France: Working conditions and occupational risks: SUMER 2010

About the survey

The Medical Monitoring Survey of Professional Risks (Surveillance médicale des expositions aux risques professionnels, SUMER (in French)) examines working conditions in France. SUMER’s main aim is to assess employees’ exposure to harmful working conditions and to analyse appropriate protection mechanisms. Thus one of the survey’s main functions is to identify occupational risks and to develop an agenda for the prevention of the most common threats. SUMER is conducted jointly by the General Directorate of Labour (DGT) and the Directorate for Research, Studies and Statistics (Dares) of the Ministry of Work, Employment, Vocational Training and Social Dialogue, and co-financed by the General Directorate of Administration and the Public Sector (DGAFP).

The study consists of interviews with employees conducted by the company medical officer during their regular compulsory medical examination This approach allows Dares to draw on the medical expertise of the doctor in charge and to ask employees to complete a questionnaire that is described as being technically challenging in parts. Another advantage of this methodology is that it permits access to a large population, allowing measurement of occurrences of relatively rare phenomena.

Fieldwork for the third edition of SUMER was carried out between January 2009 and April 2010. In addition to face-to-face interviews with the medical officer, employees filled in on their own a paper-and-pencil questionnaire dealing with their personal experience and psychosocial factors. All employees were asked to do this in 2010, with a response rate of 97% of participants, whereas only one out of two were asked in previous editions. Medical officers that were unsure about employees’ responses could ask Dares to investigate the workplace further.

More than 20% of all medical officers (2,400 people) were randomly selected to participate in the survey in 2010. They examined 53,940 employees in the period covered, with 47,983 (89%) taking part in the survey. On average, each doctor returned 22 questionnaires. The sample is deemed to be representative for the roughly 22 million employees in France.

Compared with earlier editions, the 2010 survey (in French) draws on a wider range of employees, most notably from the public sector. Whereas older versions covered essentially private sector employees (SUMER 1994 (in French)) or only some parts of the public sector (SUMER 2003 (in French)), SUMER 2010 included a large range of groups in state-owned companies and some 40% of employees in central administration, excluding all those working in national education, the Ministry of Social Affairs and the Ministry of Justice.

Despite these changes the results are comparable over time, according to a Dares report (in French, 1.28MB PDF) published in February 2013 detailing differences in exposure to occupational hazards between sector identified by SUMER 2010, although all time comparisons in the report use only the subpopulation of private sector employees. An earlier Dares report on SUMER 2010 (in French, 151KB PDF), which was published in March 2012, presents some time comparisons in less detail and is discussed in another Eurofound report on the impact of the crisis on working conditions (to be published in 2013).

All the data presented here are taken from the Dares report published in February 2013. These data have been weighted according to:

• individual characteristics of the medical officer;
• correlation between frequency of visits and frequency of exposure to hazardous working conditions (French labour law states that employees in unsafe environments must be examined more frequently);
• extent of non-response;
• characteristics of the employees and their workplace (gender, age, nationality, type of working time, socioprofessional category, sector and workplace size).

This report is divided into three broad subsections:

• work organisation (including working time);
• behavioural factors;
• exposure to potentially harmful working conditions.

Work organisation

In 2010, 18% of French employees worked more than 40 hours per week (Error! Reference source not found.). The notable difference in this indicator between men (23%) and women (11%) might be due to a greater proportion of women working part time – 20% compared with only 9% of men in 2010 according to data (in French, 61KB MS Excel) from the National Institute of Statistics and Economic Studies (Insee). Nevertheless, the figure is interesting given that statutory weekly working time in France is 35 hours (FR0001137F).

Other substantial gender differences were reported for the following indicators:

• long consecutive hours without break;
• night work, which is particularly present in male-dominated industry employment but also in public hospitals (27%) and central public administration (20%);
• uncertainty about future working time.

Note: * Including part-time work

Source: Arnaudo et al (2013, Table 1)

Many of the ‘atypical’ forms of working time arrangements, such as on-call duty, night work, work on Sundays or shift work, were found to be more common in the public sector than in the private sector. However, uncertain working time (not the same working hours every day, not knowing the working hours for the next week) was far less likely in the public sector than in the private sector.

The private sector had further substantial sectoral differences (Table 2). The short break indicator, for instance, seemed to be inflated by a high proportion of employees with long working hours without breaks in the transport and commerce sectors as well as to a lesser extent in other services. Sunday work was also significantly more likely in the former. On-call duty was substantially more common in agriculture than in any other sector. Construction work, on the other hand, was characterised by low rates of employees working at night, on Sundays or in shifts.

Note: * Including part-time work

Source: Arnaudo et al (2013, Table 1)

While these findings suggest that stressful working hours are more frequent in the public sector, the scores for work intensification indicators were mostly higher in the private sector. For example, 40% of private and 33% of public sector employees reported at least three constraints to their working rhythm (automatic replacement of a product or piece, the rhythm of a machine, other technical constraints, external demand that requires immediate response, production norms or deadlines of one hour or more, control or supervision through information technology systems). The most affected sectors were industry (50%) and commerce and transport (43%), followed by public sector hospitals (41%).

In addition, Arnaudo et al (2013) used Karasek’s (1979) model to determine psychological demands (the mental load caused by tasks) and job decision latitude for different sectors (Figure 1). The analysis shows moderate gender differences with regard to job decision latitude, whereas scores are roughly equal for the psychological demands of the job. Outliers in terms of either indicator are employees in agriculture (low job demands), construction (high decision latitude) and public hospital workers (high job demands).

Figure 1: Karasek scores for psychological job demands and job decision latitude (%)

Source: Arnaudo et al (2013, Graph 1)

A common concern of employees in all sectors (67%) was that their individual error could have severe consequences for the product or service. Not surprisingly, this was more common in the industry (78%), public hospitals (75%) and construction (73%) sectors. Workers in the construction and industrial sectors and, to a lesser extent, in agriculture were also most likely to expect their company to suffer a substantial financial loss when they make an error at work (74%, 74% and 65%, respectively and 56% for all employees). Further consequences of individual errors include health and safety concerns, which are particularly present in public hospitals (69%), agriculture (57%) and construction (56%). The overall percentage for all employees was 39%. Finally, roughly half of all employees (49%) feared the personal consequences of individual errors. This indicator was particularly high in public hospitals (65%).

However, surveyed employees expressed widespread concern about the feasibility of doing their job correctly, with 19% stating that there were insufficient employees to perform the tasks assigned. This proportion was considerably higher in the public (25%) than in the private sector (17%). In public hospitals, workers most often reported not to have sufficient amounts of material necessary for their job (27% compared with 17% in the whole workforce). In both central and local government, employees declared more often not having clearly stated and sufficient information (21% and 17%, respectively, and 15% for all employees) or sufficient and appropriate training (22% each and 16% for the entire workforce) to perform their tasks. According to DGAFP (2012), fewer days are spent on training in local government than in the public sector as a whole. In central public services, however, research shows that there are some flaws in skills management and insufficient use of the skills available (Cordier and Larmat, 2009).

Despite the recorded difficulties in performing tasks, strict quality control procedures were reported to apply to 47% of all employees, with the most notable outliers being public hospitals (76%) and the industrial sector (65%). In addition, roughly a third of the surveyed employees stated they had met precise quantitative targets with above average rates in industry (43%), commerce and transport (40%) and central public administration (36%). In general, men were more likely to have to attain quantitative targets than women.

Individual performance was assessed at annual review meetings for 56% of French workers, though again there was a substantial difference between the public and the private sector (85% and 50%, respectively). Annual individual assessment meetings are compulsory in the French public sector following legislation in 2002, 2007 and 2009.

Behavioural factors

Some 29% of all employees stated that they did not receive sufficient support in difficult situations, 15% reported that they did not receive the respect they deserved from their colleagues and 31% perceived the support from their superiors as insufficient (Table 3). There was relatively little variance in these indicators across sectors. Employees in agriculture and industry were slightly more positive about the support they received, while workers in the industrial sector were less satisfied than the average (Table 4). Men were slightly more likely to report lack of support than women (Table 3).

Source: Arnaudo et al (2013, Table 4)

Source: Arnaudo et al (2013, Table 4)

Moreover, 28% of the employees reported experiencing an unwanted change at the time of the survey (Table 3); the proportion is marginally higher in the public than in the private sector (32% and 28%, respectively). The figure is suspected to be inflated by the impact of the result of the general revision of public policies (Révision générale des politiques publiques, RGPP). RGPP was launched in 2007 and sought to achieve a far-reaching reduction of expenditure in the public sector (FR0904039I). As a consequence, the proportion of employees experiencing unwanted change was particularly high in central and regional public administration (40% and 32%, respectively) and generally higher in the public than in the private sector (32% and 28%, respectively) (Table 3). This indicator was also relatively high in industry (33%) (Table 4). The latter might reflect substantial restructuring in the sector, which might also account for the high proportion of employees who perceived their job security to be under threat.

Some three-quarters of the surveyed employees stated they were in regular contact with the public (Table 3). Not surprisingly, this was more likely in sectors that involved dealing with people in the broader sense (public hospitals, local public services, private sector services) and less likely in sectors such as agriculture and industry. Although it is expected that contact with the public might lead to tensions, this was only the case for 9% of employees (Table 3). Employees in public hospitals were most frequently affected (17%).

However, indicators that measure occasional abuse when dealing with the public score substantially higher. Public hospital employees were particularly concerned by occasional verbal abuse from their patients (29% compared with 15% of the whole workforce). The same group was also more likely to experience occasional abuse from colleagues (18% and 11% of all employees). In the private sector, service sector workers were most likely to be subject to abuse from the public (18% in transport and commerce, and 16% in other services). Those employed in industry and construction seemed to be least affected (6% each) (Table 4).

Exposure to potentially harmful working conditions

Exposure to high levels of noise

The sectors most exposed to noise were construction, industry and agriculture with exposure rates to noise levels below 85 dB(A) of 57%, 39% and 38%, respectively (Figure 2). These sectors also scored highest with regard to exposure to higher noise levels (that is, above 85 dB(A), which is considered as dangerous and requires the use of protective equipment). Since the sectors concerned are characterised by a dominantly male workforce, it is not surprising that men were generally more often exposed to noise than women. Noise exposure in the public sector was reported to be limited.

Figure 2: Exposure to noise by noise level, by sector, gender and industry (%)

Exposure to severe physical constraints

Similarly, severe physical constraints were reported to be more frequent in the construction sector and agriculture, where the rate of employees exposed to at least one severe physical constraint was 62% and 54%, respectively (38% in the total workforce). Severe physical constraints are defined as:

• working in an upright position for 20 hours or more per week;
• manual carriage of a load for 20 hours or more per week;
• repetitive movements for 10 hours or more per week;
• vibrations transmitted to other parts of the body for 10 hours or more per week;
• postural constraints for at least two hours per week (kneeling, arms in the air, cowering or bending over).

Other sectors with high exposure to severe physical constraints were commerce (45%) and public hospitals (44%). The latter was also the only sector in which the exposure of women was roughly the same as that of men. Male blue-collar workers in the private sector were most affected by physical constraints (Arnaudo et al, 2013).

Sectors less affected by these indicators may, however, have other forms of stressful working conditions. For example, the majority of surveyed employees used computers (53%) and 22% spent more than 20 hours per week in front of a computer screen or monitor. Central public administration and other private sector services were highest for this (31% and 30%, respectively), while women were more likely to spend a large proportion of their working week at a computer than men (27% and 18%, respectively).

Exposure to hazardous chemicals

SUMER data also cover exposure to chemical and biological risks, with a third of French employees stating in 2010 that they worked with chemical agents (Table 5). Sectors with high exposure to chemical risks were construction (61%), industry (45%) and agriculture (43%) (Table 6), plus also employment in public hospitals (55%). In addition, 29% of employees in the construction sector and 25% of employees in public hospitals were exposed to at least three chemical agents (Arnaudo et al, 2013, p. 10). Solvents (ethanol or another alcohol) were reported to be used frequently in hospitals (37% of employees).

Perhaps the most alarming finding was that almost a third of construction workers stated that they were exposed to at least one carcinogenic chemical agent. Agents most often mentioned by medical officers were diesel exhaust fumes, petroleum, wood dust and crystalline silica. Finally, long exposure to chemical agents for at least 10 hours per week was recorded frequently in construction (25%) and industry (19%). Long exposure to chemical agents was reported by 9% of the overall workforce.

In contrast to the predominant exposure of the private sector and men to chemical agents, women and public sector employees were considerably more likely to be in contact with biological agents at work (Figure 3). This is mainly a consequence of high exposure rates in public hospitals where more than three-quarters of respondents (76%) reported using biological agents, followed by agriculture (37% compared with 22% in the workforce as a whole). A more detailed analysis of the data suggests that the most common activities with exposure to biological agencies were in the health sector, social work, home care, hotels, restaurants and catering (Horeca), gardening, and waste management (Arnaudo et al, 2013, p. 11).

Figure 3: Exposure to biological risks, by sector, gender and industry (%)

Source: Arnaudo et al (2013, Table 6)

Source: Arnaudo et al (2013, Table 6)

Source: Arnaudo et al (2013, Table 6)

References

Arnaudo, B., Léonard, M., Sandret, N., Cavet, M., Coutrot, T. and Rivalin, R. (2012), L’évolution des risques professionnels dans le secteur privé entre 1994 et 2010: premiers résultats de l’enquête SUMER (151KB PDF), Dares Analyses 2012 No. 023, Dares, Paris.

Arnaudo, B., Léonard, M., Sandret, N., Cavet, M., Coutrot, T., Rivalin, R. and Thierus, L. (2013), Les risques professionnels en 2010: de fortes différences d’exposition selon les secteurs (1.28MB PDF), Dares Analyses 2013 No. 010, Dares, Paris.

Cordier, M. and Larmat, A. (2009), ‘Les conditions de travail des personnels du secteur hospitalier public et de la fonction publique de l’État en 2006’, in DGAFP (ed.), Rapport annuel sur l’état de la fonction publique 2008–2009, Volume 1, La documentation Française, Paris, Chapter 2.

DGAFP (Directeur Général de l’Administration et de la Fonction Publique) (2012), Rapport annuel sur l’état de la fonction publique: Édition 2012, La documentation Française, Paris.

Karasek, R.A. (1979), ‘Job demands, job decision latitude, and mental strain: Implications for job redesign’, Administrative Science Quarterly, Vol. 24, No. 2, pp. 285–308.

Sebastian Schulze-Marmeling, IRShare