Professional Documents
Culture Documents
SUMMARY
Objective To investigate why elderly persons moved to a nursing home and detect any change in their psychopathological
and cognitive profile, dependency and perceived QOL.
Method We considered 100 elderly consecutively admitted to a nursing home, within 1 week of admission and 6 months
later. Data were collected from medical and nursing records (medical and psychiatric history, ADL), interviews (MMSE,
reasons for admission) and self-report instruments (GDS, BSI, WHOQOL-brief).
Results At admission, 68 subjects were able to participate in the survey and agreed to take part. Only 5.9% (n ¼ 4) had
moved to the nursing home by their own choice, 26.5% (n ¼ 18) because they were lonely and 36.7% (n ¼ 25) because they
had no caregiver available. The reasons for admission reported by the interviewed residents only partly corresponded with
the ones noted in their records. Twenty-six subjects had an MMSE score 18 and agreed to undertake complete assessment.
The scores of 20 of the 26 subjects exceeded the cut-off on the GDS and five of the nine BSI subscales. During the follow-up
period, 19 of the 68 originally assessed residents died (33 of the entire sample of 100) and one refused to continue to take
part in the survey. The mean MMSE and ADL scores of the 48 survivors decreased from 16.87 (SD 7.32) to 14.27
(SD 7.24; t ¼ 4.89, p < 0.000) and from 8.22 (SD 3.55) to 6.39 (SD 3.56; t ¼ 5.34, p < 0.000), respectively. The scores
achieved on GDI and BSI subscales worsened in 20 survivors with MMSE 18. The score achieved on the QOL physical
health domain also worsened. Comparison of the subjects who died and survived during the follow-up period showed a
significant difference in terms of ADL (6.60 SD 4.221 vs 3.64 SD 3.773 F ¼ 11.639; p < 0.001).
Conclusion The 68 original participants expressed feelings of loneliness and marginalization, but these experiences were
not noted in their personal records. Most subjects presented psychiatric symptoms and cognitive decline. After 6 months, the
clinical condition of the sample, particularly the females, had worsened, with a 33% mortality rate. This suggests that mov-
ing to a nursing home did not bring about improvement or stabilization; rather, psychiatric symptoms worsened and quality
of life was perceived more poorly. Copyright # 2006 John Wiley & Sons, Ltd.
key words — nursing home; quality of life; follow up; elderly; reasons for admission
transfer to an NH (Black et al., 1999). In addition, part were assessed by the MMSE and asked why they
residents’ quality of life (QOL) is lower than that had been admitted to an NH.
of elderly individuals living in the community The complete test package was administered to
(Nagamoto et al., 1995). In most cases, older people willing subjects achieving an MMSE score of 18/30
move to a nursing home when their health declines or and over. The entire procedure was performed within
they loose lose their independence; this may be 7 days of admission to the NH and then 6 months
accompanied by feelings of demise and uselessness, later, for follow-up purposes.
influencing perception of quality of life (Su-Zu Tseng The interviews were carried out by two psycholo-
and Ruey-Hsia, 2001). gists from the project team, who had been trained
The aim of our study was to assess why older sub- on how to approach the elderly and administer tests.
jects moved to an NH and to evaluate any changes in Study group meetings were organized at regular inter-
their psychopathological and cognitive profile, degree vals to discuss any details and problems.
of dependency and perceived QOL, after six months The NH staff agreed to cooperate with the research-
spent in an NH. ers with a view to improving their clinical and perso-
nal knowledge of newly admitted residents.
METHOD
Between June 2002 and April 2003 we enrolled to the INSTRUMENTS
study 100 subjects aged 65 and over, consecutively
admitted to a large NH (383 beds) in Mestre-Venezia From the documentation on admission to the facility,
(Italy). we obtained demographic, family and health status
Figure 1 illustrates the recruitment procedure and data, reasons for admission to the NH and level of
assessments made in the individual groups of subjects. Activities of Daily Living (ADL: Katz et al., 1970).
Informed consent to take part in the study was The following were administered to those residents
sought from those subjects able to participate. If the able to participate and who gave their consent:
subject agreed, a willing family member could be * The Mini-Mental State Examination (MMSE;
involved in obtaining consent. Those agreeing to take Folstein et al., 1975): a widely used measure of
cognitive status;
Time 1
(within 7 days of admission) * An open question about reasons for admission.
Copyright # 2006 John Wiley & Sons, Ltd. Int J Geriatr Psychiatry 2006; 21: 281–287.
nursing home institutionalization 283
ANALYSES was 16.46 (SD 7.23; min 0.4 max 29.7) with mean
ADL scores of 7.60 (SD 3.65) (see Table 1).
Statistical analysis was performed using SPSS, ver-
The reasons for admission to NH reported by the
sion 11.5. ANOVA, the t-test or the Kruskal–Wallis
residents were: 18 (26.5%) because they ‘felt lonely’;
test, as appropriate, to investigate cognitive, psycho-
25 (36.8%) because ‘no caregiver was available’ for
pathological and quality of life profile. Bonferroni
them; ten (14.7%) because of ‘disease, disability
corrections for multiple comparisons were used for
and impossibility of being cared for at home’; five
the Test Post Hoc.
(7.4%) due to ‘disputes with relatives’; four (5.9%)
based on ‘free choice’; one resident said he had no
RESULTS house to live in, while another had been transferred
from another institution (four residents replied: ‘I
Sample at admission
don’t know’).
Description of all 100 newly admitted residents. The The reasons for admission reported in the residents’
mean age of the total sample was 83.85 (SD 8.33); records were: ‘caregivers not available’ (n ¼ 49;
70% were females, who were significantly older than 72%), ‘resident is sick, an invalid or unable to be
the males (t ¼ 2.80 p < 0.05). Over 80% of the sam- cared for at home’ (n ¼ 17; 25%); coming from ‘other
ple had a low educational level (41% had no academic facilities’ (n ¼ 2; 3%).
qualifications), but the males had a higher mean num- Twenty-seven of the 68 residents (11 males and 16
ber of schooling years than the females (5.93—SD females) had an MMSE score of 18/30 or higher (15
3.26 vs 4.16—SD 3.06; t ¼ 2.610 p < 0.01). The subjects achieved 24/30 or higher). One subject sub-
widowed accounted for 62% of the sample; on aver- sequently dropped out; hence, we ultimately assessed
age, the males had been widowed for 136.20 months 26 residents with the whole test package within 7 days
(SD 105.03) and the females for 223.91 months of admission (see Fig. 1).
(SD 163.49).
The psychiatric disorders gleaned from medical
records were as follows: 59% suffered from dementia, Comparison between admission and follow-up
19% from other psychiatric disorders (e.g. depressive
disorders, alcohol addiction and/or abuse etc); 10% of Within 6 months of admission to the nursing home, 33
the sample presented comorbid psychiatric disorders. of the 100 residents died (see Fig. 1). Of the 68 sub-
Twenty-two residents in the sample had received no jects who had participated at the time of admission,
diagnosis of psychiatric disease. 19 had died, one subject declined to take part in the
Thirteen percent of relatives of the test subjects suf- follow-up; thus, the follow-up evaluation was admi-
fered from dementia and 10% from depression and nistered to 48 elderly people. Nevertheless, the mean
alcohol addiction. MMSE score fell from 16.87 (SD 7.32) to 14.27
Eighty-seven of the residents were not self- (SD 7.24; t ¼ 4.89, p < 0.000), while the ADL
sufficient on admission to the NH, while the remain- scores decreased from 8.22 (SD 3.55) to 6.39
ing 13% were autonomous. (SD 3.56; t ¼ 5.34, p < 0.000).
Thirty-two subjects were too unwell to consent to Twenty (seven males and 13 females) scored ¼ >
continue assessment. 18 on the MMSE.
The mean GDS score rose from 15.05 (SD 5.14;
Subjects able to participate. The mean MMSE score min 7, max 26) to 16.9 (SD 5.5; t ¼ 1.92, p < 0.05).
of these 68 subjects, adjusted for age and education, At follow-up, all but one subject exceeded the cut-off
Subjects (no.) 32 41 27
Gender M/F 10/22 11/30 11/16
Mean Age (SD) 83.03 ( 10.31) 83.73 ( 7.18) 85.00 ( 7.47) 0.411–< 0.664
Mean ADL (SD) 1.41 ( 1.76) 6.68 ( 3.61) 9.00 ( 3.32) 49.59–< 0.000
Mean MMSE (SD) NA 11.55 ( 4.33) 23.90 ( 3.23) 160.3–< 0.000
Copyright # 2006 John Wiley & Sons, Ltd. Int J Geriatr Psychiatry 2006; 21: 281–287.
284 p. scocco et al.
value, while four subjects had a score below the cut- bid disorders with complex symptoms (e.g. depres-
off at the first assessment. sion and dementia; Lyketsos et al., 1997), take a
Compared to the initial assessment, at follow-up wide range of drugs and present varying degrees of
the survivors showed a poorer mean score on all disability. This is why category-based diagnostic pro-
BSI subscales and on the GDI, which exceeded the cedures are difficult and sometimes useless. More-
cut off value (indicating distress in this area). over, the diagnoses from medical records at the
At follow-up, the females had a significantly higher nursing home were not so complete and accurate. A
BSI score than the males on the following scales: dimensional approach permits evolution to be viewed
somatization (males 0.69 SD 0.24 vs females 1.76 over time even when the sample shows modest
SD 0.57; F ¼ 21.489; sig. ¼ 0.000), interpersonal changes, irrespective of the disorders or concomitant
sensitivity (males 0.75 SD 0.52 vs females 1.29 conditions underlying symptoms.
SD 0.54; F ¼ 4.651; sig. ¼ 0.045) and GDI (males We also included a population with an MMSE
0.73 SD 0.19 vs females 1.29 SD 0.40; F ¼ score of at least 18, because we believe it is important
11.970; sig. ¼ 0.003). to give voice to the demented elderly (Schnelle,
During the follow-up period, respondents’ score for 2003); we agree with Godlove Mozley et al. (1999)
perceived quality of life was significantly lower (indi- who claimed that ‘not knowing where I am doesn’t
cating a worse QOL) in the physical area, where the mean I don’t know what I like’. In that study, the
average score dropped from 53.92 (SD 13.71) at base- authors found that over 77% of subjects who scored
line to 46.07 (SD 17.76) at follow up (t ¼ 2.159 10 or over on the MMSE were ‘interviewable’. Like-
p < 0.03). wise, other authors have claimed that QOL can actu-
Comparison of the scores achieved on the tests ally be measured by self-report instruments in a large
administered at the time of admission to the NH by proportion of the nursing home population, including
those who had deceased and those who had survived cognitively impaired residents (Kane et al., 2003).
at follow-up, did not show any significant difference. Nevertheless, the decision to include a population
Only the ADL score (6.60 SD 4.221 vs 3.64 with MMSE ¼ > 18 may, in the psychometric assess-
SD 3.773 F ¼ 11.639; p < 0.001) was significantly ment, be seen as being rather an arbitrary one.
lower in the deceased than in the survivors. However, as already noted by Andresen et al.
(1999), an MMSE cut-off of 18 or lower, including
only those able to make a self-report, amounts to eval-
DISCUSSION
uating just one in four nursing home residents.
In order to track the evolution of the cognitive, psy- This study has a number of limitations. We consid-
chological and QOL profile of recently institutiona- ered a sample of 100 residents consecutively admitted
lized elderly, we used self-report, dimensional to just one large, urban NH. To date, studies on NHs
evaluation and a conservative approach (including have proven quite difficult: therefore, it is not possible
subjects with an MMSE of at least 18, who are gener- to generalize our results because of broad variability
ally not assessed with self-report instruments). Older in this setting in terms of size (e.g. number of
adults living in an NH frequently suffer from comor- beds), location (e.g. rural, suburban and urban) and
Admission Follow-up
Cut-off Mean SD No. subjects > Cut-off Mean SD No. subjs. > Cut-off t–p
Copyright # 2006 John Wiley & Sons, Ltd. Int J Geriatr Psychiatry 2006; 21: 281–287.
nursing home institutionalization 285
organization (e.g. availability of professionals, finan- toms affected 38% of demented patients within just
cial arrangements). Overall organization of services 2 weeks of admission and 36.1% after 6 months.
provided to the elderly has also witnessed many The most frequent symptoms were flattened or incon-
changes in recent years (Reichman et al., 1998; gruous affect, suspiciousness or aggressiveness and
Purandare et al., 2004). anxiety.
Data gleaned from medical records (e.g. psychiatric Psychic status was found to worsen more in female
diagnoses of patients and their relatives) were prob- than in male residents.
ably unreliable and it was not always possible to trace Moving to an institution is not always a deliberate
back to the method used to recoup available data. choice made by the elderly. In fact, it turned out that
Despite these limitations, longitudinal self-evaluation less than 6% of respondents claimed to have person-
provided significant data and information. ally decided to live in an NH. These data are con-
The Brief Symptom Inventory showed that the firmed by a study by Mattimore et al. (1997). These
symptom areas at the time of admission to an NH authors measured willingness to live permanently
were Somatization, Anxiety, Depression, Psychoti- in a nursing home on a five-point scale ranging
cism (and ‘General Distress’). At follow-up, resi- from ‘very willing’ to ‘would rather die’. Of 9105
dents’ condition had worsened, with a higher approached patients from five university medical cen-
score exceeding the threshold value in all BSI areas, ters, 3262 (36%) agreed to take part in the survey. Of
except Hostility. As to gender, females performed these, 26% said they were ‘very unwilling’, 30%
worse than males in the following areas: somatiza- would ‘rather die’; however, 7% were ‘very willing’
tion, interpersonal sensitivity and overall evalua- to live permanently in a nursing home.
tion (GDI). It also emerged that the reasons for admission
The depressive symptoms measured by the BSI- reported by the interviewed residents only partly cor-
depression subscale and GDS showed comparable responded to those noted in the records. Clearly, being
levels of severity. Previous studies have indicated a transferred to an NH because of ‘feeling lonely’ or
lower prevalence. However, the studies in question because of ‘disputes with relatives’ does not exactly
adopted different rating scales. Using the Hamilton correspond in semantic nor conceptual terms to ‘care-
Rating Scales for Depression (HDRS), Bekaroglu giver not available’. Interestingly, the records did not
et al. (1991) observed depressive symptoms in 41% indicate any instances of ‘free choice’.
of NH residents vs 29% of residents in the local com- Thus, although admission to these facilities in Italy
munity. Horigichi and Inamy (1991) interviewed 920 does not happen in a vacuum but is shaped by the
elderly residents admitted to 32 NHs and 1153 com- decisions of a multiprofessional team (although this
munity residents. They used the Zung Self-Rating rarely includes healthcare providers such as psychia-
Depression Scale (ZSRS) and observed that 61% of trists or psychologists), which evaluates each sub-
the former and 36% of the latter showed depressive ject’s needs and resources, this assessment does not
symptoms. These differences in frequency in our data seem to fully consider the elderly person’s subjective
at the time of admission may be explained by the dif- experience.
ferent evaluation method (self-rating scales, in parti- While Gonzalez-Salvador et al. (2000) claimed that
cular the GDS, overestimate depression compared to their sample of residents in a long-term care facility
scales administered by others, such as the HDRS) or exhibited a better QOL than expected, our data
by the different time frame. showed that 6 months after entry, the elderly per-
The presence of persistent depressive symptoms ceived a worse quality of life in terms of physical
after spending a period of time in an NH was also health. This was also confirmed by a worsening in
highlighted by Katz et al. (1989), although other the level of activities of daily life, assessed by ADL.
authors observed a reduction of symptoms (Payne The subjects who died prior to follow-up were also
et al., 2002). significantly characterized by lower autonomy. The
We found a high rate of psychotic symptoms. prognostic value of reduced autonomy on mortality
In fact, psychoticism and paranoid ideation were at the time of admission to an NH has been reported
observed in 50% of 26 residents evaluated at admis- several times in the literature (Fillenbaum, 1985;
sion. Morriss et al. (1990) found that 21% of 125 Lichtenstein et al., 1985).
patients newly admitted to NH had delusions. We observed a high mortality rate at 6 months. Yet,
Wancata et al. (2003) observed stabilization over there exists wide variation in the mortality rate among
6 months of any non-cognitive symptoms among NHs; this was accounted for by residents’ character-
NH residents suffering from dementia. Such symp- istics (Dale et al., 2001).
Copyright # 2006 John Wiley & Sons, Ltd. Int J Geriatr Psychiatry 2006; 21: 281–287.
286 p. scocco et al.
Copyright # 2006 John Wiley & Sons, Ltd. Int J Geriatr Psychiatry 2006; 21: 281–287.
nursing home institutionalization 287
Payne JL, Sheppard JM, Steinberg M, et al. 2002. Incidence, pre- Su-Zu T, Ruey-Hsia W. 2001. Quality of life and related factors
valence, and outcomes of depression in residents of a long-term among Elderly Nursing Home Residents in Southern Taiwan.
care facility with dementia. Int J Geriatr Psychiatry 17(3): 247– Pub Health Nursing 18(5): 304.
253. Teresi J, Abrams R, Holmes D, Ramirez M, Eimicke J. 2001. Pre-
Purandare N, Burns A, Challis D, Morris J. 2004. Perceived mental valence of depression and depression recognition in nursing
health needs and adequacy of service provision to older people in homes. Soc Psychiatry Epidemiol 36(12): 613–620.
care homes in the UK: a national survey. Int J Geriatric Psychia- Wancata J, Benda N, Meise U, Windhaber J. 2003. Non-cognitive
try 19: 549–553. symptoms of dementia in nursing homes: frequency, course
Reichman WE, Coyne AC, Borson S, et al. 1998. Nursing Home. A and consequences. Soc Psychiatry Epidemiol 38(11): 637–
survey of six states. Am J Geriatr Psychiatry 6: 320–327. 643.
Schnelle JF. 2003. Improving nursing home quality assessment: Yesavage JA, Brink TL, Rose TL. 1983. Development and valida-
capturing the voice of cognitively impaired elders. J Gerontol tion of a geriatric depression scale: a preliminary report. J Psy-
A Biol Sci Med Sci 58(3): 238–239. chiatric Residents 17: 37–49.
Copyright # 2006 John Wiley & Sons, Ltd. Int J Geriatr Psychiatry 2006; 21: 281–287.