Document jQKwB5nEkoG9N3QrY1meMOQO

/; l cc. Pr`. -l.'a \:r \ j 20 OooUotJtfo. :!~090_,, _ ... 'i\idr.i-!.ce.l b:,f.~ Center .m-o. s-6212 U . . Department or Labor tii:. :d an.:'. Constitution Avenue, LC. v;. D'.iOhingtori, L). C. 20'>10 Re; Proposed 29 CFR Part 1930 Lear Sir: Air Products and Chemicals, Inc. snorriis herewith in quadruplicate its written comments on Proposed 29 CFR Part 1930. We reserve the right to subnit supplemental written cona`?,n.l:s during the post-hearing comment period. If you have any questions about tha enclosed, or re quire further information, please contact: John T. Barr Air Products and Chemical?;, Inc. P. 0- Do:-; 530 Allentorr., Peonsylvaa tv. j 0 ^05 (215) 393--J363 \t.rv 'cur. l', your:: .. AID, r.C'YJ ' ,i ' i l '' v.': 1 './x'Jif> : r-.;jb : c !/.');JM'u': c*o! ve V i rev itc.`! t AP00048473 WRITTEN COMMENTS OF AIR PRODUCTS AND CHEMICALS, INC. ON THE OSHA PROPOSED RULE ON IDENTIFICATION, CLASSIFICATION AND REGULATION OF TOXIC SUBSTANCES POSING A POTENTIAL OCCUPATIONAL CARCINOGENIC RISK AP00048474 COMMENTS ON SPECIFIC ISSUES RAISED BY OSHA - A. The discussion o.- the proposed rules states on page 54160, that "there appears to be general agreement on some of the following major issues, and OSHA relies upon these concepts in preparing these regulations"* We do not agree fully that there is general agreement with the way in which OSHA has stated these concepts, and provide the following comments, using the numbering system, of the discussion. 1. Mammalian Species We agree that v/e must rely on animal testing in the absence of human epidemiological data, and that mammals are more appropriate than non-mammals. We emphasize, however, that selection of the appropriate model and meticulous attention to experimental design and data analysis are essential. However, where adequate human data are available, they are to be preferred over animal tests. All properly conducted tests should have equal weight. There is no rational basis for any other AP00048475 position. Above all, negative human epidemiology results, from a study that conducted and evaluated ^ t properly, should imve greater weight than a positive animal study. For example, a cohort of several hundred or a thousand humans, with proper matched controls and resonably estimated exposures, means far more in predict ing the results of occupational exposure than does the result of a test of 50 or a hundred mice at exposures near the maximum tolerable dose. The AIHC proposal speaks specifically to the many detailed criteria which must be examined for an animal study. When conflicting animal results are obtained, the Classification Panel should weigh all relevant data and determine where the preponderance of the evidence lies. OSHA is incorrect in its reliance on DE3 as support for its position. Recent data have shown that females receiving high dosages of this material show increased incidence of breast cancer. Testing at High Dose OSHA has failed to consider a substantial body of recent metabolic data that shows, for some types of chemical carcinogens, a primary metabolic pathway that -2- AP00048476 I cgji detoxify foreign materials.- There are few materials tat are not seriously toxic at some level; the concent of i a maximum tolerable dose confirms this. Results ,obtained at extremely high dosages must be suspect, in and of themselves. Here, too, the AIHC has proposed general guidelines for acceptable dosages. When the latency period exceeds the expected life span of the animal, and when the incidence of tumors cannot be distinguished from the normal background, then a no-effect level can be established, and this result has been reported in many cases. The regulatory action to be taken in such cases must contain a consider ation of the incremental risk/banefit at the various possible levels of regulation; otherwise,resources are diverted without avail. 4. Organ Specificity Organ specificity of carcinogens may provide important insights into mechanisms of action. Every effort must be made to understand such mechanisms so any regulations which follow ace yrounded on ch`;r which are relevant to man and the reel anoo-uro jiv; :tsu. The important point is the mechanism of the carciiuvjeu, whether it is a direct carcinogen or must be mediateu -3- AP00048477 b a metabolic process, and the method of exposure, Thus, it is not correct to make a blanket statement regarding all materials as the same. This disregards the facts. 5. statistical Significance We agree completely with the statement that ail results must be evaluated for statistical significance. No other approach would be appropriate. We note, however, that the proposal would accept "only suggestive" (defined as "less than persuasive, or not statistically significant") data as adequate for classification in Category IX. This may be rebutted if the data are "less than suggestive". It is difficult to imagine what is meant by less than "not statistically signifi cant". Thus, the body of the model standard is not consistent with the stated philosophy of OSHA. There is no justification for making important regulations with wide impact on the basis of insignificant data, and the proposal to do so seems ludicrous. * The proposal then states that there are further issues on which agreement is not assumed- We concur, and offer comments on these issues also. -4-* AP00048478 Weight to be Placed on Benign Tumors Any evaluation of the significance of benign tumors requires that at least three points be con sidered by the Classification Panel: .1) is there likelihood that this tumor type undergoes malignant transformation?? 2) is there evidence that the agent may also induce primary malignant disease?; 3) what is the evidence that the benign tumor might occur in man and cause injury on a mechanical or structural basis? Therefore, the Classification Panel should consider the type, location, and history of benign tumors, and the species and strain of the test animals in its evaluation of the seriousness of risk for each material under consideration. The AIHC Alternative Proposal goes into more detail on this point in its discussion of the duties of the Classification Panel. Indiscriminate weighting of all tumors equally ignores the medical knovrledge in this area and puts an unduly pessimistic interpretation on the facts. Spontaneous Tumors We agree with the position taken by OSHA that only a significant increase in naturally-occurring tumors should be considered positive, and that judgment be -5- AP00048479 mediated by a consideration of the natural frequency of oticurroi.ee. V'e cio not agree chat the h i'dier the'spontaneous rate the more appropriate the strain , because this makes the statistical analysis less precise. The most appprcpriate strain is the one from which results are best extrapolated to man. This may or may not be the "most sensitive" strain. If "most sensitive" suggests a unique genetic or metabolic constitution, then the results may not be widely applicable. 8. Route of Exposure We agree with OSHA in its two-point guiding prin ciples on the relevant routes of exposure. The NIOSH list contains many substances which are there only because of tumors induced at the site of implantation in mice. These should be removed from the list and, such "evidence" ignored. 9. Confirmation of Positive Tests We agree with OSHA that confirmation of carcino genicity in humans in adc-crufoe regulation, and wo believe that confirm-1.1;.ion. in two spec Los of mammals, or one species by a replicated tost; is sufficient to require careful consideration of the need for regulation. --o-- AP00048480 Wgdo not agree that "only suggestive11 svider.ee is ` ef~lje~gtiati fer regulation, as discussed above. CertainlyI there should be no regulation where "the evidence is inadequate to raise any concern...". It is as possible, from a statistical viewpoint, to get a falsa positive result as a false negative result. On the other hand, there may be occasion where a single, well conducted experiment could be given considerable v/eight by the Panel. Therefore, we believe that no rigid rule can be developed for this subject, and that each report must be examined carefully for its relevance. However, it should be recognized that, in the long run, the question of a no-effect level loses it rele vancy, although it will continue to be debated until our knowledge of cancer mechanism is much advanced. What is relevant is a thorough understanding of the risk/benefit proportion in each case and the detailed review of evidence for both risk and benefit. When one cannot prove that a no-effect level does or does not ist, then it is incumbent, upon 03HA to estimate the sk at low levels of exposure and determine whether or not this risk, or probability of risk, is societaliy acceptable in the light of the societal benefits of production. -7- AP00048481 In addition to the foregoing numbered issues, the Proposal requests comments on several other points. ,Our comments are given below: 1 Page 54168 requests comments on the use of the principle of molecular structure to determine toxicity. We agree with the position of OSHA that this is not an adequate basis for regulation. Many examples exist in homologous series and isomeric compounds where one member differs widely from the other in its biological properties; and each case must be studied separately. As pointed out earlier, we do not agree with the overall categorization and regulative procedure proposed by OSIIA, and urge adoption of the AIHC alternative in its place. Page 54175 requests comments on the use of reporting systems, recognizing the duplication of effort between this Proposal and other regulations. We believe that it is a proper function of an inter-agency commission to develop a sin g^S.reporting system, available to ail relevant agencies, tha refill maximize the security of 'chb information submitic-, and minimize the burden of recordkeeping and reporting for both the government and industry. -8- AP00048482 VI*. Detailed Comments on the Proposed Model Standards uatailed comments made here ere in response to the request of OSKA, ar-.d as the result of study of the proposed rules. Failure to comment on a section does not imply approval of its promulgation in the context of the regulatory approach proposed by CSHA. A particular action may be appropriate for a particular application, but we repeat our statement that we do not believe the proposed rulemaking is proper, nor does it contribute to the solution of the problem at hand. Many of the difficulties apparent in the proposed rule arise from an attempt to prepare one blanket standard for all possible future regulations, and these problems emphasize the need for a case-by-case study at the proper time. 1900.0000 (b) Definitions: Authorized Persons Ko provision has been made for entry of several categories of parsons who must enter regulated areas, and are other then employees of the operator or representatives of the empl'^Bls* These include local safety and enforcement of florals, service poisonnil tor telephones, fire protection systems, elevators, weighing devices, and the like, contractors' employees, and representatives of various regulatory agencies. -9- AP00048483 The exposure of these individuals is infrequent and unlikely to be Severe. The enforcement of medical,* h-yga iene",1 an. d -ii* other provisions of the- proposed regulation is beyond the ' power of the operator, yet the presence of such persons is necessary. (c) Permissible Exposure Limits Paragraphs (c)(1) and (3) - No exposure is permitted for uses where substitutes exist. This is tantamount to a ban, even though the preamble states that there could be a completely closed system; in actuality, such a system does not exist. Does this ban carry back to a raw material for. which some uses have suitable substitutes and some do not? How is the exposure to be ratioed? Throughout this standard all references to the `'lowest practical exposure" should be changed to the "permissible limit". Paragraph (c)(2) - "The employer shall assure that no employee is exposed to eye or skin contact...". There always will be some material present in the atmosphere, even belo^ffijpe permissible exposure limit, and therefore, there v;ilinger eye and skin contact, at least with trace vapors. The only alternative is a totally impervious suit, even when the breathing concentration is within permissible limits. -10- AP00048484 11 No contact11 should be changed to ,rno contact with liquid. statement. Further, many, if not most carcinogens are site specific, and others require metabolic activation. Thus skin contact is not a universal hazard with all carcinogens. A case-by case determination is necessary. This proposal omits a.ny mention of an action level, below which employee exposure does not trigger medical examination, routine monitoring, compliance programs, or recordkeeping. Circumstances differ greatly as to probable employee exposure depending on whether the toxic material is used in its pure form or is present as a trace impurity in one component. Furthermore, the lowest possible exposure will differ for various stages of processing or manufacturing, and one set limit will not be applicable to all operations. As we discussed above, occasionally, there are persons who are required to enter for safety reasons that will experience very short exposures. Provisions should therefore be for an action level of ej^Sjsure, as has been done in past standards, below which exem^Sdn. is granted for most provisions of the s tc'-Pdcrd- The Secretary has continued this approach in the proposed acrylonitile (43FR2586) and benezene (43FR591S) standards. AP00048485 Secretary has omitted the supporting data for the statew^t on page 54174, column 2, that the ceiling must bn no mors than five times the permissible limit for a statis tically sound protection of the employee. This conclusion should be explained, and its basis made available for comment. Comment was requested in the Preamble (page 54174) on the proposed policy for setting exposure limits on toxic materials. There can be no justification for any procedure other than a case-by-case study of the appropriate and feasible exposure to that particular material at a particular finite concentration. This can be arrived at only after full and public review of the facts at a hearing. (e) Exposure Monitoring Paragraph e(l)(ii) - OSHA persists in requiring monitor ing without regard to respirators. The purpose of monitoring is to provide knowledge of the exposure to the employee, and health data relative to that exposure, therefore, data taken under other circumstances are irrelevant and useless for documenting exposure-health correlation. Monitoring should refl^^p^the exposure by inhalation for air-borne contaminants, except' where exposure through the skin is known to be more important, and should be with regard to the air actually breathed. -12- AP00048486 (^(6) - A major consideration in establishing feasible limiuS^gill be the. availability of adequate and accepted analytical methods. The rula should reflect the need for properly demonstrated procedures. (j) Protective Clothing Paragraph j(1) - Again, reference should be to contact with material other than in a vapor form unless entry into the body through the skin is the principal hazard. (1) Waste Disposal Paragraph 1 - Disposal of large equipment in sealed bags is an obvious impossibility. The proper procedure is to decontaminate (if necessary) before removal from the regulated area. The same alternative should be allowed for waste, down to a predetermined allowable concentration. (m) Hygiene Paragraph m(2)(ii) - The question of an employer assuring that employees, and especially female employees, shcwSoffter each shift is something that cannot and should not bfFaone. Provisions for shower facilities, and a require ment to change clothes is all that can ba dene. There, and elsewhere throughout the standard, the word "assure" should -13- AP00048487 be changed to "require'1. If OSIA wishes to create a legal obligation to shower, as stated on page 54177, column 3, the regulation should be directed tov.'ard the employee. Here, too, the problem of an all-inclusive regulation is apparent. OSHA has pointed out that even with a material of the low volatility of benzene, showers are not necessary (page 54177). (n) Medical Surveillance Paragraph n(5)(vi) - The only information on prior examinations that can be provided by the employer are from those given by that employer. Any other medical history can only be obtained by the employee. (p) Signs and Labels Paragraph - Here, too, there should be an exemption for trace concentrations of raw materials or solvents remaining in an otherwise nontoxic product. The limit should be set at those residues which will not give concentrations above the permissible level at the next point of use. -14- AP00048488 (q) Recordkeeping jggragraph (q)(3)(iii) - OSHA has effected a change in .t policy- regarding delivery of medical data directly to the ; employee. Previous rules have supplied this information through the employee's physician, and this should continue. It .is completely unclear as to what is meant on page 54179, column 3, of the preamble in regard to "(d) the environmental variables that could affect the measurement of employee exposure...". Mo such requirement is contained in the corresponding section of the proposed standard itself. This should be fully explained and made subject to comment. The preamble on page 54180, second column, also has a requirement for transmittal of records by registered mail to NIOSH, while no such requirement is present in the proposed standard. This statement in the preamble should be deleted, because even a few years records are too voluminous to be accepted by the Postal Service. The discussion of this paragraph is incorrectly labelled "g" in the preamble, not "q".(i) (i) Emergency WjF emergency is defined in paragraph (b) as any "unex pected exposure... in excess of the ceiling limit". Each emergency is to be reported within 24 hours, in compliance with (d)(2), and (i) requires a written plan and an alarm. -15- AP00048489 The use of the ceiling as the threshold for emergency situations is preposterous* Each and every time the ceiling is exceeded, whether or noi the employee is actually exposed, to that amount, or is protected by respiratory equipment or clothing, will require a written report. This will undoubtedly occur frequently during the first phases of the engineering compliance program, and OSHA will be deluged with meaningless reports. The purpose of this requirement is totally obscure. The ceiling will be so far below the acute effects level that there is no medical justification, and it is well above the permissible limit, so the employee has already withdrawn or donned protective equipment, and the actual exposure is not a factor. The numerical definition of an emergency in terras of ambient concentration has been found in the case of vinyl chloride to be unworkable, and this provision should be withdrawn. -16- AP00048490 be i >j - T. Barr J. Novak R. H Schenck FEB 17 1977 : 15 February 1978 H-Wt-L- A OSHA Docket Office Docket No. S-250 Room S6212 Department of Labor 3rd Street and Constitution Ave., N.w. Washington, D.C. 20210 Sir: In response to the request for comment in the 13 December 1977 Federal Register on proposed revisions to 29 CFR part 1910, Air Products and Chemicals, Inc., provides the following comments: 1* We support the intention of the Secretary to simplify and improve part 1910 by elimination of unnecessary requirements* In particular, we support the proposed deletions in 1910.1017, Vinyl Chloride, of the requirement for prepar ing a daily roster, (e)(2), and for maintaining it 30 years (m)(2)(ii). 2. We recommend that two further deletions be made In the Vinyl Chloride standard and in other comparable standards. These are: A. Delete the phrase "without regard to respirators" in 1910.1017(d)(2) and elsewhere. Monitoring air not breathed by the employee vitiates all benefit from the exposure and record-keeping program. No possible use can be made of records that do not reflect actual personnel exposure. AP00048491 OSHA Docket Office 2 15 February 1978 B. Delete the phrase "by registered mail" from 1910.1017(m)(2)(11). The U.S. Postal Service will not accept as registered mall a package large enough to hold even a few years' records. Thus, shipment must be by a parcel service. Very truly yours. Richard Fleming Executive Vice President AP00048492