Comment: Each set of supplemental guidelines should be considered individually and collectively to assure compliance with the legislative intent of SB 1731 (e.g., reducing uncertainty and compounded conservatism and use of best available science).

OEHHA should assure that the opportunity is provided for public review of the final risk assessment process once all of the revised guidelines have been issued, and prior to approving any of the individual guidelines, or finalizing the integrated process for conducting health risk assessments under the revised AB 2588 program. In evaluating the cumulative impact of the guidelines, it will be essential for reviewers to consider the interaction between the various component guidelines (acute, chronic non-carcinogen, carcinogen, and exposure assessment). To achieve this review, OEHHA should incorporate an additional, extended public review period for the completed revised risk assessment process before any action is taken to finalize and/or approve any of the component guideline documents.

Response: Public and Scientific Review Panel (SRP) review of all the components of the guidelines is taking place. The final risk assessment guidelines will contain information from each of the Technical Support Documents, and will also undergo public and SRP review. In the end, the entire set of documents will have undergone public and peer review.

Comment: OEHHA should not direct limited resources toward developing new health values where USEPA values or other ARARs already exist.

OEHHA should focus their limited resources on developing appropriate supplemental risk assessment guidance for the areas explicitly addressed in SB 1731 (e.g., uncertainty analysis, exposure assessment). Development of exposure limits for assessing health risks for acute, chronic non-carcinogen, or carcinogen exposures should be deferred to U.S. EPA for application within specific state programs.

Response: The U.S.EPA has a limited number of cancer potency factors and chronic noncancer reference concentrations which we can use. To date, USEPA has not developed any acute reference concentrations for routine or planned releases. OEHHA has filled this void by developing acute Reference Exposure Levels for one-hour exposures.

As stated in the statute (Health and Safety Code Section 44306), "health risk assessment means a detailed comprehensive analysis prepared pursuant to Section 44361 to evaluate and predict the dispersion of hazardous substances in the environment and the potential for exposure of human populations and to assess and quantify both the individual and population-wide health risks associated with those levels of exposure." Health risk assessments require a number of inputs for risk characterization. Exposure assessment is only one of them and exposure information alone does not constitute a risk assessment. In order to characterize the risks from exposure one needs information on the toxicity of the chemicals emitted. The acute and chronic reference exposure levels for noncancer endpoints, and the cancer potency factors function as measures of the toxicity and are used quantitatively to assess health risks. Development and compilation of existing cancer potency factors and reference exposure levels is integral to the development of comprehensive risk assessment guidelines. Costs of developing the reference exposure levels and cancer potency factors were included in the cost estimations of SB 1731 legislation prepared for the legislature. It was therefore clear from the beginning that some effort was going into developing these toxicity criteria. U.S. EPA has neither the resources nor the mandate to address the chemicals subject to California’s Air Toxics Hot Spots Act.

Comment: The guidelines establish values and methodologies that will be used in the risk assessment process. They also introduce risk management elements that will bias regulatory decision making. For example, in addressing Risk Characterization (Presentation of Results, pg. 12), OEHHA uses a hazard index of 0.5 as a trigger for evaluation of, and inclusion of, the potential additive health risk from background concentrations of criteria air pollutants for respiratory irritants. This provision is not mandated by AB 2588 or SB 1731 and would limit CARB and local district discretion in making cost-effective risk management decisions. Such initiatives should be addressed independently of the revised risk assessment guidelines to allow for full consideration of potential risk management implications (e.g., increase in a facility’s risk above a district notification threshold due to the conservative default assumption of additive toxicity from background criteria air pollutants). It is critical that risk assessment and risk management activities be considered separately, so that informed regulatory decisions can be based on a full evaluation of all relevant factors. Therefore, we recommend that OEHHA remove all such risk management references from the draft. Of course, we recommend that subsequent draft guidelines also observe this critical separation between risk assessment and risk management.

Response: The proposal to consider background levels of criteria air pollutants has been dropped.

Comment: The Executive Summary of the Technical Support Document (pg. iv) states that "The RELs also have application in characterizing health risks in event of unplanned releases." This statement suggests an inappropriate broadening of the applicability of these guidelines, which are being developed under legislative mandate for the express purpose of addressing the conduct of health risk assessments for facilities under the AB 2588 program. Such expansion of the guidelines’ applicability beyond conducting AB 2588 risk assessments appears to exceed the legislative intent and authorization of SB 1731. In addition, given the questionable scientific basis of the animal data used to derive some of the REL values (see Appendix 1), it is clearly premature to contemplate broadening the regulatory applicability of the proposed RELs. Moreover, we are concerned that such statements may be cited inappropriately in tort litigation against facilities, as supporting exaggerated claims of adverse health effects simply because an REL may have been exceeded during an accidental release. Full consideration of the compounded conservatism in establishing RELs, particularly in the choice of "endpoints of toxicity" in the hazard identification phase, combined with the default application of ten-fold uncertainty factors, and the methodology proposed for addressing the relative contribution of concentration and time using a modification of Haber’s Law in the dose-response phase, clearly leads to the conclusion that the resultant acute RELs are uniformly set at levels that are likely one or more orders of magnitude below exposures that actually result in injurious effects. We recommend that OEHHA strike this statement from the draft document, and add a precautionary statement which clearly addresses such inappropriate applications of the guidelines.

Response: The proposed acute Reference Exposure Levels were already under development prior to the passage of SB 1731. The effort to develop acute RELs was started in 1989 (under a different program in conjunction with California’s Office of Emergency Services), and the guidance document thus produced was publicly reviewed in 1991. When SB 1731 was passed, we expanded our efforts to include evaluating exposures to routine emissions. OEHHA initially took the existing guidelines and modified them to fit the Air Toxics Hot Spots program risk assessment guidelines by developing RELs for one hour community exposures. OEHHA has since altered the document to emphasize its applicability in conducting risk assessments under AB 2588.

As part of our reevaluation of the acute RELs following public comment, OEHHA has analyzed the use of the benchmark concentration to reduce uncertainty and the magnitude of the uncertainty factors. Where data are adequate, we have used the benchmark concentration approach. The toxicological endpoint reflects the scientific literature and does not introduce "compounded conservatism". The application of default ten-fold uncertainty factors in a REL calculation accounts for the lack of scientific information (uncertainty) in extrapolating from animals to humans and the degree of interindividual variability in the human population. There are no data to suggest that the RELs are set at levels "one or more orders of magnitude below exposures that actually result in injurious effects". It is because inadequate data are available that uncertainty factors are used. OEHHA is currently studying the degree of protection which 10-fold uncertainty factors afford.

Comment: Health based ambient air quality standards have been established for criteria air pollutants under federal and state ambient air quality programs. Under these programs, facilities must meet stringent criteria pollutant emission standards for existing equipment and any new or modified equipment. Inclusion of criteria air pollutants in the AB 2588 program (Table 1) will overlap efforts of ambient air quality programs. Such action will increase AB 2588 compliance costs and complicate the program without a corresponding improvement in human health and environmental protection. Criteria pollutants are not listed as hazardous air pollutants (HAPS) under Section 112 (42 USC section 7412(b)) and are therefore not required to be defined as a toxic air contaminant under state law. Criteria pollutants are not among the substances required to be quantified in the AB 2588 program. In addition, SO₂, NO₂, and ozone are not included in any of the substance lists in H&SC section 44321. In view of these facts, criteria pollutants should not be listed as toxic air contaminants.

Response: There has been no attempt at listing criteria air pollutants as toxic air contaminants. Since exposure occurs simultaneously to criteria air pollutants as well as hot spots chemicals emitted from a facility, from a scientific standpoint, evaluation of the health impacts from exposure to criteria air pollutants and hot spots emissions makes sense. In spite of this, consideration of background concentrations of criteria pollutants will not be required in the Hot Spots Program.

Comment: The guidelines provide examples of facility emissions that must be evaluated in the risk assessment, including venting to flares and pressure relief valves (pp. 4). The facility risk assessment mandated by the Legislature for the AB 2588 program applies specifically to impacts due to air releases that result from routine operation of a facility, or release that are "predictable". The above-referenced examples should include language clarifying that the guidelines are only applicable to routine or predictable facility emissions.

Response: Health and Safety Code Section 44303 defines "air releases" as "any activity that may cause the issuance of air contaminants, including the actual or potential spilling, leaking, pumping, pouring, emitting, emptying, discharging, injecting, escaping, leaching, dumping, or disposing of a substance into the ambient air and that results from the routine operation of a facility or that is predictable, including, but not limited to, continuous and intermittent releases and predictable process upsets or leaks." Venting to flares and pressure relief valves are predictable occurrences. Therefore, they should be included in evaluating health risks from facility emissions.

Comment: The AB 2588 program is intended to identify risks from individual facility emissions. In discussing the legislative intent of the AB 2588 program, H&SC section 44301 makes explicit reference to "specific sources". Subsection (h) states: "it is in the public interest to ascertain and measure the amounts and types of hazardous releases from specific sources that may be exposing people to those releases, and to assess the health risks to those who are exposed." There is no legislative mandate to address multiple-facility impacts. OEHHA guidance should not overreach clear legislative intent. All references to requirements or recommendations associated with multiple facility impacts should be removed from the draft.

Response: H&SC section 44301 (d) states that toxic air releases "may create localized concentrations or air toxics "hot spots" where emissions from specific sources may expose individuals and population groups to elevated risks of adverse health effects, including, but not limited to, cancer and contribute to the cumulative health risks of emissions from other sources in the area." It is clear that the legislature is concerned about cumulative exposures. Based on this statement, a District at its discretion may do an analysis of multiple facility impacts and ask for information from facilities in order to conduct such an analysis. OEHHA never intended to imply that a facility operator do an assessment of the risks from neighboring facilities.

TECHNICAL COMMENTS
The following comments specifically address the methodology used by OEHHA in developing the draft guidelines:

Comment: The use of uncertainty factors, and their magnitude, must take into account the severity of the effect against which the public is being protected. Particularly for Level I Acute Toxicity Exposure Limits (generally the REL), the endpoint on which the exposure limit is based is generally a mild effect of questionable toxicological significance (mild respiratory irritation, mild eye irritation, minimal signs of CNS depression). In many cases an uncertainty factor far less than 10 (e.g. 2-3) would provide adequate protection.