Beskæftigelsesudvalget 2021-22
BEU Alm.del Bilag 285
Offentligt
2595048_0001.png
Downloaded from www.sjweh.fi on June 07, 2022
Scand J Work Environ Health
Online-first -article
Published online: 01 Jun 2022
doi:10.5271/sjweh.4034
Workplace bullying and risk of suicide and suicide attempts: A
register-based prospective cohort study of 98 330 participants
in Denmark
by
Conway PM, Erlangsen A, Grynderup MB, Clausen T, Rugulies R,
Bjorner JB, Burr H, Francioli L, Garde AH, Hansen ÅM, Hanson LM,
Kirchheiner-Rasmussen J, Kristensen TS, Mikkelsen EG, Stenager E,
Thorsen SV, Villadsen E, Høgh A
In this large-size sample of Danish workers, we found new evidence of
a prospective association between exposure to workplace bullying and
subsequent suicidal behavior, including suicide attempt and death by
suicide, both assessed by means of register-based data. The results
call for a need to further strengthen national and international
measures against workplace bullying.
Affiliation:
Department of Psychology, University of Copenhagen,
Øster Farimagsgade 2A, 1353 Copenhagen, Denmark.
[email protected]
Refers to the following text of the Journal:
2016;42(3):246-250
Key terms:
bullying; cohort study; death by suicide; depression;
harassment; mental health; offensive behavior; register-based study;
risk; suicidal behavior; suicide; suicide attempt; workplace bullying
This article in PubMed:
www.ncbi.nlm.nih.gov/pubmed/35648097
Additional material
Please note that there is additional material available belonging to
this article on the
Scandinavian Journal of Work, Environment & Health
-website.
This work is licensed under a
Creative Commons Attribution 4.0 International License.
Print ISSN: 0355-3140 Electronic ISSN: 1795-990X
 BEU, Alm.del - 2021-22 - Bilag 285: Orientering om nye forskningsresultater om sammenhæng mellem mobning på arbejdspladsen og risiko for selvmord, fra beskæftigelsesministeren
2595048_0002.png
O
riginal article
Scand J Work Environ Health – online first: 1 June 2022. doi:10.5271/sjweh.4034
This work is licensed under a Creative Commons Attribution
4.0 International License.
Workplace bullying and risk of suicide and suicide attempts: A register-based
prospective cohort study of 98 330 participants in Denmark
by Paul Maurice Conway, PhD,
1
Annette Erlangsen, PhD,
2–5
Matias Brødsgaard Grynderup, PhD,
1
Thomas Clausen, PhD,
6
Reiner
Rugulies, PhD,
1, 6, 7
Jakob Bue Bjorner, PhD,
6–8
Hermann Burr, PhD,
9
Laura Francioli, PhD,
1
Anne Helene Garde, PhD,
6, 7
Åse Marie
Hansen, PhD,
6, 7
Linda Magnusson Hanson, PhD,
10
Jonas Kirchheiner-Rasmussen, MSc,
6
Tage S Kristensen, DrMedSci,
11
Eva
Gemzøe Mikkelsen, PhD,
12
Elsebeth Stenager, MD,
13
Sannie Vester Thorsen, PhD,
6
Ebbe Villadsen, BSc,
6
Annie Høgh, PhD
1
Conway PM, Erlangsen A, Grunderup MB, Clausen T, Rugulies R, Bjorner JB, Burr H, Francioli L, Garde AH, Hansen ÅM, Hanson
LM, Kirchheiner-Rasmussen J, Kristensen TS, Mikkelsen EG, Stenager E, Thorsen SV, Villadsen E, Høgh A. Workplace bullying
and risk of suicide and suicide attempts: A register-based prospective cohort study of 98 330 participants in Denmark.
Scand
J Work Environ Health
– online first. doi:10.5271/sjweh.4034
Objectives
The aim of this study was to analyze whether individuals reporting exposure to workplace bully-
ing had a higher risk of suicidal behavior, including both suicide attempt and death by suicide, than those not
reporting such exposure.
Methods
Using a prospective cohort study design, we linked data from nine Danish questionnaire-based surveys
(2004–2014) to national registers up to 31 December 2016. Exposure to workplace bullying was measured by a
single item. Suicide attempts were identified in hospital registers and death by suicide in the Cause of Death Reg-
ister. Among participants with no previous suicide attempts, we estimated hazard ratios (HR) and 95% confidence
intervals (CI), adjusting for sex, age, marital status, socioeconomic status, and history of psychiatric morbidity.
Results
The sample consisted of 98 330 participants (713 798 person-years), 63.6% were women, and the mean
age was 44.5 years. Of these participants, 10 259 (10.4%) reported workplace bullying. During a mean follow-up
of 7.3 years, we observed 184 cases of suicidal behavior, including 145 suicide attempts, 35 deaths by suicide and
4 cases that died by suicide after surviving a suicide attempt. The fully-adjusted HR for the association between
workplace bullying and suicidal behavior was 1.65 (95% CI 1.06–2.58). The HR for suicide attempts and death
by suicide were 1.65 (1.09–2.50) and 2.08 (0.82–5.27), respectively. Analyses stratified by sex showed a sta-
tistically significant association between workplace bullying and suicidal behavior among men but not women.
Conclusions
The results suggest that exposure to workplace bullying is associated with an elevated risk of
suicidal behavior among men.
suicidal
behavior.
Key terms
offensive behavior; death by suicide; depression; harassment; mental health; register-based study;
1
Department of Psychology, University of Copenhagen, Denmark.
2
Danish Research Institute for Suicide Prevention, Mental Health Centre Copenhagen, Capital Region of Denmark, Copenhagen, Denmark.
3
Copenhagen Research Centre for Mental Health, Capital Region of Denmark, Copenhagen, Denmark.
4
Department of Mental Health, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, United States.
5 Centre for Mental Health Research, Australian National University, Canberra, ACT, Australia.
6
National Research Centre for the Working Environment, Copenhagen, Denmark.
7 Department of Public Health, University of Copenhagen, Copenhagen, Denmark.
8 Optum Patient Insights, Lincoln, RI, USA.
9 Department of Work and Health, Federal Institute for Occupational Safety and Health BAuA, Berlin, Germany.
10
Stress Research Institute, Stockholm University, Stockholm, Sweden.
11
Task-Consult, Gilleleje, Denmark.
12
Department of Psychology, University of Southern Denmark, Odense, Denmark.
13
Focused Research Unit in Psychiatry, Institute of Regional Health Research, University of Southern Denmark, Odense, Denmark.
Correspondence to: Paul Maurice Conway, Department of Psychology, University of Copenhagen, Øster Farimagsgade 2A, 1353 Copenha-
gen, Denmark. [E-mail: [email protected]].
Scand J Work Environ Health – online first
1
 BEU, Alm.del - 2021-22 - Bilag 285: Orientering om nye forskningsresultater om sammenhæng mellem mobning på arbejdspladsen og risiko for selvmord, fra beskæftigelsesministeren
2595048_0003.png
Workplace bullying and suicidal behavior
Globally, more than 817 000 deaths by suicide occur
annually, while the number of persons with a non-fatal
episode (ie, self-harm) might be 20 times higher (1). The
suicide rate has declined globally over recent decades
(2).
In Denmark, the suicide rate has remained relatively
stable in recent years, suggesting that new venues for
intervention are needed.
Suicidal behavior has a complex etiology, with mul-
tiple contributions from biological, psychological, clini-
cal, social, and environmental factors (3). Among social
factors, repeated exposure to psychosocial work stressors
may play a role in suicidality (4).
Workplace bullying is
an especially severe form of psychosocial stressor, involv-
ing a long-lasting exposure to repeated negative behaviors
at work, such as harassment, offending or ostracizing
individuals, or taking actions that negatively affect an
individual’s job (5).
The average prevalence of workplace
bullying worldwide is estimated as 14.6%, ranging from
11.3% – when workplace bullying is measured using
the self-labelling method (as it is the case in the present
study) – to 18.1%, when workplace bullying is instead
measured using the behavioral experience approach (6).
The available evidence supports a role of workplace bul-
lying in the onset of mental disorders, especially depres-
sion (7,
8),
and suicidal ideation (9,
10),
which are key
antecedents of suicidal behavior (11). A link between
workplace bullying and both mental disorders and sui-
cidality appears plausible, considering that exposure to
workplace bullying may cause severe psychological pain,
including feelings of hopelessness, entrapment, loss of
control, worthlessness, social exclusion, deterioration
of self-esteem, and chronic psychological distress (12).
However, prospective studies on the association between
workplace bullying and subsequent risk of suicidal behav-
ior, including suicide attempt and death by suicide, are
lacking, with the only available evidence supporting a
prospective association between workplace bullying and
suicidal ideation (13).
We aim to fill this gap by examin-
ing, in a large Danish study linking pooled survey data to
national register data, the prospective association between
exposure to workplace bullying and subsequent suicidal
behavior, including suicide attempts and death by suicide.
To avoid selective reporting and other post-hoc decision-
making biases, a protocol detailing the analytical plan was
published prior to the present study (14).
Methods
Study design and participants
We adopted a prospective cohort study design. At the
Danish National Research Centre for the Working Envi-
ronment, we created a single dataset by pooling together
questionnaire data collected from 2004–2014 in nine
Danish surveys, all of which contained an item on
self-reported workplace bullying. The surveys included
individuals employed in different occupational groups
in both the private and public sector (see supplemen-
tary material
www.sjweh.fi/article/4034,
table S1, for
details about the surveys included). The nine surveys
provided 14 waves of measurement in all, since four
of the surveys [The Danish Work Environment Cohort
Study (DWECS), Workplace Bullying and Harassment
(WBH), Social and Health Care Study (SOSU), and
Work Environment and Health (WEHD)] comprised
more than one wave. Using the unique personal identi-
fier, assigned to all Danish residents (15),
the pooled
dataset was linked to the following national registers:
the Danish Civil Register (since 2004) (16),
the Danish
National Patient Register (since 1977) (17),
the Danish
Psychiatric Central Research Register (since 1994) (18),
the Danish Register of Causes of Death (since 2004)
(19),
and the Income Statistics Register (since 2004).
Those individuals who responded to the item on
self-reported workplace bullying in these surveys were
included in the present study. With regard to the four
surveys with more than one wave, the first wave in
which participants provided a valid response to the item
on workplace bullying was used to determine exposure
status (ie, exposed or non-exposed to workplace bully-
ing). For instance, if in the three-wave SOSU cohort,
a participant provided a valid answer to the item of
workplace bullying in the second but not the first wave,
the second wave was used to classify the participant as
exposed or not exposed. The date the question on work-
place bullying was answered was considered as the date
of exposure and follow-up started on the following day.
If the response date was missing, date of exposure was
considered as the date the survey questionnaire was sent
out. With regard to the study outcomes (suicide attempt
and death by suicide), participants were followed in the
registers from the date they completed the question-
naire survey until 31 December 2016. Participants, who
migrated or died by causes other than suicide, were cen-
sored from the study at the time of the respective event.
Overall, the pooled dataset included 139 575
questionnaire responses, corresponding to 105 455
unique participants. First, we excluded 6192 (5.9%)
participants with missing data on workplace bullying
(N=5945) and the covariates sex, age, marital status,
and socioeconomic status (N=247), resulting in 99 263
participants (94.1%). Next, we excluded 930 (0.9%)
participants with previous suicide attempts. We finally
excluded three participants who were officially listed
as having migrated out of the country on the date when
they answered the questionnaire. The application of
these further exclusion criteria resulted in a final sample
of 98 330 (93.2%) participants included in the present
2
Scand J Work Environ Health – online first
 BEU, Alm.del - 2021-22 - Bilag 285: Orientering om nye forskningsresultater om sammenhæng mellem mobning på arbejdspladsen og risiko for selvmord, fra beskæftigelsesministeren
2595048_0004.png
Conway et al
analyses (figure 1). A comparison of the sociodemo-
graphic characteristics between participants in the final
sample and participants who had been excluded due to
missing values is provided in supplementary table S3.
Ethics committee approval
The study followed the principles outlined in the Decla-
ration of Helsinki. The Danish Data Protection Agency
approved the project (Capital Region of Denmark; j.nr.:
2012-58-0004) and data are stored on a secure server
at Statistics Denmark. According to Danish law, stud-
ies that use questionnaire and register data only do not
require approval from the National Committee on Health
Research Ethics.
Workplace bullying
In seven of the nine surveys, self-reported exposure to
workplace bullying was measured retrospectively with
the following questionnaire item: “Have you been sub-
jected to bullying at work within the past 12 months?”,
to be answered using a Likert-type scale with five
response options; 1=never, 2=now and then, 3=monthly,
4=weekly, and 5=daily. In the Nursing Work Environ-
ment, Well-being and Health (SATH), DWECS 2005,
and DWECS 2010 surveys, exposure frequency was
measured with a dichotomous item (1=no, 2=yes).
In WBH 2006 and WBH 2008, the format with five
response options was used but the retrospective expo-
sure time was 6 instead of 12 months. For all the sur-
veys included, the questionnaire item was preceded by
a definition of workplace bullying, which is useful to
calibrate responses by reducing the potential influence
of individual differences in the interpretation of the
exposure. To harmonize measures across surveys, we
created the following dichotomous exposure variable
that was used in all the analyses of the present study:
1=0 “non-exposed to workplace bullying” (reference);
2–5=1 “exposed to workplace bullying”.
Suicidal behavior
The primary outcome, suicidal behavior, was defined
as a first episode of a suicide attempt or death by sui-
cide during follow-up. Suicide attempts were identified
through contacts to somatic or psychiatric hospitals
in the Danish National Patient Register (17)
and the
Danish Psychiatric Central Research Register (18).
This comprised persons who had been registered with
a main or sub-diagnosis, according to the 10
th
revision
of the International Classification of Diseases (ICD),
which indicated a suicide attempt (ICD-10: X60-X84)
or that the reason for contact was a suicide attempt. All
types of hospital contacts, ie, emergency department,
inpatient, and outpatient, were included. Participants
were classified as cases of suicide attempt at the first
date of the event, and then censored (ie, repeated events
of suicide attempt were not considered). Information on
individuals who died by suicide was retrieved from the
Danish Register of Causes of Death, using the ICD-10
codes X60-X84 (19).
The date of death was considered
as the date of the outcome. Participants were censored
after the first episode of suicide behavior. Participants
registered with a suicide attempt before the baseline date
were excluded from the study.
Figure 1.
Participants’ inclusion criteria.
Scand J Work Environ Health – online first
3
 BEU, Alm.del - 2021-22 - Bilag 285: Orientering om nye forskningsresultater om sammenhæng mellem mobning på arbejdspladsen og risiko for selvmord, fra beskæftigelsesministeren
2595048_0005.png
Workplace bullying and suicidal behavior
Covariates
The following covariates were included: sex (men,
women); age (continuous variable); marital status
(unmarried; married/cohabiting/registered partnership;
divorced; widowed); socioeconomic status, coded based
on Denmark Statistics’ classification (Low, Medium,
High, Student/Other); history of mental disorders (yes/no;
identified as ICD-8 codes 290-316 or ICD-10 codes F00-
F99); history of a psychotropic drug prescription (yes/
no; identified as prescription of antipsychotics [N05A],
anxiolytics [N05B], anxiolytics, hypnotics and sedatives
[N05C], antidepressants [N06A] or psychostimulants,
agents used for ADHD and nootropics [N06B]).
Data on sex, age, and marital status were obtained
from the Danish Civil Register (16),
and information
on socioeconomic status was derived from the Income
Statistics Register. Information on history of mental dis-
orders and psychotropic drug prescriptions was retrieved
from the Danish Psychiatric Central Research Register
and the Danish National Prescription Registry (20),
respectively. Since most surveys asked about workplace
bullying during the previous 12 months, information
on history of mental disorders and psychotropic drug
prescriptions were included from 1 January 2000 up to
12 months before survey participation.
Statistical analyses
Main analysis.
Using multivariable Cox proportional
hazard models, hazard ratios (HR) and 95% confidence
intervals (CI) were calculated to estimate the association
between workplace bullying and subsequent suicidal
behavior. Robust clusters based on the survey waves
were used to account for intra-group correlations due to
the clustering of participants in different surveys (21).
The proportional hazards assumption was tested and
confirmed by Schoenfeld’s residuals (P>0.05), and the
visual inspection of the observed Kaplan-Meier survival
curves and the log-log plots (data not shown).
Next to a crude analysis, we adjusted for sex, age,
marital status and socioeconomic status (model 1). We
additionally adjusted for history of mental disorders
(model 2), considering that individuals with mental
disorders might have an elevated risk of both being
exposed to workplace bullying and engaging in suicidal
behavior (11).
Supplementary analyses.
First, we estimated the associa-
tions between workplace bullying and suicide attempt
and death by suicide as separate outcomes. Second, we
conducted sub-group analyses with respect to sex, age
dichotomized (<31 versus ≥31 years), socioeconomic
status and marital status, and calculated multiplicative
interaction terms in the fully adjusted model. Third, we
Table 1.
Characteristics of study sample (N=98 330). [SD=standard
deviation.]
Total N (%)
Sample
Age
Age (years) dichotomized
<31
≥31
Women
Participants reporting exposure to
workplace bullying
Marital status
Living alone
Cohabiting
Divorced
Widow (no)
Socioeconomic status
a
Low
Medium
High
Student/Other
Diagnoses of any mental disorder
Psychotropic drug prescriptions
a
Mean (SD)
44.5 (11.2)
98 330 (100)
13 327 (13.6)
85 003 (86.4)
62 582 (63.6)
10 259 (10.4)
12 332 (12.5)
78 568 (79.9)
6208 (6.3)
1222 (1.2)
37 595 (38.2)
≈24 900
b
(≈25)
≈19 210
b
(≈20)
16 627 (16.9)
3064 (3.1)
20 901 (21.3)
For socioeconomic status: low=employed in a job requiring skills on a
basic level; medium=employed in a job requiring skills on the mid-level;
high=leaders, both employed and self-employed with subordinates, and
participants employed in a job requiring skills on the highest level; student/
other=students and self-employed without subordinates.
b
The exact total number of participants is not reported for high and medium
socioeconomic status to avoid being able to calculate the numbers for SOSU
U by subtraction (see supplementary table S2). The symbol ≈ indicates ap-
proximate numbers.
examined the association between workplace bullying
and suicidal behavior while excluding four participants
who had a record of both suicide attempt and death by
suicide. Fourth, we examined the association between
workplace bullying and suicidal behavior adjusting for
history of psychotropic drug prescriptions, instead of
history of mental disorders. Fifth, we calculated the
association between workplace bullying and suicide
behavior stratified by cases that occurred in the first four
years of follow-up and cases that occurred later. Sixth,
we calculated the association between the covariates and
suicidal behavior.
All analyses were performed following the proce-
dure described in the pre-published protocol (14), with
the exception of the calculation of interaction terms,
which was an authors’ post-hoc decision. Other analyses
not described in the pre-published protocol (exclusion
of the four participants who were recorded both with
a suicide attempt and later with death by suicide and
stratification of suicidal behavior by cases that occurred
during the first four years of follow-up and cases that
occurred later) were included during the revision pro-
cess. All analyses were conducted using the statistical
package STATA 16.1 (StataCorp LP, College Station,
TX, USA).
4
Scand J Work Environ Health – online first
 BEU, Alm.del - 2021-22 - Bilag 285: Orientering om nye forskningsresultater om sammenhæng mellem mobning på arbejdspladsen og risiko for selvmord, fra beskæftigelsesministeren
2595048_0006.png
Conway et al
Figure 2.
Kaplan-Meier curve
for the probabilities of suicidal
behavior for participants exposed
(solid line) and not exposed (dashed
line) to workplace bullying.
Number at risk
Exposed
Non-exposed
10 259
88 071
10 160
87 337
8204
72 712
5348
51 993
4638
45 916
Results
The mean study follow-up time was 7.3 years [range
1 day to 12.1 years; median 8.1, standard deviation
(SD) 3.3 years], yielding 713 798 person-years. The
characteristics of the study sample are shown in table
1. Details for each of the 14 samples are provided in
supplementary table S2. The sample consisted of 63.6%
women, and the mean age was 44.5 years. Overall, 10
259 participants (10.4%) reported exposure to workplace
bullying. During follow-up, 184 cases of suicidal behav-
ior were identified, consisting of 145 suicide attempts,
35 death by suicide and 4 cases that first had a suicide
attempt and later died of suicide. These 4 cases were
included in both analyses on suicide attempt and death
by suicide. The mean number of years and standard
deviation between baseline and suicide events were 4.0
(SD 2.8) for suicidal behavior (N=184), 3.9 (SD 2.9) for
suicide attempt (N=149), and 4.6 (SD 2.6) for death by
suicide (N=39). Of all participants, 3.1% and 21.3% had
a history of diagnosed mental disorders and a history of
psychotropic drug prescriptions, respectively.
Main analysis
Figure 2 shows the Kaplan-Meier curves for the prob-
abilities of suicidal behavior for participants exposed
and not exposed to workplace bullying during follow-
up. The curves indicate that the proportional hazard
assumption was fulfilled and that exposed participants
showed more suicidal behavior. The corresponding HR
are presented in table 2. Participants exposed to work-
place bullying at baseline had a statistically significant
elevated risk of suicidal behavior compared with the
non-exposed (HR 1.83, 95% CI 1.20–2.78; P=0.002) in
the crude model. The association remained statistically
significant after adjusting for sex, age, marital status,
and socioeconomic status (HR 1.77, 95% CI 1.15–2.70;
model 1), and additionally for previous history of diag-
nosed mental disorders (HR 1.65, 95% CI 1.06–2.58;
model 2).
Supplementary analyses
When we analyzed suicide attempt and death by suicide
separately (table 2), exposure to workplace bullying was
associated with both suicide attempts (fully-adjusted HR
1.65, 95% CI 1.09–2.50) and death by suicide (fully-
adjusted HR 2.08, 95% CI 0.82–5.27), although the CI
for death by suicide were wide and included unity. The
Kaplan-Meier curves for suicide attempt and death by
suicide are provided in supplementary figures S1 and S2.
Table 3 shows the analyses stratified by sex, age,
socioeconomic status, and marital status. The HR for
Scand J Work Environ Health – online first
5
 BEU, Alm.del - 2021-22 - Bilag 285: Orientering om nye forskningsresultater om sammenhæng mellem mobning på arbejdspladsen og risiko for selvmord, fra beskæftigelsesministeren
2595048_0007.png
Workplace bullying and suicidal behavior
Table 2.
Hazard ratios (HR) of suicidal behavior, suicide attempt and death by suicide. [WB=workplace bullying]
N (%) of suicidal
events
a
Suicidal behavior
Non-exposed to WB
Exposed to WB
Suicide attempt
Non-exposed to WB
Exposed to WB
Death by suicide
Non-exposed to WB
Exposed to WB
a
Person-years
Rate per 100 000
person-years
23. 81
43.52
19.30
35.01
4.82
11.21
Crude HR
(95% CI)
1.00
1.83 (1.20–2.78)
1.00
1.81 (1.22–2.70)
1.00
2.34 (0.98–5.62)
Model 1
b
HR
(95% CI)
1.00
1.77 (1.15–2.70)
1.00
1.77 (1.19–2.63)
1.00
2.21 (0.87–5.64)
Model 2
c
HR
(95% CI)
1.00
1.65 (1.06–2.58)
1.00
1.65 (1.09–2.50)
1.00
2.08 (0.82–5.27)
153 (83.2)
31 (16.8)
124 (83.2)
25 (16.8)
31 (79.5)
8 (20.5)
642 571
71 227
642 571
71 227
643 183
71 391
Four participants first attempted suicide, followed by death by suicide during follow-up. These participants are included in all three analyses. For the analyses of
suicide behavior and suicide attempt, case status is determined on the date of the suicide attempt; for death by suicide, case status is determined based on the date
of death.
b
Model 1 was adjusted for sex, age, marital status, and socio-economic status.
c
Model 2 was adjusted for the same covariates as in Model 1 plus previous history of diagnosed mental disorders.
Table 3.
Hazard ratios (HR) of suicidal behavior stratified by sex, age dichotomised, socio-economic status and marital status. Results are not
reported for widowed respondents due to too few cases of suicidal events. [WB=workplace bullying]
N (%) of suicidal
events
Men (N=35 748)
Non-exposed to WB
Exposed to WB
Women (N=62 582)
Non-exposed to WB
Exposed to WB
Aged <31 (N=13 327)
Non-exposed to WB
Exposed to WB
Aged ≥31 (N=85 003)
Non-exposed to WB
Exposed to WB
Low/Student/Other socio-eco-
nomic status
f
(N=37 595)
Non-exposed to WB
Exposed to WB
Medium/High socio-economic
statusf (N=44 108)
Non-exposed to WB
Exposed to WB
Living alone (N=12 332)
Non-exposed to WB
Exposed to WB
Cohabiting (N=78 586)
Non-exposed to WB
Exposed to WB
Divorced (N=6208)
Non-exposed to WB
Exposed to WB
a
b
Person-years
Rate per 100 000
person-years
22.82
76.33
24.28
29.84
31.45
58.62
22.58
41.47
28.91
46.78
15.39
33.20
23.45
58.62
23.13
35.70
33.42
74.03
Crude HR
(95% CI)
1.00
3.32 (2.04–5.40)
1.00
1.23 (0.79–1.92)
1.00
1.84 (0.59–5.77)
1.00
1.84 (1.17–2.90)
1.00
1.61 (0.87–2.98)
1.00
2.17 (1.18–3.98)
1.00
2.48 (1.26–4.86)
1.00
1.72 (0.94–2.53)
1.00
2.23 (0.68 –7.27)
Model 1 HR
(95% CI)
1.00
3.03 (1.81– 5.06)
a
1.00
1.23 (0.79–1.91)
a
1.00
1.62 (0.54–4.87)
b
1.00
1.76(1.09–2.83)
b
1.00
1.62 (0.88–2.96)
c
1.00
2.03 (1.09–3.79)
c
1.00
2.44 (1.24–4.81)
d
1.00
1.51 (0.92–2.50)
d
1.00
2.32 (0.72–7.56)
d
Model 2
e
HR
(95% CI)
1.00
2.92 (1.74–4.91)
1.00
1.12 (0.71–1.78)
1.00
1.55 (0.51–4.72)
1.00
1.64 (1.00–2.71)
1.00
1.53 (0.83–2.81)
1.00
1.87 (0.98–3.55)
1.00
2.45 (1.25–4.82)
1.00
1.38 (0.80–2.37)
1.00
2.23 (0.69–7.21)
47 (74.6)
16 (25.4)
106 (87.6)
15 (12.4)
28 (84.8)
5 (15.2)
125 (82.8)
26 (17.2)
75 (83.3)
15 (16.7)
43 (82.7)
9 (17.3)
18 (75.0)
6 (25.0)
120 (86.3)
19 (13.7)
13 (72.2)
5 (27.8)
205 950
20 961
436 620
50 266
89 017
8 530
553 554
62 697
259 438
32 067
279 417
27 105
76 764
10 236
518 842
53 221
38 896
6 754
Model 1 adjusted for age, marital status, and socio-economic status.
Model 1 adjusted for sex, marital status, and socio-economic status.
c
Model 1 adjusted for sex, age, and marital status.
d
Model 1 adjusted for sex, age, and socio-economic status.
e
Model 2 adjusted for the same covariates as in Model 1 plus previous history of diagnosed mental disorders.
f
For the present analysis, socio-economic status was dichotomized into: Low/Student/Other and Medium/High.
6
Scand J Work Environ Health – online first
 BEU, Alm.del - 2021-22 - Bilag 285: Orientering om nye forskningsresultater om sammenhæng mellem mobning på arbejdspladsen og risiko for selvmord, fra beskæftigelsesministeren
2595048_0008.png
Conway et al
the association between workplace bullying and suicidal
behavior were similar for younger (<31 years) versus
older (≥31 years) participants (P for multiplicative inter-
action=0.969), for participants with low versus medium/
high socioeconomic status (P for multiplicative interac-
tion=0.458), and for persons living alone versus cohabit-
ing persons (P for multiplicative interaction=0.216) and
divorced (P for multiplicative interaction=0.860), with
overlapping CI. With regard to sex, the HR in the fully-
adjusted model were statistically significant among men
(HR, 2.92; 95% CI 1.74–4.91) but not among women
(HR, 1.12; 95% CI 0.71–1.78), with a P for multiplicative
interaction of 0.002, indicating effect modification by sex.
When we excluded the four individuals who were
recorded both with a suicide attempt and later with death
by suicide, estimates were slightly attenuated but in the
same direction as the estimates in the main analyses
(supplementary table S4).
When we adjusted for history of psychotropic drug
prescriptions, instead of history of diagnosed mental
disorders, the fully-adjusted HR for suicidal behavior
was comparable to the HR from the main analysis
(supplementary table S5).
When we stratified suicidal behavior by cases that
occurred during the first four years of follow-up and
cases that occurred later, the associations were slightly
stronger for cases occurring in the first four years (Sup-
plementary tables S6 and S7).
Supplementary table S8 shows the crude and
adjusted HR for the covariates. Risk of suicidal behav-
ior was higher for men than women, younger than older
individuals, individuals with a lower socioeconomic
status, and individuals with a history of diagnosed men-
tal disorders compared to those without such a history.
association between another type of offending behavior,
workplace sexual harassment, and register-based sui-
cidal behavior in the Swedish workforce (22).
We found that workplace bullying was a statistically
significant risk factor for suicidal behavior among men.
This is in accord with a few previous studies suggest-
ing that the association between workplace bullying
and mental health is stronger among men, although the
current evidence about sex-related differences remains
inconclusive (23).
A possible explanation for this find-
ing is that men may tend to make less frequent use of
health services (eg, professional psychological support)
when confronted with adverse life circumstances (24).
For instance, previous research found that masculinity
may reduce help-seeking for depression among men
(25).
Reduced support seeking might lead to untreated or
worsened mental disorders, which may in turn increase
the risk of suicidal behavior (11).
In addition, men
might rely on their work role to establish their self-
identity more than women do, which may result in
mental health being more strongly affected among men
when confronted with highly stressful events – such as
workplace bullying – that can severely threaten one’s
self-esteem (26).
Strengths and limitations
The strengths of our study are the prospective design
and the large sample size, providing sufficient statistical
power to analyze the prospective association between a
low-base rate phenomenon such as workplace bullying
and suicidal behavior. The study was based on a detailed
protocol describing the planned analytical strategy that
was published before the analyses were performed (14),
as a precaution against selective reporting and post-hoc
decision making. In addition, suicide attempts and death
by suicide were assessed using register-based data, and
unique identifiers allowed us to ensure that each indi-
vidual was included only once. Finally, by adjusting for
register-based history of diagnosed mental disorders and
psychotropic drug prescriptions, we were able to account
for the fact that individuals with previous mental disor-
ders may be at a higher risk of both being bullied and
engaging in suicidal behavior (11).
This study has also limitations. Information on his-
tory of non-treated mental disorders was not available.
Non-participation and missing data in the questionnaire
surveys could have been related to both the reporting
of workplace bullying and subsequent suicidal behav-
ior, which might have biased our estimates. We could
not account for possible confounding factors, such as
personality traits, co-occurring life- and work-related
traumatic events, and exposure to other psychosocial
work stressors. In particular, personality traits have been
associated with both the reporting of workplace bullying
Scand J Work Environ Health – online first
Discussion
Pooling data from 14 survey waves, we were able to
generate the largest study sample to date on the asso-
ciation between exposure to workplace bullying and
suicidal behavior. To our knowledge, this is the first pro-
spective study examining such an association. We found
an increased risk of suicidal behavior, including both
suicide attempts and death by suicide, among men who
had previously reported exposure to workplace bully-
ing and had no previous history of suicide attempts and
diagnosed mental disorders. The association between
workplace bullying and suicidal behavior was not sta-
tistically significant among women.
Our findings are in agreement with earlier studies
reporting associations between workplace bullying
and the onset of suicidal ideation (9,
10)
and mental
disorders (7). A recent study reported a prospective
7
 BEU, Alm.del - 2021-22 - Bilag 285: Orientering om nye forskningsresultater om sammenhæng mellem mobning på arbejdspladsen og risiko for selvmord, fra beskæftigelsesministeren
2595048_0009.png
Workplace bullying and suicidal behavior
and suicidal behavior (27,
28). Lack of analytical power,
mainly resulting from the low base-rate prevalence of
suicidal behavior, prevented us from estimating dose–
response associations between exposure to workplace
bullying of increasing frequency and suicidal behavior.
The latter might have introduced effect underestimation
since the most negative consequences on mental health
were previously observed in connection with the most
severe degrees of exposure (ie, frequent bullying) (8).
Further, it can be argued that infrequent exposure to
workplace bullying (eg, being bullied “now and then”)
does not constitute a chronic stressor. That we still found
an association with suicidal behavior suggests that work-
place bullying might lead to extreme consequences for
mental health even when the exposure is not chronic, as
it is the case when individuals report being bullied on
an infrequent basis. At present, however, this remains
speculative, and future studies with sufficient power to
perform dose–response analyses are needed to shed light
onto the impact that different frequencies of exposure
to workplace bullying have on suicidal behavior. While
in the present study a one-time exposure to workplace
bullying is associated with later suicidal behavior, we
were not able to examine if variation in exposure to
workplace bullying over time plays a role in the size of
the association. In addition, while we adjusted and strati-
fied the analyses by socioeconomic status, we could not
examine the potential confounding role of occupation.
This is a limitation because previous research suggests
that both the risk of workplace bullying (29) and the
risk of suicide might differ according to occupation (30).
Suicidal behavior, especially suicide attempt, may be
under-detected, hence possibly underestimating absolute
risks reported here. As suicide is a low base-rate occur-
rence, we had limited statistical power to detect effects
of small or moderate size in the sub-group analyses. The
generalizability of our findings should be ascertained
in future studies examining the association between
workplace bullying and suicidal behavior in other geo-
graphical contexts.
Implications for future research
To date, there is scarce empirical evidence about the
mechanisms linking workplace bullying to suicidal
behavior. Future research should address such mecha-
nisms by examining both moderators and mediators of
the association between workplace bullying and suicidal
behavior. For instance, factors such as feeling defeated
or humiliated have been reported as potential anteced-
ents to suicidal behavior if not mitigated by, for exam-
ple, effective coping strategies or social support (11).
When it comes to potential mediators, it is plausible
that mental disorders, such as depression, play a sub-
stantial role in this association (8,
11). Workplace bul-
lying might lead to enduring feelings of entrapment and
humiliation, which are associated with the development
of mental disorders (31).
In addition, workplace bullying
may provoke a severe deterioration of self-esteem and
feelings of worthlessness, which have been found in
association with elevated risk of suicidal behavior (32,
33).
From a pathophysiological perspective, due to its
long-lasting and escalating nature, workplace bullying
might represent a stressor that could lead to dysregula-
tions of the hypothalamic-pituitary-adrenal axis and
subsequent physiological changes that are linked to the
development of depression (34).
Practical implications
On a clinical level, primary care clinicians and psy-
chiatrists should be aware that patients presenting with
mental health problems related to particularly severe
work-related experiences, such as workplace bullying,
might be at elevated risk of suicidal behavior. Our find-
ings suggest the need to pay special attention to men.
Psychotherapy could be useful for targets suffering from
mental health problems. For instance, a cognitive-behav-
ioral in-patient psychotherapeutic approach specifically
developed for the treatment of individuals exposed to
workplace bullying has proven effective in improving
mental health (35).
On top of initiatives supporting indi-
viduals who have been already exposed to workplace
bullying, there is a need to implement workplace inter-
ventions aimed at reducing work-related psychosocial
factors that may increase the risk of workplace bullying
(eg, role conflicts and ambiguity (36)). Workplace level
interventions should also include the implementation of
procedures to handle cases of workplace bullying as they
occur as well as conflict management initiatives to pre-
vent conflicts from escalating into bullying (37). From
a public health perspective, the new evidence reported
in this study supports efforts to strengthen national and
international measures against workplace bullying.
Concluding remarks
In conclusion, our study showed that workplace bullying
is associated with subsequent suicidal behavior among
individuals who had not previously been recorded with
a suicide attempts, while also adjusting for history of
mental disorders and psychotropic drug prescriptions.
In sub-group analysis, this association was statistically
significant among men but not women.
8
Scand J Work Environ Health – online first
 BEU, Alm.del - 2021-22 - Bilag 285: Orientering om nye forskningsresultater om sammenhæng mellem mobning på arbejdspladsen og risiko for selvmord, fra beskæftigelsesministeren
2595048_0010.png
Conway et al
Acknowledgment
3.
Feb;364:l94.
PubMed https://doi.org/10.1136/bmj.l94.
Turecki G, Brent DA, Gunnell D, O’Connor RC, Oquendo
MA, Pirkis J et al. Suicide and suicide risk. Nat Rev Dis
Primers 2019 Oct;5(1):74.
https://doi.org/10.1038/s41572-
019-0121-0.
Milner A, Witt K, LaMontagne AD, Niedhammer I.
Psychosocial job stressors and suicidality: a meta-
analysis and systematic review. Occup Environ Med
2018 Apr;75(4):245–53.
https://doi.org/10.1136/
oemed-2017-104531.
Einarsen S, Hoel H, Zapf D, Cooper C. The Concept of
Bullying and Harassment at Work: The European Tradition.
In: Einarsen S, Hoel H, Zapf D, Cooper C, editors. Bullying
and harassment in the workplace: Theory, research and
practice. 3rd ed. Boca Raton, FL: CRC Press Taylor &
Francis Group; 2020. p. 3–53.
Nielsen MB, Matthiesen SB, Einarsen S. The impact
of methodological moderators on prevalence rates
of workplace bullying. A meta-analysis. J Occup
Organ Psychol 2010;83(4):955–79.
https://doi.
org/10.1348/096317909X481256.
Theorell T, Hammarström A, Aronsson G, Träskman Bendz
L, Grape T, Hogstedt C et al. A systematic review including
meta-analysis of work environment and depressive
symptoms. BMC Public Health 2015 Aug;15(1):738.
https://
doi.org/10.1186/s12889-015-1954-4
Gullander M, Hogh A, Hansen ÅM, Persson R, Rugulies R,
Kolstad HA et al. Exposure to workplace bullying and risk of
depression. J Occup Environ Med 2014 Dec;56(12):1258–
65.
https://doi.org/10.1097/JOM.0000000000000339.
Nielsen MB, Nielsen GH, Notelaers G, Einarsen S.
Workplace Bullying and Suicidal Ideation: A 3-Wave
Longitudinal Norwegian Study. Am J Public Health
2015 Nov;105(11):e23–8.
https://doi.org/10.2105/
AJPH.2015.302855.
All coauthors were involved in the design of the study
protocol. All coauthors participated in conceptualizing
and designing the present study. Funding for this study
was obtained by AH, AMH, PMC, AE, TC, and ES.
MBG, TC, AE, and EV performed the data management.
TSK participated in the data collection for COPSOQ II.
AH participated in the data collection for WBH 2006
and WBH 2008. AH and TC participated in the data
collection for SOSU I, SOSU II, and SOSU III. ÅMH
participated in the data collection for PRISME 2007,
WBH 2006, and WBH 2008. SWT participated in the
data collection for DANES 2008. HB participated in the
data collection for DWECS 2005 and DWECS 2010. RR
participated in the data collection for WEHD 2012 and
WEHD 2014. EV participated in the data collection for
PRISME 2007, COPSOQ II, DWECS 2005, DWECS
2010, WEHD 2012, and WEHD 2014 (see supplemen-
tary table 1 for a description of these surveys). MBG,
AE, PMC, TC and JKR carried out the formal analyses.
MBG, AE, TC and JKR curated the data visualization.
PMC administered and supervised the project in close
collaboration with AE, TC, and MBG. PMC, AE, and
MBG, and TC wrote the first draft of the manuscript,
which was critically revised and approved in its final
version by all coauthors. The corresponding author
(PMC) confirms that all listed authors meet authorship
criteria.
Conflict of interest
Authors declare no conflicts of interest.
Funding
The study was supported by the Region of South-
ern Denmark (grant number A1763) and the Danish
Working Environment Research Fund (grant numbers
20130023294 and 10-2019-03).
The funders had no role in study design, the col-
lection, analysis and interpretation of data, the writing
of the report, and the decision to submit the paper for
publication.
4.
5.
6.
7.
8.
9.
10. Nielsen MB, Einarsen S, Notelaers G, Nielsen GH. Does
exposure to bullying behaviors at the workplace contribute
to later suicidal ideation? A three-wave longitudinal study.
Scand J Work Environ Health 2016 May;42(3):246–50.
https://doi.org/10.5271/sjweh.3554.
11. O’Connor RC, Nock MK. The psychology of suicidal
behaviour. Lancet Psychiatry 2014 Jun;1(1):73–85.
https://
doi.org/10.1016/S2215-0366(14)70222-6.
12. Mikkelsen EG, Hansen ÅM, Persson R, Byrgesen MF, Hogh
A. Individual Consequences of Being Exposed to Workplace
Bullying. In: Einarsen S, Hoel H, Zapf D, Cooper C, editors.
Bullying and harassment in the workplace: Theory, research
and practice. 3rd ed. Boca Raton, FL: CRC Press Taylor &
Francis Group; 2020. p. 163–208.
13. Leach LS, Poyser C, Butterworth P. Workplace bullying
and the association with suicidal ideation/thoughts
and behaviour: a systematic review. Occup Environ
Med 2017 Jan;74(1):72–9.
https://doi.org/10.1136/
oemed-2016-103726
14. Conway PM, Erlangsen A, Grynderup MB, Clausen T,
Scand J Work Environ Health – online first
References
1.
2.
World Health Organization. Preventing suicide: a global
imperative. Geneva: World Health Organization; 2014.
Naghavi M; Global Burden of Disease Self-Harm
Collaborators. Global, regional, and national burden of
suicide mortality 1990 to 2016: systematic analysis for
the Global Burden of Disease Study 2016. BMJ 2019
9
 BEU, Alm.del - 2021-22 - Bilag 285: Orientering om nye forskningsresultater om sammenhæng mellem mobning på arbejdspladsen og risiko for selvmord, fra beskæftigelsesministeren
2595048_0011.png
Workplace bullying and suicidal behavior
Bjørner J, Burr H, et al. Study Protocol: The association
between workplace bullying and suicidal behaviour: A
register-based prospective study of 98,330 participants in
Denmark 2020.
15. Erlangsen A, Fedyszyn I. Danish nationwide registers
for public health and health-related research. Scand
J Public Health 2015 Jun;43(4):333–9.
https://doi.
org/10.1177/1403494815575193.
16. Pedersen CB. The Danish Civil Registration System. Scand
J Public Health 2011 Jul;39(7 Suppl):22–5.
https://doi.
org/10.1177/1403494810387965.
17. Lynge E, Sandegaard JL, Rebolj M. The Danish
National Patient Register. Scand J Public Health
2011 Jul;39(7 Suppl):30–3.
PubMed https://doi.
org/10.1177/1403494811401482
18. Mors O, Perto GP, Mortensen PB. The Danish
Psychiatric Central Research Register. Scand J
Public Health 2011 Jul;39(7 Suppl):54–7.
https://doi.
org/10.1177/1403494810395825.
19. Helweg-Larsen K. The Danish Register of Causes of Death.
Scand J Public Health 2011 Jul;39(7 Suppl):26–9.
https://
doi.org/10.1177/1403494811399958.
20. Kildemoes HW, Sørensen HT, Hallas J. The Danish National
Prescription Registry. Scand J Public Health 2011 Jul;39(7
Suppl):38–41.
https://doi.org/10.1177/1403494810394717.
21. Williams RL. A note on robust variance estimation for
cluster-correlated data. Biometrics 2000 Jun;56(2):645–6.
https://doi.org/10.1111/j.0006-341X.2000.00645.x.
22. Magnusson Hanson LL, Nyberg A, Mittendorfer-Rutz E,
Bondestam F, Madsen IE. Work related sexual harassment
and risk of suicide and suicide attempts: prospective cohort
study. BMJ 2020 Sep;370:m2984.
https://doi.org/10.1136/
bmj.m2984.
23. Boudrias V, Trépanier SG, Salin D. A systematic
review of research on the longitudinal consequences of
workplace bullying and the mechanisms involved. Aggress
Violent Behav 2021;56:101508.
https://doi.org/10.1016/j.
avb.2020.101508.
24. Juvrud J, Rennels JL. “I Don’t Need Help”: Gender
Differences in how Gender Stereotypes Predict Help-
Seeking. Sex Roles 2017;76(1-2):27–39.
https://doi.
org/10.1007/s11199-016-0653-7.
25. Seidler ZE, Dawes AJ, Rice SM, Oliffe JL, Dhillon HM. The
role of masculinity in men’s help-seeking for depression: A
systematic review. Clin Psychol Rev 2016 Nov;49:106–18.
PubMed https://doi.org/10.1016/j.cpr.2016.09.002.
26. Wiley MG. Gender, Work, and Stress: The Potential
Impact of Role-Identity Salience and Commitment.
Sociol Q 1991;32(4):495–510.
https://doi.
org/10.1111/j.1533-8525.1991.tb00150.x.
27. Nielsen MB, Knardahl S. Is workplace bullying related to
the personality traits of victims? A two-year prospective
study. Work Stress 2015;29(2):128–49.
https://doi.org/10.1
080/02678373.2015.1032383.
28. Brezo J, Paris J, Turecki G. Personality traits as correlates
of suicidal ideation, suicide attempts, and suicide
completions: a systematic review. Acta Psychiatr Scand
2006 Mar;113(3):180–206.
PubMed https://doi.org/10.1111/
j.1600-0447.2005.00702.x.
29. Einarsen S, Hoel H, Zapf D, Cooper C. Empirical
Findings on Prevalence and Risk Groups of Bullying in
the Workplace. In: Einarsen S, Hoel H, Zapf D, Cooper C,
editors. Bullying and harassment in the workplace: Theory,
research and practice. 3rd ed. Boca Raton, FL: CRC Press
Taylor & Francis Group; 2020. p. 105–62.
30. Milner A, Spittal MJ, Pirkis J, LaMontagne AD. Suicide
by occupation: systematic review and meta-analysis.
Br J Psychiatry 2013 Dec;203(6):409–16.
https://doi.
org/10.1192/bjp.bp.113.128405.
31. Kendler KS, Hettema JM, Butera F, Gardner CO,
Prescott CA. Life event dimensions of loss, humiliation,
entrapment, and danger in the prediction of onsets of
major depression and generalized anxiety. Arch Gen
Psychiatry 2003 Aug;60(8):789–96.
https://doi.org/10.1001/
archpsyc.60.8.789.
32. Bi B, Xiao X, Zhang H, Gao J, Tao M, Niu H et al. A
comparison of the clinical characteristics of women with
recurrent major depression with and without suicidal
symptomatology. Psychol Med 2012 Dec;42(12):2591–8.
https://doi.org/10.1017/S003329171200058X.
33. Calati R, Ferrari C, Brittner M, Oasi O, Olié E, Carvalho AF
et al. Suicidal thoughts and behaviors and social isolation:
A narrative review of the literature. J Affect Disord 2019
Feb;245:653–67.
https://doi.org/10.1016/j.jad.2018.11.022.
34. Gold PW. The organization of the stress system and its
dysregulation in depressive illness. Mol Psychiatry 2015
Feb;20(1):32–47.
https://doi.org/10.1038/mp.2014.163.
35. Schwickerath J, Zapf D. Inpatient Psychotherapy of
Bullying Victims. In: Einarsen S, Hoel H, Zapf D, Cooper C,
editors. Bullying and harassment in the workplace: Theory,
research and practice. 3rd ed. Boca Raton, FL: CRC Press
Taylor & Francis Group; 2020. p. 593–626.
36. Reknes I, Einarsen S, Knardahl S, Lau B. The prospective
relationship between role stressors and new cases of
self-reported workplace bullying. Scand J Psychol 2014
Feb;55(1):45–52.
https://doi.org/10.1111/sjop.12092.
37. Zapf D, Vartia M. Prevention and Treatment of workplace
bullying. In: Einarsen S, Hoel H, Zapf D, Cooper C, editors.
Bullying and harassment in the workplace: Theory, research
and practice. 3rd ed. Boca Raton, FL: CRC Press Taylor &
Francis Group; 2020. p. 457–96.
Received for publication: 12 July 2021
10
Scand J Work Environ Health – online first