Skip directly to search Skip directly to A to Z list Skip directly to site content

PUBLIC HEALTH ASSESSMENT

January 5, 2002 Air Release

DIAZ CHEMICAL CORPORATION
(a/k/a FMC C/O DIAZ CHEMICAL C/O FMC)
VILLAGE OF HOLLEY, ORLEANS COUNTY, NEW YORK


APPENDIX G: COVER LETTER AND ELIGIBILITY SCREENING QUESTIONAIRE FOR THE NEW YORK STATE DEPARTMENT OF HEALTH VOLATILE ORGANIC COMPOUNDS (VOC) EXPOSURE REGISTRY

Click here to view Appendix G in PDF format (PDF, 37KB)


APPENDIX H: SUMMARY OF COMMENTS RECEIVED FROM THE PUBLIC AND RESPONSES

Upon release of the public health assessment (PHA) in November 2002 until February 10, 2003,the public was invited to submit written comments about the document. This Appendix is asummary of the comments including those submitted at the December 2002 public meeting. Some comments were consolidated or grouped together to incorporate similar concerns. Wesolicited comments on the draft PHA to understand remaining community concerns and tounderstand if the information in the PHA is accurate and thorough. In return, we provide thissummary of comments and a written response. The response may be to provide additionalinformation; it may reference changes to the PHA draft including changes to therecommendations and public health action plan. The response may also be further discussion,and a written response to comments. If you have any questions, please contact the HealthLiaison Program of the New York State Department of Health (NYSDOH) at the toll-freenumber 1-800-458-1158.

Comments Primarily about Toxicology

Comment 1. The estimates of risk are biased high because the New York State Department ofHealth (NYSDOH) biased estimates of exposure high by assuming 30 year exposure and usingmaximum values for air and urine in risk calculations. This is not justifiable given that themaximum values were found in the earliest rounds of sampling and that the Public HealthAssessment (PHA) shows that levels in all media are decreasing. The risk assessment is overlyconservative and overstates the estimate of risk. Because multiple worst case scenarios areassumed, the risks are over estimated.

Response 1. We acknowledge that, in some cases, the methods we used in the PHA mayoverestimate certain aspects of the risks. We also know that there is uncertainty in our riskcharacterization as acknowledged in the PHA because of many factors including, but not limitedto, that there were other contaminants in the release and that the toxicological information about2-chloro-6-fluorophenol (CFP) is limited. However, using the stated exposure assumptions forCFP, the estimates of risk were minimal for possible noncancer effects or low to very low forpossible cancer effects (see Tables 5b, c, d). We recognize that the risk may be lower or higher.

Comment 2. Compounding the over estimation of risk was the fact that NYSDOH used2,4,6 - trichlorophenol as a surrogate for assessing cancer when the Agency for ToxicSubstances and Disease Registry (ATSDR) 2002 study determined CFP was not acarcinogen. What is the rationale for using 2,4,6-trichlorophenol as a surrogate for cancerrisks?

Response 2. This question is related to comment 3 and further information about the issueis provided in response 3. 2,4,6-Trichlorophenol was used as a surrogate to estimatepossible cancer risks for exposure to CFP because it is the only chlorophenol that isstructurally similar to CFP that has been tested and has caused cancer in an animal study.As described in the Public Health Assessment, chemicals with similar structures (e.g., size,shape) often have similar toxicological, including carcinogenic, effects. Therefore, weassumed that CFP might cause cancer in humans at the same levels of exposure that 2,4,6-trichlorophenol caused cancer in animals. Another closely related chlorophenol (2,4-dichlorophenol) did not cause cancer in animals and ATSDR's structure activity analysissuggested CFP was not likely to cause cancer. However, this was a computer-basedevaluation and CFP has never been tested for carcinogenicity in an animal study. We donot know with certainty, however, whether CFP would act more like 2,4,6-trichlorophenolor more like 2,4-dichlorophenol if people were exposed to it for a long time. To ensure thatpossible carcinogenic risk was considered, we assumed that CFP might act more like 2,4,6-trichlorophenol than 2,4-dichlorophenol. We also assumed that individuals would beexposed for 30 years over a period of 70 years (lifetime).

Comment 3. The public health implications and potential health effects from exposure toCFP was based on the assumption that chemicals with similar structures often cause similarhealth effects and the potential health effects were based on information on chlorophenols.Although this seems reasonable given the lack of toxicity information on the compound, itwould be important to get other opinions on the reasonableness of this critical assumptionand/or perhaps what other options are available.

Response 3. The fact that chemicals with similar chemical structures often have similarbiological or chemical activities is well accepted in toxicology, pharmaceutics andchemistry. Relationships between chemical structure and biological activity (as well asphysical/chemical properties, etc.) are termed structure activity relationships or SARs. (Infact, recognition of the value of SARs in predicting biological activity underliesgovernment implementation of the Toxic Substances Control Act (TSCA). For example, theInteragency Testing Committee, which independently advises the U. S. EnvironmentalProtection Agency (EPA) of chemicals in need of testing under TSCA, has a statutorymandate to consider SARs when recommending chemicals for testing. Further, TSCArequires that chemical manufacturers submit pre-manufacture notifications (PMNs) priorto manufacturing new industrial chemicals. TSCA does not require testing prior tosubmission of a PMN; therefore, few data are submitted and SARs are used to predicthealth effects.) Several very recent reviews on the value of SARs in predicting biologicalactivity of chemicals in general, and in predicting biological activity of chlorinatedchemicals specifically are available (Faustmann & Omenn 2001; McKinney et al. 2000;Cronin et al. 2003; Willes et al. 1993). Additionally, two other scientific reviewsspecifically discuss how the replacement of a hydrogen next to the hydroxyl (-OH) groupwith a fluorine would be likely to affect biological activity of a phenol (Park andKitteringham 1994; Park et al. 2001). Information described in these reviews suggests thatCFP (2-chloro-6-fluorophenol) is likely to have biological activity more similar to 2-chlorophenol, than to either 2,4-dichlorophenol or 2,6-dichlorophenol.

The view of several NYSDOH staff with appropriate expertise as well as scientists fromother agencies (ATSDR, EPA) that reviewed the PHA, reflect the consensus of scientificopinion. The consensus is that, in the absence of data, consideration of SARs is a valuableand useful means of evaluating potential biological activity for chemicals.

Comment 4. The statement about the similarity of CFP to monochloro or dichlorophenols ischemically simplistic. CFP would be similar to 2,6-dichlorophenol; but not otherdichlorophenols; fluorine has a higher electronegativity and would produce a stronger acidity forthe phenol hydroxyl.

Response 4. We agree that fluorine has a greater electronegativity therefore suggesting a greater acidity for a CFP molecule versus a two chlorine phenol. However, the CFP chemistry is not that simple as evidenced by examining the pKa of CFP and other halogenated phenols. The pKa, as we calculated using Hammett's equation for CFP, is 7.24, in comparison to a measured pKa of 6.78 for 2,6-dichlorophenol. The pKa analysis suggests that the acidity may not be increased (lower pKa) by the change to fluorine. Most likely this is because the hydrogen bonding and inductive bonding reduces the effect of the electronegativity and, therefore, mitigate changes in the acidity. In summary, we recognize that replacing chlorine with fluorine changes the chemistry of the molecule. We also recognize that the acidity is not the only characteristic of the CFP molecule that may change with the change from fluorine to chlorine.

The assumption made in the PHA that CFP is similar in chemical structure and thereforebiological activity to monochloro- and dichlorophenols is based on the widely acceptedprinciple that chemicals with similar chemical structures very frequently have similarbiological activity. CFP is very similar in structure to both 2-chlorophenol and 2,6-dichlorophenol. The only difference among these three compounds is that the atom boundat the 6 position in the phenol ring (the position next to the hydroxyl (-OH) group) isdifferent. The CFP molecule has a fluorine atom, the 2-chlorophenol molecule has ahydrogen atom, and the 2,6-dichlorophenol molecule has a chlorine atom bound there. Theeffect these atoms are likely to have on biological activity of the phenol depends uponseveral factors. Some such factors are the effect of the atom on the reactivity of the -OHhydrogen; the effect of the atom on the electron resonance of the phenol aromatic ring; theeffect of the atom on the overall size of the molecule; and, the effect of the atom on thehydrogen binding of the -OH hydrogen to chlorine or fluorine in the 2 or 6 position.Consideration of these structural and chemical properties of hydrogen, fluorine andchlorine atoms along with consideration of the effect these atoms are likely to have on thechemical properties of the phenol molecule suggests that CFP probably has characteristicsin common with both 2-chlorophenol and with 2,3-, 2,4- and 2,6-dichlorophenol. Althoughthe comparison of CFP to mono- and dichlorophenols may appear simplistic, it is based onthe fundamental principle of structure-activity relationships.

References:

Cronin, MTD; Jaworska, JS; Walker, JD; Comber, MHI; Watts, CD; Worth AD. 2003.doi:10.1289/ehp.5760 (available at http://dx.doi.org/)

Faustmann, EM; Omenn, GS. 2001.Risk Assessment. In: Casarett & Doull's Toxicology.The Basic Science of Poisons (Klaassen C.D., ed.) 6th Edition. McGraw-Hill

McKinney, JD; Richard, A; Waller, C; Newman, MC; Gerberick, F. 2000. The practice ofstructure activity relationships in SAR in toxicology. Toxicological Sciences. 56:8-17.

Park, K; Kitteringham, NR. 1994. Effects of fluorine substitution on drug metabolism:pharmacological and toxicological implications. Drug Metab Rev 26:605-643.

Park, BK; Kitteringham, NR; O'Neill, PM. 2001. Metabolism of fluorine-containing drugs.Annu. Rev. Pharmacol. Toxicol. 41:443-470.

Willes, RF; Nestmann, ER; Miller, PA; Orr, JC; Munro, IC. 1993. Scientific principles forevaluating the potential for adverse effects from chlorinated organic chemicals in theenvironment. Reg. Toxicol. Pharmacol. 18:313-356.

Comment 5. It is clear that a population of residents was acutely exposed to this mix ofcompounds and the short-term acute effects are now resolving. The long-term potential healtheffects are completely unknown and special attention should be directed to those particularly atrisk for sequelae from acute exposures, in particular children and the elderly, especially thosewith existing chronic conditions. Why have you included no infant or child exposure/riskassessment in the PHA?

Response 5. The ATSDR and the NYSDOH consider children when evaluating exposurepathways and potential health effects from environmental contaminants. Children are of specialconcern because of their greater potential health effects from environmental contaminants andbecause of their greater potential for exposure from play and other behavior patterns. Childrensometimes differ from adults in their susceptibility to the effects of hazardous chemicals, butwhether there is a difference depends on the chemical. Children may be more or less susceptiblethan adults to health effects from a chemical and the relationship may change withdevelopmental age.

In this PHA children were explicitly considered when estimating potential exposure to CFPbased on urine levels as described in Appendix F, Table F1 and Table F2. CFP exposures wereestimated assuming physiological characteristics typical of children and these exposures werethen compared to available toxicity values.

Estimated exposures of residents to CFP were compared to reference doses (RfDs) and referenceconcentrations (RfCs) (estimated from reference doses) in the PHA. RfDs reflect long-term dailyexposures that are unlikely to result in an adverse noncancer health effect. RfCs are analogous toRfD's and represent air concentrations that are unlikely to cause adverse noncancer healtheffects over long-term exposure. These toxicity values incorporate uncertainty factors thatconsider sensitive populations such as children or the elderly.

Comment 6. Much of the discussion on toxicity and potential health implications of thechemical release surrounds the chemical 2-chloro-6-fluorophenol (CFP). However, thiscomprised only 55 percent of the material that was released into the environment. It is unclearfrom the document whether any effort has been made or is currently underway to assessexposure to those other compounds released and to garner information regarding their toxicityand potential health effects.

Response 6. As noted in the PHA, toluene was the other contaminant released in fairly highconcentrations on January 5. Toluene is a highly volatile solvent that has been associated withshort term effects such as fatigue, headache and dizziness in individuals exposed to high levels(about 380 milligrams per cubic meter (mg/m3) or more in air) (ATSDR 2000). It is not knownwhether levels of toluene resulting from the release were high enough to have caused these typesof effects. If they were, these effects would have diminished quickly as toluene was brokendown and/or carried away in the wind. Toluene is rapidly broken down in the atmosphere so thateven if there were no wind, after about 3-4 days levels of toluene resulting from the releasewould be very low. This makes long term exposures to elevated levels of toluene unlikely sothat long term health effects are unlikely to occur.

As noted in the PHA, several other chemicals in addition to CFP and toluene were present in thevessel residue and in soil and wipe samples collected after the release. The analytical methodsavailable to us are able to tell us what general class these chemicals belong to, but are not able toidentify the exact chemicals. Thus, the chemicals are called "tentatively identified chemicals" orTICs. The TICs found in the reaction vessel, soil and wipe samples are most likelydichlorofluorophenols, benzyldichlorophenols, chlorofluorophenol dimers andchlorofluorophenol-dichlorofluorophenols dimers. Because the exact identity of the chemicals isunknown, it is difficult to determine whether they might be associated with health effects. Oneway to evaluate the effect of possible exposures to the TICs is to consider information forchemicals that have similar chemical structures and are therefore likely to have similar toxicity.Based on this we would expect dichlorofluorophenols to be similar to other chlorophenols andCFP in toxicity. However, because they were present in lower amounts than CFP, exposureswould probably be lower and therefore health effects would be less likely. Information on healtheffects for classes of chemicals similar to the other TICs suggests that the health risks associatedwith them are low. Although we would not expect health effects to be associated with theseTICs, there are uncertainties in using information for similar chemical to evaluate the TICs.

Reference:

ATSDR. 2000. Toxicological Profile for Toluene.

Comment 7. Unfortunately, as noted in the report, "health effects information about CFP islimited…virtually no information concerning short-term, high-level exposure and long-term,low-level exposure to CFP is available." This is because unlike chemicals used as medicines,which must be tested for safety and side effects before use, industrial chemical products andemissions historically have not routinely been screened for health effects despite widespreadexposures. We are very concerned that the NYSDOH has made its conclusions andrecommendations in the Public Health Assessment without any concrete knowledge about thelong-term health effects of CFP. Page 1 of the PHA states that "Health Effects on CFP arelimited." So no one really knows the true effects. Correct?

Response 7. (See response to comment 3). Because no information is available about CFP, thePHA relied on a structure-activity approach to draw conclusions about the likelihood ofexperiencing adverse health effects due to the release of CFP on January 5. Other governmentagencies and scientists have used a structure activity approach to evaluate the toxicity of otherchemicals such as specific polycyclic aromatic hydrocarbons (PAHs). The scientific premise ofthis method is that chemicals with similar structure evoke similar physiological outcomes orhealth outcomes.

Also, the PHA relied on the observation that CFP exposures, as measured in urine, declined afterthe initial release. Based on health effects associated with chemicals similar to CFP and based ondeclining exposures, minimal (noncancer) and low to very low (cancer) risks were estimated. The risk estimates were made based on the highest urine result, which was found in the firsturine sampling round (January 2002). The risks were calculated assuming an exposureconsistent with this urine result for a 30-year period. Subsequent urine sampling showed levelsmuch lower than those found in January 2002. Based on the lower urine results, the risks wouldbe lower and our current estimate high.

Comment 8. Has there been follow-up or more recent data for the chlorophenol family ofchemicals than 1979?

Response 8. There are several references about chlorophenols used in the PHA that werepublished after 1979. Some of the references are; a study of the carcinogenicity of 2,4-dichlorophenol in rats and mice by the National Toxicology Program in 1989 (NTP 1989), theATSDR toxicological profile for chlorophenols produced in 1999 and Lin et al's (1999) reporton pentachlorophenol. The 1989 study of 2,4-dichlorphenol found that when rats or mice werefed high levels of 2,4-dichlorophenol every day for their entire lifetime, they did not developcancer. A good overview of the information available on chlorophenols is the ATSDRtoxicological profile. This document can be found at this internet address: http://www.atsdr.cdc.gov/toxpro2.html.

Comment 9. How does the toxicology data from feeding rats relate to threat for the people ofHolley considering that the people of Holley had both dermal absorption and ingestion routes ofexposure?

Response 9. In Appendix F of the PHA, CFP in urine was used as one measure of CFP exposure.As explained in the PHA, urine levels of CFP reflect total CFP exposure resulting from allpossible pathways of exposure, including dermal absorption and ingestion routes of exposure. Urine levels are a good measure of exposure based on the results of controlled scientificexperiments in which rats were given known amounts of chlorophenols and chlorofluorophenolsand their absorption, metabolism and elimination in urine were studied.

The toxicology data from feeding rats provides information about how the bodies of mammalshandle CFP and indicated that CFP is likely to be rapidly absorbed, likely to be extensivelymetabolized in the liver to a glucuronide conjugate and likely to be rapidly excreted in the urine.CFP is unlikely to accumulate in the body. In the absence of information to the contrary,mammals are assumed to handle chemicals in very similar ways, including assuming thathumans will handle CFP similarly to how rats handle closely related chemicals.

Therefore, estimates of total individual exposure based on urine CFP levels, when expressed asmilligrams of CFP per kilogram body weight (mg/kg) are assumed to be comparable toexposures of rats expressed as mg CFP/kg body weight when they are fed diets containing CFP.

Comments about Urine Sampling and Health Effects

Comment 10. Perhaps the most important part of the public health action plan is to establish andmaintain a registry of all the residents of Holley, New York, including those exposed and notexposed to the plume release. Exposure estimates could be established based on the knowndistribution of the exposure plume, residential history and time spent in the exposed area at thetime of the release and subsequently. It would be helpful to the community to know exactly howthis exposure registry will be implemented, the time frame for implementation and how thecommunity will be solicited and informed about its objectives and goals.

Response 10. An exposure registry is a resource tool that may help us learn whether exposuresto chemicals, like CFP, are related to long-term health effects. Past and current healthinformation is gathered using a questionnaire. Then, contact with that person is maintained, sothat health status updates can be obtained periodically.

The first steps to implement the Volatile Organic Compounds (VOC) Exposure Registry in theVillage of Holley have begun. Residents throughout the Village of Holley were contacted at theend of January and again at the end of April through a screening mailing, inquiring as to theresident's interest in participation. This screening mailing also includes information about theregistry's goals and objectives, and gives a toll-free telephone number to call if the person hasquestions or concerns. We are currently receiving responses to this initial mailing. Examples ofthe letter and screening mailing have been added as Appendix G to the PHA.

After we receive screening questionnaires indicating a willingness to participate, the residentsare sent questionnaires to complete. The questionnaires ask about time spent in the Village ofHolley, past and current health status of each household member, and other factors related tohealth such as smoking history. Questions about current and past cancer diagnoses, as well asrespiratory, neurological, cardiovascular, gastrointestinal, musculo-skeletal, endocrine, andreproductive symptoms and diseases are included.

The information collection process and initial review of health status information for a new siteare projected to take from six to twelve months. A status report on the success of enrollmentefforts and a summary of health information is to be provided to the community within the next24 months, with a target of June 2005. The status report will include a written description of thegroup of people who are enrolled in the registry. This description will include the group'scurrent age range and age during residence, their length and timeframe of residence, and the typeof potential chemical exposures they experienced, based on prior sampling data. The report willalso provide descriptive information about the types and numbers of health problems reportedand will compare these, where feasible, to national, state or local data. For small populations,with small numbers of health outcomes reported, the analysis may be limited to ranking the mostfrequently reported health conditions and comparing this ranking to a ranking of these conditionsin the general population. If a high percentage (over 50%) of the eligible population participates,sufficient numbers of some types of health conditions may allow the calculation of rates perpopulation that can be compared to national or more local data. Included with the status reportwill be a description of the characteristics of VOC registrants and a summary of health status atthe time of enrollment.

Those enrolled in the registry and all residents of Holley will be kept informed of any researchresults. Information will be shared with VOC registrants and, if they wish, their health careproviders if new information becomes available that points to potential health problemsassociated with these types of exposures. All information provided by registry participants isstrictly confidential, and no individual information is provided in reports. After householdmembers have agreed to enroll in the registry and have returned the completed questionnaire,NYSDOH will contact them approximately every two years and ask for updated healthinformation for each household member.

In addition, the VOC Exposure Registry will gather and evaluate data from vital statisticsrecords in order to assess information on births and deaths occurring in the exposure area duringthe exposure timeframe and appropriate post exposure timeframe. Evaluation of birth data maybe possible, if numbers are sufficient, for birth weight and sex ratio. Births will be matched toCongenital Malformations Registry data to evaluate data on major congenital malformationsusing specific birth information. NYSDOH is currently developing resources and methods forthis project and is unable to estimate a date for completion of the vital statistics evaluation. Routinely collected information on cancer diagnoses will also be assessed for the Village ofHolley. After a waiting period of at least five years after exposure, to take account of the latentperiod between the initiation of cancer and the diagnosis of cancer, addresses within the area ofexposure will be matched to New York State Cancer Registry records for an evaluation of cancerincidence (total cancer and specific types of cancer) in the exposure area in comparison toexpected cancer incidence based on statewide averages.

Comment 11. There should be discussion about the need for continued medical surveillancegiven the unknown consequences of exposure to this substance. For example, would it beprudent to assess neurobehavioral performance in children exposed or lung function/bronchitisamong susceptible groups, i.e., asthmatics, smokers.

Response 11. As discussed in the Response 9, in 12-24 months NYSDOH will issue a statusreport summarizing the registration of individuals into the VOC registry and associated healthinformation. NYSDOH will evaluate self-reported health data collected through the VOCExposure Registry and data from birth, death and cancer records as described in Response 9 inorder to track health outcomes for the next ten years. If the health information that is gatheredsuggests ongoing unusual health problems that may be related to exposures due to the releasefrom Diaz in January of 2002, additional actions such as medical surveillance or studies may bedeveloped.

Comment 12. The PHA should describe the methods used for analyzing the self-reported healthsymptoms and determining whether such symptoms were related to pre-existing healthconditions, smoking, flu, allergies, asthma etc. The PHA should also describe the problems withthe urine sampling and describe the potential limitations of the urine data because the peoplesupplying the urine are involved in a major lawsuit against Diaz. These problems include thepossibility of individuals manipulating results by spiking their urine sample with some CFP fromthe release which had yet to be cleaned from the yards, providing a sample of urine from ananimal like their dog, or substituting another person's urine for their own.

Response 12. The methods used to analyze the self-reported health symptom data are describedin the "Health Outcome Data Evaluation" section of the PHA. Additional data tables areprovided in Appendix B, Tables 6a - 6d. We limited our analyses of health symptoms data totwo comparisons and we emphasized that strong conclusions could not be drawn for theseanalyses because of data limitations. In one comparison, we looked at the relative risk ofreporting symptoms between those persons with a greater potential for exposure to CFP andthose persons with less potential for exposure to CFP. In the other comparison, we looked at therelative risk of reporting symptoms between those persons with a detectable level of CFP inurine and those persons without a detectable level of CFP in urine.

On pages 21, 22 and 23 of the PHA, we discussed some of the limitations of the data. Specifically, on page 22, we addressed several concepts that are key to understanding these datalimitations. These include reporting bias, selection bias, and the differences in reporting stylesincluding differences such as whether people left out symptoms if they were sure they were notrelated to the release, while others reported all symptoms, no matter what they thought causedthem, even pre-existing conditions. The questionnaires did not suggest that participants shouldattribute their health symptoms to particular causes, but rather that they should report allsymptoms that occurred during the appropriate time frame.

People may have reported health symptoms related to pre-existing health conditions, smoking,flu, allergies or some other factor especially because some of these types of symptoms aresimilar to expected symptoms after exposure to a respiratory irritant such as CFP. If peoplemore concerned about the release were more likely to report these types of symptoms and werealso more likely to live in the impacted area or have a detectable level of CFP, this differencewould affect the analyses.

Other limitations identified by the commenter included the possibility of "spiking" (i.e. addingCFP) urine samples with remaining CFP found in the yards, the potential for provision of urinesamples from animals, and the potential for substituting another person's urine. These are issueswith chain-of-custody for the urine samples. NYSDOH staff refrained from witnessing urinesample collection to promote participation, facilitate ease of collection, and avoid invasion ofprivacy. It was important to us to get a large amount of participation because the greaterparticipation, the more we could learn about the potential for exposure.

Comment 13. Page 1 of the PHA does not mention that CFP is a chemical used in themanufacture of arthritis medication. The finding of CFP in urine may be a result of peopletaking arthritis medicines.

Response 13. NYSDOH does not know of any arthritis medicine produced or distributed in theUnited States of America that contains CFP. If parties know of such a drug, NYSDOH would beinterested in learning the name and manufacturer of these specific drugs and would incorporateany verified information into future interpretation of data. A request for information aboutarthritic medicines and CFP was made to Diaz. Diaz did not identify any specific medicines thatcontain CFP.

Comment 14. Additional testing of urine for continuing exposure measurements using the newvalidated approach is critical to understanding the issues regarding continued exposure in thiscommunity. Why would someone who tested negative in round 2 and 3 then test positive inround 4 if they are not living in their home? Why would someone test positive for CFP in urineand yet be told that home and soil was free of CFP? Will you be re-testing with the new method? Is it possible that people who tested negative before would have been positive with the newmethod? Does NYSDOH feel more people would have tested positive for CFP in their urine ifNYSDOH had sampled more people at the time of the first round? If no CFP was found at theschool, why did a child have CFP in their urine even though they live near the school? Have anyof the residents that moved out of Holley tested positive since they moved?

Response 14. To protect confidentiality we can not present information about urine results thatmay be used to identify individuals. For this reason we can not present information about anyspecific urine result. However, we can discuss the issues presented in the above comment in ageneral fashion.

Urine CFP is a sensitive means for detecting exposure to CFP. However, as with anymeasurement, there is variability from one measure to another, especially, at levels near thedetection limit. As an example, if you were to use a scale to weigh yourself three times, there is agood chance that at least one of the measures would be a pound or more different than the othertwo, if not all three being different from each other. The measures would most likely be just 1 to5 percent different. If you were now to use your scale to measure a one pound (2.2 kilograms)object it is possible that the scale will not show any change because one pound is near or belowthe detection limit of the scale. It is also possible that if the one pound object is placed threetimes on the scale, sometimes the scale will respond and sometimes it will not. Most of the urinesamples that contained CFP had levels of CFP near the detection limit and so one measure mayshow a positive and the next may not, similar to one measuring the pound weight on the scale forweighing yourself. Variability in making measurements is one potential reason why someonemay have tested negative in an earlier round and positive in a later round.

Some people had CFP detected in their urine even though no obvious potential exposure wasidentified during environmental sampling (e.g., air, soil) from their residence or nearbyresidences. Individuals move about in their daily lives and may have been exposed to CFP at alocation other than their residence while detectable levels of CFP were not found in theenvironmental samples at their residence. For instance, someone who has relocated couldpotentially have had undetectable levels of CFP in earlier rounds of urine sampling and thenhave a detectable level of CFP in a later round if they visited their homes or other areas whereCFP continues to exist.

It is possible that people who did not have detectable levels of CFP in prior analysis may have adetectable level of CFP using the new enzyme digestion method. Enzyme digestion separatesmore CFP from metabolites in urine than the method used in previous analyses. Consequently,the new method may provide for a more complete measurement of the total amount of CFP in aperson's urine than the old method. Samples reanalyzed with the enzyme digestion method mayshow higher levels of CFP because this method may include CFP that was part of the metabolitein the previous analysis.

We are offering reanalysis of previously collected samples with the new methodology. Thedetails of this reanalysis plan have been developed and sent to participants. Results of thereanalysis are expected by the end of 2003.

In response to the specific question about whether more people would have tested positive forCFP in their urine if NYSDOH had sampled more people at the time of the first round, if urinesampling included more people in January 2002, more people may have had detectable levels ofCFP.

Comment 15. Will more urine analyses be undertaken to monitor and evaluate the remediation process?

Response 15. NYSDOH has committed to undertaking urine sampling one week prior to andone week after a resident returns to a home vacated since the release. This offer is independentof whether remediation to the home has occurred.

Comment 16. We are very concerned that almost a year after the chemical spill, some citizensstill have detectable levels of CFP in their body. With this in mind, we recommend that thereport include a recommendation to explore with the community a long-term biomonitoringprogram to assess how persistent CFP is in the body.

Response 16. Based on review of the biomonitoring data collected to date, a long-termbiomonitoring program to assess persistence of CFP in the body is not warranted. From theurine data collected to date, CFP levels continue to decrease to non-detectable levels for almostall participants. CFP is most likely excreted quickly after intake into the body. Very fewindividuals continued to have detectable levels of CFP in their urine as of December 2002. Thelikely scenario for resident(s) with detectable levels in December 2002 was recent exposure toCFP, not persistence of CFP in the body. After we receive new data from the reanalysis of theurine samples and any potential additional sampling, we will reassess our conclusion that thedetectable levels in December 2002 were from recent exposure.

The information we gained from sampling, in conjunction with the toxicological evaluation inthe PHA, showed that the risks were minimal to low for possible noncancer effects and low tovery low for possible cancer effects (see Tables 5b, c, d).

Comment 17. The NYSDOH should explore other options to monitor CFP in resident's bodies,such as through blood samples. This is especially necessary because "the urine results suggestthat low level exposure to CFP may still be occurring."

Response 17. NYSDOH chose to do urine sampling instead of blood sampling for severalreasons. Urine is a good medium to biomonitor for CFP, it is easier to analyze than blood andcollection is non-invasive. Invasive techniques such as blood sampling often inhibit people fromparticipating. Invasive techniques also have health risks, albeit minor.

Comment 18. There are issues of the psychological impact of the disaster and the long-termrelocation that are ignored in the public health assessment. Does NYSDOH realize howfrightening the information about the CFP in urine is?

Response 18. NYSDOH and ATSDR have not developed an approach for incorporatingconsideration of psychological impacts in the PHA process. In response to this concern, we areadding a recommendation to the PHA and an item in the Public Health Action Plan to evaluatethe community's thoughts about the urine sampling program. We are aware of the importance ofthis issue and welcome suggestions on how to better characterize and address the psychologicaleffects of an event such as this.

A situation like that at Diaz can raise concerns in part because of a lack of understanding thehealth risks. The information we gained from doing the urine sampling helped us characterizethe risks. Hopefully, the results of this public health assessment presented in the PHA helpdiminish concerns.

Comment 19. No information is provided in the report with regard to participation rates for thebiomonitoring sampling program and symptom prevalence survey for critical subgroups of thepotential population at risk including those who resided in the exposure plume of the release,children and elderly subgroups of the population. It is unclear why the fourth round ofbiomonitoring excluded those people who were located outside the area of greater impact butwho were exposed when the accident occurred.

Response 19. Participation in the urine sampling program was targeted initially to people mostlikely to be exposed to CFP. In response to the findings and community interest, the samplingprogram was expanded in the fourth round. Participation was not targeted specifically tochildren or the elderly. Participation rates were not calculated as we were conducting anexposure investigation among those interested in participation, not among the entire population.As the entire population was not offered urine sampling in every round, NYSDOH can notcalculate the total number of people who had the opportunity to participate. Furthermore,NYSDOH can not calculate participation rates because individuals who worked, but did notreside, in the Village of Holley participated in some rounds of urine sampling, and no estimate ofthe total number of these workers exists. A participation rate is calculated by dividing thenumber of people who participate by the number of people who had the opportunity toparticipate.

The fourth round of biomonitoring did not exclude those located outside of the area ofgreater impact, but who were present when the accident occurred. This group participated in theurine sampling, but was only excluded in portions of the analysis, as many of these individualsroutinely returned to the area of greater impact but no longer resided there. Therefore, theexposure potential of this sub-group fell in between the two groups of exposure that weidentified: those who continuously lived in the area of greater impact and those who never livedin the area of greater impact. Because this group did not fit into either of the two exposuregroups, they were removed from the fourth round analysis.

Comment 20. Much of this risk assessment hinges on the analytic data used in estimatingexposure to CFP from urine levels. This is a new test developed by NYSDOH of unknownsensitivity and specificity. The investigators themselves note that these data are limited andmany assumptions were used in these calculations. They also state that there is considerableuncertainty associated with estimates of CFP exposure based on the CFP urine levels. It is likelythat there is considerable exposure misclassification, and the investigators readily acknowledgethis. The investigators do compare the symptom prevalence data with the exposure measuresbased on urine CFP levels and also individuals classified based on greater potential for exposure.In fact they do show that symptom prevalence reporting is higher in those with greater potentialfor exposure, whereas symptom reporting is the same or even higher in those with non-detectable levels of CFP in their urine.

Response 20. The comment is noted.

Comment 21. The demographics of the population at greatest risk is not provided. Theindividual exposure assessment relied on the biomonitoring program established by NYSDOHfor analysis of metabolites in urine and several rounds of testing were conducted. The symptomprevalence survey was also conducted only among those who participated in the urine sampling.It is not clear from the document whether those who participated in the urine sampling roundsand the symptom prevalence survey were representative of the population of Holley. Mostimportantly, it would be helpful for the investigators to specifically state the response rates forthose living in the exposure plume area, which is defined by NYSDOH as the area of greaterimpact.

Response 21. The demographics of the population with the greater exposure potential were notassessed. An estimate of the population demographics for the area of greater impact used in theanalysis has been attempted, using census blocks from the 2000 census. However after review,the area of greater potential exposure does not conform to these census boundaries and,therefore, any resulting population estimates would likely be inaccurate.

The symptom prevalence survey was conducted among a greater population than just thoseparticipating in urine sampling. At the time of the first round of sampling, 65 persons submittedan exposure survey, while only 35 urine samples were collected. We were not attempting to gaina population that was representative of the population as a whole in the Village of Holley in therounds of urine sampling or exposure surveys. As stated previously, we were conducting anexposure investigation among those interested in participation, not among the entire population. As the entire population was not offered urine sampling in every round, there is no denominatorcalculable for determining response rates.

Comment 22. The document does not indicate how many children were tested and how manytested positive?

Response 22. In order to preserve confidentiality, we are unable to release the number ofchildren who had detectable levels of CFP in urine. Unlike the other rounds of urine sampling,the fourth round of urine sampling had a large number of participants and therefore we are ableto make the following statements regarding children while preserving confidentiality. In thefourth round of urine sampling, fifty-six children younger than age 18 were sampled. Thepercentage of children with detectable levels of CFP was less than the percentage of adults withdetectable levels of CFP in this fourth round.

Comment 23. Why have you not performed a health study for the village?

Response 23. Since January 2002, NYSDOH has provided several opportunities for individualsto report health concerns. Opportunities for the public to document health problems includedprovision of exposure and health outcome surveys at the time of each round of urine sampling(January, March, April, May and December). These surveys were analyzed to the extentpossible and results are presented in the Public Health Assessment (PHA). Local physicianswere also contacted and encouraged to report any unusual signs and symptoms noted in patientsfrom Holley.

Additionally, in response to community concerns about health effects, the NYSDOH is offeringenrollment in the NYS Volatile Organic Compounds (VOC) Exposure Registry to Villageresidents. An exposure registry is a resource for research that may help us learn whetherexposures to chemicals, like CFP, are related to chronic health effects. Enrollment in theRegistry entails completion of a survey about potential exposure to CFP, the health status ofeach member in the household, and other factors related to health, such as smoking. Residentswill then be contacted on a periodic basis to update address information and monitor changes inhealth status. NYSDOH has contacted households, inviting residents to participate.

Comment 24. Why has it taken three months or more to get results from urine sampling?

Response 24. The duration between sample collection and result mailing ranged from 2 weeksto 2 ½ months (10 weeks) for the 5 different sampling rounds.

Round 1: 10 weeks
Round 2: 3 weeks - 5 weeks
Round 3: 2 weeks - 3 weeks
Round 4: 8 weeks
Round 5: 4 weeks

The methodology for CFP analysis in urine is complex, non-routine, requires specializedtraining, and uses sophisticated equipment. Sample preparation, instrument time and reading ofthe results from the instruments are all time consuming processes. After the release, the NewYork State Department of Health's Wadsworth laboratory developed the methodology necessaryfor analysis. The time spent to develop this methodology accounts for some of the time betweensample collection and mailing of results for Round 1. The large quantity of samples to beprocessed (250) and the subsequent increase in the preparation of result letters increased the timeexpenditure in Round 4. For each of the sampling rounds, the urine samples were the highestpriority in the laboratory that performed the analysis.

Comment 25. I think the NYSDOH has made a mistake in the urine analysis. The premise ofthis is that NYSDOH reported that five out of the ten tested in December 2002 tested positive forCFP and is on record in an April 2002 letter as stating that the CFP is probably quicklyeliminated in the urine. NYSDOH stated about half of what is in the body is eliminated in oneday or less.

Response 25. In the round five summary document, NYSDOH stated that "fewer than six of theten urine samples showed detectable levels of CFP in this round." While this statement can beeasily misinterpreted, it does mean that one, two, three, four, or five samples in the December2002 sampling (Round Five) had detectable levels of CFP.

To preserve confidentiality, we do not identify the exact number of samples with detectablelevels when they are fewer than six. Anyone who had detectable levels of CFP in Round 5 hadobvious exposure pathways that may easily explain why they continue to have detectable levelsof CFP in urine. Urine sampling is believed to be a more sensitive indicator of exposure to CFPthan environmental sampling. CFP is probably quickly eliminated in the urine and that mostlikely anyone with detectable levels in December 2002 had been recently exposed.

Comment 26. Every time my furnace turns on my sister is sick.

Response 26. If CFP was deposited in your home, it is possible that upon heating it moves froma solid into the air as a gas. This is why you may smell it when you turn on a furnace, enter a carthat has been warming in the sun, or when you are close to other objects that have absorbed CFPand then are warmed. This could be a possible link between the CFP release and your sisterfeeling sick when the furnace turns on. Another potential concern is that your furnace is notoperating efficiently. If the furnace is not operating efficiently, it may produce carbon monoxideand other non-desirable gases. Carbon monoxide is a colorless odorless gas which can make yousick and could even lead to death. We recommend you consider having your furnace checked orserviced by a professional. Having your furnace system cleaned may alleviate the odors yousmell.

Comment 27. What will happen when residents sell their home? Will they still be tested andwill they be put on the VOC Exposure Registry?

Response 27. Any individual who was a resident of the Village of Holley for any length of timebetween January 5, 2002 and December 31, 2002 is eligible for the VOC Exposure Registry. Any resident who is currently eligible for participation in the VOC Exposure Registry andwishes to participate, but is moving from the area, is encouraged to notify the VOC ExposureRegistry of your new address so you can continue to participate and receive materials.

NYSDOH is committed to provide urine sampling for any relocated residents that move backinto a home vacated since the release. Urine samples will be collected within the week prior toand one week after moving into the house.

Comment 28. The cancer rate in Holley is high. There is an increased number of children withlearning disabilities.

Response 28. Cancer, unfortunately, is a common disease. One of every two men and one ofevery three women will develop cancer during his/her lifetime. The number of people withcancer is increasing in most communities because more people are living to older ages, whencancer is more common.

The development of cancer is usually a lengthy process. For many types of cancer, symptoms donot occur until 10 to 30 years after exposure to cancer-causing agents. Therefore, it is unlikelythat the chemical release of CFP on January 5, 2002 was the cause of any cancers seen today. Cancer data at the ZIP code level can be analyzed for a better understanding of cancer incidencein the area of Holley, NY. ZIP codes 14470 (Holley), 14452 (Fancher), and 14429 (Clarendon)were combined for analysis in the statewide Cancer Surveillance Improvement Initiative. TheseZIP codes reveals a 15 - 49% elevation from expected in male lung cancer (4 extra cases), a 50to 100% elevation from expected in female lung cancer (5 extra cases), a breast cancer levelwithin 15% of expected (0.3 extra cases), a level 15 - 50% below expected for female colon andrectum cancer (-3.9 cases), male colon and rectum cancer within 15% of expected (1.6 extracases), and prostate cancer within 15% of expected (-0.9 cases). This analysis of the rates of thevarious cancers includes three zip codes. Upon further review of the cancer maps available fromthe NYS DOH web page (click on the button called cancer mapping) under the heading cancermapping, one can see that Holley is near the edge of areas with elevated and not elevated rates ofcancer. Smoking is the most common cause of lung cancer.

Much more research is necessary before the causes of cancer are well understood. Currentknowledge, however, suggests that the leading preventable cause of cancer is cigarette smoking. Dietary practices such as excessive alcohol consumption and the eating of high fat foods are alsobelieved to be important. In fact, tobacco and diet may account for as many as two-thirds of allcancer deaths. Other avoidable risk factors include excessive exposure to sunlight, ionizingradiation, and various occupational exposures to cancer-causing agents.

It is important to realize that many cancers can be effectively treated if they are diagnosed at anearly stage. Screening for cancers of the breast, cervix, rectum, colon, and prostate, for example,helps to identify these diseases before the onset of symptoms and at a time when they are usuallythe most curable.

The VOC Exposure Registry is a tool for gathering specific information about cancer diagnosesfrom Holley residents. In addition, New York State law requires that health care providersreport all diagnoses of cancer to the New York State Cancer Registry. Both the VOC ExposureRegistry health status updates (every two to three years for at least ten years) and data from theCancer Registry can be used to evaluate cancers that may occur among residents of Holley in thefuture. Cancer Registry data are strictly confidential, as are data from the VOC ExposureRegistry. As part of research related to the VOC Registry project, staff can apply for approval tohave NYSDOH Cancer Registry researchers check to see if any of the Registry participants havebeen diagnosed with cancer, even if they have moved to different parts of New York State.

The VOC Exposure Registry is a tool for gathering information about a range of issues ofconcern, also including learning disabilities. The Registry questionnaire asks specifically aboutlearning disabilities. If enough people participate in the Registry, the information gathered bythese questionnaires may be helpful for assessing whether there are unusual patterns of healthproblems, including conditions such as learning disabilities, among Holley residents.

Comment 29. The PHA should note that the urine analysis is non-routine and that the PHA doesnot allow the reader to ascertain if standard method validation procedures are formed.

Response 29. It is true that the method for CFP in urine had to be developed and validatedbefore participant urine could be analyzed and reported. Diaz assisted DOH Wadsworth Centerby providing a reference standard of CFP (2-chloro-6-fluorophenol) that allowed analyticalconditions to be optimized and validation performed. The method description provides someinitial validation data, but was not appended to the PHA document. Description of the analyticalmethodology is provided in one section of Appendix F. However, if more information is needed,copies of the method including documentation of validation may be obtained by calling 1-800-458-1158, ext 2-7800.

Comment 30. The discussion of the urine sampling should include an assessment of thelikelihood of finding false positives. This is especially true for the new method, which islooking for metabolites.

Response 30. False positives may result from different reasons including but not limited tosample contamination, interfering compounds and analytical error. The urine test may show afalse positive level of CFP as a result of contamination of the sample with CFP some timebetween sample collection and sample analysis. Another chemical may behave similarly in theanalysis and therefore be counted as CFP when in fact it is a different compound. This is knownas interference. An analytical error such as reading the chromatogram incorrectly can potentiallyresult in a false positive. Every effort is made to minimize the potential for a false positive bycareful laboratory practice and ongoing quality control of the analytical procedure. Qualitycontrol measures include running samples known to have no CFP in them to confirm that CFPreadings are not found.

Although not a true false positive in terms of the analysis of samples, another consideration inevaluating the urine results is recognizing that a result of a urine test may be truly positive as aresult of a CFP exposure unrelated to the Diaz chemical release. If this is occurring, a positiveurine result may be mistakenly attributed to exposure of CFP from Diaz. One possible scenariofor CFP exposure is from medications or other ingested material with CFP present in them.

Comments About the Risks and Agency Actions

Comment 31. Because there is no existing scientific information regarding the toxicity of CFP,the NYSDOH should follow the precautionary principle. The 1998 Wingspread Statement onthe Precautionary Principle summarizes the principle this way: "When an activity raises threatsof harm to human health or the environment, precautionary measures should be taken even ifsome cause and effect relationships are not fully established scientifically." The NYSDOH has aduty to prevent harm, when there is credible evidence that harm is occurring or is likely to occur-even when the exact nature and full magnitude of the harm is not yet proven.

Response 31. NYSDOH and ATSDR are aware of the precautionary principle. The use of good science and the precautionary principle by the agencies can be seen when examining many programs, even if when those programs were initiated, the term precautionary principle may not have been invoked. In New York State, the setting of comprehensive standards for organic chemicals in drinking water is an example of using a scientific basis in a process that was consistent with the precautionary principle. NYSDOH's and ATSDR's response to the January 5, 2002 release has also been consistent with the precautionary principle. Our response has been resource extensive and our risk assessment is based on estimates that are consistent with the precautionary principle. The risk assessment for cancer is based on using 2, 4, 6-trichlorophenol as a surrogate even though other phenols that may be better surrogates for CFP, including 2, 4-dichlorophenol, were not found to be carcinogenic. Our exposure estimates are based on the highest urine level found and are based on a 30 year exposure over a 70-year life. These estimates are more likely to be high than low. The results of our analysis, although with recognized uncertainty, showed that the non-cancer health risks are minimal to low and that the cancer risks are very low to low. This information has been provided to the citizens of Holley, as well as the NYS DEC and US EPA.

Comment 32. Are our homes as safe as they were prior to the release on January 5, 2002?

Response 32. The PHA document shows that there is a minimal to low health risk for non-cancer health risks and a very low to low risk for cancer from exposure to CFP as a result of theJanuary 5, 2002 release. These risks were calculated using assumptions that would more likelyoverestimate than underestimate the risk. Additionally, with time, the CFP exposures in generalhave been decreasing and therefore the risks are decreasing. However, these risks were notpresent prior to the January 5, 2002 release.

Comment 33. Diaz should pay for the cleanup, not the taxpayer. Cleanup should be doneimmediately and in both on and off site locations. Why don't we have a written plan for EPA'sclean up? We need more government oversight. Close Diaz and buy the contaminated houses.We are very concerned that relocated residents may be forced to move back into their homes,without knowing if the homes are fully clean. Residents who wish to should be grantedpermanent relocation from their homes. Because of the possibility of ongoing exposure, andbecause we know little to nothing about the long-term health effects of CFP, a permanentrelocation is warranted. We also feel that the NYSDOH needs to ensure that relocation beoffered to families outside of the splash zone, who have tested positive for CFP either in theirhomes or in their urine.

Response 33. The PHA document provides a summary of the environmental conditions and associated health risks from the January 5, 2002 accidental release. Diaz has closed the Holley facility and the USEPA is currently doing additional evaluation including sampling to determine if the Diaz site should be placed on the National Priorities List (NPL) of hazardous waste sites.

The assessment is being used by the environmental agencies to help determine what additionalwork including cleanup is needed. USEPA is performing the relocation and remediation underthe Superfund program, which also requires USEPA to identify the party(ies) responsible for thecontamination, and to recover USEPA's cost for the cleanup and relocation. The USEPAcurrently has assumed the responsibility of paying for relocation costs and remediation. Forquestions specific to relocation and remediation, contact Mr. Dwayne Harrington of USEPA at732-906-6899.

Comment 34. What is the plan for people who return home and then have a detectable level of CFP in urine? What will be done if families move back and the urine re-testing shows CFP? Who will relocate them to a new home?

Response 34. The NYSDOH has offered to do the urine testing for people who move back intotheir vacated home. If any detects or significant increases in CFP urine levels occur NYSDOHwill work with the family and the other agencies to take appropriate action.

Comment 35. The Environment and Society Institute (ESI) was asked to review and commenton the public health assessment for the Diaz Chemical Air Release in the Village of Holley inOrleans County, New York. We commend the NYSDOH and ATSDR's response to this releaseby launching the comprehensive approach to assessment of this environmental disaster throughdevelopment of analytic techniques for environmental sampling and biomonitoring, detailedenvironmental data analysis, the pathway analysis assessment and the conduct of biomonitoringin combination with symptom prevalence surveys.

Response 35. The recognition of the amount of work and resources put into the NYSDOHresponse is appreciated.

Comment 36. Without really knowing the long-term health effects how can NYSDOH tellpeople to raise children on Jackson Street or anywhere they have found the chemical? Do we letchildren lay with their heads on the carpet in our home knowing that porous materials absorbedthe CFP?

Response 36. The risk assessment in the PHA provides a good estimate of the long-term risksbased on exposure estimates that were based on the highest and average urine results. Based onthese concentrations the risks were shown to be very low to low for cancer and minimal to lowfor non-cancer effects. These estimates do have uncertainty associated with them and these arepresented in the document. The environmental data and urine data have both shown a decreasein CFP with time from the release. This decrease is expected to continue and any exposureresulting from any remaining CFP would have even lower health risks than those estimated inthe document. We hope that we have provided the information about our public healthassessment in a manner that helps individuals make decisions.

Comment 37. Why can't NYSDOH make more definitive statements about health risks in thedocument?

Response 37. The PHA is careful to not over interpret the data. There is uncertainty in justabout any analysis and that is true for our health assessment. Many residents want us to provethat there will be no long-term health effect from the release. This is in essence trying to provethe negative, which is impossible. Our approach is to make the assessment using the best scienceavailable and to make reasonable assumptions. We identify those assumptions and anyassociated uncertainty in the PHA, which is why the PHA uses language like "estimates, may be,potential and possible". One reason that this information is provided is that some individualsmay not agree with our assumptions or not have the same level of comfort with our assessment. Individuals may choose to use the information along with our uncertainty assessment to maketheir own conclusion.

Comment 38. Why does this site not qualify as an imminent health threat as a result of theJanuary 5, 2002 release?

Response 38. Page 79 of the PHA provides the different categories used by ATSDR to classifythe health risks. The Urgent Public Health Category (imminent health threat) is not appropriatebecause there are no short-term health hazards that require intervention. The IndeterminatePublic Health Hazard was selected because the toxicology of CFP and the related compoundsreleased from Diaz are not thoroughly understood and characterized.

Comment 39. Why should we move back home with so many issues remaining (dioxins, nohealth study, urine testing)?

Response 39. Although uncertainty exists, much information is available from the extensiveamount of work. The estimates of health effects based on the highest urine levels found in oursampling are minimal for possible noncancer effects or low to very low for possible cancereffects (see Tables 5b, c, d). The characterization of health effects is based on the highest urinelevels and the more recent data show lower exposure and, therefore, lower risks.

Comment 40. Why are people not told that their home is near a hazardous waste site?

Response 40. The NYS DEC publishes on an annual basis a listing of the Inactive HazardousWaste Sites in New York. Information about the annual publication and the entire inactivehazardous waste site program can be found at the NYS DEC website: http://www.dec.state.ny.us. Once at the website, look in the subject file under H to select Hazardous Waste Sites, Inactive.

Comment 41. Diaz had another spill in August 2002, when do we get peace? What is beingdone to prevent future spills?

Response 41. Diaz is required to have a permit for the air and water discharges from itsoperation. Prior to Diaz stopping its operation in the spring of 2003, the January 5, 2002 releaseresulted in USEPA and DEC conducting reviews of Diaz's air and water discharge permit. TheUSEPA and DEC were reviewing Diaz's permits to see what amendments to the permits mighthave been necessary. USEPA also has started evaluating the Diaz Facility for possible additionto the National Priorities List (NPL) of hazardous waste sites. As part of this evaluation, in June2003 USEPA, began collecting additional environmental (e.g,. indoor air, soil, indoor dust)samples. The agencies are now working to see what are actions may be necessary to clean thesite of potential residues and to make sure the groundwater remediation system Diaz wasoperating is appropriately maintained.

Comments Primarily about Environmental Sampling and Site Conditions

Comment 42. The PHA should note the efforts made by Diaz to clean properties (e.g., washingsiding, yards, vegetation) and that Diaz did much of this work voluntarily.

Response 42. The PHA assessment document indicates in the second paragraph of the summary that Diaz cleaned properties and relocated individuals. The document on Page 5 and 34 alsodiscusses Diaz's clean-up efforts.

Comment 43. The discussion in the indoor air section of the PHA indicates that air tests in 11of 23 units tested positive for CFP. This discussion is inaccurate if not misleading because only4 were above the quantitation levels.

Response 43. The text of the PHA document has been amended to indicate that a signalbelieved to be CFP was detected in 11 of the 23 units tested.

Comment 44. The PHA should recognize that dioxins and furans could not have been formedas part of the process making CFP because high temperatures are needed, temperatures that werenot reached during the CFP production and at the time of the release.

Response 44. The dioxin results from the wash sample suggest that dioxins and furans wereformed as part of the release. The document will be amended to indicate that the toluene used towash the vessel could have contained dioxins and other chemicals.

It is not clear that high temperatures are required for the production of dioxins. The reactionwith ortho-halophenols to form dioxins is reported to proceed quite quickly (<5hrs) to give 1%or better yields at temperature of 145ºC under certain conditions. Since this mixture had beenmaintained in a steam-jacketed vessel for several months, some reactions with modest reactionrates may have formed products such as dioxins. (see USEPA Report EPA-600/2-80-197).

Comment 45. The residents have expressed concerns about dioxins as a component of theexposure. The analysis of potential dioxin exposure was based on dioxin analysis of one sampleof the storage vessel solution and an assumption that the ratio of dioxin TEQ (related to 2,3,7,8-TCDD) in the storage vessel represents the ratio emitted into the environment. It would be moreprudent to measure dioxin concentrations directly on soil samples in contaminated anduncontaminated areas to validate this assumption, to allay this concern. Why has NYSDOH nottested for dioxin and furans and other compounds? To date, there has been no soil or indoortesting for dioxins and furans following the explosion. Because these were associated with therelease, this is warranted.

Response 45. Based on our analysis of potential dioxin exposure, we did not do any additionaltesting for dioxins. However, we received numerous requests from residents for dioxin andfuran sampling from residents. Based on these requests, we have been working to analyzedioxins and furans in the seven soil samples that we collected in January 2002 and analyzed forCFP. These samples have been kept in controlled environments and should be good samples touse to determine the levels in the soil as of last January. We expect to complete this work in thefall of 2003. Also, in 2003, the US EPA as part of its assessment for considering the DiazChemical Facility as a potential NPL site collected soil samples for analysis of dioxins.

Comment 46. In addition to monitoring for CFP, we request that the NYSDOH explore optionsto monitor for other chemicals released by Diaz, such as ethylene dibromide, a chemical that wasbanned by the US EPA in 1984 for use as a soil and grain fumigant.

Response 46. The US EPA, as part of their assessment for considering the Diaz ChemicalFacility as a potential NPL site, is doing sampling to look at many other compounds includingethylene dibromide. ATSDR and NYSDOH are working with US EPA and the other agencies toaddress these concerns.

Comment 47. The agencies need to re-look at the definition of the splash zone. Cars parked theday after the spill at Village Square had residues. Will there be any additional testing in theVillage in areas other than those immediately adjacent to Diaz? People outside of the splash zonehave been ignored and consequently inadequate testing has been done in these areas.

Response 47. The USEPA as part of their assessment for determining if the site will be listed onthe NPL is collecting samples from within and outside the splash zone. The splash zone or "areaof greater impact" does not mean that exposures did not occur elsewhere. Impacts did occurelsewhere. The area referred to as the area of greater impact was defined for the purposes ofhelping us to make meaningful comparisons to ascertain any differences in the potential forexposure and any possible health effects. Environmental sampling and urine sampling has beenconcentrated in the area of greater impact but sampling has been done outside this area as well. Our evaluation showed that the highest levels of contamination occurred in the area of greatestimpact and that these levels were determined to represent a minimal to low health concern fornon-cancer health effects and very low to low for cancer health effects.

Comment 48. Are we sure the sample of CFP referenced on Page 5 was pure?

Response 48. To be clear, the sentence on Page 5 refers to two samples. The "pure" CFP wasused to help develop the analytical methodology. The other sample, the wash from the tank,contained toluene as the washing agent and the other chemicals listed in Table 2 and has beenused to help characterize what was released into the environment. We did not expect it to bepure and our analyses showed that it was not.

Comment 49. How do you come up with quantification limits? If we saw the release residueshould CFP be detectable?

Response 49. Quantification limits are based on statistical interpretations of the data to give thechemist some reasonable expectation of the quantity of material present. It is possible for amethod to detect a compound but at a concentration too low to be certain of its quantity. Byconvention, this as known as "detected but below the quantitation limit" and often at the lowlevels there is an increased uncertainty of the exact identity of the chemical released. Thesamples we tested that were collected based on visible contamination all contained CFP.

Comment 50. Does the NYSDOH believe that removing soft furnishings (beds, pillows, couchand carpet) and cleaning would remove CFP from the home?

Response 50. The data show that some porous materials in homes contain CFP at low (part perbillion) concentrations and removal of these items would remove CFP from the home. Althoughremoval of these items will remove CFP from the home it is unclear if their removal will makeany measurable difference in either air concentrations or personal exposure. Based on the lowconcentrations (part per billion) of CFP found in some of the sofa cushions or bed mattressessampled, the absolute number of CFP molecules may be small. Just as important, the CFP maynever leave these porous materials with normal use and not be a source of on-going exposure. Another consideration is that our risk estimate was based on the highest urine level found. Thehighest level of CFP in urine was found in the earliest (first) round of sampling. Since then, theurine levels have been lower and likely reflect decreases in exposure from all sources includingoff gassing of CFP from the porous materials.

Comment 51. The pool referenced on Page 9 of the PHA was not removed. Diaz cleaners brokethe pool.

Response 51. The text of the document has been amended to indicate that the pool no longerexists. The original statement was intended to indicate that the pool is no longer on-site.

Comment 52. How can CFP be cleaned off property when it does not wash off? Our house wascleaned twice and it is still on the siding, basement door and maple tree. Diaz needs to becleaned first, then the homes in the nearby neighborhood. We need soils and property on-siteand off-site cleaned. Washing has not removed all of the CFP. A washed chair came backpositive and it does not come off homes.

Response 52. The sampling has shown that CFP levels decrease with cleaning and that withtime levels are decreasing. Some CFP may remain after cleaning, but at lower concentrations. The US EPA is currently assessing remedial options for cleaning CFP.

Comment 53. The summary of the PHA should reflect that the release entered homes not justambient air.

Response 53. The fifth paragraph of the summary indicates that CFP contamination occurred inair, soil, surfaces and miscellaneous household articles. Based on this comment we haveamended the text from "air "to "ambient air and indoor air" to explicitly note that CFP enteredhomes.

Comment 54. Diaz has poured illegal quantities of phenols into our Holley's water treatmentplant.

Response 54. This information should be reported to the NYS DEC and to Town officials. Ifyou have specific information, you should contact the NYS DEC office in Avon at 585-226-2466 and/or report spills to the NYS DEC spill hot-line at 1-800-457-7362.

Comment 55. The DEC and EPA needs to spend more time monitoring procedures and testingto ensure we are safe before those homes are lived in. We need NYSDOH to do more testing; weneed NYSDOH to assure us it is safe!

Response 55. NYSDOH is committed to reanalyzing urine samples already collected and tocomplete urine sampling of any residents relocated since the release one week before and oneweek after they move back into their homes. NYSDOH is also committed to analyzing soilsamples collected in January 2002 for dioxins and furans. Additional environmental sampling isnot planned by NYSDOH. The USEPA in June 2003 did more sampling, which is not specific tothe January 5, 2002 release, for the purposes of evaluating whether the Diaz facility should beincluded on the NPL.

In response to community concerns about health effects, the NYSDOH is offering enrollment inthe NYS Volatile Organic Compounds (VOC) Exposure Registry to Village residents. Anexposure registry is a resource for research that may help us learn whether exposures tochemicals, like CFP, are related to chronic health effects. Enrollment in the Registry entailscompletion of a survey about potential exposure to CFP, the health status of each member in thehousehold, and other factors related to health, such as smoking. Residents will then be contactedon a periodic basis to update address information and monitor changes in health status. NYSDOH has contacted households, inviting residents to participate.

Comment 56. The analytical methodology used and further refined demonstrated a commitmentby NYSDOH to deal with the unknown chemical aspects of the target analytic CFP. However, asshown in Table 2 of Appendix B, the target 2 chloro-6-fluorophenol constitutes only 55 percentof the compounds released and does not deal with the toluene exposure (solvent). The productdimer and unresolved isomers of dichloroflurophenol constitute another 31 percent of themixture. These minority constituents are not at levels that can be ignored. A plan to improvechemical analysis for continued analysis must be undertaken. NYSDOH should develop morespecific analyses so as to monitor the continued exposure, environmental fate and health effectsof these other compounds.

Response 56. The initial purpose of the chemical testing in the environment was to determine ifCFP could be found in the environment and to provide information about potential exposures. The biomonitoring (urine sampling) was done to learn more about exposures that occurredbefore and after the release. Although not the primary purpose of our sampling, the data alsocould be used to help describe the areas of contamination and be used to evaluate the clean-upefforts. Validated analytical methods were rapidly developed to address these issues using CFPas the marker compound. CFP was selected because it was the most abundant unique singlecompound that could be directly linked to the Diaz release of January 5, 2002. The companywas able to provide a reference sample of the purified CFP allowing validation of the methodsused for environmental and biological specimens. Analytical results were also obtained for asample of the wash liquid from the reaction vessel that produced the release which gaveestimates of the relative amounts of other compounds that were present in the commercial batchof unpurified CFP. Methods have not been developed and validated at this stage to quantitateother compounds present in the reaction mixture. For many of the potential compounds, neitherour laboratory nor other laboratories have reference standards nor are reference standardsavailable commercially. It is possible that reference standards could be developed, but thedevelopment of reference standards would be a very extensive effort and would most likely takeyears. Without standards, accurate quantitation and positive identification is difficult, if notimpossible. Toluene was not a target compound for our sampling because our first samplingoccurred 9 days after the release on January 14, 2002. Toluene is a very volatile compound andgiven that the source was an air release, it was not expected to be found in either air or soil thatmany days after the release.

Comment 57. Dichlorobenzene was shown to have a positive interference in the EPA TAGAresult. Are other analyses subject to this interference?

Response 57. Analyses done using a mass spectrometer (MS) are subject to this potentialinterference. The Trace Atmospheric Gas Analyzer (TAGA) analysis used by USEPA for thereal-time analyses is an MS. Analyses done using an MS coupled to another instrument such asa gas chromatograph (GC) are not subject to the interference. Most of the environmentalsamples and all of the urine samples analyzed by NYS DOH were done using an MS/GC systemand are not subject to this interference.

The MS instrument measures the mass of individual molecules or fragments of molecules in theair and counts the number of times that mass occurs. A CFP molecule has a mass (molecularweight) of 146, which also is the molecular weight of dichlorobenzene. The mass spectrometerwill report the number of molecules with a mass of 146, which will include both the CFP anddichlorobenzene. This is how the interference occurred in the TAGA analysis.

When the MS is coupled to another instrument called a gas chromatograph this problem isaddressed. A gas chromatograph contains a column through which the sample passes. Differentchemicals in the sample will travel through the column at different speeds because of theirdifferent chemical properties (e.g., size, shape). So, even though dichlorobenzene and CFP havethe same mass, because of their different chemical properties they will travel through a GCcolumn at different speeds and therefore be separated. Once separated, the mass spectrometercan then be used to count the number of molecules or molecule fragments at the different timesto measure the amount of CFP and dichlorobenzene. The down side of coupling the GC to theMS is that some of the chemicals are lost in the column and so the detection limits may not be as good as doing the analysis solely with a mass spectrometer.



Table of Contents

  
 
USA.gov: The U.S. Government's Official Web PortalDepartment of Health and Human Services
Agency for Toxic Substances and Disease Registry, 4770 Buford Hwy NE, Atlanta, GA 30341
Contact CDC: 800-232-4636 / TTY: 888-232-6348

A-Z Index

  1. A
  2. B
  3. C
  4. D
  5. E
  6. F
  7. G
  8. H
  9. I
  10. J
  11. K
  12. L
  13. M
  14. N
  15. O
  16. P
  17. Q
  18. R
  19. S
  20. T
  21. U
  22. V
  23. W
  24. X
  25. Y
  26. Z
  27. #