Workplace dermal exposure assessment is a complex task that aims to understand the dynamic interaction between the skin and the hazardous substances present in the surrounding environment. A European project known as RISKOFDERM gathered dermal exposure data in 85 workplaces (industrial and other types) in five countries in Europe. In order to optimize data collection and to develop a representative picture of dermal exposure, scenarios (tasks made up of a series of activities) were grouped together into dermal exposure operation units (DEOs). The allocation of scenarios to relevant DEOs was achieved on the basis of similarities of exposure routes, tasks and professional judgement. Sampling and quantification procedures were based on the approaches recommended by the OECD protocol. The laboratories involved in the analysis of the samples participated in quality assurance programmes. This exercise resulted in 419 body measurements and 437 measurements on hands expressed in terms of formulation (product) in use. Exposures for a given scenario varied by several orders of magnitude. The extent and patterns of exposure were found to be dependent on various exposure determinants, including inter- and intra-scenario variations. Hands were found to be the most contaminated parts of the body. Exposure patterns for liquid and solid contaminants were different. On the basis of the analysis of the data presented here, the averaged results (median and 95th percentile) for a given DEO unit should not be used as a representative measure of dermal exposure for all scenarios within that DEO without taking the exposure determinants into account. However, the data could be used to develop an exposure matrix (indicative exposure distributions) for different types of scenario and workplace, using determinants of exposure and a Bayesian approach to integrating expert opinion.

The Health and Safety Executive (HSE) in Great Britain (GB), in association with its stakeholders, developed the Control of Substances Hazardous to Health (COSHH)-Essentials (C-E) control banding tool in 1998. The objective was to provide a simple tool for employers, particularly small and medium-sized enterprises (SMEs), to help select and apply appropriate measures for the adequate control of exposure to hazardous substances. The tool used hazard classification information (R-phrases) to assign substances to one of five health hazard groups, each with its respective 'target airborne concentration range'. The validity of the allocation of substances to a target airborne concentration range was demonstrated at the time using 111 substances that had a current health-based Occupational Exposure Limit (OEL) in GB. The C-E control banding approach remains an important tool to complement exposure assessment/monitoring and the selection and use of suitable control measures for hazardous substances. These include engineering controls and personal protective equipment (PPE). The C-E based control banding approach has been adopted around the world. This paper extends the original validation exercise, using a greater number of chemical substances, to establish whether the target airborne concentration ranges remain appropriate. This is of particular interest in light of the introduction of the Globally Harmonized System (GHS) for classification, in which R-phrases have now been replaced by hazard-statements (H-statements). The validation exercise includes substances with OELs published by nine bodies internationally; and the Derived No-Effect Levels (DNELs) assigned by registrants under the European Union-Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH) Regulations. When compared against 8-hour TWA OELs for 850 substances drawn from nine bodies and a limited number of DNELS, the C-E target airborne concentration ranges remain valid. This comparative work also informs a wider consideration around the practicality and the applicability of the C-E generic approach to facilitate the implementation of good practice control for a wide range of substances (more than 95%) which do not have any recognized OEL.