present. First, by requiring particularfeatures in special care units, the six States’ regula-tions imply that those features are unique to or moreimportant in the care of residents with dementia thanin the care of other nursing home residents. Yet someof the required features probably are not moreimportant for residents with dementia than for otherresidents. Examples are Iowa’s and Tennessee’srequirements for an interdisciplinary care planningteam, Colorado’s requirement for sufficient staff toprovide for the residents’ needs, and Texas’ require-ment for a social worker to assess the residents onadmission, conduct family support group meetings,and identify and arrange for the use of communityresources. If these features are important for allnursing home residents, it is misleading and poten-tially harmful to residents of nonspecialized units torequire the features differentially for special careunits.

Second, by requiring particular features in specialcare units, the six States’ regulations imply thatthose features are more important in the care ofresidents with dementia than other features that arenot required by the regulations. Yet experts indementia care disagree about which features aremost important in the care of these residents. Theexisting special care unit regulations emphasize stafftraining and physical design features and place farless emphasis on specialized activity programs andprograms to involve and support residents’ families.Although there is no research-based evidence thatany of these features are more likely than the othersto produce positive resident outcomes, some expertsin dementia care would undoubtedly argue thatspecialized activity programs and family supportprograms are as important as staff training andphysical design features in the care of these resi-dents.

Third, by requiring particular features in specialcare units, the six States’ regulations imply that theresources available to the unit should be expendedfor the required features rather than other features.Since most special care units have limited resources,features not required in special care unit regulationsare likely to be neglected.

The six States’ requirements for physical designfeatures are especially troublesome, in part becausethey are so detailed. To incorporate some of therequired features involves extensive remodeling,with obvious cost implications. In some facilities,the required features cannot be incorporated, evenwith extensive remodeling. For such facilities, therequirements can lead to costly new construction ora decision by the nursing home not to establish aspecial care unit (337). If there were evidence of theeffectiveness of particular physical design features,it might be reasonable to require the features. Torequire the features without such evidence is proba-bly inappropriate.

The impact of the six States’ special care unitregulations on the growth of special care units ineach State is unclear. Anecdotal evidence suggeststhat the regulations have discouraged some nursinghomes from establishing special care units. TheStates vary in the extent to which they are enforcingtheir regulations, but several nursing homes in atleast two of the States have closed their special careunit because the unit could not meet the Staterequirements (169,267). It is possible that specialcare unit regulations could cause the closing of unitsthat provide good care for their residents, eventhough they do not meet one or more of the Staterequirements. There is no evidence to determinewhether this has occurred.

As noted earlier, Oklahoma is developing regula-tions for special care units. The regulations areintended by their supporters to set a‘ ‘basic standardof care,” rather than to define what would be ‘‘idealor high-quality care” (118). In the developmentprocess, the draft regulations have become increas-ingly detailed, moving away from what some of theirsupporters first envisioned as broad, general guide-lines that would inform families, nursing homeadministrators, and others about what constitutesbasic care. In the spring of 1992, a telephonefollowup to the 1991 survey of all U.S. nursinghomes with more than 30 beds found that someOklahoma nursing homes that had a special care unit


in 1991 reported they had since closed the unit (246).When asked why they had closed their special careunit, most of the respondents declined to give areason, but one respondent said the unit in hisfacility had been closed in anticipation of verydetailed regulatory requirements the unit would notbe able to meet. OTA has no information about thequality of care provided by this unit or any of theother special care units in Oklahoma that wereclosed between 1991 and 1992.

POLICY IMPLICATIONSFindings from the available research on special

care units and the information just presented aboutthe regulatory environment for special care units andproblems with the existing State special care unitregulations have implications for each of the policyareas addressed in this report: consumer education,research, regulation, and reimbursement.

Implications for Consumer Education AboutSpecial Care Units

The diversity of existing special care units sub-stantiates the need for consumer education. Familiesand others who make decisions about nursing homecare for individuals with dementia could reasonablyassume that all special care units are alike. Theyneed to know that special care units vary in virtuallyevery respect, including the number of residents theyserve, their patient care philosophies and goals, theirphysical design features, their staff-to-resident ra-tios, their admission and discharge policies, andtheir charges. Ideally families and others would haveeasy access to information about each of thesecharacteristics for the special care units they areconsidering. If such information is not available,families and others need to know what questions toask to obtain the information when they call or visita special care unit.

To compile information about the special careunits in a given jurisdiction would be more or lessdifficult, depending on the number of units in thejurisdiction. In jurisdictions with more than onespecial care unit, definitional issues would have tobe resolved so that information about different unitswould be comparable. Since the units are likely tochange over time, an ongoing effort would berequired to update the information.

Compiling and updating information about thespecial care units in a given jurisdiction could be aproject of an Alzheimer’s Association chapter,another private agency, or a public agency .11 Inmostjurisdictions, a local agency would be the mostappropriate organization to perform this task. Be-cause of the amount of detail involved and thenecessity for frequent updates, the information couldnot be effectively compiled and updated at theFederal level. In States with relatively few specialcare units, it probably could be compiled andupdated at the State level.

Descriptive information about the characteristicsof particular special care units would be useful tofamilies and others because the characteristics ofsome units (e.g., the units’ patient care philosophies,discharge policies, or design features) would matchtheir individual needs, preferences, and values. Itshould be recognized, however, that the availableresearch findings do not provide objective standardsto help families and others evaluate special careunits. Although some unit characteristics may seemright intuitively and match the needs, preferences,and values of some families, the available researchfindings do not prove that any particular unitcharacteristics are associated with better residentoutcomes.

Based on the available information, the messagefor consumers is that special care units vary greatly;that there is little research-based evidence of betterresident outcomes in special care units than innonspecialized units; and that although a givenspecial care unit may have better resident outcomesthan another special care unit or a nonspecializedunit, there is no research-based evidence to identifythe unit characteristics that explain the differentoutcomes. On the positive side, it can be said thatspecial care units are likely to have fewer residentsand more staff members per resident than nonspe-cialized nursing home units; that in comparison withthe residents of nonspecialized units, special careunit residents are less likely to be physicallyrestrained; and that even though there is littleresearch-based evidence of better resident outcomesin special care units than in nonspecialized units,there is much less evidence of worse outcomes inspecial care units. Consumers need to know, how-ever, that these statements refer to averages that maynot apply to a given unit. Although this message

11 IU some jfis&ctions, a public or private agency compiles and updates similar types of information about IOCd nursing homes.

36 . Special Care Units for People With Alzheimer’s and Other Dementias

does not meet the need for objective standards toevaluate special care units, it does accurately repre-sent what is known about the units.

A few States have or are developing consumereducation materials about special care units. NewHampshire has published an 8-page booklet in-tended for family members who are trying toevaluate special care units and nursing home opera-tors who are interested in establishing a special careunit (325). The booklet describes the characteristicsof an individual with Alzheimer’s disease, the needsof the individual and the family, and the characteris-tics of specialized dementia care. It provides ques-tions and a checklist that families can use to evaluatespecial care units. For nursing home operators, thebooklet lists reasons for having a special care unit,questions the nursing home operator and staff shouldconsider in establishing a special care unit, andfactors that will influence the success of the unit.

The American Association of Homes for theAging, the Massachusetts Alzheimer’s Disease Re-search Center, the National Institute on Aging’sAlzheimer’s Disease Education and Referral Center,the University of South Florida’s Suncoast Geron-tology Center, and the University of Wisconsin-Milwaukee’s Center for Architecture and UrbanPlanning Research have developed guidelines forspecial care units, and other organizations aredeveloping such guidelines (see ch. 5). The Alz-heimer’s Association released its special care unitguidelines in July 1992. Some of these organiza-tions’ guidelines are intended primarily to assistfamilies in evaluating special care units and otherorganizations’ guidelines are intended primarily toassist nursing home operators in planning andsetting up a special care unit.

OTA’s review of the various organizations’ spe-cial care unit guidelines indicates that the guidelinesare quite similar in content, despite some differencesin emphasis, format, and wording. Each organiza-tion’s guidelines cite numerous unit characteristicsthe organization considers desirable. This informa-tion is useful for families and others who are tryingto evaluate special care units, but consumers need toknow that statements about the desirability of par-ticular unit characteristics are based on expert opin-ion and that experts disagree about these matters.

Information about the theoretical concepts ofspecialized dementia care discussed earlier in thischapter may also be useful for families and others

who are trying to evaluate special care units. Theyneed to know, however, that the concepts are notimplemented in all special care units and that thesame concept may be implemented differently, withdifferent results, in different units.

Given the availability of special care unit guide-lines developed by various organizations, there is noneed for Federal agencies to develop additionalguidelines. Federal agencies that serve elderly peo-ple and their families could play a valuable role,however, in disseminating the available guidelinesand promoting their use.

As noted earlier, the task of compiling andupdating information about the characteristics ofspecial care units in a given jurisdiction is probablymost effectively performed by local agencies, in-cluding Alzheimer’s Association chapters. In somejurisdictions, however, local agencies that receiveFederal funding, such as area agencies on aging(AAAs), might be the most appropriate organiza-tions to perform the function.

In thes summer of 1992, the Alzheimer’s Associa-tion contracted for a study to identify and documentconsumer problems with special care units. Theresults of this study, which will be available in thespring of 1993, will provide useful informationabout the extent and types of problems families andothers encounter in dealing with special care unitsand may indicate a need for additional governmentinitiatives in this area.

Implications for Research onSpecial Care Units

The findings of the available special care unitstudies confirm the need for research on manyunresolved issues. For public policy purposes, themost important research issues are those pertainingto effectiveness. Evaluative research is needed toanswer three interrelated questions about the effec-tiveness of special care units for their residents:

1)

2)

3)

Do special care units improve resident out-comes?If so, is it the overall milieu or particular unitcharacteristics that are effective, and if it isparticular unit characteristics, which charac-teristics?Are special care units effective for all nursinghome residents with dementia or only certain


types of residents with dementia, and if onlycertain types, which types?

Research on the effectiveness of special care unitsfor residents’ families, unit staff members, andnondemented nursing home residents is also needed.

Descriptive information is needed to provide abetter general understanding of special care unitsand to develop descriptive topologies. Such typolo-gies, which would be based on unit and perhapsresident characteristics, are important for designingevaluative studies and understanding and generaliz-ing from their findings. To be useful for publicpolicy purposes, descriptive topologies must repre-sent the full range of existing units.

Information is needed about the cost of caring forindividuals with dementia in special care units vs.nonspecialized nursing home units. Because of thediversity of special care units, this information willbe useful only if it is developed in the context of aninclusive typology of the units.

OTA is aware of several sources of forthcomingdescriptive information that will meet some of theseneeds. One source is the 1991 survey of all nursinghomes with more than 30 beds. The survey’sfindings with respect to the proportion of nursinghomes that had a special care unit in 1991 were citedearlier in this chapter. The survey also includedquestions about the physical features of the units,their admission and discharge criteria, staff trainingprograms, staff support groups, activity programs,family programs, and sources of reimbursement.

A second source of forthcoming descriptiveinformation is the resident assessments mandated bythe nursing home reform provisions of OBRA-87.All Medicare and Medicaid-certified nursing homesare now required to assess each of their residents,including special care unit residents, at the time ofthe residents’ admission to the nursing home andannually thereafter. OBRA-87 mandated the devel-opment of a set of core items to be addressed in therequired assessment, and the core items include each

of the resident characteristics discussed in thischapter.

Lastly, as noted earlier, the Multi-State NursingHome Case-Mix and Quality Demonstration in-cludes special care unit residents among the 6800nursing home residents in the study sample. Infor-mation has been collected on more than 300residents of 20 special care units in 6 States (137). ToOTA’s knowledge, this study is the first to includea time-and-motion analysis of resource use inspecial care units.

Given the pervasive complaints and concernsabout the care provided for nursing home residentswith dementia, the extensive involvement of govern-ment in regulating nursing homes and paying fornursing home care, and the competing claims ofspecial care unit advocates and critics, one mightexpect that Federal agencies would have fundedmany special care unit studies. In 1984, the TaskForce on Alzheimer’s Disease of the U.S. Depart-ment of Health and Human Services noted the needfor this research (470). In 1986, Congress mandatedspecial care unit research (P.L. 99-660), but fundingfor the research was never appropriated. Between1986 and 1990, seven Federal agencies each pro-vided funding for one special care unit study .12Three of the studies were small pilot studies, and twowere relatively small components of large-scalenursing home studies. Two of the National Instituteon Aging’s Alzheimer’s Disease Research Centerseach provided funding for one special care unitstudy. The Alzheimer’s Association, the BrookdaleFoundation, the State of California, and threeuniversities each provided funding for one specialcare unit study. Most of the other special care unitstudies have been small pilot studies with no fundingsource. 13

In 1990, the Alzheimer’s Disease Research Centerat Washington University in St. Louis sponsored aspecial care unit conference that included workshopsfor researchers. The intent of the workshops was toidentify the problems that were obstructing progressin special care unit research. Many interrelated

12 The seven agencies and the studies for which they provided full or partial funding are: 1) ~“ “stration on Aging: “Special Care Units forAlzheimer’s Disease Patients: An Exploratory Study of Dementia Speciilc Units” (64); 2) Agency for Health Care Policy and Research: 1987 NationalMedical Expenditure Survey (249); 3) Department of Veterans Affairs: “A Comparison of Alzheimer Care Units: Veterans Administration State, andPrivate” (232); 4) Health Care Financing Administration: Multi-State Nursing Home Case-Mix and Quality Demonstration (144,382); 5) HealthResources and Services A&mm“ “stration: “Hospitalization Rates in Nursing Home Residents With Dementia: A Pilot Study of the Impact of a SpecialCare Unit” (99); 6) National Center for Nursing Research: “Nursing Evaluation Research: Alzheimer’s Care Unit” (265); and 7) National Institute onAging: “Five-State Study of Special Care Units in Nursing Homes” (194).

Is ‘r’ables 3-1% b, and c in & 3 and tables L&l ~d A-Z in Ch. A ~St tie funding Sources for ~1 the speci~ cme unit studies discussed in ~S RpOfi.


problems were identified, including the difficulty ofobtaining funding for special care unit research, thedifficulty of getting special care unit researchpublished, and numerous conceptual and methodo-logical issues in designing and conducting this kindof research (see app. B). Following the conference,the researchers formed an ad hoc group, theWorkgroup on Research and Evaluation of SpecialCare Units, to address the identified problems. Bythe end of 1991, the workgroup had over 100members (193). It has no formal sponsor and nofunding.

In the fall 1991, the National Institute on Agingfunded nine studies under anew “Special Care UnitsInitiative, ’ and the agency funded a tenth study inearly 1992. Two of the studies will develop descrip-tive topologies of special care units. Two otherstudies will compare service use and costs forspecial care unit residents and demented and nonde-mented residents in nonspecialized units in a total of24 nursing homes. Another study will compareresident outcomes in the special care units andnonspecialized units in the Multi-State NursingHome Case-Mix and Quality Demonstration.

The National Institute on Aging’s “Special CareUnits Initiative” represents a major commitment tospecial care unit research. The results of the 10studies will greatly expand knowledge about specialcare units. Moreover, the studies were funded underan arrangement that requires the 10 research teamsto collaborate on the development of commondefinitions and assessment procedures so that, al-though the studies focus on different issues, theirfindings will be comparable.

As noted earlier, the effectiveness of special careunits is the most important research issue for publicpolicy purposes. Although several of the NationalInstitute on Aging studies will evaluate the effec-tiveness of the units they are studying, the complex-ity of the policy-related questions about effective-ness means more research will be needed on thisissue. Some researchers believe that a clinical trialwith a randomized case control design will eventu-ally be needed to determine the effectiveness ofspecial care units (143,41 1). Currently funded stud-ies will provide the basis for designing such aclinical trial. The legal and ethical issues discussedlater in this chapter also raise important policy-related questions that are not addressed in theNational Institute on Aging studies.

To complement special care unit research, studiesare needed in two broad areas:

1.

2.

physical design features and care methods forpeople with dementia generally; andalternatives to special care units, includingspecial programs for nursing home residentswith dementia in nonspecialized units, specialresidential care programs inboard and care andassisted living facilities, and special adult dayand in-home services.

Studies in the first area can be conducted inspecial care units or in other residential and nonresi-dential care settings. It may be easier and moreefficient to conduct some of these studies in specialcare units, however, because all the residents havedementia.

Research on specific design features and patientcare methods may help to explain the findings ofspecial care unit research. If certain design featuresor care methods are shown to be effective orineffective in general or for certain types of resi-dents, those findings may explain the results ofspecial care unit studies. More importantly perhaps,studies of specific design features and care methodscan identify features and methods that will improvethe care of residents with dementia in nonspecializedunits and other settings as well as in special careunits.

The Robert Wood Johnson Foundation and theCleveland Foundation have funded research onvarious design features and patient care methods intwo special care units at the Corinne Dolan Alz-heimer’s Center in Chardon, OH. Studies of this kindhave also been conducted in some of the special careunits at VA medical centers (159). Three special careunits that constitute the Dementia Study Unit at theVA medical center in Bedford, MA, have been thesite for numerous studies on the care of individualswith dementia in the late stages of their illness. ToOTA’s knowledge, the Dementia Study Unit is theonly research group in the country to focus its effortson the difficult, emotionally charged, clinical issuesin late-stage and terminal care for individuals withdementia. The research group has studied swallow-ing and feeding difficulties (476), tube feeding(475), use of antibiotics vs. palliative measures totreat fever in late-stage patients (135), and use of ahospice-like approach to care for late-stage patients(474).


Implications for Government Regulation ofSpecial Care Units

The diversity of special care units, the fact thatexisting units often do not incorporate the featuresrecommended for special care units, and pervasiveclaims that some special care units just use the wordsspecial care as a marketing tool and actually providenothing special for their residents lead many Alz-heimer’s advocates, State officials, and others tosupport the development of special regulations forspecial care units. On the other hand, the lack ofagreement among experts about what features aremost important in the care of residents with demen-tia and the lack of research-based evidence showingthat any particular features are associated with betterresident outcomes make it difficult to justify theselection of particular features that should be re-quired in special care units.

The Alzheimer’s Association has developed leg-islative principles that identify 11 areas a Stateshould include when drafting special care unitlegislation or regulations: 1) statement of mission,2) involvement of family members, 3) plan of care,4) therapeutic programs, 5) residents’ rights,6) environment, 7) safety, 8) staffing patterns andtraining, 9) cost of care, 10) quality assurance, and11) enforcement (4). As described in chapter 5, thespecial care unit guidelines developed by variousorganizations identify similar areas that requirespecial consideration in the care of nursing homeresidents with dementia. Thus, there appears to besome agreement about the areas of concern.

Having agreement about areas of concern ishelpful in “thinking about the particular features thatmight be desirable or required in special care units,but agreement about areas of concern is not the sameas agreement about particular features. For example,agreement that therapeutic programs and physicalenvironment are areas of concern does not constituteagreement about which therapeutic programs orphysical design features should be required. OTAhas observed that in discussions about special careunit regulations, agreement about areas of concernoften masks considerable disagreement about partic-ular features and gives an erroneous impression thatthere is consensus about the particular features thatshould be required.

As noted earlier, OTA’s analysis of the existingState regulations for special care units indicates

several problems that are likely to arise in anyspecial care unit regulations that could be devised atpresent. First, regulatory requirements for particularfeatures in special care units imply that thosefeatures are unique to or more important for specialcare unit residents than for other nursing homeresidents. Yet many of the features that are importantfor special care unit residents are probably just asimportant for other residents. This is especially truesince most nursing home residents with dementia arenot in special care units now and may never be.

Second, regulatory requirements for particularfeatures in special care units imply that thosefeatures are more important in the care of specialcare unit residents than other features that are notrequired by the regulations and that the resourcesavailable to the unit should be expended for therequired features. Most special care units havelimited resources, so features that are not required inspecial care unit regulations are likely to be ne-glected. Yet experts in dementia care disagree aboutwhich features are most important in the care ofthese residents.

The problem of special care unit regulations thatomit features regarded as important by some demen-tia experts could be solved by expanding theregulations to require those features. The more theregulations are expanded, however, the more likelyit is that the required features will be important forother nursing home residents as well.

Given these problems, OTA concludes that OBRA-87 provides a better framework for regulatingspecial care units than any special regulations thatcould be devised at this time. The advantages ofOBRA-87 are its comprehensiveness, its emphasison individualized care, and its mandated assessmentand care planning procedures. The primary problemwith OBRA-87 for special care units is the sameproblem faced by anyone who ties to developregulations for special care units: i.e., the lack ofagreement among experts about what features aremost important in the care of residents with demen-tia and thus what should be special about special careunits. Solving this problem through support forresearch to evaluate the effectiveness of particularfeatures may eventually provide a substantive basisfor special care unit regulations. In the meantime, itis important to consider alternate ways of addressingthe concerns that have led many Alzheimer’s


advocates, State officials, and others to favor thedevelopment of special care unit regulations.

Alternatives to Special Care Unit Regulations

Alzheimer’s advocates, State officials, and otherswho favor the development of special care unitregulations often cite the need to protect individualswith dementia from poor-quality care and the needto protect these individuals and their families fromnursing homes that claim to provide special care butactually do not. Some people who favor the develop-ment of special care unit regulations also cite a needto assist nursing homes in designing their specialcare units and to assist surveyors in inspecting theunits. Each of these objectives can be achievedwithout special regulations.

In discussions about special care unit regulations,it is sometimes suggested that there are two types ofspecial care units—’ good’ units and ‘bad’ units—and that regulations are needed to eliminate the“bad’ units. In this context, it is probably more ac-curate to think about four types of special care units:

1.

2.

3.

4.

units that provide the features a given observerconsiders important for residents with demen-tia,units that do not provide those features but doprovide other features the unit operator, staff,or advisers consider important for residentswith dementia,units that claim to provide special care butactually provide nothing special for theirresidents, andunits that provide poor-quality care that wouldbe inappropriate for any nursing home resi-dent.

Anecdotal evidence suggests that there are veryfew units of the last type, and the one study that hasaddressed this issue supports that conclusion (154).OBRA-87 provides a sufficient basis for censuringunits of that type, without the need for specialregulations.

Most special care units are of the first three types.Objective classification of particular units into thesetypes would be difficult, since the classificationdepends on a given observer’s opinion about thefeatures that are important in a special care unit anda judgment about the intentions of each facility’sadministrators. Although some nursing home ad-ministrators may knowingly provide no specialservices in their special care unit, other administra-

tors probably believe erroneously that they areproviding appropriate care. One commentator refersto the latter units and their administrators as“innocent” (21).

An earlier section of this chapter discussed theneed for consumer education about special careunits. As noted there, families and others who aretrying to evaluate special care units need to knowthat existing units vary greatly. They need compara-ble information about the characteristics of thespecial care units in their geographic area andinformation about characteristics that may be impor-tant in a special care unit. Lastly, they need to knowthat experts disagree about the importance of partic-ular unit characteristics and that their personalpreferences and values are relevant in selecting aunit. These types of information will not protect allpotential special care unit residents and their fami-lies from nursing homes that provide no specialservices in their special care unit. Neither will theseindividuals be protected, however, by regulationsthat require special care units to incorporate featuresthat have not been proven to be effective.

For the purpose of consumer protection, nursinghomes could be required to disclose certain informa-tion about their special care units to potentialresidents and their families. In particular, they couldbe required to disclose what is special about the unit;how the unit differs from nonspecialized units in thesame facility; how physical restraints and psy-chotropic medications are used in the unit; whetherthere are behavioral problems that cannot be handledon the unit; whether it is expected that individualswho are admitted to the unit will be dischargedbefore their death and, if so, for what reasons. Adisclosure requirement could be mandated at theFederal level within the framework of OBRA-87 orat the State level within the framework of Statelicensing regulations. Such a disclosure requirementwould be quite different from regulations thatrequire particular features in a special care unit. Itwould make useful information available to con-sumers without suggesting that particular featuresare known to be effective. A disclosure requirementwould not eliminate the need for the other types ofconsumer information described above.

Guidelines are the best method to assist nursinghomes in designing their special care units. Severalof the guideline documents mentioned earlier in thischapter and discussed at greater length in chapter 5


are intended primarily for this purpose.14 More sothan regulations, guidelines can convey the objec-tives of specialized dementia care, the currentuncertainty about the most effective methods ofcare, and the need for innovation and evaluativeresearch in special care units.

Surveyor guidelines developed within the frame-work of OBRA-87 are the best method to assistnursing home surveyors in inspecting special careunits. Since 1989, the Joint Commission on Accred-itation of Healthcare Organizations (JCAHO) hasbeen working on guidelines to help its surveyorsevaluate special care units. JCAHO is a privateorganization that accredits hospitals, home healthagencies, mental health organizations, and about1000 nursing homes in the United States (214). Thecommission’s effort to develop guidelines evolvedfrom its surveyors’ questions about how to evaluatethe increasing number of special care units they wereseeing in nursing homes accredited by the commis-sion (434).

JCAHO’s draft surveyor guidelines provide whatis, in effect, a detailed answer to the question, ‘Whatconstitutes appropriate care for nursing home resi-dents with dementia?’ The guidelines are based onthe commission’s standards for all nursing homes(435). No changes have been made to the basicstandards. Instead, statements have been added nextto many of the standards to explain the implicationsof the standard for the care of residents withdementia and to describe the process the surveyorshould follow in scoring the special care unit on thatstandard. Although some commentators may dis-agree with some of the statements, the JCAHOguidelines provide a valuable model which could beadapted to OBRA regulations.

Waivers and Other Methods To AllowInnovation in Special Care Units

As noted earlier, special care unit operators andothers often complain that the existing regulationsand survey and certification procedures for nursinghomes discourage innovation by interfering with theuse of physical design and other features theybelieve would be effective for residents with demen-tia. From a societal perspective, one objective, and

perhaps the most important objective, of special careunits is to develop better ways of caring for nursinghome residents with dementia. To accomplish thisobjective, methods must be found to allow andencourage innovation in special care units.

One method to allow greater innovation in specialcare units is to eliminate regulations that restrictinnovative physical design and other features. Al-though this method may eventually be appropriate,the current lack of agreement about the features thatare important in a special care unit and the lack ofresearch-based evidence for the effectiveness ofparticular features make decisions to eliminateexisting regulations premature.

A better method is to create a process by whichindividual special care units could obtain waivers toimplement physical design features, patient carepractices, and other innovations they believe willbenefit residents with dementia. Most existingregulatory codes have a process for granting waiv-ers, but in some and perhaps many States, thewaivers that are granted are for relatively trivialchanges (201). The purpose of creating a waiverprocess for special care units would be to allow theimplementation and evaluation of nontrivial innova-tions. Since such innovations would change the careof individuals with dementia in significant ways, thewaivers should only be granted on a facility-by-facility basis after careful prior review by a panelthat includes health care professionals, consumeradvocates, industry representatives, architects, de-signers, surveyors, fire marshals, building inspec-tors, and others. The panel would have to determinewhether a proposed innovation was worth evaluat-ing and whether sufficient safeguards had been builtinto the proposal to protect the residents. The panelwould also have to monitor the waivered innova-tions on an ongoing basis to assure the safety andwell-being of the residents. A panel of this kindprobably would function most effectively at theState level, but the Federal Government couldencourage the development of such panels throughdemonstration grants.

At present, State efforts with respect to specialcare units are focused primarily on the development

14 E-Pl~~ of @&~c dOCW~nt~ int~~d~d to assist n~sing homes in &Si@g a special Care tit are tie American Association Of Homes fOrthe Aging’s “Best Practices” document (10); the Massachusetts Akheimer’s Disease Research Center’s “Blueprint” document (287); the Universityof Wisconsin-Milwaukee’s Center for Architecture and Urban Planning Research’s “Design Guide” (95); and the Alzheimer’s Association’s“Guidelines for Dignity,’ released in July 1992. The forthcoming VA guidelines for special care units in VA medical centers will also be useful fornursing homes that are trying to establish a special care unit.


of special regulations. To OTA’s knowledge, noState has created a process for waiving regulationsthat interfere with innovation in special care units. Afew States have provided grants to nursing homesand other facilities to create model special care units.In at least one of these States, the State’s ownregulations made it difficult for some of the facilitiesthat received the grants to implement the featuresthey considered appropriate for individuals withdementia, thus defeating the purpose of the grants. Ifspecial care units are to fulfill the societal objectiveof developing better methods of care for nursinghome residents with dementia, policies to allow andencourage innovation must receive at least as muchattention as methods to regulate and control theunits.

In addition to a waiver process, several othermethods to allow and encourage innovation inspecial care units are discussed in chapter 6. Someof the methods pertain primarily to special careunits, e.g., providing training materials and pro-grams to inform surveyors and others about prob-lems in the care of nursing home residents withdementia and the importance of developing alternateapproaches to their care. Other methods pertain to allresidential facilities for older people, e.g., simplify-ing the process for obtaining approval of new designor other features, eliminating conflicts and inconsis-tencies in the requirements of different agencies andregulatory codes, and including in any new regula-tions an explicit statement of the purpose of eachrequirement; such a statement would provide gov-ernment officials with a basis for allowing innova-tions that meet the purpose, if not the precisestipulations, of the requirement.

Fire safety regulations and interpretations of firesafety regulations are often cited as limiting the useof innovative physical design features in special careunits. A conference or invitational meeting jointlysponsored by the Alzheimer’s Association, theNational Fire Protection Association, and the Fed-eral Government would be a valuable first step indelineating this problem and identifying possiblesolutions.

Implications for Reimbursement forSpecial Care Units

Although most special care unit operators reportthat it costs more to create and operate a special careunit than a nonspecialized nursing home unit, somespecial care unit operators disagree. As noted earlier,the cost of new construction or remodeling to createa special care unit varies greatly for different units.Ongoing operating costs also vary. This variation incosts provides little justification for an across-the-board increase in government reimbursement forcare in special care units.

Ninety percent of government-funded nursinghome care is paid for by Medicaid (250). Medicaidreimbursement for nursing home care varies indifferent States. It is low in many States and very lowin some States. High-quality nursing home care forindividuals with dementia probably costs more thanMedicaid pays in these States, regardless of whetherthe care is provided in a special care unit or anonspecialized unit. High-quality nursing home carefor individuals with other diseases and conditionsprobably also costs more than Medicaid pays inthese States. To improve quality of care, it may benecessary to increase Medicaid reimbursement forall nursing home care in these States. In the contextof this OTA report, however, the question is whetherreimbursement should be increased differentially forspecial care units.15

The results of two studies cited earlier indicatethat average staff time and therefore the average costof care is higher for residents with dementia inspecial care units than in nonspecialized nursinghome units (143,413). If future studies confirm thisfinding, one could argue that government reimburse-ment should be increased differentially for care inspecial care units. If the higher average cost of carein special care units is not associated with betterresident outcomes, however, increasing governmentreimbursement will raise government expendituresand create financial incentives for the establishmentof more special care units without necessarilyimproving the care available for individuals withdementia----dearly not a desirable result. On theother hand, if the higher average cost of care in

15 Arelat~but~erentquestiOn is whethergovernment reimbursement should be increased differentially for nursing home residents witi dementkvs. nondemented residents in any nursing home unit. Two studies have found that certain types of residents with dementia (i.e., those who do not havesevere impairments in activities of daily living or extensive medical care needs) use more staff time and therefore more of a nursing home’s resourcesthan nondemented residents who have the same impairments and medical care needs (16,144). Given these findings, it would be reasomble forgovernment to differentially increase reimbursement for these types of residents with dementia.

Chapter 1--Overview and Policy Implications ● 43

special care units is associated with better outcomesfor individuals with dementia, policymakers will befaced with a difficult question of values, sinceincreasing government reimbursement for the careof demented and nondemented residents in nonspe-cialized units would probably produce better out-comes for those individuals as well.

In the past, reimbursement for nursing home careinmost State Medicaid programs was based on a flatrate system that paid nursing homes at the same ratefor each of their Medicaid-eligible residents, regard-less of differences in the resources required for eachindividual’s care. As of 1990, 19 State had switchedto case-mix systems to determine the level ofMedicaid reimbursement for nursing home care(51). Case-mix systems are intended to match thelevel of reimbursement for individual residents tothe resources used and therefore the cost of their care(142). To implement an increase in governmentreimbursement for care in special care units proba-bly would involve more complex mechanisms inStates with case-mix vs. flat rate reimbursementsystems. Such an increase is not indicated, however,unless and until there is better evidence than iscurrently available that special care units improveresident outcomes.

LEGAL AND ETHICAL ISSUES INSPECIAL CARE UNITS

Because of the cognitive impairments of specialcare unit residents, difficult legal and ethical issuesarise in connection with many aspects of their care.These issues are not unique to special care units, butthey tend to be magnified in special care unitsbecause of the concentration of individuals withdementia and the likelihood that they are in the laterstages of their illness and at least moderatelycognitively impaired.

Many of the difficult legal and ethical issues in thecare of individuals with dementia have been ana-lyzed at length in three previous OTA reports(457,458,459) and in a supplement to The MilbankQuarterly based on OTA contract documents (496).These issues are: criteria and procedures for deter-mining an individual’s decisionmaking capacity;methods of enhancing decisionmaking capacity;competency determinations; criteria and proceduresfor designating a surrogate decisionmaker; rightsand responsibilities of family members as surrogatedecisionmakers; criteria for surrogate decisions;

guardianship and conservatorship; decisions aboutfinancial matters, use of services, and medical carein the end of life; advance directives; the role ofethics committees; risk taking and professional andprovider liability; and the ethical aspects of resourceallocation. Other agencies and individuals have alsowritten extensively about many of these issues.

This section describes some of the particularlytroublesome legal and ethical issues that arise withrespect to three aspects of the care of individualswith dementia in special care units: locked units,admission and discharge, and informed consent forresearch participation. These issues and many of theissues noted above require further clarification andanalysis as they apply to special care units.

The 1991 report of the Advisory Panel onAlzheimer’s Disease includes a section on values(2), and the panel is working on a report on legalissues in the care of individuals with dementia (450).The panel’s 1991 report discusses value differencesand potential value conflicts among the four mainconstituencies involved in the care of individualswith dementia: the individuals, their families, formalservice providers, and the public. Although notfocused on special care units, the panel’s analysis ofthese value differences and potential value conflictsis relevant to some of the most difficult ethicalquestions that arise in special care units, e.g.,questions about whose interests should be givenprecedence in defining the goals of care, makingday-to-day decisions about care, and selecting theoutcomes to be studied in special care unit research.In each of these areas, nondemented nursing homeresidents constitute an important fifth constituencywhose interests must be considered.

Issues With Respect to Locked Units

At least three-quarters of existing special careunits have an alarm or locking system to keepresidents from leaving the unit unescorted or withoutstaff knowledge. Probably at least half of these unitsare locked, although the exact proportion is notknown and undoubtedly varies from State to State.

People’s attitudes about locked special care unitsdiffer (20,178). Some people regard locked units asa way of providing greater freedom and autonomyfor individuals with dementia who otherwise mightbe physically restrained or medicated to keep themfrom wandering away from the unit. At the otherextreme, some people regard locked units as a form


of involuntary confinement that restricts freedomand autonomy and violates the civil rights ofindividuals with dementia. Some people considerlocked units a necessary placement option, whereasothers consider them unnecessary and argue thatwandering residents can be managed effectively inan unlocked unit with an alarm system.

People distinguish in various ways betweenlocked units they regard as acceptable and lockedunits they regard as unacceptable. Some peopleregard locked units that provide adequate staff andactivities as acceptable and locked units that do notprovide these features as unacceptable. Likewise,some people regard as acceptable locked units thathave direct access to an outdoor area, such as anenclosed courtyard or garden, where residents canwander freely (although they are still confined),whereas they regard as unacceptable locked unitsthat do not have such an outdoor area. It is unclearwhether these differences are important from a legalor an ethical point of view.

Some people also distinguish between lockedunits and units that are not locked but have someother method of keeping residents from leaving theunit, e.g., camouflaging the exit doors or using a typeof doorknob that most people with dementia cannotfigure out how to open. Again, although somepeople regard these as distinct alternatives, it isunclear whether the distinction is important from alegal or an ethical point of view.

Units that are not locked but have another methodof keeping residents from leaving the unit are oftenreferred to as secure, secured, protected, or protec-tive units. These terms are also used—sometimes aseuphemisms-for the term locked. This semanticproblem makes it difficult for people to communi-cate clearly about the legal and ethical issues raisedby various methods of keeping residents fromleaving a special care unit.

Some States prohibit locked nursing home unitsor classify them in a different regulatory categorythan unlocked units.16 At least one State official hasargued that locked units constitute physical re-straints in the context of OBRA regulations and thus

require ongoing efforts to move the residents to aless restrictive environment (85).

Families often worry about the safety of a personwith dementia who wanders. Anecdotal evidencesuggests that one thing some families are looking forin a special care unit is assurance that the person willbe safe. They may prefer a locked unit for thisreason. On the other hand, some families may bevery reluctant to place their relative with dementia ina locked unit.

The effect of locked units on the residents isunclear. One study compared the behavior of 22special care unit residents after they encountered alocked vs. an unlocked exit door. The study foundthat the residents were much less agitated after theyencountered the unlocked door (315). Some resi-dents who encountered the unlocked door tested thedoor several times-apparently to be sure it wasunlocked-and then decided not to go out.

Issues With Respect to Admissionand Discharge

Nursing home admission for a nondementedperson raises difficult legal and ethical issues, in partbecause decisions about nursing home admission areseldom autonomous (8,307). The admission of aperson with dementia to a special care unit may raiseeven more difficult issues if the person is incapableof an autonomous decision, the unit is locked, orboth.

Many commentators have debated the similaritiesand differences between the admission of an elderlyperson to a nursing home and the admission of apsychiatric patient to a mental hospital.17 The twosituations are generally perceived as different enoughso that the legal protections that apply to mentalhospital admissions are considered unnecessary orinappropriate for nursing home admissions. In thecase of locked units and individuals who lackdecisionmaking capacity, however, some peoplebelieve additional legal protection is needed. Onepossibility is a requirement for a legally appointedguardian to give consent when a person who lacksdecisionmaking capacity is admitted to a locked

16 AS descfi~ in chapter 5, Colorado’s special care unit regulations apply OI@ to locked tit%17 s=, fore~ple, Cohen, “CaringfortheMentallyIll Elderly Without DeFacto Commitments to Nursing Homes: The Right to the LeastR@rictive

Environment” (90); Moody, “Ethical Dilemmas in Nursing Home Placement” (307); and Spring, “Applying Due Process Safeguards” (420).


unit. Another possibility is a requirement for a civilcommitment in such cases.

These requirements would provide additionalprotection for individuals with dementia and at thesame time create grave obstacles to special care unitadmission. Many families would be unwilling topursue either guardianship or a civil commitment,and some individuals with dementia have no one toinitiate the necessary legal proceedings for them. Ifbetter care is available in a special care unit, legalrequirements intended to protect potential specialcare unit residents could be seen instead as denyingthem access to better care. In fact, if better care isavailable in a special care unit, any decision not toadmit an individual to a special care unit or todischarge an individual from the unit could be seenas denying the individual access to better care. Suchdecisions could be regarded as discriminatory,depending on the basis for the decision.

Some of the difficult legal and ethical issues withregard to discharge involve a conflict between thepresumed right of the unit and its staff to determinewho will be cared for in the unit and the presumedright of residents to remain in the unit if they or theirfamilies so choose. A recent case in a Washington,DC, nursing home illustrates one such conflict. Inthis case, the family of a 91-year-old special careunit resident challenged the facility’s decision todischarge the resident from the unit (204). Thefacility, which had a formal discharge policy,wanted to move the resident to another unit because,in the opinion of the unit staff, she could no longerbenefit from the special care unit. The family arguedthat the resident, who had been in the same room forsix years, might experience ‘‘transfer trauma” as aresult of the move. The hearing examiner ruled thatthe facility could not move the resident even thoughit was clear that the resident did not meet thefacility’s criteria for placement on the unit.

A related issue pertains to special care units thatadmit but later discharge individuals who havebehavioral symptoms which, in the opinion of theunit staff, cannot be managed on the unit. Somepeople believe special care units should be expectedto and should be able to care for individuals withsevere behavioral symptoms. They suggest thatspecial care units that discharge such individualsmay be violating their formal or informal admission

agreement with the residents and the residents’families. On the other hand, the facility is liable forinjuries to other residents that may be caused by aphysically aggressive resident and responsible to theother residents and their families for the overallatmosphere in the unit, which may be negativelyaffected by behaviorally disturbed residents.

Issues With Respect to Consent forResearch Participation

Special care unit researchers report that obtaininginformed consent for research participation byspecial care unit residents is very difficult (79,411,436).Most of the residents are not capable of givinginformed consent, and many residents’ families arereluctant to give consent. As a result, studies thatrequire informed consent are likely to end up withsmall samples that may not be representative of thelarger population of residents. To address thisproblem, some special care unit studies have beendesigned to avoid the need for informed consent. Insuch studies, the researchers review the residents’medical records, observe the residents, and talk tothe unit staff, but they do not interact directly withthe residents because to do so is perceived to requireinformed consent. In contrast, record reviews, resi-dent observation, and staff interviews are not per-ceived to require informed consent.

OTA is not aware of any published analyses of theissue of informed consent for research participationby special care unit residents. Much has been writtenabout this issue, however, as it pertains to nursinghome residents in general and individuals withdementia in any setting. In addition, several re-searchers who are part of the Workgroup on Re-search and Evaluation of Special Care Units arepreparing a paper on ethical issues in special careunit research that includes a discussion of informedconsent for research participation (495).

In the late 1970s and early 1980s, the NationalCommission for the Protection of Human Subjectsin Biomedical and Behavioral Research and thePresidents’ Commission for the Study of EthicalProblems in Medicine and Biomedical and Behav-ioral Research studied and made recommendationsabout informed consent for research participation bynursing home residents (322,350). Other commenta-

46 . Special Care Units for People With Alzheimer’s and Other Dementias

tors have also made recommendations on thisissue. 18

All of these recommendations arise from seriousconcerns about the potential exploitation of nursinghome residents as research subjects. They wouldstrictly limit the types of research that could beconducted in nursing homes and the participation ofresidents who are not capable of informed consent.The National Commission for the Protection ofHuman Subjects in Biomedical and BehavioralResearch recommended, for example, that researchinvolving nursing home residents should only beallowed if it is relevant to a condition the subjectssuffer from, i.e., therapeutic research, and only ifappropriate subjects cannot be obtained in any othersetting. Cassel recommended that surrogates shouldbe formally designated to make decisions aboutresearch participation on behalf of residents who arenot capable of informed consent (74).

None of these recommendations has been incor-porated into law, and no special regulations oninformed consent for research participation bynursing home residents are now in effect. OBRA-87gives residents the right to refuse to participate inresearch (463) but does not address the issue ofinformed consent for research participation. Thus,research in nursing homes is governed by the generalFederal law which allows consent for researchparticipation by a legally authorized representativeon behalf of an incompetent person. The term legallyauthorized representative is not defined in theFederal law.

In 1981, the National Institute on Aging spon-sored a conference to explore the legal and ethicalissues with respect to informed consent for researchparticipation by individuals with dementia in anysetting (301). After the conference, a task force drewup guidelines that recommend the use of noninstitu-tionalized subjects whenever possible (302). Federallaw requires institutions that receive Federal re-search funds to have an institutional review board(IRB) to review research proposals involving humansubjects, and the task force’s guidelines cite severalcriteria IRBs could use to evaluate the informedconsent procedures to be used in a given study. Theguidelines point out that the greater the risks posedby a study and the less likely an individual subject

will benefit directly, the more stringent the informedconsent procedures should be. These guidelines arenot part of any official regulations, however.

Researchers generally turn to a nursing homeresident’s family to obtain consent for researchparticipation. It is assumed the family’s decisionwill reflect the wishes and best interests of theresident. The one published study OTA is aware ofthat has addressed families’ decisions about researchparticipation by an elderly relative casts doubt onthat assumption. The researchers asked the familiesof 168 nursing home residents with dementia toconsent to the residents’ participation in a low-riskstudy of urinary catheters (480). About half thefamilies consented. Fifty-five of the families saidthey believed their relative would not consent toparticipate in the study, but17ofthe55(31 percent)consented anyway. Twenty-eight of the families saidthey would not choose to participate in the studythemselves, but 6 of the 28 (20 percent) consentedfor their relative with dementia to participate.

The preliminary findings of a similar study beingconducted by researchers at the University ofChicago are more positive. As of the spring 1992, theresearchers had interviewed 100 noninstitutional-ized individuals with mild to moderate dementia andtheir family caregivers (395). The individuals withdementia were asked whether they would participatein several hypothetical, high- and low-risk medicalstudies. The family caregivers were asked threequestions: whether they would consent for theirrelative with dementia to participate in the studies,whether they thought their relative would consent toparticipate, and whether they would be willing toparticipate themselves. Preliminary findings fromthe study show discrepancies between the responsesof the individuals with dementia and their familycaregivers, but the family caregivers generally havenot volunteered their relative with dementia forhigh-risk studies (395). In fact, the caregivers havebeen less willing than the individuals with dementiato consent to the individuals’ participation in high-risk studies. On the other hand, the family caregivershave been more willing than the individuals withdementia to consent to the individuals’ participationin the low-risk studies.

18 see, for example, Annas Wd GkMM, “Rules for Research in Nursing Homes” (13); Cassel, “Research in Nursing Homes: Ethical Issues” (73);Cassel, “Ethical Issues in the Conduct of Research in Long Term Care” (74); and Dubler, “Lcga.1 Issues in Research on Institutionalized DementedPatients” (122).


Numerous studies that have used hypotheticalscenarios to compare treatment decisions by elderlyindividuals and their families have found discrepan-cies between their responses (119,340,404,449,45 1,523). It has been suggested that family memberswould be more likely to make a treatment decisionthe way their elderly relative would make it if theywere specifically instructed to do so, and thefindings of one study support that suggestion (449).Even when families are asked specifically to makea decision the way their elderly relative would makeit, however, the decisions are not always the same(404,449).

If the necessary descriptive and evaluative re-search is to be conducted in special care units,informed consent procedures must be devised thatwill protect the residents from exploitation and at thesame time allow the use of research methods thatrequire informed consent, e.g., methods that involvedirect interaction with the residents. Some commen-tators have suggested the use of a durable power ofattorney for this purpose (13,302). With a durablepower of attorney, a person who is still capable ofmaking decisions for himself or herself can desig-nate someone to make decisions in the future whenhe or she is no longer capable. The problem with thisapproach is that most special care unit residentsprobably are not capable of executing a valid durablepower of attorney, and many will not have executeda durable power of attorney for research participa-tion at an earlier time when they were capable ofdoing SO.

Some special care units now require individualswith dementia to have a durable power of attorneyfor health care decisions prior to their admission tothe unit. Anecdotal evidence indicates that in somecases, these documents are being executed byindividuals who are not capable of making decisionsfor themselves (156). The same problem could arisewith a durable power of attorney for researchparticipation.

Other approaches that have been proposed are theuse of a nursing home council (13), a multidiscipli-nary nursing home committee (23,74), or an inde-pendent advocacy group (29) to approve and overseenursing home research, including the proceduresthat would be used to obtain informed consent.Certainly if a panel were established to allowwaivers for special care unit research, as suggested

earlier in this chapter, that panel could perform thesefunctions.

Lastly, it must be noted that although most specialcare unit residents probably are not capable of givingvalid informed consent, some are, and they should beasked. Preliminary findings of the ongoing Univer-sity of Chicago study of informed consent forresearch participation by noninstitutionalized indi-viduals with dementia show that many of theseindividuals are able to provide helpful informationabout their values and preferences, even though theyare not capable of giving valid informed consent(395). Some and perhaps many special care unitresidents may also be capable of providing suchinformation.

OTHER ISSUES OF IMPORTANCETO NURSING HOME RESIDENTS

WITH DEMENTIAThree additional issues are important for all

nursing home residents with dementia, includingspecial care unit residents. These three issues arediscussed briefly below.

The Availability of Physicians’ Services

Physicians’ services are essential for all nursinghome residents with dementia. Yet the special careunit literature contains little discussion of the role ofphysicians in special care units. With the exceptionof the Tennessee regulations, the existing Stateregulations for special care units do not mentionphysicians except to require that a physician approvea resident’s admission to the unit and document thereason for the admission. Requirements for ongoingphysician care appear in other sections of theseStates’ nursing home regulations and in the Federalregulations for Medicare and Medicaid certificationof nursing homes. The lack of such requirements inthe special care unit regulations implies, however,that physicians’ role is limited to admission-relatedfictions.

Clearly, the appropriate role of physicians in thecare of nursing home residents with dementia goesfar beyond admission-related functions. One of themost frequent complaints about the care of theseresidents is that acute and chronic illnesses thatexacerbate their cognitive impairments and reducetheir functioning often are not diagnosed or treated.Diagnosis and treatment of these illnesses will


reduce excess disability and improve the residents’quality of life, even if the conditions that cause theirdementia are incurable and progressive. Ongoingphysician involvement is essential to identify andtreat residents’ acute and chronic illnesses.

One stated objective of some special care units isto get away from the ‘‘medical model” of care andadopt a ‘‘social model’ instead. Semantics aside,this objective is unrelated to the role of physicians,who are as essential in a social as a medical modelof care (146). In special care units, as in nursinghomes generally, the physician may be a teammember rather than the team leader (226), but thereis no question about the need for initial and ongoingphysician involvement in the care of residents withdementia in special care units and other nursinghome units.

The Availability of Mental Health Services

Many commentators have noted the lack ofadequate mental health services in nursing homes(58,175,339,393). Although Alzheimer’s diseaseand most of the other diseases that cause dementiagenerally are not considered mental illnesses, theirmanifestations include mental, emotional, and be-havioral symptoms that may respond to behaviormanagement techniques, psychotropic medications,and other mental health treatments. Psychiatrists,psychologists, psychiatric nurses, psychiatric socialworkers, and other mental health professionals withexpertise in the evaluation and treatment of thesesymptoms seldom work in nursing homes.

The lack of adequate mental health services inmost nursing homes is attributable to several factors.One factor is a lack of reimbursement. A secondfactor is the IMD exclusion. As an optional Medicaidbenefit, States may choose to provide Medicaidreimbursement for the care of individuals under age22 or over age 65—but not individuals age 22 to65—in an institution for mental diseases (IMD).Medicaid regulations define an IMD as ‘‘an institu-tion that is primarily engaged in providing diagno-sis, treatment, or care of persons with mentaldiseases, including medical attention, nursing care,and related services’ (460). If a nursing home isclassified as an IMD, it loses Medicaid funding forall its residents age 22 to 65. If the nursing home isin a State that does not provide Medicaid reimburse-ment for care in IMDs, it loses Medicaid funding forall its residents. Because of a fear of being classified

as an IMD, some nursing homes choose not toemploy mental health professionals, not to providemental health services, or both (192,205).

Medicaid regulations cite 10 criteria to be used indetermining whether a facility is an IMD. No singlecriterion is definitive; rather, the criteria are to beused together to determine whether a facility’s‘‘overall character is that of a facility established andmaintained primarily for the care and treatment ofindividuals with mental diseases” (460). Two of thecriteria are troublesome to nursing homes that carefor individuals with dementia:

1)

2)

“The facility specializes in providing psychiatric/psychological care and treatment. This may beascertained through review of patients’ re-cords. It may also be indicated by the fact thatan unusually large proportion of the staff hasspecialized psychiatric/psychological trainingor by the fact that a large proportion of thepatients are receiving psychopharmacologicaldrugs” (460).“More than 50 percent of all the patients in thefacility have mental diseases which requireinpatient treatment according to the patients’medical records” (460).

The second criterion, often referred to as the “50percent rule,” excludes residents with senility ororganic brain syndrome “if the facility is appropri-ately treating the patients by providing only generalnursing care. ” According to the regulations, resi-dents with senility or organic brain syndrome areexcluded because these conditions “are essentiallyuntreatable from a mental health point of view’(460). Residents with senility or organic brainsyndrome are not excluded from the 50 percent rule“if the facility is treating these patients for theeffects of a mental disorder, as opposed to providinggeneral nursing and other medical and remedialcare” (460).

A third factor that may discourage the provisionof mental health services in nursing homes isPreadmission Screening and Annual Resident Re-view (PASARR), a program mandated by OBRA-87that requires States to: 1) screen all nursing homeapplicants and nursing home residents to determinewhether they have mental illness or mental retarda-tion, and 2) evaluate all those who are found to havemental illness or mental retardation to determinewhether they need nursing home care and whetherthey need ‘‘specialized services” for their mental


illness or mental retardation. Mentally ill andmentally retarded nursing home applicants andresidents who are found in a PASARR evaluationnot to need nursing home care or to need “special-ized services” must be placed elsewhere. Mentallyill and mentally retarded nursing home residentswho have been in a nursing home for 30 months ormore can choose to remain in the nursing home evenif they are found not to need nursing home care or toneed “specialized services” (320).

The impact of PASARR on the availability ofmental health services in nursing homes is unclearand probably differs from State to State. Anecdotalevidence suggests that at least in some States,PASAAR has had the same effect as the IMDexclusion—that is, to cause some nursing homes notto employ mental health professionals, not toprovide mental health services, or both, because ofa fear that if the facility employs mental healthprofessionals or provides mental health services, itwill be perceived as caring for mentally ill peopleand therefore lose Medicaid funding.

The Federal regulations for Medicare and Medic-aid certification of nursing homes include provisionsthat would seem to require the involvement ofmental health professionals in assessing residents’care needs and the provision of some mental healthservices. 19 It is unclear how these provisions will beinterpreted and implemented.

The American Association of Retired Persons(AARP) is currently funding a study of barriers tomental health care in nursing homes (260). Thestudy, which will be completed in 1993, will provideinformation about regulations, reimbursement, andother factors that interfere with access to mentalhealth services by all nursing home residents,including residents with dementia.

The Use of Psychotropic Medications

As noted earlier, a large proportion of nursinghome residents receive psychotropic medications,and residents with dementia are more likely thanother residents to receive these medications. Psy-chotropic medications are frequently referred to inthe special care unit literature and elsewhere aschemical restraints or pharmacological restraints.The use of the word restraints in this context implies

that psychotropic medications are an undesirabletreatment option. This implication fits well concep-tually with the growing concern about the overuseand inappropriate use of physical restraints andpsychotropic medications in nursing homes. On theother hand, many commentators have noted thatpsychotropic medications are a valuable treatmentoption for some individuals with dementia(19,28,121,180,277,347,353,367,381,402,412). Forindividuals with depressive or psychotic symptomsor extreme agitation, psychotropic medications maybe the best treatment option. The important consid-eration in these instances is the selection of the rightmedication, in the right dose, for the right indication.

Clearly, psychotropic medications should not beused as a substitute for behavioral or environmentalinterventions that may be as effective or moreeffective and do not have the negative side effectsoften associated with psychotropic medications.Research is needed to determine the indications,dosages, and long-term effects of various psy-chotropic medications. Referring to psychotropicmedications as restraints may create an atmospherein which individuals with dementia will not receivemedications that could significantly improve theirquality of life.

ALTERNATIVES TO SPECIALCARE UNITS

As noted at the beginning of this chapter, theproliferation of special care units is occurring at thesame time as numerous other government andnongovernment initiatives that are likely to improvethe care of nursing home residents with dementia orprovide alternatives to nursing home care for them.This section briefly describes a few of these initia-tives. Each of the initiatives offers an alternate wayof accomplishing one or more of the same objectivesas special care units.

Initiatives To Reduce The Use of PhysicalRestraints for All Nursing Home Residents

OBRA-87 and related legislation require nursinghomes to reduce their use of physical restraints. Priorto and since the implementation of the OBRAregulations, many organizations have developedtraining programs and materials to help nursing

19 see sectio~ 483.20@j(2)(lll) and (vii), 483.20(f), and 483.45(a), Federal Register, Sept. 9, 1991 (463).zo s=, for example, Rader, “The JoyM Road to Restraint- Free Care” (360).


homes reduce the use of physical restraints.20 TheNational Institute on Aging has funded a 3-yearclinical trial on reducing the use of physical re-straints in nursing homes, and the Food and DrugAdministration (FDA) has increased its surveillanceof restraining devices (327).21

In 1989, the Kendal Corp. in Pennsylvaniainitiated “Untie the Elderly, ” a national program tocreate ‘restraint-free’ nursing homes. In December1989, the corporation and the Senate Special Com-mittee on Aging cosponsored a policy-orientedsymposium on reducing the use of physical re-straints in nursing homes. The corporation alsosponsors workshops to help nursing homes reducetheir use of physical restraints and publishes anewsletter that describes the successful efforts ofsome nursing homes to decrease restraint use.

In 1991, the Jewish Home and Hospital for Agedin New York City initiated a three and a half year“Restraint Minimization Project, ” with fundingfrom the Commonwealth Fund. The project isintended to demonstrate ways of reducing restraintuse in nursing homes. It is being implemented in 14nursing homes in 4 States.

Nursing homes often use physical restraintsbecause they are afraid of being sued for fall-relatedinjuries to residents who are not restrained. Yethistorically, there has been a greater risk of facilitiesbeing sued for overuse or misuse of restraints(196,224). By establishing a clear standard of care,OBRA requirements for reduced use of physicalrestraints will increase the legal risks associated withtheir overuse or misuse.

As noted earlier, several studies have found thaton average physical restraints are used far less inspecial care units than in other nursing home units.It is unclear whether this difference will be sustainedas the implementation of OBRA-87 creates pressureon all nursing homes to reduce their use of physicalrestraints. The 481 nursing homes that responded toa 1991 surwey conducted by the American Associa-tion of Homes for the Aging reported that theproportion of their residents who were physicallyrestrained had decreased from an average of 43percent in 1989 to an average of 23 percent in 1991(9). Only 13 percent of the nursing homes reported

having instituted a restraint reduction programbefore 1989, the year the pertinent OBRA regula-tions went into effect.

Dementia Training Programs for NursingHome Staff Members

One of the most frequently cited problems in thecare of nursing home residents with dementia is lackof staff knowledge about dementia. Many organiza-tions and individuals have developed training pro-grams and materials to address this problem. Onevideo training program, ‘‘Managing and Under-standing Behavior Problems in Alzheimer’s Diseaseand Related Disorders,’ was funded by the NationalInstitute on Aging and has 10 training modules, eachfocused on a different behavioral symptom (439).Other programs and materials include the following:

●

●

●

●

●

●

a training manual developed by the St. LouisChapter of the Alzheimer’s Association (39);a training manual and tape series developed bythe Wisconsin Alzheimer’s Information andTraining Center (509);a video training program developed by Com-munity Services Institute, Inc. (102);a training guide and resource manual developedfor the New Jersey Department of Health (471);a video training program developed by ChurchHome and distributed by the American Associ-ation of Homes for the Aging (86); anda training manual written by Lisa Gwyther anddistributed by the Alzheimer’s Association andthe American Health Care Association (165).

These training programs and materials are likelyto improve the care of nursing home residents withdementia generally.

In 1987, the Alzheimer’s Family Center, Inc. ofSan Diego, CA, established a School of DementiaCare which trains and certifies health care profes-sionals to work with individuals with dementia(422). In 1991, the Federal Government providedfunding to the center through the Job Training andPartnership Act to train ‘‘Certified Nursing Assist-ant Alzheimer Care Specialists’ to work withindividuals with dementia in nursing homes, adultday centers, and other settings (324).

zo SW, for example, Rader, “The Joyful Road to Restraint- Free Care” (360).21 ~ J~e 1992, he ~A propoSed a new fie tit ~o~d r~fie ~be~g of physical res~ts. me req~ed label wo~d include directions fOr use

of the restraints, a warning of potential hazards, and the phrase prescription only.

Chapter 1--Overview and Policy Implications ● 51

Specialized Programs for Residents WithDementia in Nonspecialized Nursing

Home Units

Instead of or in addition to a special care unit,some nursing homes have specialized programs forresidents with dementia in nonspecialized units. It isunclear how many nursing homes have such pro-grams. In response to a 1991 survey of all U.S.nursing homes with more than 30 beds, 13 percentof the 1463 nursing homes that said they had aspecial care unit or program for their residents withdementia reported that the program was not in aphysically separate part of the facility (247). Thus,it is likely that at least several hundred nursinghomes have specialized programs.

Some nursing homes have specialized day care oractivity programs.

22 One facility established a‘ wan-

derer’s lounge” where specialized activities areprovided several hours a day for 15 to 20 dementedresidents of the facility’s nonspecialized units (299).

Rovner established an experimental special careprogram for demented residents of nonspecializedunits in one Maryland nursing home (387). Theprogram was intended to duplicate the essentialcomponents of an apparently effective special careunit described earlier in this chapter and in chapter4 (392). The special care program consisted ofweekly visits to each resident by a psychiatrist anda nurse with the purpose of identifying residents’cognitive impairments, treating psychiatric symp-toms, reducing medication side effects, maintainingresidents’ physical health, reducing the use ofphysical restraints, and increasing the residents’participation in activities (387). Five hours ofspecialized activities were provided daily. Thespecial care program is being evaluated. Its impactwill be compared with the impact of the special careunit described earlier to determine their relative costand effectiveness.

Specialized Living Arrangements OutsideNursing Homes

Outside nursing homes, special care units andother specialized living arrangements for peoplewith dementia have been established in residentialcare facilities, assisted living facilities, mentalhospitals, and other settings. Three of the bestknown special care units in the United States are inresidential care facilities:23

. the Alzheimer’s Care Center in Gardiner, ME(303);

. the Corinne Dolan Alzheimer’s Center at HeatherHill in Chardon, OH (317), and

. Wesley Hall in the Chelsea United MethodistRetirement Home in Chelsea, MI (105).

In many discussions about special care units, nodistinction is made between these three units andother model special care units in nursing homes.From a public policy perspective, however, there areimportant differences between special care units inresidential care facilities and special care units innursing homes. Residential care facilities are muchless regulated than nursing homes. The FederalGovernment does not regulate residential care facili-ties.24 States license various types of residential carefacilities (251), but some types of residential carefacilities are not licensed in each State, and thelicensing requirements, where they exist, are lesscomprehensive and far less stringent than thelicensing requirements for nursing homes.25

Since special care units in residential care facili-ties are not subject to the same kinds of regulatoryrequirements as special care units in nursing homes,they are able to implement innovative physicaldesign features, staffing arrangements, and patientcare practices that may be difficult or impossible toimplement in a nursing home. Because of theminimal regulatory requirements, special care units

22 S= for exmple, Clentiel and Fleishell, ‘An Akheimer Day Care Center for Nursing Home Patients” (89); lkmczqk~d Ba*, “Adven~eProgram” (173); Johnson and Chapmaq “Quest for Life” (21 1); and Sawyer and Mendelovitz, “A Management Program for AmbulatoryInstitutionalized Patients With Alzheirner’s Disease and Related Disorders” (400).

23 Thetermresz2ientia/ carefacWies refers to a variety Of hvtig arrangements that provide room and board and some degree of protective supervision.Examples are retirement homes, homes for the aged, group homes, and adult foster homes.

24 me o~y F~w~ role ~ the re~ation of residenti~ Cme f~ilities is through the Keys Amendment to the social security Act. The KeysAmendment requires States to certify to the U.S. Department of Health and Human Semices that all residential care facilities in which a significantnumber of Supplemental Security Lncome (SS1) recipients reside meet appropriate standards. A 1989 GAO report found that the department does littlemore than record the receipt of the certifications and that only four States were submitting the required certifications (453).

25 Reswch Tfi~gle ~ti~te fi NoM c~ol~ is conducfig a Smdy for the us. Dep~ent Of Health ~d H~ Services of State licensingrequirements and other State regulations for residential care facilities. In addition to a 50- State review of existing regulations, the study will comparethe quality of care provided in licensed and urdicensed residential care facilites in 10 States.


in residential care facilities usually cost less toconstruct and operate than special care units innursing homes. As a result, they usually charge lessthan nursing homes.

Despite these advantages, there are serious poten-tial problems with special care units in residentialcare facilities. Anecdotal evidence suggests thatmost of these units are established outside a nursinghome in order to avoid nursing home regulations(273). This may be entirely appropriate if the intentis to avoid regulatory requirements that restrict theuse of physical design or other features the unitoperator believes will benefit individuals with de-mentia; it is clearly inappropriate if the intent is toavoid regulatory requirements that are important forthe safety or well-being of individuals with demen-tia. Many government reports have documentedwidespread abuse, exploitation, and neglect ofelderly and other individuals in residential carefacilities. 26 Given the vulnerability of individualswith dementia, the proliferation of special care unitsin minimally regulated residential care facilitiesraises the prospect of severely deficient care.

Specialized living arrangements for people withdementia are also being developed in assisted livingfacilities. The term assisted living facilities refers toliving arrangements in which a variety of supportiveservices are available to residents who each have aseparate apartment that is lockable and has its ownkitchen (501). Some people consider assisted livingfacilities a type of residential care facility, and otherpeople consider them a separate category of livingarrangements. They are less likely to be regulatedthan other residential care facilities and thereforeprobably present greater potential for deficient care.27

Psychogeriatric units in public and private mentalhospitals often serve elderly individuals with de-mentia as well as elderly individuals with acute andchronic mental illnesses, but some mental hospitalshave units that serve only individuals with dementia.Such units exist, for example, in two Virginia statehospitals (56,252).

Lastly, some organizations have developed or aredeveloping campus-like settings that provide avariety of living arrangements and other specializedservices for individuals with dementia.28 The livingarrangements available in such settings may includeapartments for an individual with dementia and hisor her spouse, residential care or assisted livingunits, and nursing home units.

In addition to programs intended to improve thecare of nursing home residents with dementia orprovide alternate residential care options for them,many services have been developed to assist individ-uals with dementia who are living at home and theircaregivers. These services include adult day care,respite care, specialized hospice programs, and avariety of other in-home and community-basedservices. All these programs and services providealternatives to special care units for some peoplewith dementia. Government policies for special careunits should be considered in relation to the fullrange of care options for these individuals.

CONCLUSION

A large number of nursing home residents in theUnited States have dementia--637,600 to 922,500according to national surveys-and almost all peo-ple with dementia will probably spend some time ina nursing home in the course of their illness. Theseindividuals may receive inappropriate care that willresult in excess disability and severely reducedquality of life.

Special care units of various types have beendeveloped and are proliferating in response to thisproblem. Special care units promise to provide bettercare for individuals with dementia than these indi-viduals would receive in other nursing home units.It is unlikely all nursing home residents withdementia will ever be cared for in special care units,but methods of care developed in special care unitscould eventually be implemented in other nursinghome units as well.

26 See, forexample, “Board and Care Homes in America: A National Tragedy” (455), and ‘Board and Care: LnsufflcientAssurances ThatResidents’Needs are Identified and Met” (453).

27 Gegon has developed special regdat.ions for assisted living facilities. In 1987, the State Medicaid progran ~gan pa@g for cm in designatedassisted living facilities for individuals who are eligible for Medicaid-funded nursing home care (501). One of these facilities serves individuals withdementia (504).

28 s=, for e~ple, Stein Gerontological ceut~, “Pathways: Program Development Plan” (423).

Chapter L-Overview and Policy Implications ● 53

Better methods of care for nursing home residentswith dementia are likely to benefit not only thoseresidents, but also their families, the nursing homestaff members who take care of them, and othernursing home residents who are not demented.Families will benefit because they will be moresatisfied with the care provided for their relativewith dementia and therefore may feel less guiltyabout having placed the individual in a nursing homeand less anxious about his or her well-being. Nursinghome staff members will benefit because the resi-dents are likely to be easier to manage. Nonde-mented nursing home residents will benefit becausethe behavioral and other symptoms of residents withdementia are often disturbing to them; better meth-ods of care are likely to reduce the incidence of thesesymptoms and thus improve the quality of thenondemented residents’ lives.

The number of nursing homes that have a specialcare unit is increasing rapidly. OTA estimates that10 percent of all U.S. nursing homes had a specialcare unit in 1991.

Existing special care units vary greatly in virtuallyall respects. Although experts agree about thetheoretical principles of specialized dementia care,the theoretical principles are implemented differ-ently in different special care units and are notimplemented at all in some special care units, andthere is considerable disagreement about the particu-lar features that are necessary in a special care unit.

Proponents of special care units make strongclaims about their effectiveness, but the availableresearch provides little support for the claims. Onlytwo of the six special care unit studies that used acontrol group found any positive outcomes forspecial care unit residents. Only one of the fourstudies that measured the impact of a special careunit on the unit staff members and only two of thefour studies that evaluated the effect of special careunits on the residents’ families found any positiveoutcomes. None of these studies is definitive byitself, but their combined findings are impressiveand suggest that we do not yet know exactly whatconstitutes effective nursing home care for individu-als with dementia.

Because of the diversity of existing special careunits, their rapid proliferation, and the widespreadperception that some special care units use the wordsspecial care as a marketing tool and actually provideno special services for their residents, there is strong

pressure to regulate special care units. On the otherhand, given the lack of agreement among expertsabout the particular features that are necessary in aspecial care unit and the lack of research-basedevidence of the effectiveness of special care units, itis difficult to determine what regulations should saybeyond general statements about goals and princi-ples and a listing of issues that require specialconsideration in the care of residents with dementia,e.g., staff training, environmental design, security,activity programs, family involvement, and residentrights.

Special care unit regulations are likely to discour-age innovation by suggesting that we already knowwhat constitutes effective care for nursing homeresidents with dementia. Regulations are also likelyto lock in for the future current beliefs about thefeatures that are important in special care units.

OTA concludes that the objective of improvingnursing home care for individuals with dementiawill be better served at present by initiatives todevelop greater knowledge and agreement about theparticular features that are important in the care ofnursing home residents with dementia than by theestablishment of regulations for special care units.Some people argue that we cannot wait for theresults of such initiatives to develop special care unitregulations. It is said that regulations are needed nowto protect individuals with dementia from poor-quality care. In contrast, OTA concludes that OBRA-87 provides a sufficient basis for censuring units thatprovide poor-quality care, without any special regu-lations. It is also said that regulations are needed toprotect individuals with dementia and their familiesfrom nursing homes that fraudulently claim toprovide special care but actually provide nothingspecial for their residents. OTA concludes thatindividuals with dementia and their families can bebetter protected from these nursing homes byinitiatives that would: 1) make available guidelinesthat describe the theoretical concepts and design andother features that are believed to be important inspecial care units, 2) make available informationabout the characteristics of special care units in localjurisdictions, and 3) require nursing homes todisclose to families and others what is special abouttheir special care unit. As noted earlier, theseinitiatives will not protect all potential special careunit residents and their families from nursing homesthat provide no special services in their special careunit. Neither will these individuals be protected by

54 . Special Care Units for People With A[zheimer’s and Other Dementias

regulations that require special care units to incorpo-rate features that have not been shown to beeffective.

The potential of special care units to developbetter methods of care for nursing home residentswith dementia is exciting. That potential cannot berealized without a greater commitment than cur-rently exists to evaluation of the units and their

impact on residents, residents’ families, unit staffmembers, and nondemented nursing home residents.Such evaluation must be pursued with the recogni-tion that some of the features that are currentlybelieved to be essential in special care units may notbe effective and that once effective methods of careare identified, they may not be unique to individualswith dementia.

Documents

Special Care Units for People With Alzheimer's and Other