The relevance of socio-demographic and occupational variables for the assessment of work-related stress risk

The assessment, prevention and control of work-related stress risk are widely recognized as one of the major challenges to occupational health and safety [1]. Work-related stress is the second most common work-related health complaint among workers in the European Union (EU) and recent findings of large European surveys are consistent with a tendency to an increasing prevalence of work-related stress risks [2, 3]. This might largely be due to the rapid changes in the last few years in technology and work organization, with increasing use of non-traditional employment contracts and non-standard working hours, but also to other socio-economic and demographic factors indirectly influencing the world of work, such as economic globalization, an aging workforce and the increasing numbers of women entering the workforce.

There is also ample evidence that high-risk conditions for work-related stress have a detrimental impact on workers’ health and safety, with a clear relation in particular with cardiovascular diseases [4], mental [5] and musculoskeletal disorders [6] and workers’ psychosocial well-being [7, 8]. Work-related stress can also affect an organization’s productivity and competitiveness, with further impact on social and economic costs [1, 9].

In the last decade several studies have provided evidence of certain key indicators of the social and occupational determinants of health [10, 11]. As a consequence, there has been widespread pressure for the inclusion of those social and individual factors as determinants of workers’ health and, according to Marmot [11], of health inequalities.

Increasing importance is being attached to socio-demographic and occupational variables in the assessment and management of work-related stress risk [1, 2]. In a large multicenter European study, gender differences in work-related stress perception appeared stable across occupational groups but varied according to the geographical provenance [12]. Different patterns of generation of job stress were also described in men and women [13]. An inverted U-shaped relationship was found between age and job stress [14]. Emotional workload in younger people and lack of social support in older employees were associated with a higher risk of mental health complaints [15]. Older age and longer length of service were also associated with higher levels of stress and emotional exhaustion in a group of family physicians [16]. Low occupational status and educational level were associated with higher work-related stress risk in a large sample of French workers [17]. A large U.S. survey suggested that education could mask the effects of work-related stress on health, possibly in a gender-specific way [18]. Only limited evidence was found that shift-work with nights increased the effects of work stress on some health outcomes [19]. A recent critical review reported a negative association between short-term contracts and stress, very likely reflecting the efficiency of national welfare systems [20].

Despite the growing knowledge about these issues, further investigation is needed on the integrated role of socio-demographic and occupational factors on work-related stress risk. The aim of this study was to explore the interaction effect among such characteristics and the perception of work-related stress risk, using a validated instrument. The study started from a larger research project to develop an integrated method for the assessment of work-related stress risk based on the UK Health and Safety Executive (HSE) Management Standards (MS) approach [21, 22] and reflecting the EU and Italian legislative and regulatory frameworks. The HSE Indicator Tool (IT) questionnaire has been proposed to evaluate seven areas of work design corresponding to specific risk factors for work-related stress: demands, control, managerial support, peer support, relationships, role, and change[21‐23]. After a first step in which the Italian version of this questionnaire was validated [24], a study was conducted on a sample of over 6,000 Italian workers in order to examine the relationships between the seven MS areas and a number of workers’ demographic and occupational characteristics. Specifically, the primary aim was to verify both main and the interactions effects of socio-demographic and occupational factors on the seven Indicator Tool risk factors for work-related stress and to produce explanatory hypotheses to test in the future. This analysis will also identify the socio-economic, demographic and occupational variables to be considered for the definition of homogeneous worker groups in the assessment of work-related stress, as required by national and EU legislation.

Table 1 describes the socio-demographic and occupational characteristics of the 6,378 workers by gender. The male to female ratio was 1.47. More than 60% of workers of both sexes were aged 31 to 50 years and most were married. About a third worked on shifts and about 16% had flexible work patterns (part-time and/or fixed-term work). Fixed-term job contracts and mainly part-time work were more common for women (respectively 12.3% and 27.5%) than men (7.2% and 3.6%).

Table 2 gives the descriptive statistics (mean and standard deviation) for the seven HSE IT areas. Mean scores for men ranged from 3.36 (change) to 4.35 (role), for women from 3.31 (change) to 4.36 (role). There were no significant differences by gender except for managerial support, relationships and change.

Women workers gave significantly lower scores on control and peer support scales and had more negative perceptions of relationships and change at work than men (Table 3), mainly among those with full-time contracts (Table 4). The oldest group (>50 years) gave a higher score for control than those younger than 30 years. Both married and divorced/widowed people felt job demands more negatively than unmarried workers, though unmarried subjects had worse scores for control and role (Table 3).

Workers with high school or university education reported more positive perceptions of control but only those with a master degree had more negative perceptions of job demands than the less educated people. Blue-collar workers and clerks/managers scored less on control and reported more negative perceptions of relationships and the management and communication of change, but blue-collars perceived even lower job demands than managers. A positive correlation emerged between control and occupational status, with a mainly large difference between blue-collar workers and managers (β = -0.51) (Table 3). Blue-collar workers and clerks with university education had less positive perceptions of demands than those with elementary/middle education, but higher scores in control. Blue-collar workers with a master degree had negative perceptions of their role compared to others whereas managers with higher education seemed to perceive a clearer role. A high educational level played a positive role in relation to relationships among white-collar workers (Table 4).

There was a weak positive correlation for role and control scores and job seniority, reaching significance only in the most experienced group (>17 years in the current job). Workers with more than five years in the current job reported more negative perceptions of demands, relationships, peer and managerial support. Fixed-term contract workers had more positive perceptions of demands and relationships than permanent workers but felt their role less clearly (Table 3).

As regards commuting time, workers who spent more than an hour traveling to and from work had more negative perceptions of all the factors except role, than those who spent less than an hour (Table 3).

The negative effects of shift-work on all subscales except role was evident from the interaction with age; the scores were significantly lower for shift and/or night workers, mainly in the oldest groups (Table 4). Control and change scores were inversely related to age. Night work was significantly associated with negative perceptions of demands and relationships, particularly among the youngest workers. Those with individual missing data had more negative perceptions of the seven IT mean scores.

This study explored the relationship between workers’ individual characteristics and perceived risk for work-related stress measured using the Italian validated version of the HSE IT. A large cross-sectional survey of Italian companies was designed, covering a large part of the national economic sectors. The findings confirmed that both socio-demographic and occupational characteristics significantly influence work-related stress risk factors.

In compliance with Italian regulations, analysis of these characteristics contributes to the identification of “homogeneous” groups of workers for the assessment and management of work-related stress risk. The identification of particularly sensitive groups of workers, considering suitable individual and organizational aspects, is a key issue for the prevention policy recently introduced in EU and Italian legislation.

The study sample was in line with the composition of the overall Italian workforce, as compared with national statistics. In Italy in 2010 about 5% of male and 30% of female employees worked part-time and the proportions of fixed-term workers varied from 11% for men to 15% for women. The percentage of subjects doing shift-work was comparable to that obtained from EU surveys among Italian workers [3].

As our findings confirmed, women had more negative perceptions of work-related stress risk factors than men, especially the full-time workers. This could well be due to gender role differentiation at work [28]. Traditionally, women have primarily a caretaking role and bear a higher burden of family-related responsibilities than men and that can make them more likely to feel strained by their job, particularly when they have trouble balancing work and family commitments, as in the case of full-time work [26]. In fact, by rebalancing work and family commitments through part- time work, the gender difference in perceived risk for work-related stress turned out to be irrelevant. It is worth noting that welfare deficiencies, exacerbated today by economic difficulties and the financial crisis, may also contribute to increasing the work-family imbalance.

We noted that married workers of both sexes perceived higher work demands. This too is likely to reflect the higher burden of family responsibilities on married workers compared to unmarried ones. As reported elsewhere [29], the weight of family responsibilities might lead to a potential work-home conflict that could reduce the individual’s ability to cope with increasing demands at work. Unmarried workers perceived less autonomy in their job and greater role ambiguity than married ones. These findings are original but are very likely due to the closer social integration and acknowledgment of married people than single ones in the Italian social and cultural contexts.

A significant interaction was found between educational level and occupational status, since blue-collar workers with the highest level of education (i.e. master degree) perceived more job demands and role ambiguity. These findings might be explained by a phenomenon known as over-qualification or under-employment, a condition in which people have a level or type of education or skill that exceeds their job requirements [30]. The current changing nature of the world of work has resulted in increased unemployment and, in some cases, in large numbers of over-qualified workers. There is evidence of the effect of under-employment on workers’ attitudes at work and physical and psychological health [27]. Moreover, blue-collar workers with higher education may perceive job demands and role ambiguity more negatively also because they are likely to be assigned more demanding jobs than their colleagues on account of their education and specialization.

Job seniority showed a positive relationship with control and role scores, confirming that the more experience a person has, the greater their autonomy and role awareness in their job [31]. At the individual level, a higher job tenure might balance the burden of responsibilities at work, emerging as higher perceived job demands. It is also worth noting that in our sample workers with higher seniority perceived a poorer quality of relationships with supervisors and colleagues and less social support at work than less senior ones. While job seniority increases, employees likely become more autonomous and the perceptions of supervisor and colleagues support may decrease. Furthermore, it is like we assist to a deterioration process of the relationships at work over time, but deeper investigation should test this hypothesis. Shift-workers perceived greater job demands, less autonomy at work and more conflicting relationships in the workplace, in line with previous findings [32]. Working on shifts and -- even more -- working nights does call for a high degree of adaptation on the biological, social and behavioral levels [33], it can affect social relationships and lead to potentially important work-home interference [34]. However, shift-work appeared to be associated with slightly higher role clarity at work. Interpretation of these findings is complex and multifaceted, possibly including the influence on the clarity of role and the absence of potential role conflicts. As for the possible combined effects of shift-work and aging on workers’ health and wellbeing, data are still contrasting. We found an interactive effect of aging and shift-work on most work-related stress risk areas. The negative perceptions among older workers might reflect their lower tolerance of shift-work, mainly involving greater sensitivity to physical health disorders and sleep problems [35]. However, age did not seem to boost the detrimental effect of shift-work, and particularly of night work, on relationships and support at work, possibly because of the higher experience among elderly workers [36].

The relationship between commuting time and perceived risk for work-related stress is widely acknowledged, like the association with physical and psychological health outcomes such as musculoskeletal disorders, cardiovascular diseases, anxiety and sleep disorders [37‐39]. Long commuting time to work also reduces productivity and job satisfaction, mostly through enhancement of the perception of work-related stress risk factors [40, 41]. In line with the literature, our findings confirm the negative role of commuting particularly on demands and control (where the management of working hours is extremely important) and on social support and relationships at work. No negative influence of commuting time was found on workers’ perceptions of their role (role clarity and conflict). This is presumably due to the fact that role is more associated with aspects related to the job design than to the job schedule.

Finally, those with a fixed-term contract had more positive perceptions of demands and relationships at work but less positive perceptions of their role than permanent workers. International studies on job contracts, fixed-term work conditions and health reported contrasting findings [20]. Several studies have indeed demonstrated that other factors (e.g. welfare policies, more flexibility in organizing work) may moderate the negative consequences on workers’ health related to fixed-term employment. However, a fixed-term contract may, in some cases, make workers feel less involved in their company, with negative consequences for role clarity.

Our findings illustrate the importance of socio-demographic and occupational variables for work-related stress risk factors. The most susceptible groups were female workers, for whom part –time work is one way of reducing the difficulties of balancing work and family commitments, better-educated workers, especially those who are over-qualified for the jobs assigned, and shift workers, especially older ones.

Finally, commuting time was confirmed as having an effect on work-related stress risk, whereas part–time work, especially for women, and fixed-term contracts had potentially protective effects.

It may be reasonable to conclude that the assessment and management of work-related stress risk should take into account the relations with all the social, demographic, and occupational factors involved. Analysis of these variables is particularly important in the light of the European legislation on safety and health at work, concerning the obligation to assess the risk of work-related stress, because it can provide useful information as a basis for identifying specific groups of workers at risk. We would warmly suggest focusing on interventions aimed at managing work-related stress risk in these groups, such as job scheduling, transportation demand, training for less experienced workers, and strategies for reducing work-home conflict.