How can it be possible that people who are theoretically protected by PPE are in fact getting contaminated? Discussions led with coverall manufacturers revealed that the coveralls currently recommended for use in farming were primarily developed for industry, with agriculture representing only a niche market, and indeed the effectiveness of the coveralls has not been tested with the active substances contained in pesticides, including some of the most widely used.
This article highlights technical and organisational flaws9 related to the assessment of the coveralls, and thus raises questions about compliance with the requirements set out in the European standard governing PPE (design, certification and commercial sale) - an issue raised previously by Mayer & Bahami (2006) but without response. Following on from Dubuc (2007), if we take an in-depth look at the manufacturer’s instructions provided with each set of coveralls, it becomes clear that the information on coverall performance and limitations is highly technical and sufficiently complex to deter workers from reading it.
Abstract
Even if global prevention approaches have been developed by prevention and health and safety institutions to reduce the risks posed by chemicals in general, and phytosanitary substances/plant protection products in particular, one can’t help noticing that the prevention measures which are most often implemented consist of using PPE.
Despite the fact that regulatory bodies and common health and safety sense clearly favour other means of prevention - such as collective prevention or administrative controls - what actually happens is that they are scarcely used.
This is due to several reasons: the costs of non PPE control measures often exceed the investment capacity of farmers, and some of the technologies are ill-suited to the agricultural industry. The effectiveness of PPE in real-life conditions is thus all the more important, as they often are the only - and not the last - resort to protect people and reduce risk. PPE effectiveness in agriculture is an occupational and public health stake and also poses the question of legal responsibility.
Prevention as a technology transfer
This article aims to present the results of an ergo-toxicologic study of the effectiveness, in real-life conditions, of the protective suits recommended by preventionists. Based on the Pestexpo study led by Isabelle Baldi, which focused on analysing the exposure of vineyard workers to plant protection products by using an ergo-toxicologic approach, this method was developed by several authors including Sznelwar, Mohammed-Brahim, Wisner and Garrigou.
The point is to identify different types of exposure by taking into account the characteristics of the real work activity (as opposed to what the operators are supposed to be doing) as a produce of personal, technical and organisational determinants.
The Pestexpo study showed that even when they were wearing PPE, people using plant protection products were contaminated, in some cases more seriously than without any protective equipment. In this paper we are going to develop the different hypotheses that may explain these results. By autumn 2006, a hypothesis was formed questioning the real effectiveness of the protective suits recommended for the use of plant protection products.
To analyse and transform the situation, a bottom-up approach was proposed, as developed by the francophone practice of ergonomics and the anthropotechnology approach developed by Wisner, and a top-down approach as Macro-ergonomics, developed by Hendricks.
An ergotoxicologic approach of phytosanitary risk
Approaching the issue of the real performance of recommended PPE, even when used properly, is a vexed question for the occupational health and prevention researcher group.
The point is to identify different types of exposure by taking into account the characteristics of the real work activity (as opposed to what the operators are supposed to be doing) as a produce of personal, technical and organisational determinants.
The Pestexpo study assessed, among other elements, the real external contamination of vineyard workers by phytosanitary substances (dithiocarbamates, in 2001 and 2002). For the spraying activities in Gironde, 96 field observation days made it possible to produce different types of data for each type of activity (preparing the product, spraying and cleaning the equipment).
The contamination was measured through a chemical analysis of 10 square cm gauze patches, which were placed directly on the operators’ skin, on different body parts. The patches were removed and replaced with sterile ones at the end of each phase of activity. See figure 1. The protocol was consistent with OECD recommendations for that type of study. For the PPE-wearing operators, the patches were placed under the PPE.
It is worth noting that through the observations operators carried out the different activities as they normally do. Some were using PPE while others weren’t. For those wearing PPE, some had appropriate equipment, others didn’t. Through the preparation stages about 2/3 of the operators used PPE, about half of them used PPE for the spraying stage and more than half for the cleaning stage. Yet, wearing PPE doesn’t mean that the PPE was suited to the task and substance, or cleaned, or not already contaminated.
The results were expressed in mg of active substance deposited on the operators’ skin thanks to an extrapolation from the patch to the considered skin surface. Figure 2 represents the median contamination (horizontal bar) and the distribution (from bottom to top: minimum, 25th percentile, median, 75th percentile, maximum). The most striking finding is that the distribution of the contamination values for operators either PPE equipped or not largely overlap. In some cases, operators wearing PPE are more exposed than operators who weren’t. The processing of the collected data leads to several observations:
- Wearing PPE does not completely avoid exposure
- In the preparation stage, wearing
- PPE limits exposure to some extent
- During the spraying and cleaning stages, the operators who used PPE are, on the whole, more exposed than those who don’t use any PPE
These results were very confusing for the different prevention institutions, as one of the main points of the recommendations they make is the use of PPE, in particular Type-4 suits, which are theoretically spray proof.
Some hypotheses to explain exposure
After numerous talks with different experts in this field (CCMSA, MSA 33, MSA 24, INRS, IFTH (French Textile Institute), UIPP (Plant Protection Products Manufacturer Association), ECPA (European Crop Protection Association), ACTA), we proposed different hypotheses to explain the observed contamination:
• Personal and collective caution know-how developed through experience. In the course of the observations, it soon became obvious that the practices and moves could vary greatly from one operator to the next. For instance, regarding the care given to opening the product container and pouring the powder into the tank. Trying to limit the dust given off by carefully opening and gently pouring the contents deep inside the tank may significantly reduce risks of inhaling the substance. Such know-how may not only limit direct contamination, but may also reduce cross-contamination. Facilitating the development and transmission of this safety know-how could then become one of the major stakes of prevention
• A pre-contamination of the PPE which can be explained by the fact that it is re-used. When the PPE is stored in contaminated premises with poor housekeeping practices, a cumulative contamination inside the suits is
most likely
• The suits just don’t work for some of the stages of the activity, for instance during cleaning, when high pressure jets and running down grey water could well cause some of the substance to migrate inside the suit
• A lack of guidance and advice on how to use, maintain and clean PPE. Generally, workers do not have the user manuals and haven’t been trained in how to take off and clean the PPE safely
• The organising and planning stages can also determine contamination. Indeed, while some workers try to plan ahead so as to avoid interruptions, or running low on products while spraying, others do not really give much thought to planning and making sure that the activities run smoothly
• The designers of spraying equipment may not have sufficiently taken into consideration the needs of the vineyard workers. For example, to fill the tank of the trailed sprayer, operators can be led to climb up on the trailer wheel and lean on the outer wall of the tank for balance while they empty the substance into the tank. Most often, this outer wall is covered with a deposit of phytosanitary substance, resulting from the previous spraying or overflowing when filling up the tank. Leaning on the tank may account for some of the contamination.
From autumn 2006 a new hypothesis questioned the real effectiveness of the suits recommended for using phytosanitary substances. Through cooperation with a suit manufacturer we were able to investigate into the permeation of the suit fabric. The manufacturing company wanted to check the efficiency of one of their suits in the protection from a herbicide and it was tested by an independent certified laboratory. The suit is a type-4 suit, widely spread on the market, and the test followed the 2004 EN 374-3 protocol. The results are alarming as they highlight a very fast permeation phenomenon for a wide range of herbicides frequently used in the agricultural sector. This means that this suit, recommended in agriculture through extrapolation from industry doesn’t provide an adequate level of protection against some of the risks present in agricultural activities.
This type of test wasn’t carried out with the substance observed in the Pestexpo study. However, we may hypothesise that there is some degree of permeation, which could account for the fact that PPE wearing operators are exposed. After discussions with PPE manufacturers it appears that the suits recommended for farm use were initially developed for the industrial sector. What’s more, as sales to farmers are only a niche, those suits may not have been tested for phytosanitary substances, including the most frequently used ones.
As of today, there is no evidence that the permeation problem does not exist in the industrial sector (for instance in the pharmaceutical/sanitary industry) and in particular in the chemical industry for other substances.
Discussion
We must consider the design and use of PPE as a transfer of technology issue. PPE was designed to be used in situations that corresponded only a little to the working conditions of the farm or vineyard workers.
We can also observe that the worker probably used the PPE for the first time because he knew that he would be observed.
We can highlight the fact he used sponges in order to protect his shoulders from the weight of the portable sprayer (about 25 kg with the preparation of agrotoxics). This physical demand means operators may compromise between the level of protection they get and reducing the demand/increasing comfort by not using PPE. In some cases, workers said they felt too hot wearing the PPE in the sun. It should also be noted that the risks and the effects of exposure may not be felt directly. Indeed, a worker explained that he couldn’t feel the substance on his skin but he could feel it in his lungs. We can then suppose that workers may favour respiratory protection over suits, even if skin contamination may actually be more important.
It is also important to note that the question of the PPE cannot solely be tackled from a technical perspective. This question involves subjective and social dimensions, as expressed in the vineyard workers’ own words: “People think I’m an astronaut”; “People think we’re contaminating the vineyards/the wine with harmful substances.” Some workers said they had stopped using PPE as their neighbours thought they were polluting the land.
All this revealed issues which we can call ‘technical and organisational failures’ in the assessment of the real effectiveness of suits, and thus in the compliance of the requirements of the European PPE standard (design, certification and marketing).
Whistleblowing
Following this research, scientists drew the conclusion that while users of PPE are expected to be familiar with notions of penetration and permeation, in reality this was rarely the case.
Furthermore, although this is the class of coverall recommended by the prevention management institutions (Ministry of Agriculture and CCMSA, 2007), it is clear that the tests on resistance to permeation by liquids for these coveralls are not conducted using active substances featured in pesticides, but with various sulphuric acid and sodium hydroxide-based solutions.
There is currently no evidence to rule out permeation as an issue in factory settings such as the pharmaceutical industry, or more specifically the chemicals industry, with other products. Furthermore, it is highly likely that the sweat generated via physical activity and trapped within the coveralls actually promotes the penetration of plant protection agents into the coveralls.
Focusing on the cleaning phase, the pressure of the water sprays together with run-off may well promote the migration of products building up on the outer shell of the coveralls.
As far as prevention solutions go, the initial response would be to recommend using Type-3 or Type-2 coveralls (see figure 4), which theoretically offer better protection. However, there is no current evidence to suggest that this would solve the permeation issue. One last point is that boosting the permeation protection level would reduce thermal comfort for workers, as the coveralls could become ‘un-wearable’. This might introduce new risks related to working in hot environments by making it impossible to “wick away” sweat, thus preventing the body’s homeostatic mechanisms for controlling core body temperature during work. This is likely to be one of the reasons why PPE is not being worn, because it soon becomes impossible to work in these conditions.
Therefore, it is not reasonable to expect pesticide users to be responsible for making sure that the protective gear they are supplied with is effectively compatible with the pesticides that they use, based on the information given in the manufacturer’s data.
The governing guidelines (Ministry of agriculture, 2007) themselves are ambiguous, both in authorship and in the policy statements of certain experts, which are liable to be detrimental to users’ health and safety. The notified bodies, for their part, seem to believe that they are playing their role if they are applying highly technical standards that are sophisticated to implement.
According to the results presented, we have decided to act as a whistleblower (Vaughan, 1996).
For strategic reasons, a process was defined for drafting and releasing a warning notice in order to mobilise as many parties as possible. This process is an integral part of a fully managed approach fuelled by the results generated through the ergotoxicology approach. It is important not to forget that the permeation results were produced through a study led by a pesticide manufacturer whom we count as a partner. It was by cross-comparing this data with Pestexpo data (see figure 5) that we were able to draw attention to the scale of the issue raised. This data nevertheless remains the property of the industry partner and as such, could not be publicly released.
The first step involved discussions with this industry partner with a view to using these results, while keeping both the company’s name and the name of the protective gear manufacturer anonymous.
The second step was to test our analysis with the various stakeholders involved. Once we had enriched the results, the decision was made to draft a warning notice in scientific article format. After informal talks with all the stakeholders concerned, and notably those tasked with PPE project sponsorship roles at the various safety and health institutions, this warning notice was sent by registered mail to the managers of the institutions concerned (six in total), and then forwarded on to various agriculture and agribusiness trade associations and labour unions (five in total).
The final stage in this whistle blowing process consisted of working together to draft a report on farmer exposure to pesticides for publication in a specialist review entitled Santé&Travail, which enabled us to produce an in-depth, fully contextual study while at the same time reaching a wider readership in touch with prevention issues - particularly farmers and/or agricultural labour associations. This report was then relayed via mass media channels such as radio and the press, which in turn has heightened general media pressure, particularly on the institutional stakeholders. The final phase was to release the warning notice.
This relatively painstaking process showed that it was very important to create conditions whereby each stakeholder is able to use their skills and expertise and interact with others beyond the traditional boundaries.
When the Ministry of Labour (tasked with monitoring the implementation of directive 89/686/EEC) received this alert on the ineffectiveness of the protective coveralls worn by farmers when working with pesticides, they looked into several courses of action, all based on the principle that the findings revealed by the whistleblowers branched back to different kinds of problems. It appeared necessary to review user practices (choice of coveralls, how they are stored, looked after and disposed of) and re-examine how the coveralls are actually engineered in relation to the content of the governing standards.
Results
Following the warning, the Ministry of Labour asked the AFSSET (Agence Française de Sécurité Sanitaire Environnement et Travail, The French Agency of environment and occupational safety and health) for support. The AFSSET then conducted research to confirm or refute the conclusions of our work.
First, they launched permeation tests for 10, class 3 and class 4 coveralls commonly sold in France (AFSSET, 2010 Le Frious & Paillat). This was to make sure that the coverall permeation performance matched the data contained in the user manual. Only two coveralls out of ten provide the level of permeation resistance displayed in the user manual. The remaining 8 coveralls vary more or less importantly with the manufacturer data, from almost immediate permeation (class 0) to a permeation resistance one class below the manufacturer’s data. These results have prompted the authorities to re-class some coveralls and even to forbid the sale of two sorts of coveralls. Manufacturers had to recall non-compliant coveralls that were already on the market.
Secondly, further tests were carried out on the best performing type 3 and type 4 suits as identified by the first series of tests. This was about testing these suits with pesticides and chemical mixtures. The tests showed that pesticides could rapidly permeate through type 4 suits. Type 3 suits had better results as they are made from materials different to those used for type 4 suits. This means that these suits are actually more resistant to permeation, although their thermal comfort is poorer. It shall be noted that the study led by AFSSET goes beyond PPE use in farming, and questions its effectiveness as a means of protection against chemicals of any kind, whatever the activity and sector.
In addition, those results question the method used, in the standards (EN 374-3-2004 for example) to assess permeation resistance in the case of pesticides.
