How to improve performance and prevent burnout in safety representatives.

BACKGROUND: The present work is part of a greater research project, aimed to examine Safety Representatives' (SRs) role, twenty years after the appointment of this figure.
OBJECTIVES: The study aims to investigate the role of some personal and organizational dimensions in the promotion of SRs' well-being, in terms of reducing burnout and improving performance.
METHODS: The study involved 455 SRs operating in North East Italy. They completed a self-report questionnaire, regarding conflicts with co-workers, ethical conflict, training satisfaction, work engagement, performance, and burnout.
RESULTS: Structural equation models show that work engagement partially mediates (γ=-0.52, p<0.001; β=0.23, p<0.01) the relationship between conflict with co-workers and performance (γ=-0.26; p<0.01), as well as partially mediating (γ=0.14, p<0.05; β=0.23; p<0.01) the relationship between training satisfaction and performance (γ=0.21, p<0.001). Moreover, it totally mediates the relationship between conflict with co-workers and burnout (γ=-0.52, p<0.001; β=-0.40, p<0.001), as well as totally mediating the relationship between training satisfaction and burnout (γ=0.14, p<0.05; β=-0.40, p<0.001). Finally, ethical conflict is positively associated with burnout (γ=0.047, p<0.001).
CONCLUSIONS: This study provides useful information about the improvement of SRs' well-being, highlighting the importance of their involvement in this role.

Introduction

The economic downturn over the last decade has brought to the forefront the topic of work and its protection, intended as the right to health and safety in the workplace. Many studies show that a company investing in its employees’ well-being is a future-oriented business, more reliable for the customer and more competitive in the global market (45, 23, 52). In this view, the fight against accidents and occupational illnesses takes on ethical resonance and a truly strategic socio-economic meaning.

The polymorphic issues of health and safety at work require multi-disciplinary skills, which, in turn, call for an ongoing dialog between the various parties involved in health and safety. These include employers, managers, people in charge of prevention and protection services, health and safety representatives (SRs), occupational physicians, and employees, in order to broaden the participatory perspective as much as possible. They all work jointly in the health and safety management system (HSMS) − a set of elements designed to achieve health and safety organizational goals which involve the organizational structure, human and productive resources, internal rules, working practices, procedures, and responsibilities (7). The HSMS integrates health and safety goals into the planning and management of the production systems, pursuing the policy of prevention in accordance with the health and safety rules. According to OHSAS 18001 (12), the HSMS is based on Deming’s cycle and consists of the well-known four steps: plan, do, check, and review (80).

In Italy, the legislative decree 81/2008 is a milestone for health and safety at work (for an accurate comparison between the two decrees, see: 81). SRs are elected or designated to represent workers’ health and safety interests. They are chosen among trade union representatives, to prevent the proliferation of representative bodies and to avoid the weakening of employees’ defence (50, 75). SRs may face critical situations, some of which concern the relational dimension.

SRs also encounter the twofold ethical conflict deriving, on the one hand, from the possible medley of the two representative roles (e.g. using their role as a trade-union negotiation device) and, on the other, from the fact that SRs may experience ethical issues in terms of conflict of interest. Indeed, they hold a boundary position, bridging the gap between employer and employees. This status could require a choice between employer’s expectations and employees’ needs. Even though the SRs’ mandate is to represent these needs, the possible repercussions on their work life pave the way for ethical conflicts. Boundary spanners share knowledge across different actors: they simplify information flows in order to reduce possible asymmetry and connect individuals (61). The effort to conciliate incongruent expectations and the relational nature of their role, which demands frequent personal interactions, concern, and care for co-workers, turns the organization into an emotionally-charged environment. Daily job activities thus require extra effort, which creates the basis for emotional distress, frustration, and anxiety (79). The latter, if protracted, can increase burnout risk (31).

Furthermore, Italian law requires SRs to participate in training activities and refresher courses. In particular, training sessions must have a minimum duration and be carried out during working hours. The SRs’ right to an ad hoc-training on health and safety is also endorsed. The training is to cover certain topics: e.g. national and Community legal principles, health and safety legislation, awareness of the actors involved and their duties, risk definition and identification, risk assessment, identification of preventive and protective measures, regulatory issues in relation to the representation activity, and basic principles of communication techniques.

The psychological literature evidences that workplace conflicts and training opportunities influence employees both in a negative and in a positive manner. Relational conflict reduces employees’ well-being in terms of work engagement (83), performance (19, 63), and burnout (5, 51, 87). Ethical conflict increases burnout and reduces work engagement and performance (41, 47, 88). Conversely, training opportunities are important, because they can increase employees’ resources (58, 68) and well-being (2, 30). Consequently, satisfaction with the training received entails positive outcomes, such as better performance (11, 10, 20), reduced burnout (9, 49, 76), and greater work engagement (1, 35, 42). The latter can increase performance (4, 36, 77, 82) and limit the effects of burnout. Indeed, even though some Authors argue that work engagement and burnout are two poles of the same continuum (57), other studies highlight the distinctiveness of the two constructs (18, 37, 71, 73). Based on this evidence, we believe that work engagement can contribute to clarify the relations investigated.

The literature offers models to explain these relations, such as the Demand-Control Model (DCM; 48), the Effort-Reward Imbalance (ERI; 78), and the Job Demands-Resources Model (JD-R model; 21). The JD-R Model is among the most useful and flexible models (6, 69, 72). It divides job characteristics into two categories: job demands – that require efforts and are associated with costs for the individual – and job resources – that stimulate personal growth, while fostering goal achievement. The former triggers a health impairment process, with significant negative outcomes on strain levels. The latter attenuate the negative consequences mentioned and give rise to a motivational process, which enhances positive outcomes – such as performance – through greater levels of work engagement.

Hence, our desire to investigate the relationships between conflict with co-workers, ethical conflict, and training satisfaction on the one hand, and performance and burnout on the other, hypothesizing the mediating role of work engagement (figure 1).

Figure 1

Hypothesized model

This study present work is part of a greater research project (16) aimed to examine SRs’ role twenty years after the appointment of this figure, to identify the actions to be promoted to overcome critical issues. The project involves SRs working in manufacturing, the services industry, schools, and healthcare in the Veneto Region1.

Methods

Procedure and Participants

We did not have a list of the local SRs. Therefore, we considered companies in the Province of Padua, located through the archive of the INAIL (Istituto Nazionale per l’Assicurazione contro gli Infortuni sul Lavoro, National Institute for Insurance against Accidents at Work) Information Flows relating to 2011, based on the assumption that each company would have at least one SR. We then carried out a simple random sampling.

The local workplace prevention, hygiene, and safety agency called every company in advance to ascertain the presence of a SR and explain the aim of the project to employers. Companies that did not have SRs were replaced through a random procedure.

Then, the agency sent a letter to each company explaining the purpose of the project and how to take part in it. The letter included an anonymous questionnaire to be filled in by the SR and a personal invitation underlining the importance of every single SR’s contribution. A self-addressed stamped envelope was enclosed for the receiver’s convenience.

The questionnaire, consisting of several scales, was individually completed by SRs. To reduce apprehension during the evaluation, participants were reminded that there are no right or wrong answers, and were encouraged to provide assessments as close to their own opinions as possible.

Four hundred fifty-five questionnaires were completed by SRs working in companies in North East Italy and in charge of employee-safety representation activities. The sample was made up of 79.3% men and 20.7% women. 70.0% of participants were 31 to 50 years old, 25.0% were over 50, and 5.0% were under 30. For an accurate description of the sample, see table 1.

Table 1

Sample characteristics

	N	Valid %		N	Valid %
Gender			Duration of service as SR
Male	348	79.3	Up to 5 years	215	48.9
Female	91	20.7	6 to 15 years	189	43.0
Missing	16		15 years or more	36	8.2
Total	455	100.0	Missing	15
			Total	455	100.0
Age			Sector
31 to 50	308	70.0	Industry/craftsmanship	287	65.7
Over 50	110	25.0	Services	88	20.0
Under 30	22	5.0	Trade	62	14.2
Missing	15		Missing	18
Total	455	100.0	Total	455	100.0
Educational level			Company size
High school	247	56.3	16 to 50 workers	143	32.6
Elementary junior high school	140	31.9	5 to 15 workers	120	27.3
University	52	11.8	51 to 250 workers	110	25.1
Missing	16		Over 250 workers	35	8.0
Total	455	100.0	Up to 4 workers	31	7.1
			Missing	16
			Total	455	100.0

Measures

The instructions for completing the questionnaire invited participants to answer referring to their specific SR role.

Conflict with co-workers was assessed with four items taken from the Qu-Bo test (17), anchored on a 6-point Likert scale (1=very much disagree, 6=very much agree). A sample item is the following: “There is a great deal of willingness to provide mutual support”. The Cronbach’s alpha for this scale is 0.65.

Ethical conflict was assessed with two items taken from the Qu-Bo test, anchored on a 6-point Likert scale (1=very much disagree, 6=very much agree). A sample item is the following: “My organization consistently takes into account customers’ needs”. The Cronbach’s alpha for this scale is 0.79.

Training satisfaction was assessed by asking: “How satisfied are you with the contents of the training received as a SR?” and “How satisfied are you with the duration of the training received as a SR?” Items were answered on a 6-point Likert scale (1=very dissatisfied; 6=very satisfied). The Cronbach’s alpha for the scale is 0.79.

Work Engagement was assessed with the Utrecht Work Engagement Scale (UWES-9; 70). The 6-point response scale ranged from 1 (very rarely) to 6 (very often). A sample item in this scale is: “At my work, I feel bursting with energy”. The Cronbach’s alpha for the scale is 0.92.

Burnout was assessed with eight items taken from the Qu-Bo test. The 6-point response scale ranged from 1 (very rarely) to 6 (very often). A sample item in this scale is: “My work is useless”. The Cronbach’s alpha for the scale is 0.84.

Self-evaluated performance was assessed with two items (28) answered on a 10-point Likert scale (from 10% to 100%). A sample item is: “We would kindly ask you to specify, using a percentage, how successful you were in reaching your work goals last year”. The Cronbach’s alpha for the scale is 0.87.

Statistical Analyses

We tested the hypotheses estimating structural equation models (SEM) with latent variables, using the Lisrel 8.80 software (46). SEM is a technique that allows the observation not only of the indicators directly measured by the researchers, but also of the latent variables, which are inferred indirectly (22). The fundamental difference is that the observed variables contain a measurement error, while the latent factors do not. Because most of the variables are normally distributed and the deviations from normality are not too extreme, we applied the Maximum Likelihood method, which is the most widely used fitting function for SEM (74).

To assess the model fit, we used the chi-square test (χ2). A model shows a good fit to the data if χ2 is nonsignificant. Given that χ2 is sensitive to sample size, we considered additional fit indices (39), representing different classes of goodness-of-fit criteria, as recommended by Mueller (62). In particular, we considered the normed fit index (NFI), the standardized root mean residual (SRMR), and the root mean square error of approximation (RMSEA). Values close to or greater than 0.90 for NFI, values close to or smaller than 0.10 for SRMR, and values close to or smaller than 0.08 for RMSEA indicate an acceptable fit (74).

To verify the significance of the indirect effects, 95% asymmetric confidence intervals were considered, based on the distribution of the multiplication term. We used this technique to manage the non-normality derived from the path a * path b multiplication, as recommended by MacKinnon’s procedure (PRODCLIN; 54, 56). If the confidence interval does not contain zero, the mediation is significant (55).

Finally, to obtain a more parsimonious solution, we fixed all non-significant paths to zero.

Before analyzing the data, we excluded participants with missing values. Consequently, the final sample consisted of 353 participants.

Results

Variables means, standard deviations (SD), and Pearson’s correlations are presented in table 2.

Table 2

Variables means, standard deviations, and Pearson’s correlations

Variable	M	SD	1.	2.	3.	4.	5.	6.
1. Conflict with co-workers	2.20	0.82	1.00
2. Ethical conflict	2.53	1.01	0.25**	1.00
3. Training satisfaction	4.58	0.91	-0.20**	-0.11**	1.00
4. Work engagement	4.38	0.81	-0.34**	-0.37**	0.24**	1.00
5. Burnout	1.65	0.66	0.40**	0.23**	-0.15**	-0.41**	1.00
6. Self-evaluated performance	6.76	1.62	-0.30**	-0.29**	0.29**	0.37**	-0.24**	1.00

N=353, ** p<0.01

Conflict with co-workers has a mean of 2.20 (SD=0.82); ethical conflict has a mean of 2.53 (SD=1.01); training satisfaction mean is 4.58 (SD=0.91); work engagement has a mean of 4.38 (SD=0.81); burnout mean is 1.65 (SD=0.66); finally, self-evaluated performance has a mean of 6.76 (SD=1.62).

We then estimated the hypothesized structural model. Fit indices show an acceptable adaptation to the data, considering χ2(89)=268.90, p=0.00; NFI=0.92; SRMR=0.06; RMSEA=0.08. In the model, conflict with co-workers is negatively associated with work engagement (γ=-0.52, p<0.001) and performance (γ=-0.22, p<0.05). Ethical conflict is positively associated with burnout (γ=0.56, p>0.001). Training satisfaction is positively associated with work engagement (γ=0.15, p<0.05) and self-evaluated performance (γ=0.21, p<0.001). Work engagement is positively associated with self-evaluated performance (β=0.25, p<0.05) and negatively associated with burnout (β=-0.49, p<0.001). The following paths are not statistically significant: conflict with co-workers – burnout (γ=-0.20, p>0.05); ethical conflict – work engagement (γ=0.007, p>0.05); ethical conflict – self-evaluated performance (γ=-0.04, p>0.05); training satisfaction – burnout (γ=-0.003, p>0.05).

We also verified the significance of the four specific indirect effects: conflict with co-workers on self-evaluated performance and on burnout, and ethical conflict on self-evaluated performance and on burnout. The asymmetric confidence interval for the relationship between conflict with co-workers and self-evaluated performance, through work engagement, does not contain zero; the unconventional estimate is -0.57, 95% CI [-1.028, -0.209]. We can now conclude that work engagement mediates the relationship between conflict with co-workers and self-evaluated performance. In addition, when controlling the effect of work engagement, the negative relationship between conflict with co-workers and self-evaluated performance remains significant (γ=-0.22, p<0.05). Therefore, we conclude that the mediation is partial.

The asymmetric confidence interval for the relationship between conflict with co-workers and burnout through work engagement does not contain zero; the unconventional estimate is 0.42, 95% CI [0.200, 0.701]. Thus, we can conclude that work engagement mediates the relationship between conflict with co-workers and burnout. Moreover, when controlling the effect of work engagement, the positive relationship between conflict with co-workers and burnout is not significant (γ=-0.20, p>0.05). We therefore conclude that such mediation is total.

The asymmetric confidence interval for the relationship between training satisfaction and self-evaluated performance through work engagement does not contain zero; the unconventional estimate is 0.08, 95% CI [0.014, 0.178]. Thus, we can conclude that work engagement mediates the relationship between training satisfaction and self-evaluated performance. Moreover, when controlling the effect of work engagement, the positive relationship between training satisfaction and self-evaluated performance remains significant (γ=-0.21, p<0.001). Consequently, we conclude that such mediation is partial.

The asymmetric confidence interval for the relationship between training satisfaction and burnout through work engagement does not contain zero; the unconventional estimate is -0.06, 95% CI [-0.121, -0.011]. Thus, work engagement mediates the relationship between training satisfaction and burnout. Moreover, when controlling the effect of work engagement, the negative relationship between training satisfaction and burnout is not significant (γ=-0.003, p>0.05). Hence, we conclude that such mediation is total.

Subsequently, to obtain a more parsimonious solution, we estimated another model (figure 2), where non-significant paths were set to zero. Given that this constrained model, which takes into consideration fewer parameters, is more parsimonious and it does not perform worse than the unconstrained model (Δχ2=4.59, p=0.33), it is preferable. Fit indices show an acceptable fit to the data, considering χ2(93)=273.49, p=0.00; NFI=0.92; SRMR=0.06; RMSEA=0.08. In this model, conflict with co-workers and training satisfaction are associated with work engagement (γ=-0.52, p<0.001; γ=0.14, p<0.05; respectively), which, in turn, is positively associated with both self-evaluated performance (β=0.23, p<0.05) and burnout (β=-0.40, p<0.001). In addition, conflict with co-workers and training satisfaction are directly associated with self-evaluated performance (γ=-0.26, p<0.01; γ=0.21, p<0.001; respectively), thus emphasizing the role of work engagement as a partial mediator. Furthermore, work engagement totally mediates the relationships of conflict with co-workers and training satisfaction with burnout. Ethical conflict is directly and positively associated only with burnout (γ=0.47, p<0.001).

Figure 2

Final model

Discussion

The SR’s role is a multifaceted and complex function. It is not limited to consulting with co-workers, nor to being familiar with all the possible organizational risks. This figure’s range of activities comprises active participation and proposing solutions addressing both possible real risks and the improvement of existing situations. Therefore, the importance of promoting SRs’ well-being – in terms of increasing performance and decreasing burnout – is undeniable.

The results display a rather positive situation, which allows room for improvement. Indeed, SRs experience limited conflict with co-workers, combined with moderate ethical conflict. Moreover, they are quite satisfied with the training received, in terms of both duration and contents. Vigor, dedication, and absorption – namely work engagement – which SRs put into the exercise of this role, appear moderate. In addition, the levels of burnout are limited and performance is moderate.

Our results confirm what the literature affirms (figure 2). We found that work engagement partially mediates the relationship between conflict with co-workers and self-evaluated performance, which confirms the literature data. Conflict with co-workers is negatively associated with work engagement (83) and performance (19, 63), the latter being increased by work engagement (8, 84). Work engagement partially mediates the relationship between training satisfaction and self-evaluated performance, an association already supported by some evidence: training satisfaction increases both work engagement (35, 42) and performance (10, 11). Additionally, work engagement totally explains the relationships between conflict with co-workers and burnout on the one hand (5, 51, 87), and training satisfaction and burnout on the other (9, 49, 76). Although not significantly associated with work engagement nor with self-evaluated performance, ethical conflict is directly and positively associated with burnout, as highlighted by Kammeyer-Muller and colleagues (47), and Wlodarczyk and colleagues (88).

Our findings show the importance of the relationships between SRs and their co-workers, and fill the gap observed in the literature (64). Previous studies analyzed in depth the problems that SRs may face only in their relationships with employers. Little was known about the relationships between co-workers. Intervening in the latter, if perceived as critical, could improve performance both directly and through work engagement. Therefore, it is important to foster positive relationships between SRs and co-workers, because it simplifies the role of representation. At the same time, SRs often complain about the lack of these relationships, especially in terms of perceived support (33).

In order to contrast burnout, both work engagement and restriction of possible ethical conflicts are important. Frequently, SRs perceive the latter in terms of conflict of interest (32, 40). Therefore, the SR should be adequately supported to prevent poor work engagement from leading to occupational distress, in the form of burnout.

Furthermore, the results provide useful information about the importance of increasing certain aspects relating to SRs’ training. In particular, these outcomes highlight the fact that adequate training increases performance, both directly and through work engagement, confirming previous evidence (59, 60). Additionally, appropriate training is also important in order not to aggravate co-workers’ workload, which could reduce the willingness to participate in training activities. Moreover, adequate training, combined with overall recognition of the role, creates the ideal background for SRs to play their role (26).

Regarding practical implications, we should take into account the need for training focusing more on relational skills, such as listening, communication, and information, in line with Article 37 of Legislative Decree 81/2008 in order to facilitate the SR’s centrality and authority, toward employees and employer alike. Therefore, it may be appropriate to provide training and awareness courses, aimed to increase personal and job resources able to promote SRs’ engagement and prevent the potential negative outcomes of high levels of conflicts.

The combined actions of the technical-professional and relational interventions aim to enhance the perception of the SR’s role. They also contribute to limiting the potential conflicts between SRs and represented employees who are not always aware of their representatives’ tasks, functions, and responsibilities.

Knowledge quickly becoming obsolete due to swift changes in the labour market, increasing organizational flexibility, and need for periodic and specific training require an integration between the current educational system and the goal of fostering self and lifelong learning. At the same time, greater resources should be provided to carry out these activities. Such resources could include the promotion of peer debates to explore best practice and reflections about the SR’s role which, if accurately understood and adequately undertaken, can not only create healthy workplaces, but also stimulate innovation, efficient use of resources, and quality employment. An effective training program is much more than mere education on health and safety issues: it involves an empowerment on these topics, based on workers’ participation (32, 38, 67). Therefore, among factors influencing SRs’ performance, the adequacy of training activities is of primary importance (27, 64). Training-intervention initiatives are also vital in fostering transformative learning (13, 53), and allow to move from a mere observer’s perspective to a culture of responsibility and safety. The establishment of Communities of Practice (CoP) is also useful to this end. CoP are spaces for sharing reflections, suggestions, and experiences with colleagues. In addition, CoP satisfy the need to compare the challenges pertaining to the SR’s role and to improve job outcomes (34).

The findings are to be read in light of the multifaceted issues on health and safety in the workplace. These topics require multidisciplinary competencies during both risk assessment and prevention measures management, in a Corporate Social Responsibility perspective (CSR; 7). In this view, the SR should be involved in a tangible and productive collaboration with other health and safety actors, in order to carry out the role effectively. Another key issue is adopting organizational models oriented to CSR not only in large companies, but also in small and medium enterprises (the majority of the sample examined). This could be achieved by using the support offered by advisory and counseling services, devised by positive organizations (14, 15).

Conclusions

The present work allows to highlight the importance of promoting action aimed to solve possible critical issues, such as conflict with co-workers, ethical conflict, and training satisfaction.

A working environment that fosters organizational well-being is a setting that goes beyond the traditional vision and fosters productivity and safety. The new perspective promotes a philosophy by which only safeguarding product quality, raw material, processes, workers’ and environmental safety can guarantee competitive long-lasting edges in an increasingly globalized market. Therefore, it is important to encourage actions aimed to develop a safety-positive culture in the workplace, from which SRs will benefit.

This research has some limitations. The first one reflects the specific limitations of self-report measures and cross-sectional studies, for example the common method variance and the impossibility of determining the direction of the relationships, respectively. However, regarding the latter, even though different explanations are possible, strong reasons support our results. Another limitation derives from the size of the companies examined that has prevented us from studying the figure of the territorial SR. Future research could extend the analysis not only to this role, but also to different contexts, such as healthcare and educational settings.

No potential conflict of interest relevant to this article was reported by the authors