Dr. Ruth J. Ward,
Dr. Mia E. McLaughlin, Dr. Martin G. Livingston
Department of Psychiatry, Gartnavel Royal Hospital, 1066 Great Western Road, Glasgow, G12 OXH
SMJ 2003: 48(2) 38-40
Abstract
In 1986, the rehabilitation of every long stay
psychiatric patient in Glasgow was assessed with a view to reducing bed numbers
and developing comprehensive community services. Ten years on, we have attempted
to trace 91 patients with a diagnosis of schizophrenia assessed at Gartnavel
Hospital, in order to repeat assessments of their psychopathology and levels of
functioning. We believe this population represents a unique group in terms of
their age, length of hospital stay and chronicity of symptoms . Only 2 patients
were untraced but 36% of the original 91 patients were deceased. Discharge to
the community and variations in standards of care appeared to have little impact
on the symptomatic presentation of the survivors over ten years. The results
provide useful information on the success or otherwise of a large scale
discharge and community care programme which is continuing in Scotland.
Key words: schizophrenia, long term care, community
psychiatry, community mental health services, community care, rehabilitation,
follow‑up study, longitudinal studies, &institutionalisation,
psychiatric status rating scales
In 1984 there were approximately 3000 psychiatric beds in
Glasgow, 2000 of which were for long stay patients. 1300 of these were occupied
by patients with major mental illnesses such as schizophrenia and severe
enduring mood disorders. The development of a city wide mental health unit and
later a mental health trust with control of its own ring fenced budget, provided
the impetus to reassess how services for Glasgow's mentally ill could be
delivered in the future.
In October 1986, a survey assessing the rehabilitation
potential of every psychiatric long stay patient in the catchment population of
Greater Glasgow Health Board was carried out (ref 1). The results showed that the long term residents tended to
be elderly - the average age at Gartnavel Hospital was 60. Approximately
one third of the patients were considered to be fit for rehabilitation and
discharge into the community. A further third were judged capable of coping in a
lower dependency unit or facility.
In 1987 GGHB published a mental health strategy which
planned further reductions in bed numbers, the development of comprehensive
community services and the provision of more staff‑supported accommodation
in the community. Prior to the initiation of the discharge process, the 161 non
dementia long stay patients at Gartnavel Royal Hospital were assessed in detail (ref
2), and were found to be highly dependent and severely disabled. 77% were
experiencing delusions, 28% had severe tardive dyskinesia and 64% had the
poorest dependency rating on the Momingside Rehabilitation Scale (ref
3). Few had social contacts outside the ward in which they lived. However by
1994, less than 30 of these patients remained in long-stay hospital care.
There is a relative dearth of well conducted outcome
studies of what has been termed 'de-institutionalisation' and none which
focus on a Scottish population. Therefore we decided to review patients with
schizophrenia from our Glasgow long-stay cohort 10 years on, in order to
evaluate the impact of discharge to the community on this group within a
Scottish context.
Method
We attempted to trace the 91 patients found to have a
DSMIUR diagnosis of schizophrenia from the original 161 non-dementia
long‑stay patients (defined as having been inpatients for more than 12
months) assessed at Gartnavel Royal Hospital in 1988. We did this by examination
of hospital computer records and case notes, by contacting GP's, Community
Mental Health Teams, nursing homes and supported accommodation facilities. Once
traced, each patient was sent a letter requesting an interview at their home, or
at the hospital as they chose. An interview which lasted approximately 45
minutes was carried out by one of two raters (table
I). Following the interviews we spoke to staff involved in caring for the
patient, if applicable, and consulted any available case notes or background
information.
Inter-rater reliability was tested by repeating
approximately 10% of the interviews with the alternative rater and found to be
satisfactory.
For those patients who had died in the intervening ten
years, further information on the cause of death was sought at the General
Register Office for Scotland in Edinburgh.
Results
Of the original 91 patients with a DSMEIR diagnosis of
schizophrenia, 32 (35%) were living in the community and 24 (26%) were
inpatients. 33 (36%) were deceased and only 2 were untraced. We were
able to collect complete data on only 15 of those we interviewed. (
table II).
The mean age of the surviving group was 66 years, the
mode being the 70-79 age group (fig1). In 1988, more than 60% of the patients
were over the age of 65 years, 40 (44%) were male and 51 (56%) were female. Ten
years later, more than 75% of patients were over 65 years, 24(43%) were male and
32(57%) female.
The length of stay of the 22 patients interviewed in
longstay wards ranged between 5 and 56 years with an average of 33 years (fig
2).
The majority of patients in the community had been discharged from hospital more
than 5 years previously (fig 3). Of those patients living in the community, the
majority lived in the Greater Glasgow catchment area, but 4 were now living in
Dunoon or Helensburgh and one lived in Stomaway. Table
III gives details of the
type of accommodation used by those in the community and the amount of support
they were receiving.
We wanted to know whether there had been any changes in
mental state over the ten year period. There were no significant differences
between mean BPRS scores in 1988 and at follow up or between those groups
of patients currently in hospital and those in the community( tables IV and table
V). We were able to trace causes of death in all 33 cases (fig.
4)
Discussion
There were some problems with the design of the study.
Some of our information was incomplete, largely because of the length and
complexity of the interview administered to an aged and disabled group of
patients. The numbers giving complete interviews are consequently small,
particularly once divided into accommodation type. However we did succeed in
assessing psychopathology in all those interviewed. The inter-rater
reliability for 1998 interviews was satisfactory. The raters in 1988 were
different and therefore the reliability of conclusions drawn from comparing 1988
with 1998 assessments is open to question.
We believe that the population of patients studied in
Glasgow represent a unique group in terms of their age, length of hospital stay
and level of functioning. The Northwick Park study (ref 11) for example, looked at a schizophrenic population with a
mean age of 42 years who were much more able than our study group as judged by
their social outcomes. The TAPS (ref 12)
group followed patients, 80% of whom had schizophrenia, with a mean age of 61
years and an average length of stay of between only 21 and 28 years over a 5
year period. The older age and high level of disability of our group explains
our difficulty completing interviews in many patients who struggled to
concentrate for the 45 minutes required for the full interview.
The majority of the community group had been discharged
at least 5 years previously. Our clinical impression was that in terms of
severity of symptoms and level of functioning there was little to choose between
those patients who were still in long‑stay hospital beds and those in the
community at follow up. This was borne out by the lack of significant
difference between average BPRS scores for these two groups. This may partly
reflect difficulties in rehabilitating some patients who were anxious about the
prospect of discharge in 1988. It may also be due to the subsequent effect of
the care they have received in the intervening ten years. Our perception was of
wide variations in standards of care provided for these patients: For
those patients in long-stay hospital care, the physical environment of the wards
was shabby, the regime was institutional and there was a lack of occupational
Although the physical environment in nursing homes that
we visited in Glasgow appeared generally superior to that of the hospital, often
little information was available about patients, many of whom had formerly spent
most of their lives in hospital. Levels of staff motivation, experience and
supervision appeared variable. Some patients had been moved to homes out with
the Greater Glasgow area due to availability of accommodation rather than for
family or clinical reasons. These patients had no psychiatric follow up.
We gained the impression that there was little expectation of improvement in the
mental health of these patients.
In contrast to the nursing homes, the supported
accommodation that we saw (two different units in the north and east of the
city) was modem, purpose built and attractive in design. Patients were
living in small group homes with higher ratios of trained staff who encouraged
self‑reliance, involvement in preparing and shopping for food and in
community integration. There were links with CPN's and outpatient services for
all but one of these patients.
All those patients living at home were in contact with
secondary care services. Two of these were also in regular contact with their
families. Although we did not formally enquire about family or social ties, our
impression was that only a small percentage of interviewed patients received
visits from family or friends. This presumably reflects the chronicity of
illness and level of social functioning of the study group overall.
Our results showed no significant change in overall BPRS
at ten year follow up. For long stay patients, these results are consistent with
Abrahamson's review (ref 13) of
outcome of patients in long-term hospital care. He found a heterogeneity
of results but concluded that progressive deterioration in the long term was not
characteristic of schizophrenia in general, but appeared to be a phenomenon of
the first five to ten years of the illness, followed by a plateau. Improvement
might occur in even the very long‑term patient.
We might have expected that those of our patients in
apparently better resourced supported community placements with higher levels of
secondary care follow up would have lower BPRS scores than those in hospital, or
in nursing homes. However, this was not the case. In this Glasgow population,
rates of psychiatric follow up of patients at home and in supported
accommodation compare favourably with Northwick Park (ref 11) data which also showed no association between intensity of
follow up and improvements in mental state over time. Some studies (ref
14, 15) do show a clear correlation between impoverished enviromnent and
worsening clinical and social functioning. Others (ref
12,16) have demonstrated improvements in mental state over time in
well resourced community care.
With regard to nursing home care, Linn et al (ref
17) carried out a controlled trial of nursing homes as an alternative to
long term hospital care for a group of 'long term ill' including
both schizophrenic and those with organic brain syndromes in the USA. They found
that those in homes deteriorated significantly in self-care, behaviour,
confusion and attitude over the trial period. The study highlighted the
importance of considering in detail the choice of community placement for these
patients.
On reviewing cause of death, it was interesting to note
that there were no recorded suicides. It is likely that the chronicity of
illness and age of the population studied together with their inpatient status
(only 8 of the 33 patients dying were not in long-stay psychiatric beds)
reduces the likelihood of suicide as a cause of death. In comparison, in the
Northwick Park study (ref 11) there
were 24 unnatural deaths, 9 as inpatient, from a cohort of more than 500 younger
schizophrenic patients.
Conclusions
Our main finding was that discharge to the community and
variations in standards of care for a long-stay cohort of patients with
schizophrenia had little impact on their symptomatic presentation over ten
years. The original study group was depleted by a 36% death rate due to physical
health factors. We believe that our patients represent a unique study group in
view of their age and chronicity of symptoms. These results provide useful
information on the success or otherwise of the large scale discharge and
community care programme which is continuing in Scotland.
References
1. Livingston M.G and Bryson A. The Glasgow
Rehabilitation Survey. British Journal of Psychiatry 1989; 154:620-624
2.Livingston M.G. Characteristics of
long‑stay psychiatric patients: implications for rehabilitation.
Proceedings of the 17th CINP, Kyoto, Japan, Volume 2 1994:262
3.Affleck J.W and McGuire RI Measurement of
psychiatric rehabilitation status: a review of needs and a new scale, British
Journal of Psychiatry 1984; 145:637-640
4. Overall J.E and Gorham D.R. The Brief
Psychiatric Rating Scale. Psychiatric Reports 1962; 10:799-812
5. Krawiecka M. Goldberg D. and Vaughn M. A
standardised assessment scale for rating psychotic patients. Acta Psychiatrica
Scandinavica 1977; 55:299-308
6. United States Department of Health, Education
and Welfare. Abnormal Involuntary Movement Scale (AIMS). ECDEU Assessment Manual
1976; 534-537
7. Barnes T.R.E. A rating scale for
drug‑induced akathisia. British Journal of Psychiatry 1989;
154:672‑676
8. Simpson G.M. and Angus JN.S. A rating scale for
extra‑pyramidal side-effects. Acta Psychiatrica Scandinavica 1970;
supplement 212:11-19
9. Withers E, Hinton J. Three forms of the clinical
tests of the sensorium and their reliability. British Journal of Psychiatry
1971; 119(548): 1-8
10. Folstein M17, Folstein SE and McHugh PR. Mini
mental State: A practical method for grading the cognitive state of patients for
clinicians. Journal Psychiatric Research 1975; 12: 189-198
11. Johnstone E.Q Frith C.D,Crow T.J, Owens D.G.C, Done
D.J, Baldwin E.J and Charlette A. The Northwick Park Functional Psychosis Study:
diagnosis and outcome. Psychological Medicine 1992; 22:331-346
12. Leff J, Thomicroft G., Coxhead N. Crawford C.
The TAPS Project 22: A five‑year follow-up of long-stay psychiatric
patients discharged to the community. British Journal of Psychiatry 1994;
supplement 25:13-7
13. Abrahamson D. Institutionalisation and the long-term
course of schizophrenia. British Journal of Psychiatry 1993; 162:533-538
14. Wing J.K and Brown G.W. Institutionalisin and
schizophrenia, A comparitive study of three Mental Hospitals. London Cambridge
University Press 1970.
15. Linn M.W, Caffey EM, Klett J et al. Hospital
vs. community (foster) care for psychiatric patients. Archives of General
Psychiatry 1977; 34:78-83
16. DeSisto M, Harding C.M, McCormick R.V, Ashikaga
T, Brooks G.W. The Maine and Vermont Three‑Decade Studies of Serious
Mental Illness II comparisons. British Journal of Psychiatry 1995; 167(3): 338-42
Longitudinal course
17. Linn M.W, Gurel L, Williford W.0 et al. Nursing
Home care as an alternative to psychiatric hospitalisation. Archives of General
Psychiatry 1985; 42:544-551