Document Xz0KeEm19X20VLLdKJRO8gNbG

Title 29 - Labor CHAPTER WIN-OCCUPATIONAL SAFETY AND HEALTH administraijoa, department of labor PART IV10 - OCCUPATIONAL SAFETY AND'HEALTH STANDARDS Standard for Excosurc to Vinvl Chloride Pursuant to Sections 6(b), 6(c), and 8(c) of the Occupational Safety and Health Act of 1?0, (S4 5tatJ593, 1596, 1599; 29U.S.C. 655, 65?) Secretary of Labor's Order No. 12-71 (36 FR 875*0 and 29 CFR Part i?ll, ! 1910.93 of Part 191C ct Title 29, Code of Federal Regulations is hereby I amended in. the manner set forth below, in .order to provide an Occupa tional Safety and Health stand or c. ceding with the cry or ore of employees to vinyl chloride. I. B a. c.h ground . (1) Vinyl Chloride. Vinyl chloride (chloroethcno) , Chemical Abstracts Service Registry No. 75014, is a synthetic organic chemical made from ethylene Oi acetylene and chlorine by any of several processes. It is the parent compound of a series of thermoplastic resin polymers and copolymers which are widely used for containers, wrapping film, electrical insulation, pipe, conduit, and a variety of other industrial and consumer products. Vinyl chloride has been made commercially in this country since 1939, and present production is in excess of seven billion pounds per year. The vinyl chloride industry divides into three segments: monomer Droduction, ASI 00023824 polymer production, ny:d fabrication. Production of the monomer is a large-scale continuous process, involving only a few firms. There are comparatively few employees in this segment of the industry, because the processes lend themselves to automation. Vinyl chloride (VC) is used primarily in the production of poly vinyl chloride (PVC) , a resin which is produced through batch processing. The conversion of the VC monomer into a polymer or copolymer is an incomplete process, i.e., not all of the monomer is reacted. PVC is fabricated by a variety of techniques, including extrusion, injection molding and calendering, to form a finished product that needs no further chemical handling. The vast majority of employees involved in the VC industry are employed by fabrication firms. Such firms range in size from those with few employees and simple equip ment to large plants involving many employees and considerable capital, Vinyl chloride (VC), a gas at ambient temperature and pressure, is a chlorinated hydrocarbon, which heretofore has been regarded as having moderate liver toxicity. The initial standard, contained in Table G-l of 1910.93, established a ceiling value of 500 parts of VC per million parts of air. (2) The emergency temporary standard. On January 22, 1974, the Occupational Safety and Health Administration (OSHA) was I ASI 00023825 2 informed by the Nation.?.] Institute for Occupational Safety and Health (NIOSH) that the B. F. Goodrich Chemical Company had reported that deatbo of several of it.e employees from a rare liver cancer (angiosarcoma) may have been occupationally related. As a result of this notification, and after consultation with NIOSH, and a joint inspection of the B. I7 . Goodrich plant by OSHA, NIOSH and the Kentucky Department of Labor, a fact-finding hearing was announced on January 30, 1974, (30 FR 3S74) and held on February 15, 1974. Information obtained from this hearing, particularly the r preliminary reports of experiments conducted by Professor Cesare Maltoni of the Institute di Oncologia, Bologna, Italy, demonstrated that vinyl chloride induced angiosarcoma in rats at levels as low as 250 ppm, and in other species at higher levels. Experiments performed at lower levels of exposure were not completed at that time. Other testimony from medical witnesses and NIOSH, and the results of autopsies, led to the conclusion that the Goodrich workers had angiosarcoma of the liver and that VC probably was the causal agent in the angiosarcomas observed. In post hearing comments, additional angiosarcoma deaths were reported among workers who had been exposed to VC in plants operated by Union Carbide Corporation, Firestone Plastics Corpora tion and Goodyear Tire & Rubber Company. 3 ASI 00023826 On the basis o o1 fact that employees were being exposed at levels around the experi- mentally observed effect level of 250 ppm, an emergency temporary ceiling of 500 ppm to a 50 ppm ceiling, and established other requirements, including, for example, monitoring and respiratory protection. It was expre y recognized that this standard limiting exposures to a 50 ppm ceiling was a tentative, interim standard, and that the whole question of exposure to VC would bs considered more fully in the light of additional information, especially tire results of experiments which were known to be underway at that time. On April 15, 1974, information and data were presented to representatives of OSHA, NIOSH, and the Environmental Protection Agency by the Industrial Bio-Test Laboratories, Northbrook, Illinois, concerning results of animal exposure studies with VC. These studies were sponsored by the Manufacturing Chemists Association. Although only preliminary in nature at that time, these results revealed that 2 out of 200 mice exposed to VC concentrations of 50 ppm for 7 hours a day, five days a week, for approximately 7 months, had developed angiosarcoma of the liver. 4 AS1 00023827 (3) The pronosod staoba rri . Based on the demonstrated evidence of VC's carcinogenicity in three, animal species (rats, mice and ha mi t ers) , and the substantial probability that VC had been the causal agent in the cases of liver angiosarcoma found in workers both here and abroad, OSIIA proposed to revise 1910,93q and published a* comprehensive proposal (39 FR 16C96) on May 10, 1974, to protect employees from hazards of exposure to VC. The proposal called for limitation of employee exposure to VC to "no detectable level," as measured by a sampling and analytical method sensitive to 1 ppm, with an accuracy of 1 ppm 50%. The proposal also called for the establishment of regulated areas and limited access to such areas to authorized persons. A require ment for monitoring of employee exposures was proposed, along with engineering and work practice controls to be implemented when exposures over the detectable limit were measured. Respiratory protection would have been required while en gineering and work practice controls were being implemented or where exposures exceeded the permissible limit even after feasi- ' / blc engineering ccntrols were instituted. In addition., the proposed standard included requirements for medical surveillance, protective clothing, emergency procedures, training, specific protection during maintenance and decontamina tion op or alio iis , transportation loading and unloading operations and recordkeeping . -5- ASl 00023828 (4) Ih ri 1 : <' on ;:u t >ia"on'we ,u . Tin- proposal, as published un May 10, 107-1, allowed 30 days for interested parties to submit written comments and to request an informal rulemaking hearing. Informal contacts with OSI1A staff and early response;', indicated that the subject was of great interest and importance to many persons Because of Live limited fame available before expiration of the six month period provided in Section 6(c)(3) of the Act for promul gation of a final standard , it was decided to hold a hearing as scon a; as possible. Accordingly, on May 24, 1974, a no'lee of a hearing was published (39 FR 1S303) , setting a hearing date of June 25, 1974, The hearing was conducted from June 25 through June 28, and again from July 8, through July 11, before Administrative Law Judge Gordon J. Myatt. All participants were given the opportunity to present testimony and to cross-examine other witnesses. Persons participating in the hearing were given until August 23, 1974, to file additional posthesring comments, including various items of information which were requested during the examination of witnesses . (5) Leone-mic and technical impact study. During the hearing, OSHA determined that additional facts would be needed to determine the practicality of certain aspects of the proposed standard. Accordingly, OSHA contacted an independent consultant, Foster D. Snell Corporation, to conduct studies of the feasibility -6ASI 00023829 of co;:'.]iI various exposure levels, including those proposed by OSIIA d ethers ud\ ed by industry spokesmen . Snell was OSHA announced that the preliminary study was available and that comments were to be submitted no later than September 6, 1974 on or before September 25, 1974 (39 FR 33009). (6) Environmental Imnac.f Statements. A notice of intent to file an environmental impact statement assessing the impact of a proposed standard on occupational ex posure to VC was published in the Fedcral Register on April 24, 1974 (39 FR 14 522) . The notice invited any person having infor- motion or data on the environmental impact to submit it to OSIIA by Ktay IV, 197-:. On Juno 12, 1974, a draft environmental impact state ment was prepared and circulated to all interested persons. Ten copies were forwarded to the Council of Environmental Quality (CEQ) , which published a notice of its filing and availability in the Federal ASI 00023830 -7- R c ;y if ter on f 2;j, 19/5, (.'>9 rk 299'.9)- A 45 dr y period was allowed lew the Cubans; ion of comments on tiec draft statement. On Sc ptember 5, 199-1, the final environmental impact statement was prepared and a copy of it and all substantive comments were Sent to appropriate governmental agencies, private organications , and other interested persons. CliCl published a notice of availability for the final statement on September 6, 1999 (39 FR 32350). The submissio of comment was invited until September 25, 1994. The final statement and'all significant comments have bec-n. carefully considered : n arriving at the fin.all standard on occupational exposure to VC . (7) The Record . The record in this proceeding is one of tin most exhaustive ever relied upon by OSHA. It consists of pre- and post-hearing 7) Ai f r>e f MVpMr ' > * * S-- 1* * * I* f v U +^ . Li* t finding and rule-making hearings, the studies and inspections con ducted by OSI1A personnel, the environmental impact statements, the economic and technical impact studies, and all other relevant information. Ju all, over 600 written comments have been received, with more than 200 separate oral and written submissions made with regard to the two hearings. The record itself exceeds 4,000 pages. Employers, employees, labor unions, public health groups, inde pendent experts, physicians, research scientists, and specialists in many fields have been invited to submit information and have, made their views, knowledge and cxpcrr.se available to Oh HA. The AS I 00023831 -8- entire . nccmwsiit' these submissions was fhorougbb reviewed end evaluated in reaching the determinations set forth below II. bindings rcforci:r' carcinogenicity , exposure levels and feasibility . Cl) Csrciner enicitv of vinyl chloride. The carcinogenicity of vinyl chloride for three animal species (rat, mouse, hamster) lias been documented or. the record by the studies of Maltoni and Bio-Test Laboratories. Moreover, Maltoni's investigations have demonstrated a dose-dependent relationship for induction of tumors (i.e., more tumors occur at higher exposure levels) including angiosarcoma of the liver, in rats. The investigations of Industrial Bio-Test Labot atories have demonstrated a similar relationship for both, rat? and mice These investigators have induced angiosarcoma of the liver in rats and mice at exposure concentrations of 50 ppm, and in hamsters at higher concentrations of exposure. Additional tumors involving other organs, including the kidneys, lungs, and skin of exposed animals, were also observed in frequencies much in excess of control animals .__The incidence oi tumors in mice in the Industrial Bio-Test Laboratories investigations is particularly pertinent. Of 200 mice (100 males, 100 females) exposed to 50 ppm of vinyl chloride by inhalation, for eleven months, 100 died. Sixty-four animals died w.tiOtau. ui ... usiii pumui*uc .* *ocin c? porio,, rmcQ . Of the 36 remaining animals for which a gross postmortem pathologic ASI 00023832 -9- skin tumors, and one- with a hidr.ey tumor. c 9 (25u) with VC by inhalation--a route comparable to employee exposure. In the liver among employees exposed to VC, a particularly significant number in view of L 11 C2 t ,\u L rity of this cancer in ihe U.3. adult male population (testimony of Dr. Marcus Key, Director of NIOSH, at the rulemaking hearing). The findings of angiosarcoma of the liver in both experimental animals and exposed employees is compelling evidence that exposure of humans to vinyl chloride induces this tumor . Industry spokesmen, at the her.rirg. conceded that VC is carcinogenic for hurnans (ci.g. testimony ot Dr .-McBurney , Rulemaking hearing , 1041). Accordingly, it is concluded that VC must be regarded as a human carcinogen, and the probable causal agent of angiosarcoma of the liver, and that exDosurc of employees to VC must be controlled. -10- ASI 00023833 Acldilion.'tl evidence of tumor induction in a variety of other organs, including lung, kidney, brain and akin, as well as non-malignant alterations, such as fibrosis and connective tissue deterioration, indicate? additional oncogenic and toxicologic 'properties of vinyl chloride, which must tee considered in establishing control regulations. (See testimony and results of studies by Bio-Test Laboratories, Tabersbaw-Coopcr, Maltoni, MIOSH, and Eelikoff) . (2) Expo sun-c limits . Upon finding that exposure of employees to vinyl chloride may create a carcinogenic hazard, the amount,of exposure which is hazardous must be determined. The Surgeon General's Ad Hoc Committee referred to above concluded that safe exposure levels for carcinogenic substances cannot be scientifically determined. This position is supported by the testimony of HI OSH at the hearing, its recommendations for a standard of no detectable level, and by the testimony of expert witnesses from the National Cancer Institute. Several witnesses and persons who submitted comments have taken a contrary vbwj.r.d have suggested that man is less sensitive to biologic ' aberrations induced by vinyl chloride exposure than experimental animals. Proponents of this position have argued that if humans were as sensitive as rodents, an "epidemic" of cancer resulting from VC exposures should have already been discovered among employees. They also argue that the employees in whom tumors have been observed are ASI -11- 00o23834 those who have considerable employment experience as polymcriza- it is argued that the lower levels currently found in the workplace argument. The fact that approximately three -quarters of those employees with the longest exposure to VC (greater than 20 years since initial exposure) have not yet been located, makes it impossible to determine the actual number of affected employees. The eases of liver tumors observed to date have an average latency period, since induction of angiosarcoma is a dose-related phenomenon, and if employees engaged in cleaning reactors did, in fact, receive larger doses of vinyl chloride, it would be expected that such tumors would be observed earlier for this employee population. For this reason, the significance of presumed lower doses cannot be accurately assessed until a longer period of time has passed, as a longer inducrio.o period would be expected. Initiation, of exposure to chemical carcinogens and induction of cancer are not necessarily synchronous events. Because of the physiologic complexities involved with carcinogenesis, induction of tumors does not occur in all employees with similar exposure -12t AS I 00023835 histories. For example, Dr. Schneiderian of the National Cancer Institute emphasized during Ids testimony that only about a fifth of longer-term heavy smokers develop lung cancer. Accordingly, the industry contention that exposure levels have been dramatically reduced since the 1910's is not reliable evidence that current level; of exposure arc safe. Some industry spokesmen also suggested that the apparent nonrandom distribution of observed cancer in employees may indicate an exposure thr cshold for tumor induction, based on variations in ii the workplace d osign or practice and resultant`employee exposures (testimony and questioning by Tenncco Chemicals, Inc,). It has also been emphasised that in only 3 of S polymerization plants where employees have been exposed to VC for more than 20 years have any employees developed angiosarcoma of the liver. This argument is very similar to that raised concerning variability of past employee exposure. Although geographic and workpractice differences may ultimately be demonstrated to be factors in distribution of angio sarcoma, sufficient information is unavailable to exclude from consideration of risk those employees in workplaces for which cases of angiosarcoma have not been observed. It has also been suggested that the absence of cancer in a popu lation of 335 Dow Chemical Company polymerization employees monitored over a period of 7 years, indicates that exposure to vinyl -13- ASI 00023836 chloride at concentrations of less than 200 pur; is safe, (See study by iI Dr, Cook, submitted at the hearing by Dow Chemical Company.) However, the group surveyed did not include all workers who had been exposed, and the missing employees included many who had the longer term (over 20 years) exposures. Moreover, the statistically insignificant size of the sample population decreases the possibility that tumors would be observed. Dow also presented preliminary data in testimony at the hearing on the possible metabolic pathways of VC. The hypothesis presented was that VC may exert its carcinogenic effect by a metabolite, and that the metabolite is produced only when VC is metabolized by a secondary metabolic pathway operating only when enz3*mes regulating the primary pathway are saturated, as would be the result at higher exposures. The preliminary data indicated the possibility of an additional pathway for metabolism of VC in rats exposed to concentra tions of VC in excess of 220 ppm . However, the occurrence of angio sarcoma in both rats and mice at VC exposure concentrations of 50 ppm indicates that if a metabolite of VC is the ultimate carcinogen, then it must be generated at lower exposure concentrations in these species. Although this research may be helpful to the thorough understanding of the carcinogenicity of VX, it appears that it does not yet offer evidence which can assist in determination of safe exposure concentrations for employees, or even that such sate exposures exist. AS | 00023837 -14- a. A number of wi tnesi.es rcproseniin g employer.*; hive stressed Hint there is no evidence of cancer, either in employees or experimental animals-., at exposure concentrations of VC less than 50 ppm. (See e.g. testimony of F ires ton a, Tenncco Chemicals.) The conclusion of these witnesses was that r;c decision con be made concerning risk of exposure: to VC at concentrations less than 50 ppm. On the oilier hand, the testimony of most expert witnesses, including some industry biomedical experts, stated that quantification of a safe exposure concentration is not possible with the present state of scientific knowledge. (See e.g. testimony of Selik'off, Firestone, .NCI, and NIOSH.) In our view, the demonstration of cancer induction in humans *5U+ T> r*\ f'oc-1i.- *-- r pi a v 1 o'vr/C- 'l ^ r* i i w W f-w* i ie-- *11! In *tL u^ i W -> a I1- h11 Ait LL a substance represents a cancer hazard for humans at that level. It would be imprudent to assume man to be less sensitive to VC exposure than experimental animals in the absence of conclusive evidence. If would also be unfounded to assume that animals will not develop tumors when exposed at concentrations of VC of less than 50 ppm. Should a sufficiently large number of experimental animals ^ be exposed to VC at concentrations of less than 50 ppm, Schnciderrnan t-- ------------------------ -------- --------- ------------------------ -- -------- -----~ " '~ said that it would be expected that some would develop VC induced tumors. -15- ASl 00023838 (3) Feasibility . There is virtually no dispute that most, if not all, fabricate:':; ere currently capable of reaching exposure level: of! ppm. through c-nginecring control:;, These employers employ well over 9 5c, of all employees exposed to VC . Indeed, several fabricators are already operating vt this level (see SPI testi mony) . However, industry spokesmen have universally claimed that it is infeasible, lor the VC and the PVC industries to remain below 1 ppm consistently, using engineering controls. In addition, the Snell study on technical feasibility concluded that a 1 ppm ceiling is not feasible for the VC and PVC industries with present technology, but that the VC industry could currently attain lower exposure <____________ ________ ___________________ " ------------- ------- --- levels than the PVC industry. Labor union spokesmen and the Health Research Croup, Inc., however, have sucocsted that such a level is attainable. Since there is no actual evidence that any of the VC or PVC manufacturers have already attained a 1 ppm level or in fact instituted all available engineering and work practice controls/ any estimate as to the lowest feasible level attainable must necessarily involve subjects judgment. Likewise, thepprojactions of industry, labor, and others concerning feasibility are essentially conjectural. Indeed, as Firestone has suggested, it is not possible to accurately predict the degree of improvement to be obtained from engineering changes until A^^^* T-< /H.U -S . i,, _ 11 vy. 1j./ x i -16- AS 1 00023839 Wc hl'I'c-c that li.e.PVC arid VC establishments will not be able to attain a 1 ppm TWA level tor all job classifications in the near future. V'e do believe, however, that they will, in time, be able to attain . i -------------------- ---------------------___ levels of 1 ppm TWA for most job classifications most of th-.: time. It is apparent that reach ing such levels may require some new technology and work practices. It may also be necessary to utilize technology presently used in other industries, In any event, the VC and PVC industries have already made great strides in reducing exposure levels. (Sqe testimony of Dow Chemical Co, , '| TPv 973.) For example, B.F. doodrich testified (TR 1120) that it has reduced average exposure levels in several PVC plants from 35-40 ppm early this year to 12 13 ppm at the time of the hearing. Y/e are confident that industry trill continue to do so. (4) Conclusions . The conclusions below are based on a thorough review and evaluation of all thq evidence submitted. Where decisions can be based on record evidence, this has been done. Where, how ever, factual certainties are lacking or where the facts alone do not provide an answer, policy judgments have been made. Ihcre is little dispute that VC is carcinogenic to'man and we so conclude. However, the precise level of exposure which poses a hazard and the question of whether a "safe" exposure level exists, -J7- ASI 0023S40 cannot be definitively answered on the record. Nor ir it clear to what extent exposures cap] be feasibly reduced. We cannot wait i until indisputable answers toj thc-se questions are available, because lives of o-mployees are at stable. Therefore, we have had to exercise our best judgment on the basis of the. best available evidence. These judgments b.ave required a balancing process, in which the overriding consideration has been the prpte.ctlon of employees, even those who rr.a- have regular exposures to VC! throughout their wording lives. ! t j Eased on the available evidence and in view of'the above consider; tions, including feasibility, we believe (hat employee exposures to VC must be reduced to a 1 ppm tlnjc-wcightcd average: (TWA) . We also i believe that PVC and VC establishments will, in time, be able to attain that level through engineering controls, and that fabricators can do so in the immediate future. In addition to the TWA requirement, we have established a 5 ppm ceiling (averaged over a 15~m3nute period) in order to prevent exposure of employees to unacceptably high excursions. From an I operation standpoint, this ceiling level is realistic because minor excursions up to the ceiling level are likely to occur- on a regular basis. -18- AS 1 00023841 III. The Pink Sfr.nd.rd (1) Scene end air/lic alien. i Both the RTS end the; propose! would apply the standard to the entire VC industry, hiclpding numufacturers of VC and PVC and i fabricators, but exek'dif.g employers handling, or using fabricated f products made from VC. There is no dispute that a. standard is required for the 2r.ono2r.er and'polymcr industries.! However, the Society of Plastics Industry (SPI) ana various fabricators (see testimony of Goodyear, Central i Cable, etc.) rccoir.mcndjed that fabricators be excluded from the I> standard, or that a separate requirement be established for them < because many of them wjero already at or below the proposed I ceiling level. The record evidence* cstablish.es that at least some employees in the fabricating industry are exposed in excess of the permissible control limits (See testimony, TR 106; Robintech TR 642.) In these circumstances, v.t believe that it is imprudent to grant a blanket exemption for ajll fabricators. Therefore, the final standard is applicable to the fabrication industry, as well as the mor.omnr and polymer industries. Eipployers who, in fact, are substantially below j the exposure limit will |jc subjected to only minimal burdens by virtue cf the "action level" to be discussed below. -191 ASI 00023842 Where employers in the fabficating industry have exposures approaching the permissible lii nit, they will appropriately be subject to the standard. Employers ha idling or using fabricated products maade of PYC were not included in the ETS or the proposal and arc ;xeluded from the final stance: d. This conclusion is based on the absence of adequate evidence of exposure to VC in these opera tions. The final standard clarifies the exemption by defining a fabricated product as a product made wholly or partly from PVC which docs not require further processing at temperatures, and 1 tor times, sufficient to cause mass melting of the PVC. SPI and others (cf TR. 34-1) requested that PVC resins with less than 0,1% residual monomer be exempted from the regulation now, ar.d that the exemption level be reduced to 0.01% in three years. SPI suggested that the exemption of materials with less than 0.1% cf 14 carcinogens from 29 CFR 1910.93p (39 FR 3756) was an appropriate precedent. The cases arc not comparable, because no attempt had been made to set air concentration limits for the 14 carcinogens. The record did not include information that reliable monitoring and measuring techniques were available. Moreover, the exemption did not exempt airborne traces of carcinogens. The administrative cutoff was pro vided to avoid regulation of materials about which there was no health hazard information, and which would have broadly extended the application of the regulation beyond the record. Herein, no information was presented to show safe concentration -20- AS I 00023843 results from the use of re.`..ins with specific kn'cls. Indeed, the I / proposal to change the level late)-, when improved technology would permit such reduction, would seem to indicate that SPI has doubts about the safety of O.lf; residue level. Diamond Shamrock (Exhibit 142) testified that there is no direct relation. They indicate that the airborne concentration is more related to the physical form of the resin and the ventilation provided. Also, monitoring data from industry (ci.. Exhibits 1,31, 16S, 1/0) and -OSHA (Exhibit 151) indicate that levels in excess of 1 ppm may be found in fabrication operations. il In view of these facls and of the opportunity for employers to dis continue many duties upon a showing of no exposures above the action level, it does not appear that any residue exemption is either justified or necessary at this time. This course also agrees with a number of industry proposals (cf, TR 660). SPI (TR 345), among others, asked that compounded PVC pellets be exempted from the standard on the grounds that the pellets had too low a residue to cause harmful or measurable emissions. While it appears that PVC pellets would have a lower residue level than virgin PVC, the fact that the pellets must be heated to a molten mass at the same temperature as PVC, for further processing, indicates that a potential for release of the residue still exists. It appears that the exemption of fabricated products should be limited to just those items -21- AS I 00023844 II which will net undergo such mass heating. Further, the opportunity to demonstrate that exposures are. below the action level, and thus, discontinue many duties of the standard, provides a more positive control rind an adequate relief. ' , (2) Permissible exposure limit. The standard sots an exposure limit of 1 ppm averaged over any 8 hour period, and a ceiling of 5 ppm averaged over any period not exceeding 15 minutes. As more fully .discussed above, this limit is based on an evalua tion of the best available evidence and on a judgment that the health and safety of employees must be protected to the fullest extent feasible. In view of the fact that releases of VC in the VC and PVC manufactur ing processes arc variable, the 1 ppm ceiling level provided in the proposal would require maintenance of an average level significantly more difficult to attain through feasible engineering controls. There fore, the exposure limit prescribed in the proposal has been rejected. (3) Action level The final standard, unlike the ETS and the proposal, provides for an "action level" of 0.5 ppm TWA, one-half of the permissible expo- i i , - ---- sure limit. The purpose of the action level is to minimize the impact of the standard on the employers who have attained exposure levels well below the normissihle limit. Tims, where the vociiltc; nf moni toring under paragraphs (d)(1) or (d)(2) demonstrate that no AS1 00023845 -22- C7nployat' is in ex cosi Oi 0. a ppm h\v,''t, cmpiovcrs may, in effect, be exemptedfroin ::o;n: provisions of the standard. For example, fabricators who arc below the action level are not required to provide: medical surveillance or to monitor again, unless the employer lies reason to suspect that any employee is exposed in excess of the action, level, hi our judgment, exposures below the action level do not present a sufficient hazard to warrant applica tion of the entire standard to the many employers who are or will be below that level. (4) Monitoring The final standard, lihe the proposal, requires that individual employee exposure levels bo determined. This may be accomplished by person?.! or are? monitoring . Some witnesses and persons who sub mitted comments did not understand the meaning of the term "95 percent confidence level" in the proposal. Essentially it means that the employer is required to take a sufficient number of measurements so that the results obtained are statistically valid. IVe have modified the proposa.l to establish accuracy range require cents for various measurement levels. These ranges are narrow enough to ensure that a determina tion of compliance can be made, and broad enough to allow the application of a variety of technologies. All covered employers are required to conduct initial monitoring. -23- ASI 00023846 V.'he re monitoring and measuring re.si; arc: at or below the action level, no further monitoring is required unless the employer has .reason to suspect that any employee is exposed in excess of the action level, or unless changes have been made in production, process, control, type of resin, etc. Where the exposure level, without regard to respirators, exceeds the permissible levels, monitoring must be conducted at least monthly, Where exposures arc less than the permissible levels, but greater than the action level, monitoring must occur at least quarterly, (5) Methods of compliance. j The standard, like the proposal, requires that employers immed iately institute feasible engineering and work practice controls to reduce exposures to at or below the permissible exposure limit. Where feasible engineering and workpraetice controls will reduce exposures below the permissible levels, they must be instituted. Where such controls will not reduce exposures below the permissible level, they must nonetheless bo implemented to reduce exposures to the lowest practicable level, and be supplemented by the use of respirators, tc provide the necessary protection. Thereupon, a continuing program of engineering and work practice controls must be instituted to reduce exposures to the lowest practicable level. When exposures arc at or below the permissible exposure limits, the program may be discontinued. -24- AS I 0002384 In addition, a plan for achieving control by engineering and work practice methods must be drawn up and be made available, upon request, to representatives of OSHA and NI0E1I . We recognize that many employers covered by the standard can not currently achieve compliance with the permissible exposure limit solely by tire use of feasible engineering and work practice controls. The record also reflects broad generic distinctions between the compliance capabilities of the VC and PVC industries, Some industry spokesmen, including SPI (TR. 358-362), recom mended that a schedule of different permissible exposure limits .and compliance, dates be established for the VC, PVC and fabricating segments of the industry. This view assumes that the ability and the time required to feasibly reach increasingly lower control levels is similar within each industry, but differs markedly between industries. While the record does suggest that such differences do exist between industries, as noted above, It is clear that intra-industry differences also exist. Thus, the ability and time required by each employer to attain lower control levels may depend upon such factors as the-climate in which the plant is located, the age of equipment, the size of reactors, or the type cf resin manufactured or used. (Snell study, Firestone testimony, etc.) -25- AS 1 00023848 variations. (Sec e. g Dow, Firestone, Tcnnaco.) As noted above, the standard requires all employers to compliance is achieved. We have not established any deadlines for full compliance through engineering controls because we arc presently unable to determine when it will be feasible for most establishments to reduce exposure levels to the permissible level. We also believe that the requirement that each employer reduce airborne concentrations to the permissible level, or to the lowest and intra-industry technological differences which do exist, and will avoid the setting of separate industry standards on the basis of the general situation and conditions in each industry. (6) Regulated areas. The proposed standard would have required that regulated areas be established, that access be limited to authorized employees, and that daily rosters or summaries of those entering be kept for at least 20 years. In objection to these requirements, it was asserted that such control of access was not necessary from a health standpoint. Secondly, 1 J i A v f. -26- AS 1 00023849 by preventing r\cccss of needed employees or non-employees such as contractors, truck drivers, customers and consultants. The purpose cf establishing regulated areas in the proposal was to limit the risk of exposure to as few employees as possible. This concern is, still paramount, and thus the limited access feature remains. The final standard amends the proposal slightly to allow "authorized persons" to enter regulated areas. This change, it' is felt, will allow operations to continue without undue interference. The' final standard has also increased the length of time daily rosters must be maintained from 20 to 30 years. ! This' change was based largely on epidemiological considerations (See NIOSH testimony, tr. 119). (7) Respiratory protection. The final standard, like the proposal, requires the use of respirators where employee exposures exceed the permissible control level. Industry representatives made a number of objections to proposed requirements for respiratory protection. They stated that the "no detectable level" would effec tively i equine continuous wearing of respirators in PVC and VC plants, and that this is not feasible because respirators are cum bersome, present a safety hazard, and employees would not use them. We would agree that respirators have many drawbacks; the pro posal did not contemplate their s a final solution. The record shows 1^f f a fr~' tt' 'I i i 1^ vlir rv* "ii* r* d plant environmental levels can be reduced so that respirators arc "27- AS I 00023850 necessary only occasionally. However, we cannot agree that respira tory protection should not be required simply because it is incon venient, may require additional personnel, interferes with production, or may require extensive retraining of employees and restructuring of work practices. We have carefully considered all the objections, and . --- .... | , ... ........- -- have concluded that if the environmental level is not controlled to the permissible exposure limit, then employees must be afforded respiratory protection. While exposures in excess of the permissible level do constitute a hazard, we believe that it is necessary to mitigate some of the problems associated with implementing a program of respiratory protection while employees are being fitted and trained in respirator use, and while other adjustments which may be required are implemented. Therefore, until January 1, 1976, where exposures are not in excess of a 25 ppm ceiling, each employer must provide O' v\ V each employee with an appropriate respirator. However, employees \ 7> v* \< whose exposures do not exceed a 25 ppm ceiling, may decline to use the respirator, in which case the employer is not obligated to require its u~e. During this adjustment period, employees will be ---- 1 --V1 - '--* trained in the uses, purposes and limitations of respirators, and the hazards' of exposure to vinyl chloride. Moreover, each employee will be notified in writing if he has been exposed in excess of the permissible exposure limit. -28- AS1 00023851 `Where exposures .exceed a 25 ppm ceiling, respiratory protection ------ - YvwrU is mandatory in light of our judgment that t->--.arc associated will; such exposures. The provisions in the final standard regarding the selection and use of respiratory protective devices differ from those in the proposal. The descriptions of atmosphere-supplying respirators have been revised to indicate more clearly the types of devices in" tended, and the maximum permissible concentration level for each device. Moreover, the number of types of atmosphere-supplying .I devices has been increased. * Edwin C. At the hearing Mr./Hyatt, an OSHA consultant, made suggestions regarding the use of particular respiratory devices. We have concluded that his suggestions are meritorious. Therefore, the. provisions for selection of atmosphere-supplying devices follow closely the recommend ations contained in his testimony. Industrial spokesmen also agreed with Hyatt's suggestions. (See e.g. testimony of SPI and B. F. Goodrich) (TR R5 ft) We had originally omitted air-purifying respirators because none had 1 ecn approved by NIOSH for use against VC, principally because they lacked indicators to signal the expiration of the service life'of the sorbent. Hyatt and other witnesses discussed in detail the desirability of being able to use canisters or cartridge AS I 00023852 air-purifying respirators, provided a sorbent could be shown to effectively absorb vinyl chloride with an adequate service life. Recently. OSHA has received respiratory data from laboratories regarding the effectiveness of commercially available canisters and cartridges for vinyl chloride. These evaluations were conducted separately by KIOSK and by the B. F. Goodrich Company and sub mitted to OSHA in post-hearing comments. The results indicate that certain presently available canisters and cartridges effectively absorb vinyl chloride at relatively low concentrations. In discussions of these findings with NIOSH, it has indicated that it is willing to consider on an expedited basis the approval of air-purifying respirators for use against VC. Consequently, we have included three types of air-purifying respirators in the list of acceptable units, subject to the approval of such units by NIOSH. The maximum concentration for which each respirator may be used is based upon our evaluation of the data submitted by NIOSH and Goodrich. Because air-purifying respirators do not indicate sorbent exhaustion or bi"1 ah through of VC, and because VC has no inherent warning properties at levels for which these devices are used, strict adminis trative controls will be required for their use. Such controls include a program to assure timely replacement of canisters or cartridges and an alarm system to alert employees when vinyl chloride ASI 00023853 concentrations exceed the concentrations allowed for the particular type of respirator in use. (8) Haza rdous operations. This is a new section within the final standard. It encompasses essentially the proposal's requirernents for maintenance and decontamination but has restated them in terms of performance language to allow greater flexibility for employers to deal with such operations. The intent of the new section is to protect employees engaged in activities that present a risk of exposure to vinyl chloride in excess of the permissible levels. An example would be the cleaning of a filter where resin containing high residual monomer is trapped. The proposal's requirement for full-body, impervious clothing has been replaced by the direction to use impervious garments suited to the particular situation and probable extent of exposure. Thus, full-body clothing is not always necessary, and is therefore not required where less protection is adequate. Since vessel a entry falls within the definition of/hazardous operation, the vessel entry section of the proposal has been deleted from the final standard,. (9) Emergency situations . The definition of emergency has been recast in terms of an unexpected massive release. The main objection to the section on emergency situations in the proposal was -31- AS I 00023854 that, as the term was defined, many ordinary leaks or operations resulting in a small release of vinyl chloride would be considered emergencies. This was not the intent of the proposal. The final standard has been clarified to correct this ambiguity. It should be noted that the written operational plan required by the standard - . ___ _ --,--... 1 1 "" ' ' need not be developed for minor excursions above the permissible exposure limit, and that such excursions need not be reported. (10) Signs and labels . The thrust of the signs and labels section is to apprise employees of the cancer and fire hazards. No objections have been raised with respect to informing employees- of the fire hazard. However, a number of objections were raised at the hearing and in written submissions to the requirement that the word "cancer" appear on all signs and labels. The principal argument advanced against its use was that the term "cancer" or "cancer-suspect agent" scares employees and that instead, the message should contain instructions on how to deal with the substance (TR. 347) . We believe that a diluted form of warning will not suffice. We appreciate the concern of employers with the reaction of their employees. But we consider it imperative . that a worker be fully informed, and that he realize the possible risks involved in his occupation. Coupled with the training requirement in the standard, we believe that the signs and labels required will adequately inform employees of the hazard. In - 32 - AS I 00023855 addition, ouch signs will warn unauthorized personnel to keep out of regulated areas. The proper application of most protective measures requires an amount of training and indoctrination of employees that cannot easily be- conveyed on a sign or label. Also, the variety of measures that could be prescribed would result in an unwieldy or excessively detailed legend. Consequently, the required message on signs and labels will not include information on precautions, relevant symptoms, etc. The addition of suitable information by the employer would be permitted, providing it does not detract in any way from the required statement. The requirement in the proposal for labeling containers of vinyl chloride has been amended by deleting the reference to the possible hazard of violent polymerization. Very little information was developed on this hazard during the standard-setting procedure. It does appear that this hazard is essentially under control and that the fire and carcinogenic hazards at present are the most significant. Since labeling or placarding that is in compliance with the U.S. Department of Transportation regulations (49 CFR Part 173, Subpart H) already * warns of the fire hazard, only a statement concerning the carcino genic hazard need be added to the Department of Transportation labels. (11) Medical surveillance. The principal questions that have been raised regarding medical surveillance arc the necessity and efficacy of requiring certain specific scrum enzyme determinations (SMA-12 - 33 - AS1 00023856 I scries) and the application of medical examination requirements to the fabrication segments of the industry where employees arc exposed to lower levels of VC, The objection, has also been raised that the specification of tests and procedures interferes with tire application of advances in medical knowledge. A particular difficulty in considering medical surveillance is that the most commonly discussed lesion, angiosarcoma of the liver, currently cannot be diagnosed until the victim is terminal and, usually, within months of death. Precursor physiologic alterations, which might be reversible, have not yet been directly associated with the lesion. Consequently, there are no specific diagnostic tests which can be prescribed which will determine presence or absence of this tumor at an early stage of development. However, most medical witnesses indicated that the medical tests proposed are currently the only ones available which are useful for medical surveillance (TR 121, Exh. 95. TR 589'591)- Consequently, the specific blood tests proposed have beer, retained as a minimum requirement to assist the examining physician in determining fitness of potential employees for assignment to workplaces involving VC exposure. In addition, alternative medical examinations may be used where the examining physician determines that they are at least as good as those specified by the standard. -34- AS I 00023857 1 The Tabershaw-Cooper r.iudy and the various animal experiments suggest that VC may produce a wide spectrum of malignant and nonmalignant disorders. The general scope of the required medical exami nation has, therefore, been broadened to include kidneys, skin, connective tissue, snlcen, and pulmonary system, as well as the liver. No additional specific procedures or tests are required, but recommendations have been included in the Appendix to assist the examining physician. Because of the nonspecific nature of the required medical tesits , it is not appropriate to prescribe timing, or type of followup tests, or to mandate withdrawal from exposure based solely on results of the tests. Instead, the employer is required to obtain a statement from the examining physician of the employee's suitability for continued exposure, when the examining physician has completed such tests as he considers appropriate. The employer is required to withdraw an employee only when this statement indicates that the employee may be at added risk from continued VC exposure. As with monitoring, there; appears to be no basis for complete exemp tion of the fabrication industry from the requirement for`medical examina tion. The record docs show fabricating establishments-with concen trations of VC monitored considerably above the action level. In these instances, medical surveillance of affected employees will provide baseline data for future evaluation of their health, even if both -35- ASI 00023858 monitoring ond medical surveillance arc discontinued because improved controls reduce concentrations below the action level. Where exposures are below the action level, the medical surveillance requirements do not generally apply. (12) Trn ining. A separate provision for employee training has been added to the final standard rather than including it within the section on emergency situations as in the proposal. The new para graph provides for training of employees concerning the carcinogenic hazard of VC, emergency procedures, the need for monitoring and an annual review of the standard. It also provides for.training of employees concerning the purpose for, proper use of, and limitations connected with respiratory protection. (13) Records and Reports . The provisions for recordkeeping contained in the final standard require the preparation and maintenance of essentially the same information required by the proposal. The major change from the original proposal is the requirement for maintenance of monitoring records and daily roster sheets of authorized persons for 30 years, instead of 20 years. Additionally, the employer is required- to maintain medical records for the duration of an employee's employment plus 20 years, or 30 years, whichever is longer. The original proposal called for only 20 years. This change has been implemented because the latency period for induction of angiosarcoma ranges up to 30 years from initial exposure. -36- ASI 00023859 Therefore, as a minimum, medical records must be maintained for at least that long. It should be noted that spokesmen for both labor and industry recommended that this change be made. The reporting requirements are not significantly different from those in the original proposal. However, instead of the requirement for reporting incidents winch result in the release of VC into areas where employees may be exposed, the final standard clarifies our original intent by staling that only emergencies must be reported. Also the requirement for filing a detailed, written report within 15 days has been deleted. It has been concluded that submission, within 24 hours, of an initial report that includes facts immediately available, would ordinarily be sufficient. However, if the OSHA Area Director requests further information relevant to the emergency, the employer will be required to furnish such information. 04) D clctcd portions of the proposal. The proposal contained provisions requiring that shower facilities and change rooms be provided, and that storage or consumption of food be prohibited in regulated Aieas. We have deleted these % provisions because it is our conclusion they arc no longer necessary. Showering facilities are not required because protective clothing, where required by the final standard, should protect employees from skin absorption by direct contact with VC and because -37- ASI 00023860 there is no reliable evidence that VC vapor is absorbed through e skin. In addition, since we anticipate that most employees will not be wearing protective clothing and that employees who wear pm toctivr clci'hrv' will change, such clothing infrequently, we are not requiring that change rooms be provided. In addition, we feel that there is inadequate evidence showing that hazardous amounts of VC can be absorbed through ingestion. For this reason, the requirement prohibiting tine storage or consumption of food in regulated areas has been deleted. The proposal also contained provisions on maintenance and decontamination, transportation loading and unloading, and polymer 4andling operations. These requirements arc not mentioned in the final standard because attention to these items is implicit in the requirement that each employer reach the permissable exposure limit or attain the lowest feasible level. (15) Effect! ve date. In order to ensure that affected employers and employees will be informed of the- existence of these provisions and that employers affected rrr given an opportunity to familiarize themselves and their employees with the existence of the new requirements, the effective date of the amendment to 1910.93q will be January 1, 1975. To provide continued protection for employees until that date, the provisions currently contained in 1910.93q are icrcby promulgated, pursuant to section 6(b), 6(c) and 8(c) of -38- ASI 0002386 the Occupational Safety and Health Act, as an occupational safety and health, standard effective October 4, 1974. the amendment to 1910,93q set out below will supersede these provisions as of January 1, 197 5. Accordingly, upon consideration of the whole record of this proceeding, Part 1910 of Title 29 Code of Federal Regulations is amended, effective January 1, 1975, by revision 1910.93q to read as follows: 1 -39- AS I 00023862 t