BEU, Alm.del - 2022-23 (2. samling) - Bilag 67: Orientering om nye forskningsresultater fra NFA

Beskæftigelsesudvalget 2022-23 (2. samling)
BEU Alm.del Bilag 67
Offentligt

Journal of Affective Disorders 319 (2022) 79–82

Contents lists available at

ScienceDirect

Journal of Affective Disorders

journal homepage:

www.elsevier.com/locate/jad

Short communication

Workplace discrimination and onset of depressive disorders in the Danish

workforce: A prospective study

Thomas Clausen

a, *

, Reiner Rugulies

a, b

, Jian Li

c, d

National Research Centre for the Working Environment, Copenhagen, Denmark

Department of Public Health, University of Copenhagen, Denmark

Department of Environmental Health Sciences, Fielding School of Public Health, University of California, Los Angeles, United States

School of Nursing, University of California, Los Angeles, United States

A R T I C L E I N F O

Keywords:

Longitudinal study

Mental health

Negative acts

Occupational health

Psychosocial work environment

Stress

A B S T R A C T

Objectives:

Experiences of discrimination at work are a long-standing problem, but research on its mental health

effect is sparse. The purpose of this study was to examine the prospective association between workplace

discrimination and onset of depressive disorders among Danish workers.

Methods:

The prospective cohort study comprised 2157 workers, all free of depressive disorders at baseline. Using

logistic regression models, we estimated the association between workplace discrimination at baseline and

depressive disorders at 6-months follow-up, adjusted for demographics, health behaviors, job group, educational

attainment and other psychosocial working conditions.

Results:

At baseline, 103 participants (4.8 %) reported workplace discrimination during the previous 12 months.

Among the 103 exposed participants and the remaining 2054 unexposed participants, onset of depressive dis-

orders during follow-up occurred in 16 (15.5 %) and 88 (4.3 %) participants, respectively. After adjustment for

all covariates, the odds ratio was 2.73 (95 % confidence interval: 1.38–5.40) comparing exposed to unexposed

participants.

Limitations:

All measures were self-reported, entailing risk of common methods bias, and we also cannot rule out

selection bias.

Conclusions:

Exposure to workplace discrimination is a risk factor for onset of depressive disorders. Eliminating or

reducing workplace discrimination may contribute to the prevention of depressive disorders in working

populations.

1. Introduction

Discrimination is defined as differential treatment of individuals due

to actual or perceived membership in particular groups (Williams

et al.,

1994),

such as sex, age, ethnicity, religion, health or sexual orientation.

Although evidence indicates that experiences of discrimination can

severely affect mental health (Vargas

et al., 2020),

only few studies have

examined the mental health-related consequences of discrimination in

the workplace. These studies found that workers reporting workplace

discrimination had lower levels of psychological well-being than

workers not reporting discrimination (Hammond

et al., 2010; Kim et al.,

2022; Lee et al., 2016; Marchiondo et al., 2019; Schütte et al., 2014).

Most of these studies, however, were based on cross-sectional designs,

severely limiting the possibility for causal inference and rendering the

studies vulnerable to common methods biases and inflated effect esti-

mates (Podsakoff

et al., 2003).

To the best of our knowledge, only one

study has prospectively investigated the association between workplace

discrimination and mental health (Marchiondo

et al., 2019).

This study

found that perceived age discrimination at work predicted elevated

depressive symptoms over time. However, the study from

Marchiondo

et al. (2019)

examined changes in depressive symptoms but not onset of

depressive disorders and, therefore, the present study contributes with

new knowledge by analyzing the prospective association between

workplace discrimination and onset of depressive disorders.

2. Methods

The study is based on a prospective cohort study (Clausen

et al.,

* Corresponding author at: National Research Centre for the Working Environment, Lersoe Parkalle 105, DK-2100 Copenhagen, Denmark.

E-mail address:

[email protected]

(T. Clausen).

https://doi.org/10.1016/j.jad.2022.09.036

Received 1 July 2022; Received in revised form 12 September 2022; Accepted 15 September 2022

Available online 17 September 2022

BEU, Alm.del - 2022-23 (2. samling) - Bilag 67: Orientering om nye forskningsresultater fra NFA

T. Clausen et al.

Journal of Affective Disorders 319 (2022) 79–82

2019).

The baseline questionnaire was distributed in the spring of 2015

to 8958 employees in Denmark. The study population was stratified in

14 job groups representing a variety of positions in the Danish labor

market. Procedures for sampling and questionnaire interviews are

described in detail elsewhere (Clausen

et al., 2019).

Of the 8958 invited individuals, 4340 responded (48.4 %). Six

months after baseline, we sent follow-up questionnaires to respondents

from the baseline study and obtained response from 2540 (58.5 %). We

excluded 194 respondents with missing values on key study variables

and 189 respondents with indications of prevalent depressive disorders

at baseline (MDI-score

≥21;

see next section), yielding a study sample of

2157 employees.

An analysis of non-response showed that women and older in-

dividuals were significantly more likely to participate in the baseline

study than men and younger individuals. An analysis of dropout from

baseline to follow-up showed no statistically significant differences

when comparing baseline levels of workplace discrimination and MDI-

scores for participants and non-participants at follow-up (Online sup-

plementary Table 1).

2.1. Workplace discrimination

To assess workplace discrimination, we asked participants this

question from the Danish Psychosocial Work Environment Question-

naire (DPQ) (Clausen

et al., 2019):

Have you within the last 12 months

experienced discrimination or been treated poorly in your workplace due to e.

g. your sex, age, ethnicity, religion, health or sexual orientation?

Response

options were: (1) Yes, daily or almost daily, (2) yes, weekly, (3) yes,

monthly, (4) yes, now and then, and (5) no, never. For the analyses, we

collapsed these response options into two categories: (1)

exposed

(daily

or almost daily, weekly, monthly, and now and then), and (2)

not

exposed

(never).

2.2. Onset of a depressive disorder

We measured prevalence (at baseline) and incidence (at follow-up)

of depressive disorders using the

Major Depression Inventory

(MDI)

(Bech

et al., 2001; Olsen et al., 2003; Rugulies et al., 2012).

A detailed

description of the MDI has been published elsewhere (Bech

et al., 2001).

Briefly, the MDI consists of 12 items assessing the presence of depressive

symptoms during the last two weeks on a scale ranging from 0 (the

symptom has not been present at all) to 5 (the symptom has been present

all of the time). The MDI sum score ranges from 0 to 50 points, as for two

pairs of items only the higher score is considered. A clinical validation

study of the MDI had previously shown that an MDI-score

≥21

indicates

the presence of a depressive disorder (Bech

et al., 2015).

Consequently,

we used this cut-off point for defining the presence of depressive dis-

orders in our study.

2.3. Covariates

Since workers' mental health may also be influenced by other factors

in the psychosocial work environment than discrimination (such as job

demands, control, and social support) (Theorell

et al., 2015),

adjusted the analyses for the following measures from the DPQ (Clausen

et al., 2019):

Quantitative demands

(four items; sample item:

How often is

it the case that you do not have time to complete all your work tasks?;

Cronbach's

: 0.84),

Influence at work

(four items; sample item:

Is it

possible for you to make important decisions about your work?;

Cronbach's

: 0.87),

Social support from colleagues

(four items; sample item:

Can you

talk to your colleagues about it if you experience difficulties at work?;

Cronbach's

: 0.82), and

Leadership quality

(four items; sample item:

your immediate supervisor good at motivating the employees?;

Cronbach's

0.91). Data on age, sex and job group were retrieved from national

registers. We collected data on cohabitation and educational attainment

in the study questionnaire.

2.4. Statistical analysis

Using logistic regression analysis, we calculated odds ratios (OR) and

95 % confidence intervals (95 % CI) to estimate the association between

workplace discrimination reported at baseline and risk of onset of

depressive disorders at follow-up. We calculated crude estimates and

adjusted for covariates in three models (see

Table 1).

In model 1, we

adjusted associations for age (as a continuous variable), sex, job group,

and educational attainment. In model 2, we additionally adjusted for

smoking, cohabitation with partner, and cohabitation with children, and

in model 3, we additionally adjusted for psychosocial working condi-

tions (quantitative demands, influence at work, social support from

colleagues, and leadership quality). To assess the robustness of the es-

timates, we conducted a sensitivity analysis, where we excluded re-

spondents with a baseline MDI-score

≥15,

i.e. baseline scores that were

close to the cut-off point for defining a depressive disorder (MDI-score

≥21).

Data were analyzed using the LOGISTIC procedure in SAS 9.4 (SAS

Inc., Cary, US).

3. Results

Table 1

shows descriptive statistics for the main study variables. The

baseline population consisted of 53 % men and 47 % women. The mean

age was 48 years (SD

11.0). Exposure to workplace discrimination was

reported by 103 participants (4.8 %) and was more often reported

among women (6.0 %) than among men (3.7 %). We did not observe

obvious differences in the prevalence of self-reported discrimination

across age groups, educational level, or job groups.

Table 1

also shows

that participants reporting workplace discrimination reported higher

quantitative demands, lower influence at work, and poorer relations to

colleagues and managers than participants who did not report work-

place discrimination.

Of the 2157 participants, 103 had a new depressive disorder at

follow-up (4.8 %).

Table 2

shows the estimates for the association be-

tween workplace discrimination and risk of onset of depressive disor-

ders. The cumulative incidence rates of depressive disorders were 15.5

% and 4.3 % among exposed and non-exposed workers, respectively. In

the crude analysis and in Models 1 and 2, odds ratios for incident

depressive disorders were about 4, when comparing respondents

reporting workplace discrimination at baseline with the unexposed

reference group. After additionally adjusting for other psychosocial

working conditions in Model 3, the odds ratio attenuated to 2.73 (95 %

CI: 1.38–5.40).

When we excluded participants with a baseline MDI-score

≥15

points in a sensitivity analysis, the association between workplace

discrimination and onset of depressive disorders became stronger (On-

line supplementary Table 2, Model 3: OR: 4.53; 95 % CI: 1.91–10.76).

4. Discussion

In this prospective cohort study, self-reported exposure to workplace

discrimination predicted onset of depressive disorders in a population

free of depressive disorders at baseline. The strength of the association

between the predictor and the outcome attenuated when we adjusted for

psychosocial working conditions, suggesting that factors in the psy-

chosocial work environment may play a role in the association between

workplace discrimination and onset of depressive disorders as either

instigators, mediators, or moderators

–

an issue that warrants further

research. However, even after adjustment for other psychosocial work

environment factors, the estimate for the association between workplace

discrimination and onset of depressive disorders remained considerably

and statistically significant, both in the main analysis and in the sensi-

tivity analysis, suggesting that workplace discrimination is an important

predictor of onset of depressive disorders.

Following the perspective of

Semmer et al. (2007),

acts of workplace

BEU, Alm.del - 2022-23 (2. samling) - Bilag 67: Orientering om nye forskningsresultater fra NFA

T. Clausen et al.

Journal of Affective Disorders 319 (2022) 79–82

Table 1

Descriptive statistics for main study variables.

Workplace discrimination at

baseline

Yes

Workplace discrimination at baseline (% (n))

Depressive disorders (MDI-score

21) at

follow-up (% (n))

Sex

Female (% (n))

Male (% (n))

Age (mean (SD))

18–34 years (% (n))

35–44 years (% (n))

45–54 years (% (n))

55 years or older (% (n))

Psychosocial working conditions

Quantitative demands (mean (SD))

Influence at work (mean (SD))

Social support from colleagues (mean

(SD))

Leadership quality (mean (SD))

Job group

Office workers (% (n))

Technical draughtsmen (% (n))

Teaching and research staff in universities

(% (n))

Health care helpers (% (n))

Primary school teachers (% (n))

Medical doctors (% (n))

Mail carriers (% (n))

Slaughterhouse workers (% (n))

Smith workers (% (n))

Engineers (construction) (% (n))

Sales assistants in shops (% (n))

Private bankers (% (n))

Business managers (% (n))

Police officers (% (n))

Educational level

Low (primary, lower secondary) (% (n))

Middle-low (upper and post-secondary) (%

(n))

Middle-high (first stage tertiary) (% (n))

High (second stage tertiary) (% (n))

Cohabitation with partner

Yes

Cohabitation with children

4.8 (103)

15.5 (16)

6.0 (61)

3.7 (42)

48.3

(12.4)

5.9 (17)

4.1 (18)

3.9 (27)

5.6 (41)

56.4

(18.1)

52.3

(22.0)

59.8

(20.0)

46.5

(23.5)

8.0 (12)

4.0 (7)

5.6 (8)

4.7 (5)

6.0 (11)

5.7 (9)

8.1 (10)

5.9 (8)

–

(<5

)

2.4 (5)

4.4 (5)

5.7 (10)

–

(<5

)

3.8 (7)

6.6 (14)

3.4 (17)

5.3 (51)

4.1 (20)

4.3 (72)

6.6 (27)

95.2

(2054)

4.3 (88)

94.0

(953)

96.3

(1101)

48.3

(11.0)

94.1

(269)

95.9

(418)

96.2

(674)

94.4

(693)

49.0

(19.3)

66.2

(20.3)

71.4

(16.8)

58.9

(22.3)

92.0

(139)

96.0

(170)

94.4

(135)

95.3

(101)

94.0

(173)

94.3

(150)

91.9

(113)

94.1

(127)

96.8

(119)

97.6

(201)

95.6

(108)

94.3

(165)

98.9

(177)

96.2

(176)

93.4

(197)

96.6

(480)

94.7

(905)

95.9

(465)

95.8

(1621)

93.4

(384)

Table 1

(continued )

Workplace discrimination at

baseline

Yes

<0.0001

0.0109

96.3

(1000)

94.3

(1005)

3.8 (39)

5.7 (61)

0.9767

We excluded respondents with depressive disorders at baseline from the

analyses.

The four measures of psychosocial working conditions are measured on a

scale ranging from 0 to 100, with a score of 100 indicating the highest level of

the measured dimension.

Results for groups smaller than five persons cannot be reported due to data

protection regulations.

0.0002

<0.0001

0.1887

discrimination may be considered offenses to the ‘self’ of the targeted

individuals. Accordingly, such offenses to the self may have a negative

impact on the self-worth of the targeted employees and in cases of

extended exposure to acts of workplace discrimination, the exposure

may consequently have an adverse impact on the mental health of the

targets. Indeed, in discussing their prospective results,

Marchiondo et al.

(2019)

conclude that extended exposure to age discrimination may

overwhelm the ability of exposed workers to cope successfully with

these exposures, which ultimately may lead to adverse health-related

outcomes, such as depressive disorders.

The results are in accordance with previous cross-sectional studies

(Hammond

et al., 2010; Kim et al., 2022; Lee et al., 2016; Schütte et al.,

2014)

and one prospective study (Marchiondo

et al., 2019)

reporting

that exposure to workplace discrimination is associated with lower

levels of psychological well-being. To our knowledge, the present study

is the first one providing prospective evidence on the association be-

tween workplace discrimination and onset of depressive disorders.

Accordingly, the present study meets the call for research on the asso-

ciation between discrimination and health (Williams

et al., 2008),

and

the call for longitudinal studies on the association between discrimina-

tion and mental health (Vargas

et al., 2020).

Finally, other studies showed that the

global burden of disease

to mental health issues is on the rise (GBD

2019 Mental Disorders Col-

laborators, 2022).

Efforts to reduce the prevalence of depressive disor-

ders may, therefore, take its' point of departure in combatting work-

related acts of offensive behavior in general and acts of workplace

discrimination in particular.

4.1. Limitations

It is a weakness of the study that it was not possible to identify the

precise type of discrimination that the respondents reported being

exposed to (e.g., discrimination due to age, sex, or ethnicity). This limits

the possibilities for developing targeted interventions on the basis of the

present study and, hence, limits the practical applicability of the study

results. It is also a limitation of the study that all variables were based on

self-reported measures as this entails risk of common method bias.

Further, the use of a self-reported measure of depressive disorders

lowers the validity of the outcome measurement, compared to a clinical

diagnostic interview. Although we cannot rule out that the observed

association between workplace discrimination and onset of depressive

disorders may be inflated because of common methods biases (Podsakoff

et al., 2003),

it is likely that the prospective design of the study has

reduced such biases in the present study. The low response rate in the

baseline study and the attrition from baseline to follow-up may consti-

tute a source of selection bias, and the analysis of non-response at

baseline showed that women and older individuals were more likely to

participate in the baseline study. However, we found no differences

between participants and non-participants at follow-up in their baseline-

levels of workplace discrimination or MDI-score. Finally, the follow-up

0.1914

0.0466

0.0338

BEU, Alm.del - 2022-23 (2. samling) - Bilag 67: Orientering om nye forskningsresultater fra NFA

T. Clausen et al.

Journal of Affective Disorders 319 (2022) 79–82

Table 2

Odds ratios (OR) and 95 % confidence intervals (95 % CI) for the association between workplace discrimination at baseline and onset of a depressive disorder during six

months follow-up (n

2157).

At risk

Cases n/%

Risk of onset of depressive disorder after six months of follow-up

Crude model

Workplace discrimination at baseline

Model 1

3.94

95 % CI

2.13 to 7.30

Reference

Model 2

3.80

95 % CI

1.99 to 7.28

Reference

Model 3

2.73

95 % CI

1.38 to 5.40

Reference

95 % CI

2.31 to 7.30

Reference

Yes

103

2054

16/15.5

88/4.3

4.11

Crude model: Unadjusted.

Model 1: Adjusted for age, sex, job group, and educational attainment.

Model 2: Model 1 plus smoking, cohabitation with partner, cohabitation with children.

Model 3: Model 2 plus psychosocial working conditions: quantitative demands, influence at work, social support from colleagues, and leadership quality.

period was only six months long, which is short for studying the onset of

a depressive disorder. It is possible that most of the new cases at follow-

up occurred among participants who had already substantial mental

health problems at baseline, raising concerns about inflated estimates.

However, when we excluded participants with MDI-scores of 15 to 20

points at baseline, i.e. scores close to the threshold for identifying

depressive disorders (MDI

≥

21 points), the association between base-

line discrimination and onset of depressive disorders during follow-up

did not become weaker but actually stronger. Thus, the association be-

tween baseline discrimination and onset of depressive disorders during

follow-up was not driven by individuals who were close to a prevalent

depressive disorder at baseline.

5. Conclusions

The results of this study indicate that workplace discrimination

constitutes a serious stressor with potentially damaging effects on the

mental health of exposed individuals. Eliminating or reducing work-

place discrimination may contribute to prevent cases of depression in

the working population.

CRediT authorship contribution statement

All authors contributed to the design of the study. T. Clausen per-

formed the data analysis and wrote the first draft of the manuscript. All

authors have contributed significantly to the preparation of the manu-

script and to the interpretation of the results. All authors have approved

the final version of the manuscript.

Conflict of interest

The authors have no conflicts of interest to disclose.

Acknowledgements

The work was supported by a grant from the Danish Working Envi-

ronment Authority Research Fund (grant no. 10-2019-03). The funder

had no influence on the conceptualization of the paper, data collection,

interpretation of results, writing of the paper, and the decision to submit

for publication.

Appendix A. Supplementary data

Supplementary data to this article can be found online at

https://doi.

org/10.1016/j.jad.2022.09.036.

References

Bech, P., Rasmussen, N.A., Olsen, L.R., Noerholm, V., Abildgaard, W., 2001. The

sensitivity and specificity of the Major Depression Inventory, using the Present State

Examination as the index of diagnostic validity. J. Affect. Disord. 66 (2–3), 159–164.

Bech, P., Timmerby, N., Martiny, K., Lunde, M., Soendergaard, S., 2015. Psychometric

evaluation of the Major Depression Inventory (MDI) as depression severity scale

using the LEAD (Longitudinal Expert Assessment of All Data) as index of validity.

BMC Psychiatry 15, 190.

https://doi.org/10.1186/s12888-015-0529-3.

Clausen, T., et al., 2019. The Danish Psychosocial Work Environment Questionnaire

(DPQ): development, content, reliability and validity. Scand. J. Work Environ.

Health 45 (4), 356–369.

https://doi.org/10.5271/sjweh.3793.

GBD 2019 Mental Disorders Collaborators, 2022. Global, regional, and national burden

of 12 mental disorders in 204 countries and territories, 1990-2019: a systematic

analysis for the Global Burden of Disease Study 2019. Lancet Psychiatry 9 (2),

137–150.

https://doi.org/10.1016/s2215-0366(21)00395-3.

Hammond, W.P., Gillen, M., Yen, I.H., 2010. Workplace discrimination and depressive

symptoms: a study of multi-ethnic hospital employees. Race Soc. Probl. 2 (1), 19–30.

https://doi.org/10.1007/s12552-010-9024-0.

Kim, S., et al., 2022. Gender discrimination in workplace and depressive symptoms in

female employees in South Korea. J. Affect. Disord. 306, 269–275.

https://doi.org/

10.1016/j.jad.2022.03.050.

Lee, B.J., Lamichhane, D.K., Jung, D.Y., Moon, S.H., Kim, S.J., Kim, H.C., 2016.

Psychosocial factors and psychological well-being: a study from a nationally

representative sample of Korean workers. Ind. Health 54 (3), 237–245.

https://doi.

org/10.2486/indhealth.2015-0191.

Marchiondo, L.A., Gonzales, E., Williams, L.J., 2019. Trajectories of perceived workplace

age discrimination and long-term associations with mental, self-rated, and

occupational health. J. Gerontol. B Psychol. Sci. Soc. Sci. 74 (4), 655–663.

https://

doi.org/10.1093/geronb/gbx095.

Olsen, L.R., Jensen, D.V., Noerholm, V., Martiny, K., Bech, P., 2003. The internal and

external validity of the major depression inventory in measuring severity of

depressive states. Psychol. Med. 33 (2), 351–356.

Podsakoff, P.M., MacKenzie, S.B., Lee, J.Y., Podsakoff, N.P., 2003. Common method

biases in behavioral research: a critical review of the literature and recommended

remedies. J. Appl. Psychol. 88 (5), 879–903.

Rugulies, R., et al., 2012. Bullying at work and onset of a major depressive episode

among Danish female eldercare workers. Scand. J. Work Environ. Health 38 (3),

218–227.

Schütte, S., Chastang, J.F., Malard, L., Parent-Thirion, A., Vermeylen, G.,

Niedhammer, I., 2014. Psychosocial working conditions and psychological well-

being among employees in 34 European countries. Int. Arch. Occup. Environ. Health

87 (8), 897–907.

https://doi.org/10.1007/s00420-014-0930-0.

Semmer, N., Jacobshagen, N., Meier, L., Elfering, A., 2007. Occupational stress research:

the "Stress-as-offense-to-self" perspective. In: Houdmont, J., McIntyre, S. (Eds.),

Occupational Health Psychology. Nottingham University Press, Nottingham,

pp. 43–59.

Theorell, T., et al., 2015. A systematic review including meta-analysis of work

environment and depressive symptoms. BMC Public Health 15 (738), 738.

https://

doi.org/10.1186/s12889-015-1954-4.

Vargas, S.M., Huey, S.J., Miranda, J., 2020. A critical review of current evidence on

multiple types of discrimination and mental health. Am. J. Orthopsychiatry 90 (3),

374–390.

https://doi.org/10.1037/ort0000441.

Williams, D.R., Lavizzo-Mourey, R., Warren, R.C., 1994. The concept of race and health

status in America. Public Health Rep. 109 (1), 26–41 (Washington, DC: 1974).

Williams, D.R., Neighbors, H.W., Jackson, J.S., 2008. Racial/ethnic discrimination and

health: findings from community studies. Am. J. Public Health 98 (9 Suppl),

S29–S37.

https://doi.org/10.2105/ajph.98.supplement_1.s29.