Full text of DOI:10.1093/geront/40.4.422

Copyright 2000 by
The Gerontological Society of America
This study contrasts rates of mortality and of relocation to higher levels of care as well as
trajectories of cognitive status, functional ability, depression, and subjective health of
residents of an assisted living facility with those of a nursing home. Data were collected from
medical records and face-to-face interviews with 158 residents at baseline and 4, 8, and 12
months later. All participants lived on a single long-term care campus. Logistic regression
revealed that facility was not a significant predictor of mortality or relocations due to
declining health. A repeated measures analysis of variance found that outcomes for people
living in the two facilities did not change at different rates. These consistent findings suggest
that although the assisted living and nursing home environments claim to have different
philosophies of care, health outcome patterns for people living in the two environments
were similar.
The Gerontologist
Vol. 40, No. 4, 422–428
Key Words: Assisted living, Nursing homes, Mortality, Relocation
The Effect of Long-Term Care Environments on
Health Outcomes
Rachel A. Pruchno, PhD,1 and Miriam S. Rose, MEd2
Assisted living facilities have become an increas-
ingly popular form of long-term care. Developed in
response to the growing need for an intermediate
level of structured care and marketed toward seniors
who require more care than is usually provided at
home, but less care than is traditionally provided in a
nursing home, the number of assisted living facilities
in this country has grown exponentially in the past
decade. Although assisted living facilities are often
promoted as cost-effective environments for older
people who need help and cannot live on their own
(Fonda, Maddox, Clipp, & Reardon, 1996; Morton,
1995), little is known about how people living in as-
sisted living environments actually fare when com-
pared with those who live in nursing homes. Assisted
living facilities are frequently billed as promoting in-
dependence and health maintenance, with the impli-
cation that their morbidity and mortality rates may be
lower than those in nursing homes. Whether assisted
living facilities are actually superior to nursing homes
in terms of outcomes such as mortality and morbidity
remains a matter of debate, however. Nyman (1994),
for example, maintains that research has not yet dem-
onstrated that assisted living residents exhibit slower
decline in physical and cognitive abilities than nurs-
ing home residents. In addition, the shift away from
the medical model exemplified by assisted living fa-
cilities could result in “deteriorated health conditions
and increased morbidity and mortality of residents,”
(p. 7) because assisted living facilities are not as
highly regulated as nursing homes (Kane & Wilson,
1993). In cases in which this results in fewer and less
qualified staff, there may be detrimental effects on
frail older people. The analyses that follow contrast
the mortality rates and need for higher levels of care
as well as the trajectories of cognitive status, depres-
sion, functional ability, and subjective health of peo-
ple residing in an assisted living facility with those of
people living in a traditional nursing home.
The concept of assisted living facilities is both old
and new. Although policy makers and social plan-
ners for years accepted the rather simplistic dichot-
omy between independent living and long-term care
institutionalization, in 1963 congregate housing sur-
faced as a combination of communal living and ser-
vices for low-income, frail elderly people. The early
congregate housing environments were positioned
between the more protective environments of skilled
and intermediate care facilities and largely indepen-
dent environments such as retirement communities,
private houses, and apartments. They were devel-
oped in response to a trend of gradually declining
levels of physical, functional, and mental ability of
many older people. However, the majority of these
facilities did not have the capacity to meet the ever-
increasing declines experienced by many of their res-
idents. The new assisted living facilities of the 1990s,
providing a combination of housing and services, fall
under the umbrella of congregate housing, yet serve
a more frail group of people than did congregate liv-
ing facilities of the past.
This research was supported by grants from the Retirement Research
Foundation and the David and Inez Myers Foundation, and by Menorah
Park Center for Senior Living. We thank the administrators, staff, and resi-
dents of Stone Gardens and Menorah Park for their cooperation with the
study.
Freed from many of the regulations that constrain
nursing homes, assisted living facilities are positioned
to move away from the “medical model” of care, to-
ward a new paradigm for providing care to frail older
persons. Assisted living is a model encompassing
housing with services provided differently than in a
1Address correspondence to Rachel A. Pruchno, PhD, Center for Work
& Family, Boston College, 885 Centre Street, Newton, MA 02459. E-mail:
pruchno@bc.edu
2Myers Research Institute, Menorah Park Center for Senior Living,
Beachwood, OH.
422
The Gerontologist

nursing home. Such facilities seek to market them-
selves as noninstitutional, homelike environments
with a philosophy of maximizing autonomy, choice,
privacy, well-being, independence, and continuation
of normal lifestyles. Yet, some critics fear that the
move from the medical model could be accompa-
nied by declining health and increased mortality of
residents because of the possibility of less supervision
and less qualified staff. Advocates for older disabled
people are concerned that this could result in a
stripped-down, less well-staffed version of traditional
nursing home care without the compensating ameni-
ties and benefits.
In the analyses that follow, we focus first on exam-
ining patterns of mortality and health declines requir-
ing more intense care for people living in an assisted
living facility and those living in a nursing home,
controlling for initial differences in the two groups.
We then contrast trajectories of cognitive status,
functional ability, depression, and subjective health
for people living in an assisted living facility with
those living in a nursing home.
Methods
Setting and Sample
Literature regarding assisted living facilities is scant,
with much of it describing the philosophy of assisted
living (Kalymun, 1990; Mollica, Ladd, Dietsche, Wil-
son, & Ryther, 1992) or the architectural and interior
design of such facilities (e.g., Regnier, Hamilton, &
Yatabe, 1991). Although to date there have been no
studies focusing on adaptation to life in assisted liv-
ing facilities, several studies have evaluated the ef-
fects of earlier forms of congregate housing on resi-
dents. Research by Sherwood, Morris, and Ruchlin
(1986) concluded that congregate housing had been
effective in preventing unnecessary institutionaliza-
tion and was a cost-effective alternative to institutional
care. Several other investigations have confirmed the
positive effects of planned housing for elderly adults
using a variety of measures, including housing satis-
faction, general life satisfaction, involvement in com-
munity and on-site activities, and the quality of socio-
behavioral relations (Carp, 1976; Hinrichsen, 1985;
Lawton, 1982; Lawton & Cohen, 1974; Sherman, 1985;
Sherwood, Greer, & Morris, 1979).
The effect of environment is important to examine
in light of the relationship between the environment
and a person’s “competence” in areas such as health
and social role performance, as described by the “en-
vironmental docility” hypothesis (Lawton & Simon,
1968). According to this hypothesis, as people age
and their competence declines, more of their behav-
ior may be attributable to environmental, as opposed to
personal, characteristics (Lawton & Nahemow, 1973).
This hypothesis and the finding by Kane and Wilson
(1993) that some residents of assisted living facilities
and some residents of nursing homes may be similar
to one another in terms of their physical and mental
health, combine to provide a natural laboratory for
examining the effects of environment on behavior.
A factor that is often responsible for determining
whether an individual who meets the admission cri-
teria for assisted living receives care in a nursing
home or in an assisted living facility is the ability to
pay for the needed care. Most residents of assisted
living facilities pay privately for their accommoda-
tions, as Medicaid reimbursement is largely unavail-
able for assisted living. As such, in many states those
frail older people who cannot afford to pay privately
for care in an assisted living facility may be forced to
reside in a nursing home, even though they do not re-
quire 24-hr nursing care and could be cared for ap-
propriately in an assisted living environment.
This study was conducted on the campus of a sub-
urban Cleveland, Ohio, nonprofit organization offer-
ing a continuum of care for older adults, including a
350-bed nursing home and a newly opened 66-apart-
ment assisted living residence. The nursing home and
assisted living facilities are located within a city block
of one another on the same campus, which has one
executive director and board of directors. The facili-
ties’ staff are separate, staffing patterns are distinct,
and each environment has its own managing admin-
istrator and maintains a separate waiting list. (The as-
sisted living facility residents are given priority over
community residents on the nursing home waiting
list, however.) The purpose of the assisted living facil-
ity is specifically stated to be providing services “. . .
for people who need assistance managing some as-
pects of daily living but who do not need nursing
home care” and the facility is “. . . dedicated to [resi-
dents’] independence, their well-being and their dig-
nity.” The nursing home’s philosophy of care, which
does not speak to independence, emphasizes quality
of life within a health care context.
Criteria for residence in the assisted living facility
and on the nursing pavilion for high-functioning indi-
viduals from which participants for this study were
drawn were similar. They included independence in
mobility with or without the use of assistive devices;
bowel and bladder continence; need for minimal as-
sistance with activities of daily living (ADLs); general
independence in care needs; a Mini-Mental State
Exam (MMSE; Folstein, Folstein, & McHugh, 1975)
score of at least 15; and alertness and orientation to
person, place, and time. Facility staff used both clini-
cal judgment and standardized assessment measures,
such as MMSE scores and ADL abilities, to establish
whether residents met these criteria. Those who were
eligible for admission to either facility, along with their
family members, were able to choose between them,
following the opening of the assisted living facility.
All participants in the study were long-stay residents;
none were admitted for short-term rehabilitation.
The analyses that follow are based on data col-
lected from 158 residents, 76 from the nursing home
and 82 from the assisted living residence. All persons
participating in the study were White and Jewish.
People meeting eligibility requirements for the study
were invited to participate, only a minority of whom
declined to do so. A group of 13 research assistants
Vol. 40, No. 4, 2000
423

were trained by senior researchers to administer the
study protocol. Face-to-face interviews were con-
ducted with each resident at the start of the study and
then 4 (Time 2), 8 (Time 3), and 12 months after the
baseline interview (Time 4). Some participants were
assessed by different interviewers during the study
period because of personnel changes. Over the course
of time, 23 residents of the nursing home and 19 res-
idents of the assisted living facility were lost to the
study because of either death (57%) or a refusal to
continue to participate (43%). Contrasts of these 42
people with the 116 for whom complete longitudinal
data were available indicated that the groups were
similar in terms of age, gender, pay status (private vs.
Medicaid), education, marital status (widowed or not),
length of stay, depression at baseline, and subjective
health at baseline. People lost to follow-up had signifi-
cantly lower cognitive ability, t(120) ϭ Ϫ2.91, p Ͻ
.01, and lower functional ability, t(128) ϭ Ϫ2.12, p Ͻ
.05, at baseline than people who participated in all
four waves of data collection.
Pay status was coded as either private pay (0),
characterizing 75% of the sample, or Medicaid (1),
characterizing 25% of the sample. This information
was derived from resident records. Slightly more than
a quarter of the sample (27%) had not graduated
from high school, 42% were high school graduates,
and 31% had furthered their education beyond high
school. Length of stay ranged from 1 day to 13.3
years. Because the variable was highly skewed, it
was dichotomized and treated as a categorical vari-
able with values of 0–14 days (68%) and longer than
2 weeks (32%).
Relocation of participants was tracked by project
staff using facility census reports. These permanent
room changes were made in both the assisted living
facility and the nursing home when a resident’s func-
tional and health status declined. Resident deaths
were also tracked by project staff using facility census
reports. Census reports were reviewed for a period of
15 months, enabling the number of outcome experi-
ences to be maximized in the relatively small sample.
Results
Measures
As indicated in Table 1, residents of the two facili-
ties were similar in terms of their age, gender, and
marital status (widowed or not, given the prevalence
of widowhood among older persons). The partici-
pants in the sample ranged in age from 60 to 101
years; 75% were female; 73% were widowed. Resi-
dents of the two facilities differed with respect to ed-
ucation, length of stay, and pay status (private vs.
Medicaid). Residents of the assisted living facility
were more likely to be better educated and privately
responsible for paying for their long-term care. The
entire assisted living sample consisted of new admis-
sions, compared with 34% of the nursing home resi-
dents. There were no differences in the cognitive sta-
tus of residents of the two facilities at baseline.
Residents of the assisted living facility had higher
functional ability scores at baseline than did residents
of the nursing home, however, and there were also
significant differences between residents of the two
facilities with respect to depression, with nursing
home residents indicating more depressive symp-
toms. In addition, residents of the assisted living facil-
ity rated their overall health as better than residents
of the nursing home.
Cognitive status was measured using the MMSE
(Folstein et al., 1975). Scores on the scale could
range from 0 to 30, with a higher score indicating
better cognitive functioning. Internal consistency for
the scale as measured by Cronbach’s alpha ranged
from .73 to .85 over the four points in time.
Functional ability was measured by the Multilevel
Assessment Instrument (MAI) developed by Lawton,
Moss, Fulcomer, and Kleban (1982). The scale con-
sists of seven ADLs (eating, dressing, grooming, get-
ting around home, getting in and out of bed, bathing,
and toileting) and eight instrumental ADLs (house-
work, laundry, preparing meals, grocery shopping,
getting to places out of walking distance, using the
telephone, managing money, and taking medica-
tion). Residents were asked if they could perform
each activity without help, with some help, or if they
were completely unable to perform it. Scores at each
time could range from 15 to 45, with higher scores
indicating better functional ability (i.e., less help
needed to perform the activities). Internal consistency
for the scale as measured by Cronbach’s alpha
ranged from .80 to .86.
Depression was measured using the Center for Ep-
idemiological Studies Depression Scale (CES-D; Rad-
loff, 1977). Higher scores indicate more depressive
symptoms. Coefficient alpha for the scale ranged
from .86 to .88 over the four points in time.
Subjective health was measured using a 4-item
scale developed as part of the MAI (Lawton et al.,
1982). These questions asked residents to rate their
overall health, compare the state of their health to
that of 3 years ago, assess the extent to which health
problems prevented them from doing the things they
wanted to do, and rate their health in comparison to
peers. Higher scores indicate better self-rated health.
Cronbach’s alpha for the scale ranged from .53 to .70
over the four points in time.
Mortality and Relocation
Ten residents of the assisted living facility and 14
residents of the nursing home died over the 15-
month period. The mortality rates for the two facili-
ties were not significantly different from one another,
2
␹ (1, N ϭ 158) ϭ 1.19. In the assisted living facility,
20 people were relocated over the course of 15
months, all to a nursing home. In the sample of nurs-
ing home residents, 16 people were relocated, all to
units within the nursing home providing significantly
greater assistance with resident care than the unit in
which they resided when the study began. The two fa-
424
The Gerontologist

Table 1. Baseline Contrasts of Nursing Home and Assisted Living Facility Residents
Total sample
Nursing home
Assisted living
Variable
(N ϭ 158)
(n ϭ 76)
(n ϭ 82)
Difference test
Age (in years)
t(156) ϭ 1.26
M
SD
86.82
6.04
87.45
6.31
86.24
5.75
Women (%)
t(156) ϭ 1.02
M
SD
75
43
79
41
72
45
Widowed (%)
t(156) ϭ 1.52
M
SD
73
44
79
41
68
47
Education (in years)
M
SD
Length of stay (in days)
M
SD
Private pay (%)
M
SD
MMSE
M
SD
Functional ability
M
SD
CES-D
M
SD
t(156) ϭ Ϫ3.23**
t(75) ϭ 6.05***
t(75) ϭ 9.13***
t(120) ϭ Ϫ0.64
t(128) ϭ Ϫ6.94***
t(125) ϭ 2.09*
t(131) ϭ Ϫ2.22*
12.21
3.35
11.34
3.02
13.01
3.46
369.81
848.89
764.68
1097.29
3.83
3.19
75
44
47
50
100
0
23.30
4.13
23.09
4.06
23.57
4.25
37.42
4.93
34.89
4.36
40.03
4.06
15.94
11.12
17.98
11.10
13.92
10.85
Subjective health
M
SD
8.47
2.08
8.07
1.94
8.86
2.15
Notes: MMSE ϭ Mini-Mental State Exam; CES-D ϭ Center for Epidemiological Studies Depression Scale.
*p Ͻ .05; **p Ͻ .01; ***p Ͻ .001.
cilities’ rates of relocation also were not significantly
different from one another, ␹ (1, N ϭ 158) ϭ .25.
predicted by education (more), pay status (private
pay), functional ability (worse), and cognitive status
(worse) at baseline. Facility was not a significant pre-
dictor of these relocations.
2
Logistic regression analysis was used to estimate
the effects of facility on relocation to a higher level of
care and mortality during the 15-month follow-up
period. These analyses took into account the influ-
ence of variables on which residents of the assisted
living facility and nursing home differed from one an-
other at baseline (education, length of stay, pay sta-
tus, functional ability, subjective health, and depres-
sion) as well as variables that were correlated at the
bivariate level with the mortality or relocation out-
comes (age and cognitive status). Each logistic regres-
sion analysis included two steps. In the first step, all
of the variables on which residents of the two facili-
ties differed, as well as the variables having a signifi-
cant bivariate relationship with the outcome vari-
ables, were entered. In the second step, a variable
designating facility (assisted living vs. nursing home)
was entered. Cases with missing data for any of the
variables used in these analyses were excluded.
Outcome Trends
Preliminary trend analyses examined the way in
which the four health outcome variables were impacted
Table 2. Logistic Regression Analysis Predicting Mortality
(n ؍ 109)
Step 2
Variable
B
SE B
Education
Pay status
Ϫ0.05
Ϫ0.96
0.00
.10
.82
.00
.08
.18
.03
.06
.08
.88
Length of stay
Functional ability
Subjective health
Depression
Age
0.02
Results of the logistic regression analyses, pre-
sented in Tables 2 and 3, were nearly identical at
each step. As indicated, the sole significant predictor
of mortality was age. Facility (nursing home vs. as-
sisted living) had no effect on mortality. Relocation
resulting from the need for a higher level of care was
Ϫ0.07
Ϫ0.02
0.14*
Ϫ0.14
Ϫ1.50
Cognitive status
Facility
*p Ͻ .05.
Vol. 40, No. 4, 2000
425

Table 3. Logistic Regression Analysis Predicting Relocation
Information regarding the distribution of scores on
cognitive status, functional ability, depression, and
subjective health at each time period may be found
in Table 4. Results from the repeated ANOVAs are
presented in Table 5. These analyses suggest that
there were no time, Time ϫ Facility, or facility effects
for cognitive status and depression. For functional
ability, there was a significant main effect for facility,
with residents of the assisted living facility having
higher (better) mean scores across time than residents
of the nursing home. For subjective health, there was
a significant linear main effect of time, with residents
of both the nursing home and the assisted living facil-
ity reporting that they perceived their health to be
better over time. There were no significant interac-
tion effects involving facility and any of the outcome
variables.
(n ϭ 109)
Step 2
Variable
B
SE B
Education
Pay status
0.17*
Ϫ1.73*
0.00
Ϫ0.16*
0.15
Ϫ0.03
0.00
Ϫ0.18*
0.39
.08
.87
.00
.07
.14
.03
.05
.07
.74
Length of stay
Functional ability
Subjective health
Depression
Age
Cognitive status
Facility
*p Ͻ .05.
Discussion
by the demographic variables on which residents of the
assisted living facility and the nursing home differed
at baseline (education, length of stay, and pay status).
For example, in one such analysis, education—cate-
gorized as less than 12 years, high school diploma,
or beyond high school—was the between-subjects
factor, and cognitive status scores at the four points
in time were the within-subjects factors and depen-
dent variables. In this case, whether the interaction of
Time ϫ Education was significant was noted.
These repeated analyses of variance (ANOVA) in-
dicated no interaction effects over time between edu-
cation and cognitive status, F(3,6) ϭ .22, p ϭ .97,
functional ability, F(3,6) ϭ 2.07, p ϭ .06, depression,
F(3,6) ϭ .08, p ϭ .99, and subjective health, F(3,6) ϭ
.98, p ϭ .44. Length of stay was not found to be re-
lated over time to cognitive status, F(3,3) ϭ .04, p ϭ
.99, depression, F(3,3) ϭ 1.46, p ϭ .23, or subjective
health, F(3,3) ϭ 2.46, p ϭ .06. There was a signifi-
cant interaction, however, between length of stay
and functional ability measured over time F(3,3) ϭ
3.76, p Ͻ .05, such that new admissions experienced
a decline in functional ability, whereas veteran resi-
dents did not. Pay status had no interactive effects
over time with cognitive status, F(3,3) ϭ 1.52, p ϭ
.21, functional ability, F(3,3) ϭ 2.60, p ϭ .053), de-
pression, F(3,3) ϭ .44, p ϭ .72, or subjective health,
F(3,3) ϭ 1.79, p ϭ .15. Results from these analyses
suggested the importance of including length of stay,
but not education or pay status, as a covariate in the
subsequent trend analyses.
Taken together, these data suggest very similar
outcomes over time for these nursing home and as-
sisted living facility residents, all of whom met similar
eligibility criteria for admission. Facility did not influ-
ence patterns of mortality or relocation due to declin-
ing health, or trends in cognitive ability, functional
ability, depression, and subjective health.
Although these findings are compelling in their
consistency, their generalizability is limited and their
statistical power is restricted for a number of reasons.
First, data for this study derive from a small, homoge-
neous sample (all were White and Jewish). Second,
with both facilities being part of a larger parent orga-
nization, it is possible that the assisted living and
nursing home environments were not actually as dis-
Table 4. Outcome Measure Scores for Nursing Home and
Assisted Living Residents (n ϭ 116)
Nursing home
Assisted living
(n ϭ 53)
(n ϭ 63)
M
SD
M
SD
Cognitive status
Baseline
Time 2
Time 3
Time 4
23.13
23.33
22.53
22.60
3.68
4.16
4.64
4.52
24.12
24.40
23.88
23.64
4.06
5.32
4.33
5.40
Functional ability
Baseline
Time 2
Time 3
Time 4
34.65
35.65
35.22
35.59
4.59
4.35
4.22
3.79
40.72
39.63
39.18
39.23
3.62
3.71
3.98
3.77
To examine the way in which people living in the
nursing home and in the assisted living facility
changed over time and to address directly the ques-
tion of how these living environments impacted cog-
nitive status, functional ability, depression, and sub-
jective health, a separate repeated ANOVA was
conducted for each outcome measure. These analy-
ses examined the main effect of time as well as the
interaction of Time ϫ Facility for each outcome vari-
able. Significant findings for the interaction terms
would suggest different trends for residents of the
nursing home and those of the assisted living facility.
Depression
Baseline
Time 2
Time 3
Time 4
16.38
15.91
14.15
13.74
11.63
11.76
10.96
10.70
13.69
12.05
12.77
13.08
10.58
8.07
8.17
10.28
Subjective health
Baseline
Time 2
Time 3
Time 4
8.03
8.50
9.18
8.53
1.90
2.00
2.24
2.43
8.63
9.24
9.37
9.46
2.17
1.43
1.79
1.69
426
The Gerontologist

Table 5. Repeated Analysis of Variance for Cognitive Status, Functional Ability, Depression, and Subjective Health (n ϭ 116)
Cognitive status Functional ability Depression Subjective Health
Source
df
F
df
F
df
F
df
F
Within subjects
Time
3
3
3
1.33
0.14
0.18
3
3
3
2.01
1.46
2.43
3
3
3
1.04
0.94
0.25
3
3
3
5.45**
2.13
0.89
Time ϫ Length of Stay
Time ϫ Facility
Error (time)
156
(4.18)
222
(5.04)
210
(48.37)
216
(1.52)
Between subjects
Length of stay
Facility
Error
1
1
52
0.26
1.12
(70.41)
1
1
74
0.01
14.91***
(49.70)
1
1
70
0.84
0.02
(278.76)
1
1
72
0.14
0.61
(10.85)
Note: Values in parentheses represent mean square errors.
**p Ͻ .01; ***p Ͻ .001.
similar from one another as their stated philosophies
of care might lead one to believe. Even more serious
is the fact that the parent institution is one of the pre-
mier long-term care facilities in the United States.
Third, results from this study are generalizable only
to that subset of older persons who would qualify for
both assisted living care and nursing home care. Per-
sons at this intersection are the highest functioning
nursing home residents and the lowest functioning
residents of assisted living facilities.
Another potential difficulty with these analyses is
the effect due to length of stay. Although new admis-
sions constituted the entire assisted living sample,
they represented only about a third of the nursing
home sample. Even though length of stay had a sig-
nificant interaction with the functional ability out-
come over time, it was not a significant predictor in
any of the analyses. It may be argued that length of
stay and facility are hopelessly confounded, yet the
fact that neither variable predicted any of the out-
comes adds credence to the lack of findings of facil-
ity differences.
It may also be argued that relocation could have
been influenced by the availability of a continuum of
care on the parent organization’s campus. Some be-
lieve that an assisted living facility affiliated with an
organization offering multiple levels of care would
be more likely to relocate residents whose health de-
clined, compared with a freestanding facility trying to
maintain full occupancy. In a national study of as-
sisted living, however, administrators of facilities on
multilevel campuses reported a resident case mix
similar to those of freestanding facilities (Hawes,
Rose, & Phillips, 1999). In addition, because the two
study facilities had substantial waiting lists, which for
the nursing home consisted of both community resi-
dents and current residents of the assisted living facil-
ity, and because the facilities both maintained full
occupancy, it was unlikely that the management of
the assisted living facility felt compelled to keep resi-
dents after their health declined markedly. At the
same time, because the assisted living residents were
almost all paying privately for their care, there does
not seem to have been a financial incentive for them
to be relocated. All of the relocated assisted living
residents moved to a nursing home, where the care is
more expensive and private resources would be ex-
hausted more quickly.
Nonetheless, results from this unique opportunity
to contrast the nursing home and assisted living expe-
riences in a controlled environment set the stage for
similar studies. Future research of this nature involv-
ing larger numbers of residents in multiple facilities
has the potential to add to our understanding of the
fit between older people and different long-term care
environments and its effect on outcomes. If similar in
their findings, such studies might also provide a com-
pelling rationale for altering existing funding mecha-
nisms for long-term care. With similar morbidity and
mortality trajectories characterizing residents of as-
sisted living and nursing home facilities, shifting
Medicaid funding from nursing homes to assisted liv-
ing facilities could be a promising strategy for reining
in runaway long-term care costs. Less intensely regu-
lated than nursing homes, assisted living facilities are
considerably less expensive to operate and seem to
have similar effects on their residents. Kingsley and
Struyk (1991) suggested that the 1990s would be an
important time in the evolution of U.S. housing pol-
icy. Central to the policy goals is enabling the frail
elderly to remain in the community for as long as
possible. If, as these data suggest, people living in as-
sisted living facilities and nursing homes have similar
trajectories of morbidity and mortality, care in as-
sisted living facilities would seem to be a viable alter-
native to nursing homes for some individuals.
References
Carp, F. M. (1976). Housing and living environments of older people. In
R. H. Binstock & E. Shanas (Eds.), Handbook of aging and the social
sciences (pp. 244–271). New York: Van Nostrand Reinhold.
Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). Mini-Mental State:
A practical method for grading the cognitive state of patients for the cli-
nician. Journal of Psychiatric Research, 12, 189–198.
Fonda, S. J., Maddox, G. L., Clipp, E., & Reardon, J. (1996). Design for a lon-
gitudinal study of the impact of an enhanced environment on the
functioning of frail adults. Journal of Applied Gerontology, 15, 397–413.
Hawes, C., Rose, M., & Phillips, C.D. (1999). A national study of assisted
Vol. 40, No. 4, 2000
427

living for the frail elderly; Executive summary: Results of a national sur-
vey of facilities. Beachwood, OH: Menorah Park Center for Senior
Living, Myers Research Institute.
(1992). Building assisted living for the elderly into public long term
care policy: A guide for states. Portland, ME: Center for Vulnerable
Populations, National Academy for State Health Policy and Brandeis
University.
Hinrichsen, G. A. (1985). The impact of age-concentrated, publicly as-
sisted housing on older people’s social and emotional well-being.
Journal of Gerontology, 40, 758–760.
Kalymun, M. (1990). Toward a definition of assisted living. In L. A. Pasta-
lan (Ed.), Optimizing housing for the elderly: Homes not houses. New
York: Haworth Press.
Morton, A. (1995). Camouflaging: For assisted living providers, the chal-
lenge is providing care discreetly. Contemporary Long Term Care, 18(7),
40–47.
Nyman, J. A. (1994). Assisted living: Will it reduce long-term care costs?
Journal of Aging & Social Policy, 6(4), 33–51.
Kane, R. A., & Wilson, K. B. (1993). Assisted living in the United States: A
new paradigm for residential care for older persons? Washington, DC:
American Association of Retired Persons.
Radloff, L. S. (1977). The CES-D scale: A self-report depression scale for re-
search in the general population. Applied Psychological Measurement, 1,
385–401.
Kingsley, G. T., & Struyk, R. J. (1991). Housing policy in the United States:
Trends, future needs, and implications for congregate housing. Journal
of Housing for the Elderly, 9, 39–57.
Lawton, M. P. (1982). Environments and living arrangements. In R. H. Bin-
stock, W. Chow, & J. H. Schulz (Eds.), International perspectives on
aging: Population and policy challenges. New York: United Nations
Fund for Population Activities.
Lawton, M. P., & Cohen, J. (1974). The generality of housing impact on the
well-being of older people. Journal of Gerontology, 29, 194–204.
Lawton, M. P., Moss, M., Fulcomer, M., & Kleban, M. H. (1982). A re-
search and service-oriented Multilevel Assessment Instrument. Journal
of Gerontology, 37, 91–99.
Lawton, M. P., & Nahemow, L. (1973). Ecology and the aging process. In
C. Eisdorfer & M. P. Lawton (Eds.), The psychology of adult development
and aging (pp. 619–674). Washington, DC: American Psychological
Association.
Regnier, V., Hamilton, J., & Yatabe, S. (1991). Best practices in assisted liv-
ing: Innovations in design, management, and financing. Los Angeles:
Long Term Care National Resources Center, University of California,
Los Angeles, and University of Southern California.
Sherman, S. R. (1985). Housing. In A. Monk (Ed.), Handbook of geronto-
logical services. New York: Van Nostrand Reinhold.
Sherwood, S., Greer, D. S., & Morris, J. N. (1979). A study of the Highland
Heights apartments for the physically impaired and elderly in Fall
River. In T. O. Byerts, S. C. Howell, & L. A. Pastalan (Eds.), Environ-
mental context of aging. New York: Garland Press.
Sherwood, S., Morris, J. N., & Ruchlin, H. (1986). Alternative paths to long-
term care: Nursing home, geriatric day hospital, senior center, and do-
miciliary care options. American Journal of Public Health, 76, 38–44.
Lawton, M. P., & Simon, B. (1968). The ecology of social relationships in
housing for the elderly. The Gerontologist, 8, 108–115.
Received August 31, 1999
Accepted January 28, 2000
Mollica, R. L., Ladd, R. C., Dietsche, S., Wilson, K. B., & Ryther, B. S.
Decision Editor: Vernon L. Greene, PhD
428
The Gerontologist