LUND UNIVERSITY
PO Box 117
221 00 Lund
+46 46-222 00 00
Formal support, mental disorders and personal characteristics: a 25-year follow-up
study of a total cohort of older people.
Samuelsson, Gillis; Sundström, G; Dehlin, Ove; Hagberg, Bo
Published in:
Health & Social Care in the Community
DOI:
10.1046/j.1365-2524.2003.00416.x
2003
Link to publication
Citation for published version (APA):
Samuelsson, G., Sundström, G., Dehlin, O., & Hagberg, B. (2003). Formal support, mental disorders and
personal characteristics: a 25-year follow-up study of a total cohort of older people.
Health & Social Care in the
Community
,
11
(2), 95-102. https://doi.org/10.1046/j.1365-2524.2003.00416.x
Total number of authors:
4
General rights
Unless other specific re-use rights are stated the following general rights apply:
Copyright and moral rights for the publications made accessible in the public portal are retained by the authors
and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the
legal requirements associated with these rights.
• Users may download and print one copy of any publication from the public portal for the purpose of private study
or research.
• You may not further distribute the material or use it for any profit-making activity or commercial gain
• You may freely distribute the URL identifying the publication in the public portal
Read more about Creative commons licenses: https://creativecommons.org/licenses/
Take down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove
access to the work immediately and investigate your claim.
Health and Social Care in the Community
11
(2), 95–102
© 2003 Blackwell Publishing Ltd
95
Abstract
The present study was designed to describe the pattern of long-term formal
support received by people with mental disorders, and to investigate the
relationship between the medical, psychological and social characteristics of
the participants and types of formal support. This study is based on a cohort
(
n
= 192) of people born in 1902 and 1903 in a community in Southern
Sweden. The research participants were assessed using interviews,
psychological tests and medical examinations. Information was collected
about the use of primary healthcare and social services. The first assessment
took place when the cohort was aged 67 years, and then on eight further
occasions until they were 92. The participation rate ranged from 72% to
100%. During the observation period of 25 years, 53% of people with
dementia eventually received both home help and institutional care
compared to 34% of people with other psychiatric diagnoses and 12% of
people with good mental health. The last group had all physical health
problems and/or problems with activities of daily living. However, 35% of
the dementia group, 46% with other psychiatric diagnoses and 52% of
people with good mental health did not receive any formal support. Males
and self-employed people were significantly less likely to use formal
support. The institutionalised group reported loneliness significantly more
often than the other two groups. In a logistic regression analysis, loneliness,
low social class, high blood pressure and low problem-solving ability were
predictors of formal support use. People with a mental disorder, including
dementia, were significantly more likely to use formal support compared to
people with good mental health. Social factors were the main factors
predicting formal support.
Keywords:
formal support, longitudinal, mental disorders, multidisciplinary
Accepted for publication
5 October 2002
Blackwell Science, Ltd
Formal support, mental disorders and personal characteristics: a 25-year
follow-up study of a total cohort of older people
G. Samuelsson
PhD
1
, G. Sundström
PhD
2
, O. Dehlin
MD
3
and B. Hagberg
PhD
1
1
Lund University, Lund, Sweden,
2
Institute of Gerontology, Jönköping, Sweden and
3
Department of Community Medicine,
Malmö, Sweden
Correspondence
Gillis Samuelsson
Lund University
Gerontology
Baravägen 3
Box 187
S-22100 Lund
Sweden
E-mail: gillis.samuelsson@geron.lu.se
Introduction
In the 1980s, the importance of providing social and
medical care (including diagnostic facilities) for elderly
people with dementia was recognised (Annerstedt
1995). In addition, special attention has been given to
the situation of family members caring for relatives
with severe, long-term dementia. Research on geriatric
mental disorders over the past 10–15 years has focused
on dementia, and comparatively few studies have
analysed the entire spectrum of psychiatric disorders
experienced by old people.
The present authors studied the full range of mental
disorders and formal support using a longitudinal pro-
spective study in a cohort of elderly people. For the pur-
poses of this paper, formal support is defined as the use
of home help services, old age home and/or nursing
home services. The authors are not aware of any other
longitudinal study of elderly people that investigates
over a 25-year follow-up period the whole range of
G. Samuelsson
et al.
96
© 2003 Blackwell Publishing Ltd, Health and Social Care in the Community
11
(2), 95–102
mental disorders and their relationship to formal sup-
port patterns. Findings from the present study have
been reported earlier (Samuelsson & Sundström 1988,
Hagberg
et al
. 1991, Samuelsson
et al
. 1993, Samuelsson
et al
. 1994, 1998). However, none of these earlier
research reports included analyses of different mental
disorders and formal support extending over the entire
25-year period.
Calsyn & Winter (2000) reviewed the research about
predictors of service use by elderly people, and found
that living alone and state of physical health consist-
ently predicted the use of home help service. The
review was much less conclusive about the importance
of social variables. This finding might be because of
differences between the studies which were reviewed
in terms of the length of the follow-up periods; for
example, longer follow-up periods tend to provide
more diverse information about diseases and changes
in support. Living alone consistently predicted use of
community services when other network supports
were taken into account.
Mental disorders seems to be another determinant
of service use. In a population study of people suffering
from dementia in Stockholm, Fratiglioni
et al
. (2001)
reported that 47% of the study participants (age
≥
77 years) were institutionalised at the beginning of
their study and 93% 7 years later. Smith
et al
. (1999)
found that functional status and mental morbidity were
major contributors to nursing home placement.
A Norwegian population study of people aged
≥
80 years followed from 1981 until they were all dead
in the year 2000 (Romoeren 2001) reported that 48%
eventually used home help services and that 73% ended
their lives in institutional settings. These rates are much
higher than coverage rates for home help and institu-
tional care during this time period in Norway.
In a previous paper based on the same research
project as the one presented here, information about
symptoms, diagnoses and contacts with the primary
healthcare system from the ages of 67 to 84 years
was reported (Sundström
et al
. 1991). Participants
with mental disorders tended to establish contact
earlier than mentally healthy persons, and to maintain
contact with their general practitioner over a longer
time period. Dementia patients had also more contact
with the primary healthcare system than non-demented
people. Sjöbeck (1994) confirmed this finding in her
study.
The aims of the present study were to:
1
describe formal support (home help and institutional
care) patterns in relation to people with mental
disorders between the ages of 67 and 92 years,
including those who died before the age of 92; and
2
investigate the relationships between the medical,
psychological and social characteristics of
participants when they were 67 years and use of
formal support during a 25-year follow-up
period.
Subjects and methods
Research participants
The cohort (
n
= 192) comprised every 67-year-old
person living in the Dalby Primary Health Care District
during the year 1969–1970. Dalby is situated 40 km
from Malmö, the third largest city in Sweden. Follow-
up began in 1970 with repeated examinations and
interviews, usually every other year up to 1985. Further
assessment or examination was undertaken in 1994–
1995, when the survivors had reached 92 years of
age. The rate of participation at 67 years of age varied
from 76% to 81% in the medical, psychological and
social examinations. In the eight following waves,
only one person dropped out. At 92 years of age, all
the survivors participated. Eighty-nine per cent of
the study group died between the ages of 67 and
92 years.
Thirty-eight people refused to participate in the
examination at 67 years of age or died (six men) prior
to the examinations. Information was collected on mental
diagnoses and formal care for the non-participants,
as well as for the 12 people who moved out of the health-
care district during the 25-year period. The non-
participants had lived previously in urban rather than
in rural areas. There were no differences between
participants and non-participants in terms of gender,
marital status, mobility during earlier life and social
class. Non-participant men had higher income levels
than participating men, whilst the opposite was true
for women (Samuelsson
et al
. 1994). Among non-
participants, 11% had dementia, 11% had symptoms of
other mental disorders and 79% were mentally healthy
according to health records.
The general level of education was approximately
the same in the cohort as in the corresponding age
group for the whole country. Most of the men had
worked in manual occupations (86%), and most
women (88%) in household work or manual work.
This is a fairly typical pattern for a rural, turn-of-the-
century cohort in Sweden. Two per cent lived in old
people’s homes at 67 years of age. The rest lived in
the community, and 15% lived alone. Males were
clearly over-represented in the first examination (59%
were men and 41% women), a pattern which was
typical for older rural populations in that period
(Quensel 1945).
Formal support, mental disorders and personal characteristics
© 2003 Blackwell Publishing Ltd, Health and Social Care in the Community
11
(2), 95–102
97
Methods
The medical information and psychological data were
collected at the local healthcare centre by a physician
and a team of psychologists. These examinations took
approximately 1.5 days to complete. The medical data,
collected by a physician, included: a full history of cur-
rent and earlier diseases; blood tests, urine analysis,
electrocardiograms and laboratory tests; as well as the
results of a standard physical examination. Psychologi-
cal tests measured personality, cognition, needs, atti-
tudes, adjustment and intelligence. In addition, a social
worker interviewed the subjects at home about their
earlier and present life situation. This took approxi-
mately 2 hours (Samuelsson 1981). Furthermore, all
contacts with medical and social services over the years
of the study were recorded and information was
extracted from death certificates.
Predictors of support
Social measures
A questionnaire with items based on previous
Scandinavian studies (Stehouwer & Ostergard 1967,
Johansson 1970) of elderly people was used to assess
social factors. The following variables are included in
the present analysis: social class, economic status,
urban/rural living (during the life span), educational
level, marital status, having children, social support and
loneliness.
Social class was coded according to the Swedish
Socioeconomic classification (SEI) coding system: self-
employed (including farmers and business owners),
white-collar and blue-collar workers (Statistics Sweden
1982). Economic status was obtained from taxable
income from 65 to 68 years of age. Educational level
was measured in number of school years. Information
about years living in an urban/rural area from birth
to 67 years of age was collected from parish records.
Loneliness was assessed on a four-point Likert scale
and then dichotomised: (1) never/seldom; or (2) some-
times/often. Information about marital status and
loneliness was collected at 67 years of age and before an
individual moved into an institution.
Psychological measures
Cognition was assessed by psychometric tests which
measured: logical inductive reasoning; verbal, numeric
and spatial abilities; intelligence; and motor speed
(Lindberg
et al
. 1980). A factor analysis produced
indicators of four cognitive functions labelled: theory,
motor, problem-solving and knowledge. These factors
were used in the present analyses. Life adjustment
was based on a clinical interview and coded from 1 to 3:
(3) ‘good’ adjustment to present life situation – satisfac-
tory social integration, spends time on interesting
activities; (2) ‘fair’ adjustment – most things work well,
conflicts in some areas (e.g. social interaction, few
activities, and some worries); and (1) ‘poor’ adjust-
ment – conflicts in life (e.g. fear of the future, isolation,
negative or aggressive behaviour). Attitude scales
based on the semantic differential method (Osgood
et al
. 1967) were used to assess the degree to which
participants viewed their future in positive or negative
terms, and the extent to which they held a positive
self-concept.
Medical variables
The presence or absence of the following disorders
was assessed at 67 years of age: diabetes mellitus,
rheumatoid arthritis, angina pectoris, lung disease,
gastrointestinal disease, liver disease, kidney disease or
cancer, and any of the symptoms dyspnoea, vertigo and
headache (together called dizziness): (1) no diseases;
and (2) one or more diseases. Use of sleeping pills, anal-
gesics and sedative/hypnotic medication was also
assessed.
Data on type of mental disorder was collected
for each person, and date of death for those who died
before 92 years of age. Four men and 13 women sur-
vived beyond the age 92 years. Diagnostic classifica-
tion systems (National Board of Health and Social
Welfare 1970) have changed during the follow-up
period, but the changes were just changes of the codes
attributed to each mental disorder. However, the
underlying definitions of disorder were the same.
Incidence of mental disorders was calculated for the
entire 25-year period.
The main outcome variable was the use of formal
support services, as indicated in records from home
healthcare and social institutions. The type of formal
support was coded as: (1) no formal support (2) only
home help, and (3) home help and institutional care.
Statistical analysis
The Pearson chi-square test was used to identify any
associations between social, psychological and medi-
cal variables, and the formal support category. The
Kruskal–Wallis test was used to test for differences
in ordinal or (possibly skewed) interval variables
between the three support categories. A logistic regres-
sion analysis included all 23 variables to identify
whether or not any were significant predictors of the
use of different types of formal support services. A
goodness-of-fit measure was obtained by the Hosmer
and Lemeshow test that gave a non-significant
P
-value
of 0.31.
G. Samuelsson
et al.
98
© 2003 Blackwell Publishing Ltd, Health and Social Care in the Community
11
(2), 95–102
Results
Formal support and mental disorders between 67 and
92 years of age
Significantly more people with mental disorders, includ-
ing dementia, received formal support (institutional
care and home help) compared with mentally healthy
persons. The proportion of people who received home
help and were later relocated to institutions increased
with age and the increase was especially evident for
people with dual diagnoses. The mean age of mentally
healthy persons who were relocated to institutions
was 82.6 years, and this was 76.8 years for people with
both dementia and other mental disorders, 80.2 for
other mental disorders and 81.2 years for persons with
dementia. However, 35% of people with dementia and
44% of people with other mental disorders did not
receive any formal support during the 25-year period
compared to 53% of the mentally healthy subjects
(Table 1).
One person with dementia who also had other men-
tal problems received only home help. The correspond-
ing figure for people with other mental diagnoses was
21%, and among mentally healthy persons, 23%. A com-
bination of home help and institutional care was used
much more frequently by people with dementia or with
other mental disorders (mainly depression and intellec-
tual disability) than by the mentally healthy group. A
direct move from home to an institutional setting with-
out previous use of home help was common only
among people with dementia (35%) or other mental dis-
orders (17%). Some people with mental disorders
refused to accept home help.
Medical, psychological and social characteristics, and
their relationship with formal support
In the bivariate analysis, some social variables appeared
to be significantly related to receipt of formal support.
Females received home help and institutional care
much more often than men; 58% of the men did not
receive formal support compared to 33% of women.
White- and blue-collar workers received formal sup-
port significantly more often than the self-employed.
Two out of three people with dementia were married at
67 years of age. Most of the people who did not receive
any formal support (35%) stayed at home and received
informal support.
The institutionalised group experienced loneliness
at the age of 67 years more often than any other support
group (see Table 2). Thirty-two per cent felt lonely
‘sometimes or often’ before moving to an institution.
None of the other social variables was significantly
related to formal support.
None of the psychological variables which were
assessed at age 67 (using four factors of cognitive tests,
future view, self-concept and life adjustment) predicted
later patterns of support or service use (
P
-values varied
between 0.47 and 0.70).
None of the eight medical variables which were
assessed at 67 years of age were significantly asso-
ciated with receipt of services at follow-up (
P
-
values = 0.06–0.38). Compared to other groups, more
people who were relocated to institutions were diag-
nosed as having hypertension at the age of 67 years
(
P
= 0.06). An update of the information in medical
journals just before the time of relocation to an institu-
tion showed that all mentally healthy persons had
problems with activities of daily living (ADLs) and
physical diagnoses, sometimes in combination with
social problems.
The regression analysis tested the extent to which 23
medical, psychological and social variables predicted
formal support. Included in the final model were social
class, blood pressure, the cognitive factor ‘knowledge
function’ and loneliness. White- and blue-collar work-
ers were three times more likely to be recipients of
formal care. Loneliness increased the risk for receipt of
formal support by two and a half times; low knowledge
function had an odds ratio value of 1.20, indicating
increased risk for receipt of formal support; and normal
blood pressure decreased the risk of using formal serv-
ices (odds ratio value = 0.36) (Table 3).
Table 1 Mental disorders and formal support during a 25-year period (numbers and row percentage)
No formal support Home help Home help and institution Total
Diagnosis Number Percentage Number Percentage Number Percentage Number Percentage
Dementia 6 35 – – 11 65 17 100
Other mental disorder 15 44 7 21 12 35 34 100
Both dementia and other
mental disorder
3 38 1 12 4 50 8 100
Mentally healthy 50 53 23 24 22 23 95 100
Total 74 48 31 20 49 32 154 100
