Rise in psychosocial risk factors at the workplace

Increasingly diverse working conditions

2005 Working Conditions Survey

Working conditions surveys are carried out by the Research, Analyses and Statistics Department (Direction de l’Animation de la Recherche, des Études et des Statistiques, DARES) of the Ministry of Labour, Social Relations, Family Affairs, Solidarity and Urban Affairs (Ministère du Travail, des Relations Sociales, de la Famille, de la Solidarité et de la Ville) since 1978. They are repeated every seven years – the last four working conditions surveys date from 1984, 1991, 1998 and 2005. The collated answers refer to working conditions as perceived by the respondents who belong to all sectors of the economy, including the public sector.

The surveys are conducted at home on a representative sample of the population in metropolitan France aged 15 years or over and in employment. They supplement the Labour Force Survey (Enquête Emploi) published by the National Institute for Statistics and Economic Studies (Institut de la Statistique et des Études Économiques, INSEE) – see Annex for more details on survey methodology.

In the various survey waves, the questionnaire has undergone many changes. Very focused on the world of industry in 1978, the Working Conditions Survey broadened its scope in 1991 to describe mental load factors as well as workers’ room for manoeuvre in task execution through demands such as ‘must leave a task for another’, ‘unable to change deadlines’ and ‘what to do in case of unexpected event’. In 1998, the survey was enriched with questions pertaining to the fragmentation or unpredictability of work schedules and accidents at the workplace. In 2005, new questions appeared concerning the mental load at work, the prevention of occupational risks and the ability to carry out the same job until the age of 60 years.

Employment status impacts on working conditions

According to a recent analysis on working conditions and precarious employment (in French) published by DARES in July 2009, some 27% of workers were in precarious employment in 2005 (see methodology note in Annex for a definition of precarious employment). The breakdown of workers in precarious employment is as follows:

• 10% hold fixed-term employment contracts;
• 2% hold temporary work agency contracts;
• 4% feel underemployed and would like to work more;
• 10% hold an open-ended or permanent employment contract but fear losing their job within the year.

Working conditions change according to a worker’s employment status. Fixed-term and temporary agency contract holders have less autonomy in their work and less readily receive help from their co-workers, even if they are not required to work quite as hard. They have more often a variable work schedule, just like part-time workers with an open-ended contract who would like to work more. For workers with an open-ended contract who fear for their job, working conditions are characterised by limited autonomy, hard work and many physical burdens.

Generally, workers in precarious employment do not attend to usual risk prevention trainings as easily as workers in stable employment.

The various forms of precarious employment under examination apply to diverse population groups and career profiles:

• workers on fixed-term employment contracts tend to be young people, women, less qualified workers employed in offices, trade or other services – notably personal services, social work and administration;
• temporary agency workers tend to be young and are mostly employed in construction and manufacturing;
• underemployed workers who hold an open-ended employment contract tend to be women with low qualifications working in the services sector, notably in personal services;
• vulnerable workers with an open-ended employment contract tend to be in the intermediate age group and work in the manufacturing industry. In economic sectors on the decline, such as the textiles industry, up to 40% of workers with an open-ended employment contract are in a vulnerable employment situation. The proportion amounts to 25% in industry sectors facing strong competitive issues, such as metalworking, domestic appliances, electrical and electronic components.

Atypical work schedules linked to particular demands

While 37% of workers have a ‘normal’ working day, almost two out of three wage earners work on a so-called ‘atypical’ work schedule (refer to Annex for definition). Some 19% of the working population is affected by regular night or weekend work, and 10 % by occasional night and weekend work. These workers, and more specifically those working regularly on weekends or at night, suffer from several working schedule demands such as irregular work weeks and the lack of a 48-hour weekly rest period. Overall, 80% of these workers are in contact with the public, typically in healthcare, social work, policing, trade, and hotels and restaurants. Employees working regularly at night and on weekends must also bear the many demands of work organisation – such as the need to hurry up, lack of resources and no support from managers – as well as physical risk factors. Some 60% of workers working regularly on weekends or at night are exposed to more than three demands, compared with 41% in the rest of the working population surveyed.

Night workers are exposed to particularly difficult working conditions as outlined in Table 1.

Notes: (*) Among the following risk factors: frequently walking long distances, moving heavy loads, painful or tiring positions, prolonged standing, painful or tiring movements, sustaining jolts or vibrations, being unable to hear someone two or three meters away.

(**) Among the following demands: keeping eyes on work at all times, reading small type characters, monitoring hard to detect visual or sound signals.

Source: INSEE-DARES, Working Conditions Survey, 2005

For 7% of workers, working hours vary according to the time of year. Those concerned work mainly in economic sectors with seasonal peaks – agriculture, food processing, construction and consumer goods. Their working conditions are characterised by tight schedules and a rapid work pace, along with little employment security (fear of losing their job within the year).

More than 10% of workers face long working weeks, exceeding 40 hours a week, characterised by flexible working schedules and work impinging on non-working life (they may be contacted while being out of work). These workers are mainly managers and technicians, and mostly men. They have both high work demands – time pressures and objectives to achieve – and high control over their own work.

Finally, 17.5% of workers – 5% of men and one third of all women – work part-time, they are concentrated in services and jobs in contact with the public. One third of part-time workers wish to work more and they generally have poorer working conditions and a less advantageous employment status – that is, atypical employment contracts, irregular work schedules and a high demand for flexibility as compared to full-time workers (see the EWCO article on Considerable variations in the situation of part-time workers).

Another survey on employment offers and recruiting (in French) (see Annex on methodology), carried out in 2005 by DARES and INSEE, provides many additional elements on the respective situations of men and women in the labour market in France (Garner and Magnien, 2009). In 2005, men accounted for 59% of workers recruited on an open-ended employment contract, and 66% of workers recruited were aged between 30 and 39 years. Women are more often hired for part-time work: some 46% of women hired on fixed-term employment contracts and 31% on open-ended contracts are working part-time, compared with respectively 16% and 6% of men. Overall, women have higher qualifications than men: 60% of women recruited on an open-ended employment contract hold a secondary level qualification or above, compared with 43% of men. Regarding open-ended employment positions, women fall behind from the very beginning of their career. Some 25% of women under the age of 30 years who are university degree holders – two years or more of education at a third level institution – find employment in low-qualified blue-collar or white-collar occupations, compared with 18% of men.

A gender bias found in career guidance at school and in female recruitment for certain sectors of the economy and occupations – teaching, personal care, reception, accounting and administration – partly explains this situation. One quarter of employers surveyed also show a lack of gender neutrality while recruiting – in 70% of cases, they indicate preferring to hire men than women.

Some jobs viewed as unsustainable

Based on the results of the 2005 Working Conditions Survey, DARES compiled a study (in French) on the psychosocial factors impacting on workers’ perception of whether they will be able to do their current job when they reach 60 years of age (Coutrot, 2008; see also FR0901039I highlighting the study’s findings). The study revealed a small gender difference with 36% of female and 32% of male respondents replying ‘no’ to the question ‘Do you feel able to do your current job at the age of 60 years?’. This question was introduced for the first time in the 2005 Working Conditions Survey. Figure 1 highlights the distribution by gender and occupational categories.

When work entails physical strain, more respondents would find it difficult to carry on doing the same job until they reach 60 years of age. However, other job characteristics also contribute to those feelings, namely psychosocial risk factors: work that offers no opportunity to learn, inadequate resources to do the job well, tensions with the public or management, irregular work schedules (for men) or frequent overtime without any compensation (for women).

Figure 1: Feeling unable to do the current job at the age of 60 years, by gender and occupational category (%)

Source: INSEE-DARES, Working Conditions Survey, 2005

Feeling unable to do the current job at the age of 60 years, by gender and occupational category (%)

Impact of psychosocial factors in the workplace

The growing impact of psychosocial risk factors in the workplace is the second major development to emerge from recent working conditions surveys.

The 2003 edition of the SUMER Medical Monitoring of Risks (Surveillance Médicale des Risques) survey, also carried out by DARES (see Annex for methodological details), included psychosocial exposure measurements by means of an additional questionnaire (see the EWCO 2006 review of working conditions in France for a summary of the survey findings). The questionnaire results were further analysed in publications by DARES in 2007 and 2008.

Perceived impact of job strain on health

Job strain – as defined by the ‘Karasek model’ – is a health risk in which autonomy at work is too limited to allow workers to live up to work demands as they see them. To measure autonomy at work, Karasek referred to the degree of control or decision latitude of workers (refer to Annex for more details on the Karasek model and questionnaire).

As illustrated in Figure 2, women are more at risk to suffer from job strain than men.

Figure 2: Psychological demands and control at work, by gender and occupational category

Note: → Men;  Women.

The graph is structured around two axes representing the median values of the psychological work demands, on the one hand, and workers’ control – that is, their autonomy – on the other. The highest risk of job strain occurs in jobs characterised by high psychological demands and low control or decision latitude.

Source: Guignon et al, DARES, May 2008, p. 2 (based on data from the 2003 SUMER survey)

Psychological demands and autonomy at work, by gender and occupational category

Low-skilled white-collar and blue-collar workers are also more affected by job strain than high-skilled workers because they do not enjoy the same level of autonomy at work.

Job strain is a scientific definition, and it cannot be used as a synonym for ‘stress’ as it is used in daily language: a large proportion of workers define themselves as being exposed to a high level of stress even if their situation is not strictly that of ‘job strain’. This is mainly the case for professionals and managers, who face strong work demands but have a high level of autonomy.

However, workers who face high psychological demands at work, but have low levels of autonomy at work run a higher risk of job strain. A greater proportion of workers subjected to job strain declare being in poor health – 22% compared with 12% of the total working population. These workers are also more likely to feel that ‘their work is not good for their health’ – 43% compared with 28% in the working population as a whole.

Beyond these two dimensions – psychological demand and control or decision latitude – other psychosocial factors have a measurable impact on work-related health outcomes. In particular, the various forms of harassment and violence that workers experience at work.

According to an analysis of the SUMER survey findings (in French) (Bué and Sandret, 2008), psychological violence or bullying – consistently hostile behaviour coming from one person or more – affected 17.3% of the employed population in 2003, with 18% of working women being affected compared with 16.3% of working men. Bullying occurs through disrespectful conduct (6.5% of the workforce), denying work achievements or competence of the worker (9%) or degrading attacks (2%).

Hostile conducts tend to develop in environments with high work demands from superiors and a high work pace. Also, those behaviours are more likely to occur if workers judge work organisation to be at fault and indicate lack of support from their co-workers and superiors.

These situations impact on workers’ perceptions of health at work and give rise to the feeling that ‘working is bad for health’ in the same proportions as job strain exposure.

As more workers increasingly come in contact with the public during work – 70.9% in 2003 compared with 63.3% in 1994 – new work situations arise such as the development of atypical work schedules, work emergencies, many interruptions but also more decision latitude.

In general, women work in contact with the public more often than men – 80% of women compared with 63% of men. In this instance, they get verbally and physically assaulted by the public more often than men (Table 2).

Source: Bué and Sandret, DARES, April 2007 (based on data from the 2003 SUMER survey)

Postal workers, bank employees, security personnel and health professionals are the highest occupational risk categories. Assault from the public is more common for workers subjected to atypical work schedules, high work intensity and a demanding work organisation (Bué and Sandret, April 2007; Nicot, June 2006).

Mental health at work under epidemiological monitoring

Besides working conditions issues, psychosocial factors in the workplace constitute a public health challenge and a field of research that needs to be further developed.

At the national level, the French Institute for Health Surveillance (Institut de veille sanitaire, InVS) launched the Samotrace programme in 2006. The programme aims to estimate the prevalence of mental health disorders according to employment and its associated occupational exposures (see Annex on methodology).

The initial results of the chapter on epidemiological monitoring in the workplace were published in early 2009 (covering a limited area, please refer to annex for the methodological details). What emerges is a prevalence of psychological distress at work among 24% of men and 37% of women. An imbalance in terms of effort reward at work and an over-commitment in workers are significantly linked to psychological distress regardless of sex. Furthermore, psychological distress is also linked to the exposure to violence and harassment – either physical or psychological – and to work practices that go against professional ethics (see InVS publications (in French) and EWCO article on Workers in contact with public more exposed to aggression).

Psychosocial factors at work and public health

In a broader perspective, work has become one of the health factors analysed in the large public health surveys of the general population such as the 10-year health survey of INSEE or the Health Barometer of the National Institute for Prevention and Health Education (Institut national de prévention et d’éducation pour la santé, INPES). As a result, when mental health is examined, useful links with occupation can be explored since the large number of respondents makes a representative sample of the working population.

The Mental Health in the General Population Survey provides an insight on the prevalence of mental health disorders and their occupational consequences.

The outcome is as follows:

• mood disorders – bouts of depression, dysthymia, manic episodes – concern about 11.2% of men and 15.9 % of women in the general population (respectively 9.9% and 14.2% in the economically active population);
• anxiety disorders – generalised anxiety disorder (GAD), agoraphobia, social phobia, panic attacks and post-traumatic stress disorder (PTSD) – affect 17.4% of men and 25.4% of women in the general population (respectively 17.3% and 25.7% in the economically active population);
• alcohol issues – addiction and binge drinking – affect 7.2% of men and 1.5% of women in the general population (respectively 7.4 % and 1.6 % in the economically active population);
• drug issues concern 3.9% of men and 1.2% of women in the general population (respectively 3.2% and 1.1% in the economically active population).

Whatever the pathology under study, the less well-off occupational categories, namely low-skilled blue-collar and white-collar workers, are affected the most. They are generally followed by self-employed craftspeople and shopkeepers.

Workers suffering from mental health disorders took sick leave with proportions ranging from 12% (GAD for men) to 37% (dysthymia for women) . Irrespective of gender, workers in managerial positions are the most likely to take sick leave, mainly for anxiety disorders or depression. In the craft and retail trade sectors, self-employed workers seldom take time off work, in spite of the significant discomfort felt by many.

It is worth noting that these results paint a picture of the situation but do not provide clues as to underlying occupational causes of the troubles observed.

The INPES Health Barometer 2005 sheds some light on the links between health and psychosocial factors in the workplace.

The survey results show that women as a group have significantly lower health scores than men as a group; these findings hold true for all scores: physical health, mental health or general health.

Professionals and knowledge workers have a better general health score than unskilled workers, machine operators and skilled workers. On the other hand, professionals, knowledge workers, clerical workers together with services and sales workers are worse off in terms of mental health.

While 83% of the workforce indicate relatively high levels of job satisfaction, 80.1% express feelings of work strain, either physical strain or moral strain. Regarding occupational stress, 63.9% of the employed workforce report being affected or stressed by their work. To cope with stress, the following strategies are enacted:

• 44.4% of workers declare eating more than usual – often or occasionally;
• 77.9% of smokers are prompted to smoke more;
• 17.5% of men and 6.4% of women declare having increased their alcohol intake;
• 13% of men and 22.9% of women declare resorting to psychoactive medication;
• 3.6% declare using other – illicit – drugs.

Over half of the people who feel they suffer from occupational stress indicate that they had sleeping trouble in the previous eight days – 51.4% compared with 33% for the non-stressed population. Some 5.6% of people suffering from occupational stress mention that they had thoughts about suicide, while 8.6% had a serious case of depression during the previous 12 months, compared with respectively 3.1% and 5.6% of the non-stressed population.

In 2008, the growing concern over psychosocial risks at work prompted the government to set up a task force to propose a set of monitoring indicators on this issue. Experts will publish their conclusions by the end of 2009.

References

Bué, J. and Coutrot T., Direction de l’Animation de la Recherche, des Études et des Statistiques (DARES), Horaires atypiques et contraintes dans le travail : une typologie en six catgories, Premières Informations Premières Synthèses, No. 22.2, May 2009.

Bué, J. and Sandret, N., DARES, Contact avec le public : près d'un salarié sur quatre subit des agressions verbales, Premières Informations Premières Synthèses, No. 15.1, April 2007.

Bué, J. and Sandret, N., DARES, Un salarié sur six estime être l’objet de comportements hostiles dans le cadre de son travail, Premières Informations Premières Synthèses, No. 22.2, May 2008.

Cohidon, C., Prévalence des troubles de santé mentale et conséquences sur l’activité professionnelle en France dans l’enquête Santé Mentale en Population Générale : images et réalités, Paris, InVS, August 2007.

Cohidon, C., Arnaudo, B., Murcia, M. et al, ‘Mal-être et environnement psychosocial au travail : premiers résultats du programme Samotrace, volet entreprise, France’, Bulletin Épidémiologique Hebdomadaire, InVs, No. 25–26, June 2009.

Coutrot, T., DARES, Se sentir capable de faire le même travail jusqu’à 60 ans : le poids des facteurs psycho-sociaux, Premières Informations Premières Synthèses, No. 29.2, July 2008.

Garner, H. and Magnien, M., DARES, Hommes et femmes recrutés en 2005 : les préférences des employeurs, Premières Informations Premières Synthèses, No. 31.3, July 2009.

Guignon, N., Niedhammer, I. and Sandret N., DARES, Les facteurs psychosociaux au travail. Une évaluation par le questionnaire de Karasek dans l’enquête SUMER 2003, Premières Informations Premières Synthèses, No. 22.1, May 2008.

Institut de veille sanitaire (InVS), reports (in French) available online at: http://www.invs.sante.fr/publications/default.htm.

Juanico, R., European Working Conditions Observatory (EWCO), A review of working conditions in France, Dublin, European Foundation for the Improvement of Living and Working Conditions (Eurofound), March 2006.

Karasek, R. A., ‘Job demands, job decision latitude, and mental strain: implications for job redesign’, Administrative Science Quarterly, Vol. 24, 1979, pp. 285–308.

Karasek, R. A. and Theorell, T., Healthy work: stress, productivity and the reconstruction of the working life, New York, Basic Books, 1990.

Nicot, A.-M., EWCO, Expectations about working capacity at age 60, Dublin, Eurofound, January 2009.

Nicot, A.-M., EWCO, Workers in contact with public more exposed to aggression, Dublin, Eurofound, June 2006.

Léon, C. and Ménard, C., ‘Activité professionnelle et santé, quels liens ?’, in Beck, F., Guilbert, P. and Gautier, A. (ed.), Baromètre santé 2005 Attitudes et comportements de santé, Paris, INPS, 2007, p. 415–444.

Leymann, H., ‘The content and development of mobbing at work’, European Journal of Work and Organizational Psychology, Vol. 5, 1996, pp. 165–184.

Niedhammer, I., David, S. and Degioanni, S., ‘Violence psychologique au travail et santé mentale : résultats d'une enquête transversale en population salariée en région Paca, France, 2004’, Bulletin Épidémiologique Hebdomadaire, InVS, No. 25–26, June 2009.

Niedhammer, I., David, S., Degioanni, S. and 143 occupational physicians, ‘Association between workplace bullying and depressive symptoms in the French working population’, Journal of Psychosomatic Research, Vol. 61, 2006, pp. 251–259.

Niedhammer, I., David, S., Degioanni, S. and 143 occupational physicians, ‘Economic activities and occupations at high risk for workplace bullying: results from a large-scale cross-sectional survey in the general working population in France’, International Archives of Occupational Environmental Health, Vol. 80, No. 4, 2007, pp. 346–353.

Rouxel, C., DARES, Conditions de travail et précarité de l’emploi, Premières Informations Premières Synthèses, No. 28.2, July 2009.

SUMER survey questionnaire and methdological details, available online at (in French): www.travail-solidarite.gouv.fr/etudes-recherche-statistiques-dares/statistiques/sante-au-travail/enquetes/sumer.html.

Working Conditions Survey questionnaires and data, available online at (in French): www.travail-solidarite.gouv.fr/etudes-recherche-statistiques-dares/statistiques/conditions-travail/.

Annex: Survey methodologies

Working Conditions Survey 2005

The 2005 Working Conditions survey is the first survey to complement the permanent Labour Force Survey (LFS). The questionnaire was submitted to all persons in employment among the respondents of the sample population, amounting to 19,000 people in 2005.

The working conditions surveys aim to figure out the reality of work as perceived by workers and not as described or stipulated by the company or the employer. To this end, the survey is carried out at workers’ homes and each worker in the household answers the questionnaire individually. The questions asked do not mention objective measurements – such as quotations or ergonomic studies – nor do they tackle personal views on work, the focus is on a concrete description of the job, its organisation and conditions from various angles: what is prescribed, autonomy, cooperation, work rate, physical strain and risks incurred.

Questions are as factual as possible. For instance, on the topic of noise: workers indicate whether they can hear someone two or three meters away from their workstation without them raising their voice, thus allowing to spot exposure to a noise level of 85 decibels.

Further, the questionnaire attempts to grasp certain dimensions of work by cross-checking questions. The study of work rate strains relies both on a question about assembly line work and a review of possible determinants of pace of work.

Understanding the level of autonomy in the workplace presupposes assessing both the prescriptive nature of instructions and to which extent workers can manage incidents on their own. Regarding physical or mental strain at work, reference to thresholds and figures were avoided on purpose. The idea is to collect the interviewee’s estimation regarding physical exertion, exposure to extreme temperatures and risk of errors or work intensity.

Definition of job insecurity in Working Conditions Survey analysis

A report on defining indicators in matters of employment, unemployment and job insecurity by a workgroup from the National Council for Statistical Information (Conseil national de l’information statistique, CNIS) underlines that precarious employment is a ‘difficult notion to work out’.

The Working Conditions Survey provides no direct indicator of a worker’s employment instability. The analyses quoted here chose a varied approach, using the criteria set out in the CNIS report: they reviewed the type of employment contract (fixed term, temporary work), dissatisfaction with working time (underemployment) and feeling insecure about employment (fear of losing one’s job within the year). Underemployment is incorporated because job dissatisfaction is a contributing factor of job instability – often ending with worker resignation.

Definition of atypical work schedule in Working Conditions Survey analysis

The Working Conditions Survey analysis establishes the five groups of workers through an automatically generated ‘hierarchical classification’ of full-time workers. The classification is based on the following 17 variables, for which these groups present homogeneous features:

• working 40 hours or more during the week preceding the interview;
• working everyday or often overtime without any compensation in the form of wage or time off in lieu;
• being contacted by phone for professional reasons during non-working time (more than 10 times in the past year);
• being subjected to on-call work;
• taking work home (every day or often);
• starting work before 07.00;
• ending work after 20.00;
• working usually at night;
• working occasionally at night;
• working usually on Saturdays;
• work occasionally on Saturdays;
• working usually on Sundays;
• work occasionally on Sundays;
• being unaware of working time schedule for the next week;
• free working time schedules;
• working time schedules determined by employer;
• during the year, weekly working time shorter on some periods and longer on others.

Survey on employment offers and recruiting 2005

The survey targeted private sector companies which had initiated a recruitment process for at least one employee. The questionnaire aims to describe what took place in a particular recruitment process involving a particular person from the point of view of the employer.

The companies were sampled from the company register SIRENE (Système d’Identification du Répertoire des Entreprises et de leurs Établissements) run by INSEE, then selected by telephone on the basis that they had run a recruitment process in the 12 months preceding the survey. A total of 4,052 face-to-face interviews were carried out over metropolitan France between January and July 2005. In 546 interviews out of 4,052 the last worker recruited was a manager, 3,038 interviews concerned a last non-manager hire and in 468 cases the last recruitment attempt had failed. The sample is stratified by company size and sector of activity (32 strata).

The survey was designed by the Ministry of Labour’s Research, Analyses and Statistics Department DARES, the National Employment Agency (Agence nationale pour l’emploi, ANPE now Pôle Emploi), the Centre for Employment Studies (Centre d’études de l’emploi, CEE) and the Research Centre on Qualifications (Centre d’études et de recherches sur les qualifications, Céreq).

SUMER 2003 – Medical Monitoring of Risks

The SUMER Medical Monitoring of Risks (Surveillance Médicale des Risques) survey was jointly launched and coordinated by the Occupational Health Department (Inspection médicale du travail) of the General Labour Directorate (Direction générale du travail, DGT) and DARES. It drew up an inventory of worker exposure to the main occupational risks in France. The strength of this survey is based, on the one hand, on the expertise of occupational doctors (whom the workers visit regularly) who can manage highly technical questionnaires and, on the other hand, on the very large sample of workers interviewed which allowed quantifying exposure to relatively uncommon risks. The survey was carried out from June 2002 until the end of 2003, involving 1,792 occupational doctors, that is to say more than 20% of all practicing occupational doctors who drew lots to select 56,314 workers, of whom 49,984 responded.

In 2003, the survey covered employees and workers in the agricultural sector, workers employed in public hospitals and some large public organisations; however, it did not take into account public sector employees and the armed forces. Out of 21.7 million workers, 17.5 million or 80% of workers are represented in the SUMER 2003 survey.

Survey questionnaires

In 2003, a self-completion questionnaire known as the Karasek questionnaire was added to the main questionnaire filled out by occupational doctors. The self-completion questionnaire was offered to one out of every two workers surveyed. Some 24,486 people chose to answer the self-completion questionnaire. Particular adjustments had to be made because the shape of answers left blank differed significantly from the main questionnaire.

The so-called ‘Karasek questionnaire’ is an internationally validated tool to assess the following three dimensions of work: psychological demand, level of control and social support.

The psychological demand of work is assessed through a set of nine questions referring to:

• work load and pace of work;
• complexity and intensity of work;
• fragmentation of work and unpredictability.

A worker’s level of control is evaluated through nine questions covering three subject areas:

• decision latitude or ‘room to manoeuvre’;
• use of skills or competences in job;
• development of competencies.

Regarding social support, the questionnaire distinguishes between emotional and professional support given by colleagues and/or superiors.

Besides the Karasek questionnaire, the SUMER survey included questions concerning:

• appreciation of the reduction of working time;
• self-assessment of health, job and the influence of work on health;
• sick leave and occupational accidents;
• violence in the workplace and hostile behaviour.

The next issue of the survey in 2009 will be expanded to include questions on additional psychosocial exposures at work, as well as on mental health through a scale of anxiety and depression symptoms.

Samotrace Programme

The Samotrace programme – a monitoring system for mental health at work – consists of three separate projects developed in two pilot geographic areas: France’s Centre region and surrounding areas, and the Rhône-Alpes region in southeast France.

The first instrument is the monitoring of workers’ health in companies in order to obtain mental health indicators according to occupation and sector of activity, as well as evaluations of exposure to risk factors by occupational doctors participating in a voluntary capacity.

Part of the second instrument is compiling a knowledge base of monographs on mental health. The purpose of this instrument is to provide a detailed qualitative analysis of mental health suffering and psychopathological disorders of workers. A network of occupational doctors with experience in mental health and the writing of monographs on this subject lead this part of the programme.

The third instrument of the programme focuses on recording medical and administrative data on all cases of workers who have been declared disabled on the ground of mental health problems. This part of the project relies on the cooperation of the consultant-doctors of the National Health Insurance Fund (Caisse nationale d’assurance maladie, CNAM).

Regarding the monitoring of workers’ health in the Centre region, the findings of which are presented in this report, the population sample has been selected by drawing lots among all economically active people in employment. Some 110 occupational doctors supervised the procedure. The population sample comprises a total of 6,056 employees, 57% of whom are men. The General Health Questionnaire (GHQ28) was used to assess workers’ mental health. The occupational psychosocial factors under scrutiny related to effort-reward imbalances and overinvestment in work, as well as to working in a way that goes against work ethics.

Mental Health in the General Population Survey

Between 1999 and 2003, the international transversal survey on Mental Health in the General Population: images and realities (Santé mentale en population générale : images et réalités, SMPG) was conducted in France by the Northern Association for Psychiatric Epidemiology (Association septentrionale d’épidémiologie psychiatrique, ASEP) together with the national collaborative centre of the World Health Organization (WHO) (Centre Collaborateur de l’Organisation Mondiale de la Santé pour la recherche et la formation en santé mentale, CCOMS) in Lille in northern France. The survey aims to assess the common view of mental health in the general population by investigating the perception of mental disorders as well as of the care services and professionals in place. Moreover, the survey seeks to estimate the incidence of different psychiatric disorders. The survey questionnaire also included several questions on employment.

SMPG is an international transversal survey run by many centres. The French part of the survey was carried out in cooperation with the Direction of research, evaluation and statistics of the Ministry of Labour (Direction de la recherche, des études, de l’évaluation et des statistiques, DREES). The population sample in France included a total of about 36,000 French individuals (selected with the quota methodology) and is representative of the French population aged over 18 years in terms of gender, age, level of education, occupation and employment status.

In charge of analysing mental health according to the occupational characteristics provided by the survey, the occupational health department of the French Institute for Health Surveillance (Institut de veille sanitaire, InVS) took the opportunity to provide a first inventory of the incidence of mental health disorders in different occupations in France. First, the InVS analysis outlines the incidence of certain mental health disorders according to employment status and occupational categories. Secondly, it examines the impact of mental health disorders on working life.

Health Barometer 2005

The Health Barometer is a repeated transversal phone survey, based on a 30,000-strong random sample of the general population between 18 and 75 years of age. Over nearly 15 years, Health Barometers have followed the main behaviours, attitudes and perceptions regarding risk-taking attitudes and health of the population residing in France: use of tobacco, alcohol and other drugs; vaccination practices; sexual behaviour; cancer screening; physical activity; quality of life; and healthcare use.

Until now, Health Barometers studied health determinants which were limited to social, economic and demographic characteristics. Starting with Health Barometer 2005 and in collaboration with DARES, new variables concerning working conditions and additional questions were introduced to expand the questionnaires and the scope of the analysis.

Anne-Marie Nicot, National Agency for the Improvement of Working Conditions (Agence nationale pour l’amélioration des conditions du travail, ANACT)