850 Article Summary

frank2012

militaryhierarchy.pdf

Home >Psychology homework help >850 Article Summary

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Occupational Medicine 2012;62:182–187 Advance Access publication on 7 March 2012 doi:10.1093/occmed/kqs006

Military hierarchy, job stress and mental health in

peacetime

L. C. X. Martins 1,2

and C. S. Lopes 1

1 Department of Epidemiology, Institute of Social Medicine, State University of Rio de Janeiro (IMS/UERJ), Rua São Francisco

Xavier 524, 7� andar, Rio de Janeiro, RJ 20550-900, Rio de Janeiro, RJ 20550-900, Brazil, 2Department of Post-Graduation/School of Physical Education of Brazilian Army, Escola de Educacxão Fı́sica do Exército—Secxão de Pós-Graduacxão, Avenida João Luı́z Alves, S/Nr—Urca, CEP 22291-090, Rio de Janeiro, RJ, Brazil.

Correspondence to: L. C. X. Martins, Department of Post-Graduation/School of Physical Education of Brazilian Army, Escola de

Educacxão Fı́sica do Exército—Secxão de Pós-Graduacxão, Av. João Luı́z Alves, S/Nr—Urca, CEP 22291-090, Rio de Janeiro, RJ, Brazil. Tel: 155 21 2586 2297; fax: 155 21 2295 7745; e-mail: [email protected] and C. S. Lopes, Department of Epidemiology,

Institute of Social Medicine, State University of Rio de Janeiro (IMS/UERJ), Brazil Rua São Francisco Xavier 524, 7� andar, Rio de Janeiro, RJ 20550-900, Brazil. Tel: 155 21 2334 0472; fax: 155 21 2334 2152; e-mail: [email protected]

Background Most studies of mental health in the armed forces focus primarily on post-traumatic stress disorders

among military personnel in combat situations.

Aims To evaluate the prevalence of common mental disorders (CMD) and job stress, and the association

between the two, among military personnel in peacetime. Additionally, it sought to identify occupa-

tional subgroups with higher prevalences of CMD.

Methods The study participants were 506 military personnel from a Brazilian army directorate in Rio de Janeiro

City. CMD were evaluated using the 12-item version General Health Questionnaire. Job character-

istics were measured using the effort-reward imbalance (ERI) model and by categories of military

rank. Prevalence ratios (PRs) were estimated by Poisson regression to obtain robust (95%) confidence

intervals (CIs).

Results The prevalence of CMD was 33% (95% CI 29–37). After adjusting for age, education, income, life-

style and other occupational characteristics, ERI was associated with CMD (PR 5 2.03; 95% CI 1.3–

3.1). Overcommitment proved to be an important component of job stress. Independently of socio-

economic, demographic, lifestyle and job stress variables, the rank of lieutenant associated strongly

with CMD (PR 5 2.02; 95% CI 1.2–4 0.1).

Conclusions This study found that job stress among armed forces personnel is associated with CMD. In addition,

the specific occupational characteristics of the military environment can lead to a higher prevalence of

CMD among those holding the rank of lieutenant.

Key words Effort-reward imbalance model; job stress; military personnel; mental health.

Introduction

The mental health of military personnel has been studied

in developed countries, particularly those involved in

war situations [1,2], focussing on the occurrence of

post-traumatic stress disorder. Few studies have inves-

tigated mental health among the military in peacetime

[2].

The work process of military personnel has character-

istics that can have an impact on patterns of illness. Their

working environment is strongly marked by peculiarities

inherent to the military profession: rigid hierarchy,

heightened competition and the ever-present possibility

of changes occurring against the individual’s wishes.

These characteristics have been considered important

sources of job stress [2].

The relationship between state of mental health and

work environment is complex and multifaceted, and an

unfavourable work environment is associated with higher

prevalence of mental disorders [3]. Stress generally has

been pointed to as one important risk factor for the

occurrence of common mental disorders (CMD) [4], a

situation characterized by symptoms of anxiety and de-

pression, in addition to a series of non-specific and somatic

complaints. The literature indicates that service personnel

with high stress levels [2,5] or stress-related disorders suf-

fer more physical and mental morbidity, miss more days

work, are less productive and make more frequent use

at E vans L

ibrary P eriodicals on M

ay 10, 2013 http://occm

ed.oxfordjournals.org/ D

ow nloaded from

http://occmed.oxfordjournals.org/

of health services [6]. Job stress can have an overall

effect on people, leading to anxiety, exhaustion, de-

pression and even substance abuse. In addition, the

constant possibility of changes at work, against the

individual’s wishes, can precipitate additional stress

[7]. The disability caused by stress is as major as that

resulting from work accidents or other common medical

conditions [6].

Few studies have investigated mental health among

armed service personnel in peacetime. Studies in the

UK have shown prevalences of CMD ranging from

20% [8,9] to 27%, depending on the cut-off point of the

instrument used. In the US armed forces, the prevalence

was 28% [1].

Despite the peculiarities of the work process of military

personnel while performing their normal military duties

in peacetime, little is known about the role of job stress

in their mental health. Studies, conducted mainly in

the USA and UK, have shown an important association

between exposure to job stress and the occurrence of

mental disorders [1,2]. In addition, those occupying

lower ranks in the military hierarchy display more likeli-

hood of mental disorders [10].

In Brazil, the only study of stress and mental health

among armed forces personnel was performed on the

return of one contingent deployed on a peacekeeping

mission in Haiti, and it focused on post-traumatic stress

disorder [11]. In this context, as far as we know, this is the

first study to investigate the prevalence of CMD and its

association to job stress, according to the effort-reward

imbalance (ERI) model and rank among active service

personnel in peacetime.

The aims of this study were to evaluate the prevalence

of CMD and its association with job stress and rank

among military personnel in peacetime.

Methods

This was a cross-sectional census-type study of armed

forces personnel serving at a Brazilian army directorate

and its subordinate military organizations. Data

collection took place from October 2009 to February

2010.

Written informed consent was obtained from all par-

ticipants. Ethical approval was obtained from the Ethics

Committee of Rio de Janeiro State University.

Data were collected using self-completed question-

naires on socio-economic, demographic and occupational

characteristics, physical activity, self-reported health con-

ditions and self-perceived health. CMD (outcome) was

evaluated using the GHQ-12, validated in Brazil [12].

The questionnaire asks about the presence of CMD dur-

ing the previous 2 weeks. It is scored by designating each

item as absent (0) or present (1); those scoring $3 (of 12)

were classified as ‘cases’ [13].

The exposure variables were job stress and rank. Job

stress was assessed by the ERI model, which states that

an adverse work environment is characterized by both

high effort and low reward and displays a reciprocity def-

icit between ‘costs’ and ‘gains’. Such an imbalance may

lead to a state of active distress, by evoking negative emo-

tions. This model also assumes that overcommitment

(OC) (a personality characteristic) intensifies the process;

highly overcommitted workers will respond to an ERI

with more strain reactions than less overcommitted

workers [7,14]. The instrument, adapted and validated

in Brazil [15], comprises 23 questions, and the results

are expressed as scores covering three facets: effort (E),

reward (R) and OC. Each facet was analysed separately,

as was the effort-reward ratio (E/R) and the total score

(TS) from the instrument, TS 5 (E/R 1 OC) [7]. The

quadrants of the ERI model were set-up as follows:

the two lower terciles of the scores were coded as

0 (low) and the upper tercile as 1 (high), from which

the following quadrants were formed: ‘high effort and

high reward’, ‘high effort and low reward’, ‘low effort

and low reward’, ‘low effort and high reward’. ‘Low effort

and high reward’ was taken as the reference category.

With the same methodology, TS and E/R were also

dichotomized into high stress (1) and low stress (0).

Rank was categorized as follows: ‘Superior officers and

captains’ (captain, major, lieutenant-colonel, colonel and

general), ‘Lieutenants’, ‘Sergeant-Majors and Sergeants’

and ‘Corporals and Privates’.

The covariates were age, education, income, marital

status and lifestyle variables. Per capita family income

(total family income divided by the number of family

members living on that income) and categorized in Brazil-

ian minimum wages. For lifestyle, the variables investi-

gated were consumption of any alcohol in the previous

2 weeks and current smoking (dichotomous) and physical

activity. The latter was evaluated by Baecke’s Question-

naire [16], a self-completed instrument that evaluates

physical activity levels in three dimensions (occupational,

sport/leisure exercise and other leisure-time physical ac-

tivities) and expresses the levels in the form of scores. This

questionnaire was validated in Brazil [17]. For some

analyses, the total score for physical activity was dichot-

omized, considering the two upper terciles as ‘High’ and

the lower tercile as ‘Low’.

Pretest and pilot studies were conducted, with test-

retest reliability assessment, in order to guarantee the

quality of the information. For the three scales, the find-

ings show high reliability ranging from substantial to

perfect [18].

Descriptive analyses were conducted. From the

bivariate analysis (x 2

), the variables that showed a statis-

tical association (P # 0.25) with CMD were selected

for multiple analysis in order to control for possible

confounders.

L. C. X. MARTINS AND C. S. LOPES: MILITARY HIERARCHY, JOB STRESS AND MENTAL HEALTH 183

at E vans L

ibrary P eriodicals on M

ay 10, 2013 http://occm

ed.oxfordjournals.org/ D

ow nloaded from

http://occmed.oxfordjournals.org/

To evaluate the unadjusted and adjusted prevalence

ratios (PRs) and their respective 95% confidence intervals

(CIs) for the association between ERI, the E and R

quadrants, rank and CMD, Poisson regression was used,

via Generalized Linear Models, as that yields more robust

CIs. Data modelling fit was tested by the goodness-of-fit

method. Absent data were excluded from the analyses.

Data entry was performed in MSAccess/2000. Statis-

tical analyses were performed using the R Platform.

Results

There were 654 military personnel serving at the director-

ate. Of these, 68 (10%) had been deployed away from the

directorate indefinitely and were excluded. Forty-six

(8%) refused to take part and of the remaining 540, 34

(6%) were women who, being so few in number, were

withdrawn from the study. The study population there-

fore comprised of 506 participants. Their characteristics

are given in Table 1. The prevalence of at least one self-

reported physical morbidity was 42%, but none displayed

any statistically significant association with CMD.

Eighty-two (16%) respondents reported suffering from

job stress (high effort and low reward) (Table 1). Figure 1

shows the distribution of job stress by rank. The highest

prevalences can be seen among lieutenants (40%), corpo-

rals and privates (37%).

Prevalence of CMD was 33% (95% CI 29.1–37.3).

Table 1 shows this prevalence by socio-economic and

demographic characteristics, self-perceived health, life-

style components, rank and job stress in the study

population. Self-perceived health (P , 0.001) and occu-

pational physical activity (P , 0.01) were positively asso-

ciated with CMD, while sports/leisure exercise (P , 0.05)

was inversely associated with CMD (Table 1).

Table 2 shows the prevalence and the prevalence ratios

of CMD by job stress levels and by rank.

The unadjusted models showed strong association

between the rank of lieutenant and CMD. This associa-

tion did not change even after adjustment for other work

characteristics, self-rated health and lifestyle (full model)

(PR 5 2.02; 95% CI 1.2–4.1). No association was found

between other military ranks and CMD.

As regards job stress, the combination of high effort

and low reward (ERI) displayed the strongest association

with CMD in both the unadjusted and adjusted models.

In comparison with the reference quadrant (low effort

and high reward), and after adjustment for all variables

(full model), high effort and low reward displayed twice

the prevalence of CMD (full model: PR 5 2.01; 95% CI

1.3–3.1), followed by high effort and high reward (PR 5

1.87; 95% CI 1.2–2.9).

The E/R showed no association with CMD, but TS of

the ERI model did show association (PR 5 1.13; 95% CI

1.1–1.2).

Discussion

The results showed that one third of the male service per-

sonnel evaluated displayed CMD and that high effort and

low reward at work and the rank of lieutenant in the

Table 1. Characteristics of the study population and prevalence of

CMD

Characteristic n CMD P Cases (%)

Age NS

Up to 25 years 216 74 (34)

From 26 to 35 years 139 55 (40)

From 36 to 45 years 119 30 (25)

Above 45 years 29 8 (28)

Unknown 3

Education NS

More than 9 years of study 479 159 (33)

Up to 9 years of study 27 9 (33)

Income a

Up to 2 149 48 (32)

More 2–5 177 67 (38)

More 5–10 119 39 (33)

.10 41 9 (22)

Unknown 20

Rank ,0.001*** Superior officers and captains 95 21 (22)

Lieutenants 113 56 (50)

Sergeant-majors and sergeants 107 28 (26)

Corporals and privates 191 63 (33)

Lifestyle b

Physical activity c

Occupational ,0.01** High 354 130 (37)

Low 147 34 (23)

Sports/exercises at leisure-time ,0.05* High 344 101 (29)

Low 156 63 (40)

Other physical activity

during leisure and commuting

High 380 119 (31)

Low 119 45 (38)

Self-perceived health ,0.001*** Very good 344 85 (26)

Good 169 69 (44)

Regular 24 13 (54)

Bad 1 1 (100)

Job stress d

,0.001*** Low efforts and high rewards 284 58 (22)

Low efforts and low rewards 94 31 (35)

High efforts and high rewards 75 34 (48)

High efforts and low rewards 82 45 (55)

Characteristics of the study population and prevalence of CMD (N 5 506), 2010.

a Income monthly.

b Alcohol consumption and smoking were not statistically significant (P . 0.05).

c Levels determined by terciles.

d CMD.

e Quadrants of ERI model.

Statistical significance: *P ,0.05; **P , 0.01; ***P , 0.0001.

184 OCCUPATIONAL MEDICINE

at E vans L

ibrary P eriodicals on M

ay 10, 2013 http://occm

ed.oxfordjournals.org/ D

ow nloaded from

http://occmed.oxfordjournals.org/

military hierarchy were associated with these disorders.

These associations were independent of age, schooling,

income, marital status, self-perceived health, lifestyle

and other occupational characteristics. These findings

are important in that the occurrence of CMD can entail

consequences such as reduced productivity at work and

higher rates of absenteeism [6].

The strengths of this study are that it is one of the few

to have investigated mental health in serving members of

Brazil’s armed forces and that it achieved a very good

participation rate (92%). The study population was

homogeneous (socio-economic and demographic char-

acteristics), which reduced the likelihood of residual con-

founders. In addition, the census of service personnel at

the directorate revealed a sample population with the

same socio-economic and demographic characteristics Figure 1. Job stress according to military rank.

Table 2. Prevalence and prevalence ratios of CMD according to levels of job stress and rank (N 5 506), 2010

Exposure levels N (%) PR unadjusted (95%) Model 1 Model 2 a

Adjusted PR (95%) Adjusted PR (95%)

Rank

Superior officers and

captains b

21 (22) 1.00 1.00 1.00

Lieutenants 56 (50) 2.24 (1.4–3.7) 2.04 (1.2–3.5) 2.02 (1.2–4.1)

Sergeant-majors and

sergeants

28 (26) 1.18 (0.7–2.1) 1.14 (0.7–2.4) 1.33 (0.7–2.4)

Corporals and privates 63 (33) 1.49 (0.9–2.5) 1.3 (0.8–2.4) 1.57 (0.8–3.2)

Job Stress by quadrants c

Low efforts and high

rewards

58 (22) 1.00 1.00 1.00

Low efforts and low

rewards

31 (35) 1.59 (1.0–2.5) 1.59 (1.1–2.5) 1.60 (1.1–2.5)

High efforts and high

rewards

34 (48) 2.17 (1.4–3.3) 2.04 (1.3–3.2) 1.87 (1.2–2.9)

High efforts and low

rewards

45 (55) 2.49 (1.7–3.7) 2.18 (1.4–3.3) 2.01 (1.3–3.1)

Job stress by facets d

Effort (E) 79 (52) 1.10 (1.1–1.1) 1.06 (1.1–1.1) 1.05 (1.1–1.1)

Reward (R) 92 (28) 1.07 (1.0–1.1) 1.03 (1.0–1.1) 1.03 (1.0–1.1)

OC 69 (47) 1.15 (1.1–1.2) 1.07 (1.0–1.1) 1.07 (1.0–1.2)

Job stress by ERI

E/R e

47 (28) 0.85 (0.6–1.2) 0.86 (0.6–1.2) 0.88 (0.6–1.2)

TS of ERI model f

80 (47) 1.13 (1.1–1.2) 1.12 (1.1–1.2) 1.11 (1.1–1.2)

Unadjusted and adjusted PRs and respective 95% CI for the association between levels of job stress, rank and CMD: in military personnel of the Brazilian Army (N 5 506)

in 2010. Model 1: adjusted for lifestyle (alcohol and physical activity). Model 2: adjusted for lifestyle, age, self-rated health and other job characteristics. Job stress set by

quadrants, effort and reward were not included in the model (full model).

a Rank and job stress mutually adjusted.

b Reference category.

c Quadrants of the ERI model.

d Facets of ERI model; categorized in ‘high’ and ‘low’ levels.

e Effort–reward ratio.

f TS of the ERI model: TS 5 E/R 1 OC.

L. C. X. MARTINS AND C. S. LOPES: MILITARY HIERARCHY, JOB STRESS AND MENTAL HEALTH 185

at E vans L

ibrary P eriodicals on M

ay 10, 2013 http://occm

ed.oxfordjournals.org/ D

ow nloaded from

http://occmed.oxfordjournals.org/

as the ground forces overall [19], enabling the results to

be extrapolated to the whole Brazilian army.

Some limitations of the study must be considered.

Firstly, at initial sampling (the census), the proportion

of women was far higher (6%) than in the army overall

(1.3%), which could weaken the power of the study

and we therefore excluded women from the analyses.

However, our results remained comparable to the studies

cited here because they analysed the outcomes controlling

by gender, although they do not present all results sepa-

rately by gender. Secondly, this study is based on cross-

sectional data, which are unlikely to determine the

temporal nature of the phenomena examined, restricting

the inference concerning a causal direction of the associ-

ation of job stress and military hierarchy with CMD. We

cannot exclude reverse causation where respondents with

CMD perceived their work environment as more stress-

ful. However, longitudinal studies have consistently

shown that an adverse psychosocial environment at

work is a predictive factor for psychiatric morbidities

[20,21,22]. Our findings are in agreement with the

literature. Thirdly, individuals who are suffering psycho-

logically may report working conditions more negatively

than individuals who do not display such symptoms [23].

This bias can be avoided through the strategy of clustering

the workers into occupational categories under homoge-

neous working conditions and then aggregating cases and

non-cases by these categories [24]. Those conditions

were met by the characteristics inherent to military occu-

pational categorization, thus meeting the requirements

proposed [24] for less biased analysis. Lastly, the study

was based on the GHQ-12, a screening instrument, rather

than on a standardized clinical interview, which could

yield psychiatric diagnoses. Studies have shown that asso-

ciations between work characteristics tend to be greater in

studies using standardized clinical interviews [29].

According to the ERI model, the combination of high

effort with low reward (job stress condition) [7] increases

the risks for future health outcomes and defines adverse

psychosocial work conditions. That claim has been con-

sistently corroborated in relation to mental health [25].

Our findings agree with the literature [26,20,14], showing

that ERI was associated with higher prevalence of CMD

(PR 5 2.01). A study conducted among 7766 military

[10] using the demand-control model demonstrated that

service personnel displaying job strain showed greater risk

of psychological symptoms. Other studies have shown

that military personnel with high levels of occupational

stress had significant rates of mental disorder and loss

of productivity. In the same direction, Pflanz and Ogle

[5] found that both job stress and depression were signif-

icantly associated with reduced job performance, more

days of absenteeism, greater morbidity and more negative

perceptions of superiors’ abilities.

The total score of the ERI model showed association

with CMD, indicating that OC is an important compo-

nent in overall perception of job stress. OC occurs out

of a desire for esteem and approval and can, consequently,

result in emotional exhaustion and also aggravate the

adverse effects of the situations investigated by the ERI

model [7]. ERI plus OC represents a risk for several

diseases [20], including mental disorders [3].

Our findings show that, independently of socio-

economic, demographic, lifestyle and job stress variables,

the rank of lieutenant was strongly associated with CMD,

with prevalence (50%) more than twice that of the other

officer ranks (22%). Fear et al. [10] showed that military

rank is associated with mental disorders. They found

that the category ‘commissioned officers’ was associated

with lower risk for psychological symptoms than ‘non-

commissioned officers’ and other ranks, which differs

somewhat from our findings. In an eminently hierarchical

environment, the higher the rank, the greater the plan-

ning role and, the lower the rank, the more executive ac-

tivities predominate. In such a context, the rank of

lieutenant displays certain specific characteristics: (i) it

is the lowest commissioned officer rank, the first

rank above the non-commissioned officer category

(sergeant-majors and sergeants); (ii) it has more superiors

to answer to than the other officer ranks; (iii) it is directly

responsible for the performance of duties to be carried out

by non-commissioned officers, corporals and privates;

(iv) it is the only officer rank that does guard duty, i.e.

remains in the barracks overnight on a duty roster basis,

safeguarding the security of the military unit and (v) it is

the rank most liable to undesirable geographical or func-

tional changes at work. These characteristics indicate that

there is a possibility of stress in certain differing and con-

comitant directions. In relation to the hierarchy, the need

for esteem and approval may be directed towards both

superiors and subordinates, given that it is quite common

for sergeants and sergeant-majors to have served many

more years in the armed forces than lieutenants, nearly

all of whom have been performing military duties for only

a few years. Also, the possibility of undesirable changes

at work, which is greater at the rank of lieutenant, is

established in the literature as a cause of stress [7].

The prevalence of CMD encountered (33%) was

slightly higher than found in the armed forces of devel-

oped countries: in the UK, the prevalence was 27%

[27] and in the USA, 28% [2]. The prevalence of

CMD in the overall population of developed countries

is slightly lower than in developing countries [3,28].

In Brazil, the prevalence of CMD in the general pop-

ulation ranges from 17 to 35% and varies by region. In

Rio de Janeiro, one study of 3754 civil servants found

a prevalence of 27% among men [23]. The higher prev-

alence of CMD among male military personnel can be

explained by factors such as the possibility of forced

changes at work that entail changes of function or even

relocation and can also be related to the military hierarchy,

as found in our results.

186 OCCUPATIONAL MEDICINE

at E vans L

ibrary P eriodicals on M

ay 10, 2013 http://occm

ed.oxfordjournals.org/ D

ow nloaded from

http://occmed.oxfordjournals.org/

Acknowledgements

Our thanks to the members of the Brazilian army who agreed to

take part in this study, and to the Director and Commanding

Officers of the military units involved who made it possible to

carry out this study by offering all necessary support.

References

1. Pflanz SE. Work stress in the military: prevalence, causes, and

relationship to emotional health. MilMed 2002;167:877–882.

2. Hourani LL, Williams TV, Kress AM. Stress, mental health,

and job performance among active duty military personnel:

findings from the 2002 Department of Defense Health-

Related Behaviors Survey. Mil Med 2006;171:849–856.

3. Chopra P. Mental health and the workplace: issues for

developing countries. Int J Ment Health Syst 2009;3:4.

4. Goldberg DP, Huxley P. Common Mental Disorders.

A Bio-Social Model. London: Routledge, 1992.

5. Pflanz SE, Ogle AD. Job stress, depression, work perfor-

mance, and perceptions of supervisors in military personnel.

Mil Med 2006;171:861–865.

6. Kalia M. Assessing the economic impact of stress—the

modern day hidden epidemic. Metabolism 2002;51:49–53.

7. Siegrist J. Adverse health effects of high-effort/low-reward

conditions. J Occup Health Psychol 1996;1:27–41.

8. Jones M, Rona RJ, Hooper R, Wesseley S. The burden of

psychological symptoms in UK Armed Forces. Occup

Med (Lond) 2006;56:322–328.

9. Iversen A, Dyson C, Smith N etal. ‘‘Goodbye and good luck’’:

the mental health needs and treatment experiences of British

ex-service personnel. Br J Psychiatry 2005;186:480–486.

10. Fear NT, Rubin GJ, Hatch S etal. Job strain, rank, and mental

health in the UK Armed Forces. Int J Occup Environ Health

2009;15:291–298.

11. Souza WF, Figueira I, Mendlowicz MVet al. Negative affect

predicts posttraumatic stress symptoms in Brazilian volun-

teer United Nations peacekeepers in Haiti. J Nerv Ment Dis

2008;196:852–855.

12. Mari JJ, Williams P. A comparison of the validity of two

psychiatric screening questionnaires (GHQ-12 and SRQ-

20) in Brazil, using Relative Operating Characteristic

(ROC) analysis. Psychol Med 1985;15:651–659.

13. Goldberg D, Williams P. The User’s Guide to the General

Health Questionnaire. Windsor, UK: Nfer-Windsor, 1988.

14. van Vegchel N, de Jonge J, Bosma H, Schaufeli W. Reviewing

the effort-reward imbalance model: drawing up the balance of

45 empirical studies. Soc Sci Med 2005;60:1117–1131.

15. Chor D, Werneck GL, Faerstein E, Alves MG,

Rotenberg L. The Brazilian version of the effort-reward im-

balance questionnaire to assess job stress. Cad Saude Publica

2008;24:219–224.

16. Baecke JA, Burema J, Frijters JE. A short questionnaire for

the measurement of habitual physical activity in epidemio-

logical studies. Am J Clin Nutr 1982;36:936–942.

17. Florindo AA, Latorre MRDO. Validation and reliability of

the Baecke questionnaire for the evaluation of habitual

physical activity in adult men. Revista Brasileira de Medicina

do Esporte 2003;9:129–135.

18. Landis JR, Koch GG. The measurement of observer agree-

ment for categorical data. Biometrics 1977;33:159–174.

19. BRASIL. Decreto n.� 5170, de 5 de agosto de 2004, Altera os Quadros V e VI do Anexo ao Decreto no 4.967, de 30 de

janeiro de 2004, que dispõe sobre os efetivos do pessoal mil-

itar do Exército, em servicxo ativo, a vigorar em 2004. Diário Oficial da Republica Federativa do Brasil, Brası́lia, DF,

n. 152, p. 3, Secxão 1. 9 ago, 2004. 20. Stansfeld S, Candy B. Psychosocial work environment and

mental health—a meta-analytic review. Scand J Work Envi-

ron Health 2006;32:443–462.

21. Kivimaki M, Hotopf M, Henderson M. Do stressful work-

ing conditions cause psychiatric disorders? Occup Med

(Lond) 2010;60:86–87.

22. Stansfeld SA, Fuhrer R, Shipley MJ, Marmot MG. Work

characteristics predict psychiatric disorder: prospective re-

sults from the Whitehall II Study. Occup Environ Med

1999;56:302–307.

23. Lopes CS, Araya R, Werneck GL, Chor D, Faerstein E. Job

strain and other work conditions: relationships with psycho-

logical distress among civil servants in Rio de Janeiro, Brazil.

Soc Psychiatry Psychiatr Epidemiol 2010;45:345–354.

24. Kolstad HA, Hansen AM, Kaergaard A et al. Job strain and

the risk of depression: is reporting biased? Am J Epidemiol

2010;173:94–102.

25. Siegrist J. Chronic psychosocial stress at work and risk of

depression: evidence from prospective studies. Eur Arch

Psychiatry Clin Neurosci 2008;258(Suppl 5):115–119.

26. Kivimaki M, Virtanen M, Elovainio M, Kouvonen A,

Vaananen A, Vahtera J. Work stress in the etiology of coro-

nary heart disease—a meta-analysis. Scand J Work Environ

Health 2006;32:431–442.

27. Iversen A, van Staden L, Hughes J et al. The prevalence of

common mental disorders and PTSD in the UK military:

using data from a clinical interview-based study. BMC Psy-

chiatry 2009;9:68.

28. Patel V, Kleinman A. Poverty and common mental disor-

ders in developing countries. Bull World Health Organ

2003;81:609–615.

29. Dragano N, He Y, Moebus S, Jockel KH, Erbel R, Siegrist J.

Two models of job stress and depressive symptoms. Results

from a population-based study. Soc Psychiatry Psychiatr Epi-

demiol 2008;43:72–78.

Key points

• The prevalence of common mental disorders in the Brazilian army in peacetime was 33%, which

was higher than those found in other studies.

• The effort-reward imbalance (high effort and low reward) at work in military personnel was strong

and independently associated (prevalence ratio 5

2.01; confidence interval 95% 1.3–3.1) to common

mental disorders.

• Common mental disorders were twice as prevalent at the rank of lieutenant as at the other ‘commis-

sioned officer’ ranks.

L. C. X. MARTINS AND C. S. LOPES: MILITARY HIERARCHY, JOB STRESS AND MENTAL HEALTH 187

at E vans L

ibrary P eriodicals on M

ay 10, 2013 http://occm

ed.oxfordjournals.org/ D

ow nloaded from

http://occmed.oxfordjournals.org/