MONDAY September 14, 1992 Part IV Environmental Protection Agency 40 CFR Parts 156 and 170 Receipt of and Notification of United States Department of Agriculture Comments on Agricultural Worker Protection Standards; Final Rule (This reprint was prepared from the electronic file that accompanied the original signed documents transmitted to the Office of the Federal Register. This file was certified to be a true copy of the original.) (This document appeared at 57 FR 42471-42480.) Federal Register / Vol. 57, No. 178 / Monday, September 14, 1992 / Rules and Regulations ENVIRONMENTAL PROTECTION AGENCY 40 CFR Parts 156 and 170 [OPP - 300164A; FRL - 4161 - 9] RIN 2070 - AA49 Final Rule; Receipt of and Notification of United States Department of Agriculture Comments on Agricultural Worker Protection Standards AGENCY: Environmental Protection Agency (EPA). ACTION: Final rule; receipt and notification of USDA comments. SUMMARY: The United States Department of Agriculture has requested that EPA publish its March 27, 1992 comments on EPA's June 1991 final regulations for the Worker Protection Standard for Agricultural Pesticides. Those final rules and associated notices for public comment were published in the Federal Register of August 21, 1992. ADDRESSES: The docket has been given the document control number OPP-300164A and is available for public inspection from 8:30 a.m. to 4 p.m., Monday through Friday, excluding legal holidays, at the Office of Pesticide Program's Document Control Office, Rm. 1132, CM #2, 1921 Jefferson Davis Hwy., Arlington, VA 22202. The docket contains the final regulations published in the Federal Register of August 21, 1992, and a copy of the draft final rules to which USDA responded and to which references are made in this document. A detailed EPA response document to USDA's comments is also contained in the docket. FOR FURTHER INFORMATION CONTACT: By mail: James J. Boland, Acting Chief, Occupational Safety Branch (H7506C), Field Operations Division, Office of Prevention, Pesticides and Toxic Substances, Environmental Protection Agency, 401 M St., SW., Washington, DC 20460. Office location and telephone number: Rm. 1114, CM #2, 1921 Jefferson Davis Hwy., Arlington, VA, (703) 305-7666. SUPPLEMENTARY INFORMATION: Electronic Availability: This document is available as an electronic file on The Federal Bulletin Board the day of publication in the Federal Register. EPA's Agricultural Worker Protection Standards, published as a Separate Part III in the Federal Register of August 21, 1992 (57 FR 38102), are currently available on The Federal Bulletin Board. By modem dial 202-512-1387 or call 202-512-1530 for disks or paper copies. These files are available in Postscript, Workperfect 5.1, and ASCII. EPA issued final regulations on the Worker Protection Standard for Agricultural Pesticides (40 CFR part 170 and subpart K of 40 CFR part 156) in the Federal Register of August 21, 1992 (57 FR 38102). A summary of USDA's comments on the June 1991 draft version and EPA's responses to them was presented in the August 21, 1992 final regulations (see especially the section entitled ``Statutory Review--U.S. Department of Agriculture''). The USDA has requested that EPA publish USDA's comments in their entirety. Therefore, EPA is issuing the substantive text of a letter from Daniel D. Haley, Administrator, Agricultural Marketing Service, USDA, dated March 27, 1992, in reply to EPA's June 1991 draft regulations on agricultural worker protection standards as follows: The United States Department of Agriculture (USDA) has reviewed the draft final rule on the Worker Protection Standards, 40 CFR parts 156 and 170, of the Environmental Protection Agency (EPA) and offers the following comments and suggestions. We commend EPA for its efforts to establish standards for the protection of agricultural workers and pesticide handlers from unreasonable risk of pesticide exposure. USDA supports EPA's Congressional mandate ``to protect the public health and the environment from unreasonable pesticide risks while permitting the use of necessary pest control technologies.'' However, we believe there are important issues in the draft final rule and its supporting documents which need to be addressed. These include: --Desirable alternatives are available for certain provisions which would protect workers, but reduce the compliance burden of agricultural employers. --There is insufficient information provided for adequate analysis of the costs and benefits of some provisions. --There is no indication that EPA has considered the costs of this rule to agricultural workers. --From a purely economic perspective based on the Regulatory Impact Analysis (RIA), dated March 31, 1991, the estimated pecuniary costs and anticipated pecuniary benefits of this rule do not appear to be supportable. This analysis relates only to the economic costs and benefits and does not take into account social and environmental costs and benefits. The comments are arranged in the following manner: I. Issues - A. Training--General B. Training--Cost Analysis C. Training--USDA's Cooperative Extension Service (CES) D. Agricultural Employer Responsibilities E. Thirty-Day Listing of Pestcides F. Emergency Assistance G. Decontamination H. Decontamination--Cost Aanalysis I. Reentry Restrictions J. Regional Climate-Based Reentry Intervals K. Personal Protective Equipment L. Cost/Benefit Analysis M. Cost/Benefit Analysis--Acute Poisoning N. Cost/Benefit Analysis--Chronic Illnesses O. Conclusion II. Appendix III. Conclusion USDA urges the EPA to address the concerns detailed below. If EPA is unable to modify these deficiencies, USDA requests that these comments, including the appendix, be included in the publication of the final rule in the Federal Register. I. Issues A. Training--General USDA supports the concept of providing training to workers who may be exposed to potentially dangerous pesticides. However, the manner in which the draft final rule requires such training is unreasonably burdensome. The draft final rule requires training by a qualified trainer, using specified training materials, for each worker who enters a treated area on an agricultural establishment for which a restricted entry interval is in effect or has expired within the past 30 days. EPA expects that such training will often be redundant. Most farms do not have any person who meets the trainer qualifications within their work complement and may find it necessary to hire outside trainers. In effect, the training provision will require that an employer provide training at the beginning of the season, and then each time an additional worker or replacement worker is hired. Generally, outside trainers must be hired each time the training program is provided. Given the extremely high variability and turnover within labor intensive agricultural work groups (1,000 percent is not uncommon), this procedure is unreasonably burdensome and would frequently result in virtually continuous training of small groups of new hires by each employer. USDA also questions whether a general requirement for training for a period of 30 days after the expiration of the restricted entry interval is reasonable. Herbert Nigg and his colleagues have noted, ``[d]elayed fieldworker reentry illnesses have clustered in the central valley of California. In particular, the longer (3 to 60 days) acute reentry incidents have been limited to California.''Nigg, H.N., Henry, J.A., and Stamper, J.H., Regional Behavior of Pesticide Residues in the United States, 85 Residue Review, page 257 (1983).EPA acknowledges that ``[t]he risks from pesticide residues should become negligible by the expiration of the restricted-entry interval.''Draft final rule, page 118.Thus, a reasonable alternative would be a requirement for training during a 30-day period in arid areas only, and then only where the restricted-entry interval is established by the generic reentry interval rather than by longer permanent, interim, or State-established reentry intervals where any risk has become ``negligible.'' Another more reasonable alternative would be to require employers to provide training to previously untrained workers during periods of high employment levels, but not more than twice per growing season. This would substantially reduce costs by training workers in larger groups and thereby reducing the required number of training sessions. Some replacement workers would miss the initial round of training, but the redundancy of training would result in virtually universal training in a short period. This delay for a relatively small number of workers is acceptable because these workers would be exposed to pesticide residues only, and then only after the expiration of restricted-entry intervals. EPA acknowledges that ``[t]he risks from pesticide residues should become negligible by the expiration of the restricted-entry interval.''Draft final rule, page 118.Unemployed workers, concerned that they might not be included in the employer-provided training and who desire training prior to employment, would not be denied the opportunity to be trained because they could take advantage of the ancillary training being developed by EPA. Further, the training program should include a mechanism to verify employee training. As part of this process, a card could be issued by an EPA approved organization(s) upon completion of the training. The card could be presented by the potential employee upon application for employment. USDA welcomes the opportunity to work with EPA to develop such a verification program. B. Training--Cost Analysis EPA estimates the cost of training to be zero, based upon an assumption that the Occupational Safety and Health Administration (OSHA) Hazard Communiction Standard (HCS) already imposes this requirmement.Revised Regulatory Impact Analysis, page 46-47. This assumption is not warranted because HCS does not impose training requirements related to pesticides. The U.S. Court of Appeals for the District of Columbia Circuit ruled that the promulgation of regulations by EPA precluded the Secretary of Labor from issuing regulations under the Occupational Safety and Health Act relating to farm workers' exposure to pesticides.Organized Migrants in Community Action, Inc., et al, v. Brennan, et al., 520 F.2d 1161 (D.C. Cir. 1975).Moreover, the draft final rule requires different trainer qualifications and training materials than those in the HCS. Thus, agricultural employers would experience significant additional incremental costs to meet EPA's training requirement. USDA believes EPA should recalculate its estimated incremental cost to take into account that the requirement for training does not duplicate OSHA requirements and to include the costs associated with EPA's trainer qualifications and training materials requirements. C. Training--USDA's Cooperative Extension Service If EPA anticipates that USDA, through its Cooperative Extension System (CES), will be able to meet some of the training requirements of the draft final rule, additional funding will be required. D. Agricultural Employer Responsibilities USDA agrees that the agricultural employer does have a responsibility to train workers on the proper use, maintenance, cleaning and storage of personal protective equipment (PPE). However, workers should be held responsible for their actions or lack of action after having received training and instructions from the employer. Sections 170.112(c)(6), 170.240(e)(1) and (2), and 170.240(f) give the responsibility solely to the agricultural employer. It has been argued that in industry, under OSHA regulations, the employer is ultimately responsible for workers. This is not an acceptable argument for farmers as they deal with a variety of workers, many of whom are never on their payrolls, but work for contractors. Agriculture in most cases is not confined within a controlled or monitored area like many other industries. Making agricultural employers responsible for employees' own safety actions is unrealistic. E. Thirty-Day Listing of Pesticides USDA does not support the requirement to display for 30 days after the restricted-entry interval has expired, in a central location, a listing of all pesticides which have been applied. The display period is inappropriate as many crops will be harvested before the display period has expired. The previously treated area may be in production of another crop. We believe that the display and maintenance of the display provides very little actual benefit to the worker or handler when compared with the cost to the agricultural employer. Approximately 89 percent of the agricultural producers have less than 10 employees. The case where small agricultural producers would be required to maintain such a display for one or two employees must be questioned. Although USDA recognizes EPA's concern that workers and handlers have information on possible exposure to pesticides, we believe a better method would be to make the information available upon request by a worker or handler. F. Emergency Assistance Section 170.160 of the draft final rule requires the agricultural employer to make available to the worker prompt transportation to an appropriate emergency facility. This provision needs clarification to indicate the extent of the agricultural employer's responsibility. USDA interprets this provision to be applicable only while the worker is at the workplace. G. Decontamination The decontamination provisions contained in 170.150 of the draft final rule are unreasonably burdensome to agricultural employers because they require potable water for decontamination purposes, including handwashing. The requirement that potable water be available for washing purposes is unnecessary.``Potable water'' means water that meets the standards for drinking established by State or local authority having jurisdiction or water that meets the National Interim Primary Drinking Water Standards in part 141 of this chapter. [Draft, page 216].Clean water should be sufficient, and is readily available from farm and irrigation wells, whereas potable water may not be. Changing the decontamination provisions from the use of potable water to permit the use of clean water would greatly reduce the burden and expense to agricultural employers without significantly reducing worker protection. An appropriate standard might be the regional or local standard for recreational water, i.e., water suitable for close bodily contact with a possibility for ingestion. EPA explained its requirement of potable water for decontamination as ``[EPA] believes that since OSHA uses the potable water standard for its Field Sanitation Standard, it would be easier for employers to comply with one water standard than with two.''Draft, page 107.This explanation ignores the fact that the employers who commented and opposed the ``potable'' water standard are in a better position to evaluate what may be easier for them than EPA. EPA also ignores the fact that, by its own estimate, 89 percent of agricutural employers have fewer than 10 workers and are thus exempt from the OSHA standard upon which EPA relies.OSHA has exempted agricultural establishments where 10 or fewer workers are employed in hand labor activities on a given day from its Field Sanitation Standard. 29 CFR 1928.110(a).EPA's explanation is clearly not responsive to this group of employers. EPA should provide a well reasoned explanation for the need for potable water, rather than clean water, if this requirement is implemented. USDA also questions the reasonableness of the general requirement for decontamination facilities for a period of 30 days after the expiration of the restricted-entry interval. USDA previously noted the studies of Herbert Nigg and his colleagues: ``delayed fieldworker reentry illnesses have clustered in the central valley of California. In particular, the longer (3 to 60 day) acute reentry incidents have been limited to California.''Nigg, N.H., Henry, J.A., and Stamper, H.H., Regional Behavior of Pesticide Residues in the United States, 85 Residue Review, page 257 (1983). EPA acknowledges that ``[t]he risks from pesticide residues should become negligible by the expiration of the restricted-entry interval.''Draft final rule, page 118. Thus, a reasonable alternative would be a requirement for decontamination facilities during the 30-day period following expiration of the restricted-entry interval in arid areas only, and then only where the restricted-entry interval is established by the generic reentry interval rather than by the longer permanent or interim reentry intervals which take pesticide degradation into account. USDA also believes that the term ``decontamination facilities'' does not describe appropriately the type of facilities that are to be made available after the expiration of the restricted-entry interval when the risk of pesticide exposure is negligible. A more appropriate term would be ``personal hygiene facilities'' or simply ``washing facilities.'' H. Decontamination--Cost Analysis EPA estimates the aggregate first year cost of its decontamination provision to be $79.5 million, but then discounts this cost by 50 percentRevised Regulatory Impact Analysis, page 46.on the grounds that ``[t]he OSHA field sanitation requirements cover an estimated 50.0 percent of all hired farmworkers.''This adjustment conflicts with the fact that the OSHA standards do not require eye-flush dispensers as does the draft final rule. Thus, the cost estimate should be increased to account for 100 percent of the cost of eye-flush dispensers for the 450,000EPA estimate of number of covered farms. 1987 Regulatory Impact Analysis, page 55.farms with hired workers. The draft final rule requires decontamination facilities with eye-flush dispensers for all hired farm workers. It is unclear whether EPA has attributed any cost to the provision of eye-flush dispensers to farm workers other than pesticide handlers. Table 21 of the Revised Regulatory Impact Analysis attributes the entire decontamination cost to change of clothing, soap, water, and towels. EPA's cost estimate details no accounting for the cost of the testing of potable water as required by the National Interim Primary Drinking Water Standards. EPA correctly notes that decontamination costs should be calculated, in part, on a per establishment basisDraft final rule, page 185.and that 89 percent of farms have less than 10 workers.1988 Regulatory Impact Analysis, page 54.Since 89 percent of farms are exempt from the OSHA field sanitation requirement, the aggregate cost of water containers, their maintenance, and the cost of testing should be discounted by 11 percent, rather than 50 percent. EPA calculated costs on the basis of the number of workers covered by the OSHA field sanitation standards and not on a per establishment basis. EPA should recalculate the decontamination provision cost, for other than soap and towels, on an establishment basis. The cost of eye-flush dispensers should be accounted for, and an appropriate estimate for the provision and testing of potable water should be attributed to the cost of this rule. I. Reentry Restrictions The prohibition of reentry for tasks requiring manual labor during the generic, permanent, and interim restricted-entry intervals is unreasonable for many crops. EPA contracted with Development Planning and Research Associates (DPRA) for an ``Analysis of Proposed Reentry Interval Regulations Under FIFRA,'' which determined that a number of crops would be adversely affected by an up to 48-hour restricted-entry interval. The problems identified in the DPRA study would be greatly exacerbated under the longer intervals and the total proscription of activities contained in the draft final rule. Provision, similar to the reentry provisions of the proposed rule, should be made to permit necessary agricultural activities if PPE is utilized. Appropriate PPE would be that required for the treatment which caused the reentry restriction. Such a reentry provision is particularly necessary where restricted-entry intervals exceed the up to 72-hour generic reentry interval. It would also be reasonable to apply a reduced standard for necessary activities which entail minimal contact with treated surfaces. USDA is unable to evaluate precisely the effect of the reentry prohibition, because EPA was unable to provide a complete list of the products for which permanent and interim restricted-entry intervals have been established. A partial list was provided by EPA at a recent meeting with USDA, but it was given to USDA with the admonition that publication deadlines precluded adequate time for its evaluation. USDA notes that a limited exception to the reentry restriction is provided for cut flowers and ferns, but we observe that there are a number of other crops which would also be significantly affected by the reentry restrictions. At a minimum, a provision similar to that for cut flowers and ferns should be made for the other crops which would be significantly affected. A number of these crops were identified in the EPA-commissioned ``Analysis of Proposed Reentry Interval Regulations under FIFRA'' prepared by DPRA. However, this study underestimated the impact of this rule because it excluded certain crops, excluded certain areas for some crops, and contemplated less restrictive restricted-entry intervals. 1. Reentry cost estimates.The generic reentry intervals required in the draft final rule represent a substantial increase over that proposed. Although EPA states that reentry costs are based on the level of disruption to agriculture,Draft final rule, page 44.the costs were not revised to reflect the new reentry restrictions. USDA does not understand why EPA cost estimates for three quite dissimilar options are identical (after adjustment for EPA's reduction of their greenhouse cost estimate). See Table 1 below. EPA estimated its reentry costs in the proposed rule based upon ``allowing routine entry for unlimited time to areas under a restricted-entry interval''Draft final rule, page 53.if PPE, decontamination and training were provided. The draft final rule states: ``[EPA] intends to prohibit all routine hand-labor tasks in areas treated with pesticides until the restricted-entry interval has expired.''Ibid.Despite this substantially greater restriction, the reentry cost estimates were unchanged. Table 1--Reentry Options and Costs Considered in the Farm Worker Protection Standard Development\1\ ÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ Proposed Low Option Proposed Rule Final Rule ÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ Tox I = 24 h Tox I = 48 h for organophosphates and carbamates Tox I = 72 h for organophosphates areas Tox I = 24 h Tox I = 24 h for others Tox I = 48 h for others Tox II = spray dry and dust settled Tox II = 24 h for organophosphates and carbamates Tox II = 24 h Tox II = spray dry and dust settled Tox II = spray dry and dust settled Tox II = 24 h Tox III = spray dry and dust settled Tox III = spray dry and dust settled Tox III = 12 h Tox IV = spray dry and dust settled Tox IV = spray dry and dust settled Tox IV = 12 h (Early reentry with PPE allowed) (Early entry with PPE allowed) (No early reentry, with narrow exemptions for limited time) Total = $33.7 million Less greenhouse cost\2\ - 27.5 million Cost = $6.2 million Adjusted cost = $6.2 million Cost = $6.2 million ÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ \1\ Sources: Final rule, p. 196; 1988 Regulatory impact analysis, pp. 24,27; Revised regulatory impact analysis, p.46. \2\ Greenhouse reentry restriction reduced in final rule. 00000.TBLPart 156 retains all previously established ``permanent'' and ``interim'' restricted-entry intervals that are longer than the generic reentry intervals to be established by this rule.Draft final rule, pages 41 and 42.This, in and of itself, represents no change from previous regulation. But the effect of this change, due to the prohibition of ``all routine hand-labor tasks'' during a restricted-entry interval, is substantial. Rather than EPA's explanation: ``the disruption to agriculture, and thus the cost, should be minimal; pesticides could be applied in the evening, and worker entry would be allowed the following morning''Draft final rule, page 44.; the prohibition and disruption could last many times longer. EPA provided a partial list of interim restricted-entry intervals from which USDA has identified interim reentry intervals as long as 30 days. In our judgment, the disruption to agriculture by prohibiting all routine hand labor tasks for as long as 30 days may be substantial. USDA is concerned that the access prohibition contained in this draft final rule will cause States to also prohibit access during restricted-entry intervals because FIFRA disallows State standards less stringent than those established by EPA. If States applied such prohibition to their present State-established restricted-entry intervals, access to fields would be prohibited by as much as 90 days. The draft final rule does not indicate how such standards should be applied. USDA believes it is imperative that EPA clarify in its explanation of the rule precisely the circumstances under which access should be prohibited and that it would be inadvisable to extend generally the access prohibition to longer State-established restricted-entry intervals because it could be tantamount to banning the use of such materials. EPA expects ``that most agricultural management practices can be carried out after the restricted-entry interval expires * * * .''Draft final rule, page 52.USDA believes the pertinent issue is not whether most practices could be delayed in this manner but rather whether certain essential practices could be. The effect of the reentry restrictions will vary depending upon the commodity and the growing conditions. The DPRA study noted a number of manual labor tasks which must be performed at shorter intervals than some of the restricted-entry intervals which would be established by the draft final rule. DPRA determined that, for crops with multiple harvests, ``[t]he interval between harvests ranged from one-half day (asparagus) to six weeks (avocadoes). Most crops fell into the three- to ten-day range, however.'' [Emphasis added]. USDA agrees with the DPRA study that there are many critical practices required in a number of crops which conflict with the up-to-30-day prohibitions to be established by this draft final rule. USDA can provide additional examples upon request. Few employers could maintain an idle labor force for 30-day periods. The prohibition of access could be expected to cause layoffs and a significant level of lost wages, agricultural production losses, and reduced employment due to the potential for the reduced competitiveness of U.S. agricultural producers. USDA believes these costs to farmers and farm workers should be taken into account by EPA. There is insufficient information furnished in the RIA to replicate EPA's reentry cost estimate. However, EPA estimated the difference between a 48/24/12 (Toxicity categories I/II/III) hours reentry provision and a 72/48/ 24 hours provision to be over $160 million [Table 1]. Since reentry cost is a function of the disruption to agriculture, the cost of prohibiting access by farm workers for as much as 30 days must be considerably more than that associated with the proposed rule, and this additional cost should be reflected in the draft final rule. EPA's RIA estimates the cost of prohibiting worker access to nurseries and greenhouses to be zero. Clearly there are costs associated with such prohibition. EPA previously estimated, in the RIA for the proposed rule, that the cost of an absolute prohibition of access to treated greenhouses under the proposed rule to be more than four times that of all farms combined. Under the draft final rule, the reentry restrictions would be reduced from that contained in the proposed rule, but access would still be prohibited to treated areas in greenhouses. While the new provision for greenhouses is less burdensome, it is not reasonable to assume the entire cost associated with this prohibition has been eliminated. The cost estimate for the proposed rule was based upon 1.8 million hired workers on 550,000 farms with hired workers, and apparently excluded family farms. The RIA prepared in conjunction with the proposed rule stated ``[p]esticide usage situations involving only a family farm setting without hired labor are not covered by the proposed regulation.'' It is unclear how the cost of the reentry requirements of the draft final rule was calculated. However, since the cost estimated for the draft final rule is identical (after adjustment for the change in greenhouse restrictions) to that of the proposed rule, USDA concludes the cost estimate for the draft final rule also excludes family members. Since the reentry restrictions apply also to family members on family farms, USDA believes this additional cost should be taken into account. USDA estimates there are 2.9 million operators and unpaid workers (primarily family members) who perform farm work on plant crops,Oliveira, Victor J. and Cox, E. Jane, The Agricultural Work Force of 1987: A Statistical Profile, pages 11 and 13 (May 1989).so the number of workers affected by the reentry provision would be 4.7 million rather than 1.8 million. The inclusion of farms with no hired farm workers also doubles the number of establishments affected by the reentry provisionFarms where only family labor is used are not to be covered by most provisions of the worker protection proposals. This exemption eliminates from the effects of this regulation about 550,000 out of about 1,000,000 farms which grow crops. [1987 Regulatory Impact Analysis, page 54].and substantially increases the affected acreage compared to farms with hired workers only. It is unclear whether EPA estimated the cost of its reentry provision on a per worker, per establishment, or per acre basis. However, it is clear that whatever the basis, there should be an accounting for the cost of including family members in the reentry provision. USDA believes that hand labor disruptions may be longer under the reentry levels imposed by the States. This area of regulation needs additional clarification by EPA to assess the effect of longer intervals beyond the EPA standard. As part of this effort, EPA does consider protective clothing for handlers, but fails to consider the benefits of PPE relative to agricultural workers. EPA should revise its reentry cost estimates in order to reflect the increased cost of prohibiting access to treated areas, the longer generic restricted-entry intervals, the permanent and interim reentry intervals that are longer than the generic reentry intervals, the inclusion of establishments with no hired workers, the cost of workers due to layoffs during the restricted-entry intervals, and the costs expected to be incurred by nurseries and greenhouses. J. Regional Climate-Based Reentry Intervals USDA believes that reentry intervals, and the need for decontamination provisions and training, should be based on the pesticide persistence expected in a particular region. EPA's explanation indicates the provisions of the draft final rule are based upon the experience in California. EPA analysts have noted that: Due to climatic conditions and an intensive pesticide use pattern, California has unique problems in farm worker safety.Zweig, Gunter, Adams, James, D., and Blondell, Jerome, U.S. Environmental Protection Agency, Residue Reviews, Vol. 75 (1980). World Resources Institute said: [S]cientific evidence suggests that pesticide degradation rates are dramatically affected by ambient moisture, humidity, temperature, and rainfall. In hot, dry areas like California, where crops may be grown only with extensive irrigation, residues remain on foilage for longer periods, unwashed by dew or rains. In other places, such as Florida and much of Texas, frequent precipitation reduces the danger to human beings.Field Duty: U.S. Farmworkers and Pesticide Safety, World Resources Institute, page 8 (1985). As previously noted, Herbert Nigg and his colleagues found: Delayed fieldworker reentry illnesses have clustered in the central valley of California. In particular, the longer (3 to 60 day) acute reentry incidents have been limited to California.Nigg, N.H., Henry, J.A., and Stamper, J.H., Regional Behavior of Pesticide Residues in the United States, 85 Residue Review, page 257 (1983). And, in the proposed rule, EPA itself said: The [reentry] problem exists throughout the country, although it appears to be greatest in California: * * * Researchers have clearly determined there are wide differences in the rate of pesticide degradation or disappearance which cannot be explained by time alone and that climatic factors should also be taken into account. Because of this, it is inappropriate to establish national reentry intervals based upon only time and experience in California. USDA believes EPA's mandate ``to protect the public health and the environment from unreasonable pesticide risks while permitting the use of necessary pest control technologies''Senate Appropriations Committee Report No. 102-107, July 11, 1991.could be better fulfilled by scaling reentry intervals to regional weather characteristics. California officials have recognized this need and established different districts within the State and provided exceptions or restrictions when certain amounts of rainfall have occurred or overhead irrigation has been used. This procedure does not appear to be overly burdensome to the California agency and it accommodates the requirements of agricultural production. In 1972, the Federal Working Group on Pest Management appointed a Task Group on Occupational Exposure to Pesticides which recommended: 1. The Federal Government should require pesticide registrants to develop and submit data sufficient to enable Federal officials to promulgate safe reentry intervals for each crop for which any new organophosphorus pesticide is to be registered. 2. Responsible Federal Agencies should pay due regard to significant geographical differences in the prevalence of the worker reentry problem and consider implementation of pesticide reentry intervals in those regions.Milby, T.H. (Chairman), Occupational Exposure to Pesticides. Report to the Federal Working Group on Pest Mangement from the Task Group on Occupational Exposure to Pesticides, page 51 (January 1974). USDA believes that reentry intervals, as well as the need for decontamination provisions and training, should be based on the pesticide persistence expected in a particular region. National standards based upon experience in California which EPA analysts describe as ``unique'' are an unreasonable burden upon farmers and farm workers in most other States. USDA is also concerned that such regulation beyond the harvest interval could be misinterpreted in a manner which would generate unwarranted food safety concerns. Considering that EPA has been gathering pesticide disappearance data for over 15 years in its registration process, and a body of research exists on the effects of climatic differences on pesticide persistence, it appears that regional climate-based reentry intervals are both feasible and appropriate. K. Personal Protective Equipment The cost estimate for PPE is based upon ``[t]otal handlers estimated at 527,300, including 170,000 commercial handlers.''Table 3, Cost Effectiveness Summary, 1991 Revised Regulatory Impact Analysis.This estimate does not include EPA's estimated 980,000 family-member handlers for whom PPE must also be provided.Draft final rule, page 248.The 1991 Revised RIA (page 7) states that ``[p]esticide usage situations involving a family farm setting without hired labor are exempt from the proposed (sic) regulation.'' Thus, the draft final rule and its RIA are contradictory, and there is no accounting for the number of unpaid workers for whom PPE is required. If the estimated cost for PPE were adjusted in proportion to the additional 980,000 family-member handlers, the total first year cost would be $315.5 million instead of $110.7 million; the first year incremental cost would be increased from $222.14 million to $63.09 million. L. Cost/Benefit Analysis The draft final rule states that ``the benefits in decreasing the number and severity of pesticide-related illnesses and injuries to agricultural employees far exceed the modest costs of the rule to agricultural employers, pesticide handler employers and registrants.''Draft final rule, page 24.The draft final rule also states that ``substantial'' benefits will accrue to employers as the result of this rule by reduction in lost time from the work force, reduced medical expenses, reduced insurance costs, and overall increased productivity from having a work force less affected by pesticide exposure.Draft final rule, page 184. The Cost Evaluation of EPA's RIA does not quantify any benefits to be realized, but merely states that the benefits from reduction of lost time, reduced medical costs, reduced insurance expenses, and increased productivity from having a workforce less affected by pesticide poisoning ``will substantially outweigh the costs of this proposed (sic) regulation.''Revised Regulatory Impact Analysis, page 9. However, the Cost Effectiveness Evaluation of the RIA prepared in conjunction with the proposed rule estimated the average cost per poisoning case, consisting of hospitalization and lost time, to be $580. If EPA's $580 estimate were correct, the number of illnesses which must be avoided to break even on a cost basis could be calculated by dividing $580 into the incremental cost estimate of $139,000,000. Thus, using EPA's estimates, over 239,000 cases of hospitalization must be avoided for the economic benefits of the draft final rule to outweigh its costs. Since 239,000 is twelve times EPA's highest estimate of medically diagnosed cases,Revised Regulatory Impact Analysis, page 24. it is not clear to USDA that substantial benefits will accrue to employers as a result of this rule. Furthermore, EPA should consider that only a fraction of medically diagnosed cases result in time lost from work, and only a small fraction of these require hospitalization. EPA should provide information supporting its assertion that the economic benefits of this rule will outweigh its costs, and include a description of the assumptions used in making its calculation. Otherwise, the public will be left with the conclusion that the assertion is not supportable. Farm workers may also realize additional costs associated with this rule due to layoffs and reduced employment which may result from any diminishment of the competitiveness of the U.S. agricultural industry. USDA believes EPA should take these costs into account in its cost effectiveness evaluation. M. Cost/Benefit Analysis--Acute Poisoning EPA measures the cost-effectiveness of the rule in terms of cost per poisoning case avoided: ``[a]ssuming that 80% of the current acute illness and injury incidents in agricutural employees caused by occupational exposures to pesticides are prevented through compliance with this new rule, each worker incident avoided in the first year will cost between $820 and $12,330, depending on the actual number of acute incidents.''Draft final rule, page 23; Revised Regulatory Impact Analysis, Appendix, Cost Effectiveness Summary, Table 7.The Revised RIA states: ``[e]stimates of annual acute poisoning cases from pesticides range from 20,000 to 300,000.''Revised Regulatory Impact Analysis, page 59.This range of ``cases'' is used by EPA to calculate the cost per incident avoided. If, as discussed below, the cost of the rule is understated or the number of cases is overstated, the range of $820--$12,330 cost per pesticide case avoided must also be understated. The RIA dated December 8, 1987, for the proposed rule estimated the number of medically treated agricultural pesticide poisonings ``ranges from 5,000 to 20,000 per year with 12,500 being the best estimate.'' The draft final rule states ``EPA estimates that 20,000 to 300,000 acute illnesses and injuries * * * occur annually to agricultural employees * * * .'' Although the RIA for both the proposed rule and the draft final rule cite the same source, no explanation is given for EPA's new estimates. The estimate of 300,000 ``acute poisoning cases'' appears to be an assumption that undiagnosed symptoms such as ``headaches, muscle aches, fatigue and others which may be mistaken for the common flu'' are equivalent to acute poisoning cases if the worker has been where pesticide residues may exist. To arrive at this estimate, data reported in California for 1979 are expanded to reflect a national labor force estimate of 4 million (including about 1.6 million livestock workers and others not covered by the draft final rule). This inflated national estimate is then multiplied by 100 to compensate for misdiagnosis, underreporting, and the possibility of minor symptoms for which no medical treatment was sought. USDA questions whether this estimation procedure is valid because other published data do not approximate this estimate. A detailed discussion of the extrapolation used by EPA is included in the Appendix. N. Cost/Benefit Analysis--Chronic Illnesses In the draft final rule EPA states: [T]he Agency is convinced that a substantial, but undetermined, number of additional incidents caused by delayed-onset illnesses will be prevented through compliance with this new rule. Any costs of this rule attributed to expected avoidance of delayed-onset illnesses in workers and handlers would reduce the costs per incident projected above for acute incidents. EPA regards these likely costs as modest.Draft final rule, page 24. USDA does not consider EPA's estimated cost per chronic case avoided to be modest. In its cost evaluation EPA said: For chronic health cases, with costs apportioned (50 percent), the incremental cost per case avoided could range from about $150 thousand to $150 million in the first year and about $100 thousand to $100 million in an out year, depending upon level of risk.Revised Regulatory Impact Analysis, Appendix, Cost Evaluation, page 3. If the cost per acute incident avoided is to be reduced by attributing a portion of the total cost of the rule to the avoidance of chronic illness, USDA believes the estimated cost to avoid chronic illnesses should be disclosed in the rule. O. Conclusion The above comments and suggestions are based upon the interim draft of the final rule transmitted on June 14, 1991, which bears the date of March 13, 1991. USDA understands that EPA intends to revise this draft prior to its publication in the Federal Register. We offer our assistance in revising the RIA for this rule as well as our assistance in clarifying the rule, where appropriate. Included in this assistance would be a review of the final draft of the final rule. II. Appendix Poisoning Incidence Based on information provided by the Environmental Protection Agency (EPA) and gathered by USDA, we do not believe that the EPA estimate of 300,000 pesticide poisoning incidents per year on farms is supportable. In the Regulatory Impact Analysis (RIA) for its draft final rule for Worker Protection Standards, EPA increased its estimate of pesticide poisoning incidents from a range of 5,000 to 20,000 per year with a ``best estimate'' of 12,500 to a range of 20,000 to 300,000. EPA makes no reasoned explanation for the change in their estimates and no new studies are cited. This revision is best illustrated by examination of the editing changes made in the text of the RIA: EPA's review of the available data conservatively estimates that the number of medically diagnosed [``treated'' deleted] agricultural pesticide poisonings ranges up [``from 5,000'' deleted] to 20,000 per year. [``with 12,500 being the best estimate.'' deleted] In addition to the number of poisonings for which medical treatment was sought, there are [``believed to be'' deleted] numerous instances where exposure to pesticides cause (sic) symptoms that are not brought to the attention of a physician for treatment. Symptoms of pesticide poisoning may include headaches, muscle aches, fatigue and others which may be mistaken for the common flu or other illnesses that are debilitating to some degree. [``One private research organization'' deleted] World Resources Institute (1985) estimates [``estimated'' deleted] that as many as 300,000 persons annually suffer some symptoms of pesticide exposure. This estimate, however, is not well documented. For evaluation purposes, therefore, this analysis will assume a range of 20,000 to 300,000 potential acute pesticide poisoning cases per year * * *.1987 RIA, page 22; Revised RIA, page 24. (Rule form; Additions underlined, deletions struck through). If new data exist indicating that an estimate of 20,000 medically diagnosed pesticide poisonings per year is appropriate, the source should be cited. EPA's estimate of 300,000 unreported instances of pesticide-related symptoms is not adequately supported. World Resources Institute (WRI), an environmentalist policy research organization, did not estimate that 300,000 persons suffered symptoms of pesticide exposure each year. Instead, WRI stated: [S]pecialists still disagree about the risks posed by pesticide use. Reviewing information from California, for example, epidemiologist Molly Coye has recently estimated that perhaps as many as 313,000 farmworkers in the U.S. may suffer the effects of pesticide-related illness each year, including such symptoms as dizziness, vomiting, `pin-point pupils,' and severe skin rashes. As Tables 2 and 3 suggest, however, regional differences in crop production and pesticide application rates almost certainly give rise to different patterns of pesticide exposure.Wasserstrom, Robert F. and Wiles, Richard, Field Duty: U.S. Farmworkers and Pesticide Safety, World Resources Institute, page 3 (July 1983).[Emphasis added.] Although WRI acknowledged that their description of the extreme estimate of pesticide poisoning was ``almost certainly'' inaccurate, EPA notes only that the estimate is ``not well documented.'' Furthermore, a review of the cited treatise indicates that Ms. Coye did not estimate 313,000 poisoning cases either. She merely described an inapposite 1976 study in California and made a hypothetical extrapolation; her own estimate of farmworker illnesses was lower by a factor of 24. In Ms. Coye's words: [T]he director of pesticide programs for the Department of Health Services estimated that as little as 1 percent of all pesticide-related illness in farmworkers [is] reported in California--despite the fact that state regulation requires physicians to report such cases to their county Health Officer within 24 hours of diagnosis (Kahn, 1976). In 1982, 235 cases of pesticide-related illness among farmworkers were reported; if this represents one percent of actual illness, the ``true'' prevalence would be 23,500 cases among the estimated 300,000 farmworkers in the state. If this rate is applied to a conservative estimate of the national farmworker labor force, or 4 million workers, the prevalence may be 313,300 cases; if we apply the rate only to hired seasonal farmworkers who are at greatest risk for field residue exposure, the prevalence would be 156,000.Coye, Molly Joel, Health Effects of Agricultural Production, presented at a symposium of the National Academy of Sciences, page 180 (1986). [Emphasis added.] Dr. Ephraim Kahn's 1976 study, cited by Ms. Coye, is obsolete and inapposite. The present California Health and Safety Code was enacted by the State Assembly in 1977 and amended in 1979 for the purpose of strengthening their system of reporting suspected pesticide injuries and illnesses. Under current law (and that which existed in 1982, the year of Ms. Coye's data), reports by physicians are cross checked against workers' compensation claims and a $250 fine may be imposed on physicians who fail to report pesticide incidents. Under this procedure, the California reporting system today functions much better than it did in 1976. Thus, the circumstances described by Dr. Kahn in 1976 are not comparable to either the circumstances at the time of Ms. Coye's hypothesis or of the present. In constructing her hypothesis, Ms. Coye may have drawn more from Dr. Kahn's article than he intended. He had said: In California we think the number of officially reported cases of residue-related illness is probably only a small fraction of the actual number, possibly no more than 1 or 2%. This, of course, is only a guess, but there is epidemiological evidence to back it up. [Emphasis added.] Dr. Kahn summarized that there was ``considerable indirect evidence'' of undetected and unreported symptoms, most of which were classified as ``possibly pesticide related,'' that farm workers are adversely affected by pesticide residues and that the true magnitude of the problem is uncertain. Thus, the basis for Ms. Coye's hypothesis is very weak at best. Indeed, had she elected to use the high rate (2%) of Dr. Kahn's guess rather than 1 percent, her extrapolation would have been reduced by half. In a personal communication with Dr. Kahn in September 1991, he stated that his article had been misconstrued and was inapplicable to pesticide handlers. Kahn's article referred to field workers only, and noted that a much higher proportion of handlers was reported. In its cost evaluation, EPA uses the same extrapolation for pesticide handlers as is used, based upon Dr. Kahn's 1976 guess, for farm workers. California's current pesticide-related illness reporting system is considered to be significantly better than the county Health Officer reporting system which was discussed by Dr. Kahn in 1976. As Robert I. Krieger, Chief/Supervising Toxicologist of the Worker Health and Safety Branch of the Department of Pesticide Regulation of the California Environmental Protection Agency, stated to USDA Economic Analysis Staff in a personal communication, ``[i]f you believe there is underreporting by doctors of workers' compensation claims, then you must also believe they don't care about being paid for their services.'' Dr. Krieger also stated that the extrapolation of Dr. Kahn's guess to estimate a national incidence rate of pesticide-related illnesses was ``wholly inappropriate.'' Although EPA believes pesticide-related farm worker illnesses are underreported, there is evidence to the contrary. One study found that field workers claimed pesticide symptoms 15 times more than a control group,Wasserstrom, Robert F. and Wiles, Richard, Field Duty: U.S. Farmworkers and Pesticide Safety, World Resources Institute, page 17 (July 1983). yet, field workers represent only 17 percent of the verified pesticide-related illnesses found under California's reporting system.``Summary of Illnesses and Injuries Reported by California Physicians as Potentially Related to Pesticides--1988;'' California Department of Food and Agriculture, Division of Pest Management, Environmental Protection and Worker Safety, Worker Health and Safety Branch, page 15 (April 30, 1990).This may reflect a higher sensitivity to pesticide concerns by California farm workers as a result of the multi-million dollar campaign mounted by the United Farm Workers Union to persuade consumers that boycott-targeted fruits and vegetables are contaminated by dangerous pesticides. The most reliable reporting system in the nation, according to EPA, is that of California. The most recent ``Summary of Illnesses and Injuries Reported by California Physicians as Potentially Related to Pesticides--1988,'' disclosed that only two thirds (2,118 of 3,144) of the potential pesticide illnesses reported by physicians could be considered definite, probable, or even possible pesticide-related illnesses. The numbers of reported and confirmed farm worker cases is not given; however, a measure of the sensitivity to pesticide danger to farm workers may be found by comparing the number of ``possible'' illnesses to the number confirmed or deemed ``probable'' across the various categories of workers. Of 19 categories, only agricultural field workers were reported to have a greater number of ``possible'' pesticide illnesses than ``definite or probable.'' Among field workers, only 36 percent of the reported potential pesticide illnesses were confirmed to be definite or probable; among all other workers 71 percent were definite or probable. These data suggest the level of reporting of pesticide-related illnesses of farm workers is higher than any other category, including categories such as pesticide applicators and handlers who are at far greater risk. Thus, the reasonable inference is that farm worker pesticide incidents are overreported. There may be a bias toward overreporting by physicians of pesticide illnesses for patients with symptoms similar to pesticide exposure, because after the first visit, pesticide illnesses are eligible for physicians' payments by workers compensation whereas similar symptoms (headaches, muscle aches, fatigue, flu, etc.) for nonoccupational illnesses are not. Ms. Coye's own hypothesis reveals that the 313,300 projection is not suited for estimating the pesticide-related illnesses among hired farm workers. She projected the number of hired farm worker illnesses to be 156,600. However, even 156,600 is not appropriate for EPA's analysis because the draft final rule concerns a subset of all hired farm workers which excludes livestock workers. Livestock workers represent about 48 percent of the U.S. hired labor force. Thus, if the Coye extrapolation is applied to the workers covered by this rule, her hired farm worker projection should be reduced by 48 percent to 81,432. Although WRI included Ms. Coye's hypothesis of 300,000, they did not consider it to be an estimate of farm worker illnesses. Instead, they said: Since 1973, when improved surveillance methods were put into effect, the number of poisoning incidents among fieldhands has risen each year by 13.9 percent. According to Molly Coye of the National Institute of Occupational Safety and Health in San Francisco, these fieldhands currently suffer the highest rates of occupationally related illnesses in the state: 7 cases per 1,000 for full-time workers (more than twice the overall average). Over 40 percent of these incidents involve only five compounds: parathion, diazinon, Phosdrin/Mevinphos, methomyl, and Omite Propagite.Field Duty: U.S. Farmworkers and Pesticide Safety, World Resources Institute, page 41 (1985). [Emphasis added] The 7 per 1,000 rate in California, if applied to EPA's estimated 1.8 million hired workers, computes to be 12,600. This number is remarkably similar to EPA's original ``best estimate'' of 12,500. The above quotation suggests that World Resources Institute (WRI), (and derivatively, EPA) confused all-occupational illness rates and pesticide-related illnesses. An examination of the underlying document by Ms. Coye indicates she referred to ``occupational illnesses,'' rather than pesticide-related illnesses.Coye, Molly Joel, Health Effects of Agricultural Production, Presented at a Symposium of the National Academy of Sciences, page 176 (1986).The underlying document also shows her data are not current but represent a 1979 survey. Although here they said ``occupationally related illnesses,'' WRI had earlier discussed the 40 percent incidence rate of five particular compounds in a clearly pesticide-related context, not of all occupational illnesses.Field Duty: U.S. Farmworkers and Pesticide Safety, World Resources Institute, page 38 (1985). If the draft final rule is to be based upon the WRI's report and Ms. Coye's treatise, it would be more appropriate to use her actual estimate of farm worker illnesses of 7 per 1,000 full-time workers.Coye, Molly Joel, Health Effects of Agricutural Production, Presented at a Symposium of the National Academy of Sciences, page 176 (1986). If the Coye/World Research Institute 7 per 1,000 full-time workers estimate is used, it should be recognized that most hired farm workers do not work on a full-time basis. In 1987, hired farm workers worked an average of 112 days or 45 percent of full-time.Oliveira, Victor J. and Cox, E. Jane, The Agricultural Work Force of 1987: A Statistical Profile, page 5 (May 1989). 0.45 X 1.8 million X 0.007 = 5,670 ÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄÄ 00001.TBL