In 2012, The Synergist published a series of articles about developments in chemical risk assessment. The series–aptly named “Risk Assessment’s New Era”–highlighted research and policy initiatives shaping the future of risk assessment in industrial hygiene. The articles detailed new tools and approaches for characterizing occupational health risk related to chemical exposures. Now, ten years later, another phase of the evolving practice of risk assessment has emerged. A key driver is the implementation of the June 2016 Frank R. Lautenberg Chemical Safety for the 21st Century Act, which amended the Toxic Substances Control Act of 1976. The amended TSCA defines new responsibilities for EPA in evaluating and managing risks posed by existing chemicals in commerce in the United States, including those in the occupational environment. However, there are differences between EPA’s methodologies and those employed in the traditional practice of industrial hygiene. These differences may create confusion regarding the optimal risk management strategies for specific chemicals, but one must remember that all occupational risk assessors share a common goal of ensuring worker health. Describing the context and nuance of different approaches is critical to improving the practice of occupational risk assessment and in communicating to decision-makers and stakeholders.

Since passage of the Lautenberg Act, EPA has evaluated the risk of 10 “high priority” chemicals, concluding that unreasonable risks exist in the manufacture and utilization of chemicals for some “conditions of use.” The agency arrived at these conclusions by applying the assumptions and methods in its published risk evaluations. In this article, we explain the key elements of EPA’s current methodology and highlight how it differs from traditional OEHS risk assessments. Our goal is to help practitioners understand where potential disconnects exist and to prepare OEHS professionals for the kinds of requests for information that EPA may issue to support its risk assessments. Ultimately, analyzing differences in methodology provides an opportunity to increase awareness among all stakeholders of various data, methodology, and assessment toolkits that can enhance occupational risk assessment practice. Ultimately, the ideal outcome of such efforts would be increased alignment, where appropriate, of current practices in occupational risk assessment methods with EPA’s characterization of chemical exposures in the workplace under the TSCA program. FIT FOR PURPOSE The similarities and differences among occupational risk assessment approaches were the theme of an online workshop held by AIHA last August that focused on the various goals of risk assessments conducted at NIOSH, OSHA, and EPA and in traditional industrial hygiene practice. Before we can appreciate how EPA’s risk assessments differ from many traditional IH risk assessments, it is helpful to acknowledge what they have in common. All risk assessments have three main elements: characterizing the hazard qualitatively and quantitatively (dose-response), assessing exposure, and quantifying and describing potential risk (risk characterization). At EPA, the risk assessment can only be carried out after a problem formulation phase, which defines the specific question that the risk assessment is designed to answer and outlines preliminary methods. This common foundation can support assessments developed for any context, but each assessment must be designed to be “fit for purpose”—that is, adapted to a specific scenario or chemical risk assessment question, similar to the way tools are designed for specific purposes. For example, you wouldn’t use a Phillips screwdriver to tighten a flathead screw. Examples of various types of risk assessment activities were presented in the workshop, as discussed below. At NIOSH, for example, some risk assessments support agency exposure control recommendations to protect workers from exposure to chemicals, some are undertaken as part of the process for evaluating health hazards at work sites, some are components of epidemiological studies, and some are intended to assess the efficacy of engineering controls in specific industries. From a dose-response perspective, health-based dose-response risk assessments that support NIOSH recommended exposure limits (RELs) are designed to protect workers from occupational exposure even if they are exposed every day for a working lifetime of 45 years. Even within an agency there are many different types of occupational risk assessment activities that rely on different data and analysis approaches. At OSHA, risk assessments undertaken in support of agency standards are intended to estimate how many workers are exposed within various exposure ranges for each industry sector. These estimates inform the agency’s calculation of the potential reduction in illnesses that will result from implementation of its risk-based standards that are developed through the health risk assessment approach. The amended TSCA requires EPA to undertake risk assessments of existing high priority chemicals, which the agency identifies from the tens of thousands of chemicals in commerce. For existing high priority chemicals, the agency is legally required to have 20 risk assessments in process at any one time. TSCA risk evaluations include assessment of occupational exposures that potentially affect both workers and occupational non-users. If EPA determines that a particular use of a chemical represents an unreasonable risk, the agency must determine how to manage those risks (for example, through regulation). Given EPA’s authority over occupational chemical exposures under TSCA and its requests for occupational data, there is a need for enhanced communication between EPA assessors and company IH professionals. HOW RISK ASSESSMENTS DIFFER The human health risk evaluations carried out by EPA must cover the wide variety of industries and workers that may handle a particular chemical at a national level. In contrast, an IH assessment is highly specific to a particular task at a particular location, with the objective of determining the needs of exposure control or mitigation. Consequently, it is important to bear in mind that these two types of assessments are meant to fulfill different purposes and take place in different decision contexts. The different approaches to exposure assessment can lead to different conclusions about risk that are important to understand and resolve​.

In this section, we discuss three main variables across risk assessments. These variables include the kinds of data the assessments are based on, exposure groups of interest, and the role of occupational exposure limits. Data the Assessments Are Based On In its risk evaluation approach, EPA is trying to use as much data as possible to make risk determinations and recommend risk management measures. The agency requests industry to provide occupational exposure data historically collected for industrial hygiene programs. Industrial hygienists follow a well-developed and widely adopted strategy to anticipate, recognize, evaluate, control, and confirm protection from workplace hazards, along with a prioritization scheme. This approach has been applied and recommended by OSHA, NIOSH, and AIHA, and is the basis for how occupational exposure data is integrated and analyzed. The AIHA publication A Strategy for Assessing and Managing Occupational Exposures calls for a tiered approach in which a qualitative assessment is used to identify groups who are potentially exposed under similar conditions and prioritize the quantitative assessment of the potential exposure those groups experience. Occupational exposure assessments are typically carried out first on the activities deemed to pose the highest potential for exposure, and not necessarily to characterize the universe of potential exposures in the workplace. Therefore, most of the quantitative exposure assessments submitted to EPA are likely to be of groups with greater potential for exposure, and the distribution of exposures is likely biased high. The data IHs have is not representative of all workers’ activities at the work site, nor is it representative of all workers in the same industry. Even the mean of the distribution should be considered as a conservative—if not high-end—estimate of the potential for exposures among all workers. For its risk assessments, EPA requests information associated with the process, production volumes, operating conditions, and presence and efficiency of engineering controls to predict potential exposures in the workplace. Data related to these factors are not typically collected in occupational exposure assessments, which are site-specific and not intended to be used for data aggregation. Although the site-specific nature of IH risk assessments represents a significant limitation for data aggregation, the opportunity exists for EPA to use the information to identify differences between industries, activities, or determinants of exposures, as well as to screen out processes that do not present unreasonable risk. Following the example of other regulatory agencies, EPA could leverage industry expertise and information from downstream users to help predict exposures.

A paper that describes in detail some of the commonalities and differences in OEL derivation methods was published in the Journal of Occupational and Environmental Hygiene in 2015. As explained in the JOEH paper, both risk science and risk policy differences exist. For example, new data, methods, or analytical protocol defaults can generate large differences in OEL values (see Figure 1). In terms of policy, differences in the requirement to include consideration of feasibility, broader impact, and acceptability of residual risk can impact OELs.