Mental symptoms, psychotropic drug use and alcohol consumption in immigrated middle-aged women. The Women’s Health in Lund Area (WHILA) Study

This study aims to analyse mental symptoms, psychotropic drug use and alcohol consumption,
in immigrant women born in Finland, the other Nordic countries, Eastern Europe, Western
Europe and countries outside Europe, compared with Swedish-born women, and furthermore,
to study if age at immigration may have an influence. All women (n/10,766) aged 50
59 years
and living in the Lund area of southern Sweden received a postal invitation to a health survey
named the Women’s Health in Lund Area; 64.2% (n/6917) participated. The participants
answered a questionnaire including prevalence of mental symptoms during the past 3 months,
regular use of psychotropic drugs, alcohol consumption during an average week, country of
birth and age at immigration. Severe mental symptoms were more common among most
immigrant groups compared with native Swedes, but the association to country of birth was not
significant after adjustment for possible confounders. Regular use of hypnotics was more
common among Nordic immigrants only (odds ration). East European and non-
European immigrants less often were alcohol consumers . Heavy
drinking was more common among non-Nordic immigrants who immigrated at a younger age
than at an older age. Furthermore, it was found that although East European and non-European
immigrants had a higher educational level, they were less often gainfully employed compared
with native Swedes. In middle-aged women, country of birth as well as age at immigration are
important factors to consider in relation to alcohol consumption, but these factors may be of
less importance considering mental health.

During the last decade, alcohol consumption and
long-term sick-leave due to mental diagnoses have
increased dramatically in the Swedish population, particularly
among women (1, 2). Alcohol consumption and
mental health problems in immigrated middle-aged
women have been insufficiently studied, but increased
knowledge is of vital interest, since Sweden, like many
European countries, is turning into a multi-ethnic
society. In general, immigrants may be considered a
vulnerable group as they face a number of stressors such
as acculturative, socio-economic and minority stress.
In a female population 50-59 years old, we have
previously investigated mental symptoms and alcohol
consumption, and found that immigrants more often
had severe mental symptoms but less often drank
alcohol compared with native Swedes (3, 4). In these
studies, we only analysed whether the woman was an
immigrant or not. In the present study, we will further
investigate the findings by focusing on the immigrated
woman’s country of birth and age at immigration.
This study aims to analyse mental symptoms, psychotropic
drug use and alcohol consumption in immigrant
women born in Finland, the other Nordic countries, East
Europe, West Europe and countries outside Europe,
compared with Swedish-born women, and in addition
to study if age at immigration may have an influence.
Material and Methods
Participants
All women (n/10,766) aged 50
59 years, born between
2 December 1935 and 1 December 1945, and by 1
December 1995 living in the Lund area of southern
Sweden (172,005 inhabitants) were invited to the Women’s
Health in Lund Area Study (WHILA). This study
is based on a self-administered questionnaire, which was
received by mail, filled out at home and returned at a
medical examination where a nurse/midwife assisted to
rule out information bias; an interpreter assisted when
needed. The examinations took place from 2 December
1995 until 3 February 2000. A total of 6917 women
(64.2%) participated, among whom 601 (8.7%) were
immigrants. An attrition analysis has been presented
previously (3). Informed consent was obtained from all
participants. The ethics committee at Lund University
approved the study.
We lack information about country of birth among
non-participants, but according to Statistics Sweden, of
all women aged 50
59 years and in 1995 living in the
Lund area, 89.4% (n/9649) were native Swedes, 3.5%
(n/379) born in another Nordic country, 5.5% (n/595)
born in another European country and 1.5% (n/167)
born outside Europe. In comparison with our figures
(presented under Results), Nordic and non-European
immigrants were representative, whereas European immigrants
participated less often (PB/0.001).
Questionnaire
NATIVE COUNTRY
Based on stated country of birth the women were
divided into six categories: ‘‘native Swede’’, ‘‘Finnish
immigrant’’, ‘‘Nordic immigrant’’ (i.e. Denmark, Iceland
and Norway), ‘‘East European immigrant’’ (also
including former Soviet Union), West European immigrant
(also including countries in southern Europe) and
‘‘non-European immigrant’’.
AGE AT IMMIGRATION
Based on their age at immigration to Sweden the
immigrated women were divided into: B/18 years (lowest
quartile), 18
34 and ]/35 years old (highest quartile).
MENTAL SYMPTOMS
The Gothenburg Quality of Life instrument was used to
measure prevalence of 10 mental and 19 physical
symptoms. The women answered ‘‘yes’’ or ‘‘no’’ as to
whether the symptom had troubled her during the past
3 months (5). Based on the sum of mental symptoms, we
classified severity of mental symptoms: ‘‘absent/slight’’
(0
1 symptom, lowest quartile), ‘‘moderate’’ (2
6
symptoms) and ‘‘severe’’ (7
10 symptoms, highest
quartile).
PSYCHOTROPIC DRUG USE
The women reported what medications they regularly
use. In this paper, we examined their use of the
psychotropic drugs*anxiolytics, hypnotics and antidepressants.
ALCOHOL CONSUMPTION
The women reported the quantity (glass/bottles specified
in centilitres) of wine, beer and liquor respectively that
they drink during an average week, or chose the option
‘‘no alcohol’’. We converted total alcohol intake into
grams of alcohol and created four categories: none, low
(1
83 g), moderate (84
167 g) and heavy (]/168 g) (4).
SOCIAL SITUATION
The women stated household composition (with partner,
alone, with partner and child/ren, single parent or with
parent/other), highest level of education (comprehensive
school, upper secondary school or university education),
employment status (full-time/part-time work, unemployed,
disability pension, long-term sick leave or
housewife) and if she visited friends at least once a
month.
PHYSICAL HEALTH
The women stated whether they use hormone replacement
therapy. As described above the prevalence of 19
physical symptoms was asked for. Based on the sum of
physical symptoms we classified severity of physical
symptoms: ‘‘absent/slight’’ (0
2 symptoms, lowest quartile),
‘‘moderate’’ (3
7 symptoms), and ‘‘severe’’ (8
19
symptoms, highest quartile).
Statistics
Calculations were performed using the computer software
SPSS 12.0. The chi-square test was used to analyse
differences in proportions and when required Fishers
exact test was used. P-values B/0.05 were considered
statistically significant. In calculation of employment,
old-age pensioners and students were
excluded.
When analysing the country of birth, 21 women were
excluded as they stated being born abroad but did not
state the country. Each immigrant group was compared
with native Swedes (Table 1).
Four separate multivariate logistic regression analyses
were performed: severe mental symptoms vs.
absent/slight symptoms ; use of hypnotics vs. no use ; use of antidepressants (n/325)
vs. no use ; and non-drinking (now/moderate/heavy drinking (n 4901). In the first
three analyses, we adjusted for age, alcohol consumption,
household, level of education, employment, visiting
friends, physical symptoms, and use of hormone replacement
therapy in the first block, and country of birth in
the second block. In the last-mentioned analysis (nondrinking)
we adjusted for age, severity of mental
symptoms, use of anxiolytics, use of hypnotics, use of
antidepressants, household, level of education, employment,
visiting friends, physical symptoms, and use of
hormone replacement therapy in the first block, country
of birth in the second block. Hosmer and Lemeshow
tests were performed: 0.816-0.843.
When analysing the influence of age at immigration,
64 immigrants were excluded: 21 that did not state
country of birth and 43 that did not state when they
immigrated. Due to the small number of women in each
age group, Finnish and Nordic immigrants were merged
into ‘‘Nordic immigrants’’ and East and West European
immigrants into ‘‘European immigrants’’. Age at immigration
within Nordic, European and non-European
immigrants respectively was compared (Table 2).
Results
East European, West European and non-European
immigrants more often were troubled by severe mental
symptoms compared with native Swedes; Nordic immigrants
more often used hypnotics and antidepressants,
and non-European immigrants significantly more often
were non-drinkers of alcohol (Table 1).
All immigrant groups except Nordic immigrants
differed regarding the social situation in comparison
with native Swedes. Finnish immigrants more often lived
alone, 27% vs. 18%, or as single parents, 8% vs. 4% (PB/
0.01, 4 df), were disability pensioners, 20% vs. 8%
(PB/0.01, 4 df), and less often visited friends, 18% vs.
8% (PB/0.001, 1 df). East European immigrants more
often had university education, 62% vs. 34% (PB/0.001,
2 df), were disability pensioners, 14% vs. 8%, or housewives,
6% vs. 2% (PB/0.001, 4 df). West European
immigrants more often lived alone, 25% vs. 18%
(PB/0.05, 4 df), had upper secondary school education,
51% vs. 41% (PB/0.05, 2 df), and less often visited
friends, 18% vs. 8% (PB/0.001, 1 df). Non-European
immigrants differed most; they more often lived alone,
30% vs. 18%, as single parents, 16% vs. 4%, or with
parent/other, 4% vs. 1% (PB/0.001, 4 df). They more
often had university education, 50% vs. 34% (PB/0.01,
2 df), were unemployed, 15% vs. 4% (PB/0.001, 4 df),
and less often visited friends, 15% vs. 8% (PB/0.05, 1 df).
All immigrant groups except Nordic immigrants more
often were troubled by severe physical symptoms in
comparison with native Swedes; Finnish immigrants,
28% vs. 17% (PB/0.01, 2 df), East European immigrants,
35% vs. 17% (PB/0.001, 2 df), West European immigrants,
29% vs. 17% (PB/0.001, 2 df), and non-European
immigrants 33% vs. 17% (PB/0.001, 2 df).
Furthermore, non-European immigrants less often
used hormone replacement therapy, 24% vs. 45% (PB/
0.001, 1 df).
Four separate multivariate logistic regression analyses
were performed as described in the Statistics section.
Independently of age and other possible confounders,
Nordic immigrants were associated with use of hypnotics
,
East European and non-European immigrants were
associated with non-drinking of alcohol . East European, West
European and non-European immigrants did not remain
associated with severe mental symptoms, nor did Nordic
immigrants remain associated with use of antidepressants.
Age at immigration
Discussion
In a total population of women aged 50
59 years and
living in a geographically defined area, we analysed
whether mental symptoms, psychotropic drug use and
alcohol consumption were associated with country of
birth and age at immigration.
There are some weaknesses in the study*primarily
that the immigrant women were divided into rather
broad categories based on their country of birth. The
categorization could lead to inaccurate generalizations,
but was necessary due to the limited number of
participants from different countries (except from Finland).
To analyse every country specifically could lead to
inaccurate generalizations, as each country contains a
heterogeneous group of ethnic groups, social classes,
religions and cultures. Another weakness in the study
was that the woman’s reason for immigration was not
requested. The major types of immigration to Sweden
are labour migration and forced migration, and these
characteristics have varied between countries and over
time (6). Tentatively we can assume that Finnish, Nordic
and West European immigrants are labour migrants (as
only three women from former-Yugoslavia had immigrated
after the outbreak of war), whereas East European
and non-European immigrants are refugees.
Consequently, the results in this study, as for most
studies of immigrants, must be interpreted with caution.
In the present study, we found that East European,
West European and non-European immigrants were
more often troubled by severe mental symptoms compared
with native Swedes, but the association to country
of birth was not statistically significant after control for
possible confounders. Several studies have stressed
poorer mental health among immigrants compared to
native Swedes in terms of psychosomatic complaints,
psychological distress, longstanding psychiatric illness,
attempted suicide and suicide . European immi-
grants B/18 years at immigration were less often
troubled by severe mental symptoms. Young age at
migration has been found favourable for mental health,
suggesting that immigrants may overcome the nativity
disadvantages found for emotional distress with increased
duration of residence (12).
When an immigrant patient seeks help for mental
symptoms, it may well be easier for the general practitioner
to prescribe psychotropic drugs than to offer
psychosocial or psychotherapeutic treatment. Hjern
concludes that immigrants’ higher consumption of
sedatives and hypnotics indicates a differential treatment
of minor psychiatric disorders of members of ethnic
minorities in the Swedish healthcare system (13). In our
study, however, we found that only Nordic immigrants
used psychotropic drugs more frequently. This is an
interesting finding, which warrants further investigation.
In contrast to prior Swedish studies, we did not find that
non-European immigrants had a higher use of psychotropic
drugs (9, 13, 14).
Furthermore, we found that non-European immigrants
were more often non-users of alcohol, which
was independent of age and other possible confounders.
Many non-European immigrants were born in countries
where female drinking is uncommon. Conceivably, some
were Muslim who tend to refrain from alcohol. Furthermore,
East European immigrants more often were nonusers,
but this association was rather weak. Alcohol
consumption did not differ between Finnish immigrants
and native Swedes. Prior studies have found more
alcohol-related problems, higher hospital admission
because of an alcohol-related disorder and higher
alcohol-related mortality among female Finnish immigrants
(15
17). Among European and non-European
immigrants, subjects ]/35 years old at immigration were
more often non-drinkers, whereas younger women were
more often heavy-drinkers. When immigrants get assimilated
to their new society their alcohol consumption
usually increases to approximate that of the native
population (18
21). However, our results suggest that
for women a young age at migration may involve an
increased risk of heavy drinking in their middle age.
That Lund is a pronounced university town can
explain the generally high level of education amongst
participants, which may be one reason for the lack of
agreement with results in prior studies. Another reason
may be that none of those studies have focused on
middle-aged women.
Another finding was that although East European
and non-European immigrants had a higher level of
education than native Swedes, they were less often
gainfully employed. A similar result was found in a
Swedish study of younger immigrant women (7). The
reason for this is likely multi-factorial and needs further
attention. We found that all immigrant groups, except
the Nordic, differed from native Swedes regarding their
social situation. Non-European immigrants had the
poorest social situation. A report from Statistics Sweden
shows that labour migrants from wealthy countries soon
attain the same standard of living as native Swedes,
whereas many refugees never do (22).
To sum up the findings in this study; severe mental
symptoms were more common among most immigrant
groups compared with native Swedes, but the association
to country of birth was not significant after adjustment
for possible confounders. Regular use of hypnotics was
more common among Nordic immigrants only. East
European and non-European immigrants less often
consumed alcohol. Heavy drinking was more common
among non-Nordic immigrants who immigrated at a
younger age than at an older age. Although East
European and non-European immigrants had a higher
educational level they were less often gainfully employed
compared with native Swedes.
We conclude that in middle-aged women country of
birth as well as age at immigration are important factors
to consider in relation to alcohol consumption, but these
factors may be of less importance considering mental
health.