• Public Affairs

  • ACOEM Comments on NIOSH Carcinogen Classification Document

    February 12, 2014

    NIOSH Docket Office
    Robert A. Taft Laboratories
    4676 Columbia Parkway, MS C–34
    Cincinnati, Ohio 45226

    Re: CDC–2013–0023; NIOSH 240–A

    To Whom It May Concern:

    The American College of Occupational and Environmental Medicine (ACOEM) welcomes the opportunity to comment on the External Review Draft of the Current Intelligence Bulletin: Update of NIOSH Carcinogen Classification and Target Risk Level Policy for Chemical Hazards in the Workplace. This document was prepared by an ACOEM Presidential Task Force chartered to review and comment on the proposed draft and includes input from the ACOEM Board of Directors and other ACOEM members.

    ACOEM applauds the efforts of the National Institute for Occupational Safety and Health in developing this document. We do believe that the proposed carcinogen policies are consistent with the current scientific knowledge of toxicology, risk assessment, industrial hygiene, occupational cancer, and principles of carcinogenicity. Application of the proposed approach to classification and following the resulting recommended exposure limits (RELs) will lead to reduced risks to workers in settings in which they are potentially exposed to carcinogens. While the revised RELs will not be regulatory limits, they should provide an impetus for appropriate changes to the Occupational Safety and Health Administration’s permissible exposure limits (PELs) and for organizations to better control exposures to carcinogens.

    We concur with the NIOSH plan to rely upon the classifications of agents put forward by authoritative bodies, specifically the U.S. National Toxicology Program (NTP) in their Report on Carcinogens (RoC), the U.S. Environmental Protection Agency (EPA) utilizing their Guidelines for Cancer Risk Assessment, and the International Agency for Research on Cancer (IARC). As NIOSH indicates, the use of existing evidence-based classifications of agents precludes the need for NIOSH to duplicate this effort, thus allowing them to focus on worker protection efforts, including setting of appropriate RELs. We endorse the NIOSH proposed approach to utilize the most health-protective classification from these authoritative bodies with the potential exceptions that have been noted. We concur with the approach that NIOSH suggests regarding agents that are likely relevant to workplace settings but that have not been evaluated by the authoritative bodies, such as IARC or EPA, i.e., to nominate them for NTP study or to conduct an internal NIOSH assessment. While we understand the resources involved for NIOSH if they are to develop their own science-based carcinogen classification of an agent in this setting, we feel it is important that NIOSH attempt to “fill the void” in knowledge for occupationally important chemicals/agents. Also, for the purposes of harmonizing classification, we are comfortable with the plan for NIOSH to include their determination of the appropriate GHS (Globally Harmonized System of Classification and Labeling of Chemicals) category, but the original risk categorizations of IARC, NTP and EPA should be retained when they are available for an agent. We do generally agree with the validity of the NIOSH correspondence table (Table 2) and its usefulness as a guide to determine GHS hazard categories.

    The IARC and EPA classification systems provide categories that effectively correspond to designating an agent as a known, a probable, or a possible human carcinogen based upon available scientific evidence (NTP in the RoC does not include the latter designation). We agree with NIOSH that the term, “potential occupational carcinogen,” does not accurately describe the state of knowledge regarding occupationally relevant known human carcinogens, such as benzene and asbestos. It would be appropriate to describe these agents, for which there is sufficient evidence of carcinogenicity in humans, as occupational carcinogens. However, in the approach proposed by NIOSH, an agent would be designated as an occupational carcinogen, if it were to fall into any of these three levels of evidence groups and if it were occupationally relevant. The decision by NIOSH to label all “occupationally relevant” agents that fall into any one of these categories as occupational carcinogens tends to blur the evidence-based distinctions indicated by these agency classification systems, even though NIOSH intends to list the authoritative body determinations after the occupational carcinogen label. While this appears to be a laudably health-conservative approach, it may “deflate” the perceived importance of the label and may result in misallocation of limited preventive resources by treating a known human carcinogen, such as benzene, in the same fashion as a possible human carcinogen, such as phenyl glycidyl ether or titanium dioxide. Furthermore, this “leveling” may impede the ability to place risks into proper perspective in risk communication efforts directed to workers. Intuitively, it seems reasonable to focus more energy on prevention of exposure for known human carcinogens than for probable or possible human carcinogens, particularly for agents in the latter group with only limited evidence for carcinogenicity in experimental animals or for which the mechanism of carcinogenicity in animals likely does not apply to humans.

    We agree with the proposed use by NIOSH of quantitative risk assessments to determine the cancer risk from working lifetime exposures to low concentrations (doses) of occupationally relevant agents, including the central and 95% lower confidence limit estimates of risk. NIOSH may want to consider relying upon other well-supported cancer risk assessments, e.g., from EPA, for this purpose rather than developing their own assessment. We agree that primacy should be given to selecting data stemming from high-quality epidemiologic studies or animal studies using relevant exposure routes (for use in developing these risk estimates). These estimates can then be used in developing cancer RELs for these agents. Because the resulting estimates from quantitative risk assessments will vary based upon the selected study data source and assumptions used in mathematical modeling, NIOSH should attempt to select the most appropriate study data and risk assessment approach to utilize in setting the REL. Doing so will result in RELs which will be health-protective but not necessarily the most health-conservative (if the latter would be less relevant to the occupational setting). We suggest that NIOSH specify in this document how they will make this selection between alternative risk assessment approaches or studies. Similarly, NIOSH should specify, in the material supporting the REL for a specific agent, the basis for the selection of the risk assessment approach utilized.

    While we understand the potential legal/regulatory and practical considerations that led NIOSH to recommend a target risk level of 1 in 1,000, we believe that that proposed level may be insufficiently protective, particularly in consideration of the likely presence of susceptible individuals and subgroups of individuals in the workplace. With this in mind, we respectfully suggest the following approach for setting the target risk levels. For at least those occupationally relevant carcinogens for which there is sufficient evidence to be categorized as known human carcinogens (based upon authoritative body or NIOSH determinations), we recommend using a target risk level of 1 in 10,000. For those agents that are categorized as probable or possible human carcinogens, i.e., those for which there is insufficient human evidence of carcinogenicity, we think it may be appropriate to use the proposed target risk level of 1 in 1,000. In general, based on both scientific concerns and ethical/philosophical grounds, we urge NIOSH to adopt a more protective posture in selecting the appropriate target risk level for carcinogens.

    We understand and agree in principle with the proposed approach by NIOSH to set the REL at the limit of quantitation (LOQ) of the sampling and analytical method, the “REL-AF” (the analytically feasible REL), in those cases in which it is not analytically feasible to measure the concentration of the agent at the level of the health-based REL. However, in some cases, it may not be technically feasible to measure the concentration of the agent with adequate precision at levels as low as the LOQ (i.e., within ±25% of the true value 95% of the time). Accordingly, we recommend that NIOSH set (and publish) the “REL-AF” at the lowest level above the health-based REL at which measurements can be made with adequate precision. We recommend this approach because we believe that the “REL-AF” should be feasible and implementable. We recommend that NIOSH also publish the health-based REL in these situations. This approach would provide the greatest amount of useful information: a target goal to which NIOSH and organizations can aspire (should technical methodology improve), while also providing a practical and implementable REL for current use.

    For occupationally relevant agents which are known, probable, or possible carcinogens (as determined above) and for which there is reasonable scientific evidence for dermal absorption as a route of exposure (based upon animal or human evidence), we recommend that NIOSH indicate this potential risk when establishing an REL. The determined REL could then have a “skin” designation, akin to the approach used by the American Conference of Governmental Industrial Hygienists (ACGIH) in publishing threshold limit values (TLVs). For these agents with potential for dermal absorption, NIOSH would then recommend the use of appropriate personal protective equipment that would prevent skin exposure to workers.

    In terms of setting a REL, NIOSH should also consider how they would address certain types of “agents,” such as shift work involving night work and occupations that are known, suspected, or possible risk factors for occupational cancer (even though the specific agent responsible for the increased risk may not have been identified). In these cases, it does not seem that one could set a recommended exposure limit, at least not in a fashion similar to that for specific chemical carcinogens.

                                                                             * * *

    ACOEM, an organization of more than 4,000 occupational physicians and other health care professionals, provides leadership to promote optimal health and safety of workers, workplaces, and environments.

    Thank you for your consideration of our comments. Please do not hesitate to contact me or Dr. Michael Fischman, the Chair of ACOEM’s Task Force, at 925-283-2366, should you have any questions.


    Ronald R. Loeppke, MD, MPH, FACOEM