Safety Science 49 (2011) 1394–1403

Contents lists available at ScienceDirect

Safety Science
journal homepage: www.elsevier.com/locate/ssci

Fatal workplace accidents in outsourced operations in the manufacturing industry

Sanna Nenonen ⇑
Center for Safety Management and Engineering, Department of Industrial Management, Tampere University of Technology, P.O. Box 541, FI-33101 Tampere, Finland

a r t i c l e i n f o a b s t r a c t

Article history: Several studies have indicated that outsourcing increases the risk of accidents and presented some expla-
Received 5 July 2010 nations for this phenomenon. For example, higher accident proneness of external employees has been
Received in revised form 5 January 2011 presented, but the common causes of accidents have not been reviewed in depth. This paper provides
Accepted 15 June 2011
information about typical accidents, the contributing factors, and preventive measures of fatal occupa-
Available online 8 July 2011
tional accidents that occurred in outsourced manufacturing tasks. This paper also compares whether
these factors differ from accidents that occurred when tasks were performed in-house in the manufactur-
Keywords:
ing industry. The focus is on operations executed in the factory area for or by an organization operating in
Outsourcing
Manufacturing industry
the manufacturing business. The accident analysis is based on information gathered from accident
Fatal accidents reports for fatal workplace accidents that occurred in Finland during 1999–2008. At outsourced opera-
Shared workplace tions in manufacturing, accidents occur most commonly when installations or work preparations are
Accident statistics being performed. According to the reports, dangerous work practices and insufficient hazard identifica-
tion most frequently contributed to accidents. In order to prevent typical accidents, e.g., occupational
instructional and guidance, hazard identification, work practices, supervision, and task planning should
be improved. Statistical differences between outsourced and in-house operations were also found, mainly
within the contributing factors. Therefore, the safety of outsourced manufacturing operations should be
considered in detail in order to prevent accidents and ensure occupational safety also when operated
with other performers.
Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction
1.1. Workplace accidents in the manufacturing industry
The manufacturing industry is one of the most dangerous
branches in light of the frequency of occupational accidents. In
2007, the number of accidents causing more than 3 days’ absence
at manufacturing companies in EU-15 countries and Norway totaled
about 942,000. In the same year, 667 fatal accidents were recorded
in the manufacturing industry (Eurostat, 2009c). According to Euro-
stat (2009a), about 4.5% of the workers employed in manufacturing
businesses faced accident(s) at work in a 12-month period. These
figures rank manufacturing second in the accident statistics right
after construction (see Eurostat, 2009a).
In Finland during 2007, about 30,000 accidents occurred at
manufacturing workplaces, of which slightly more than 14,000
caused a minimum of 4 days’ absence. Six of the accidents were
fatal. Thus, about every fourth workplace accident involved
employees performing manufacturing operations. During the same
year, in Finnish manufacturing workplaces, the accident frequency
of accidents causing at least 4 days’ absence (workplace accidents
⇑ Tel.: +358 40 849 0356; fax: +358 3 3115 2671.
E-mail address: sanna.nenonen@tut.fi
per million work hours) was about 20, and about 43 if calculated
from all reported accidents. According to the accident frequency,
the manufacturing industry was the third most risky branch after
construction and private households with employed persons. The
number of accidents in the manufacturing industry was the same
as the previous year, but the frequency decreased a little as the
number of work hours increased (FAII, 2009a).
The most common working processes at the time of accidents at
Finnish manufacturing workplaces in 2007 were production,
manufacturing, processing (56% of the accidents). The most com-
mon modes of injuries were contact with sharp, pointed, rough,
coarse material agents (23% of the accidents) and the most common
injuries resulting from the accidents were wounds and superficial
injuries (47% of the accidents) (FAII, 2009a).
1.2. Outsourcing in the manufacturing industry
Increased competition, globalization, and a rapidly changing
operational environment have turned outsourcing into an ordinary
part of companies’ operations and management. Especially manu-
facturing companies favor outsourcing to an increasing extent (see
Eurostat, 2009b). According to the outsourcing trends, Finnish com-
panies are some of the most eager outsourcers in Europe (Ekström,
2007; Eurostat, 2009b; Rikama, 2008). Of the Finnish companies

0925-7535/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ssci.2011.06.004
 S. Nenonen / Safety Science 49 (2011) 1394–1403
operating in the manufacturing industry and employing more than
10 employees, almost two thirds outsourced some of their opera-
tions during 2000–2006 (Ali-Yrkkö, 2007). In Finnish industry,
outsourcing has been most common among companies operating
in the manufacture of metals, the chemical industry, and the forest
industry. In these lines of business, more than 80% of the companies
have outsourced some of their operations. On the contrary, food
industry companies have used external providers the most seldom
(fewer than every second one of the companies). Almost every stud-
ied Finnish manufacturing company that has subcontracted work to
other performers has operated with domestic companies, but every
sixth company has also contracted operations abroad (Ali-Yrkkö,
2007). Larger Finnish manufacturing companies (at least 100
employees) have contracted internationally to an even larger extent
(Eurostat, 2009b). The share of outsourced operations has been pre-
dicted to increase at the end of the first decade of 2000 (Ekström,
2007).
Finnish companies commonly outsource services supporting
primary business, but the execution of core services is also sifted
to service providers (see Ali-Yrkkö, 2007; Eurostat, 2009b). Accord-
ing to the study conducted by Ali-Yrkkö (2007), Finnish manufac-
turing companies with more than 10 employees have most
commonly outsourced service operations (e.g., financial manage-
ment, property maintenance, machine maintenance and guarding).
About every second company operating in the manufacturing
industry has outsourced current operations since the beginning
of the 2000s. Every third manufacturing company has outsourced
manufacturing operations and slightly lesser R&D operations.
Service operations have been acquired almost solely from Finnish
providers, but about a tenth of R&D operations and manufacturing
is outsourced outside Finland (Ali-Yrkkö, 2007). Compared with
other European countries, the international outsourcing statistics
of Finnish large companies are somehow average from the view-
point of operation types and the frequency of their externalization
(Eurostat, 2009b).
According to Nordic outsourcing research, the most common
reasons for outsourcing among Nordic companies are focusing on
core operations and obtaining special knowhow. In addition, fluctu-
ation in personnel demand, reduction of costs, and a means to
change the fixed costs to variable ones were mentioned as common
reasons (Ekström, 2007). The main motives for outsourcing abroad
are, in addition to cost efficiency and special skills, improvement of
competitiveness and access to new markets (Rikama, 2008) as well
as groups’ strategic decision and competitors/clients example
(Eurostat, 2009b). All these factors correspond with those reported
in several international studies to explain why companies out-
source their operations (see Hätönen and Eriksson, 2009).
1.3. Research on effects of outsourcing on workplace accidents
In several circumstances, outsourcing has been indicated to
increase the possibility of accidents and even the number of work-
place injuries (e.g., Blank et al., 1995; Kochan et al., 1994; Mayhew
et al., 1997; Quinlan, 1999; Rousseau and Libuser, 1997; Salminen,
1995). However, the magnitude of the influence is difficult to assess
due to limitations in accident statistic data (Hämäläinen, 2010).
Nevertheless, studies indicate that the hazardousness of multi-
employer worksites is derived not just from one operator’s actions
but also from the other party’s performance. Although most victims
of shared workplace accidents worked for an external employer,
accidents were usually attributable to actions by both the customers
and the providers (Rantanen et al., 2007a). Several reasons for the
subcontracted employees’ higher accident rate have been reported.
Higher accident rates have been explained to be a consequence of,
e.g., subcontractors’ unfamiliarity with the working conditions and
practices at the worksite (Clarke, 2003), external employees’ short
1395
duration of visits at the work site (Salminen, 1995), tendency of sub-
contractors to perform high-risk tasks (Blank et al., 1995; Kochan et
al., 1994), deficiencies in external performers’ safety training and
experience (Kochan et al., 1994), insufficiencies in subcontractors’
hazard identification (Salminen, 1995) and safety awareness (Hon
et al., 2010), blurred responsibilities (Clarke, 2003), economic pres-
sures (Quinlan and Mayhew, 2000) and poor disorganization of work
processes (Quinlan and Mayhew, 2000).
Even though the effect of outsourcing on accident rates has
been noticed, comprehensive and detailed research relating to
the accidents has been extremely concise. Blank et al. (1995) has
studied the nature of injuries and type of accidents that occurred
between 1986 and 1990 in the mining industry in Sweden. Accord-
ing to that study, contractor employees seem to suffer from more
frequent and more severe injuries than mining companies’ own
employees. Similar results were reported by Salminen (1995),
who compared accidents that involved the main contractors and
subcontractors operating in Finnish construction during 1988
and 1989. The most recent study related to the subject has studied
how the fact that workplace is shared has influenced in general and
in construction occupational accidents occurred between 1999 and
2004 in Finland (see Rantanen et al., 2007b). More research on the
subject, especially on what kind of factors contribute to accidents,
is needed in order to promote occupational safety and health at
shared manufacturing workplaces.
1.4. Objectives of the article
This paper presents the results of a study conducted regarding
fatal occupational accidents in outsourced services in the Finnish
manufacturing industry. The aims of the paper are to
(1) study during which work tasks as well as what kind of fatal
accidents occur when employees perform outsourced
operations,
(2) determine the factors that contribute to the fatal workplace
accidents in outsourced operations,
(3) review what kind of factors are recommended as preventive
measures in the case of fatal accidents that occur in outsour-
ced operations, and
(4) find out if contributing and corrective factors differ from
those related to accidents that occurred within in-house
operations.
The focus in this study is on operations carried out for a customer
operating in the manufacturing industry. The operations reviewed
include maintenance and repair, property maintenance, installa-
tions, cleaning, and loading/unloading in a factory area.
2. Materials and methods
2.1. Accident reports
The data used in this study has been gathered from investigation
reports of fatal workplace accidents that occurred in Finland. The
Federation of Accident Insurance Institutions (FAII) and labor orga-
nizations coordinate the compilation of the reports that are drawn
up after fatal accidents occur in Finnish workplaces or in corre-
sponding conditions involving employees or entrepreneurs insured
according to the Employment Accidents Insurance Act. Reports are
not compiled in cases where an accident has occurred in external or
commuting traffic or the death was caused by an occupational dis-
ease. Reports are used to inform the workplaces where the fatalities
occurred. Reports are also publicly available in Finnish via FAII’s
web pages (FAII, 2009b).
1396 S. Nenonen / Safety Science 49 (2011) 1394–1403
The accident reports have been compiled based on investiga-
tions conducted by a group of experts. The group consists of repre-
sentatives from the FAII and from labor, officials’, and employers’
organizations in the line of the business in question. In addition,
participants from research institutes and insurance companies
may be used. The aim of the investigations is to determine what
happened, why an accident occurred, and how similar accidents
could be prevented. The people responsible for the accident are
not investigated (FAII, 2009b).
The accident reports are about five pages long verbal descrip-
tions but they also include a chain-of-events graph based on the
Finnish accident investigation model (see Saari et al., 2001) and pic-
tures and drawings clarifying the accident. In addition, the reports
include background information on the injured party, employer and
accident scene. The reports also contain the accident classification
information based on the ESAW-classification. The classifications
indicate the victim’s age, workstation, occupation, working process,
physical activity, deviation, mode of injury, branch of the injured
party’s employer and company size.
The investigation system has been in place since 1985 in the
case of employees and since 1998 in the case of entrepreneurs.
By the end of July 2008, 811 accidents had been investigated, in
which 786 employees and 39 entrepreneurs died (FAII, 2009b).
2.2. Data in focus and analyzing methods
The focus in this study is on fatal occupational accidents involv-
ing employees during the 10-year time period 1999–2008. Acci-
dent reports used to gather the data have been obtained from
the FAII’s accident database. The accident reports were selected
as a data source because unlike other accident statistics they in-
clude reliable information on whether the victim was employed
by the customer or the service provider company and whether
the accident occurred at the customer’s site or somewhere else.
Furthermore, the accident reports include corrective actions rec-
ommended by the accident investigation group aimed at prevent-
ing similar accidents.
The data presented in this paper were compiled as a part of a
research project at Tampere University of Technology by reviewing
all the accident reports from the period under review. Accidents,
that occurred when employees performed industrial operations
for the manufacturing industry at the industrial site, were selected.
From the selected reports, the following information was gathered:
– type of the operation (in-house or outsourced),
– operation-related factors (working process and physical
activity),
– accident factors (deviation and mode of injury),
– factors contributing to the accidents,
– recommended corrective actions,
– background information (e.g., service providers’ branches, com-
pany size and age, profession and experience of the victim).
Information on the type of operation, factors contributing to the
accidents, recommended corrective actions and experience of the
victim were obtained from the accident descriptions. Data on the
other variables were available as ESAW-codes. However, some of
the accident reports did not contain information on all of these
factors.
During the period under review, 274 fatal workplace accidents
occurred involving employees at Finnish workplaces. Summaries
were available for all of these (representing information about
operation type, operation-related factors, accident factors, and
some background information) and full-length reports for 262 acci-
dents. Of these accidents, 83 (30%) corresponded to the target
group: employees who died while performing manufacturing
industry operations at an industrial site. The production area,
factory, workshop; maintenance area, repair workshop; area used
principally for storage, loading, unloading, and other similar types
of work environments are considered industrial sites (ESAW codes
010–019). Of these relevant cases, 34 (41%) occurred when
employees were performing outsourced operations. The cases
selected were from the following lines of businesses: manufactur-
ing industry (SIC code D), transport, storage and communication
(I), construction (F), real estate, renting and business activities
(K), electricity, gas and water supply (E), wholesale and retail trade
(G), and other community, social, and personal service activities
(O). Detailed information on the cases selected is presented in
Table 1.
The data was analyzed using the statistical program SPSS 15.0.
Descriptive statistics were calculated to summarize the basic fea-
tures of the data. To identify dependencies between variables,
the data was subjected to statistical tests (Fisher’s exact test and
Mann–Whitney U-test).
3. Results
3.1. Background information
During 1999–2008, 34 fatal workplace accidents occurred at
outsourced operations and 49 at in-house operations in the Finnish
manufacturing industry. The annual number of accidents varied
between 1 and 6 in the case of outsourced operations and between
3 and 10 for in-house operations. One accident that occurred at
outsourced operations claimed three victims, but other cases
involved one victim. The share of the fatal workplace accidents that
occurred at outsourced operations compared with all fatal
accidents in Finnish manufacturing is illustrated in Fig. 1.
Victims performing outsourced operations were on average
43.2 years old (18–69 years). The age did not differ statistically sig-
nificantly between outsourced operations even though the average
age of victims in the case of in-house operations was slightly high-
er, 47.2 years (23–62 years). The majority (about one third) of the
victims belonged to the age group of 45–54 years in both operation
types. Compared with in-house operations in the group of outsour-
ced tasks, the share of victims was greater among the under
25-year-olds and smaller among the ages 55–64.
The most common occupations of the victims in the case of
outsourced operations were truck driver (21% of the victims),
machine driver (15%), electrician (12%), and machine repairer (9%).
Other occupations were cleaner, plate maker, machine operator,
mechanic, forklift driver, welder, insulation worker, carpenter, engi-
neering worker, and engineer. Victims were in general experienced
in the work tasks they were performing at the time of the accident.
Of the external employees, 73% had at least 5 years’ experience or
were described as experienced in the accident reports. Less than
every tenth external victim had at most only 1 year of experience.
Victims performing in-house operations were competent as well,
because almost 90% were considered experienced and only 3% had
only a little experience. The differences between outsourced and
in-house operations were not statistically significant.
Half of the provider companies for whom the victims worked
employed at least 50 employees. However, the proportion of smal-
ler companies employing nine employees at most was also remark-
able (40%). In-house operations were commonly performed for
large companies, because three of four companies were large (at
least 50 employees), and only one tenth employed at most nine
employees. The differences in the size of the victim’s employer
company between outsourced and in-house operations were
statistically significant (p = 0.003). The victims of accidents that
occurred when employees were performing outsourced operations
 S. Nenonen / Safety Science 49 (2011) 1394–1403 1397

Table 1
Information related to the cases selected for the study.

Economic activity of the employer (SIC No. of cases included (% of fatal accidents Operations of the cases included
code) within the business)
Manufacturing (D) 67 (88%) Manufacturing, cargo handling, maintenance and repair, adjustment,
installation, inspection and transfer
Transport, storage and communication (I) 9 (20%) Loading and unloading
Construction (F) 2 (2%) Cleaning and electrical installation
Real estate, renting, and business 2 (2%) Cleaning machine and repair
activities (K)
Electricity, gas, and water supply (E) 1 (20%) Maintenance
Wholesale and retail trade (G) 1 (8%) Maintenance
Other community, social, and personal 1 (13%) Cleaning
service activities (O)

12 70 % to the accidents that occurred at outsourced manufacturing opera-

tions. In almost every third report, this factor was recorded as a

Proportion of accidents
Number of accidents

10 60 %
deviation. Other deviations that frequently took place during a per-
50 %
8 formance of outsourced tasks were slipping, stumbling, falling, or
40 % falling of persons (24% of cases) and shock, fright, violence, aggres-
6
30 % sion, threat, or presence (18%). In the case of in-house operations,
4 the most common factor was different, the loss of control of a ma-
20 %
chine, handheld tool, object, or animal. This was marked as a devi-
2 10 % ation in every fourth report discussing accidents that occurred
0 0% when employees performed in-house manufacturing operations.
1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 However, the difference was not statistically significant. Detailed
Number, Outsourced operations Number, In-house operations
information on the deviation data is presented in Appendix.
Proportion, outsourced Linear (Proportion, outsourced)

Fig. 1. Fatal workplace accidents at manufacturing operations in Finland during 3.2.4. Mode of injury
1999–2008. Almost every second accident report that discussed accidents
that occurred at outsourced manufacturing operations mentioned
injuries being caused due to trapping or crushing. In addition, hor-
were mainly from service providers (81% of the cases). One acci-
izontal or vertical impact with or against a stationary object; struck
dent claimed victims from both types. Most of the accidents were
by or collision with object in motion and contact with electrical
caused by the victim’s own actions. Another person contributed to
voltage, temperature, or hazardous substances were mentioned
the accident only in 6% of accidents within outsourced tasks and in
in the context of several accidents within outsourced operations.
2% in the case of in-house operations. The difference was not sta-
The different modes of injuries were almost as common in the case
tistically significant.
of in-house as outsourced operations, so statistical significances
were not found. Comprehensive information on the percentages
3.2. Working process, specific physical activity, deviation, and mode of of modes of injuries in outsourced and in-house operations is pre-
injury sented in Appendix.

3.2.1. Working process 3.3. Contributing factors

The most common working processes executed at the time of
accidents that occurred at outsourced manufacturing operations The most common factors contributing to fatal workplace acci-
were installations and preparations (27% of the cases), mainte- dents in outsourced operations in manufacturing related to dan-
nance and repairs (18%), and cleaning tasks (15%). The working gerous work practices (e.g., working in a danger zone or utilizing
processes performed differed notably between outsourced and work methods involving a considerable possibility of errors), insuf-
in-house operations even though statistical significances were ficient hazard identification (e.g., default on risk assessment or
found only in the cases of production (p = 0.044) and monitoring underrating the significance of identified hazards), human error
(p = 0.019) that were more common in-house tasks. Detailed infor- (e.g., miscalculation in the workings of machinery or defective
mation on the percentages and examples of the working processes assembly) and deficiencies in instruction and guidance (e.g., intro-
performed is presented in Appendix. duction to work tasks or worksite neglected or carried out defec-
tively). The first two factors were mentioned as contributors
3.2.2. Specific physical activity more than in every second accident report discussing accidents
The majority of the victims among the outsourced operations that occurred when employees performed outsourced operations.
were moving at the site while the accident occurred (41% of the The latter two had an influence on every third shared workplace
cases). Other common activities were working with hand-held tools accident. Other common factors contributing to the accidents were
(16%) and handling objects (16%). Some slight differences in per- inadequate supervision (e.g., supervision at the worksite was
centages between outsourced and in-house operations can be ignored or supervision of compliance with instructions was inade-
found, but those were not statistically significant. The particular quate), defective communication (e.g., risks of work tasks, changes
percentages related to victims’ activity are displayed in Appendix. in work performance, simultaneous work tasks or employees on
site were not known to all parties involved), insufficient safety de-
3.2.3. Deviation vices (e.g., safety devices were not provided or they were removed
Breakage, bursting, splitting, fall, or collapse of the material or switched off), and inappropriate warning signs (e.g., warning
agent was most commonly mentioned as a deviation contributing signs on machinery were defective or ambiguous or danger zones
1398 S. Nenonen / Safety Science 49 (2011) 1394–1403
were not marked). These factors were listed as accident contribu-
tors in every fourth outsource-related accident report. Several
other factors were also mentioned in the accident reports studied.
When the contributing factors related to accidents that oc-
curred in outsourced operations were compared to those that took
place in-house, several statistically significant differences were
found. Inappropriate warning signs (p = 0.003) and inadequate
flow of information (p = 0.012) were mentioned more commonly
in the accidents reports discussing outsourced operations than in
reports related to in-house operations. On the other hand, machine
malfunction was considered contributing to accidents more rarely
in the outsourced cases than in the in-house ones (p = 0.018). In
addition, differences that can be considered as indicative emerged
within insufficient hazard identification, instruction, and guidance
and safety devices. From these, safety device insufficiencies were
more common among the accidents that occurred in in-house
operations. Two other factors were mentioned more commonly
in the reports that considered outsourced operations. The list of
the contributing factors mentioned at least in every tenth report
discussing outsourced operations, their frequency, and the differ-
ences from in-house operations are illustrated in Fig. 2.
3.4. Recommended corrective actions
The reports discussing accidents that occurred during the exe-
cution of outsourced operations in the manufacturing industry rec-
ommended several actions to prevent similar accidents. Most
commonly, reports brought out the importance of proper occupa-
tional instruction and guidance (e.g., introduction to safe work
methods, performance in abnormal situations and proper use of
tools and safety devices given to all employees regardless of the
employer) as well as sufficient hazard identification (e.g., system-
atic identification of hazards and execution of corrective actions re-
lated to established work practices, abnormal situations and
different work phases). Occupational instruction and guidance
were mentioned as a recommended corrective action on more than
80% of the reports related to outsourced tasks. In the case of hazard
identification, the proportion was only 10% smaller. The next com-
mon recommended preventive factors were the utilization of safe
work practices (e.g., execution of work phases in correct order, ver-
ifying that the installation is dead and avoidance of danger zones),
Dangerous work practice
Insufficient hazard identification
Human error
Deficiencies in instruction and guidance
Inadequate supervision
Inadequate flow of information
Insufficient safety devices
Inappropriate warning signs
Ignorance of rules and instructions
Insufficient task planning
Machine malfunction
Working while machine is running
Inappropriate education or experience
Inappropriate working plane or passage
Deficiencies in personal protective equipment
0%
Significance: *** p < 0.010, ** 0.010 ≤ p <0.050, * 0.050 ≤ p < 0.100
Fig. 2. The most common contributing factors related to fatal workplace accidents in the Finnish manufacturing industry presented in the accident reports during 1999–
2008.
sufficient supervision (e.g., supervision of compliance with safe
work practices, interference with faults, ensuring possibility to
perform tasks safely), task planning (e.g., planning of proper execu-
tion order of work phases, working in fault situations, and execu-
tion of simultaneous, short-term and short notice work tasks),
consideration of safety issues in machine design (e.g., by taking
into account safety aspects, different work tasks, and operation
conditions), compliance with rules and instructions (e.g., to follow
work orders, utilize given safety devices, and operate according to
safety regulations), sufficient safety devices (e.g., flash barriers,
backup beepers, and safety rails that are installed and utilized
appropriately), appropriate warning signs (e.g., excavations’ seal-
ing-off line markings, valves’ setting markings, and switchboards’
voltage signs), and written work instructions (e.g., for work tasks
performed rarely, requiring notable recalling and performed by
inexperienced employee). These factors were mentioned at least
in every third report but even in every two of the three reports
relating to outsourced tasks. In addition, several other actions were
recommended as preventive measures among outsourced opera-
tions in the Finnish manufacturing industry. Factors that are men-
tioned at least in every tenth outsourced-related report are
presented in Fig. 3.
The recommended corrective actions were more or less the
same in the reports discussing outsourced and in-house opera-
tions. Differences were found merely in the case of two factors,
and additionally, these differences can be considered only indica-
tive. Indicative differences emerged in the case of occupational
instruction and guidance and task planning. Both measures were
mentioned more frequently in the reports reviewing accidents that
occurred at outsourced operations than in the reports of in-house
operations.
4. Discussion
4.1. Evaluation of the data
The data used in this study consists of 83 accident cases; about
40% of these cases occurred at outsourced operations. The size of
the data is only moderate, but even so, it covers all fatal workplace
accidents that occurred during the period under review involving
Finnish employees working in the manufacturing industry.
*
*
**
*
***
*
**
Outsourced
operations (n=33)
In-house operations
(n=48)
10 % 20 % 30 % 40 % 50 % 60 %
 S. Nenonen / Safety Science 49 (2011) 1394–1403
Occupational instruction and guidance
Hazard identification
Safe work practices
Supervision
Task planning
Machine design
Compliance with rules and instructions
Safety devices
Appropriate warning signs
Written work instructions
Communication strengthening
Determination of responsibilities
Sufficient education or experience
Machinery inspections and maintenance
Appropriate working plane and passage
Development of working environment
Adequacy of personal protective equipment
Appropriate tools and machines
0 % 10 % 20 % 30 % 40 % 50 % 60 % 70 % 80 % 90 %
Significance: *** p < 0.010, ** 0.010 ≤ p <0.050, * 0.050 ≤ p < 0.100
Fig. 3. The most commonly recommended corrective actions for preventing fatal occupational accidents in the manufacturing industry presented in the accident reports
during 1999–2008.
However, due to the small number of data, all possible differences
in variables between accidents that occurred at outsourced and at
in-house operations may not have come up with the statistical
tests used. By reviewing the presented results, it is still possible
to get a clear picture of the nature of the fatal accidents. Another
noteworthy issue is that the accident reports used as a source of
the data are mainly qualitative descriptions of the accident cases
that occurred in the Finnish manufacturing industry. The opera-
tion-related factors and accident factors, which are working pro-
cess, physical activity, deviation and mode of injury, as well as
the victim’s profession and a branch and the size of the victim’s
employer company, are presented in the reports according to the
ESAW codes. All other information included in the data has been
gathered from the description and categorized by the researchers
performing the study. In order to ensure appropriate data collec-
tion, only those factors mentioned unambiguously in the descrip-
tions are endorsed to the data. No further interpretations are
made in order to avoid misunderstandings and to ensure uniform
processing of the accident reports. Third, the accident reports
compiled about fatal workplace accidents are made based on an
accident investigation carried out by a case-specific group of
experts. The differences in the investigator groups and the investi-
gation processes may cause variation in the aspects presented in
the accident reports. From the scientific point of view, variation
in the reports makes the gathering of comprehensive and congru-
ent information from all the accident reports at least burdensome
and complicated but in some cases even impossible. However,
relatively quality data was obtained with an extensively careful
dissection of the reports.
The focus in this paper has been only on fatal workplace acci-
dents, because such data about non-fatal occupational accidents
that occurred in Finland that includes information about the
employer status is not available. Thus, the most comprehensive
occupational accident database does not include information that
can be used to define whether the work task performed at the time
of the accident was outsourced or not. However, it could be
1399
*
*
Outsourced
operations (n=33)
In-house operations
(n=48)
assumed that the results obtained by studying fatal accidents can
be also widened to the non-fatal accident context. In light of a
study discussing fatal and severe non-fatal accidents in mainte-
nance, the accident types and accident sources of severe and fatal
accidents are quite similar. The clearest difference between these
two accident categories is that the machines and devices are
emphasized as an accident source in the case of severe non-fatal
accidents. Otherwise, the accident information is relatively uni-
form regardless of the severity of the accident (see Lind, 2008).
By taking into account the facts that maintenance tasks are one
of the most commonly outsourced industrial operations and that
the types and sources of maintenance accidents presented in Lind’s
(2008) paper correspond with those obtained in this study, it is
likely that the results of this article can be applied to the whole
range of manufacturing operations and prevention of accidents in
varying severity. In addition to the focus on fatal accidents, this pa-
per discusses only those accidents that occurred in the Finnish
manufacturing industry. However, the results can be applied to
other industrialized countries due to similarities in outsourcing
characteristics (see Eurostat, 2009b; Hätönen and Eriksson, 2009)
and resulting safety management problems (see Nenonen, 2010).
When compared with European statistics, the accident factors re-
ported in this paper describe typical accidents that occur in the
manufacturing industry rather well (see European Commission,
2009), although some variation exists between different countries.
Due to similar safety management problems encountered in multi-
employer worksites in different countries, the contributing factors
to accidents and the recommended corrective actions in particular
reported in this paper are relevant to the safety of every multi-em-
ployer worksite in the manufacturing industry.
4.2. Review of the results
4.2.1. Accident trend
Compared with all fatal accidents in the manufacturing industry,
the proportion of accidents that occurred at outsourced operations
1400 S. Nenonen / Safety Science 49 (2011) 1394–1403
seems to have risen during the period of review. However, the num-
ber of accidents in outsourced operations remained at about the
same level during the same time span. Thus, the number of acci-
dents that occurred when employees performed in-house an oper-
ation has notably decreased. The most probable reason for the
upward trend of the accident proportion within outsourced opera-
tions is the expansion of outsourcing during the past decade. Unfor-
tunately, no specific data about the extent of the use of external
service providers was available. Thus, within this study, it cannot
be found whether the increase of the accident proportion among
outsourced manufacturing operations and the steady level in acci-
dent numbers differing from the development within the in-house
tasks can be explained totally by the increased outsourcing num-
bers or whether the growth originates partially from the develop-
ment of poorer safety that in the group of in-house operations.
4.2.2. Working processes
Even though the differences in working processes between out-
sourced and in-house operations were statistically significant only
in the case of production and monitoring, the differences in the
proportion of setting up, maintenance, and cleaning were also
remarkable. Production and monitoring were more common in-
house processes and the latter more frequent among outsourced
operations. This reflects the companies’ current tendency to focus
on their core operations and outsource the operations supporting
core functions. The differences in the executed working processes
between outsourced and in-house operations may have some kind
of influence on other variables reviewed. For example, it is clear
that some of the differences between outsourced and in-house
operations in the proportions of the specific physical activity, devi-
ation, and mode of injury can be explained by the different work-
ing processes. Thus, it is also probable that working processes may
have an effect on the factors contributing the accidents. However,
most of the contributing factors relate to the organizational and
human factors that are not very task specific. Therefore, it could
be assumed that the working process’s influence is not that signif-
icant from the perspective of the utilization of the results. How-
ever, the assumptions and the magnitude of the possible
influence cannot be checked due to the limited number of data.
4.2.3. Contributing factors
In the case of accidents that occurred at outsourced operations,
inappropriate warning signs, insufficient hazard identification,
deficiencies in instruction and guidance, inadequate flow of infor-
mation, and insufficient task planning were emphasized as con-
tributing factors more commonly than at in-house operations. An
explanation for this is that ensuring the adequacy of these issues
is considered difficult when working at work sites operated by
other performers (Nenonen, 2009). The complexity of managing
the factors in question may relate to the problems in the service
providers’ own safety management but as well as to unworkable
arrangements and cooperation between partners. According to
the literature, e.g., weaknesses in external companies’ hazard
awareness may spring from hazard identifications external compa-
nies made unsystematically (see Trethewy et al., 2003) or equally
from host companies’ poor communications of hazard identifica-
tion results (see Mynttinen, 2006). From the viewpoint of instruc-
tion and guidance, insufficiencies may result because host
companies have not considered that external employees are not
as aware of and experienced with common practices and proce-
dures as the host’s own employees (see Clarke, 2003). On the other
hand, the flow of information may be inadequate, e.g., due to un-
clear procedures causing breaks in the information flow or work-
ers’ obliviousness about what information should be shared with
partners (see Kuitunen et al., 1999). Furthermore, problems in task
planning may arise because the task should be executed at short
notice or at a work site that is not familiar to the performer (Lind
and Nenonen, 2008). In addition, providers’ poor commitment in
managing safety at customers’ site (see Holmes et al., 1999), lim-
ited resources available for safety management (see Lin and Mills,
2001), and long supply chains with confusing relations (see Loo-
semoore et al., 2003) may cause insufficiencies in the factors con-
tributing to the accidents.
The contributing factors of the accidents that took place when
employees performed in-house operations differing from those
that occurred in outsourced operations were insufficient safety de-
vices and machine malfunction. These two factors seem to relate
closely to each other. In-house operations include more machine
failure removals made by machinists than outsourced tasks. Re-
moval of machine malfunction, especially if the failures are recur-
rent, may lead to bypassing or out-taking of safety devices in order
to quicken the execution of the tasks.
4.2.4. Recommended corrective actions
The only significant differences in the recommended corrective
actions between outsourced and in-house operations were found
in occupational instruction and guidance and task planning. This
may partially be a consequence of the way the accident reports
are compiled. Many reports recommend along with the case-spe-
cific corrective actions more universally applicable measures.
These universal actions may not relate just directly to the identi-
fied primary contributing factors but are measures that affect more
generally safety performance and prevention of accidents. Because
of this reporting style, several corrective actions are mentioned in
every report, and thus, some of the recommendations presented in
the reports may be somewhat similar.
4.2.5. Exploiting the results of this study
According to the results presented in this paper, even though
most of the contributing factors between outsourced and in-house
manufacturing operations are similar, several differences in the
frequency of the contributing factors exist. From these results
can be drawn a conclusion that mainly similar measures to manage
safety in outsourced operations can be utilized than in in-house
ones. However, when employees operate at another company’s
site, the focus in safety management and accident prevention
should be directed at the contributing factors most common to
the outsourced operations. Also noteworthy is the transition from
the execution of manufacturing operations in-house to acquiring
these operations from external companies. This also shifts the haz-
ards and possible accidents to the service-providing companies.
However, the responsibility of managing the safety cannot be off-
loaded merely to providers but should be done with the coopera-
tion of all performers operating on the site in order to ensure the
safety of the employees of the service-providing companies and
other employees operating at the shared workplace.
5. Conclusions
This article discussed fatal workplace accidents that occurred in
the Finnish manufacturing industry during 1999–2008. The pur-
pose of this review was to identify what kinds of accidents occur
when employees performed outsourced manufacturing operations
and study the factors contributing to these accidents. In addition,
the purpose was to compare accidents that occurred at outsourced
operations with those that took place at in-house operations. The
study was carried out by analyzing accident reports made based
on accident investigations executed by a group of experts.
According to the analysis of the accident reports of the Finnish
manufacturing workplaces, every year several accidents claim the
lives of the employees working for the service providers or the
 S. Nenonen / Safety Science 49 (2011) 1394–1403 1401

main companies. About 40% of these accidents occur when
employees perform outsourced operations. The largest share of
the fatal workplace accidents involves middle-aged employees
who are experienced in the tasks they are performing at the time
of the accident.
The most common working processes in the case of fatal acci-
dents that occurred at outsourced operations were installations
and preparations as well as maintenance and repairs. The most
common specific physical activity, deviation and mode of injury
were movement, breakage, and impact against a stationary ob-
ject, respectively. When outsourced operations were compared
with in-house tasks, some differences in the shares among the
studied variables were found. However, these differences were
statistically significant only in the case of a couple working
processes.
The most common factors and deficiencies that were mentioned
in the accident reports as contributing to the accidents that oc-
curred at outsourced operations were dangerous work practice,
insufficient hazard identification, human error, and deficiencies
in instruction and guidance. These contributing factors differed be-
tween outsourced and in-house operations in several cases. How-
ever, the recommended corrective actions in the reports were
very similar.
The analysis discussed in this paper has yielded previously
quite shortly reviewed information related to the factors influenc-
ing the fatal accidents at outsourced operations in the manufactur-
ing industry. Results of the analysis presented in the paper can be
used among others in risk assessment, accident prevention, and
orientation of safety measures at shared workplaces in the manu-
facturing industry. The results gathered during the analysis
presented in this paper are also examined in a research project
executed at the Department of Industrial Management at the
Tampere University of Technology. During the project, an
operational model of safety management for service providers
operating in the manufacturing industry was developed. With
the support of the practical information and tools of the model,
industrial providers can enhance safety when operating with cus-
tomer companies.
Acknowledgements
The author acknowledges the financial support received from
the Finnish Work Environment Fund, the Finnish Maintenance
Society Promaint and the group of Finnish industrial companies
for the execution of the research project ‘Safety Management of
Industrial Services.’ In addition, the writer would like to sincerely
thank Mr. Jouni Kivistö-Rahnasto, Prof., Ms. Noora Hintikka, M.Sc.
(Eng.), and Marileena Koskela, M.Sc. (Eng., Econ.), for valuable com-
ments on previous drafts of the article.

Appendix A

Accident-related variable proportions of fatal workplace accidents at outsourced and in-house operations in the manufacturing industry
in Finland during 1999–2008.

Variable (ESAW-code) Examples of variables presented in the accident Outsourced In-house Sig.
reports operations operations
(%) (%)
Working process (n = 33) (n = 49)
Production, manufacturing, processing (11) Working in a production line, manufacturing of 3 18 –**
products, and removal of a process malfunction
Storing (12) Load of a cargo and transfer of supplies at a store 6 6
Remodeling, repairing, extending, building Repair of a valve 3 0
maintenance (24)
Setting up, preparation, installation, mounting, Unloading of a cargo, installation of lights, and 27 20
disassembling, dismantling (51) assembly of scaffolding
Maintenance, repair, tuning, adjustment (52) Repair of a loader, fault correction of process 18 6
equipment, and survey of electrical connections
Cleaning working areas, machines (53) Cleaning of a production area, a pipeline, and a 15 4
conveyor
Waste management, disposal, waste treatment Discharging of a silo, loading of a garbage truck, and 6 2
of all kinds (54) transferring waste to a waste container
Monitoring, inspection of manufacturing Oil level survey, inspection of a production line 0 16 –**
procedures, working areas, means of operation, and visit to a production area
transport, equipment (55)
Other auxiliary activity (59) Adjustment of a crane cable and removal of a pipe 6 12
blockage
Movement (61) Moving between work sites and the transfer of a 12 12
forklift
Other working processes (99) Lifting of a container and inspection of a process 3 2
operation

Specific physical activity (n = 32) (n = 46)

Operating machine (10) Usage of a band saw, a baking chamber, and a 9 9
grinding machine
Working with hand-held tools (20) Usage of a pressure washer, a torque wrench, and a 16 15
vacuum cleaner

(continued on next page)

1402 S. Nenonen / Safety Science 49 (2011) 1394–1403

Appendix A (continued)
Variable (ESAW-code) Examples of variables presented in the accident Outsourced In-house Sig.
reports operations operations
(%) (%)
Driving/being on board a means of transport or Usage of a wheel loader, weight transfer by a bridge 9 13
handling equipment (30) crane, and moving in a factory area by bicycle
Handling of objects (40) Unfastening of a valve’s packing and opening a door 16 7
Carrying by hand (50) Pulling a bulkhead and moving a box on a 3 7
production line
Movement (60) Walking on a deck and climbing to a platform 41 33
Presence (70) Presence near machinery, at a dock, and in a factory 6 17
area

Deviation (n = 33) (n = 49)

Deviation due to electrical problems, explosion, Inflammation of turpentine, electric arc from an 12 12
fire (10) element, and explosion of a reactor
Deviation by overflow, overturn, leak, flow, Escape of hot alkaline cooking alcohol and release of 6 4
vaporization, emission (20) tire pressure
Breakage, bursting, splitting, fell, collapse of Falling of a rolling door, sliding of a container cover, 30 20
material agent (30) and catapulting of a grindstone piece
Loss of control of machine, means of transport Unexpected work movement of a machine, sudden 6 24
or handling equipment, handheld tool, spin of a superconductor rod, and sudden
object, animal (40) movement of a truck
Slipping, stumbling and falling, fall of persons Falling of person through an uncovered opening 27 16
(50) and slipping in icy woodchips field
Body movement without any physical stress Hand being caught by a conveyor 0 4
(60)
Shock, fright, violence, aggression, threat, Working or moving in a danger zone 18 18
presence (80)

Contact – mode of injury (n = 34) (n = 49)

Contact with electrical voltage, temperature, Exposure to an electric arc, high temperatures due 12 14
hazardous substances (10) to a fire, and hot alkaline cooking alcohol
Drowned, buried, enveloped (20) Being buried in ash 3 0
Horizontal or vertical impact with or against a Getting struck in an asphalt field or on the floor 24 18
stationary object (30)
Struck by object in motion, collision with (40) Getting stuck by a tipping truck side, a falling 12 16
rolling door, a forklift, and a flying grindstone piece
Contact with sharp, pointed, rough, coarse Contact with high-pressure water 3 0
material agent (50)
Trapped, crushed etc. (60) Getting crushed between a truck cabin door, in a 44 51
waste press, and under a wheel loader
Other contacts (99) Getting struck on an ash hopper 3 0
⁄
0.050 6 p < 0.100.
⁄⁄
0.010 6 p < 0.050.
⁄⁄⁄
p < 0.010.

References Eurostat, 2009c. Number of Accidents at Work by Economic Activity and Size of
Enterprise. <http://nui.epp.eurostat.ec.europa.eu/nui/show.do?dataset=hsw_
aw_nnasz&lang=en> (retrieved 07.01.10).
Ali-Yrkkö, J., 2007. Outsourcing in Finnish Manufacturing – Does Industry Matter.
FAII – The Federation of Accident Insurance Institutions, 2009a. National Database
ETLA, The Research Institute of Finnish Economy, Discussion Papers No. 1070,
of Occupational Accidents and Diseases.
Helsinki, Finland. http://www.etla.fi/files/1715_Dp1070.pdf (retrieved
FAII – The Federation of Accident Insurance Institutions, 2009b. FAII’s Web Pages.
31.12.09).
<http://www.tvl.fi/www/page/tvl_www_6391> (retrieved 17.12.09).
Blank, V.L.G., Andersson, R., Lindén, A., Nilsson, B.-C., 1995. Hidden accident rates
Hämäläinen, P., 2010. Global Estimates of Occupational Accidents and Fatal
and patterns in the Swedish mining industry due to involvement of contractor
Work-Related Diseases. Doctoral Thesis. Tampere University of Technology,
workers. Safety Science 21 (1), 23–35.
Finland.
Clarke, S., 2003. The contemporary workforce, implications for organisational safety
Hätönen, J., Eriksson, T., 2009. 30+ years of research and practice of outsourcing –
culture. Personnel Review 32 (1), 40–57.
exploring the past and anticipating the future. Journal of International
Ekström, P., 2007. Ulkoistaminen laajentaa strategisia vaihtoehtoja. In: Lumijärvi,
Management 15 (2), 142–155.
O.-P. (Ed.), Huipulla. Miten yrityksen menestysyhtälö ratkaistaan, WSOY,
Holmes, N., Lingard, H., Yesilyurt, Z., DeMunk, F., 1999. An exploratory study of
Helsinki, Finland, pp. 195–219.
meanings of risk control for long term and acute effect occupational health and
European Commission, DG Employment, Social affairs and Equal opportunities,
safety risks in small business construction firms. Journal of Safety Research 30
2009. Causes and Circumstances of Accidents at Work in the EU. Luxembourg;
(4), 251–261.
Office of Official Publications of the European Communities.
Hon, C.K.H., Chan, A.P.C., Wong, F.K.W., 2010. An analysis for the causes of accidents
Eurostat, 2009a. 8.6% of Workers in the EU Experienced Work-related Health
of repair, maintenance, alteration and addition works in Hong Kong. Safety
Problems. Statistics on Focus, 63/2009. <http://epp.eurostat.ec.europa.eu/cache/
Science 7 (48), 894–901.
ITY_OFFPUB/KS-SF-09-063/EN/KS-SF-09-063-EN.PDF> (retrieved 07.01.10).
Kochan, T.A., Smith, M., Wells, J.C., Rebitzer, J.B., 1994. Human resource strategies
Eurostat, 2009b. International Sourcing Statistics. <http://epp.eurostat.ec.europa.
and contingent workers: the case of safety and health in the petrochemical
eu/statistics_explained/index.php/International_sourcing_statistics> (retrieved
industry. Human Resource Management 33 (1), 55–77.
04.01.10).
 S. Nenonen / Safety Science 49 (2011) 1394–1403
Kuitunen, K., Räsänen, P., Mikkola, M., Kuivanen, R., 1999. Developing Enterprise
Network. Supplier Networks as a Source of Competitiveness and Knowledge.
Research Notes 1976, Technical Research Centre of Finland, Espoo, Finland.
<http://www.vtt.fi/inf/pdf/tiedotteet/1999/T1976.pdf> (retrieved 7.11.09).
Lin, J., Mills, A., 2001. Measuring the occupational health and safety performance of
construction companies in Australia. Facilities 19 (3/4), 131–138.
Lind, S., 2008. Types and sources of fatal and severe non-fatal accidents in industrial
maintenance. International Journal of Industrial Ergonomics 38 (11–12), 927–933.
Lind, S., Nenonen, S., 2008. Occupational risks in industrial maintenance. Journal of
Quality in Maintenance Engineering 14 (2), 194–204.
Loosemoore, M., Dainty, A., Lingard, H., 2003. Human resource management in
construction projects - strategic and operational approaches. Taylor and Francis,
London. Cited in: Loosemore, M., Andonakis, A. (2007). Barriers to
implementing OHS reforms – The experiences of small subcontractors in the
Australian Construction Industry. International Journal of Project Management,
25(6), pp. 579-588.
Mayhew, C., Quinlan, M., Ferris, R., 1997. The effects of subcontracting/outsourcing
on occupational health and safety: survey evidence from four Australian
industries. Safety Science 25 (1–3), 163–178.
Mynttinen, M., 2006. Safety in Shared Workplaces. Bachelor’s Thesis, Jyväskylä
University of Applied Sciences, Finland. <http://www.tyosuojelu.fi/upload/
2is9hffo.pdf> (retrieved 06.10.09).
Nenonen, S., 2009. Problems in safety management in shared workplaces in
industry – the challenges for service providers. In: Proceedings of the 17th
World Congress on Ergonomics IEA 2009, Peking, China, p. 9. (14.08.09).
Nenonen, S., 2010. Safety management problems encountered by industrial service
providers. In: Proceedings of IEEM 2010, Conference on Industrial Engineering
and Engineering Management, Macau, p. 7. (10.12.10).
1403
Quinlan, M., 1999. The implications of labour market restructuring in industrialized
societies for occupational health and safety. Economic and Industrial
Democracy 20 (3), 427–460.
Quinlan, M., Mayhew, C., 2000. Precarious employment, work re-organisation and
the fracturing of OHS management. In: Frick, K., Jensen, P.L., Quinlan, M.,
Wilthagen, T. (Eds.), Systematic Occupational Health and Safety Management:
Perspectives on an International Development. Pergamon, Amsterdam, pp. 175–
198.
Rantanen, E., Lappalainen, J., Mäkelä, T., Piispanen, P., Sauni, S., 2007a. Accidents at
shared work places – lessons to be learned. Työ ja ihminen 21 (4), 364–379.
Rantanen, E., Lappalainen, J., Mäkelä, T., Piispanen, P., Sauni, S., 2007b. Yhteisten
työpaikkojen turvallisuus. TOT-raporttien analyysi. VTT Technical Research
Centre of Finland, Research Report, Tampere, Finland. <http://www.vtt.fi/inf/
julkaisut/muut/2007/VTT-R-02095-07.pdf> (retrieved 10.03.10).
Rikama, S., 2008. Suomalaisyritykset ovat ulkomaille ulkoistamisen etujoukkoa.
Tieto&Trendit, 4–5, Statistics Finland. <http://www.stat.fi/artikkelit/2008/
art_2008-07-10_003.html> (retrieved 31.12.09).
Rousseau, D.M., Libuser, C., 1997. Contingent workers in high risk environments.
California Management Review 39 (2), 103–123.
Saari, J., Räsänen, T., Levä, K., Mäkeläinen, J., Ruotsalainen, M., Seppänen, S., 2001.
Työkirja tapaturmien ja vaaratilanteiden tutkimiseen.
Tapaturmavakuutuslaitosten liitto, Työturvallisuuskeskus, Työterveyslaitos ja
Turvatekniikankeskus, Helsinki, Finland.
Salminen, S., 1995. Serious occupational accidents in the construction industry.
Construction Management and Economics 13 (4), 299–306.
Trethewy, R.W., Atkinson, M., Falls, B., 2003. Improved hazard identification for
contractors in the construction industry. Journal of Construction Research 4 (1),
71–85.