PLOS ONE

RESEARCH ARTICLE

The negative impact of long working hours on

mental health in young Korean workers
Sungjin Park1☯, Hyungdon Kook ID2☯, Hongdeok Seok3, Jae Hyoung Lee4, Daeun Lim5,
Dong-Hyuk Cho2, Suk-Kyu Oh ID2*
1 Department of Occupational and Environmental Medicine, Cheonan Medical Center, Cheonan, Korea,
2 Division of Cardiology, Department of Internal Medicine, Korea University Medical Center, Seoul, Korea,
3 Department of Occupational and Environmental Medicine, Busan Adventist Hospital, Sahmyook Medical
Center, Busan, Korea, 4 Medical School for International Health, Ben Gurion University, Beersheba, Israel,
5 Department of Biological Sciences, University of Seoul, Seoul, Korea
a1111111111
☯ These authors contributed equally to this work.
a1111111111 * ohsukkyu@hanmail.net
a1111111111
a1111111111
a1111111111
Abstract
Long working hours are known to have a negative effect on health. However, there is no
clear evidence for a direct link between mental health and long working hours in the young
OPEN ACCESS adult populations. Therefore, we aimed to determine whether long working hours are associ-
Citation: Park S, Kook H, Seok H, Lee JH, Lim D, ated with mental health in young adult workers. Data were collected from a 2012 follow-up
Cho D-H, et al. (2020) The negative impact of long survey of the Youth Panel 2007. A total of 3,332 young adult employees (aged 20 to 35)
working hours on mental health in young Korean
were enrolled in the study. We analyzed stress, depression, and suicidal thoughts by multi-
workers. PLoS ONE 15(8): e0236931. https://doi.
org/10.1371/journal.pone.0236931 variate logistic regression analysis based on working hours (41 to 50, 51 to 60 and over 60
hours, compared to 31 to 40 hours per week), which was adjusted for sex, age, marriage
Editor: Sergio A. Useche, Universitat de Valencia,
SPAIN status, region, and educational level. From the 3,332 young adult employees, about 60% of
the workers worked more than 40 hours and 17% of the workers worked more than 50 hours
Received: July 29, 2019
per week. In a Chi-square test, stress level, depression, and suicidal thoughts increased
Accepted: July 17, 2020
with increasing working hours (p-value <0.001, 0.007, and 0.018, respectively). The multi-
Published: August 4, 2020 variate logistic regression model showed that, compared to the 31 to 40 hours per week
Copyright: © 2020 Park et al. This is an open group, the adjusted odds ratios of the 41 to 50, 51 to 60, and over 60 hours per week groups
access article distributed under the terms of the for stress were 1.46(1.23–1.74), 2.25(1.79–2.83) and 2.55(1.72–3.77), respectively. A simi-
Creative Commons Attribution License, which
lar trend was shown in depression [odds ratios: 2.08(1.23–3.53), 2.79(1.44–5.39) and 4.09
permits unrestricted use, distribution, and
reproduction in any medium, provided the original (1.59–10.55), respectively] and suicidal ideation [odds ratios: 1.98(0.95–4.10), 3.48(1.48–
author and source are credited. 8.19) and 5.30(1.61–17.42), respectively]. We concluded that long working hours were
Data Availability Statement: All relevant data are associated with stress, depression, and suicidal ideation in young employees, aged 20 to
within the manuscript and its Supporting 35.
Information files. Youth Panel data is open data. All
the original data could obtained from https://
survey.keis.or.kr without restriction. We attached
Supporting Information files that statistically
analyzed the original data.

Funding: The authors received no specific funding Introduction

for this work. Long working hours are known to affect health negatively. One of the reasons is the unhealthy
Competing interests: The authors have declared behaviors associated with working overtime, such as increased alcohol consumption and lack
that no competing interests exist. of exercise [1, 2]. In addition, employees working long hours may not have the time to seek

PLOS ONE | https://doi.org/10.1371/journal.pone.0236931 August 4, 2020 1 / 13

PLOS ONE Long working hours and mental health

proper medical treatment when they fall ill [3]. Furthermore, working long hours may induce
hypertension, diabetes, and metabolic syndrome and is closely linked to ischemic heart dis-
ease, stroke, and increased mortality [2, 4–7].
In addition to the effects on physical health, adverse effects on mental health were observed
in employees working long hours. An excessive workload increases workers’ fatigue and
thereby negatively affects the subjective perceptions of health [8, 9]. In addition, anxiety and
depression are more common in the group working long hours [10].
Negative impact on mental health is manifested in many ways, including stress, depression,
and suicidal ideation. These are important issues not only for the individual but also for the
society. Depression may occur in people who are frequently exposed to stress, leading to devel-
opment of disease and poor quality of life, and may eventually induce serious suicidal
thoughts. In turn, suicidal impulses have a significant impact on the quality of mental and
physical health and can contribute greatly to the worldwide disease burden [11].
Many countries aim to limit the working hours of their employees, as it is known that lon-
ger working hours have a negative impact on health [12]. Modern society tends to emphasize
the quality of life, since people hope for stable employment with enough time for rest and lei-
sure rather than simply earning a lot of money [13]. On the other hand, the time for leisure
and rest are limited for employees with long working hours, which ultimately puts stress on
individuals and may have a negative impact on their mental health [14–16].
According to the data obtained by the Organization for Economic Co-operation and Devel-
opment (OECD), the average annual working hours in OECD countries has decreased from
1,881 hours in 1990 to 1,734 hours in 2018. These data also showed that the average annual
working hours in South Korea decreased from 2,677 hours in 1990 to 1,993 hours in 2018, but
this number is still high compared to that reported for other OECD countries [17]. As of 2018,
South Korea has the third highest number of annual working hours among OECD countries
following Mexico and Costa Rica. Compared with the 1,363 hours reported for Germany, the
OECD country with the lowest number of annual working hours, South Korea has 1.5 times
longer working hours.
Although negative effects of long working hours on the health of workers are already
known, not many studies have been conducted to establish the association between long work-
ing hours and mental health with specific parameters such as depression and suicidal ideation,
especially in Korea. We therefore aimed to investigate the impact of an excessive workload on
mental health using the Youth Panel and its follow-up surveys.
In geriatric groups, it is hard to attribute a health problem to an independent factor because
there are a lot of pre-existing medical conditions [18, 19]. We specifically investigated youth,
because we could associate mental health with working hours as an independent variable. In addi-
tion, in Korea, the young adult population is the group that works the longest [20]. Therefore, this
group is the most suitable for finding out the link between long working hours and mental health.
We examined the differences in stress levels, depression, and suicidal thoughts in the young
adult population with respect to mental health problems resulting from long working hours.

Methods
Study population
We did a cross-sectional analysis of the data from the Youth Panel (hereafter “YP”) 2007. YP
was conducted by Korea Employment Information Service (KEIS, https://survey.keis.or.kr/
eng/yp/yp01.jsp). This panel survey was aimed to collect basic data on school life, social and
economic activities, and household background of young people to contribute to the develop-
ment of employment policies and related research in decreasing youth unemployment. The

PLOS ONE | https://doi.org/10.1371/journal.pone.0236931 August 4, 2020 2 / 13

PLOS ONE Long working hours and mental health

YP was a longitudinal survey initially conducted in 2007 with annual follow-up surveys until
2012. It was the first individual panel survey conducted on the Korean population. The YP
2007 surveyed 10,206 young people aged between 15 and 29 by establishing a nation-wide
sample of young residents. Of the initial survey participants, 91.2% completed the second sur-
vey (first follow-up) in 2008; 86.5%, the third survey (second follow-up) in 2009; 81.7%, the
fourth survey (third follow-up) in 2010; 78.9%, the fifth survey (fourth follow-up) in 2011; and
76.8%, the sixth survey (fifth follow-up) in 2012. In the sixth YP survey, 7,843 participants
remained, of which 7,057 actually completed the survey as some participants did not respond
to the whole questionnaire.
These surveys were conducted via interview visits using Computer-Assisted Personal Inter-
view (CAPI). In addition, some questionnaires were also answered online (Computer-Assisted
Web Interview, CAWI) or by telephone (Computer-Assisted Telephone Interview, CATI) in
cases where the respondents rejected the survey method or had difficulty coming in for a per-
sonal interview. Of the total respondents, 94.6% used CAPI, 3.6% used CAWI, and 1.8% used
CATI. The interview period lasted from August 6, 2012 to November 30, 2012, but for some
responders, information was gathered in early December.

Study design
This study was conducted on a sample of participants from the YP 2007. From the 10,206 indi-
viduals who took part in the initial survey, only 7,057 participants remained as 3,149 did not
respond in 2012. Of the 7,057 remaining participants, 5 high school students, 2,064 college,
university, or graduate students and 1,258 unemployed individuals were excluded. From this,
a further 398 participants were excluded from the study: 43 employers, 226 self-employed, 28
unpaid family workers, 4 people who did not respond to the working hours question, and 97
participants who worked less than 30 hours per week and were not considered to be full-time
employees in Korea [3, 4]. Finally, 3,332 participants were selected for the study (Fig 1). We
excluded part-time workers because of their irregular working time, time-shifting and part-
time working status itself on Korean social background could affect mental health [21]. Institu-
tional Review Board of Korea University Medical Center Anam Hospital specially approved
this study (IRB number: 2019-AN-0204).

Working hours
Data on working hours were obtained from answers to economic activity questions listed in
Part B of the sixth YP. This part of the survey contained questions on working hours, working
days, and wages. For example, question B44 was as follows: "How many hours do you work on
average per week in your current job?". Possible answers were "regular working hours per
week: ___ hours" and "average overtime hours per week: ___ hours". The average working
hours per week were calculated as the sum of regular working hours and average overtime
hours per week. The average working hours per week were divided into increments of 10
hours, yielding four groups, namely “31 to 40 hours per week”, “41 to 50 hours per week”, “51
to 60 hours per week”, and “more than 60 hours per week” [3, 4].

Mental health
Part H of the sixth YP was evaluated to characterize mental health issues. Question H22 was
"How often do you feel stressed in your daily life?", with possible answers of “very often",
“often", “sometimes", and " rarely". Answering "very often" and "often" was considered to indi-
cate a high stress level, while participants answering "sometimes" and "rarely" were considered
not feeling very stressed.

PLOS ONE | https://doi.org/10.1371/journal.pone.0236931 August 4, 2020 3 / 13

PLOS ONE Long working hours and mental health

Fig 1. Study population & design.

https://doi.org/10.1371/journal.pone.0236931.g001

Depression was assessed in question H23: "Have you ever felt sadness or feeling of despair
that interfered with your daily life for two consecutive weeks over the past year?", with possible
answers being “yes” or “no.” Participants answering “yes” to this question were considered
depressed, while those answering “no” were considered not depressed.
Suicidal ideation was identified in question H24: "Have you ever thought about wanting to
die in the last year?", with participants answering “yes” classified as those with suicidal
thoughts and ones answering “no” as those without suicidal thoughts.

Covariates
Both men and women participants between the ages of 20 to 35 years were recruited. Marriage
status was assessed as married, unmarried, or divorced. Residential areas were classified as spe-
cial cities, metropolitan cities, or other provinces. The category assessing the highest level of
education was divided into high school graduation or lower and college degree or higher.

PLOS ONE | https://doi.org/10.1371/journal.pone.0236931 August 4, 2020 4 / 13

PLOS ONE Long working hours and mental health

Statistical analysis
We conducted a chi-square test for general characteristics according to working hours, stress,
depression, and suicidal ideation. We analyzed stress, depression, and suicidal ideation by
multivariate logistic regression analysis generating an odds ratio (OR) and 95% confidence
interval (CI) for those working between 41 hours and 50 hours, between 51 and 60 hours, and
over 60 hours per week relative to those working between 31 and 40 hours with crude model
and adjusted model with associated covariate such as sex, marriage, region, and educational
level.
SAS software version 9.2 (SAS Inc., Cary, NC, USA) was used for statistical analysis and a
p-value below 0.05 was considered to be statistically significant.

Results
General characteristics of participants relative to working hours
In the sixth YP, 1,339 (40.2%) workers responded to working between 31 and 40 hours per
week, 1,425 (42.8%) between 41 and 50 hours, 450 (13.5%) between 51 and 60 hours, and 118
(3.5%) more than 60 hours (Table 1, S1 and S2 Tables).
In terms of stress levels, 23.0% of employees working between 31 and 40 hours per week
responded that they felt a lot of stress, 30.3% of those working between 41 and 50 hours, 39.8%
of those working between 51 to 60 hours, and 42.4% of those working more than 60 hours per
week reported stress. Therefore, the proportion of workers who felt a lot of stress increased
with increasing working hours.
Of those participants working between 31 and 40 hours per week, 1.6% had self-reported
depression, while 3.1% of those working between 41 and 50 hours reported that they felt
depressed, 3.8% of those working between 51 to 60 hours, and 5.1% of those working more
than 60 hours per week reported depression. Hence, an increase in weekly working hours was
associated with higher levels of depression.
A similar trend was observed for participants with suicidal thoughts over the past year
(0.8% of those working between 31 and 40 hours, 1.5% of those working between 41 and 50
hours, 2.4% of those working between 51 to 60 hours, and 3.4% of those working more than 60
hours per week).

Table 1. General characteristics of participants relative to working hours.

Working hours, n (%)
Total 31–40 41–50 51–60 Over 60 p-value
Gender Male 1629 569 (34.9) 702 (43.1) 275 (16.9) 83 (5.1) <0.001
Female 1703 770 (45.2) 723 (42.4) 175 (10.3) 35 (2.1)
Marriage status Married 707 290 (41.0) 296 (41.9) 99 (14.0) 22 (3.1) 0.812
Unmarried or divorced 2625 1,049 (40.0) 1,129 (43.0) 351 (13.4) 96 (3.6)
Residential area Special or metropolitan city 2028 805 (39.7) 867 (42.7) 285 (14.1) 71 (3.5) 0.685
Other province 1304 534 (40.9) 558 (42.8) 165 (12.7) 47 (3.6)
Educational Level High school graduation or below 701 232 (33.1) 273 (38.9) 145 (20.7) 51 (7.3) <0.001
College degree or above 2631 1,107 (42.1) 1,152 (43.8) 305 (11.6) 67 (2.5)
Stress level High 969 308 (31.8) 432 (44.6) 179 (18.5) 50 (5.1) <0.001
Low 2363 1,031 (43.6) 993 (42.0) 271 (11.5) 68 (2.9)
Depression Present 88 21 (23.9) 44 (50.0) 17 (19.3) 6 (6.8) 0.007
Absent 3244 1,318 (40.6) 1,381 (42.6) 433 (13.3) 112 (3.5)
Suicidal thoughts Present 48 11 (22.9) 22 (45.9) 11 (22.9) 4 (8.3) 0.018
Absent 3284 1,328 (40.4) 1,403 (42.7) 439 (13.4) 114 (3.5)
https://doi.org/10.1371/journal.pone.0236931.t001

PLOS ONE | https://doi.org/10.1371/journal.pone.0236931 August 4, 2020 5 / 13

PLOS ONE Long working hours and mental health

When analyzed by gender, 569 (42.5%) men and 770 (57.5%) women worked between 31
hours and 40 hours, 702 (49.3%) men and 723 (50.7%) women between 41 hours and 50
hours, 275 (61.1%) men and 175 (38.9%) women between 51 hours and 60 hours, and 83
(70.3%) men and 35 (29.7%) women more than 60 hours per week. Hence, the long working
hours led to a higher proportion of men working compared to women.
In terms of the educational level, of the participants working between 31 and 40 hours, 232
(17.3%) were high school graduates or had a lower level of education and 1,107 (82.7%) had a
college degree or a higher level of education, While this proportion was 273 (19.2%) to 1,152
(80.8%) in the group working between 41 hours and 50 hours. In the group of those working
between 51 hours and 60 hours, 145 (32.2%) had a lower education level and 305 (67.8%) had
a higher education level. While this proportion was 51 (43.2%) to 67 (56.8%) amongst those
working more than 60 hours per week. Therefore, longer working hours led to a higher pro-
portion of those with a lower education level.

General characteristics of participants in terms of mental health

In the sixth follow-up survey of the YP 2007, 969 (29.1%) participants responded that they felt
very stressed and 2,363 (70.9%) said they did not feel stressed. When analyzed by gender,
29.3% of males and 28.9% of females felt high levels of stress. (Table 2, S3 and S4 Tables)
Eighty-eight (2.6%) of the interviewed employees reported that they felt depressed and
3,244 (97.4%) did not. Of all participants, 1.9% of men and 3.3% of women were depressed
(p = 0.013).
Forty-eight (1.4%) participants answered that they had suicidal thoughts, and 3,284 (98.6%)
responded that they did not. Of all participants, 0.9% of males and 2.0% of females had suicidal
ideation, showing a statistically significant difference (p = 0.009).
There is no significant difference in stress, depression and suicidal ideation in terms of edu-
cational level, residential area, and marriage status.
Putting stress levels in relation to depression, only 27.7% of respondents who were not
depressed felt very stressed. In contrast, 80.7% of depressed participants also felt very stressed
(p<0.001). Most participants without suicidal ideation did not feel very stressed (28.3%), while
the majority of participants with suicidal thoughts (83.3%) had high self-reported stress levels
(p<0.001)

Table 2. General characteristics of participants in terms of mental health.

Stress level, n (%) Depression, n (%) Suicidal ideation, n (%)
High Low p-value Present Absent p-value Present Absent p-value
Gender Male 477 (49.2) 1,152 (48.8) 0.833 31 (35.2) 1,598 (49.3) 0.013 14 (29.2) 1,615 (49.2) 0.009
Female 492 (50.8) 1,211 (51.2) 57 (64.8) 1,646 (50.7) 34 (70.8) 1,669 (50.8)
Marriage status Married 204 (21.1) 503 (21.3) 0.918 15 (17.0) 692 (21.3) 0.402 5 (10.4) 702 (21.4) 0.096
Unmarried or divorced 765 (78.9) 1,860 (78.7) 73 (83.0) 2,552 (78.7) 43 (89.6) 2,582 (78.6)
Residential area Special or metropolitan city 613 (63.3) 1,415 (59.9) 0.076 58 (65.9) 1,970 (60.7) 0.383 32 (66.7) 1,996 (60.8) 0.500
Other province 356 (36.7) 948 (40.1) 30 (34.1) 1,274 (39.3) 16 (33.3) 1,288 (39.2)
Educational High school graduation or 208 (21.5) 493 (20.9) 0.734 19 (21.6) 682 (21.0) 1.000 12 (25.0) 689 (21.0) 0.617
level below
College degree or above 761 (78.5) 1,870 (79.1) 69 (78.4) 2,562 (79.0) 36 (75.0) 2,595 (79.0)
Working hours 31 to 40 308 (31.8) 1,031 (43.6) <0.001 21 (23.9) 1,318 (40.6) 0.007 11 (22.9) 1,328 (40.4) 0.018
41 to 50 432 (44.6) 993 (42.0) 44 (50.0) 1,381 (42.6) 22 (45.9) 1,403 (42.7)
51 to 60 179 (18.5) 271 (11.5) 17 (19.3) 433 (13.3) 11 (22.9) 439 (13.4)
Over 60 50 (5.1) 68 (2.9) 6 (6.8) 112 (3.5) 4 (8.3) 114 (3.5)
https://doi.org/10.1371/journal.pone.0236931.t002

PLOS ONE | https://doi.org/10.1371/journal.pone.0236931 August 4, 2020 6 / 13

PLOS ONE Long working hours and mental health

Putting self-reported depression in relation to stress levels, only 0.7% of the participants
who did not feel stressed felt depressed, but of those reporting high stress levels, 7.3% also felt
depressed (p<0.001). Only 1.8% of participants without suicidal thoughts were depressed. In
contrast, 58.3% of participants with suicidal ideation had depression (p<0.001).
In terms of stress levels, only 0.3% of those not feeling stressed had suicidal thoughts, but
4.1% of those with high stress levels stated that they thought of suicide (p<0.001). Only 0.6%
of the employees without depression responded that they had suicidal thoughts. In contrast,
31.8% of participants with depression had suicidal ideation (p<0.001).

Relationship between working hours and mental health

In the crude model, we determined the OR (95% CI) of experiencing high stress levels in terms
of working hours. The OR was 1.46 (95% CI: 1.23–1.73) for employees working between 41
and 50 hours per week, 2.21 (95% CI: 1.76–2.78) for those working between 51 and 60 hours,
and 2.46 (95% CI: 1.67–3.62) for those working over 60 hours, relative to those working
between 31 and 40 hours per week (Table 3 and Fig 2).
The OR (95% CIs) for the depression in terms of working hours were 2.00 (95% CI: 1.18–
3.38) for employees working between 41 and 50 hours per week, 2.46 (95% CI: 1.29–4.71) for
those working between 51 and 60 hours, and 3.36 (95% CI: 1.33–8.50) for those working more
than 60 hours per week, relative to employees working between 31 and 40 hours per week.
In suicidal ideation, the OR 1.89 (95% CI: 0.91–3.92) was obtained in crude model for the
employees working between 41 and 50 hours per week, 3.03 (95% CI: 1.30–7.03) for those
working between 51 and 60 hours per week, and 4.24 (95% CI: 1.33–13.52) for those working
more than 60 hours per week, relative to employees working between 31 and 40 hours.
In adjusted model, which was adjusted for gender, marriage status, region, and educational
level, we determined the following ORs (95% CI) of experiencing high-stress level: 1.46 (95%
CI: 1.23–1.74) for employees working between 41 and 50 hours per week, 2.24 (95% CI: 1.78–
2.83) for those working between 51 and 60 hours, and 2.55 (95% CI: 1.72–3.77) for those work-
ing over 60 hours, relative to those working between 31 and 40 hours.
Using adjusted model, the OR (95% CI) for the depression was 2.09 (95% CI: 1.23–3.54) for
those working between 41 to 50 hours per week, 2.77 (95% CI: 1.43–5.35) for those working
between 51 and 60 hours, and 4.11 (95% CI: 1.54–10.60) for those working over 60 hours a
week, relative to employees working between 31 and 40 hours per week.
In analysis of suicidal ideation with adjusted model, the OR for employees working between
41 and 50 hours per week was 1.97 (95% CI: 0.95–4.09), 3.49 (95% CI: 1.48–8.23) for those
working between 51 and 60 hours, and 5.28 (95% CI: 1.61–17.36) for those working over 60
hours per week, relative to those working between 31 and 40 hours per week.

Table 3. The association between working hours and mental health.

Stress Depression Suicidal ideation
Odd Ratio (95% Confidence Interval) Odd Ratio (95% Confidence Interval) Odd Ratio (95% Confidence Interval)
Crude Adjusted model Crude Adjusted model Crude Adjusted Model
Working Hours 31–40 1.00 1.00 1.00 1.00 1.00 1.00
41–50 1.46 (1.23–1.73) 1.46 (1.23–1.74) 2.00 (1.18–3.38) 2.08 (1.23–3.53) 1.89 (0.91–3.92) 1.98 (0.95–4.10)
51–60 2.21 (1.76–2.78) 2.25 (1.79–2.83) 2.46 (1.29–4.71) 2.79 (1.44–5.39) 3.03 (1.30–7.03) 3.48 (1.48–8.19)
>60 2.46 (1.67–3.62) 2.55 (1.72–3.77) 3.36 (1.33–8.50) 4.09 (1.59–10.55) 4.24 (1.33–13.52) 5.30 (1.61–17.42)
�
Adjusted model 1: 1 adjusted for gender, marriage status, residential area, and educational level.

https://doi.org/10.1371/journal.pone.0236931.t003

PLOS ONE | https://doi.org/10.1371/journal.pone.0236931 August 4, 2020 7 / 13

PLOS ONE Long working hours and mental health

Fig 2. Relationship between working hours and mental health.

https://doi.org/10.1371/journal.pone.0236931.g002

PLOS ONE | https://doi.org/10.1371/journal.pone.0236931 August 4, 2020 8 / 13

PLOS ONE Long working hours and mental health

Discussion
In this study, we investigated the association between working hours and mental health. We
revealed that as the working hours increase, the risk of stress, depression and suicide ideation
tends to increase. All of these three mental health parameters had a linear dose response pat-
tern correlating with increasing working hours.
In assessing the general characteristics of study participants in terms of their respective
working hours, we discovered that the proportion of male employees increased with increasing
working hours. This may reflect the fact that women typically take on multiple roles including
housework and cannot commit as many hours to their employment as their male counterparts.
Similar results were obtained in previous studies [22].
The increase in working hours was proportional to the stress levels in a dose-response man-
ner, in that more work led to higher observed stress levels. These results remained the same
after adjusting for sex, age, marriage status, region, and educational level. We observed that an
increase in working hours, particularly those that were unintentional or unwanted, leads to
higher stress levels in the employees. Stress in turn leads to unhealthy behaviors [23–25] and is
known as a link to many diseases [26]. High stress levels induced by long working hours are
known to influence unhealthy behaviors such as lack of exercise, higher smoking rate, and
increased alcohol consumption [23–25]. In addition, stress not only affects the subjective
health status and sleep duration but also impairs mental health, leading to depression or even
suicidal thoughts [27, 28]. Stress due to long working hours is one of the major causes of health
deterioration.
Significant differences were observed in the depression between sexes, in agreement with
previous studies [29]. As mentioned above, depression has been shown to be related to subjec-
tive health status, sleep duration, and stress [27, 28]. A dose-response relationship was
observed between depression and working hours in that employees with longer working hours
increasingly reported feeling depressed. These results did not change after adjusting for sex,
age, marriage status, region, and educational level. In agreement with previous studies, we
determined that extended working hours, particularly those that are unintentional or
unwanted, result in more depression [14, 15].
Women tended to have suicidal ideation more frequently than men, which corroborates
the results of previous studies [30]. Suicidal ideation is known to be linked to subjective health
status, sleep duration, stress, and depression [31]. Suicidal ideation was related to working
hours in a dose-response manner, as the rate of depression increased with increasing working
hours. These results were maintained after adjustment for gender, age, marriage status, region,
and educational level. We conclude that extended working hours cause stress, which in turn
may lead to depression or even suicide. These results are consistent with previous studies
[14, 16].
Previous studies showed the association of depression and anxiety with working hours.
Shields et al. showed that depression was increased in women who worked over 40 hours per
week [32]. Workers were divided into only two groups based on 40 hours per week work
hours. Therefore, that study could not show a dose-response pattern of working hours and
increasing mental health problems. Virtanen et al. showed that depression and anxiety were
increased in employees working over 55 hours per week compared with the 35 to 40 hours
group [10]. However, working hours were divided into 3 groups and employees who worked
41 to 55 hours per week did not show significant difference in both depression and anxiety.
Kim et al. also showed a similar result that employees who worked over 60 hours per week had
more depressive symptoms than those working less than 52 hours [14]. However, there was no
significant difference in the 52 to 60 hours per week group. Our study divided the working

PLOS ONE | https://doi.org/10.1371/journal.pone.0236931 August 4, 2020 9 / 13

PLOS ONE Long working hours and mental health

hours into more groups, which allows for a more significant conclusion that reducing working
hours leads to reduced depression.
There were also investigations done on the relationship between suicidal ideation and
working hours. Kim et al. showed that suicidal ideation was more frequent in over 60 working
hours per week group than in the 40 to 51 working hours group [33]. Interestingly, this study
pointed out that less than 40 working hours per week group had numerically more frequent
suicidal thoughts compared to the 40 to 51 working hours group. Yoon et al also demonstrated
that the risk of suicidal ideation increased in over 60 working hours per week group compared
with less than 52 working hours group [16]. However, this study also failed to prove that more
working hours leads to more frequent suicidal ideation. Compared with the other studies, our
study showed a dose-response relationship between working hours and suicidal ideation. In
our study, it was confirmed that about 60% of the employees work more than 40 hours a week,
and 17% work more than 50 hours a week in the Korean young adult population. As the pro-
portion of working overtime was high, working time could be stratified into four groups con-
firming the quantitative relationship between working hours and mental health. In previous
studies of Korean employees, working hours were divided in a similar way and the result was
analyzed by stratified groups [3, 4].
Many hypotheses have been put forward to explain the negative effects of extended working
hours on physical and mental health [15]. Long hours at work necessitate sufficient recovery
from fatigue, but the amount of time spent at home is limited [34]. This in turn leads to insuffi-
cient sleep. Even after adjusting the leisure time of individual workers, long working hours
deteriorate the quality and duration of sleep [35]. Insufficient sleep is known to be associated
with mental illnesses such as depression and anxiety and makes recovery impossible with det-
rimental effects on physical health [27, 36, 37].
In terms of suicidal ideation, same detrimental effect could be applied. Long working hours
not only increase physical, psychological, and emotional demand of recovery, but also decrease
the recovery time. [2, 34] Working hours also affect the quality and duration of sleep [35].
Therefore, long working hours aggravate anxiety, depression and burnout and, as a result,
long working hours increase suicidal ideation [16].
Another hypothesis to explain the negative effects of long working hours on health incorpo-
rates the imbalance models of demand and adjustment and of effort and compensation [38,
39]. Long hours at work reflect an existing demand for work, but an individual worker cannot
adjust such working environments or situations. If there is no adequate compensation for
overtime work, stress and depression may ensue [15].
The strength of our study lies in the fact that it is a large-scale epidemiological study with
over 3,300 wage workers based on panel surveys. In addition, participants of the YP are aged
between 20 and 35 and are comparatively similar to each other, thereby reducing errors caused
by age differences. Furthermore, we based our evaluation on the YP and therefore had the
advantage of identifying the influence of independent variables on health in a particular age
group. With increasing age, it becomes more difficult to attribute health problems to precise
independent variables because of increasing incidences of underlying chronic diseases. In our
study, all participants were 35 years old or younger and we could investigate mental health
with working hours as the independent variable. In addition, we were able to evaluate all three
typical indicators of mental health, namely stress, depression, and suicidal thoughts, which is
more meaningful than analyzing only one indicator. Another strength of our study is the divi-
sion of workload per week into four groups (31 hours to 40 hours, 41 hours to 50 hours, 51
hours to 60 hours, and over 60 hours work per week), rendering results that identified dose-
response relationships.

PLOS ONE | https://doi.org/10.1371/journal.pone.0236931 August 4, 2020 10 / 13

PLOS ONE Long working hours and mental health

Nevertheless, this study has some limitations. This study was based on the YP 2007 but was
designed as a cross-sectional study assessing only the sixth follow-up survey of the panel. This
cross-sectional approach has the disadvantage of making a precise causal relationship difficult
to confirm. Furthermore, the independent variables of this study were obtained with a ques-
tionnaire and were therefore self-reported, which may have caused recall bias or reporting
bias. Nevertheless, it is challenging to accurately calculate the actual working hours, and most
similar studies have used individual questionnaires as well [3, 14, 16]. The three indicators of
mental health were also self-reported reflecting the subjective information on the individual’s
health status and did not include sufficient objective information. Nevertheless, we can over-
come these limitations by focusing on variables such as subjective health and mental health.
Another limitation is that we performed statistical analysis with adjustment to gender rather
than stratification of gender, even though there were significant differences of mental prob-
lems in each gender. Finally, in our study, the prevalence of depression and suicidal ideation
was 2.6% and 1.4%, respectively. This low percentage could render certain outcomes unstable.

Conclusion
Our study has shown that an extensive weekly workload has a negative impact on mental
health. We discovered that the longer working hours, the higher level of stress and higher prev-
alence of depression and suicidal ideation. All the three mental health parameter showed dose
response pattern with working hours. We concluded that long working hours were associated
with stress, depression, and suicidal ideation in young employees, aged 20 to 35.

Supporting information
S1 File.
(ZIP)
S1 Table. General characteristics of participants relative to working hours in males.
(DOCX)
S2 Table. General characteristics of participants relative to working hours in females.
(DOCX)
S3 Table. General characteristics of participants relative to mental health in males.
(DOCX)
S4 Table. General characteristics of participants relative to mental health in female.
(DOCX)

Acknowledgments
The author thanks all of the officers who collected the YP Data.

Author Contributions
Conceptualization: Sungjin Park.
Formal analysis: Hyungdon Kook.
Investigation: Hongdeok Seok.
Methodology: Hyungdon Kook, Dong-Hyuk Cho.
Software: Hongdeok Seok.

PLOS ONE | https://doi.org/10.1371/journal.pone.0236931 August 4, 2020 11 / 13

PLOS ONE Long working hours and mental health

Supervision: Suk-Kyu Oh.

Validation: Hongdeok Seok, Suk-Kyu Oh.
Visualization: Dong-Hyuk Cho.
Writing – original draft: Sungjin Park, Hyungdon Kook.
Writing – review & editing: Jae Hyoung Lee, Daeun Lim, Suk-Kyu Oh.

References
1. Virtanen M, Jokela M, Nyberg ST, Madsen IE, Lallukka T, Ahola Ket al. Long working hours and alcohol
use: systematic review and meta-analysis of published studies and unpublished individual participant
data. BMJ. 2015; 350:g7772. https://doi.org/10.1136/bmj.g7772 PMID: 25587065
2. Bannai A, Tamakoshi A. The association between long working hours and health: a systematic review
of epidemiological evidence. Scand J Work Environ Health 2014; 40(1):5–18. https://doi.org/10.5271/
sjweh.3388 PMID: 24100465
3. Seok H, Won JU, Lee TI, Kim YK, Lee W, Lee JHet al. A dose-response relationship between long work-
ing hours and unmet need for access to hospital facilities. Scand J Work Environ Health 2016:135–143.
https://doi.org/10.5271/sjweh.3551 PMID: 26862879
4. Yoo D, Kang M, Paek D, Min B, Cho S. Effect of Long Working Hours on Self-reported Hypertension
among Middle-aged and Older Wage Workers. Ann Occup Environ Med. 2014; 26:25. https://doi.org/
10.1186/s40557-014-0025-0 PMID: 25852938
5. Kivimäki M, Jokela M, Nyberg ST, Singh-Manous A, Fransson EI, Alfredsson Let al. Long working
hours and risk of coronary heart disease and stroke: a systematic review and meta-analysis of pub-
lished and unpublished data for 603 838 individuals. The Lancet. 2015; 386(10005):1739–1746.
6. Kobayashi T, Suzuki E, Takao S, Doi H. Long working hours and metabolic syndrome among Japanese
men: a cross-sectional study. BMC Public Health. 2012; 12(1):395.
7. Virtanen M, Heikkila K, Jokela M, Ferrie JE, Batty GD, Vahtera J et al. Long Working Hours and Coro-
nary Heart Disease: A Systematic Review and Meta-Analysis. Am J Epidemiol. 2012; 176(7):586–596.
https://doi.org/10.1093/aje/kws139 PMID: 22952309
8. Park J, Kim Y, Chung HK, Hisanaga N. Long working hours and subjective fatigue symptoms. Ind
Health. 2001; 39(3):250–254. https://doi.org/10.2486/indhealth.39.250 PMID: 11500001
9. Song JT, Lee G, Kwon J, Park KW, Choi H, Lim S. The association between long working hours and
self-rated health. Ann Occup Environ Med. 2014; 26(1):2. https://doi.org/10.1186/2052-4374-26-2
PMID: 24472333
10. Virtanen M, Ferrie JE, Singh-Manoux A, Shipley MJ, Stansfeld SA, Marmot MGet al. Long working
hours and symptoms of anxiety and depression: a 5-year follow-up of the Whitehall II study. Psychol
Med. 2011; 41(12):2485–2494. https://doi.org/10.1017/S0033291711000171 PMID: 21329557
11. Fairweather-Schmidt A, Batterham P, Butterworth P, Nada-Raja S. The impact of suicidality on health-
related quality of life: A latent growth curve analysis of community-based data. J Affect Disord. 2016;
203:14–21. https://doi.org/10.1016/j.jad.2016.05.067 PMID: 27285722
12. Lee S., McCann D., & Messenger J.C. 2007. Working time around the world: Trends in working hours,
laws and policies in a global comparative perspective. Geneva, Switzerland: International Labor
Organization.
13. Lee KH, Kim JE, Kim YK, Kang DM, Yun MJ, Park SG et al. Long working hours and emotional well-
being in Korean manufacturing industry employees. Ann Occup Environ Med. 2013; 25(1):38. https://
doi.org/10.1186/2052-4374-25-38 PMID: 24472652
14. Kim I, Kim H, Lim S, Lee M, Bahk J, June KJet al. Working hours and depressive symptomatology
among full-time employees: Results from the fourth Korean National Health and Nutrition Examination
Survey (2007–2009). Scand J Work Environ Health. 2013; 39(5):515–520. https://doi.org/10.5271/
sjweh.3356 PMID: 23503616
15. Bannai A, Ukawa S, Tamakoshi A. Long working hours and psychological distress among school teach-
ers in Japan. J Occup Health. 2015; 57:20–27. https://doi.org/10.1539/joh.14-0127-OA PMID:
25422128
16. Yoon JH, Jung PK, Roh J, Seok H, Won JU. Relationship between Long Working Hours and Suicidal
Thoughts: Nationwide Data from the 4th and 5th Korean National Health and Nutrition Examination Sur-
vey. PLoS One. 2015; 10(6): e0129142. https://doi.org/10.1371/journal.pone.0129142 PMID:
26080022

PLOS ONE | https://doi.org/10.1371/journal.pone.0236931 August 4, 2020 12 / 13

PLOS ONE Long working hours and mental health

17. OECD, Average annual hours actually worked per worker. https://stats.oecd.org/index.aspx?
DataSetCode=ANHRS, [Accessed: 10th June, 2020]
18. Park J-I, Park TW, Yang J-C, Chung S-K. Factors associated with depression among elderly Koreans:
the role of chronic illness, subjective health status, and cognitive impairment. Psychogeriatrics 2016;
16(1): 62–69 https://doi.org/10.1111/psyg.12160 PMID: 26450373
19. Rapp SR, Parisi SA, Wallace CE. Comorbid psychiatric disorders in elderly medical patients: a 1-year
prospective study. J Am Geriatr Soc 1991; 39: 124–31. https://doi.org/10.1111/j.1532-5415.1991.
tb01613.x PMID: 1991943
20. Korean Statistical Information Service (KOSIS), Wage and working hours by age. https://kosis.kr/
statHtml/statHtml.do?orgId=118&tblId=DT_118N_LCE0004 [Accessed: 20 Apr 2020]
21. Mousteri V, Daly M, Delaney L. Underemployment and psychological distress: Propensity score and
fixed effects estimates from two large UK samples. Soc Sci Med. 2020; 244:112641 https://doi.org/10.
1016/j.socscimed.2019.112641 PMID: 31706192
22. Grosch JW, Caruso CC, Rosa RR, Sauter SL. Long hours of work in the US: associations with demo-
graphic and organizational characteristics, psychosocial working conditions, and health. Am J Ind Med.
2006; 49(11):943–952. https://doi.org/10.1002/ajim.20388 PMID: 17036350
23. Kouvonen A, Kivimäki M, Virtanen M, Pentti J, Vahtera J. Work stress, smoking status, and smoking
intensity: an observational study of 46 190 employees. J Epidemiol Community Health. 2005; 59(1):63–
69. https://doi.org/10.1136/jech.2004.019752 PMID: 15598729
24. Lipton R. The relationship between alcohol, stress, and depression in Mexican Americans and non-His-
panic whites. Behav Med. 1997; 23(3):101–111. https://doi.org/10.1080/08964289709596366 PMID:
9397282
25. Norris R, Carroll D, Cochrane R. The effects of physical activity and exercise training on psychological
stress and well-being in an adolescent population. J Psychosom Res. 1992; 36(1):55–65. https://doi.
org/10.1016/0022-3999(92)90114-h PMID: 1538350
26. De Kloet ER, Joëls M, Holsboer F. Stress and the brain: from adaptation to disease. Nat Rev Neurosci.
2005; 6(6):463. https://doi.org/10.1038/nrn1683 PMID: 15891777
27. Kaneita Y, Ohida T, Uchiyama M, Takemura S, Kawahara K, Yokoyama Eet al. The relationship
between depression and sleep disturbances: a Japanese nationwide general population survey. J Clin
Psychiatry. 2006; 67:196–203. https://doi.org/10.4088/jcp.v67n0204 PMID: 16566613
28. Niti M, Ng TP, Kua EH, Ho RCM, Tan CH. Depression and chronic medical illnesses in Asian older
adults: the role of subjective health and functional status. Int J Geriatr Psychiatry. 2007; 22(11):1087–
1094. https://doi.org/10.1002/gps.1789 PMID: 17407107
29. Stapley JC, Haviland JM. Beyond depression: Gender differences in normal adolescents’ emotional
experiences. Sex role 1989; 20(5–6):295–308.
30. Galaif ER, Chou CP, Sussman S, Dent CW. Depression, suicidal ideation, and substance use among
continuation high school students. J Youth Adolesc. 1998; 27(3):275–299.
31. Wunderlich U, Bronisch T, Wittchen HU, Carter R. Gender differences in adolescents and young adults
with suicidal behaviour. Acta Psychiatr Scand. 2001; 104(5):332–339. https://doi.org/10.1034/j.1600-
0447.2001.00432.x PMID: 11722313
32. Shields M. Long working hours and health. Health Rep. 1999; 11(2):33–48. PMID: 10618741
33. Kim KU, Park SG, Kim HC, Lim JH, Lee SJ, Jeon SH, et al. Association between long working hours
and suicidal ideation. 2012; 24(4):339–46.
34. Jansen N, Kant I, van Amelsvoort L, Nijhuis F, van den Brandt P. Need for recovery from work: evaluat-
ing short-term effects of working hours, patterns and schedules. Ergonomics. 2003; 46(7):664–680.
https://doi.org/10.1080/0014013031000085662 PMID: 12745680
35. Nakashima M, Morikawa Y, Sakurai M, Nakamura K, Miura K, Ishizaki Met al. Association between long
working hours and sleep problems in white-collar workers. J Sleep Res. 2011 Mar; 20(1 Pt 1):110–6
https://doi.org/10.1111/j.1365-2869.2010.00852.x PMID: 20561174
36. Nakata A. Work hours, sleep sufficiency, and prevalence of depression among full-time employees: a
community-based cross-sectional study. J Clin Psychiatry. 2011; 72(5):605–614. https://doi.org/10.
4088/JCP.10m06397gry PMID: 21658347
37. Ikegami K, Ogyu S, Arakomo Y, Suzuki K, Mafune K, Hiro Het al. Recovery of cognitive performance
and fatigue after one night of sleep deprivation. J Occup Health. 2009; 51(5):412–422. https://doi.org/
10.1539/joh.l8127 PMID: 19602843
38. Karasek RA Jr. Job demands, job decision latitude, and mental strain: Implications for job redesign.
Administrative Science Quarterly. 1979:285–308.
39. Siegrist J. Adverse health effects of high-effort/low-reward conditions. J Occup Health Psychol. 1996; 1
(1):27. https://doi.org/10.1037//1076-8998.1.1.27 PMID: 9547031

PLOS ONE | https://doi.org/10.1371/journal.pone.0236931 August 4, 2020 13 / 13