Expert perspective on health and safety determinants in wineries

1 Introduction

The scientific literature contextualizes health and safety (H&S) in two major models, the first is based on a comprehensive perspective of H&S and all in-company processes, and the second approaches H&S management from a more operational perspective that transforms tacit knowledge into explicit knowledge.

In this study, a panel of risk management experts with extensive experience in the industrial sector worked together to identify the main factors involved in H&S management in wineries.

Wineries account for 14% of the food and beverage industry in Spain, according to a 2017 census. They provide direct employment to 24,051 people, and involve processes that can cause accidents or occupational diseases [1, 2].

Wineries are part of the industrial sector, which, with 5,264 lost-time accidents per 100,000 workers in 2017, ranks second in the number of occupational accidents in Spain [3]. This shows the need to perform a risk assessment to determine the aspects that must be improved to control or eliminate H&S risks.

Wineries, like all companies, have a duty to integrate H&S measures in their management system through a Health and Safety Plan. Under EU Directive 89/391 [4], one of the most important strategies in this plan is H&S planning Therefore, companies aiming to eliminate or mitigate and controlling safety risks, thereby reducing the cost of occupational incidents, accidents and illnesses, will need to implement such a plan [5].

The different H&S models found in the scientific literature are discussed below.

Proper management involves methodically arranging interdependent activities and related procedures under a department that coordinates and controls corporate activities [6] and H&S planning.

Some authors [7] believe risk assessment to be an integral part of H&S management, insofar as it allows managers to take informed decisions and priorities interventions. In the same vein, [8] details the type of scientific and societal information required for the best risk control decisions.

In addition to specific measures to mitigate occupational risks, H&S planning also includes emergency measures and health surveillance, employee information and training, together with any regulatory obligations that affect working conditions. To ensure correct execution, employers need to define and implement systems to monitor and control the implementation and effectiveness of their H&S measures [9].

An H&S management system designed to prevent and control occupational risks must allow organizations to integrate structured, coordinated and integrated safety measures into all corporate activities and decisions. This has compelled organizations to demand certifiable management models, and international standards such as OHSAS 18001 [10] or ISO 45001:2018 [11], which offers the possibility of certification, are now widely used [12, 13].

It should be added that integrating quality, safety, and management tools such as ISO 45001:2018 into organizational learning can give companies a strategic advantage, as some researchers have shown in their studies [14–17].

Other authors [18] warn that overloading small and medium sized enterprises (SMEs) with information and standards carries the risk of over-bureaucratizing H&S, and call for the creation of accident scenario tools that distinguish between relevant and irrelevant events, thus only generating information for operational solutions [19].

In fact, there is evidence that effective accident prevention in SMEs is achieved more through operational decision-making than through the formalization of regulatory obligations [20].

Similarly, [21] in a conceptual review on machine systems risk assessment, places particular emphasis on the need for ongoing agreements between employers and workers, with expert support, to reach a consensus on the classification of each risk, together with the instigation of a culture of learning to achieve sustainable safety that converts H&S management into a continuous improvement cycle (Fig. 1).

OPEN IN VIEWER

Fig. 1

Health and safety model. Continuous improvement cycle. Source: www.insht.es (National Institute for Safety and Higiene at Work-INSHT).

Iterative and continuous improvement approaches, such as control banding, have the potential to facilitate knowledge management in H&S strategies [22].

Knowledge management involves creating, transferring and using knowledge in decision-making, and using the accumulation of tacit wisdom converted into explicit knowledge to drive change [23–25].

In line with the above, various studies, for example [26], have drawn attention to shortcomings in risk management in wineries, and propose preventive and improvement measures.

Another important study is focused on active H&S management, as mentioned in the risk assessment carried out by [27].

Also important, is the contribution made by several studies to continual improvements in H&S systems in wineries, which have led to the publication of H&S and surveillance plans [9, 28–33].

The contribution made by studies that investigate the most common workplace accidents in order to use these experiences to improve corporate H&S is particularly important [34–39]. The publication linking the risk of falls and dim lighting [40] is of interest, since winemaking processes require specific humidity, temperature and lighting conditions [41].

Winery-related papers published by other authors have allowed experts to build on the knowledge acquired and design and implement corrective measures to prevent H&S failings in the workplace and substantially improve the health of workers [42–45]. Also important in this regard are studies in olive oil mills by Parejo-Moscoso et al. [46], due to their similarity with the wine industry.

A review of the literature shows that several authors attach great importance to the control and surveillance of individual and collective protective equipment [26, 48].

According to the analysis of certain studies in the different phases of the winemaking process, substantial and varying risks arise, such as being buried beneath the mass of harvested grapes, being run over by vehicles in transit areas, falling into vats, entrapment and severed limbs. Crushing, cuts or electric shocks may occur during the grape pressing phase. In the vatting of the grape mass in tanks (commonly referred to as fermentation or vinification) and in the maceration and alcoholic fermentation phase, dangerous atmospheres may be generated with concentrations of Carbon Monoxide (CO), Carbon Dioxide (CO₂), Sulfur Dioxide (SO₂), Hydrogen, Methane, Acetylene, Air, etc. In the storage phase there are numerous activities such as cleaning, disinfection, maintenance and upkeep of tanks) [41]. These risks are also covered by the guide produced by the Autonomous Region of Madrid [49] and the main Spanish trade unions, and also the Occupational Risk Prevention Institute of Andalusia [9].

However, as far as the authors are aware there is little research as to occupational risk management processes at wineries, and the little that has been found emphasizes the analysis conducted at 32.6% of wineries in the Region of Murcia, Spain, the conclusion being a lack of integration of risk prevention at the heart of the companies analyzed, with 80% of those analyzed having risk assessments that are less complete than they should be. With regard to documentation, they revealed serious defects such as the absence of written working procedures; 93.3% of companies did not have adequate resources to measure dangerous atmospheres, and at 73.3%, the tanks were inadequately ventilated [50].

In light of the foregoing, the authors note a shortcoming in the research which suggests that a risk prevention culture has not been taken on board at wineries, and therefore, we decide to leave philosophical debates on H&S to one side and asked a panel of experts to identify the planning-related factors they consider most important to effectively improve H&S practices and reduce workplace risks in wineries.

Recent research reveals that even university education requires an increase in both the quality and the quantity of teaching about Health and Safety at Work [51].

Meanwhile, other research also shows that official accident reports need to be improved so as to enhance overall learning and optimize resources [52].

In the third trimester of 2017, we asked a panel of experts to identify the key elements involved in the management, organizational structure, and H&S strategies in wineries using the Delphi method. The Delphi technique is defined as “a method for the structuring of a group communication process which is effective in allowing a group of individuals to address a complex problem as a whole”. The experts’ considerations are built up over successive, anonymous rounds, in an attempt to achieve consensus, while maintaining the utmost autonomy on the part of the participants [53].

The panel was made up of 11 experts in wineries and risk management training in Andalusia (Spain). Seven panel members had between 10 and 20 years’ experience in the sector; 2 were agronomists with 7 to 10 years’ experience in risk prevention in wineries, and 2 had been working for 3 years as H&S consultants in wineries.

This selection of experts was based on their thorough knowledge of such a specific and significant sector as wineries [1, 2] belonging to an industrial sector with a high occupational accident rate [3].

On this basis, we considered the panel to have the expertise required to take part in the project, and to be large enough [54] to obtain consistent results. The experts needed 2 rounds to reach consensus [55] with a 100% response rate.

Following Gubiani et al. [37], who used checklists to identify risks in wineries, we developed a questionnaire (Fig. 2) based on prevailing Spanish legislation on H&S, Order TAS/2926, 21 of November 2002 [56] and Directive 89/391 [4], which establishes new H&S measures and provides mechanisms for electronic filing. This same methodology has been validated and published in other studies into occupational risks at wineries [2], and therefore struck us as appropriate for this research.

OPEN IN VIEWER

Fig. 2

Questionnaire. Drawn up by a panel of experts using EuropeAid Evaluation Guideline methodology based on the provisions of the Manual of Good Practices in Occupational Risk Prevention in the Wine Sector (European Commission, 2017).

In the questionnaire, experts were asked to assess the importance and suitability of factors included in 4 dimensions: documentation, organizational structure, activities, and H&S organization models. They were asked to rate their responses using a 5-point Likert scale (“very low” to “very high”) [57] and the answers they provided are set out in Fig. 2.

The dimensions and variables were both drawn up by a panel of qualified experts whose expertise was accepted by the remaining panelists and was not involved in the evaluation. The questionnaire was drawn up following EuropeAid Evaluation Guideline methodology [58] based on the Manual of Good Practices in Occupational Risk Prevention in the Wine Sector [9].

The questionnaire was circulated twice to implement the method: An initial round was conducted, and following the collation and statistical analysis of the results, the interquartile range was used as the statistical response of the group, meaning that all the opinions expressed by the experts were reflected and conveyed to the rest of them [46]. The gap between the two rounds was 12 days, and the response rate was 100%. (Table 1).

At the end of the first round, the results were analyzed on the basis of median scores, and transmitted to the group. Two rounds were needed, using the interquartile range to indicate consensus: responses that scored either side of the 25th and 75th percentile were considered out of consensus [59]. A second round was carried out to improve consensus, following the methodological recommendations of Landeta Rodríguez [2, 55]

Delphi analysis was performed using SPSS Windows 15.0 (SPSS Inc., Chicago, IL, USA).

The authors decided to use as our research method a qualitative focus which in principle could reveal a limitation of internal validity, although as in the case of any research method it reveals benefits and drawbacks, and in this case the benefit is that as other researchers make comments [60] this achieves a more in-depth understanding of complex phenomena lying outside the potential scope of quantification, or knowledge through other studies [61].

Table 1 shows the results of the 2 Delphi rounds, with first quartile (P25), second quartile (Me = median) and third quartile (P75) scores, and the relative median VR (Me).

Variation is seen to decrease significantly in the second round, with the interquartile range (RQ2-RQ1) variation being negative in most cases (Table 1).

The mean scores were used in statistical tests to facilitate identification of the most and least important factors (see Table 2).

The importance given to different H&S factors tends to be high overall. Mean scores range from 3.27 (medium) for the existence of “health and safety deputies” within the organizational structure, and 4.45 (high) for the existence of “training activities” and “control of individual and collective protective equipment” (See Table 2).

In the documentation category (see Table 2), the highest score in both rounds was given to “risk assessment”, with a mean of 4.27, followed by “emergency plan”, which scored an average of 4.09 in the first round and 4 in second.

Regarding the category of H&S organizational structure (see Table 2), the experts gave the greatest importance to “Employer” and “Departmental heads”, with an average score of 4.18 in first round and 4 in second, while “health and safety officers”, with a mean score of 3.27 in the second round, were considered more expendable.

Regarding the importance of activities related to risk prevention management (Table 2), in the second round the experts rated 6 of the 8 variables important or very important, that is, with a score of over 4.

The results show that experts gave “training” an average score of 4.36 in the first round and 4.45 in the second round. It is interesting to note that the score for “control of individual and collective protective equipment” increased from 4.27 in the first round to 4.45 in the second, while “emergency drills” obtained the lowest average score, with 3.45 in the first round and 3.55 in the second (Table 2).

In the second round, both “accident investigation” and “coordinated health and safety measures” obtained a score of 4.27 (see Table 2).

“Self-inspection” with a mean score of 3.73 in the second round, was also given relatively little importance.

In terms of H&S models (Table 2), “external health and safety service” obtained an average score of 3.55 in the second round, while “Employer” obtained an average score of 2.36 in both rounds. The experts justified this score by indicating that the type of H&S model implemented depends on the size of the company and the volume of business. For example, they indicated in the interview that an in-house health and safety department is the best choice for a company with a large average workforce, while smaller companies should rely on external H&S services.

The panel agreed on the most important factors involved in risk management, such as documentation, organizational structure, activities and others, and health and safety model, and within these categories there was overall agreement on the importance of the employer and departmental heads performing risk assessment.

The results of our research likewise indicate that the experts place particular importance on risk assessment, correlating with other research which likewise focuses on this aspect on identifying and evaluating risk factor in wineries [9, 28–33], to which can be added Parejo-Moscoso et al. [46], in olive oil mills, due to its similarity to this sector, and as indicated by a review of the literature, the process of winemaking is complex, with many phases in which highly significant risks requiring assessment may be identified [40, 41].

With regard to organizational structure, the experts place greater importance on managerial positions at the company than the workers’ health and safety representatives, which deviates from the conclusions of other studies such as Etherton’s conceptual review on machine systems risk assessment [21], in which ongoing agreements between management and workers on the classification of risks were particularly important.

Our study likewise highlights the importance of workers being familiar with and knowing how to use personal protective equipment, and these results are consistent with studies that consider control of individual and collective protective equipment to be the most important factor, insofar as it can protect against risks derived from the fermentation process, such as asphyxiation, poisoning, exposure to carbon dioxide [30, 47,48], and to the different chemicals used in wine making [28–33], an aspect of vital importance since, as indicated by the contributions of various studies, dangerous atmospheres are a risk associated with this type of facility, which may result in fatal accidents [41, 49]

These results are in line with the iterative and continuing improvement approaches taken by other authors [22–25]. However, they contrast with the results of the surveys carried out by Castellanos [61], which indicate that workers do not always receive H&S training an essential part of workplace practices, and one that is even more important in sectors with a high accident rate such as farming, mining and construction, although practical research reveals that the culture of accident prevention and occupational safety is not incorporated within daily management at wineries [50], and furthermore, as other studies conclude, this aspect needs to be underpinned in training and academic education, and in official wine industry reports. Our results are also in line with authors [7] who consider risk assessment to be an integral part of the risk management process, others who also found both scientific and comprehensive information to be important in decision-making [8], and other studies that highlight the risk of over-bureaucratizing H&S [18, 19].

The high accident investigation scores are incorporated within the body of research in line with the results of other studies which have suggested the need to prioritize corrective measures for avoiding failures that call for corrective measures to avoid interrupting the cycle of continuing improvement [42–45], and recommend building on experience to improve corporate H&S strategies [34–39].

Nevertheless, although our experts gave “emergency plans”, they did not rate “emergency drills” as highly. This contrasts with Gubiani et al. [37], who considered that emergency plans and drills in wineries are often poorly designed, and are therefore one of the most critical factors to be addressed. Employers sometimes regard initiatives such as “drills” or “self-inspection” as a waste of time, although several authors consider organizational learning to be an important management tool [23–25].

5 Conclusions

Based on the studies analyzed, we can distinguish 2 general approaches to H&S management: on the one hand, a holistic perspective with H&S processes in all company departments, and on the other, a more operational perspective that involves decision-making based on tacit wisdom or the creation of risk management drills that distinguish between relevant and irrelevant events.

This study enriches the existing literature by contributing expert opinions on H&S practices to specifically improve working conditions and employee safety in wineries.

The experts gave high scores to the “activities” dimension, and others such as “control of personal and collective protective equipment” and “training”, together with “accident investigation”,“coordinated health and safety measures” and “signage” are associated with capitalizing on knowledge gained in situ and promoting organizational learning to convert tacit wisdom into explicit knowledge, prompting specific risk prevention measures such as:

–Control of dangerous atmospheres: with evaluation of the interior atmosphere by means of multi-gas detector measurements; personal protective equipment of the filtering and insulating type; ventilation, or as far as possible relocation of winemaking tanks to the exterior of the wineries; surveillance from the exterior with equipment using at least two people to perform the task; communication between interior and exterior spaces with appropriate resources and perfect upkeep.

–Control of tank cleaning, maintenance, extraction of pomace and collection of samples.

Based on the interviews carried out, we can conclude that the number of workplace accidents in the sector can only be reduced by raising awareness of H&S and providing workers with adequate training. This will plant the seed of change to an H&S culture throughout the wine industry in Spain. For example, there is no use in storing personal protective equipment if it is not used. All employees, from management to the shop floor, need to assume their individual responsibility in respect of health and safety.

6 Limitations

This research reveals one limitation in that it includes only experts from the region of Andalusia, although in our opinion this is offset by their thorough knowledge of the sector in terms of their years of experience in both winery management and occupational risk prevention in the sector, and the study may therefore be extrapolated to any location in Spain or in other countries.

Another limitation is the problem concept which could have been addressed from a multidimensional perspective by addressing specific measurements. This would be optimal, but would ideally require a larger sample of experts, or possibly a focus more resembling a quantitative technique than a Delphi study, which is essentially a qualitative technique, suggesting possible lines of research in the future.

In our study the experts state that occupational risk management is important, giving particular importance to training and the control of personal protective equipment, although authors such as Guillemin & Horisberger [26] analyzed a fatal accident using the fault tree method, and showed deficiencies in H&S management in these types of companies.

7 Lines of future research

It would be interesting to complete the data collected in this study with data on accidents in the wine industry and psychosocial risks to workers.

It would be interesting to complement this qualitative study with a quantitative study in the future.