At the end of October, I travelled over to Helsinki for a few days. I’d been invited to deliver a short course on “Risk management measures in a REACH context” to personnel working for the European Chemicals Agency* (ECHA) who are responsible for evaluating the dossiers chemical manufacturers have to submit under the European REACH Regulation. Over 30 people attended the course, a much bigger number than I’d normally prefer. However, they were a really nice group of people who were keen to listen and contribute to the discussion making it an enjoyable experience for me, and, I hope, for the attendees. It was good to have the opportunity to put forward my perspective on how effective risks from hazardous substances are controlled in practice in the “real world of industry.
The principle objective of REACH is to protect human health and environment from chemical hazards, ensuring that risks from the use of chemicals are properly controlled. To achieve this, manufacturers have to undertake risk assessments for all “exposure scenarios” where their products are used and produce “extended safety data sheets” for substances, which must include appropriate “risk management measures”. We’re now starting to see these new style data sheets coming through to users.
During the training session, we looked at how exposures to chemicals vary and the practicalities of obtaining adequate data for the risk assessment process. However, the main discussion centred on the realities of how “risk management measures” are implemented in industry, based on my experiences helping companies to control the risks from using hazardous substances.
I think that there is a widespread impression that controls are much more effective than they are in practice. There are lots of reasons for this, which I’ve discussed in some previous posts. Problems can occur during all the key steps involved in the design and implementation of controls – see my Slideshare presentation and this post for some examples.
It’s a particular problem with local exhaust ventilation systems. In my experience they are rarely well designed and, in practice, their influence on exposure is considerably less than the users (and designers/suppliers) believe. The “lower tier” exposure models commonly used to prepare the REACH risk assessments can assume that LEV is up to 90% effective. The system would have to be well designed and properly used and maintained for this to be the case and I think that it is rare for it to be achieved in practice. Consequently, exposure modelling with lower tier models can considerably overestimate the reduction in exposure achieved by LEV.
Manufacturers and importers of chemicals need to make judgements about the effectiveness of controls when carrying out their risk assessments and also when deciding on what risk management measures are needed. The danger of overestimating how good they are could compromise their risk assessments and result in risk management measures being specified that won’t adequately control exposure. It’s important, then, to have a realistic appreciation of the “real world” effectiveness of common controls.
Those extended data sheets I’ve seen so far seem to specify realistic controls for the exposure scenarios. However, they are phrased in very general terms. Again this is likely to be a particular problem with LEV. For many industrial organisers “LEV” means a captor hood – often the flexible “swinging arm” type. As I’ve discussed in a previous post, these are largely ineffective at controlling contaminants. But in many cases if a company follows a general recommendation to install LEV, this is what they’ll buy. I think that if REACH is really to achieve it’s objective of improving control, then we need to ensure that the advice on risk management measures is as specific as practicable. So with LEV enough details needs to be provided to make sure that the design of the extraction hood is appropriate.
Another problem I’ve noticed with the new style safety data sheets I’ve seen is that where personal protective equipment is recommended the advise is too general. For example, recommending “wear suitable gloves”. This really isn’t any improvement on the older style sheets. Downstream users need more specific advice on what type of gloves are needed, particularly what they should be made of. The reality is that most users don’t have the expertise to select “suitable gloves” and in most cases the gloves used are made of an inappropriate material and are not used and managed properly. This point is also relevant to other types of personal protective equipment. Again, I’d like to see more specific details provided.
For me a good model for user friendly control advice is the COSHH Essentials control sheets. These provide good, concise advice on control for common processes on a maximum of 2 sides of A4. Where LEV is recommended specific details on the hood design, including an outline diagram, is provided. These sheets aren’t perfect – their advice on personal protection is too vague, for example – but I think that overall they strike the right balance between brevity and the usefulness of the information.
A lot of work has to go into carrying out the risk assessments. It’s important that the output – i.e. the information on risk management measures – should be detailed enough to ensure that controls are properly designed and implemented. Unless this happens there’s a real danger that all the effort will be in vain and an opportunity to substantially improve control of hazardous substances at work will have been missed.
*ECHA’s role is to manage and coordinate the registration, evaluation, authorisation and restriction processes and to ensure consistency in the management of chemicals across the European Union.