AMERICAN JOURNAL OF INDUSTRIAL MEDICINE 50:470–480 (2007)

Job Stress and Depressive Symptoms

Among Japanese Fire Fighters

1
2 3
Yasuaki Saijo, MD, PhD, Takeji Ueno, MD, PhD, and Yoshihiro Hashimoto, BA

Background Associations between job stresses, as assessed by theoretical job stress

model and depressive symptoms among fire fighters have not been fully investigated. The
purpose of this study is to clarify the factors of job stress that influence the depressive
symptoms in Japanese fire fighters.
Methods The subjects involved 1,672 fire fighters from a local government. The
questionnaire comprised age, gender, job type, job class, martial status, smoking, and
drinking habit, the Center for Epidemiologic Studies Depression Scale (CES-D), and The
National Institute for Occupational Safety and Health (NIOSH) generic job questionnaire.
Results A group showing depressive symptoms (CES-D  16) included 373 subjects
(22.3%). In a multivariate logistic regression analysis, high variance in workload, high
intergroup conflict, high role conflict, and low self-esteem had significantly higher odds
ratio for depressive symptoms.
Conclusions High variance in workload, high intergroup conflict, high role conflict,
and low self-esteem were significantly related to depressive symptoms among Japanese fire
fighters. Further prospective studies are needed to clarify the influence of these stress
factors on other health outcomes, and to elucidate whether alleviation of these
stress factors improve the mental health among fire fighters. Am. J. Ind. Med. 50:470–
480, 2007. ß 2007 Wiley-Liss, Inc.

KEY WORDS: job stress; depressive symptoms; fire fighters; variance in workload;
intergroup conflict; role conflict; self-esteem

mental health of professional workers should be well studied.

INTRODUCTION The work of Japanese fire fighters includes various
responsibilities, such as firefighting, emergency services,
The importance of work for mental health has been and rescue work. These emergency workers have a 24-hr
widely acknowledged. In particular, factors affecting the duty. After working day-duty, they must work 2-hr shifts
at night in order to handle unexpected emergency calls.
This night shift system, widely adopted at most Japanese
1
Department of Health Science, Asahikwa Medical College, Midorigaoka, E2-1-1-1, fire stations, can affect their health status [Takeyama
Asahikawa, Hokkaido, Japan et al., 2005]. Additionally, the number of calls for
2
Department of Health Sciences, Hokkaido University School of Medicine, Hokkaido, ambulances has recently been increasing, as Japan is
Japan
3
Sapporo Fire Bureau, Japan prone to earthquakes, typhoons, intensive rain, or volcanic
Contract grant sponsor: Ministry of Education, Culture, Sports, Science, and Technology of eruptions leading to disaster thereby, increasing the burden of
Japan; Contract grant number: 18590585. fire fighters.
*Correspondence to: Yasuaki Saijo, Department of Health Science, Asahikawa Medical
College, Midorigaoka, E2-1-1-1, Asahikawa, Hokkaido 078-8510, Japan. Firefighting may be related cardiovascular disorder
E-mail: y-saijo@asahikawa-med.ac.jp [Melius, 1995]. It has been reported that workload, physical
demand, and exposure to fire smoke during their emergency
Accepted11February 2007
DOI 10.1002/ajim.20460. Published online in Wiley InterScience affect the health of fire fighters [Dutton et al., 1978; Guidotti,
(www.interscience.wiley.com) 1992; Booze et al., 2004; Bos et al., 2004]. Symptoms of post-

2007 Wiley-Liss, Inc.

 Stress and Depression Among Fire Fighters 471

traumatic disorder have been reported among fire fighters in ‘‘Battalion Chief.’’ Job type was grouped into ‘‘firefighting,’’
other countries as well [Wagner et al., 1998; Heinrichs et al., ‘‘ambulance work,’’ ‘‘rescue work,’’ ‘‘other shift work’’
2005]. including fire inspector and command work, and ‘‘daytime
Job stresses estimated by theoretical job stress model work’’ including general affairs and fire prevention. The
questionnaires relate to depressive symptoms [Tsutsumi subjects classified in ‘‘firefighting,’’ ‘‘ambulance work,’’
et al., 2001; Pikhart et al., 2004; Godin et al., 2005]. However, ‘‘rescue work,’’ and ‘‘other shift work’’ engaged in a 24-hr
the associations between job stresses, as assessed by shift work (from 8:45 to 8:55 the next day), and rest was
theoretical job stress model and depressive symptoms among 4 days per 2 weeks.
fire fighters have not been fully investigated. The purpose of This study was approved by the Institutional Ethical
this study is to clarify the factors of job stress, as estimated by Board for Epidemiological Studies of Aahikawa Medical
The National Institute for Occupational Safety and Health School.
(NIOSH) generic job questionnaire, that influence the
depressive symptoms of Japanese fire fighters. Statistical Analysis

Students’ t-test or analysis of variance (ANOVA) was

MATERIAL AND METHODS performed to compare the mean of each measure of the
NIOSH generic job stress questionnaire and the CES-D score
The subjects included 1,767 fire fighters aged 18– across the categories of background factors.
60 years ranked at section chief or lower from a local Logistic regression analyses were used to estimate the
government. The self-administered questionnaires were univariate ORs of age, gender, job class, job type, martial
distributed to the subjects from September through October status, smoking and drinking habit, and each measure of the
2005. However, answers to the questionnaire were obtained NIOSH generic job stress questionnaire for depressive
from 1,731 fire officers (response rate: 98%). symptoms (CES-D  16). Each measure, except for non-
A total of 65 subjects were excluded for the following work activity of the NIOSH generic job stress questionnaire,
reasons: loan employee from a city office (n ¼ 9) and missing was tertilized. Since non-work activity had a narrow
information (n ¼ 50). The final study group consisted of distribution, it was dichotomized. Subsequently, multivariate
1,672 subjects. ORs for depressive symptoms including all above-mentioned
The questionnaire was composed of age, gender, job independent variables were estimated.
type, job class, martial status, smoking and drinking habit, P-values less than 0.05 were considered to be statisti-
the Japanese version of the Center for Epidemiologic Studies cally significant. All analyses were conducted using the SPSS
Depression Scale (CES-D) [Radloff, 1977; Shima et al., software package Version 14.0 for Windows (SPSS, Inc.,
1985], and the Japanese version of the NIOSH generic job Chicago, IL).
questionnaire [Hurrell and McLaney, 1988; Haratani et al.,
1993, 1996]. The Japanese version of the NIOSH generic job
questionnaire showed consistently high level of internal RESULTS
reliability (Cronbachs alpha, 0.68–0.95), and test–retest
reliability over 1 year (r, 0.44–0.71) [Haratani et al., 1993, Table I shows the characteristics of the subjects and the
1996]. From the NIOSH generic job stress questionnaire, the mean scores of the NIOSH generic job stress questionnaire.
following measures were chosen: quantitative workload, A group showing depressive symptoms (CES-D  16)
variance in workload, cognitive demand, job control, included 373 subjects (22.3%).
intragroup conflict, intergroup conflict, role ambiguity, role Table II shows the comparison of the each measure of the
conflict, social support from supervisor, social support from NIOSH generic job stress questionnaire and the CES-D score
coworker, social support from family/friend, non-work according to individual factors. Between males and females,
activity, and self-esteem. The standard CES-D Scale uses a CES-D score, cognitive demand, and self-esteem were
cutoff score of 16 points for depressive symptoms [Radloff, significantly different. Among the age categories, all
1977; Shima et al., 1985]. measures of the NIOSH generic job stress questionnaire
Age was categorized as <30, 30–39, 40–49, and except for cognitive demand were significantly different.
>50 years. The subjects were classified by smoking habits Between married and single subjects, job control, role
into ‘‘non-smoker’’ including never- and ex-smokers, and ambiguity, social support from coworkers, social support
‘‘current smoker’’ groups. Drinkers were defined as those from family/friends, non-work activity, self-esteem, and
with a weekly consumption of alcohol or more. Martial CES-D were significantly different. Between ex-/non-
status was grouped into ‘‘unmarried’’ and ‘‘married.’’ Job smokers and current smokers, role conflict was significantly
class included ‘‘Fire Fighter,’’ ‘‘Assistant Fire Sergeant,’’ different. Between non-drinkers and drinkers, cognitive
‘‘Fire Sergeant,’’ ‘‘Fire Lieutenant,’’ ‘‘Fire Captain,’’ and demand, non-work activity, and self-esteem were signifi-
472 Saijo et al.

TABLE I. Characteristics of the Subjects (Japanese Fire Fighters) except for job control and non-work activity were signifi-
cantly associated with depressive symptoms. In the multi-
Male 1,626 (97.2%)
variate logistic regression analysis, high variance in
Age (years) 43.6  10.4
workload (vs. low: OR ¼ 2.05, 95% CI ¼ 1.29–3.25), high
<30 221 (13.2%)
intergroup conflict (vs. low: OR ¼ 1.91, 95% CI ¼ 1.26–
30^39 320 (19.1%)
2.88), high role conflict (vs. low: OR ¼ 1.87, 95%
40^49 501 (30.0%)
CI ¼ 1.24–2.80), and low self-esteem (vs. high: OR ¼ 5.78,
>50 630 (37.7%)
95% CI ¼ 3.93–8.50) had significantly higher ORs for
Martial status
depressive symptoms. Middle job control had a significantly
Married 1,407 (84.2%)
lower OR for depressive symptoms (vs. low: OR ¼ 0.64, 95%
Current smoker 958 (57.3%)
CI ¼ 0.43–0.94).
Drinker 1,248 (74.6%)
Job class
DISCUSSION
Fire fighter 541 (32.4%)
Assistant Fire Sergeant 11 (0.7%)
This study documents significant relationships between
Fire Sergeant 515 (30.8%)
high variance in workload, high intergroup conflict, high role
Fire Lieutenant 374 (22.4%)
conflict, and low self-esteem and depressive symptoms
Fire captain 186 (11.1%)
among Japanese fire fighters. To our knowledge, however,
Battalion Chief 45 (2.7%)
there has been no report about stress among fire fighters
Type of job
estimated by the NIOSH generic job stress questionnaire.
Firefighting 868 (51.9%)
Of the 1,672 fire fighters 372 (22.3%) had depressive
Ambulance work 284 (17.0%)
symptoms (CES-D  16). In a recent Japanese survey
Rescue work 165 (9.9%)
involving 24,896 individuals, 25.6% of males aged 20–
Other shift work 87 (5.2%)
59 years and 29.5% of females aged 20–59 years had
Daytime work 268 (16.0%)
depressive symptoms (CES-D  16) [Kaneita et al., 2006]. A
Quantitative workload 33.7  6.4
study among 3,711 Koreans aged 20–79 reported that 23.1%
Variance in workload 8.3  3.1
of males and 27.4% of females had depressive symptoms
Cognitive demand 15.6  2.4
(CES-D  16) [Cho et al., 1998]. Thus, the CES-D score of
Job control 41.9  12.0
the subjects in this study does not vary significantly from
Intragroup conflict 18.5  5.2
the general Asian population. Meanwhile, in a study of
Intergroup conflict 19.5  4.1
Cincinnati fire fighters in 1985–1986 revealed that 33.1%
Role ambiguity 18.3  5.5
had depressive symptoms (CES-D  16), and the fire fighters
Role conflict 26.8  7.9
had poor mental health test scores compared with the general
Social support from supervisor 15.7  3.3
work population [Boxer and Wild, 1993]. Differences in
Social support from coworker 16.7  2.7
roles and ways of working between fire fighters in different
Social support from family/friend 16.8  3.0
countries could not allow direct comparison of the CES-D
Non-work activity 1.0  0.7
data.
Self-esteem 33.6  6.2
Variance in workload was significantly related to the
CES-D 16 373 (22.3%)
fire fighters’ depressive symptoms. A study of Japanese
Variables are presented as number (percentage) or mean  SD. workers (their job type was unknown) showed a significant
relationship between variance in workload and depressive
temperament [Sakai et al., 2005]. Recently, the workload
on Japanese fire official has increased, for example, by an
increasing number of calls for ambulances, and expectations
cantly different. Among the job class categories, all measures for the speedy resolution of disasters and bioterrorism.
of the NIOSH generic job stress questionnaire except for Thus, measures to improve variance in workload should be
cognitive demand, intra- and intergroup conflict were taken.
significantly different. Among the job type categories, all Intergroup conflict was significantly related to de-
measures of the NIOSH generic job stress questionnaire pressive symptoms, but intragroup conflict was not. Among
except for role conflict, social support from supervisor, non- job types, ambulance workers usually have the highest
work activity, and self-esteem were significantly different. emergency turnout [Takeyama et al., 2001]. In this study,
Table III shows the result of unvariate and multivariate daytime workers had a significantly higher OR for depressive
logistic regression analyses. In the univariate analyses, all symptoms (vs. other shift workers: OR ¼ 2.14. 95%
measures of the NIOSH generic job stress questionnaire CI ¼ 1.09–4.20). Initially, most fire fighters enjoy the
 TABLE II. Comparison of the NIOSH Generic Job Stress Questionnaire and CES-D Score According to Individual Factors Among the Japanese Fire Fighters

Quantitative workload Variance in workload Cognitive demand Job control

n Mean SD P-value Mean SD P-value Mean SD P-value Mean SD P-value

Gender
Male 1,626 33.7 6.4 0.67 8.3 3.1 0.80 15.7 2.4 0.00 42.0 12.0 0.30
Female 46 33.3 6.0 8.4 3.3 14.6 2.4 40.1 11.1
Age (years)
<30 221 33.4 5.6 <0.001 8.4 2.9 <0.001 15.5 2.3 0.80 36.1 10.7 <0.001
30^39 320 35.1 7.1 9.1 3.2 15.6 2.6 41.3 10.9
40^49 501 34.1 6.7 8.5 3.2 15.6 2.3 44.1 11.7
>50 630 32.7 5.9 7.7 2.9 15.7 2.3 42.6 12.5
Martial Status
Married 1,407 33.7 6.4 0.94 8.3 3.1 0.42 15.6 2.3 0.43 42.8 11.8 <0.001
Single 265 33.7 6.5 8.4 3.1 15.5 2.5 37.2 11.6
Smoking habit
Non-smoker 714 33.5 6.3 0.24 8.2 3.1 0.33 15.5 2.4 0.19 42.4 12.1 0.17
Current smoker 958 33.8 6.5 8.4 3.1 15.7 2.3 41.6 11.9
Drinking habit
Non-drinker 424 33.5 6.6 0.42 8.2 3.1 0.61 15.8 2.3 0.03 41.2 12.0 0.13
Drinker 1,248 33.8 6.3 8.3 3.1 15.6 2.4 42.2 12.0

473
Job class
Fire Fighter 541 32.3 5.4 <0.001 7.7 2.9 <0.001 15.5 2.3 0.32 35.8 10.2 <0.001
Assistant Fire Sergeant 11 27.6 5.0 6.4 2.5 15.1 1.9 41.5 9.3
Fire Sergeant 515 34.5 6.5 8.7 3.1 15.6 2.4 39.7 9.9
Fire Lieutenant 374 34.6 6.9 8.7 3.2 15.7 2.5 45.5 10.2
Fire Captain 186 34.4 7.1 8.5 3.2 15.9 2.4 55.5 10.3
Battalion Chief 45 32.5 4.8 7.6 2.3 16.0 2.1 55.4 8.6
Type of job
Firefighting 868 31.8 5.5 <0.001 7.4 2.7 <0.001 15.4 2.3 <0.001 41.3 12.2 <0.001
Ambulance work 284 36.7 6.4 9.9 3.0 16.2 2.2 41.0 10.5
Rescue work 165 33.5 4.8 8.3 2.9 15.8 2.4 39.0 11.3
Other shiftwork 87 36.1 7.5 10.0 3.3 16.4 2.2 42.3 11.6
Daytime work 268 36.0 7.2 8.8 3.3 15.3 2.6 46.6 12.0
Intragroup conflict Intergroup conflict Role ambiguity Role conflict

n Mean SD P-value Mean SD P-value Mean SD P-value Mean SD P-value

Gender
Male 1,626 18.5 5.1 0.34 19.5 4.1 0.26 18.2 5.5 0.42 26.8 7.9 0.30
Female 46 17.8 5.9 18.8 4.7 18.9 5.5 25.6 7.0
Age (years)
<30 221 18.3 5.3 0.03 18.9 4.4 <0.01 19.0 5.3 <0.001 27.9 8.0 <0.01
30^39 320 17.8 5.1 19.1 3.9 19.3 5.3 27.4 7.8
 40^49 501 18.7 5.4 19.5 4.1 17.8 5.2 26.9 7.9
>50 630 18.8 4.9 20.0 4.1 17.8 5.8 26.0 7.8
Martial Status
Married 1407 18.6 5.1 0.16 19.6 4.1 0.13 18.1 5.5 <0.01 26.7 7.9 0.65
Single 265 18.1 5.4 19.2 4.3 19.1 5.5 27.0 7.9
Smoking habit
Non-smoker 714 18.6 5.1 0.65 19.6 4.1 0.28 18.2 5.6 0.54 26.3 7.8 0.04
Current smoker 958 18.5 5.2 19.4 4.1 18.3 5.4 27.1 7.9
Drinking habit
Non-drinker 424 18.5 5.3 0.97 19.6 4.2 0.62 18.6 5.5 0.11 26.7 8.2 0.83
Drinker 1,248 18.5 5.1 19.5 4.1 18.1 5.5 26.8 7.8
Job class
Fire Fighter 541 19.0 5.2 0.050 19.8 4.2 0.14 19.3 5.5 <0.001 27.1 7.8 <0.01
Assistant Fire Sergeant 11 17.9 5.4 18.2 4.4 19.3 4.8 24.5 7.1
Fire Sergeant 515 18.6 5.4 19.6 4.1 19.2 5.0 27.2 7.8
Fire Lieutenant 374 18.2 5.0 19.4 4.2 17.9 5.6 27.0 8.1
Fire Captain 186 17.8 4.6 18.9 3.8 14.5 4.2 25.2 7.3
Battalion Chief 45 17.7 3.7 19.1 3.4 12.7 4.4 23.4 8.3
Type of job
Firefighting 868 19.0 5.1 <0.01 19.8 4.2 <0.01 18.1 5.6 <0.01 26.5 7.9 0.37
Ambulance work 284 17.6 4.8 18.7 3.8 19.3 5.0 26.5 7.9
Rescue work 165 18.1 5.1 19.2 3.8 18.5 5.4 27.6 8.0
Other shift work 87 18.8 5.7 19.7 4.6 17.9 5.6 27.1 7.8

474
Daytime work 268 18.2 5.5 19.7 4.2 17.6 5.7 27.3 7.7
Social support from Social support from Social support from
supervisor coworker family/friend Non-work activity

n Mean SD P-value Mean SD P-value Mean SD P-value Mean SD P-value

Gender
Male 1,626 15.7 3.3 0.78 16.6 2.7 0.14 16.8 2.9 0.44 1.0 0.7 0.96
Female 46 15.6 4.0 17.2 3.2 17.2 3.6 1.0 0.8
Age (years)
<30 221 16.0 3.1 <0.01 17.4 2.7 <0.001 17.6 2.5 <0.001 0.6 0.6 <0.001
30^39 320 16.3 3.4 17.2 2.9 17.0 3.1 0.9 0.7
40^49 501 15.5 3.1 16.5 2.5 16.4 2.9 1.2 0.7
>50 630 15.5 3.4 16.2 2.6 16.8 3.0 1.1 0.7
Martial Status
Married 1,407 15.7 3.3 0.50 16.6 2.7 <0.01 16.9 2.8 <0.001 1.1 0.7 <0.001
Single 265 15.8 3.4 17.1 2.9 16.3 3.8 0.8 0.8
Smoking habit
Non-smoker 714 15.6 3.2 0.29 16.5 2.7 0.053 16.8 2.9 0.67 1.0 0.8 0.70
Current smoker 958 15.8 3.3 16.8 2.7 16.9 3.0 1.0 0.7

ðContinuedÞ
 TABLE II. (Continued)

Social support from Social support from Social support from

supervisor coworker family/friend Non-work activity
P-value
n Mean SD P-value Mean SD P-value Mean SD P-value Mean SD
Drinking habit
Non-drinker 424 15.9 3.4 0.28 16.8 2.6 0.20 16.9 3.0 0.64 16.9 3.0 0.02
Drinker 1,248 15.7 3.2 16.6 2.7 16.8 3.0 16.8 3.0
Job class
Fire Fighter 541 15.4 3.4 <0.01 16.5 2.9 <0.02 17.2 2.8 <0.001 0.9 0.7 <0.001
Assistant Fire Sergeant 11 18.2 1.8 17.5 2.6 17.6 2.2 0.8 0.8
Fire Sergeant 515 15.7 3.4 17.0 2.7 16.8 3.1 1.1 0.8
Fire Lieutenant 374 16.0 3.3 16.5 2.6 16.6 3.0 1.1 0.8
Fire Captain 186 15.9 2.8 16.5 2.4 16.4 2.9 1.2 0.6
Battalion Chief 45 16.7 2.7 16.4 2.4 16.6 2.8 1.0 0.7

Type of job
Firefighting 868 15.7 3.3 0.76 16.5 2.7 0.01 16.9 3.0 <0.01 1.0 0.7 0.97
Ambulance work 284 15.6 3.1 16.8 2.5 17.0 2.7 1.1 0.8
Rescue work 165 15.6 3.3 17.2 2.9 17.2 2.8 1.0 0.7
Other shift work 87 15.7 3.1 17.0 2.4 16.5 3.0 1.1 0.8

475
Daytime work 268 15.9 3.5 16.5 2.8 16.2 3.1 1.0 0.7
Self-esteem CES-D

n Mean SD P-value Mean SD P-value

Gender
Male 1,626 33.6 6.2 0.01 12.0 7.2 0.01
Female 46 31.3 6.4 14.8 10.0
Age (years)
<30 221 32.5 6.9 0.02 11.8 8.3 0.19
30^39 320 33.9 6.5 11.4 7.0
40^49 501 34.0 6.5 12.1 7.7
>50 630 33.5 5.6 12.4 6.7
Martial status
Married 1,407 33.8 6.1 <0.001 11.8 6.9 <0.001
Single 265 32.3 6.9 13.3 8.9
Smoking habit
Non-smoker 714 33.4 6.6 0.38 11.8 7.3 0.20
Current smoker 958 33.7 6.0 12.2 7.3
Drinking habit
Non-drinker 424 32.4 6.7 <0.001 12.6 7.8 0.07
Drinker 1,248 34.0 6.0 11.8 7.1
476 Saijo et al.

prospect of their work; however, when they do not receive

role clarity, daytime workers lose interest and become
dissatisfied with non-firefighting tasks [Yamashita et al.,
2005]. Sectionalism and workload imbalance among job
types in the same fire station lead to higher intergoup conflict,
therefore an adequate cooperative structure in a fire station is
important.
In addition, high role conflict was significantly asso-
ciated with depressive symptoms. Relationships between
high role conflict and low job satisfaction and emotional
exhaustion were reported in the new manufacturing facility
workers [Zahrly and Tosi, 1988]. High role conflict between
nurses and civil engineers was related to low job satisfaction
and exhaustion [Bacharach et al., 1991]. Furthermore, since
fire fighters have to adhere to strict discipline, conflict
between regulations and appropriate measures for life saving
must exist.
Influence of low job strain to mental health has been
reported [Stansfeld et al., 1999; de Jonge et al., 2000;
Laaksonen et al., 2006]. One study reported that low job
control was related to low subjective symptoms among
Japanese male workers [Sakano et al., 1995]. Another study
showed that low job control was related to low mental health
0.03

0.16

problems among Japanese male permanent night workers

[Fujino et al., 2001]. In this study, the mid job control group
had a significantly low OR (0.64: vs. high) for depressive
11.3
7.2

6.8
7.2
4.9

8.0
7.5

8.0
7.4
7.4

7.4

symptoms, and the low job control group had an insignificant

but low OR (0.77: vs. high). A different reaction to job control
between Japanese workers and those from Western countries
has been speculated [Sakano et al., 1995]. Further studies are
11.0
12.7
12.6
11.3
11.7

11.9
12.5

12.2

12.3
12.0

9.0

needed to clarify the existence of an inverse or U-shaped

association between job control and mental health among the
Japanese population.
<0.001

0.21

Low self-esteem had the highest OR for depression. Low

self-esteem is related to depression [Price and Mueller, 1986;
Whisman and Kwon, 1993; Dori and Overholser, 1999].
Studies on fire fighters have reported self-esteem as being
6.2

6.3

5.3

6.3
6.2

6.6
6.4

5.9
6.4

6.4
5.4

significantly inversely related to stress [Petrie and Rotheram,

1982]. It is argued that an understanding of the development
of self-esteem, its outcomes, and its active protection and
32.5
32.6

34.5
33.2
33.5
33.7
33.7

36.8

33.7

33.1
35.1

promotion are critical to the improvement of both mental

and physical health [Mann et al., 2004]. Self-esteem is an
important factor for the evaluation of fire fighters’ mental
45

868
284
541
11
515
374
1,86

165
87
268

health, and measures of increasing their self-esteem should

be taken.
There are several limitations to our study. Since this
study was cross-sectional in design, the causal order of the
Assistant Fire Sergeant

association between each stress measure and the depressive

symptoms cannot be determined with certainty. Moreover, a
Ambulance work

Other shift work

small sample size of female subjects restricted gender-

Fire Lieutenant

Battalion Chief

Daytime work
Fire Sergeant

Rescue work
Fire Captain

specific analyses. If non-responding subjects had higher

Firefighting
Fire Fighter

Type of job

stress and depressive symptoms, the ORs would be under-

Job class

estimated. However, the high response rate of this study was

minimally affected by the selection bias. The job of Japanese
 TABLE III. ORs of Independent Variables for Depressive Symptoms (CES-D 16) Among the Japanese Fire Fighters

Univariate analysis Multivariate analysis*

n % OR 95 %CI P-value OR 95 %CI P-value

Female (vs. male) 46 2.8 2.09 1.14 to 3.85 0.02 2.18 0.99 to 4.79 0.053
Age
<30 221 13.2 1.04 0.72 to1.51 0.82 0.79 0.46 to1.34 0.38
30^39 320 19.1 1.03 0.74 to1.42 0.87 0.93 0.60 to1.46 0.76
40^49 501 30.0 1.14 0.86 to1.51 0.35 1.08 0.77 to1.52 0.67
>50 630 37.7 Reference
Married (vs. single) 1407 84.2 0.71 0.53 to 0.96 0.03 0.69 0.46 to1.03 0.07
Current smoker (vs.non-smoker) 958 57.3 1.16 0.92 to1.46 0.22 1.35 1.02 to1.78 0.03
Drinker (vs. non-drinker) 1248 74.6 0.82 0.63 to1.06 0.12 0.90 0.67 to1.21 0.49
Job class
Fire Fighter 541 32.4 Reference
Assistant Fire Sergeant 11 0.7 0.72 0.15 to 3.40 0.68 1.74 0.32 to 9.42 0.52
Fire Sergeant 515 30.8 0.96 0.72 to1.28 0.77 0.96 0.67 to1.36 0.80
Fire Lieutenant 374 22.4 0.86 0.62 to1.18 0.35 0.80 0.53 to1.23 0.31
Fire Captain 186 11.1 1.04 0.70 to1.54 0.84 1.21 0.69 to 2.14 0.50

477
Battalion Chief 45 2.7 0.32 0.11to 0.90 0.03 0.46 0.13 to1.58 0.22
Type of job
Fire fighting 868 51.9 1.05 0.60 to1.82 0.87 1.68 0.88 to 3.18 0.11
Ambulance work 284 17.0 1.37 0.76 to 2.49 0.30 1.58 0.81to 3.11 0.18
Rescue work 165 9.9 0.99 0.51to1.91 0.98 1.50 0.70 to 3.20 0.29
Other shift work 87 5.2 Reference Reference
Daytime work 268 16.0 1.66 0.92 to 3.00 0.09 2.14 1.09 to 4.20 0.03
Quantitative workload
Low 662 39.6 Reference Reference
Mid 457 27.3 1.36 1.00 to1.85 0.049 1.14 0.78 to1.66 0.50
High 553 33.1 2.32 1.76 to 3.06 <0.0001 1.44 0.94 to 2.21 0.09
Variance in workload
Low 566 33.9 Reference Reference
Mid 591 35.3 1.64 1.21to 2.22 <0.01 1.43 0.98 to 2.09 0.06
High 515 30.8 2.62 1.95 to 3.53 <0.0001 2.05 1.29 to 3.25 <0.01
Cognitive demand
Low 787 47.1 Reference Reference
Mid 525 31.4 1.31 1.00 to1.72 0.047 1.25 0.91to1.72 0.16
High 360 21.5 1.61 1.20 to 2.16 <0.01 1.41 0.97 to 2.04 0.07
 Job control
Low 602 36.0 1.26 0.94 to1.70 0.12 0.77 0.50 to1.17 0.22
Mid 652 39.0 0.89 0.66 to1.21 0.45 0.64 0.43 to 0.94 0.02
High 418 25.0 Reference Reference
Intragroup conflict
Low 630 37.7 1.50 1.11to 2.02 <0.01 0.91 0.64 to1.31 0.62
Mid 558 33.4 2.90 2.17 to 3.87 <0.0001 1.10 0.74 to1.63 0.65
High 484 28.9 Reference Reference
Intergroup conflict
Low 701 41.9 Reference Reference
Mid 532 31.8 2.17 1.61to 2.92 <0.0001 1.58 1.11to 2.24 0.01
High 439 26.3 3.67 2.73 to 4.93 <0.0001 1.91 1.26 to 2.88 <0.01
Role ambiguity
Low 633 37.9 Reference Reference
Mid 579 34.6 1.75 1.30 to 2.35 <0.001 1.15 0.81to1.64 0.44
High 460 27.5 3.01 2.24 to 4.04 <0.0001 1.27 0.86 to1.88 0.22
Role conflict
Low 564 33.7 Reference Reference
Mid 602 36.0 2.64 1.90 to 3.67 <0.0001 1.62 1.11to 2.36 0.01
High 506 30.3 4.41 3.18 to 6.11 <0.0001 1.87 1.24 to 2.80 <0.01
Social support from supervisor
Low 685 41.0 2.23 1.61to 3.07 <0.0001 1.00 0.63 to1.61 0.98

478
Mid 614 36.7 1.19 0.84 to1.68 0.33 0.77 0.49 to1.22 0.27
High 373 22.3 Reference Reference
Social support from coworker
Low 753 45.0 2.49 1.85 to 3.35 <0.0001 1.51 0.97 to 2.36 0.07
Mid 435 26.0 1.33 0.94 to1.89 0.11 1.14 0.73 to1.77 0.57
High 484 28.9 Reference Reference
Social support from family/friend
Low 685 41.0 2.00 1.45 to 2.75 <0.0001 1.33 0.90 to1.98 0.16
Mid 593 35.5 1.43 1.02 to 2.00 0.036 1.43 0.97 to 2.13 0.07
High 394 23.6 Reference Reference
Non-work activity
Low 294 17.6 1.30 0.97 to1.73 0.078 1.11 0.79 to1.56 0.54
High 1378 82.4 Reference Reference
Self-esteem
Low 647 38.7 7.19 5.08 to10.16 <0.0001 5.78 3.93 to 8.50 <0.0001
Mid 498 29.8 1.89 1.27 to 2.80 <0.01 1.69 1.10 to 2.58 0.02
High 527 31.5 Reference Reference

*All listed variables were included in the model.

 Stress and Depression Among Fire Fighters 479

fire fighters includes other work in addition to firefighting Haratani T, Kawakami N, Araki S. 1993. Reliability and validity of the
such as emergency services and rescue work. Fire fighters in Japanese version of NIOSH Generic Job Stress Questionnaire. Jpn J Ind
Health 35 Suppl :S214 (in Japanese).
the United States also have emergency medical responsi-
bilities. However, this result cannot be directly compared Haratani T, Kawakami N, Araki S, Hurrell JJ, Jr, Sauter SL, Swanson
NG. 1996. Psychometric properties and stability of the Japanese version
with other countries’ fire fighters who are engaged purely in of the NIOSH job stress questionnaire. The 25th International Congress
firefighting. on Occupational Health Book of Abstracts Pt2:393.
In conclusion, high variance in workload, high inter- Heinrichs M, Wagner D, Schoch W, Soravia LM, Hellhammer DH,
group conflict, high role conflict, and low self-esteem Ehlert U. 2005. Predicting posttraumatic stress symptoms from
were significantly related to depressive symptoms among pretraumatic risk factors: A 2-year prospective follow-up study in
firefighters. Am J Psychiatry 162:2276–2286.
Japanese fire fighters. Further prospective studies are needed
to clarify the influence of these stress factors on other health Hurrell JJ, Jr, McLaney MA. 1988. Exposure to job stress—A new
psychometric instrument. Scand J Work Environ Health 14 (Suppl
outcomes, and to elucidate whether alleviation of these stress 1):27–28.
factors improves mental health among fire fighters.
Kaneita Y, Ohida T, Uchiyama M, Takemura S, Kawahara K, Yokoyama
E, Miyake T, Harano S, Suzuki K, Fujita T. 2006. The relationship
between depression and sleep disturbances: A Japanese nationwide
general population survey. J Clin Psychiatry 67:196–203.
ACKNOWLEDGMENTS
Laaksonen M, Rahkonen O, Martikainen P, Lahelma E. 2006.
Associations of psychosocial working conditions with self-rated general
This work was supported in part by Grants-in-aid for health and mental health among municipal employees. Int Arch Occup
Scientific Research from the Japan Ministry of Education, Environ Health 79:205–212.
Culture, Sports, Science, and Technology. Mann M, Hosman CM, Schaalma HP, de Vries NK. 2004. Self-esteem in
a broad-spectrum approach for mental health promotion. Health Educ
Res 19:357–372.
REFERENCES
Melius JM. 1995. Cardiovascular disease among firefighters. Occup
Med 10:821–827.
Bacharach SB, Bamberger P, Conley S. 1991. Work-home
conflict among nurses and engineers: Mediating the impact of Petrie K, Rotheram MJ. 1982. Insulators against stress: Self-esteem and
role stress on burnout and satisfaction at work. J Organ Behav 12:39– assertiveness. Psychol Rep 50:963–966.
53.
Pikhart H, Bobak M, Pajak A, Malyutina S, Kubinova R, Topor R,
Booze TF, Reinhardt TE, Quiring SJ, Ottmar RD. 2004. A screening- Sebakova H, Nikitin Y, Marmot M. 2004. Psychosocial factors at work
level assessment of the health risks of chronic smoke exposure for and depression in three countries of Central and Eastern Europe. Soc Sci
wildland firefighters. J Occup Environ Hyg 1:296–305. Med 58:1475–1482.
Bos J, Mol E, Visser B, Frings-Dresen M. 2004. Risk of health Price JL, Mueller CW. 1986. Absenteeism and Turnover of Hospital
complaints and disabilities among Dutch firefighters. Int Arch Occup Employees. Greenwich: JAI Press.
Environ Health 77:373–382.
Radloff LS. 1977. The CES-D scale: A self-report depression scale for
Boxer PA, Wild D. 1993. Psychological distress and alcohol use among research in the general population. Appl Psychol Measurement 1:385–
fire fighters. Scand J Work Environ Health 19:121–125. 401.
Cho MJ, Nam JJ, Suh GH. 1998. Prevalence of symptoms of depression Sakai Y, Akiyama T, Miyake Y, Kawamura Y, Tsuda H, Kurabayashi
in a nationwide sample of Korean adults. Psychiatry Res 81:341– L, Tominaga M, Noda T, Akiskal K, Akiskal H. 2005. Temperament
352. and job stress in Japanese company employees. J Affect Disord
85:101–112.
de Jonge J, Bosma H, Peter R, Siegrist J. 2000. Job strain, effort-reward
imbalance and employee well-being: A large-scale cross-sectional Sakano J, Yamazaki Y, Sekiya E, Uehata T. 1995. The relation between
study. Soc Sci Med 50:1317–1327. job characteristics, job-related stress, and health related symptoms
among middle-aged male workers in Japan. J Sci Labour 71:1–12.
Dori GA, Overholser JC. 1999. Depression, hopelessness, and self-
esteem: Accounting for suicidality in adolescent psychiatric inpatients. Shima S, Shikano T, Kitamura T, Asai M. 1985. New self-rating scales
Suicide Life Threat Behav 29:309–318. for depression. Clin Psychiatry 27:717–723.
Dutton LM, Smolensky MH, Leach CS, Lorimor R, Hsi BP. 1978. Stress Stansfeld SA, Fuhrer R, Shipley MJ, Marmot MG. 1999. Work
levels of ambulance paramedics and fire fighters. J Occup Med 20:111– characteristics predict psychiatric disorder: Prospective results from
115. the Whitehall II Study. Occup Environ Med 56:302–307.
Fujino Y, Mizoue T, Izumi H, Kumashiro M, Hasegawa T, Yoshimura T. Takeyama H, Tachi N, Sakamura O, Niwa S, Suzumura H, Itani T. 2001.
2001. Job Stress and Mental Health among Permanent Night Workers. Job stress and fatigue among ambulance paramedics and fire fighters.
J Occup Health 43:301–306. Job Stress Res 8:223–231 (in Japanese, English abstract).
Godin I, Kittel F, Coppieters Y, Siegrist J. 2005. A prospective study of Takeyama H, Itani T, Tachi N, Sakamura O, Murata K, Inoue T,
cumulative job stress in relation to mental health. BMC Public Health Takanishi T, Suzumura H, Niwa S. 2005. Effects of shift schedules on
5:67. fatigue and physiological functions among firefighters during night
duty. Ergonomics 48:1–11.
Guidotti TL. 1992. Human factors in firefighting: Ergonomic-,
cardiopulmonary-, and psychogenic stress-related issues. Int Arch Tsutsumi A, Kayaba K, Theorell T, Siegrist J. 2001. Association
Occup Environ Health 64:1–12. between job stress and depression among Japanese employees
480 Saijo et al.

threatened by job loss in a comparison between two complementary job- Yamashita Y, Sato E, Ishizawa K, Hoshino N, Pak S, Owada Y,
stress models. Scand J Work Environ Health 27:146–153. Matsura S, Hanada N, Sugawara C, Murakami K, Fujimori Y. 2005.
Mental health of fire officials. J St. Marianna Med Inst 5:63–66 (in
Wagner D, Heinrichs M, Ehlert U. 1998. Prevalence of symptoms of Japanese).
posttraumatic stress disorder in German professional firefighters. Am J
Psychiatry 155:1727–1732. Zahrly J, Tosi H. 1988. Comparative analysis of the influence of stress-
related variables on adaptation to a new work setting. Psychol Rep
Whisman MA, Kwon P. 1993. Life stress and dysphoria: The role of self- 63:767–777.
esteem and hopelessness. J Pers Soc Psychol 65:1054–1060.