Working Life and Health A Swedish Survey

A changing working life
Working Life
and Health
A Swedish Survey
Bengt Järvholm (Ed.)
Production group
Editor: Lars Grönkvist
Cover typography: Lena Karlsson
Cover illustration: © Diana Ong/Great Shots
Layout: Eric Elgemyr
Figures: Marit Skoglund
© The Swedish National Board
of Occupational Safety and Health,
The National Institute for Working Life,
& The Swedish Council for Work Life Research
National Institute for Working Life
S-171 84 Solna, Sweden
Tel: (+46)-8-730 91 00 Fax: (+46)-8-730 19 67
ISBN 91-7045-386-1
Printed by AB Boktryck, Helsingborg
Working life and health. A Swedish survey
A changing working life
This review is an assessment of contemporary working life in
Sweden and its effects on health. It is designed to be used by
decision-makers as a basis for developing strategies and setting
priorities for improvement measures, research, and supervision
of work environments over the next several years. The work was
done at the joint request of three Swedish authorities: the National
Board of Occupational Safety and Health, the National Institute
for Working Life and the Council for Work Life Research. A similar
review was published in 1990 by a governmental committee, the
Swedish Commission on Working Conditions, and that analysis has
been used as a model. The review starts with the present situation
and attempts to identify opportunities and expected problems.
Work environments can be greatly changed by decisions at all
levels. This report makes it clear that a good work environment
is an important part of “the good job” and that working life has a
great influence on health and illness.
National data have been used when available. Regional data have
been used when there are no national data and when the information
is regarded as relevant to other regions of the country. Assessment
of trends has also been attempted, but insufficient data often makes it difficult to quantify the changes over time. In these cases
trends have been estimated. Quantitative differences in health and
ill health between different groups are explained on the basis of
present theories whenever possible.
Working life and health. A Swedish survey
Since July 1994, the National Board of Occupational Safety and
Health has had responsibility for keeping statistics on the work
environment. A further purpose of this review has been to assess
the usefulness of these statistics.
More than twenty different researchers collaborated in writing this review. The editor was Professor Bengt Järvholm. The
researchers have also written more extensive reviews which will
be published and presented elsewhere. This translation includes
mainly quantitative data and conclusions. Those wishing to know
more about a particular area reviewed can contact the researcher
who wrote that review. A list of the contributors, including their
addresses, is found at the end of the book. Chapters that have not
been translated are also included in the list.
Members of the steering committee for this review were Anders
Englund, National Board of Occupational Safety and Health,
(chairman), Carl A Asklöf, Swedish Work Environment Fund
(until June 30, 1995), Gunnar Aronsson, National Institute for
Working Life, Elisabet Broberg, National Board of Occupational
Safety and Health, Gunnel Färm, Swedish Council for Work Life
Research (as of July 1, 1995), Christer Hogstedt, National Institute
for Working Life, and Folke K Larsson, National Institute for
Working Life and Swedish Council for Work Life Research.
National Board of Occupational Safety and Health
National Institute for Working Life
Swedish Council for Work Life Research
A changing working life
1 A changing working life............................................................7
2 Physical, chemical and biological factors............................... 15
3 Psychosocial factors................................................................ 21
4 Mortality in occupational groups........................................... 27
5 Work environment and myocardial infarction...................... 29
6 Occupation and cancer........................................................... 33
7 Respiratory diseases................................................................ 41
8 Skin diseases........................................................................... 45
9 Work-related musculoskeletal disorders................................ 49
10 Threats, violence, sexual harassment and suicide................ 57
11 Accidents at work................................................................. 63
12 Work and gender.................................................................. 67
13 Elderly persons in working life............................................ 73
14 Immigrants........................................................................... 79
15 Disability............................................................................... 83
16 Professional drivers.............................................................. 87
17 Construction workers........................................................... 91
18 Working life in the future.................................................... 97
19 Summary............................................................................. 101
List of contributors.................................................................. 107
Working life and health. A Swedish survey
A changing working life
1 A changing working life
Hardly any period since World War II has seen faster, larger and
sometimes more surprising changes in Swedish working life than
the beginning of the 1990s. The changes have affected both the
labour market and the organisation of work. In 1990 a Swedish
governmental committee, the Swedish Commission on Working
Conditions wrote that the lack of labour would be substantial
during the 1990s and that those who were employed would consequently have higher demands regarding work and working
conditions. Three years later there was widespread unemployment
rather than a lack of labour. Between 1990 and 1993 the number of
employees decreased by more than 200,000 in the manufacturing
industry and by 75,000 in the construction industry. The number
of employees also decreased in the public sector, which had been
growing for some decades.
Change of employment
The population of Sweden has increased steadily, and in the past
hundred years grown from a little less than five million to nearly nine
million. The proportion aged 16 to 64 has remained fairly constant,
between 60 and 65 percent. The number of employed women has
increased. In 1890 less than 20 percent of Swedish women had a job
outside the home (excluding maids); by the 1980s the proportions
of employed men and women were of similar size.
The largest sectors today are manufacturing and health care.
Each employs approximately 20 percent of the working population.
Health care workers are predominantly women, while employees in
the construction and manufacturing industries are predominantly
Working life and health. A Swedish survey
men (Table 1.1). In Sweden there are about 640,000 workplaces,
270,000 of which have employed persons. the other just have
self-employed persons.
Table 1.1. Percentages of men and women employed in 1995, by industry
(Survey of Labour).
Men Total
Farming, forestry and fishing 2
Construction 110
Manufacturing 112920
Public administration, defence
Trade 121413
Education, research11
Health care
Restaurants and hotel 32
Financing, insurance, business services
Recreation, culture and sport222
Numbers (in thousands)1,9272,061
Except for the above-mentioned increase of unemployment during
the 1990s, unemployment has been about one to three percent since
World War II (Figure 1.1). The number of long-term unemployed
increased during the 1990s.
Figure 1.1. Unemployment rates in Sweden, 1945–1995. For 1985–1995 the
proportion of the labour force in government programs for the unemployed is
also shown.
A changing working life
The unemployment rate is higher among men (8.4%) than among
women (6.9%) (Labour Survey, 1995). Adolescents and young
adults have a higher unemployment rate: 14 percent among
persons 16 to 24 years old, compared to 4 to 10 percent in other
age groups.
Organisation and work
Three dimensions have been identified as closely related to the
health and well-being of workers: psychological stress and demands,
control, and social support. Some changes in the organisation of
work can benefit both the worker and the employer. Flatter organisations, development of competence, and greater individual
responsibility and authority, for example, are regarded as increasing
competitiveness and efficiency.
There are two general strategies for increasing the flexibility
of an organisation. In one, permanent staff is kept to a minimum
and other personnel are engaged as needed. This may be done by
term employment, contracting of personnel or services, and/or use
of consultants. This method makes it more difficult to transfer a
worker to another job, for example in cases of prolonged illness,
since the number of different jobs decreases in a company with a
highly specialised production. The increased number of persons
with non-permanent jobs may be a consequence of this strategy
for more flexible organisation.
A second trend is broader competence. Workers who have
competence in several areas of production can rarely be replaced
by short-term employees. This has positive consequences for the
workers, since they have more control over their work and more
opportunities to learn during work. One negative aspect is increased
psychological stress due to higher demands. Another is that it is
difficult to expand the work force quickly if the company wants to
increase production. There has therefore been a lot of overtime
work during the past year even though the unemployment rate
has been high.
Non-permanent employees
The number of permanent jobs decreased from 3.7 million to
about three million between 1987 and 1995. Half of the permanent
jobs that disappeared were jobs for persons below the age of 25.
Working life and health. A Swedish survey
Non-permanent employment accounted for about 50 percent of all
new hires in 1990. In 1994 this had grown to 70 percent. In 1995
about 500,000 persons had jobs that were not permanent. About
180,000 were substitutes, 90,000 were project workers, 80,000
were employed on demand, 45,000 were on probation and 30,000
had seasonal employment. About 75,000 were trainees or had
holiday jobs. All groups except those who work on demand, who
have increased since 1987, fluctuate more or less with conditions
on the market. A questionnaire was sent to a random sample of
permanent and non-permanent employees in 1995. The results
showed that those who do not have permanent jobs know less
about occupational safety and health than those with permanent
positions. Project workers had more opportunity to control their
own work, even compared to those who had permanent jobs,
whereas substitute workers, seasonal workers and people who
work on demand have less control over their work than those with
permanent positions. Further, the number of days for training on
paid time was highest among those who had project employment:
15 days for men and 8 days for women; for persons with permanent
positions it was 9 days for men and 7 days for women. Persons
who worked on demand had the lowest number of paid days for
training: 4 days for men and 2 days for women.
There are several Swedish authorities and organisations concerned
with working conditions and occupational health. A few are briefly
reviewed here.
Occupational Health Services have changed considerably during
the 1990s. The causes are the recession, the terms of a collective
agreement, and the end of government funding for the occupational
health centres. In January of 1993, when the government funding
stopped, it covered 27 percent of the cost of the occupational
health centres. The number of employees in the occupational
health centres decreased from 10,000 to 7,500 between 1992 and
1994. In total there are about 700 occupational health centres in
Sweden. Occupational Health Services is a larger organisation than
the Labour Inspectorate, National Institute for Working Life or
the Departments of Occupational Medicine (Table 1.2).
A changing working life
Table 1.2. Number of persons in some organisations in the occupational health
area in 1995.
National Institute for Working Life
National Board of Occupational Safety and Health
Labour Inspectorate
Occupational Health Services
Swedish Council for Work Life Research
Departments of Occupational Medicine*250
* includes persons working in environmental medicine.
The National Board of Occupational Safety and Health and the
Labour Inspectorate are in the same organisation. The Labour
Inspectorate is divided into eleven districts. It enforces the law
and inspects the workplaces. Its main task is to see to it that employers work systematically to improve the work environment and
that they make proper provisions for rehabilitation. The Labour
Inspectorate is legally empowered to give notice to a company to
correct observed shortcomings in the work environment. It may
also issue prohibitions or injunctions. Notices were given to 23,000
companies in 1993, and injunctions or prohibitions were issued to
455. There are 113,000 safety representatives in Sweden. They
are authorised to stop work if they consider it to be dangerous.
In total there were 40 such stops in 1994.
The National Institute for Working Life was founded in July
1995, in a merger of the National Institute of Occupational Health
and the Swedish Institute for Work Life Research. The Institute
conducts and promotes research, education, development and
international collaboration on research and development to improve working life.
The Swedish Council for Work Life Research was founded in
1995, replacing the Swedish Work Environment Fund. It supports
research and development, and has a budget of approximately 265
millions SEK per year.
There are several sources that can be used to evaluate the work
environment and its influence on health. In a recent review Forsberg
(1994) listed 17 different sources of information. The registers
used in this evaluation are briefly described below.
Working life and health. A Swedish survey
Occupational Injury Information System (ISA)
All on-the-job accidents and occupational injuries or illnesses are
to be reported to this register. Reports are sent by the employer
to the social insurance office. Not all reported cases are evaluated
by the office regarding eligibility for compensation. Because of
changes in the rules for compensation, the frequency of reporting
has varied over the years even though the reporting is compulsory.
Most employers also have additional private insurance for accidents
and occupational diseases, negotiated between the unions and the
employer. The statistics from that organisation (TSI) are used in
this evaluation to examine the occurrence of accidents.
The Census-linked Death Register (CDR)
All deaths in Sweden are reported to a central register containing
the person’s national registration number, name and cause of death.
The death register contains no information about occupation,
but by combining the information with that given in a census the
mortality in different occupations can be described. In this survey
the census of 1980 has generally been used.
The National Survey of the Working Environment (NWE)
Since 1989, Sweden has had a national survey every second year in
which a random sample of the population answers questions about
their work and working conditions. The questions cover both the
physical, chemical and psychosocial work environment. There is also
an annual population-based survey of work-related symptoms.
The Swedish Survey of Living Conditions (ULF)
Since the 1970s, Statistics Sweden has interviewed a random sample of
the Swedish population about their living conditions. The interviews
also contain some questions on work and working conditions.
The Cancer-Environment Register (CER)
Sweden’s national Cancer Register was begun in 1958. The reporting of any diagnosed case of cancer is mandatory. The Cancer
Register contains no information on occupation, but by linking
this register with the censuses of 1960 and 1970 the incidence of
cancer in different occupations can be calculated.
A changing working life
The Level of Living Survey (LNU)
This is a register compiled and kept by the Swedish Institute for
Social Research at Stockholm University (SOFI). It is based on
interviews (1968, 1974, 1981, 1991) of a random sample of the
population. About 3,900 persons were interviewed on each occasion. The register contains some information on occupation and
working conditions.
Working life and health. A Swedish survey
A changing working life
2 Physical, chemical
and biological factors
This chapter deals with noise, vibration, electromagnetic fields,
work at computers, chemical hazards and contagious diseases.
The quantitative data are mainly from questionnaires. There is
little information available on concentrations of chemicals – even
for the most common substances, there are no data from sampling
About 50 percent of male and 20 percent of female skilled workers
reported in 1993 that during at least one fourth of the workday
the noise level at their workplace was so high that they could not
carry on an ordinary conversation. Among professionals 6 percent
reported such noise. This means that about 800,000 Swedes, during
at least one fourth of their workday, are exposed to such noise that
they cannot converse in a normal tone of voice. According to the
law, workers should be protected against noise above 85 dBA (AFS
1986:5). A government committee suggested that the noise level in
production areas should not exceed 70–75 dBA; for office work 35
dBA was recommended. There is no study of typical noise levels
in Swedish workplaces, but in 1993–94 the Labour Inspectorate
measured noise at 34 workplaces where they had reason to assume
that levels were high. In all of them the maximum was exceeded
(> 85 dBA).
Working life and health. A Swedish survey
Hand-arm vibrations may cause white fingers and nerve damage. In
a Swedish study 66 percent of welders and 45 percent of platers
reported numbness in their hands, versus 6 percent in controls. In
the national surveys made by Statistics Sweden between 1989 and
1993, 220,000 to 300,000 men and 30,000 women reported hand
vibrations for at least one fourth of the workday.
Whole-body vibrations are common during the driving of heavy
vehicles. The National Board of Occupational Safety and Health
has estimated that about 500,000 drivers daily are exposed to
whole-body vibrations. In the national survey in 1993, 12 percent
of men and 1 percent of women reported whole-body vibrations
for at least one fourth of the workday.
Electromagnetic fields
There is no threshold limit value for occupational exposure to
magnetic fields. A criteria group concluded in 1995 that available knowledge provided an insufficient basis for establishing a
maximum allowable level. In a survey of 1,098 Swedish men in
different occupations it was found that 40 percent were exposed to
an average above 0.2µT. The daily average for welders was 1.9µT.
Loggers had the highest value, 2.5µT, due to the exposure from
chain saws. Programmers and systems analysts who often work
with computers had a daily average of 0.2µT. For a little more than
50 percent of 90 different occupational groups, average exposures
exceeded 0.2µT. However, since exposures within occupational
groups varied widely, job titles were usually a poor indication of
Work at computers
Ten years ago about one person in four worked at a computer. In 1995,
54 percent of men and 48 percent of women used computers in their
work. Computers are most common in administrative and office
work, where more than 90 percent worked at a computer in 1995.
In farming and forestry 20 percent of the workers used computers.
About 10 percent of all employed persons reported in 1995 that they
used a computer at home for their work. About 20 percent of those
had a modem. About 1 percent of all employed persons worked at
least half time at computers in their homes in 1995.
A changing
biological factors
Chemical products
Chemical products that are manufactured or imported in quantities
above 100 kilograms per year must be reported to the Product
Register of the National Chemicals Inspectorate. In 1995, 50,000
such products were registered, 80 percent of which were used only
in workplaces. About 130,000 different substances occurred in these
products. The companies have classified about 1,600 products as
toxic or extremely toxic.
There is no systematic register of the concentrations of individual
substances in the work environment. The National Board of Occupational Safety and Health has determined maximum allow­able
concentrations for about 370 substances. A control of randomly
selected small workplaces in Stockholm in 1994 revealed that
about 30 percent of them used more than 20 kilograms of toxic
products per year. The risk of dangerous exposures was high or
fairly high in 11 percent of the workplaces.
There has been a national register of persons occupationally
exposed to lead in Sweden. The register was closed in 1990. An
analysis of lead concentrations in blood between 1978 and 1989
showed that the number of persons with blood lead levels above
2.5 µmoles per litre had not decreased (Figure 2.1).
Figure 2.1. Number of persons in Sweden with a blood lead concentration above
2.5 µmoles per litre, 1978–1989. (Source: Lead Register, National Board of
Occupational Safety and Health.)
Working life and health. A Swedish survey
Persons with a blood lead concentration above 2.5 µmoles per litre
are not allowed to work with lead. For women below the age of
50 the limit is 1.5 µmoles per litre (AFS 1992:17).
In the national survey a random sample of the population
answered a questionnaire about exposures (NWE). It can be estimated from the answers that about 200,000 persons, mostly men,
were exposed to oils or organic solvents for at least one fourth of
their working time (Table 2.1). The number of exposed persons
de­creased by 25 percent between 1989 and 1993. A separate
analysis of the engineering industry showed that the proportion
exposed to chemicals changed very little between 1989 and 1993.
The decrease in number of exposed persons is thus mostly due
to the decreased number of persons employed in the industry.
From the national surveys it can be estimated that about 450,000
persons were exposed to organic dust for at least one fourth of
their working time. The numbers of exposed men and women
were similar. Since this information comes from question­naires
it gives no indication of the concentrations.
Table 2.1. Some representative occupational exposures, 1989–1993. Information
is based on interviews in which persons have reported that they are exposed to
the substance for at least one fourth of their working time.
Number of exposed persons (in thousands)
Oil on skin
Organic dust
Organic solvents
To estimate the use of some chemicals, a questionnaire was sent
in 1995 to chemists and occupational hygienists in each district
of the Labour Inspectorate. They were asked the number of
persons exposed to some substances and whether the use of these
substances had increased or decreased. They responded that for
most substances use was apparently decreasing, but they indicated
an increased use of isocyanates. The questionnaire also showed
that the average number of exposed persons per workplace was
ten or fewer.
A changing
biological factors
The national survey revealed that the number of persons exposed
to environmental tobacco smoke (ETS) decreased by 40 percent
between 1989 and 1993. In 1993, 215,000 women and 306,000
men reported that they were exposed ETS at work for at least one
fourth of their working time.
Table 2.2. Some hazardous substances: Number of exposed persons, number
of companies where the substance is used, and the trend in use estimated by
the Labour Inspectorate. The information was collected in 1995.
Number of exposed
Number of companies
Cutting fluids15,000–30,0002,000–4,000
Infectious material and other biological risk factors
Hospital personnel report most cases of infection during work
to the occupational injuries information system (ISA). Reported
cases of infection due to occupational exposure to hepatitis viruses
(type B or C) have been below five per year since the late 1980s.
In some cases it is difficult to determine whether the infection
occurred at work. There is no known case in Sweden of a person
being infected by HIV during work.
In Sweden there are about 600 cases of tuberculosis per year.
Since 1990 there have been one to four cases reported per year
involving hospital workers infected by tuberculosis at work.
Personnel at pre-school and day care centres often catch respira­
tory infections, but these cases are not reported to ISA. Another
risk, for persons working at slaughterhouses for hens and chickens,
is infection by Campylobacter, but there are no statistics. There are
no survey data on exposure to mould, bacteria and endotoxins, or
other microbiological risk factors in farming, water purification
plants etc. However, there is information indicating that exposure
to mould and spores in sawmills is decreasing.
Working life and health. A Swedish survey
A changing working life
3 Psychosocial factors
Psychological demands and the freedom to make one’s own decisions at work are important psychosocial factors (Karasek and
Theorell, 1990). Negative stress arises when the demands are high
and the freedom is low. Active work is defined as high psycho­logical
demands in combination with high control over the work situation.
Psychological demands and control over the job differ between
blue-collar and white-collar workers. High-level professio­nals
usually have higher psychological demands and greater freedom to
plan their own work. For unskilled manual workers the situation
is the opposite. The nature of psychosocial factors was therefore
investigated in four different groups: unskilled workers, skilled
workers, low-level salaried employees and mid- or high-level
salaried employees (here called “professionals”). The frequency
of stressful (high demand, low control) and active (high demand,
high control) jobs in different industries was determined for each
group. The importance of workplace size, employment sector
(private, state, municipal), shift work, supervisory responsibi­lities
and social support were analysed. Jobs with high mental demands
and good conditions for individual decisions are more common
in small workplaces, in the public sector and among non-shift
workers. Managers and supervisors more often have work with
negative stress, but they also have more active work than other
employees. The psychological demands on foremen and managers are usually rather high, so their freedom to make their own
decisions largely determines whether they have an active job or
Working life and health. A Swedish survey
one with negative stress. Social support generally lowers the risk
of negative stress.
The occurrence of negative stress and active work were analysed
through the Surveys on Living Conditions (ULF) and the National Surveys of Work Environment (NWE), both conducted by
Statistics Sweden. These surveys are based on interviews and/or
question­naires given to large samples representative of the Swedish
population. Persons who reported high psycho­logical demands and
low control were classified as having negative stress and persons
who reported high demands and high control were classified as
having active work. For each of the four groups, the variation
across industries in the percentages of persons reporting negative
stress and active work was estimated. The reference category is
“manufacturing industry excluding engineering.” Estimates above
1.0 mean that negative stress (or active work) is more common
than in the reference category, and estimates below 1.0 mean that
it is less common than in the reference category.
Negative stress
For unskilled workers, there are five industries with an elevated
risk of negative stress: health services, retail trade, hotels and
restaurants, transport and communications (Table 3.1). All these
are in the service sector. There are five occupational groups with
high risk of negative stress: assistant nurses, kitchen assistants, post
office clerks, drivers and shop assistants (especially those working
at the cash registers).
For skilled workers, there is a higher than average risk of negative stress in health services and hotels and restaurants. The risk of
negative stress is approximately three times higher in the hotel and
restaurant business than in the manufacturing industry (excluding
engineering). In health services the risk is two to four times higher
than in the reference group. Occupational groups where negative
stress is common are waiters, cooks and assistant nurses.
For low-level salaried employees, negative stress is most common
in health services, transport, communications, and banking and
insurance. Dental nurses, salesmen, receptionists, computer opera­
tors, telephone operators and assistant accountants are occu­pational
groups where negative stress is relatively common.
A changing working
factors life
For professionals, negative stress is more common in health
services and hotels and restaurants. Occupational groups with
an elevated risk of negative stress are nurses, doctors, dentists,
catering supervisors and some categories of teachers.
In summary, negative stress is more common in the service
sector than in manufacturing. In health services and hotels and
restaurants the risk of negative stress is considerably elevated for
all four groups.
Table 3.1. Occurence of negative stress in different industries by occupational
category (unskilled workers, skilled workers, salaried employees professionals).
Relative risks expressed as ratios between the proportion of stressful jobs in
a given category and the proportion in the reference category “manufacturing
industry excluding engineering”. Ratios above (below) 1.0 mean that the proportion
of employees in the industry who have stressful jobs is higher (lower) then the
corresponding proportion in the reference category. Results from both the ULF
and the NWE surveys are presented in the table (*** = too few cases).
Wholesale trade1.1/0.6
Retail trade1.0/1.51.8/1.21.0/1.81.4/1.7
Hotels and restaurants1.6/1.52.7/3.4
Banking and insurance
Public administration
Health services2.0/1.1
Other services
Active work
For unskilled workers, active work is five to ten times more common
in social care than in the manufacturing industry. In some industries
where negative stress is relatively common, active work is also common: this is the case in health services, hotels and restaurants, retail
trade, transport and communi­cations. Active work is also relatively
common in the con­struction industry and in wholesale trade. Occupational groups which often have active jobs are children’s nurses,
home helpers and assistant occupational therapists. For skilled
workers, active work is four times as common in social care as in
Working life and health. A Swedish survey
the manufacturing industry. Active work is also common in hotels
and restaurants, health services, and wholesale and retail trade. For
white-collar workers, there are generally small differences across
industries in the prevalence of active work.
Vulnerable groups
On average, psychosocial working conditions are worse for women
than for men. In 1992/93, 12 percent of women had work with
negative stress compared to 8 percent of men. However, if the
occurrence of negative stress is compared within occupational
groups there is no difference between male and female employees.
Hence, the relatively large proportion of women experiencing
negative stress at work is due to the fact that women tend to have
other types of occupations than men.
The difference in occurrence of active work is small between
men and women, on average. However, there is a large difference,
to the disadvantage of women, if men and women are compared
within the same occupational group or industry.
Compared to native Swedes, immigrants have a higher percentage
of jobs with negative stress and a lower percentage of active jobs.
The difference cannot be explained by differences in occupational
distribution. In 1992/93, 15 percent of immigrants had stressful
jobs compared to 9 percent of Swedes. Active work was reported
by 18 percent of immigrants and 25 percent of native Swedes.
During the past decade, the proportion of unskilled workers has
decreased while the proportion of professionals has increased.
Manufacturing employs a decreasing percentage of the population
and the social and health care sector has grown. Because of these
changes, work with high psychological demands and high control
has become more common, i. e. active work has increased. On top
of this structural change, psychological demands have increased
markedly. In 1979, 29 percent of all employees described their work
as hectic and mentally demanding. This figure had increased to
35 percent by 1992/93. The trend is especially marked in social
care, health services and education. This rise in psychological demands of structural change, has resulted in an increasing number
A changing working
factors life
of stressful jobs. All in all, then, the 1980s and early 1990s have
witnessed a polarisation of jobs with respect to their psychosocial
character. Both active work and negative stress have become more
Karasek R, Theorell T. Healthy work. Stress, productivity, and the reconstruction
of working life. New York, Basic Books, 1990.
Working life and health. A Swedish survey
Psykisk ohälsa, våld och trakasserier
4 Mortality in occupational
Socioeconomic differences in age-adjusted mortality have increased
in several Western countries during the past few decades. The
increase is most prominent in men. In Sweden the mortality of
middle-aged manual workers was virtually unchanged between 1961
and 1985, whereas it decreased considerably for other groups.
The mortality in different social classes has been analysed by comparing the Cause-of-Death-Register with the censuses of 1980 and
1985. The sex-specific and age-adjusted mortality for workers aged
20 to 64 was calculated for the periods 1981–1985 and 1986–1990.
Mortality had decreased in all groups of men (Table 4.1).
Table 4.1. Age-adjusted mortality according to social class among occupationally acitve persons (cases per 100,000 persons and year), 1981–85 and
1981–85 1986–90 1981–85 1986–90
High-level non-manual employees 154
Middle level non-manual employees 142
Lower non-manual employees
Skilled workers
Semi- and unskilled workers
The largest decrease was in high-level non-manual workers, who
already had the lowest mortality. The previously increasing difference between manual workers and other groups slowed down
Working life and health. A Swedish survey
during the late 1980s. Thus, mortality can decrease while the
difference between classes remains. The changes for women are
much smaller than for men. The difference in mortality between
women and men has decreased in all groups. The cause of this is
Occupation and mortality
Mortality was calculated for men and women in 245 occupational
groups and for two periods of time. In 66 combinations of sex and
occupational group there was a statistically significant high or low
mortality during both time periods.
Elevated mortality was found for 31 combinations of sex and
occupational group: both male and female bookbinders, joiners,
cleaners, printers and toolmakers, for example, had elevated mortalities during both periods. For men, waiters, cooks and kitchen
assistants had an elevated mortality during both periods, whereas
there was no increase for females in these jobs. The cause of this
difference between the sexes is unknown.
In a previous analysis, the Swedish Commission on Working
Conditions identified drivers as a group with high mortality. Relative risks were 1.5 for male drivers and 1.35 for female drivers
during the period 1986–1990.
Occupational groups with a low mortality during both periods
were engineers, dentists, judges and teachers. Male business administrators, salesmen, buyers and shop managers had a lower
than average risk. Female shop assistants, physiotherapists and
nurses had low mortality. The causes of the differences in mortality
between occupational groups are largely unknown. Probably both
occupational and non-occupational factors are of importance.
Vågerö D, Lundberg O. Socioeconomic mortality differentials among adults
in Sweden. In Lopez A, Caselli G, Valkonen T (eds). Adult Mortality in
Developed Countries. Oxford: Clarendon Press, 1995.
Psykisk ohälsa, våld och trakasserier
5 Work environment
and myocardial infarction
Approximately 1,900 Swedish men and 400 Swedish women
below the age of 65 die every year of myocardial infarction. The
mortality increases rapidly with increasing age. Less than 300
men and 50 women below the age of 50 die annually of myocardial infarction. Both morbidity and mortality due to myocardial
infarction have varied in recent decades. There are no national
statistics on morbidity. In Stockholm county morbidity increased
by approximately 2 percent per year during the 1970s and then
decreased at approximately the same rate. The decrease occurred
among both women and men, younger and older, although it is
less pronounced for younger persons. There are similar trends
in mortality due to myocardial infarction, but the causes of these
trends are unknown. Changes in smoking and dietary habits are
often suggested as explanations, as is medical treatment of high
blood pressure and high blood lipids.
The occurrence of myocardial infarction in Sweden shows some
geographical variation. The risk is higher in the north. There are
also differences between socioeconomic groups. The risk is approximately 50 percent higher among blue-collar workers than among
white-collar workers, for example (Alfredsson et al, 1995).
The geographical and sociological differences in occurrence of
myocardial infarction are too large to be explained by recognised
Working life and health. A Swedish survey
risk factors such as tobacco smoking, high blood pressure or high
blood lipids.
Risk of myocardial infarction in different occupations
The Cause of Death Registers for 1981–86 and 1987–92 were investigated to determine the mortality due to ischemic heart disease
in different occupations. The relative risks were calculated by
comparing the mortality in an occupational group with the average
mortality for all gainfully employed. A relative risk of 1.0 means
that the occupational group has a risk similar to that of the average
Swedish worker. In the analyses of the occupational groups the
differences in age were taken into account. Occupational groups
with higher or lower than average risk of death due to ischemic
heart disease were identified (Tables 5.1 and 5.2).
Table 5.1. Occupational groups with a high risk of myocardial infarction
Nurses’ assistants
Welfare workers
Forestry workers
Ship’s deck and engine-room workers
Railways engine drivers
Truck dispatchers
Furnace men
Forge and foundry workers
Metal platers and coaters
Cabinetmakers and joiners
Laminated wood and fibreboard workers
Sawmill workers
Concrete workers
Plastic product makers
Unskilled workers
Earth-moving and other heavy equipment operators
Forklift truck and conveyor operators
Warehouse workers
Cooks and kitchen assistants
Other service workers
ohälsa, våldand
och myocardial
Table 5.2. Occupational groups with a low risk of myocardial infarction,
Graduate engineers
Managers of machine shops
Chemists and physicists
School principals
College teachers
High school teachers
Grade school teachers
Systems analysts, programmers
Legislators and administrators
Business managers
Advertising salesmen
Sales representatives
Shop managers
There were 30 male occupational groups with an elevated risk.
Four had a risk above 2.0: ship’s deck and engine-room workers,
railway engine drivers, laminated wood and fibreboard workers
and “other” service workers. Of the 30 male occupational groups
with an elevated risk, 17 were in the manufacturing industry, 5
in transport, storage and communication, and 5 in the service
sector. Among professional drivers there were in total 961 deaths
during the studied period. One fourth of them, or 254 cases, could
be attributed to the elevated risk of the occupation. There were
777 deaths among toolmakers: one fourth, or 183 cases; could be
attributed to the elevated risk in this occupation.
There were five female occupational groups with higher than
average risk: professional drivers, toolmakers, storage and warehouse workers, kitchen assistants and cleaners.
Twenty male occupational groups with a lower than average
risk were identified. Physicians and dentists had the lowest risk
(relative risk = 0.5). Twelve of the groups were in the sector natural
sciences, technical and social sciences, humanities or arts, three
in the administrative sector and three in the commercial sector.
The other two were farmers and policemen.
Two female occupational groups with a low risk were identified:
nurses and primary education teachers.
Working life and health. A Swedish survey
Analyses for 1981–86 show the same pattern for men and women, low risk in university graduates and high risk in workers in
the manufacturing industry and service sector.
An earlier report by the Swedish Commission on Working
Conditions showed an elevated risk for drivers and supervisors
in the transport sector. This is less obvious in the analyses for
1981–86 and 1987–92.
For both men and women, there is a fivefold difference between
the highest and the lowest risk categories. Most of the low-risk occupations require a university degree. Most of the high-risk occupations are in the manufacturing and service sectors, e.g. blue-collar
workers and office workers with low education. The explanation for
these differences is not known, but it is reasonable to assume that
they result from a combination of lifestyle and chemical, physical
and/or psychosocial factors in the work environment.
Alfredsson L, Hammar N, Gillström P. Increasing differences in myocardial
infarction incidence between socio-economic groups in Stockholm. Nutr
Metab Cardiovasc Dis 1995;5:99–104.
Hammar N, Ahlbom A, Theorell T. Geographical differences in myocardial
infarction incidence in eight Swedish counties, 1976–1981. Epidemiology
Psykisk ohälsa, våld och trakasserier
6 Occupation and cancer
The Swedish Cancer Committee estimated at the end of the 1970s
that approximately 2 percent of all cases of cancer would not have
occurred if the known causes of cancer in work environments had
been eliminated. Most cases that would occur would be among
men, and the factor that caused most cases would be asbestos. Two
percent of cancer in Sweden today is approximately 800 cases.
Studies in northern and southeastern Sweden indicate that approximately every tenth case of lung cancer in men could have
been prevented by work environment interventions in the 1970s.
A study in Gothenburg in the mid-1980s showed that if asbestos
had not been used earlier, approximately 15 percent of cases of
lung cancer would have been eliminated in men below the age of
75. The IARC has evaluated the carcinogenic potential of approxi­
mately 1,000 substances and exposures. Substances and exposures
that the IARC has classified as definitely or probably carcinogenic
and which occur in Sweden in substantial amounts are listed in
Table 6.1. Job-exposure matrices were used to estimate the number of persons exposed to carcinogenic substances in the work
environment. A matrix shows the combinations of occupation and
industry in which there is a risk of exposure to a certain carcinogenic substance. For each combination of occupation in industry
the number of exposed persons is estimated. The estimates in the
table are based on the census of 1990. When a maximum allowable
concentration is changed, the National Board of Occupational
Safety and Health estimates number of exposed persons. This
Working life and health. A Swedish survey
information was compared to the numbers from the job-exposure
matrix. The results in the matrix were also compared to the Finnish
register of persons exposed to carcinogenic substances. There was
a fair concordance between these different sources.
Table 6.1. Estimates of the numbers of persons in Sweden exposed to some
carcinogens. The projected numbers of cancer cases are based on some assumptions (see text).
Number of exposed 1990
Men Women
Cases of cancer/year at RR=1.1
Type of tumour
Men Women
30 lung cancer
30 lung cancer
5,000 leukaemia
Diesel exhaust
1,300 lung cancer
Ethylene oxide
100 leukaemia/lymphoma
Glass industry work
250 lung cancer
stomach cancer
large intestine cancer
Foundry work
100 lung cancer
Rubber industry work
30 lung cancer
urinary bladder cancer
100 lung cancer
Chromiun (6+)
100 lung cancer
400 lung cancer
800 lung cancer
Chlorinated organic
1,100 liver cancer
Leather dust
(shoe manufacturing)
Soot, tar and
combustion products 9,000
Strong acids
Wood dust
Vinyl chloride
non-Hodgkin lymphoma
lung cancer
nasal cancer
skin cancer
lung cancer
urinary bladder cancer
larynx cancer
nasal cancer liver cancer 0.06
0.2 0.3
0.1 0.6
The number of cancer cases in the future
Estimates were made of the number of cancer cases that could have
been avoided if the known causes of cancer in the work environment
had been eliminated. These estimates are based on the following
assumptions: the number of new cancer cases per year is constant
and the same as in 1992; the age distribution of the population is
constant; and the relative risk does not increase before the age of
Psykisk ohälsa,
och trakasserier
and våld
30 but thereafter is the same for all age groups and calendar years.
Two combinations of relative risk and occupational turnover were
used. Occupational turnover is defined as the proportion that leave
an exposed occupation/branch of industry during one year. In one
combination the re­lative risk was set to 1.1 and the occupational
turnover to 10 percent, and the results are given in Table 6.1.
Comments on some exposures
Some common exposures are discussed. The number of cancer
cases today can be estimated rather well for some agents, such as
asbestos, but for substances such as diesel exhausts the estimates
are much more uncertain.
The Cancer Register shows that pleural mesothelioma has increased (Figure 6.1). Peritoneal mesothelioma is far less common
(fewer than 20 cases per year) and has increased little over the past
few decades. A high proportion of the cases of pleural mesothelioma
are probably caused by occupational exposure to asbestos. There
is no other known cause of mesothelioma in Sweden: erionite, for
example, does not occur in Sweden.
Figure 6.1. Swedish import of asbestos in tons and cases of pleural mesothelioma.
Working life and health. A Swedish survey
The import of asbestos decreased rapidly in the mid-1970s and
by the beginning of the 1990s it was below 500 tons per year. Mesothelioma rarely occurs less than 20 years after the first exposure.
The risk then increases even if the exposure stops. The number
of mesothelioma cases in Sweden has thus increased although
exposure to asbestos virtually stopped in the 1970s. If the increase
continues at the present rate approximately 150–200 men and
40–60 women per year will develop mesothelioma in 2010.
It is much more difficult to estimate the number of lung cancers
that could have been avoided if there has been no asbestos in the
work environment. The risk is also reduced if exposed persons
stop smoking. A few studies also indicate a possibility that the
risk for lung cancer might decrease if exposure stops. Judging
from earlier Swedish studies, it is a reasonable assumption that
the number of lung cancer cases in men would have been at least
5 percent lower if asbestos had never been used in Sweden. For
women the effect would hardly be detectable. Five percent of all
lung cancer cases among men corresponds to about 100 cases per
year in Sweden today.
Present exposures to asbestos would increase the risk for lung
cancer and mesothelioma by barely more than 10 percent among
those approximately 1,500 persons who are still exposed to asbestos.
That means less than one case per year is caused by the present
exposure to asbestos in Sweden (Table 6.1).
Diesel exhaust
Long-term and heavy exposure to diesel exhaust probably increases
the risk of lung cancer. Workers in bus garages who have been
exposed to high concentrations of diesel exhaust for a long time
have approximately double the average risk of developing lung
cancer. An elevated incidence has also been observed in Swedish
longshoremen, probably caused by exposure to diesel exhaust. If
the present exposure to diesel exhaust increases the risk of lung
cancer by 10 percent, that will be approximately 10 cases per year
in Sweden.
It is well known that benzene may cause some types of leukaemia.
Benzene occurs almost exclusively in petrol for motor vehicles,
Psykisk ohälsa,
och trakasserier
which in Sweden contains 3 to 4 percent benzene. It is impossible
to estimate the number of persons exposed to benzene by using a
job-exposure matrix. The National Board of Occupational Safety
and Health estimated that in Sweden 40,000 to 50,000 persons
are exposed to benzene. According to Table 6.1, a relative risk of
1.1 is equivalent to less than one case of cancer per year.
Many studies have indicated that painters have a greater than average risk of lung cancer. One cause may be exposure to asbestos.
Other causes may be silica, pigment, binders and solvents. There
are also some data indicating that painters smoke more than the
population as a whole. In a Swedish study, an elevated incidence
of lung cancer was found in painters who started to work in the
1930s and earlier, but no such tendency was seen in younger
cohorts. Swedish painters are no longer exposed to asbestos, and
the number of cancer cases will therefore probably be lower than
that indicated by a relative risk of 1.1 (Table 6.1).
Persons with silicosis have an increased risk of developing lung
cancer. However, silicosis is now a rare disease in Sweden. It is
less certain whether exposure to silica increases the risk of lung
cancer. The job-exposure matrix indicates that approximately
12,000 Swedish workers are exposed to silica, but the risk of lung
cancer at the present exposure level is hard to estimate. If the assumption in the model is correct, a few cases per year may occur
due to present exposure to silica.
Soot, tar and combustion products
Soot, tar and combustion products contain carcinogenic substances
such as polyaromatic hydrocarbons (PAH). The risk of skin cancer
from exposure to mineral oils with a high concentration of PAH has
now been almost eliminated in Sweden. Coal tar and pitch are rarely
used in Sweden today. Swedish chimney sweeps have an elevated
risk of lung cancer. Asphalt containing coal tar may increase the
risk of lung cancer, but the asphalt now used in Sweden is based on
bitumen, and it is uncertain whether such exposure increases the risk
of lung cancer. Chimney sweeps, gas oven workers and workers in
Working life and health. A Swedish survey
aluminium smelters have an elevated risk of bladder cancer . This
is probably an effect of exposure to PAH.
It is difficult to estimate the number of cancer cases caused in
Sweden by present exposures to PAH in combustion products
and soot. A few cases per year of lung cancer and bladder cancer
would probably be prevented if all exposure to soot and combustion
products stopped. The number of skin cancers prevented would
probably be of the order of one case per year or lower.
Other environmental factors
UV radiation increases the risk of skin cancer. The Swedish
Cancer-Environment Registe shows that Swedish farmers and
fishermen have more than double the average risk of developing
cancer of the lip (EpC, 1994), probably because of exposure to UV
radiation. Hepatitis increases the risk of liver cancer. In Sweden
approximately five persons per year catch hepatitis from exposure
at work. Consequently, very few cases of liver cancer will result
from such infections.
Environmental tobacco smoke in workplaces may increase the risk of
lung cancer, but there is evidence that this exposure is decreasing.
If exposure to all the substances listed in Table 6.1 were eliminated, less than a hundred cases of cancer among men and less
than ten cases among women would be prevented each year. The
best estimate of the total is somewhere between 10 and 100 cases
per year. Today there are 200 to 250 cases of cancer caused by
previous asbestos exposure, and the estimate of 10 to 100 cases per
year assumes that present asbestos exposure will cause very few
new cases. Compared to the 1950s and 1960s, another few hundred cases of cancer have probably been prevented by elimination
of asbestos. Work to decrease the risk of occupational cancer has
thus been successful.
Continuing research may lead to detection of new cancer
risks. In the future, more exposures will probably be regarded
as carcinogenic. If all substances which have been suggested as
carcinogenic had been included in Table 6.1 the number of cases
due to present exposures would obviously be higher. The risks of
electrical and magnetic fields have been discussed in recent years.
It is impossible to make a firm conclusion as to whether such fields
increase the risk of cancer.
Psykisk ohälsa,
och trakasserier
There are several current studies attempting to identify genetic
markers indicating if a person has a higher or lower than average
risk of developing cancer. However, it is unlikely that such genetic
markers will be routinely used to prevent occupational cancer
during the next five years.
Cancer Environment Register 1960–70. EpC-rapport 1994;4, Socialstyrelsen,
Stockholm, 1994.
Working life and health. A Swedish survey
Psykisk ohälsa, våld och trakasserier
7 Respiratory diseases
Asthma and hay fever (allergic rhinitis) have increased in Sweden
during the past few decades. The increase is most obvious in children; the trend is less clear for adults. In the beginning of the 1990s
6 to 8 percent of adolescents living in northern Sweden had asthma.
The prevalence among adults is approximately 5 percent. Chronic
obstructive pulmonary diseases (COPD), including emphy­sema,
occur mainly in smokers. In the beginning of the 1990s 8 percent
of men in the age group 60-69 years had COPD, compared to 4
percent of women.
In 1990-1992, 300 persons per year claimed their asthma to be
occupational asthma (ISA). If the frequency of reporting had been
the same in Sweden as in Finland, approximately twice that number of cases would have been reported. Among men, occupational
asthma was most often reported by bakers, furnace men, welders
and spray painters. Among women the most common occupations
were wood workers, plastic product makers and livestock, dairy
and poultry farm workers.
Mortality due to asthma has been calculated for different occupations, with adjustment for smoking habits in different occupational groups, by combining information from the Census and
the Cause of Death Register. There are few cases per occupation
and the random variations in relative risk are large. Farmers and
agricultural workers were the only male occupational groups that
had a significantly higher mortality due to asthma between 1981
Working life and health. A Swedish survey
and 1992. Hairdressers were the only female occupational group
with an elevated mortality due to asthma (8 cases versus 1.8 expected). There were four occupational groups with elevated mortality
from COPD between 1981 and 1992 (Cause of Death Register,
Census data and adjustments for smoking habits). The groups
were horticultural workers (16 cases versus 7 expected), postmen
(13 cases versus 5.7 expected), truck drivers (22 cases versus 10.5
expected) and storage and warehouse workers (64 cases versus
30.5 expected). The only female group with elevated mortality was
storage and warehouse workers (12 cases versus 5 expected). The
higher risk in these occupational groups is probably due largely to
occupational turnover for health reasons. COPD develops over a
long period and the patients may have been transferred or applied
for other jobs when they developed symptoms.
Table 7.1. Occupational groups with the highest frequencies of asthma during
1990–1992 (ISA).
Frequency (cases/100,000/year)
Furnace men
Spray painters
Chemical processing workers
Foundry workers
Wood processing workers
Woodworking machine operators
Plastic products makers
Paper mill workers
Wood workers
Plastic products makers
Livestock, dairy and poultry
farm workers
Machine shop workers
Laboratory technicians
Dressmakers and seamstresses
Number of cases
In the ISA statistics of occupational injuries and diseases it is
difficult to distinguish between COPD and chronic bronchitis.
våld och trakasserier
In 1990–1992 about 50 men and 25 women per year reported
those diseases. The five male occupational groups reporting the
highest frequencies were workers in the steel and metal industry,
non-specific production workers, chemical processing workers,
rubber product makers, painters, floor layers and carpenters/
cabinet­makers. In women these diagnoses were most common in
chemical processing workers and rubber product workers. Few
cases of rhinitis were reported to ISA as an occupational disease
during these three years: 32 male cases and 13 female cases. All
but two men and one woman were bakers.
During the 1990–1992 period 51 cases of allergic alveolitis were
reported by men and 11 cases by women. Thirty-three cases were
farmers: 30 men and 3 women. Six cases occurred among carpenters and cabinetmakers. For this group the cases reported to the
register have decreased during the past decade from about 80 to
20 cases per 100,000 persons per year.
New causes of respiratory disorders
Changes in the work environment influence the risk of respiratory
diseases. Asbestosis and silicosis have been virtually eliminated
since the use of asbestosis has been prohibited, and exposure to
silica has decreased considerably.
Increased use of paints and lacquers containing isocyanates may
increase the risk of asthma unless measures are taken to reduce
exposure. Paper pulp is now bleached with ozone, which may be
a new cause of occupational asthma. The shift from solvent-based
to water-based paints may increase the risk of respiratory disorders,
since the water-based paints contain biocides. An increased use
of glues and paints based on reactive chemicals may increase the
risk of respiratory diseases in the manufacturing and construction
A few new factors that can cause respiratory disorders have been
identified during the past decade, including latex dust in hospitals,
aziridrin in some paints and the environment in pig farming.
Working life and health. A Swedish survey
ohälsa, våld och trakasserier
8 Skin diseases
In 1993, 935 men and 1,423 women reported skin disorders caused by their work (ISA). These cases account for 4 percent of all
reported occupational diseases. Ninety percent were eczema cases,
and in nine of ten of these the skin of the hands was affected. Skin
disorders are reported more often by women than by men, and are
more common in young than in elderly persons (Figure 8.1).
Figure 8.1. Frequency of skin disorders reported to the ISA, 1980–1992.
Working life and health. A Swedish survey
Reports have declined in number during the 1990s, especially
among younger persons. The decrease may be an effect of the
lower number of persons employed in the manufacturing industry.
It was found in previous studies that reported cases on average
require 15 weeks of sick leave and that two thirds of those who
were on sick leave for more than 90 days due to skin disorders had
applied for compensation for occupational disease.
Only a small percentage of hand eczema cases are reported to
the ISA. In a cross-sectional study made in Gothenburg during the
1980s it was found that approximately 10 percent of the population
had had hand eczema during the previous year (Meding, 1990).
Hand eczema was twice as common among women as among
men. Occupations with an elevated risk of occupational skin
disease were identified by combining information from the ISA
for 1990–91 and the census of 1990 (Table 8.1). The highest risks
were in occupational groups having frequent skin contact with
water, detergents and oils. In half of the reported cases water and
detergents are considered the cause of the hand eczema.
Table 8.1. Occupational groups with an elevated incidence of skin disorders,
Occupation (gender)
Number of reported cases
Hairdressers (women)
Machine fitters (women)
Cooks (women)
Dental assistants (women)
Toolmakers, machine tool setters and operators (men)
Cooks (men)
Cleaners (women)
Kitchen assistants (women)
Home helpers (women)
Machine assemblers (men)
Assistant nurses and hospital orderlies (women)
Packers (women)
Machine fitters (men)
Painters (men)
Female hairdressers have the highest risk. It is estimated that approximately 20 percent of them have to change their occupation
due to hand eczema (Meding, 1990).
The most common occupational skin disorder is irritant contact
dermatitis caused by water and detergents. Jobs with this kind
Skin diseases
ohälsa, våld och trakasserier
of exposure are common among women: hairdressers, cleaners,
kitchen assistants, home helpers, dental assistants etc. The high
incidence of hand eczema among working women is probably an
effect of exposures both at work and at home.
The most common causes of allergic contact dermatitis are nickel,
rubber chemicals, colophony (rosin), biocides, chromates, synthetic
resins and perfumes. Ten percent of women and 1 percent of men
in Sweden are allergic to nickel. Sensitization to nickel is often
not of occupational origin but due to cheap jewellery and other
metal objects in close contact with the skin-watch bands, buttons
etc. The risk of sensitization from occupational exposure to nickel
in tools is under investigation in Sweden.
Cement used to be a common cause of contact allergy to chromates. As a preventive measure, for the past 10 or 15 years ferrous sulphate has been mixed into cement produced in Sweden
to reduce the amount of sensitizing chromate.
The most common cause of contact allergy to rubber is the use
of rubber gloves. Rosin originates from coniferous woods. When
used in soldering it may cause allergic contact dermatitis on the
face. Rosin also occurs in paper, and there are a few reports of
allergic contact dermatitis in pulp and paper mill workers, but the
size of this problem is unknown. The rosin content is higher in
mechanical pulp than in chemical pulp (Karlberg et al, 1995).
Contact urticaria may be either allergic or non-specific. Since
the symptoms occur within a few minutes of the exposure the association is usually obvious, and the victims rarely seek medical
attention. The incidence of these symptoms is unknown, but a
study now in progress indicates that a few percent of all workers
have or have had work-associated contact urticaria. Allergic contact
urticaria is often caused by proteins and may sometimes develop
into chronic eczema, “protein dermatitis.” Common occupational
causes of contact urticaria are exposure to latex rubber in gloves,
handling food and working with animals. Latex allergy is a problem
mostly for hospital workers who use gloves. Approximately 6 million pairs of surgical gloves and 50–60 million examination gloves
of latex rubber are used every year in Sweden. At the beginning of
the 1990s, three percent of workers in dental care and in operation
theatres in Stockholm were allergic to latex (Wrangsjö, 1993).
Working life and health. A Swedish survey
Atopy increases the risk of hand eczema. At least 25 percent of
all Swedes are atopics. Several Scandinavian studies have shown
that persons who had atopic dermatitis in childhood are three
times as likely as non-atopics to develop hand eczema in adult
life (Meding, 1990). An analysis shows that at least 40 percent of
Swedes aged 16–24 who report occupation-related skin disorders
had atopic dermatitis in childhood.
Several international studies have revealed that atopic symptoms
are increasing rapidly in most countries. Since atopic dermatitis
increases the risk of hand eczema, an increased occurrence is expected. The prevalence of hand eczema in Sweden in the 1980s
was double that of twenty years earlier.
A patient will probably be able to continue working at the same job
if exposure to the cause of the eczema is eliminated. In a survey made
in Gothenburg, 8 percent of persons with hand eczema reported
that they had changed jobs because of the skin disease (Meding,
1990). Those most likely to change jobs were hairdressers, bakers,
dental assistants, cleaners, kitchen assistants, cooks and machine
Karlberg A-T, Gäfvert E, Lidén C. Environmentally friendly paper may increase
risk of hand eczema in rosin-sensitive persons. J Am Acad Dermatol
1995; 33:427–432.
Meding B. Epidemiology of hand eczema in an industrial city. Acta Derm Venereol 1990; suppl 153.
Wrangsjö K. IgE-mediated latex allergy and contact allergy to rubber in clinical
occupational dermatology. Arbete och Hälsa 1993:25.
Psykisk ohälsa, våld och trakasserier
9 Work-related
musculoskeletal disorders
The national surveys contained questions on working positions and
movements such as bending and twisting (Table 9.1). In 1993, 11
percent of men and 5 percent of women reported that they had heavy
lifts during at least half the workday. Repeated twisting and bending,
on the other hand, was more common among women. Heavy work
and heavy lifting were most common among construction workers.
Painters were the occupational group who most often worked with
the hands above shoulder level. Almost 80 percent of the painters
reported such work at least two hours per day.
Table 9.1. Frequencies of physical risk factors for musculoskeletal disorders
among employed persons in 1984, 1989 and 1993.
1984 1989 1993 1984 19891993
Heavy work at least half the time Twisting and bending in the same
way several times per hour every day Bending forward without support
from the hands or arms at least half
of working time Twisted work posture at least
half the time
Hands at or above shoulder level
at least half the time
* not available.
Working life and health. A Swedish survey
Low control and repetitive work increase the risk of disorders
in the musculoskeletal system. In 1993, 51 percent of women
and 37 percent of men reported that they could control their
work tasks for at most half of their working time (Table 9.2).
Repetitive­ work was more common among women, and the
changes in frequencies between 1984 and 1993 are rather
Table 9.2. Frequencies of psychosocial risk factors among employed persons
in 1984, 1989 and 1993.
198419891993 198419891993
Can decide the work pace during at
most half of working time
Repeat simple movements several times
per hour for at least half of working time 34
Usually cannot decide for themselves
when task should be done
Usually not in the position to decide
about planning of own work
48 51
42 39
48 49
30 28
* not available.
Sick leave, disability retirement and occupational injuries
Musculoskeletal disorders are the most common type of occupational injury and the primary cause of sick leaves and early
disability retirements. In 1994 these disorders accounted for 40
percent of all early retirements in men and 53 percent in women.
In 1990, 37 percent of all days of sick leave were due to musculo­
skeletal disorders. In 1994, 73 percent of all occupa­tional diseases
were musculoskeletal disorders; 37 percent occurred in the neck and
shoulders, 22 percent in the arms and 19 percent in the back.
In the ISA statistics, musculoskeletal disorders are divided into
injuries and diseases. The number of reported diseases varies with
occupational group; they are most common in occupational groups
with high physical load (Table 9.3). Occupational groups with a
low risk included engineers and teachers.
Injuries to the musculoskeletal system have a pattern different from
the diseases. About 20 percent of all reported injuries in 1993 were
overloads of some part of the body. They were usually due to lifting,
Psykisk ohälsa,musculoskeletal
våld och trakasserier
often in moving a patient. These injuries are most common among
women, especially those working in health services (Table 9.4).
Table 9.3. The five occupations with the highest frequencies of occupational
disorders caused by work load. Only occupational groups with at least 1,000
workers are included.
Frequency per 1,000 workers
Glass-, pottery, tile work
Butchers and meat packers
Machine fitters
Welders and flame cutters
Metal processing workers
Butchers and meat packers
Foundry workers
Floor layers
Glass formers and cutters
Table 9.4. The occupational groups with the highest frequencies of injuries due
to overexertion, 1990–91.
Frequency per 1,000 workers
Due to lifting
Home care workers
Nurses’ assistants for mentally retarded
Nurses’ assistants and auxiliary nurses
Nurses’ assistants in psychiatric care
Concrete workers
Dairy workers
Cast concrete pruduct makers
The sick leave after an occupational injury provides some information
about the seriousness of the consequences. The patient’s absence
depends on both the seriousness of the disorder and the work environment. The risk of sick leave due to a work-related musculoskeletal
disorder (disease or injury) was highest in heavy industry (Table
9.5). Women in the construction industry had a lower frequency
than men, since women in this industry usually have office work.
In all other groups, the consequences are generally more serious
Working life and health. A Swedish survey
for women than for men. This is partly an effect of the fact that
women more often have repetitious manual work.
Table 9.5. The industries with the highest number of days of absence due to
musculoskeletal disorders (diseases or injureis).
Number of days per worker
men women
Rubber and plastic manufacturing
Sea transport
Steel/metal processing
Metal products manufacturing
All industries
Occupational groups with high risk
Construction workers run a high risk of work-related diseases or
symptoms in the musculoskeletal system. Regardless of whether sick
leaves, symptoms, early retirements or reports to the ISA are studied,
the pattern is the same. The job is physically demanding. Other occupational groups with physically demanding work, such as butch­ers,
loggers and welders, also have an elevated risk. For women, highrisk jobs include nurses’ assistant, auxiliary nurse, shop assistant,
dressmaker, butcher and some machinists. High risks are usually
associated with high physical demands and repetitious, monotonous
work. Monotonous work at cash registers and in packing may partly
explain why female shop assistants have a high risk.
Drivers have an elevated risk of musculoskeletal problems, possibly due to whole-body vibration, prolonged sitting and heavy
lifting during loading and unloading.
Dental care workers, including dentists, are the only college-educated professional group reporting a high prevalence of symptoms that
influence their ability to work. No other occupational group has such a
high percentage of symptoms involving the neck, shoulders and arms.
The cause is probably precision work in combination with difficult
working positions. An additional problem is doing piecework.
Teachers, systems analysts, programmers, engineers and office
workers in banks and insurance companies are examples of occupational groups with a low frequency of symptoms, sick leave
Psykisk ohälsa,musculoskeletal
våld och trakasserier
and early retirement. Their jobs are neither strenuous nor monotonous and do not require fixed or difficult working positions.
These workers also usually have control over their work.
The difference in risk between women and men
The frequency of reported work-related diseases in musculoskeletal system in 1990–91 was 13 per 1,000 among male machine
fitters, compared to 39 per 1,000 among female machine fitters.
The women more often have monotonous work and less control
over it. The women also have a double load, since they also do
housework. Physiological differences between men and women
may also have importance, e. g. for tendinitis due to repeated
movements using power grips.
The work environment surveys from Statistics Sweden indicate that
the factors that may influence the risk of musculoskeletal disorders
remained about the same between 1984 and 1993 (Tables 9.1 and
9.2). A similar trend is seen in the Statistics Sweden study of
work-related musculoskeletal symptoms (Table 9.6). There was
a slight increase in the proportion of persons with symptoms
between 1984 and 1989, but the proportion in 1989 was about
the same as in 1993.
Table 9.6. Symptoms reported in national surveys, 1984–1993. Percentage of
all employed men and women.
Symptoms every week
Hands and wrists
Shoulders and arms
Upper back or neck
Lower back
Working life and health. A Swedish survey
Statistics from the ISA give a different picture. The number of
reported occupational diseases and injuries increased during the
1980s but than declined considerably (Figures 9.1 and 9.2). The
peak in reported occupational diseases in 1993 is due to a change
in the law on compensation for occupational diseases.
Figure 9.1. Frequency of reported occupational accidents resulting in musculo­
skeletal injury since 1980.
Figure 9.2. Frequency of reported occupation-related dis­eases in musculoskeletal
systems since 1980.
våld och trakasserier
It is not known whether the decline in reported occupation-related disorders of the musculoskeletal system during the 1990s is
an effect of a decreased number of persons suffering from these
disorders or an effect of the changed rules for compensation.
Another hypothesis is that fewer persons report musculoskeletal
symptoms and pain when unemployment is high, since they are
afraid they may lose their jobs if they do so.
There are still several physically heavy jobs, especially in the service sector. The increased number of elderly persons has increased
the demand for health care. More patients and elderly people
will be staying in their homes, where the options for assistance,
ergonomically adjusted work and aids are limited. It is therefore
probable that heavy lifting will continue to be a problem in many
jobs in the health care sector.
Working life and health. A Swedish survey
Psykisk ohälsa,
at work
våld och trakasserier
10 Threats, violence,
sexual harassment and suicide
There is little knowledge or research on the association between
the work environment and mental diseases. During work with this
report we tried to find out about sick leave due to mental diseases in
different occupational groups and industries, but found it to be impossible. In this report violence and threats at work, bullying, sexual
harassment and suicide are related to the work environment.
Violence and threats at the workplace
During recent years violence and threats have become more widely
recognised as an occupational health problem. The cases in the
ISA records were usually reported because of mental illness or
injuries. In 1985–94, 600 to 800 such cases were reported among
women and 400 to 500 cases among men. Approximately half of
the reports came from the health care sector. Other occupational
groups with a high reported frequency were watchmen and policemen (Table 10.1).
In the 1989–93 national surveys by Statistics Sweden (NWE),
a random sample of the working population was asked whether
they were threatened or physically attacked in their work. Nine
percent of the women and 5 percent of the men reported that they
were threatened or injured by violence at least twice per month.
The difference between men and women is statistically significant.
Threats were most common in the health care and social sectors,
Working life and health. A Swedish survey
where 22 percent of the women and 30 percent of the men reported
such problems. Men are more often employed in psychiatric care,
where violence and threats are more common.
Table 10.1. Occupational groups with a high frequency of reported occupational
injuries due to threats or violence. Only cases with sick leave or tooth injury.
Number of cases per year
Attendants in psychiatric care Watchmen Hospital aides for mentally disabled Policemen Tram and underground drivers Ticket collectors Social workers 71
Relative frequency
Men Women
Policemen are the occupational group where threats and violence
are most common (Table 10.2). Half of those who report violence
or threats have feelings of apprehension about going to work. This
feeling is much more common among men. Those who have been
threatened also report more fatigue and more sleep disorders.
Table 10.2. Occupational groups where threats and violence are common. The
figures give the percentages of all men and women who report that they suffer
from violence or threats at least twice per month.
Attendants in psychiatric care
Hospital aides for mentally disabled
Social workers
Bus and taxi drivers
Assistant nurses
Post office clerks
In 1992 the occurrence of threats and violence was surveyed in the
Swedish Confederation of Professional Associations. On average
11 percent reported that they had been threatened or had suffered
from violence at work during the previous two years. This was
most common among occupational therapists, social workers and
physicians, about 25 percent of whom had been exposed to threats
or violence. In 1993, 4 percent of prosecutors reported suffering
from violence at work during the previous three years and 32
Psykisk ohälsa,
och trakasserier
harassment and suicide
percent reported being threatened. Violence directed towards the
prosecutor’s family was reported by 2 percent, and threats against
the family were reported by 11 percent.
About 250 women and 100 men per year reported bullying as an
occupational disorder in 1992–93. In a 1989–93 national survey,
5 percent of women and men reported that they suffered from
personal harassment at least twice per month. This analysis indicates that differences between occupations and industries are rather
small. In comparing men and women in different occupational
groups, only women in the manufacturing industry and men in
teaching were found to suffer a significantly higher frequency of
In 1991, 75 percent of men suffering from bullying reported that
they were attacked by other men and 3 percent only by women.
Forty percent of women were mobbed by other women, 30 percent
by men. Others were attacked by both men and women. In the
national surveys in 1989 and 1993 bullied persons often reported
health disorders, fatigue, headaches or stomach pains.
Sexual harassment
In a national survey made by Statistics Sweden in 1993, two percent
of women and one percent of men reported sexual harassment.
Other Swedish and international studies have reported higher frequencies for women, from 13 to 23 percent (Lagerlöf, 1993). The
differences are probably due to a combination of differences in
interview technique and definitions of harassment. In an analysis
of the work environment survey of 1993, sexual harassment was
more common for women if they worked in restaurants or hotels or
drove a bus or taxi. More than half the women and men who have
been harassed have feelings of apprehension when they go to work.
They also more often report sleep disturbance and fatigue.
About 1,500 deaths per year were registered as suicide during
the 1980s. One third were women. For more than another 500
Working life and health. A Swedish survey
deaths per year it is difficult to determine whether the death was
accidental or suicide. Studies have shown that most of these cases
are probably suicide, and they are therefore often included in
suicide statistics. There are no national statistics on suicide attempts. It is estimated that there are approximately ten attempts
per completed suicide.
To study the occurrence of suicide in different jobs, the 1980
census-linked Death Register for two periods, 1981–86 and
1987–92, was investigated. A similar analysis was made previously
for 1971–75 and 1976–79.
For men, ten occupational groups show an elevated frequency of
suicide during at least three of the four time periods (Table 10.3).
Table 10.3. Occupational groups with an elevated risk of suicide among men,
Attendants in psychiatric care
Hospital aides
Horticultural workers
Forestry workers
Ship’s deck and engine room workers
Machine tool setters and operators
Chemical process workers
Unskilled workers
Warehouse and storage workers
For women, suicide is more common among physicians and nurses.
Some occupational groups have a lower than average frequency of
suicide during both 1981–86 and 1987–92 (Table 10.4). Among
men, engineers and managers, and among women teachers, shop
assistants and children’s nurses, have low frequencies.
Table 10.4. Occupational groups with a lower than average frequency of suicide
during both 1981–86 and 1987–92. The analysis includes only groups with at
least ten cases.
Electrical and electronics workers
Managers in machine shops
Systems analysts and programmers
Other business administrators
Primary school teachers
Shop assistants
Children’s nurses
Psykisk ohälsa,
och trakasserier
harassment and suicide
Just why suicide is more or less common than average in certain
occupations is unknown. The cause of the high risk among persons working in health care is unknown, but mental strain and
knowledge of how to commit suicide have been suggested as
contributing factors. In an inquiry in Stockholm, 12 percent of
male physicians reported that they had considered committing
suicide during the previous year, compared with 2 percent of male
engineers or managers. Among women there were no obvious
differences between occupational groups.
The number of occupation-related mental disorders reported to
ISA increased between 1985 and 1993. However, it is not known
whether this increase is due to changes in work environments
or to a difference in willingness to report such an occupational
disease. It will not be possible to use ISA records to study mental
disorders in the future because they are usually not covered by
the new compensation laws.
Figure 10.1. Reported occupation-related mental illnesses, 1985–93 (ISA).
There is no information for 1988–89.
The national surveys for the 1989–93 period show a slight increase
in the percentage of persons who reported that they had been
Working life and health. A Swedish survey
threatened or suffered violence. The frequency had increased
from 8.3 to 9.6 percent for women and from 4.7 to 5.9 percent
for men. Both increases are statistically significant. The only single
occupational group with a statistically significant increase was female
shop assistants, where the frequency increased from 3 percent in
1989 to 9 percent in 1993. However, no corresponding increase
in reported injuries due to threats or violence is indicated in the
ISA records.
The number of men who reported bullying in the national survey increased from 4 percent in 1989 to 5 percent in 1993. Five
percent of women reported bullying in 1989 and 1993. There is
not enough register data to indicate whether the number of persons who suffer from sexual harassment increased or decreased.
The incidence of suicides declined between 1980 and 1993 for
both men and women in all economically active ages. For male
physicians there was an elevated incidence from 1971 to 1979.
From 1981 to 1992 there is no statistically significant increase in
risk. However, a study of different age groups showed that during
this period male physicians born between 1921 and 1940 had an
elevated risk. Between 1987 and 1992 the relative risk for suicide
in male physicians was 1.1. For other occupational groups with
a higher than average risk, there were no significant changes in
relative risk between the different periods.
Threats, bullying and sexual harassment have quite recently been
recognised as occupational health hazards. The Work Environment
Act stipulates that the employer must plan and organise the work
so that both threats and bullying are prevented, and must make
it clear that bullying is not acceptable at the work site. Bullying
is considered to be related to shortcomings in work organisation
and leadership. The equality act makes it the responsibility of the
employer to prevent sexual harassment of employees.
Lagerlöf E. Women, work and health. Ministry of Health and Social Affairs,
1993 (Ds 1993:38), Stockholm, 1993.
Psykisk ohälsa, våld och trakasserier
11 Accidents at work
Statistics on accidents at work have been kept in Sweden since
1906. There are now three national registers. The Information
System on Occupational Injuries, ISA, which covers all employed
persons, is based on the accident reports submitted by employers
to the social insurance system. The ISA is administered by the
National Board of Occupational Safety and Health.
There is an additional national private insurance system (TSI)
which includes those who are insured. Membership in TSI is
negotiated between unions and employers.
Statistics Sweden makes an annual survey of a random sample of
the population aged 16 to 64. They are asked whether they have
symptoms caused by their work and if these have caused absence
from work. Approximately 25,000 persons have answered these
questions every year since 1991.
The reporting to ISA has changed since 1992, when it became
a rule that employers must pay the first 14 days of sick leave. However, it seems that it is accidents in small companies and with
brief absences that are not reported to ISA to the same extent as
previously. As 80 percent of the labour force works in companies
with at least ten employees, the statistics are not much influenced
by the decreased reporting from small firms. The number of persons covered by TSI has declined during the 1990s.
Fatal accidents
There have been 100 to 150 fatal on-the-job accidents per year
among employees and self-employed persons since 1990. The
incidence is 4.9 cases per 100,000 employed men and 0.4 cases for
Working life and health. A Swedish survey
employed women. Since the late 1980s the risks have decreased
for men, but are unchanged for women. The largest decreases
among men occurred in the oldest ages. The incidence of fatal
accidents at work is low in Sweden compared to other countries,
including the other Nordic countries. A little less than half of the
deaths during the 1990s occurred in accidents between vehicles or
aircraft. Falling caused 14 percent of the deaths, falling trees and
other falling objects 14 percent, and violence against persons was
the cause of two percent. The ISA statistics of deaths are reliable
but the number of cases is small, which makes it difficult to relate
the deaths to particular occupations or industries. During the past
years pilots, fishermen, miners and ship deck and engine room
workers have had the highest risks of fatal occupational accidents.
During the 1990s there has been a change in the age distribution
of fatal accidents. The youngest workers now have the highest
frequencies, whereas in earlier years older workers always had
the higher frequencies.
Figure 11.1. Number of fatal occupational accidents per 100,000 employed in
France, the USA and Sweden.
Decline in accidents at work
The decreased tendency to report accidents applies mostly to
milder cases. The analysis here is therefore based on cases with
Psykisk ohälsa,
våld och trakasserier
at work
at least 30 days of sick leave. Approximately 10,000 such accidents
per year have been reported to the ISA in recent years. The rates
approximately halved between 1986 and 1993, with decreases
among both men and women and in all age groups. The statistics
from TSI show that serious accidents decreased by 22 percent
between 1989 and 1992.
Figure 11.2. Number of occupational accidents with at least 30 days of sick
leave per 1,000 employed, 1986–93.
Overloads to the musculoskeletal system are the type of accident
which has shown the largest decrease. Between 1990 and 1993
the number of reported cases per 10,000 employed have halved
for both men and women. For women, machine accidents were
halved and falling accidents decreased by 20 percent. The largest
decrease occurred among younger women. For men, machine
accidents and falling accidents each decreased by one third. For
both sexes, the oldest workers have the highest incidences of accidents requiring long sick leaves.
There are many explanations for the decreasing frequency of
occupational accidents. Safety work is constantly going on at the
workplaces. The work of the Labour Inspectorate and stricter
requirements regarding internal control by the employer may
influence the decrease. The number of persons working in heavy
industry has also declined. There are also fewer recently employed
Working life and health. A Swedish survey
young men, who have a higher risk of accidents than older workers. However, it is not possible to judge the relative importance
of these different factors.
The decrease in the total number of on-the-job accidents does
not mean that the risk for a single worker has decreased in a similar
way. In hospital services, for example, the frequency of occupational
accidents for assistant nurses and hospital aides was the same in
1990 as in 1980, and during this period the number of accidents
among nurses increased. There is a similar pattern among home
helpers and children’s nurses.
Occupational accidents in heavy industry
The risks for occupational accidents in 290 different occupations
were compared. Among men there were 84 occupations with a
statistically significant higher risk for accidents compared to all
working men and women. Most of them, 57 occupations, were
in the manufacturing industry; there were 9 groups in transport
and communication and 7 groups in farming and fishing. Among
women there were 24 occupations with higher than average risk:
14 were in the manufacturing industry and four in agriculture,
forestry and fishing. The 15 occupational groups with the highest
relative risks were all male. A different pattern appears if specific
types of accidents are analysed; electricians, for example, have a
high risk of accidents involving electric power.
The highest risks were among ship deck workers, carpenters
doing construction work, firemen and wood processing workers.
There was no occupational group of women with a relative risk
of three or higher compared to all employed persons. The study
of occupational accidents during 1985–86 showed the same pattern.
Disabling accidents were studied in TSI, and the pattern is
similar to that in ISA. The cost of an injury can also be analysed
in TSI. The occupations which had the highest payment from
TSI per worker are wood processing workers, miners, farmers
and forestry workers.
The Work Environment Act also covers school pupils, but there
is no reporting system for accidents in schools. New studies have
shown that the risk of accidents is higher for school pupils than
Psykisk ohälsa, våld och trakasserier
12 Work and gender
Sweden has a higher percentage of employed females than most
countries. In 1995, 76 percent of women and 80 percent of men
aged 16 to 64 were gainfully employed.
Horizontal segregation
Many occupational groups are held predominantly by either men
or women (Figure 12.1). This is called horizontal segregation. In
1990 about 40 percent of women worked in occupations where the
proportion of women was at least 90 percent. About 45 percent
of the men had jobs where the proportion of men was at least 90
Figure 12.1. The 15 largest occupational categories in percent.
Working life and health. A Swedish survey
Many women work in health services and home care. In most other
countries home care is part of housework and is therefore not included in the labour statistics. The Swedish labour market therefore
appears to be more sex-segregated than that of other countries.
Figure 12.2. Percentage of women in the five most common jobs dominated
by men.
Figure 12.3. Percentage of men in the five most common jobs dominated by
våld och trakasserier
The proportion of occupations with an even distribution be­tween
the sexes increased from 21 to 30 percent over the past ten years.
The changes from 1975 to 1990 in proportions of men and women
in the five largest occupations dominated by men or women are
shown in Figures 12.2 and 12.3. The proportion of women among
sales representatives increased from 12 to 25 percent, while the
proportion of men working as secretaries decreased from 13 to
9 percent. The proportions of women increased most strongly
in male-dominated jobs requiring a university degree: biologist,
lawyer, chemist, physicist, physician, dentist, veterinarian etc.
Vertical segregation
It is common that within an occupation women have lower status.
This is called vertical segregation. The lower status of tasks done
by women also means that women’s salaries are lower than those
of men. A comparison between women and men who finished high
school in 1982–1985 and worked full time in 1990 showed that
the median annual salary for women was approximately 30,000
SEK lower.
Table 12.1. Women working as managers in the private and public sector,
Privat sector
Public sector
county councils
Percentage of women Percentage of women
among managers
among all employees
The difference has several causes: women work in occupations with
lower salaries (horizontal segregation) and women within the same
occupation as men have lower income (vertical segregation). In 1968
men had on average 27 percent higher salary than women in the
same sector with similar education, experience and positions. By
1981 this difference in salaries had decreased to 12 percent . During
the 1980s the difference did not change, and in 1991 men had 13
percent higher salary than women with similar jobs.
Working life and health. A Swedish survey
Paid and unpaid working time
Women work part time more often than men do. Almost 40 percent of women worked part time in 1995, compared to about 10
percent of men. In 1992, 28 percent of absence for women and 3
percent for men was due to taking care of children.
On average, women used 26 hours per week for paid employment and 19 hours per week for housework in 1991. Men used
40 hours per week for paid employment and only five hours per
week for housework.
Differences in work environment
The sex-segregated labour market also means differences in
working conditions. Men are more often exposed to oils, organic
solvents, vibrations, noise and heat. Men work more often in the
manufacturing or construction industry where such exposures are
common. Women are more often exposed to violence, threats and
bullying. They also more often have musculoskeletal disorders.
The national surveys of the work environment in 1989–93
were analysed for information on working conditions in jobs held
predominantly by either men or women.
Table 12.2. Working conditions reported by white-collar workers in jobs held
predominantly by women or men or with even sex distribution, percent.
Predom. women
women men
Difficult working
Monotonous work
Heavy lifting
Can influence
own work
Can learn new things
and improve at work
Predom. men
women men
Even sex distribution
For the analysis the material was divided into blue-collar and
white-collar workers, and a predominantly men’s (women’s)
job was defined as a job where at least 75 percent of the workers were men (women). Other jobs were classified as jobs with
even sex distribution. Female white-collar workers more often
had difficult working positions and lower control over their own
våld och trakasserier
work. They also had fewer opportunities to learn and improve
at work.
In blue-collar jobs held predominantly by women the frequency
of difficult working positions was similar for men and women.
However, also in these jobs, men had more control over their
working time and their work: 56 percent of the men, compared
to 44 percent of the women, could themselves decide when a job
should be done. Men also reported less stress.
Table 12.3. Working conditions reported by blue-collar workers in jobs held
predominantly by women or men or with even sex distribution, percent.
Predom. women
women men
Difficult working
Monotonous work
Heavy lifting
Can influence
own work
Can learn new things
and improve at work
Predom. men Even sex distribution
women men
women men
The result shows that working conditions for men and women are
different, whether they are blue-collar or white-collar workers.
The sex that is in the minority is not always discriminated against.
Men in jobs held mostly by women, for example, had in general
greater control over their job and workplace than women with
the same jobs.
The horizontal segregation is decreasing slowly. The decrease
will probably be faster in male-dominated jobs that require a higher education. Women will probably obtain more well-paid jobs,
while men will probably not choose jobs in occupations dominated
by women because such jobs have lower salaries, less good career
prospects, and fewer opportunities to improve competence.
Working life and health. A Swedish survey
Psykisk ohälsa, våld och trakasserier
13 Elderly persons
in working life
There is no commonly accepted definition of an elderly worker.
The ILO has defined an old worker as one 55 years old or more
(Schneider, 1995). Some Swedish authorities use the same age
limit. If both the employers’ expectations and early signs of biological ageing are considered, a limit of 45 years may be more
realistic. The proportion of persons aged 45–64 is increasing in the
working population. In 1985, 38 percent were in that age group;
in 1995 it was 41 percent; and it is estimated to be 45 percent in
2005. The proportion of the population that is working has been
weakly decreasing in elderly men while it has increased for elderly
women (Figure 13.1).
Figure 13.1. Proportions of employed men and women aged 55–59 and 60–64,
Working life and health. A Swedish survey
During the recession in the 1990s it decreased for both men
and women. The decreased proportion of elderly persons that
are working in Sweden is an effect of the rise in unemployment
and an increase in early retirements. Early retirement is approximately five times more common among persons 60–64 years of
age than among those 45–54 years of age. Monotonous work is
associated with a 1.5 to two times higher than average risk for
early retirement in persons aged 45 or older. The most common
reasons for these early retirements are musculoskeletal disorders
(approximately 30%) and cardiovascular diseases (7%). Many
elderly persons neither work nor have early retirement pensions.
About every third woman and every fifth man 64 years of age in
1992 was in that group.
Figure 13.2. Proportions of employed men and women 60–64 years of age in
different countries in 1993 (source ILO).
Among business sectors, hotels and restaurants have the lowest
proportion of elderly workers and farming and forestry have
the highest (Table 13.1). These differences probably reflect both
working conditions and cultural and temporal shifts in selection
of occupation. Industries which are declining often have higher
proportion of elderly workers, whereas expanding industrial sectors
have a higher proportion of younger persons. The proportion of
elderly men in the manufacturing industry is remarkably low, 9
Psykisk persons
ohälsa, våld
in working
och trakasserier
percent, which indicates that the working conditions are poorly
adjusted to the capacity and demands of this group.
Table 13.1. Elderly persons (55–64 years of age) in the labour force in different
industries in 1994 (percentages).
Farming and forestry
Manufacturing and mining
Restaurants and hotels
Wholesale and retail trade
Pubblic administration
Health care
Demands in working life and the abilities of the elderly
Jobs with high physical demands exist in farming, forestry, mining,
construction and manufacturing, but also in the service sector
– health care, cleaning, hotels and restaurants, and storage and
transportation. Many elderly women have service jobs. In 1989–93,
44 percent of women 50–64 years of age who worked in health care
reported that they had physically demanding work; 54 percent of
women in service jobs and 67 percent of men in manufacturing
and mining also reported physically demanding work.
In some industries and occupations the physical demands are
probably too high for elderly people to cope with. However, there
are no measurements that give a representative picture of both the
job demands and the employee’s abilities. In the national surveys
of 1989–93 approximately equal proportions of young and elderly
workers reported that they were physically tired after work at least
once a week. The difference between men and women was large,
indicating that the women had work that was closer to the limits of
their physical capacity. Recent research indicates that psychological
and mental capacity declines with age at a much slower rate than was
previously believed. Moreover, the elderly can compensate for losses
of cognitive capacity and psychomotor speed by more strategic
ways of planning and performing their work. The productivity of
the elderly has been shown in several studies to be as high as that
of younger employees.
Working life and health. A Swedish survey
Elderly people are more likely to have chronic diseases. Thus,
cardiovascular diseases were reported in 22 percent of persons
aged 55–64 in 1992/93, but in only 3 percent of persons aged
35–44. Decreased agility was reported by 28 percent of men and
20 percent of women aged 55–64, but by only 5 percent of men
and 3 percent of women aged 35–44. Elderly workers thus are
more likely to require adjustment of their work to compensate
for chronic disease.
The lower physical capacity of elderly persons will probably
in the future have less importance in traditionally heavy industry
as physically demanding work becomes easier through mechani­
sation. Some jobs, such as construction work, manual garbage
collection and firefighting (especially smoke diving), have such
high physical requirements that a change of work in higher ages
is more the rule than the exception. If there is no other work or
labour market for these groups there is a high risk that they will
become unemployed.
In the future, work will continue to be physically demanding in
the service sector, especially home health services. Elderly women
are a vulnerable group, as they make up a large proportion of
health care workers and have many heavy duties. Approximately
every fourth women 55–64 years of age has lost some agility, and
there is consequently a need to adapt work or provide an alternative
labour market for this group. Elderly persons can handle heavy
work easier if they can decide for themselves when and how it
should be done. Work organisation is therefore important. Moreover, simple ergonomic improvements such as introduction of
lifting aids, better lighting and lower noise levels will reduce the
total physical stress on the elderly employee.
For persons born before 1950 the number of years in primary
and secondary school is correlated to age. Older persons have fewer years of schooling. Elderly people have a vulnerable situation
in the labour market due to this poor schooling. Many of them
went to school 25 to 45 years ago and their knowledge is partly
out of date. Moreover, the demands for education and training
are increasing in working life. The proportion of men aged 55–64
who had jobs requiring at least three years of education increased
Psykisk persons
ohälsa, våld
in working
och trakasserier
from 17 percent in 1984 to 26 percent in 1993. A critical factor is
the elderly person’s ability to adjust fast enough to new demands
on the labour market. “Lifelong learning” is therefore necessary,
especially considering the present rapid rate of change. During
recent years the term “lifelong learning” has been used especially
to refer to learning and development in close connection with
the workplace. This type of learning is extremely important and
probably extremely efficient for elderly persons.
Schneider G. Ageing societies: Problems and prospects for older workers. World
Labour Report 1995. ILO, Geneva, 1995, pp 31-54.
Working life and health. A Swedish survey
Psykisk ohälsa, våld och trakasserier
14 Immigrants
Immigration to Sweden has been substantial since World War
II, but its composition has shifted. For example, towards the end
of the war the immigrant population was composed largely of
refugees from Estonia, Denmark and Norway. During the boom
in the 1940s, skilled male workers were recruited from Italy and
Hungary. This recruitment of skilled male workers continued
during the 1950s, but most of the immigrants to Sweden during
that decade were women, mainly from Germany and Finland,
who worked as housemaids. In the 1960s the Swedish economy
enjoyed an upturn and a great many workers from abroad were
recruited, mainly from Finland and southern Europe. Since the
beginning of the 1970s, refugees have become the predominant
immigrant group.
Immigrants are not a homogeneous group. Their occupations and
employment opportunities differ, depending on what nationality they
have and when they arrived in Sweden. Some immigrants, such as
well-educated specialists from Western European countries, have
a strong position on the labour market, comparable to that of native-born Swedes.
Eight percent of those who were gainfully employed in Sweden
in 1994 were born abroad, and 4 percent were foreign citizens.
The proportion of immigrants who are employed varies according
to country of birth (Table 14.1). The differences are larger among
women than among men; e.g. less than half of the women from
Iran and Turkey have jobs.
Working life and health. A Swedish survey
Table 14.1. Percentages of people aged 16­–64 who were gainfully employed
in Sweden in 1994. By country of birth.
Total foreign-born
Total including those who
were born in Sweden
Immigrants more often have jobs in occupations that do not require
high levels of education and training, e.g. in the manufacturing
industry and the service sector, or else they work in occupations
that require high educational levels but are not dependent on
specific ties to Swedish culture, e.g. engineers, physicians and
scientists. If job demands include fluency in the Swedish language, e.g. certain administrative positions, then the proportion of
immigrants is usually low.
Immigrants constitute a high proportion of the work force in
certain occupations (Table 14.2.). Male immigrants work as cleaners, in the textile industry, and in hotels and restaurants more
often than Swedes do. This tendency has not changed during the
downturn of the economy in the 1990s: it was the same in 1994 as
in 1989. Although immigrant women work in the manufacturing
industry more often than Swedish women, female immigrants from
the other Nordic countries and Germany are exceptions.
Temporary employment became more common during the recession in the 1990s. On the average, 17 percent of the immigrants
in Sweden had only temporary employment in 1994, compared
to 9 percent for the total Swedish labour force.
Immigrants are more often found in jobs that impose negative
stress. This is the case even if the comparison is adjusted for occupation. For example, handicapped immigrants have jobs that
are physically and mentally more stressful than their Swedish
Psykisk ohälsa, våld och trakasserier
Table 14.2. Occupations with the highest over-representation of immigrant men
and immigrant women in 1994. Parentheses enclose an index which is the ratio
between the number of immigrants in that occupation and the expected number.
Only occupations in which at least 5,000 men or women are employed have
been included.
Country of birth
textile workers (4.3)
glass workers (4.0)
metal workers (3.7)
textile workers (3.7)
Norway health care (5.2)
dental care (3.4)
transportation (5.8)
precision manu-
facturing work (4.3)
cleaners (8.1)
textile workers (7.1)
Poland cleaners (5.9)
textile workers (5.5)
Born outside Europe* cleaners (8.2)
hotel and restaurant (7.0)
All foreign born
agricultural Work (3.8)
systems analysts (3.1)
security guards/
watchmen (2.5)
machine fitters (1.5)
systems analysts (3.1)
road workers (3.1)
wood workers (10.8)
precision manu-
facturing work (5.2)
provision workers (9.9)
electronics workers (9.0)
electronics workers (4.8)
graphics workers (3.8)
textile workers (4.5)
electronics workers (3.5)
cleaners (4.3)
machine fitters (2.4)
hotel and restaurant (3.6) textile workers (2.3)
* Except Australia, Canada, New Zealand and the USA (Source: Survey of
Labour, 1994).
Men who were born abroad have shift work more often than
native Swedish men do, 14 percent and 6 percent respectively.
The difference for women is smaller, 5 percent and 3 percent
The frequency of occupational accidents per 1,000 people was approximately 20 percent higher among immigrants, and occupationrelated illness was 30 percent higher among immigrants, during
the 1988–92 period. These analyses are adjusted for gender, age
and occupation. The reason for the higher incidence of accidents
and illnesses among immigrants is still unclear.
Future trends
On average, immigrants do not have less education than people
who were born in Sweden, often the reverse. However, a long
education does not guarantee a job that requires one. Immigrants
less frequently have jobs with high educational requirements,
Working life and health. A Swedish survey
although there are large differences between immigrants from
different countries. People born in Sweden receive more training
during working hours than immigrants do.
The increasing wage differentials between jobs requiring high
education and jobs requiring little education will probably increase
the current disparities between people who were born in Sweden
and those who were not.
Psykisk ohälsa, våld och trakasserier
15 Disability
A disability is a restriction that makes a person unable to perform
an activity in a normal way. It can be present from birth or due
to disease or injury. A handicap is defined as the consequence
of a disability. The handicap is always relative and occurs in the
interplay between the individual and the environment. A disabled
person can be handicapped in some tasks but not in others. It is
estimated that about one fourth of the population of the European
Community has some type of disability. The description of the
situation for disabled persons is difficult in the absence of longterm statistics. The Labour Market Board has 68,000 persons
registered for various measures due to disability affecting work:
46,000 had part of their salary paid by the government, 29,000
worked in a public company for disabled persons (Samhall), 17,800
worked in special government programs and 4,800 worked in the
public sector with support.
According to Statistics Sweden, 40 percent of all persons with
impaired vision had employment in the ordinary labour market
and 27 percent of all persons with impaired hearing had disability
pensions. According to a Swedish study made in 1991–92, 28
percent of all persons with impaired hearing had not told their
employer about their impaired hearing. There has been a disability
ombudsman for disabled persons in Sweden since 1994. It is the
duty of this authority to protect the rights of disabled persons and
Working life and health. A Swedish survey
to follow relevant international developments, especially within
EC. The disability ombudsman should also spread information,
influence public opinion, give advice about rights, and monitor
laws and regulations.
Samhall is a publicly owned group of companies which employ
29,000 disabled persons. The objective is for them to leave Samhall
and obtain employment in ordinary companies. In 1990–94 less
than 4,000 persons, or 2 to 5 percent of the employees, left Samhall
for other jobs. Half of those who got jobs in ordinary companies
were employed in a smaller private company, and 37 percent got
the same type of job they were doing in Samhall.
Work environments
In the national survey of living conditions (ULF), disabled persons
more often report that they have monotonous work, awkward working positions, work with vibrating tools, and noisy and dirty jobs
(Table 15.1).
Table 15.1. Working conditions of disabled persons, 1990–93, percent (source:
Little opportunity to influ-
ence working conditions
Highly unsatisfied Proportion
with type of job
of population*
Impaired hearing
Impaired vision
Heart disease
Highly impaired
work capacity
38 29
27 31 13
Total population
*Proportion in percent of all who answered the survey (ULF) and reported impairment.
The work is described as busy and monotonous. However, there is
little difference between disabled and non-disabled persons in their
reports about mental strain, social relations and the occurrence of
accidents (Table 15.2). A comparison between women and men
shows that there is the same difference between the sexes among
disabled persons as in the total working population. Men more
often have noisy, dirty jobs and work more often with vibrating
tools. Women more often have monotonous work.
Psykisk ohälsa, våld och trakasserier
Table 15.2. Working conditions reported by persons with impairments and by
the total population 1990/93, percent (Source: ULF).
Monotonous, repeated movements
Awkward working positions
High noise levels
Very dirty work
Busy and monotonous work
Mental strain
Close friends at work
Isolated at work
Accident at work during the past 12 month
A comparison between disabled Swedes and disabled immigrants
shows that differences in working conditions are rather small but
statistically significant. Disabled immigrants, more often than
disabled Swedes, have work with physical and mental strain. The
trends in working conditions for disabled persons can be studied
in the national surveys made between 1976 and 1990/93 (ULF)
(Table 15.3).
Table 15.3. Changes in working conditions between 1976 and 1990–93,
percent (Source: ULF).
Monotonous movements
Awkward working postures
High noise levels
Busy and monotonous work
Mental strain
Disabled persons
Persons with
impaired hearing
Total employed
Working life and health. A Swedish survey
There is a remarkably high frequency of awkward working positions among disabled persons, and high noise levels around persons
with impaired hearing. These had decreased somewhat but were
still higher than for the average worker. These findings indicate
that persons with these disabilities tend to stay in the environment
where the disability occurred. There is no indication of a tendency
for persons with a disability to seek a work environment which is
as good as or better than average. The trend seen between 1976
and 1990/93 is rather the opposite. A possible explanation is that
persons with impairments remain in manual work.
Future trends
Persons with disabilities have a weak position on the labour market.
There is therefore a risk that they will work in environments which
have a higher than average risk. An example is persons with hearing
impairment, who work in a noisy environment more often than
others. This further decreases their ability to understand speech
in the workplace, and their hearing impairment may even increase
more rapidly due to their exposure to noise.
Technological advances have made many tasks less physically
demanding. The development of aids can increase the disabled
person’s possibilities of finding work in ordinary companies. On
the other hand, demands on flexibility, social competence and
ability to adapt to change makes it more difficult for a disabled
person to find a job.
Backenroth G. Social interaction in deaf/hearing bicultural work groups. In
Conference Proceedings from the XII World Congress of the World
Federation of the Deaf: Towards Human Rights. Austria Vienna 6–16
July, 1995, (in press).
Backenroth G. Social interaction in deaf/hearing bicultural working groups. Int
J Rehabil Res (accepted).
Construction workers
16 Professional drivers
At the census of 1990, there were approximately 110,000 professional drivers in Sweden – not including drivers of forklift trucks and
earth-moving equipment (Table 16.1). The driver usually works
alone. The jobs have high demands but little opportunity to control
the work. Many drivers have sedentary work with little chance to
take a break. Heavy lifts, poor working postures, vibrations and
air pollution make up the working conditions of many drivers.
Table 16.1. Distribution of professional drivers in 1990, men and women (percent).
Lorry and pickup drivers
Bus and taxi drivers
Railway engine drivers
Tram and underground drivers
Men (N=100,857) 67
Women (N=8,870)
Health effects
Several studies in Sweden and elsewhere have shown that professional drivers have a higher than average risk of cardiovascular
disease. A review of 19 different Nordic studies showed that bus
drivers and taxi drivers have the highest risks. The findings for
lorry drivers are equivocal.
The high risk for cardiovascular disease is probably an effect
both of lifestyle and work environment (Hedberg et al, 1993).
Several studies indicate that professional drivers smoke more than
the general population, but they are smoking less than they used
Working life and health. A Swedish survey
to. A comparison between 1981 and 1990 shows that the regular
smokers had decreased from 47 to 31 percent.
In studies of professional drivers several of them, especially bus
and taxi drivers, report that they have a poor psychosocial work
environment (Figure 16.1). Tight schedules, hectic traffic and
impatient passengers increase the strain on bus drivers, especially
in large cities. Taxi drivers usually work on contract and have little
chance to influence their incomes. The risk of violence in combination with working alone increases the feeling of poor social
support. Shift work and variable working hours may also increase
the risk of cardiovascular disease.
Figure 16.1. Demands and control of work for professional drivers (n=352) and
a control group of employed men (n=650). Shown according to the model by
R Karasek.
Drivers have sedentary work, and generally do not exercise much
in their leisure time. This, in combination with eating habits, may
be the reason why drivers tend to weigh more than the general population. A Swedish study found that professional drivers eat more
fat food and less vegetables and fruits than the average Swede. The
irregular work schedules also influence their eating habits.
It was found in a Swedish study that drivers working for moving
companies or as garbage collectors have pain in arms and hands
more often than other drivers (Hedberg et al, 1988). Tank truck
drivers more often have pain in arms, ankles and feet. They have
some heavy tasks, such as pulling the hose to the tank that is to
be filled. Drivers of light trucks in local service more often have
pain in the lower back and legs.
A Swedish study of taxi, bus and lorry drivers revealed that different groups had different risk of cancer (Jakobsson et al, 1994).
The risks also showed geographical differences. Drivers of taxis
and lorries in large cities had a higher risk of lung cancer than
the same groups in other areas. In Stockholm county the risk of
cancer is highest for drivers of delivery vans. However, that study
revealed no increased risk of lung cancer in bus drivers.
Traffic accidents are the most common type of fatal occupational
accident. In 1990 there were 4,166 on-the-job accidents involving
professional drivers. On average, 4 percent of professional drivers
have an accident each year. Although there has been a decrease since
1987, the incidence is still double the average for all sectors.
There is a Swedish study of all occupational accidents resulting in
disablement or death. It shows that professional drivers have a higher
proportion of severe or disabling injuries than other occupational
groups. One reason may be a low usage of safety belts.
The use of drugs increases the risk of accidents. In a Swedish
study it was found that 8 percent of the drivers who died in accidents during working time had alcohol in their blood (Bylund
et al, 1995).
Future trends
Professional drivers today usually have a fairly low level of edu­
cation. The demands on technical competence and language skills
are increasing, and this will affect the education and recruitment
of drivers.
New technology will change the work environment for railway
engine drivers. The job will increasingly be a matter of supervising
instruments. The stress will probably increase, as the work will become less independent and the new trains will travel much faster.
For most professional drivers the new information technology
means that their control over their work decreases. The mental
strain will therefore increase. This increased stress is especially
serious for bus and taxi drivers, who already have a high risk of
cardiovascular disease.
Working life and health. A Swedish survey
Preventive measures
Measures are being taken in Sweden to change the lifestyle of the
professional drivers: physical exercise, anti-smoking campaigns
and information on good eating habits to decrease the risk of cardiovascular diseases. Preliminary results from a study indicate that
informing the drivers about good eating habits and suggesting to the
restaurants along their routes how they can provide more healthful
food has had a good effect.
The use of safety belts was made mandatory in Sweden in 1975,
but does not apply to drivers of lorries and taxis. Approximately
every third drivers of a light truck or lorry uses a safety belt. About
10 percent of taxi drivers use safety belts. The use of safety belts
is even lower among drivers of heavy lorries. Buses have very
poor security in collisions, since there is no impact zone around
the driver. A limit on driving times and working times especially
during the night will decrease the risk for accidents due to dozing
at the wheel.
There is a good outlook for decreasing the number of accidents
among professional drivers, since the best known technology has
not yet been applied.
Jakobsson R, Gustafsson P, Lundberg I. Lung cancer among male bus, taxi and
truck drivers in Sweden. 10th International Symposium on Epidemiology
in Occupational Health ISEOH, Como, Italy, September 20–23, 1994.
Hedberg G. The period prevalence of musculoskeletal complaints among Swedish
professional drivers. Scand J Soc Med 1988;16:5–11.
Hedberg G; Jacobsson KA, Janlert U, Langendoen S. Risk indicators of ischemic
heart disease among male professional drivers in Sweden. Scand J Work
Environ Health. 1993;19:326–333.
Construction workers
17 Construction workers
In 1995 there were 240,000 workers in the construction industry
in Sweden, 4 percent of whom were women. This industry accounted for almost 10 percent of the GNP. The number of persons
employed in the construction industry has changed substantially
during the past decade. The present depression is longer and
deeper than earlier depressions.
Construction workers used to be employed on a defined project.
In the middle of the 1980s there was a general agreement specifying
permanent employment with conditional tenure. A large number
of workers have been laid off during the depression, and these were
mainly younger workers. The average age of construction workers
is therefore higher today than it was during the 1980s.
Construction work makes high mental and physical demands.
Many persons therefor leave the industry before the ordinary
retirement age. Less than ten percent of employed construction
workers continue as construction workers until 65 years of age.
In the mid-1980s the largest construction firms and local building
contractors established a special company, Galaxen, for rehabilitation of construction workers. The company was supported by the
government, which paid part of their salaries. One fourth of the
workers in Galaxen have returned to work in ordinary construction
firms. A substantial proportion have been able to continue to work
in Galaxen until they reach 65, the ordinary age for retirement.
In most countries there are three types of health risks in the
construction industry (Ringen et al, 1995): accidental falls, musculoskeletal disorders and toxic substances, notably asbestos, silica
Working life and health. A Swedish survey
and organic solvents. Hearing impairments due to noise are also
common. The risks of construction workers can be assessed by
comparing data from the Death Register, the Cancer Environment
Register, the information system for occupational diseases and
injuries and the national surveys made by Statistics Sweden. There
is also a computerised register from the occupational health centre
for construction workers (Bygghälsan).
In 1993, 24 accidents per 1,000 construction workers were reported to the ISA. This is approximately double the average for all
workers. During recent years accidents in the construction industry
have decreased much like all other industrial accidents. The most
common accidents are falling (28%), accidents from objects in
motion (28%), hand injuries (18%), and accidents due to overexertion (15%). The highest frequencies occur among the oldest
and youngest construction workers. Sick leaves for construction
workers are longer than the average, 34 days versus 27 days.
The frequency of fatal accidents for construction workers is approximately double the average for all workers. However, there are
too few fatal accidents in the construction industry to allow a more
extended analysis of single occupations or year-to-year variations.
The decline in fatal accidents in the construction industry is similar
to the decline in Swedish industry as a whole. The frequency in
Norway and Finland was double that in Sweden (Table 17.1).
Table 17.1. Average frequency of fatal accidents in the construction industry in
some countries during the 1980s. (Source: ILO, 1994.)
Country Number of deaths per 100,000 workers
The Netherlands
Great Britain
West Germany
* Recalculated from number of cases per million working hours.
Construction workers
The Netherlands has a frequency similar to Sweden, whereas in Great
Britain it is approximately the same as in Finland and Norway. The
frequencies in West Germany, France and Belgium were several
times higher than in Sweden. The differences are probably an
effect of national differences in preventive measures.
Musculoskeletal disorders
In 1994 musculoskeletal disorders accounted for 73 percent of
all reported occupational diseases among construction workers.
National surveys have shown that construction workers have musculoskeletal symptoms resulting from difficult working positions
about twice as often as the average worker. The frequency of heavy
lifting is three times higher in the construction industry, and
construction workers report double the average frequency of back
and hip problems. For painters, who frequently work with their
hands above shoulder level, the frequency of symptoms involving
the neck, shoulders, arms and hips is double that of the average
employee. Health examinations of construction workers made
be­tween 1988 and 1992 revealed a positive correlation between
high physical strains and symptoms involving the musculoskeletal
system: painters, for example, often work with their hands above
shoulder height, which increases the risk of disorders in shoulders
and neck.
Chemical substances
Chemical substances are the cause of one case of occupational
disease per 1,000 construction workers, as reported to the ISA.
In every third case asbestos was reported as the cause.
An analysis of the computerised register from the occupational
health service for construction workers shows that insulators, plumbers and sheet metal workers, who previously had high exposure
to asbestos, have an elevated risk of pleural mesothelioma. They
also have a two to three times higher incidence of lung cancer.
However, the incidence of lung cancer in wood workers who were
rarely exposed to asbestos is lower than the average for all Swedes.
Painters who began to work during the 1930s have a higher risk
of lung cancer (relative risk = 1.5), but there is no increase among
painters who began in later years.
Working life and health. A Swedish survey
The analysis of the computerised register also shows that concrete
workers have a somewhat increased risk of lip cancer, probably due
to exposure to ultraviolet radiation during outdoor work.
In 1994 one case of impaired hearing due to noise was reported to
the ISA per 1,000 construction workers. Health examinations have
shown that hearing impairments have decreased among construction workers. For concrete workers and platers, the frequency of
hearing impairment due to noise dropped by half between 1970
and 1980. This is probably an effect of preventive measures.
Psychosocial conditions
Uncertain employment conditions used to be a characteristic of the
construction industry. However, construction work has generally
been regarded as good work with a fair amount of autonomy and
good opportunities to develop competence and skills. During the
1990s building time was shortened, primarily to decrease credit
costs. The delays in the later phases of a building project, e.g.
painting, had to be reduced. Stress has therefore probably increased
considerably during recent years. However, in the national survey
made by Statistics Sweden in 1995 only 3 percent of workers in the
construction industry reported stress and mental strain, compared
to 5 percent among the total working population.
Foremen in the construction industry work under heavy mental
strain. Several foremen have had myocardial infarctions and reported being under high stress before they got ill. In a few cases they
have applied for compensation, but the decisions have varied.
Construction workers who were given medical examinations in 1971
and 1979 and followed through 1992 showed lower mortality than
the average in Sweden. The relative risk of death from cardiovascular
disease was 0.8. Deaths from causes usually associated with alcohol,
such as cirrhosis, were also low, with a relative risk of 0.6.
Future trends
Experience has shown that proper planning of a building project
is an important step in preventing accidents. Good planning for
transport of materials can also prevent accidents. Accidental falls
Construction workers
may be prevented by regular safety rounds where rails, coverage
of holes etc. are inspected
One possible future threat is a building boom with rapid recruitment
of personnel who have little experience of construction work and its
health hazards. Construction workers have to plan their own work
to a much greater extent than other industrial workers. They must
have a basic knowledge of ergonomics if they are to avoid injuries.
Extreme specialisation, e.g. that a worker only nails cornices on the
roof, may increase the risk of occupational injuries.
The occupational health service for construction workers, Bygghälsan, was the result of a special agreement between the employers
and the unions. This comprehensive occupational health centre and
the agreement behind it no longer exist.
ILO Yearbook of labour statistics 1994. ILO, Geneva, 1994.
Ringen K, Englund A, Welch L, Weeks J, Seegal J (ed). Construction, Safety and
Health. Occupational Medicine: State of the art reviews. 1995:10.
Working life and health. A Swedish survey
Construction workers
18 Working life in the future
Working life is continually changing. During the last ten years
the number of persons in the manufacturing and construction
industries has decreased by 20 to 25 percent. In the engineering
industry the proportion of middle-aged men has increased and
the proportions of women and young persons have decreased,
i.e. segregation according to age and gender has increased. As a
consequence of the reduced employment in traditional industries
the proportion of workers in the service sector, public sector and
business has increased. Unemployment has increased, and there
are more workers with non-permanent jobs.
Increasing automation in the manufacturing industry will mean
that fewer persons will be exposed to physical loads and chemicals. It
is more difficult to predict whether the exposure levels for those who
are still exposed to chemicals will increase, decrease or be unchanged. New processes may mean that new substances will be used, or
that hazardous substances will be used to a greater extent, e.g. more
persons will be exposed to isocyanates due to changed production
technology and restrictions on the use of organic solvents.
The service sector will probably grow further, and companies
will continue to seek improvements in efficiency, which often
means that mental strain increases. Negative stress will increase if
the demands on the workers rise and their control over their work
simultaneously diminishes. On the other hand, if this control is
extended the higher demands may have a positive effect.
Working life and health. A Swedish survey
The difference between salaried employees and blue-collar
workers will become smaller and more diffuse. The sharp distinction between industrial work and office environments will also
become weaker. The consequences of these changes are difficult
to predict, but they will surely influence our attitudes toward life,
learning and stress. The definition of work environment will also
become more diffuse when the difference between work and time
off becomes less clear.
A possible consequence of a high unemployment rate is an artificial decrease in labour turnover. People may stay with a job for
too long, when they should change jobs in order to preserve their
health and well-being. Increased economic insecurity may bring
more overtime and a black market where unemployed persons
are recruited for dangerous work. The present mechanism for
“signals” about problems in the work environment will not work
under such circumstances.
The widespread concern about the environment and increased
emphasis on recycling will influence the work environment primarily for those who handle garbage and recycle goods.
In the late 1980s the Swedish Work Environment Commission
observed that the difference in health between blue-collar workers
and salaried employees seemed to be increasing. There is still a
large difference in mortality between men in these two groups,
but the difference does not seem to have increased during the last
decade. The present risk of myocardial infarction varies between
occupational groups, and it is impossible to predict whether these
differences will increase or decrease.
The risk of occupational accidents varies with occupational
group. High risks occur mainly in heavy industry. The frequency
of accidents has been decreasing for several years but there is still
a great potential to reduce the risks even farther. However, lower
risks are usually more difficult to detect and sometimes more difficult to prevent.
The new information technology introduces both possibilities and
problems. Some monotonous and repetitious work can be automated, which may lead to a decrease of musculoskeletal disorders. The
new technology may also create monotonous work. Monitoring a
control panel is monotonous work and may lead to negative stress
life in
the future
e.g. for railway engine drivers. Information technology can also
be used for supervising workers, with the consequence that their
control over their own work is decreased. For some occupational
groups, e.g. bus drivers and taxi drivers, the lack of control already
means an increased risk of cardiovascular disease.
The Swedish labour market is divided between a male and female
sector. There is a trend toward equalisation, which is most rapid in
well educated groups and in occupations which are predominantly
held by men. The proportion of women in the manufacturing
industry has dropped even lower during the 1990s recession and
there is a risk for even more segregation, since demands in the
manufacturing industry will probably increase and few women have
higher technical competence.
Some groups have not got their fair share of the general improvements in the work environment. Disabled persons have on
average more physically and mentally demanding environments
than other groups, and the differences increased between 1970
and 1990. Today, the labour market requests flexibility, social
competence, adaptability and willingness to work overtime. Immi­
grants, disabled persons and elderly with low formal education
may have the less attractive jobs, which usually means the worst
work environments.
During the last decades asthma and allergies have become
increasingly common among children. When they grow up and
enter working life more workers will be sensitive to work in wet
environments, exposure to irritants etc. If they are to be free to
choose a job they want, there are high demands on the work environment. Otherwise, they will probably be discriminated against
on the labour market.
In summary, the changes in working life will increase the differences in health between different groups. Jobs with good psychosocial
conditions will tend to be more common. This means that working
conditions and health will be improved for some groups. On the
other hand there is the risk that some vulnerable groups will have
unchanged or even worse working conditions.
Working life and health. A Swedish survey
Construction workers
19 Summary
There have been large changes in Swedish working life during the
last decade. The number of persons employed in farming, forestry,
manufacturing and construction has decreased by 20 to 25 percent.
There is a trend towards a higher proportion of persons with
non-permanent employment. In Sweden today, 13 percent of the
labour force, or half a million persons, have non-permanent jobs.
Their working conditions vary. Persons with project work have
good prospects for training and control over their own work, while
workers who are employed on demand have worse than average
working conditions. Approximately 50 percent of Swedes now use
a computer in their work, and one percent work with a computer
from home for more than half their working time.
There is segregation according to gender on the Swedish labour
market; more than one woman in three works in the health care
sector, compared to about one man in twenty. The construction
and manufacturing industries employ 10 percent of the women
and 40 percent of the men.
Vulnerable groups
Women more often work part time. Their work has on average
lower status than the men’s work and requires less education and
training. Women on average also have less control of their work.
They have more repetitious work and more often have work-related musculoskeletal disorders. Elderly women working in the
Working life and health. A Swedish survey
health service and care sector are an especially vulnerable group.
Their physical capacity decreases with increasing age, while the
demands of the job include heavy lifting.
Disabled persons have more monotonous, repetitious, dirty jobs,
and work more often with vibrating tools than the average worker.
They also work more often in noisy environments.
Immigrants are not a homogeneous group and their options on
the labour market depend on where they come from and when
they arrived in Sweden. Men from non-European countries are
seven or eight times more likely to work as cleaners or in hotels and
restaurants than men born in Sweden. For women born in nonEuropean countries work in the textile and electronics industries
is four times more common than for women born in Sweden.
Chemical and physical factors
There are about 130,000 registered chemical substances in Swedish work environments. Fewer persons are exposed to organic
solvents, oils and lead than 10 years ago. However, it is unclear
whether the exposure levels have decreased for those who have
the highest exposures. Information from the Lead Register (which
was closed in 1990) indicates that the number of persons most
heavily exposed to lead did not decrease during the 1980s. If the
trend is unchanged there are about 150 persons in Sweden with
blood lead levels above 2.5 mmoles per litre.
Noise is still a major problem at many Swedish workplaces.
About 200,000 persons are exposed to hand and arm vibrations
for at least one fourth of their working time.
Changing psychological demands
If the mental demands are high and the control over the job is
low the worker is exposed to negative stress. Employees in hotels,
postal services, restaurants, health care and communications have
more negative stress than employees in other industries such as
manufacturing. The mental demands have increased in many
sectors and jobs during the past 20 to 25 years. The increase is
largest in the health care and education sectors. In the manufacturing industry the changes are small.
Construction workers
Different patterns of ill health
The risk of myocardial infarction varies with occupational group.
University graduates have the lowest risks. In some groups the
risk is only half of the average for all employed persons. In some
occupational groups in the manufacturing industry and service
sector the risk for myocardial infarction is twice the average.
Individual factors such as smoking or high blood pressure may
explain part of these differences. It is unclear which factors in the
work environment are the most important causes of the decreased
or increased risk and what occupational health measures would
most effectively prevent cardiovascular disease. However, the large
differences between occupational groups and sectors indicate that
work to prevent myocardial infarction can be very successful if
the negative and positive factors can be identified.
Work-related musculoskeletal disorders are the most common
occupational diseases and the primary cause of early retirements:
40 percent of men who were forced to retire before age 65 have
a musculoskeletal disorder. The number of work-related diseases
and injuries in the musculoskeletal system reported to the ISA has
decreased considerably during the 1990s, but about 20,000 cases
are still reported each year.
An estimated 150,000 men and 10,000 women in Sweden are
exposed to known carcinogens. The most common are benzene,
wood dust, diesel exhausts and other combustion products. Between
10 and 100 future cases of cancer per year could be prevented if all
occupational exposures to known carcinogens were stopped.
In Sweden 300 new cases of asthma are reported as work-related
each year. More than 2,000 cases of skin disorders are reported as
having occupational origin. The most common occupational cause
of hand eczema is wet work, such as hairdresser or cleaner. About
every fifth hairdresser has to change job due to hand eczema. Two
of three persons with long-term sick leaves due to a skin disorder
have reported it as an occupational disease.
The serious occupational accidents are decreasing. The number
of fatal accidents has halved during the last fifteen years. Accidents
resulting in more than thirty days of sick leave also halved between
1986 and 1993. The occupations with the highest risks are the same
Working life and health. A Swedish survey
today as they were ten years ago: firemen, construction worker,
foundry workers, miners and wood processing workers.
Mental illness
Violence and threats at work have been reported more often
during recent years. About 10 percent of women and 5 percent
of men reported in 1993 that they suffered from violence or were
threatened at work at least twice per month.
It is unclear why suicide is more common in certain occupational groups. For men, seamen, forestry workers, unskilled manual
workers and workers in health services have a higher risk of suicide.
High-risk occupations for women are physician and nurse.
New diseases and risk groups
Vulnerable groups have been receiving more attention in recent
years. It can be expected that more people will have allergies when
they start their working life. The strong horizontal and vertical
segregation between women and men is getting more attention than
it did ten years ago. High unemployment and high immigration
during the 1990s made it more difficult for immigrants to get jobs.
The high unemployment among immigrants has therefore been
addressed. The fact that immigrants seem to have worse working
conditions than other groups, however, has not been given the
same attention in discussions and evaluations.
During the 1980s the mental demands of several jobs increased.
Some groups with a high degree of control over their work, e.g.
several professionals, some construction workers and workers in
day care centres, have experienced positive changes. In health
services the psychological pressure has increased considerably. For
some this has meant an increase in negative stress, and for others
– those who also obtained more control over their work – it has
meant a more challenging and satisfying job.
Better work environments
Many work environments and working conditions are improving.
Preventive measures have yielded many positive results. Many
classical occupational diseases, e.g. silicosis, asbestosis and metal
poisoning, have virtually disappeared in Sweden. Fewer persons
are exposed to chemicals.
Construction workers
Several cancer risks have been eliminated. Today fewer persons
are exposed to known carcinogens than was the case ten years ago.
Exposure to asbestos has declined since the mid-1970s and has
now virtually ceased. Exposures to carcinogenic mineral oils and
dyes has similarly decreased. The Swedish Cancer Committee
estimated in 1984 that about two percent of all new cancer cases
are caused by factors in the work environment, with exposures
occurring during the 1960s and earlier. Two percent of cancers
in Sweden today is equivalent to about 800 cases per year. It is
estimated that the known carcinogens present in Swedish work
environments today cause between 10 and 100 cases per year.
The work done to prevent accidents at work has been very successful. No other European country have such a low incidence of
fatal occupational accidents as Sweden. The fatal accidents among
construction workers and several other groups have de­creased by
approximately 75 percent over the past 25 years.
The psychosocial work environment has improved for several
occupational groups as their control over their jobs has increased.
The knowledge that good working conditions also increase productivity and efficiency has been a major driving force behind
improvement work.
The good job
It is possible to identify circumstances and conditions that should
not exist in a good job. The good job has no exposure to hazardous substances, no monotonous and repetitious movements, and
a minimum of negative stress. It provides opportunities to learn
and develop. It optimises benefit for both the employer and the
employee. Social relations are good. All these factors are important
to a worker’s well-being and health.
Work is an important part of life, and it is affected by the worker’s
other roles in family and society. A good job should have a positive
effect on the worker’s actions in these other roles. Good social support at the workplace is a characteristic of a good job. This implies
that the unions and employers will endeavour to provide job security,
act with justice and otherwise exert a positive influence.
Good jobs include demands and challenges and the authority
and responsibility required to meet them. They offer room to
Working life and health. A Swedish survey
decide when and how tasks will be done and provide opportunities
to learn and grow. Such jobs are found in several occupations for
college graduates. These groups have the lowest mortalities and
the lowest risks of myocardial infarction. With the exception of
some jobs in health services, they have also a low risk of suicide.
These good psychosocial environments also exist in some other
jobs – carpenters, electricians, platers and agricultural workers.
This review shows that many present trends in working life are
basically positive. Fewer persons are exposed to hazardous chemicals
and fewer suffer from severe occupational accidents. The jobs of
many workers are being expanded, with greater responsibility
and authority and more opportunities to learn and develop. For
others, e.g. some groups with non-permanent employment, there
are no such positive trends.
Construction workers
List of contributors
Many researchers have contributed to this report. This list contains the main
contributors who may be contacted for further information. The list also contains contributors who wrote on subjects that were not included in the English
translation of the report.
Professor Anders Ahlbom, Department of Environmental Medi­cine, Karolinska Institutet, Box 210, S-171 77 Stockholm; fax +46 8 31 39 61
(myocardial infarction).
Gunnar Ahlborg, consultant, Department of Occupational and Environmental
Medicine, Örebro Regional Hospital, S- 701 85 Örebro;
fax +46 19 12 04 04 (teratology).
Professor Gunnar Aronsson, National Institute for Working Life, S-171 84
Solna; fax +46 8 653 17 50 (organisation, future working life, non-permanent employees).
Associate Professor Gunnel Backenroth-Ohsako, Department of Psychology,
Stockholm University, S-106 91 Stockholm; fax +46 8 15 93 42 (disabled workers).
Elisabet Broberg, Head of Division, Occupational Injury Statistics Division,
Swedish National Board of Occupational Safety and Health, S-171 84
Solna; fax +46 8 730 19 67 (work-related accidents).
Professor Mats Ekholm, Högskolan i Karlstad, S-650 09 Karlstad,
fax +46 54 83 84 61 (teachers and school children).
Göran Engholm, statistician, National Board of Health and Welfare,
S-106 30 Stockholm fax +46 8 783 32 52 (construction workers).
Anders Englund, Director, Medical and social department, Swedish National
Board of Occupational Safety and Health, S-171 84 Solna;
fax +46 8 730 19 67 (construction workers).
Professor Mats Hagberg, National Institute for Working Life, S-171 84
Solna; fax +46 8 730 19 67 (musculoskeletal disorders).
Working life and health. A Swedish survey
Associate Professor Anne Hammarström, Department of Family Medicine,
Umeå University, S-901 85 Umeå; fax +46 90 77 66 83 (mental disorders).
Associate Professor Gudrun Hedberg, National Institute for Working Life,
Box 7654, S-907 13 Umeå; fax +46 90 16 50 27 (professional drivers).
Lars-Gunnar Hörte, Head of sector, Forensic Medicine, Uppsala University,
Dag Hammarskölds väg 17, S-752 37 Uppsala; fax +46 18 55 90 53
Urban Janlert, lecturer, Department of Epidemiology and Public Health,
Umeå University, S-901 85 Umeå; fax +46 90 13 89 77 (unemployment).
Professor Bengt Järvholm, Department of Occupational and Environmental
Medicine, University Hospital, S-901 85 Umeå; fax + 46 90 10 24 56
(editor, physical and chemical factors).
Professor Åsa Kilbom, National Institute for Working Life, S-171 84 Solna;
fax +46 8 730 19 67 (elderly workers).
Hanne-lotte Kindlund, Riksförsäkringsverket, S-103 51 Stockholm;
fax +46 8 786 95 80 (rehabilitation).
Elisabeth Lagerlöf, National Institute for Working Life, S-171 84 Solna;
fax +46 8 730 19 67 (working conditions and the EU).
Associate Professor Ingvar Lundberg, Department of Occupational Health,
NVSO, Karolinska Hospital, S-171 76 Stockholm; fax +46 8 33 43 33
(occupational cancer).
Associate Professor Birgitta Meding, National Institute for Working Life,
S-171 84 Solna; fax +46 8 730 19 67 (skin disorders).
Kjell Torén, MD, Department of Occupational Medicine, Sahlgrenska University Hospital, St Sigfridsgatan 85, S-412 66 Göteborg;
fax +46 31 40 97 28 (respiratory diseases).
Associate Professor Michael Tåhlin, Department of Sociology, Stockholm
University, S-106 91 Stockholm; fax +46 8 612 55 80 (psychosocial
Professor Eskil Wadensjö, Institute for Social Research, Stockholm University, S-106 91 Stockholm; fax +46 8 15 46 70 (immigrants).
Professor Denny Vågerö, Department of Sociology, Stockholm University,
S-106 91 Stockholm; fax +46 8 612 55 80 (mortality).
Piroska Östlin, medical sociologist, National Institute of Public Health,
Box 27848, S-115 93 Stockholm; fax + 46 8 783 35 05 (work and