Work-related psychosocial risk factors and psychiatric disorders: A cross-sectional study in the French working population.

PURPOSE: The study estimates the prevalence of probable psychiatric disorder in the working population, determines the proportion of people presenting a probable psychiatric disorder among people exposed to work-related psychosocial risk factors (PSRFs), and identifies which PSRF has the strongest association with having a probable psychiatric disorder.
METHODS: A cross-sectional study conducted in March 2018 involved a representative sample of the French working population. The General Health Questionnaire 28 (GHQ-28) was used to estimate the prevalence of probable psychiatric disorder and 44 items were gathered from theoretical models of PSRFs. We used multiple logistic regression to estimate the association of each PSRF with having a probable psychiatric disorder, adjusted on individual, health, and job confounders.
RESULTS: This study involved 3200 French participants. The proportion of probable psychiatric disorder was 22.2% [20.6; 24.0]. Ten PSRFs were significantly associated with it. The strongest association was for having problems handling professional and personal responsibilities (reported by 15% of the study population) (OR = 1.97 [1.52; 2.54]), with 45% pathological GHQ-28 scores (potential psychiatric cases) for people exposed to this PSRF versus 18% non-exposed. The next strongest association was lack of support of colleagues (reported by 28%) (OR = 1.63 [1.29; 2.06]). The third strongest association was feeling sometimes afraid when doing the job (reported by 63%) (OR = 1.53, [1.21; 1.93]).
CONCLUSIONS: Our study identified 10 PSRFs associated with psychiatric disorder, with substantial exposure rate among the population. The results of our research could help develop recommendations to improve work environment.

Introduction

Treatable common psychiatric disorders affected 17.3% of people in European Union in 2018. Anxiety disorders (5.4%) and depressive disorders (4.5%) were the most prevalent [1]. Psychiatric disorders are also an important part of the global burden of disease: for instance, major depressive disorder was one of the five leading causes of years lived with disabilities in 2016, contributing to 4.2% of the global disease burden [2].

Psychiatric disorders seem to have multiple causes, and the impact of work-related psychosocial risk factors (PSRFs) on mental health outcomes have been investigated for 40 years. For example, conceptual models such as the job demand-control-support model, the effort-reward imbalance model and the organizational justice model theorized the link between job determinants and mental health outcomes [3-5]. Epidemiological studies have investigated the association between PSRFs and a range of common mental disorders such as generalized anxiety disorder or alcohol abuse [6,7]. A recent meta-analysis showed that being exposed to PSRFs such as “poor supervisor and colleague support” or “job insecurity” was correlated with suicidality outcomes [8].

However, when the association between psychiatric disorders and work-related PSRF was investigated, it was mostly for specific disorders and with a specific population, mainly sick people [9]. Studies of representative samples of the working population are required for at least two reasons: they allow for a robust generalization of results as well as robust analyses of the association between work-related PSRFs and mental health disorders because they include not just sick individuals.

This STROBE-compliant cross-sectional study of a representative sample of the French workers aimed to estimate the prevalence of probable psychiatric disorders in the working population, determine the proportion of people presenting a probable psychiatric disorder among people exposed to work-related PSRFs, and determine which PSRF has the strongest association with having a probable psychiatric disorder, after adjusting for individual health and job characteristics.

Material and methods

The report of the study is compliant with the STROBE checklist for cross-sectional studies [10].

Design and population

A cross-sectional study was performed between February 27 and March 6, 2018 with 3200 individuals living in France and representative of French workers. Individuals aged 18 to 80 years who declared currently having a job (even a part-time job) whatever their occupation or status (employee or self-employed) were eligible. We excluded students, unemployed individuals, housewives/husbands and retired people. People were recruited in the “Ipsos Access Panel” and were invited by e-mail to participate in the study. Representativeness was achieved by using quota sampling for sex, age, occupation and residence locality and the rim weighting method (raking) [11]. All data were collected by using a standardized questionnaire (S1 Table) administered during a computer-assisted web interview.

Ethics

The “IPSOS Access Panel” is acknowledged by the French National Institute for Data Security (CNIL) and is compliant to the ESOMAR international code of conduct. All participants answered the questionnaire voluntarily after giving their informed consent online and could unsubscribe from the survey without influencing panel membership. Since the study did not involve any human experimentation and the online questionnaire was proposed to volunteers of the IPSOS Access Panel, in respect with data protection standards of the French law, we did not seek for the approval of an ethic committee.

Data collection

Measurement of probable psychiatric disorder

We used the General Health Questionnaire– 28 (GHQ-28), a self-administered screening questionnaire designed to detect a probable psychiatric disorder in primary care settings [12]. The GHQ-28 is meant to estimate prevalence in a large population more than for individual cases. The questionnaire explores four dimensions—depressive symptoms, anxiety symptoms, social dysfunction and psychosomatic symptoms—with a list of 28 items, each one rated “better than usual”, “same as usual”, “worse than usual” and “much more than usual”. The Likert scoring is recommended (scoring each item from 0 to 3 with a global score ranging from 0 to 84) [13]. A total score of 24 is the threshold for detecting probable psychiatric disorder with sensitivity 79.8% and specificity 78.5% [14]. The French version of the questionnaire, used in our study, has been validated [15].

Measurement of psychosocial risk factors (PSRF)

PSRF were evaluated with a list of 44 questions (S1 Table) were compiled by the scientific committee according to major theoretical works from Karasek, Siegrist and Greenberg. In 1979, Karasek showed that a high job demand associated with low job control (decision latitude) is associated with high mental strain as well as medical outcomes [4]. This model was enriched in 1990 with social support that appeared to be stress-buffering [5]. In 1996, Siegrist added the effort-reward imbalance model to assess adverse health effects of stressful experience at work [16]. Another complementary theoretical framework is the model of organizational justice that assess fairness perceptions of organizations [3,17]. All these models describe the individual perception of the work environment. Moreover, we took into account other factors identified in the literature such as organizational change, job insecurity, workplace conflict and bullying [18]. The 44 items were rated “utterly true”, “quite true”, “quite false”, “utterly false”. The scoring method was to 1 for “utterly true” and “quite true” and 0 for “quite false” and “utterly false”. Therefore, participants with a score of 1 to a given PSRF are called “participants exposed” to this PSRF in the results section. Among the 44 questions, 35 concerned only employees and not self-employed individuals.

Individual and job characteristics

Several groups of variables about individual factors were measured to describe the population:

Socio-demographic factors (8 variables): age, sex, residency, marital status, number of children, highest educational degree obtained, household annual income, occupational status

Health and life hygiene conditions (6 variables): chronic disease (>6 months), psychotropic medication in the past 12 months, medical work cessation over 1 week due to a psychiatric condition in the last 12 months, alcohol consumption, smoking, illegal drug use in the past 12 months.

Job characteristics (10 variables): activity sector, employed or self-employed status, duration of the job (months), work duration (hours per week), weekend work, night work, staggered hours, commuting duration, teleworking, previous experience with unemployment.

Work environment (2 variables): workplace (e.g., open space, co-working, at home, etc.), company/public institution size.

Statistical analysis

The study outcome was the binary variable of having or not a probable psychiatric disorder defined by the individual GHQ-28 score ≥ 24 [14]. We assessed the prevalence of probable psychiatric disorder: the distribution of the GHQ-28 global score, mean and median for the sample were calculated and compared to the cut-off ≥ 24. We also assessed exposure to psychosocial risk factors and prevalence of probable psychiatric disorder: the descriptive statistics for all variables are given as percentages for categorical variables or mean (SD) for continuous variables.

For each PSRF, we calculated the proportion of people exposed and not exposed to the PSRF, and for each category (exposed/not exposed), the prevalence of probable psychiatric disorder. The differences in prevalence between exposed and non-exposed groups were tested by chi-square test. P<0.05 was considered statistically significant with Holm-Bonferroni correction.

For individual and job-related variables, we calculated the proportion of people with GHQ-28 score ≥24 (potential psychiatric cases) and tested the difference in proportion of probable psychiatric disorder between exposed and non-exposed groups for each variable by chi-square test. P<0.05 was considered statistically significant with Holm-Bonferroni correction. For significant variables with more than two categories, a Pearson residuals table [(observed–expected)/sqrt(expected)] was computed to identify the modality with a significantly higher proportion of GHQ-28 score ≥ 24. This modality is the one with the higher Pearson residual for GHQ score ≥ 24.

To investigate the association between each PSRF and probable psychiatric disorder, we modeled the binary outcome GHQ-28 score ≥24 as a logistic regression of some PSRFs, adjusted for possible confounders found in the literature: age, sex, chronic medical condition, working on the weekend, staggered work hours, working at night, previous experience with unemployment, work duration per week, and commuting duration [18,19]. To select which PSRF belonged to the model, we used a fast-backward procedure [20]. This method is extremely efficient because the model is not refitted for every variable removed. The model was used for employees, excluding self-employed individuals because the latter group had 8 PSRFs less measured.

Because sex was previously identified as a confounding factor for psychiatric disorder, we calculated the interactions between sex and the 11 PSRFs selected by the fast-backward procedure and included in the model [19]. We also calculated the 55 interactions between the 11 PSRFs included in the model by using the procedure proposed by Cox and Wermuth [21]. This procedure gives an idea of the overall importance of the interaction term in the model by estimating and testing all interaction terms desired, one at a time. We corrected the p value with Holm-Bonferroni.

Results

Description of the sample and prevalence of psychiatric disorders

Among the 40514 eligible people from the “IPSOS Access Panel” invited by e-mail to participate in the study, 10660 agreed and 3487 were included (S1 Fig). Because 287 did not complete the questionnaire to the end, data for 3200 people were analyzed. For the following analysis, the sample was weighted to be representative of the general population (S2 Table).

For the whole sample (Table 1), the mean (SD) mean age was 41.4 (11.1) and 48.2% were women. Overall, 42.6% of participants had a work duration of more than 39 hours a week; in France, the legal work duration is 35 hours a week. Also, 28.5% declared night work, 64.2% weekend work and 61.4% staggered hours; 19.1% reported more than 1 hour commute time to work. In all, 51% reported no previous experience with unemployment, whereas 28.2% had ≥ 2 experiences. Also, 12.7%, had a sickness work cessation over 1 week in the last 12 months due to a psychiatric condition. Moreover, 17.2% reported having taken psychotropic medication (e.g., antidepressants, anxiolytics) and 6.3% reported using illegal drugs in the past 12 months. Furthermore, 23.9% reported a chronic disease (physical or mental). Finally, 9.2% reported drinking > 10 glasses per week, which is a cut-off for high risk of all-cause-mortality recently identified [22]. In addition, 3.1% declared drinking > 21 glasses of alcohol per week, which is the classical cut-off for the WHO.

Table 1

Characteristics of the whole sample and of participants with a GHQ-28* score ≥ 24.

	Whole sample (n = 3200)	GHQ28 score ≥24 (n = 688)
Sex
Male	51.6	18.8
Female	48.2	25.8**
Age, years	41.4 (11.1)	39.8 (10.9)
Number of children
None	52.6	21.5
1	21.7	23.7
2	19.9	23.9
3	4.8	27.2
> 4	0.9	15.3
Marital status
Married/free union	71.5	21.2
Alone	28.5	24.8
Highest educational degree
< baccalaureate degree	21.3	20.3
Baccalaureate degree	21.7	20.1
Baccalaureate degree +2 years	24.2	23.6
Baccalaureate degree+ ≥ 3 years	32.8	23.8
Annual household income (euros)
< 15 000	11.2	30.9**
15–24 000	20.5	24.2
24–36 000	26.2	22.5
> 36 000	29.3	19.3
No information	12.8	17.4
Occupational status (independent worker)	N = 400
Farmer	18.3	15.9
Craftsman, shopkeeper, entrepreneur	47.5	21
Managerial and professional occupation	18.3	17.8
Technician and associate professional	15.9	20.3
Occupational status (employees)	N = 2800
Managerial and professional occupation	17.2	20.6
Technician and associate professional	27.7	24.6
Service and office worker	31.5	22.8
Worker	23.6	21.5
Self employed
Yes	11.2	19.9
No	88.8	22.5
Size of company (no. of employees)
< 10	22	21.9
10–49	22.6	23
50–99	10.6	25.3
100–499	20.4	21.1
> 499	16.4	22.3
No information	7.9	19.6
Activity sector
Industry	12.7	21.3
Building	11.2	20.2
Trading	11.8	20.9
Transport	5.6	26.3
Insurance and real estate	3.4	28
Education, health and social work	23.5	23.9
Other services	31.8	21.2
Duration of job
< 6 months	5.6	19.7
6 months–5 years	33.6	24.1
6–10 years	21.7	21.3
> 10 years	39.1	21.4
Workplace
Individual office	23.1	19.8
Open space	17.5	24.8
Flex office	2.4	33.8
At home	5.9	22.3
Production or workshop	9.1	22.1
External sites	8.3	16.8
At customer	7	26.9
Commercial permises	8.1	22.6
Other	18.6	21.7
Teleworking
Yes	12.4	25
No	81.6	21.8
Work duration (hours per week)
< 39	57.4	22.5
39–50	37.5	20.1
> 50	5.1	34.4**
Working on the week-end
Yes	64.2	23.1
No	35.8	20.5
Night work
Yes	28.5	25.3
No	71.5	21
Staggered hours
Yes	61.4	23.9
No	38.6	19.4
Commuting duration
< 30 min	59	19.6
31 min–1hr	21.9	24.1
> 1hr	19.1	27.9**
Previous experience with unemployment
Never	51	20.3
1 episode	20.8	21.5
≥ 2 episodes	28.2	26.2**
Sickness absence over 1 week due to psychiatric conditions in the last 12 months
Yes	12.7	50.3**
No	87.3	18.1
Chronic disease (> 6 months)
Yes	23.9	36.8**
No	76.1	17.6
Psychotropic medication in the past 12 months
Yes	17.2	47.7**
No	82.8	16.9
Alcohol consumption per week
> 10 glasses	9.2	25.8
> 21 glasses	3.1	21.5
Illegal drug use in the past 12 months
Yes	6.3	44.4**
No	92.9	20.5
Smoking
Yes	24.9	24.8
No	75.1	21.3

Data are proportions (%) for categorical variables and mean ± SD for continuous variables. Differences in proportions were tested by chi-square test. For significant variables with more than two categories, a Pearson residuals Table [(observed–expected)/sqrt(expected)] was computed to identify the modality with a significantly higher proportion for GHQ-28 score ≥ 24. This modality had the higher Pearson residual for GHQ score ≥ 24.

*GHQ-28 = General Health Questionnaire– 28

** indicates significance according to Holm-Bonferroni correction.

The distribution of scores for the GHQ28 is represented in Fig 1. The mean (SD) GHQ-28 score for the whole sample was 18.3 (13.9) and the median score was 15 [14]. With the threshold score of 24 recommended in the literature the prevalence of probable psychiatric disorder was 22.2% (95% CI [20.6; 24.0]) [14]. The proportion of probable psychiatric disorder was increased for women (25.8%) and people with an annual household income < 15 000 euros (30.9%) as well as among people with a chronic disease for more than 6 months (36.8%). This proportion was also increased for people working > 50 hours a week (34.4%), those with a commuting duration > 1 hour (27.9%) and those with ≥ 2 previous experiences with unemployment (26.2%).

Fig 1

Distribution of the GHQ-28 score among the whole sample (n = 3200).

The vertical line represent the cut-off 24 we chose according to literature to identify probable psychiatric disorders (GHQ-28 score ≥24).

Exposure to work-related PSRFs

Fig 2 presents the ranking of the PSRFs according to the absolute difference between the proportion of probable psychiatric disorder among individuals exposed and non-exposed to the PSRFs as well as the proportion of exposure to PSRFs for the whole sample. For all PSRFs, the difference in proportion of probable psychiatric disorder between exposed and non-exposed individuals was significant, except for “The job I do requires that I constantly adapt to new things” and “My job often puts me in contact with clients/users”. These two PSRFs were the most reported, with 75.7% of the sample reporting having a job that required constantly adapting to new things and 70.5% reporting often being in contact with customers/clients/users.

Fig 2

Ranking of the absolute difference between the proportion of probable psychiatric disorders among exposed and non-exposed individuals.

Proportions of probable psychiatric disorders are given for each group (exposed and non-exposed) as well as the proportion of participants exposed in the whole sample. For example, 15.4% of individuals reported having problems handling professional and private life. Among them, 44.6% may be cases as compared with 18.1% among non-exposed individuals. Hence, this psychosocial risk factor is ranked first place with the largest absolute difference (44.6–18.1).

In contrast, “I have problems handling personal and professional responsibilities” was the least reported PSRF (15.4% of the whole sample). However, for this PSRF, the absolute difference between the prevalence of probable psychiatric disorder among exposed (44.6%) and non-exposed (18.1%) individuals was the largest among the 44 PSRFs. The second largest absolute difference in proportion of cases between exposed and non-exposed individuals (21.3 points) was for “My work environment is unpleasant” (22.4% of the sample) and “sometimes I feel afraid when I do my job” (27.8% of the sample). The proportion of probable psychiatric disorder among people reporting an unpleasant work environment was 38.7% and only 17.4% among those who reported a pleasant environment. The proportion of probable psychiatric disorder among people who declared feeling sometimes afraid when they did their job was 37.6% versus 16.3% for those not exposed. The third largest absolute difference (20.3 points) between the proportion of probable psychiatric disorder among exposed and non-exposed individuals was “My job does not makes me feel useful nor gives me self-esteem” (20.7% of the sample) and “I don’t get well along with my hierarchy” (18.7% of the sample). In all, 38.3% of people reporting not feeling useful were probable psychiatric disorder and only 18% among non-exposed individuals. Overall, 39.1% were probable psychiatric disorder among people reporting not getting along with their hierarchy, and only 18.8% among non-exposed individuals.

Association of work-related PSRFs and probable psychiatric disorder (GHQ-28 score ≥24)

The fast-backward procedure identified 11 PSRFs to be included in the model (Table 2) and the results of the adjusted logistic regression of the 11 PSRFs on the outcome GHQ-28 score ≥24 are in Table 2 (S2 Table for the ORs and 95% CIs for the adjustment variables). Ten PSRFs had ORs > 1, with 95% CIs strictly over 1. Three PSRFs had ORs > 1.5: “problems handling professional and private responsibilities” (OR = 1.97 [1.52; 2.54]), “can’t depend on colleagues” (OR = 1.63 [1.29; 2.06]) and “sometimes feeling afraid when doing the job” (OR = 1.53 [1.21; 1.93]). The other PSRFs were “being fearful of the professional future” (OR = 1.44 [1.15; 1.78]), “not satisfied with the job compensation” (OR = 1.42 [1.15, 1.77]), “placed in trying emotional situations during work” (OR = 1.43 [1.13; 1.79]), “unsatisfied with the communication and information exchange within the company” (OR = 1.39 [1.11; 1.75]), “a job that requires long periods of intense concentration” (OR = 1.35 [1.08; 1.70]), “a job consisting of monotonous and repetitive tasks” (OR = 1.29 [1.04; 1.60]) and finally “a job that does not make one feel useful or give self-esteem” (OR = 1.32 [1.03;1.69]).

Table 2

Logistic regression of psychosocial risk factors on the outcome GHQ-28 score ≥24 for employees.

Psychosocial risk factors	OR	95% CI		P value	Proportion of people exposed (n = 3200)
I have problems handling my professional and private responsibilities	1.97	1.52	2.54	<0.001	15.4
I know that I can’t depend on the people I work with	1.63	1.29	2.06	<0.001	27.6
Sometimes I feel afraid when I do my job	1.53	1.21	1.93	<0.001	27.8
I am fearful for my professional future	1.44	1.15	1.78	0.001	43.1
My job puts me into trying emotional situations	1.43	1.13	1.79	0.002	42.0
I am not satisfied with the compensation I receive for my job	1.42	1.15	1.77	0.001	52.9
The communication and information exchange process within my company are not satisfactory	1.39	1.11	1.75	0.004	43.3
My job requires long periods of intense concentration	1.35	1.08	1.70	0.009	63.4
My job doesn’t make me feel useful and nor gives me self-esteem	1.32	1.03	1.69	0.029	20.7
My job consists of monotonous and repetitive tasks	1.29	1.04	1.60	0.021	35.9
My work environment is unpleasant	1.26	0.99	1.61	0.061	30.1

OR = odds ratio; 95% CI = 95% confidence interval

These PSRFs were previously selected by a fast-backward procedure. Odds ratios of the adjustment variables are in S2 Table.

For the interactions between sex and the 11 PSRFs, none was significant (S3 Table). For the interactions between the 11 PSRF themselves, 9 were significant (p<0.05) but none remain significant after Holm-Bonferroni correction (S4 Table).

Discussion

This study allowed us to estimate the prevalence of probable psychiatric disorder (GHQ-28 score exceeding the threshold 24) in the working population in France: 22.2% [95% CI 20.6; 24.0]. This prevalence was congruent with 17.2% of participants having taken psychotropic medication during the 12 last months (antidepressants or anxiolytics) (Table 1). As a comparison, in a meta-analyses of epidemiological data between 1980 and 2013, the prevalence of mental disorders was estimated at 17.6% (95% CI 16.3–18.9) during the 12 months preceding the assessment [19]. This is the first French study made on a representative sample of the working population since previous study focused on specific population [7].

Our study found that probable psychiatric disorder were more numerous among women than men and those with a long commute time, working more than 50 hours a week and earning less than 15 000 euros per year (Table 1), which is consistent with the literature [18,19]. Atypical working time (weekends, staggered hours, night work) affects mental health [18]. In our study, the differences were significant, as for marital status (being alone), but not after correction by Holm-Bonferroni. We decided to use a stringent criterion to decide significance in order to avoid false discovery, and it could have also increased the β risk of ignoring a true difference.

In our study, the proportion of probable psychiatric disorder was higher among people exposed to all PSRFs as compared with those non-exposed, except for two PSRFs: being required to constantly adapt to new things and placed in contact with customers/clients/users. This finding could probably be explained by the unclear meaning of these sentences are ambiguous: it can be positive and negative to be placed in contact with customers for instance.

The imbalance of work and personal life embodied the strongest association with psychiatric disorder (OR = 1.97 [95% CI 1.52; 2.54]). These results are consistent with the literature from several developed countries about work–family conflicts [23-26]. Interpreting these results regarding the exposure, we found 44.6% of probable psychiatric disorder among exposed individuals but 18.1% among non-exposed, but 15.4% of the sample reported being exposed to this PSRF. Even if the exposure level of this factor is the lowest, its association with probable psychiatric disorder was the most important.

Social support in the workplace was associated with having a probable psychiatric disorder (OR = 1.63 [95% CI 1.29; 2.06], 27.6% of the sample being exposed) as well as communication (OR = 1.39 [1.11; 1.75], 43.3% exposed), which confirms the model of Theorell and Karasek [5]. The relevance of the job effort/job demand theory was confirmed by the strong significant association of the cognitive demand of the job (OR = 1.35 [1.08; 1.70], 63.4% exposed), performing monotonous and repetitive tasks (OR = 1.29 [1.04; 1.60], 35.9% exposed) and the importance of having a fair recognition of the work by financial (OR = 1.42 [1.15, 1.77], 52.9% exposed) or symbolic compensation (OR = 1.32 [1.03;1.69], 20.7% exposed) [4]. More than half of the sample was exposed to long periods of intense concentration and unsatisfactory job compensation, both PSRFs with a high OR (about 1.4), so they are critical PSRFs for mental health. For the latter, 28.3% of exposed individuals were probable psychiatric disorder, and only 15.4% were non-exposed.

Experiencing job insecurity (OR = 1.44 [95% CI 1.15;1.78], 43.1% of the sample being exposed) significantly affected the mental health outcome in our study, which is consistent with international findings in Europe and the United States [27-29]. Among people exposed, 32.4% were probable psychiatric disorder, and 14.5% were non-exposed.

Finally, we found an effect of emotional burden at work on mental health (OR = 1.43 [95% CI 1.13; 1.79], 42.0% exposed), which is consistent with other studies [30]. The latter is one of the most-investigated factors because of the increasing literature on burn-out. Since its first description in 1974, burn-out emerged as a specific syndrome related to work, with emotional exhaustion one of the three components [31]. However, it seems difficult to consider burn-out as a global indicator of the mental health status of workers. First, the relevance of distinguishing burn-out from depression by its attribution to work is debated. In fact, there is a strong overlap of its symptoms with those of depression [32]. Second, burn-out has no clear common definition and include a variety of symptoms. Finally, it is not recognized as a mental disorder: it does not appear in any classification (Diagnostic and Statistical Manual of Mental Disorders-5 or International Classification of Diseases, 11th revision). Therefore, we chose a global indicator of mental health, the GHQ-28, rather than a burn-out scale to investigate the association of PSRFs and mental health.

In this study we used a self-reported questionnaire about mental health (GHQ-28) to estimate the prevalence of psychiatric disorders in the sample. We choose a self-assessment instead of a hetero-assessment for feasibility reason (3200 participants). Knowing that only a standardized interview by a clinician can diagnose mental disorder, we use the wording “probable psychiatric disorder”. Moreover, the positive predictive value of the GHQ-28 compared to the Composite International Diagnostic Interview is 48.8% to 66.4% [14]. For France, this PPV was 49.9% in France in 1997 for a threshold lower than our threshold of 24. Hence, for our study, the “true” prevalence of psychiatric disorder would be at least 11%. Considering that the GHQ-28 measures symptoms of depression and anxiety (in fact its four dimensions are somatic symptoms, anxiety and insomnia, social dysfunction and severe depression), this “true” prevalence of our study is consistent with the proportion of 9.9% for depression and anxiety reported by the Organisation for Economic Co-operation and Development [1].

Most of the studies investigating the association of PSRFs and mental disorders are cross sectional. As in our study, they allow for identifying correlations between work-related factors and mental disorders. However, the determination of causality would require a longitudinal study, for instance, with a large cohort, to determine whether the exposure really preceded the mental disorder [18]. This investigation is particularly important for mental disorders such as depression in which negative cognitive bias and dysfunctional thoughts could bias the perception of the environment and thus the subjective restitution of work conditions.

Conclusion

Our study estimated that one in five French workers have a probable psychiatric disorder. Moreover, we identified 10 PSRFs associated with having a probable psychiatric disorder. Among them, work–family conflict was the most important in terms of the intensity of the association, but had the lowest exposure rate in the sample. Lack of support from the company, job insecurity, and emotional burden of the job were also associated with poor mental health outcome, as was unfair job compensation or high cognitive demanding tasks. These results could offer useful suggestions for decision makers to manage and prevent mental disorders in the workplace. However, prospective studies are required to explore the causal effect of the psychosocial factors identified as strongly associated with probable psychiatric disorder and the efficacy of preventive interventions on these PSRFs.

Flow chart of participants in the study.

(PDF)

Click here for additional data file.

Psychosocial risk factors assessed in the study.

(PDF)

Click here for additional data file.

Raw database, weighting factors and weighted database.

(PDF)

Click here for additional data file.

Results of the multiple linear regression estimating the weight of the covariates on the GHQ-28 score for employees and self-employed individuals.

(PDF)

Click here for additional data file.

Interactions between the 11 PSRFs and between sex and the 11 PRSFs.

(PDF)

Click here for additional data file.

(XLSX)

Click here for additional data file.

23 Apr 2020

PONE-D-20-02569

Work-related psychosocial risk factors and prevalence of psychiatric disorders: A cross-sectional study in the French working population.

PLOS ONE

Dear Dr. Chevance,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by Jun 07 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.

A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.

An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Raoul Belzeaux, M.D.

Academic Editor

PLOS ONE

Journal requirements:

When submitting your revision, we need you to address these additional requirements:

1.    Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at http://www.plosone.org/attachments/PLOSOne_formatting_sample_main_body.pdf and http://www.plosone.org/attachments/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please include your tables as part of your main manuscript and remove the individual files. Please note that supplementary tables (should remain/ be uploaded) as separate "supporting information" files

3. Thank you for stating the following in the Competing Interests section:

"“None”: Oumou Daouda, Mounia N. Hocine, Yannick Morvan and Alexandre Salvador report no conflict of interest

Astrid Chevance received a PhD grant from La Fondation pour la Recherche Médicale (FMD20170637634)

Raphaël Gaillard has been a member of a scientific board for Janssen, Lundbeck, Roche, SOBI and Takeda. He has been a consultant or speaker for Astra Zeneca, Boehringer-Ingelheim, Pierre Fabre, Lilly, LVMH, Lundbeck, MAPREG, Otsuka, Pileje, Sanofi, Servier and has received professional fees and has received funding for research from Servier. He is a founding member of Regstem. He is the president of the Fondation Pierre Deniker.

Gilbert Saporta is currently a consultant for Ipsos.

Patrick Légeron is a founding member of Stimulus."

Please confirm that this does not alter your adherence to all PLOS ONE policies on sharing data and materials, by including the following statement: "This does not alter our adherence to  PLOS ONE policies on sharing data and materials.” (as detailed online in our guide for authors http://journals.plos.org/plosone/s/competing-interests).  If there are restrictions on sharing of data and/or materials, please state these. Please note that we cannot proceed with consideration of your article until this information has been declared.

Please include your updated Competing Interests statement in your cover letter; we will change the online submission form on your behalf.

Please know it is PLOS ONE policy for corresponding authors to declare, on behalf of all authors, all potential competing interests for the purposes of transparency. PLOS defines a competing interest as anything that interferes with, or could reasonably be perceived as interfering with, the full and objective presentation, peer review, editorial decision-making, or publication of research or non-research articles submitted to one of the journals. Competing interests can be financial or non-financial, professional, or personal. Competing interests can arise in relationship to an organization or another person. Please follow this link to our website for more details on competing interests: http://journals.plos.org/plosone/s/competing-interests

4.  We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This is a very interesting manuscript, very well written, and bringing original evidence.

This is indeed the first epidemiological French study made on a representative sample of the working population. Representativeness was achieved by using quota sampling. The methodology is very sound and follows international standards. The authors conducted a STROBE-compliant cross-sectional study of a representative sample of the French workers, allowing to estimate the prevalence of probable psychiatric disorders in this population and among workers exposed to work-related psychosocial risk factors.

I have few comments:

- It is surprising to include individuals aged 18 to 80 years, in a study on active workers in a country where the age of retirement is around 60 to 62 years old. Moreover, the study excluded retired people. Nearly 5% of the study sample is aged more than 60. Can you clarify this point please ?

- In this study, 42.6% of participants had a work duration of more than 39h a week, while the legal work duration is 35h a week. Who's is concerned by this ? Workers, independent, managers ?

- One major result is that the imbalance of work and personal life embodied the most strongest association with psychiatric disorder. This is a clearly very relevant result even from clinical point of view, as it is frequently observed in workers suffering from burnout.

Reviewer #2: This is a very interesting study about the estimation of the prevalence of probable psychiatric disorders in the working population, and the assessment of the proportion of people presenting a probable psychiatric disorder among people exposed to work-related psychosocial risk factors. A strength of the study is that it is conducted in a large sample representative of the French population.

Only few points should be clarified.

First, authors should define patients exposed vs unexposed regarding the psychosocial risk factors (PSRF). I think that a patient who is exposed to one PSRF given is a patient with a score of 1 regarding this PSRF in the list of 44 questions. But, authors should write it clearly.

Second, I do not understand well this sentence: “These two PSRFs were the most reported, with 75.7% and 70.5% of the sample reporting having a job that required constantly adapting to new things and 70.5% reporting often being in contact with customers/clients/users.” This sentence should be re written.

Except these two points, this manuscript is very clear with a consistent statistical analysis.

Moreover, results are very interesting.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Wissam El-Hage

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

1 May 2020

Responses to the reviewers:

We thank all reviewers for their careful reading. Their comments greatly helped us improve our work. Please find in blue our comments and in green the modification of the manuscript.

Reviewer #1:

This is a very interesting manuscript, very well written, and bringing original evidence.

This is indeed the first epidemiological French study made on a representative sample of the working population. Representativeness was achieved by using quota sampling. The methodology is very sound and follows international standards. The authors conducted a STROBE-compliant cross-sectional study of a representative sample of the French workers, allowing to estimate the prevalence of probable psychiatric disorders in this population and among workers exposed to work-related psychosocial risk factors.

We thank Reviewer 1 for the positive comments and encouragements.

I have few comments:

- It is surprising to include individuals aged 18 to 80 years, in a study on active workers in a country where the age of retirement is around 60 to 62 years old. Moreover, the study excluded retired people. Nearly 5% of the study sample is aged more than 60. Can you clarify this point please ?

The eligibility criteria were:

- Individuals aged 18 to 80 years AND

- Having a job (e.g. any kind of job, including part-time/self-employment)

Exclusion criteria:

- Students

- Unemployed individuals

- Housewives/husband

- Retired people

A report of the French National Institute for Statistical and Economical Studies from 2018, showed that 5% of people aged [65-74] have an occupation (mainly men, with a good health condition, high education and living in the Parisian area). Top managers but also self-employed and farmers are overrepresented. Four profiles are described: employees with low education with part-time job, people with high education living in big cities, shopkeepers and farmers. (https://www.insee.fr/fr/statistiques/3646000?sommaire=3646226).

To collect data on the senior workers, we included people over the legal retirement age, who declared an occupation. They represented 4.5% of our whole sample (see supplementary table 2 page 3) which is consistent with the national statistics.

- In this study, 42.6% of participants had a work duration of more than 39h a week, while the legal work duration is 35h a week. Who's is concerned by this ? Workers, independent, managers ?

We performed supplementary statistics on this population of workers who reported working more than 39h per week in the table hereunder. Our results are consistent with the report of the French National Institute for Statistical and Economical Studies from 2017 with a mean work duration of 50.5 hours a week for self-employed and 39.1 hours a week for employees. (Please see figure 5 in https://www.insee.fr/fr/statistiques/3676634?sommaire=3696937)

Work duration ≥ 39 hours / week (n=1423) Work duration < 39 hours / week (n=1777)

Sex

Male 62.9 42.9

Female 37.1 57.1

Age, years 42.0 (10.7) 40.9 (11.4)

Number of children

None 50.1 54.5

1 23.4 20.5

2 21.2 18.9

3 4.6 4.9

> 4 0.6 1.2

Highest educational degree

baccalaureate degree 18.2 23.6

Baccalaureate degree 19.2 23.5

Baccalaureate degree +2 years 24.6 23.9

Baccalaureate degree ≥ 3 years 38.0 28.9

Annual household income (euros)

< 15 000 7.8 13.7

15–24 000 17.4 22.8

24–36 000 25.1 27.0

> 36 000 37.7 23.0

No information 11.9 13.5

Occupational status

Independent worker 16.2 7.4

Employees 83.8 92.6

Size of company (no. of employees)

< 10 21.6 22.3

10–49 20.0 24.5

50–99 9.7 11.2

100–499 19.3 21.3

> 499 17.1 16.0

No information 12.3 4.7

Activity sector

Industry 11.1 14.0

Building 15.2 8.3

Trading 12.1 11.7

Transport 5.5 5.6

Insurance and real estate 3.6 3.2

Education, health and social work 19.5 26.4

Other services 33.1 30.9

Duration of job

< 6 months 3.7 7.0

6 months–5 years 33.2 33.9

6–10 years 21.6 21.8

> 10 years 41.5 37.3

Commuting duration

< 30 min 55.6 61.6

30 min–1hr 21.4 22.2

> 1hr 23.0 16.2

- One major result is that the imbalance of work and personal life embodied the most strongest association with psychiatric disorder. This is a clearly very relevant result even from clinical point of view, as it is frequently observed in workers suffering from burnout.

We agree with the reviewer that the imbalance of work and personal life is a major result of the study that echoed clinical practice. In this study we choose to focus on mental disorders and not on burn-out which is not part of the current mental disorder classification (neither the DSM-5 nor the ICM-10), as mentioned in the discussion.

**********************

Reviewer #2:

This is a very interesting study about the estimation of the prevalence of probable psychiatric disorders in the working population, and the assessment of the proportion of people presenting a probable psychiatric disorder among people exposed to work-related psychosocial risk factors. A strength of the study is that it is conducted in a large sample representative of the French population. Only few points should be clarified.

We thank the reviewer for acknowledging our work with positive comments.

First, authors should define patients exposed vs unexposed regarding the psychosocial risk factors (PSRF). I think that a patient who is exposed to one PSRF given is a patient with a score of 1 regarding this PSRF in the list of 44 questions. But, authors should write it clearly.

We add a sentence in the method section (paragraph 2) Measurement of psychosocial risk factors PSRF):

“Therefore, participants with a score of 1 to a given PSRF are called “participants exposed” to this PSRF in the results section.”

Second, I do not understand well this sentence: “These two PSRFs were the most reported, with 75.7% and 70.5% of the sample reporting having a job that required constantly adapting to new things and 70.5% reporting often being in contact with customers/clients/users.” This sentence should be re written.

We corrected the sentence as followed (results section, “exposure to work-related PSRFs):

“These two PSRFs were the most reported, with 75.7% of the sample reporting having a job that required constantly adapting to new things and 70.5% reporting often being in contact with customers/clients/users. “

Except these two points, this manuscript is very clear with a consistent statistical analysis.

Moreover, results are very interesting.

Response to the editor:

- We formated the manuscript according to Plos One standard.

- We decided to make the the data set available (open access) as supporting information.

- We add the specific sentence to the conflict of interest statement.

Submitted filename: Response to reviewers.docx

Click here for additional data file.

6 May 2020

Work-related psychosocial risk factors and psychiatric disorders: A cross-sectional study in the French working population.

PONE-D-20-02569R1

Dear Dr. Chevance,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Raoul Belzeaux, M.D.

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Reviewers' comments:

12 May 2020

PONE-D-20-02569R1

Work-related psychosocial risk factors and psychiatric disorders: A cross-sectional study in the French working population.

Dear Dr. Chevance:

I am pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please notify them about your upcoming paper at this point, to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

For any other questions or concerns, please email plosone@plos.org.

Thank you for submitting your work to PLOS ONE.

With kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Dr. Raoul Belzeaux

Academic Editor

PLOS ONE