OSHA rules for injury

AGENCY: Occupational Safety and Health Administration (OSHA), U.S. Department of Labor.

SUMMARY: The Occupational Safety and Health Administration (OSHA) is revising its rule addressing the recording and reporting of occupational injuries and illnesses (29 CFR parts 1904 and 1952), including the forms employers use to record those injuries and illnesses. The revisions to the final rule will produce more useful injury and illness records, collect better information about the incidence of occupational injuries and illnesses on a national basis, promote improved employee awareness and involvement in the recording and reporting of job-related injuries and illnesses, simplify the injury and illness recordkeeping system for employers, and permit increased use of computers and telecommunications technology for OSHA recordkeeping purposes.

This rulemaking completes a larger overall effort to revise Part 1904 of Title 29 of the Code of Federal Regulations. Two sections of Part 1904 have already been revised in earlier rulemakings. A rule titled Reporting fatalities and multiple hospitalization incidents to OSHA, became effective May 2, 1994 and has been incorporated into this final rule as § 1904.39. A second rule entitled Annual OSHA injury and illness survey of ten or more employers became effective on March 13, 1997 and has been incorporated into this final rule as § 1904.41.

The final rule being published today also revises 29 CFR 1952.4, Injury and Illness Recording and Reporting Requirements, which prescribes the recordkeeping and reporting requirements for States that have an occupational safety and health program approved by OSHA under § 18 of the Occupational Safety and Health Act (the "Act" or "OSH Act").

FOR FURTHER INFORMATION CONTACT: Jim Maddux, Occupational Safety and Health Administration, U.S. Department of Labor, Directorate of Safety Standards Programs, Room N-3609, 200 Constitution Ave., NW, Washington, DC 20210. Telephone (202) 693-2222.

The following is a table of contents for this preamble. The regulatory text and appendices follow the preamble. Documents and testimony submitted to the docket (Docket R-02) of this rulemaking are cited throughout this preamble by the number that has been assigned to each such docket entry, preceded by the abbreviation "Ex.," for exhibit.

II. The Occupational Safety and Health Act and the Functions of the Recordkeeping System

E. Subpart E. Reporting Fatality, Injury and Illness Information to the Government.

II. The Occupational Safety and Health Act and the Functions of the Recordkeeping System

The Occupational Safety and Health Act (the "OSH Act" or "Act") requires the Secretary of Labor to adopt regulations pertaining to two areas of recordkeeping. First, section 8(c)(2) of the Act requires the Secretary to issue regulations requiring employers to "maintain accurate records of, and to make periodic reports on, work-related deaths, injuries and illnesses other than minor injuries requiring only first aid treatment and which do not involve medical treatment, loss of consciousness, restriction of work or motion, or transfer to another job." Section 8(c)(1) of the Act also authorizes the Secretary of Labor to develop regulations requiring employers to keep and maintain records regarding the causes and prevention of occupational injuries and illnesses. Section (2)(b)(12) of the Act states Congress' findings with regard to achieving the goals of the Act and specifically notes that appropriate reporting procedures will help achieve the objectives of the Act.

Second, section 24(a) of the Act requires the Secretary to develop and maintain an effective program of collection, compilation, and analysis of occupational safety and health statistics. This section also directs the Secretary to "compile accurate statistics on work injuries and illnesses which shall include all disabling, serious, or significant injuries and illnesses, whether or not involving loss of time from work, other than minor injuries requiring only first aid treatment and which do not involve medical treatment, loss of consciousness, restriction of work or motion, or transfer to another job."

After passage of the Act, OSHA issued the required occupational injury and illness recording and reporting regulations as 29 CFR part 1904. Since 1971, OSHA and the Bureau of Labor Statistics (BLS) have operated the injury and illness recordkeeping system as a cooperative effort. Under a Memorandum of Understanding dated July 11, 1990 (Ex. 6), BLS is now responsible for conducting the nationwide statistical compilation of occupational illnesses and injuries (called the Annual Survey of Occupational Injuries and Illnesses), while OSHA administers the regulatory components of the recordkeeping system.

This revision of the Agency's recordkeeping rule is firmly rooted in the statutory requirements of the OSH Act (see the Legal Authority section of the preamble, below). OSHA's reasons for revising this regulation to better achieve the goals of the Act are discussed in the following paragraphs.

Occupational injury and illness records have several distinct functions or uses. One use is to provide information to employers whose employees are being injured or made ill by hazards in their workplace. The information in OSHA records makes employers more aware of the kinds of injuries and illnesses occurring in the workplace and the hazards that cause or contribute to them. When employers analyze and review the information in their records, they can identify and correct hazardous workplace conditions on their own. Injury and illness records are also an essential tool to help employers manage their company safety and health programs effectively.

Employees who have information about the occupational injuries and illnesses occurring in their workplace are also better informed about the hazards they face. They are therefore more likely to follow safe work practices and to report workplace hazards to their employers. When employees are aware of workplace hazards and participate in the identification and control of those hazards, the overall level of safety and health in the workplace improves.

The records required by the recordkeeping rule are also an important source of information for OSHA. During the initial stages of an inspection, an OSHA representative reviews the injury and illness data for the establishment as an aid to focusing the inspection effort on the safety and health hazards suggested by the injury and illness records. OSHA also uses establishment-specific injury and illness information to help target its intervention efforts on the most dangerous worksites and the worst safety and health hazards. Injury and illness statistics help OSHA identify the scope of occupational safety and health problems and decide whether regulatory intervention, compliance assistance, or other measures are warranted.

Finally, the injury and illness records required by the OSHA recordkeeping rule are the source of the BLS-generated national statistics on workplace injuries and illnesses, as well as on the source, nature, and type of these injuries and illnesses. To obtain the data to develop national statistics, the BLS and participating State agencies conduct an annual survey of employers in almost all sectors of private industry. The BLS makes the aggregate survey results available both for research purposes and for public information. The BLS has published occupational safety and health statistics since 1971. These statistics chart the magnitude and nature of the occupational injury and illness problem across the country. Congress, OSHA, and safety and health policy makers in Federal, State and local governments use the BLS statistics to make decisions concerning safety and health legislation, programs, and standards. Employers and employees use them to compare their own injury and illness experience with the performance of other establishments within their industry and in other industries.

The OSH Act authorizes OSHA to require employers to keep records and to report the recorded information to OSHA. However, the Agency only requires some employers to create and maintain occupational injury and illness records. Those employers who are required to keep records must report on those records only when the government specifically asks for the information, which occurs exclusively under limited circumstances that are described below.

Employers covered by the recordkeeping regulations must keep records of the occupational injuries and illnesses that occur among their employees. To do so, covered employers must complete two forms. First, the employer must maintain a summary form (OSHA Form 200, commonly referred to as the "OSHA Log," or an equivalent form) that lists each injury and illness that occurred in each establishment during the year. For each case on the Log, the employer also prepares a supplementary record (OSHA Form 101, or an equivalent), that provides additional details about the injury or illness. Most employers use a workers' compensation First Report of Injury in place of the 101 form. The Log is available to employees, former employees, and their representatives. A Summary of the Log is posted in the workplace from February 1 to March 1 of the year following the year to which the records pertain. The Log and summary, as well as the more detailed supplementary record, are available to OSHA inspectors who visit the establishment.

The employer is only obligated to record work-related injuries and illnesses that meet one or more of certain recording criteria. In accordance with the OSH Act, OSHA does not require employers to record cases that only involve "minor" injuries or illnesses, i.e., do not involve death, loss of consciousness, days away from work, restriction of work or motion, transfer to another job, medical treatment other than first aid, or diagnosis of a significant injury or illness by a physician or other licensed health care professional.

The language of the OSH Act also limits the recording requirements to injuries or illnesses that are "work-related." The Act uses, but does not define, this term. OSHA has interpreted the Act to mean that injuries and illnesses are work-related if events or exposures at work either caused or contributed to the problem. Work-related injuries or illnesses may (1) occur at the employer's premises, or (2) occur off the employer's premises when the employee was engaged in a work activity or was present as a condition of employment. Certain limited exceptions to this overriding geographic presumption were permitted by the former rule.

Although the Act gives OSHA the authority to require all employers covered by the OSH Act to keep records, two major classes of employers are not currently required regularly to keep records of the injuries and illnesses of their employees: employers with no more than 10 employees at any time during the previous calendar year, and employers in certain industries in the retail and service sectors.

Although the Act authorizes OSHA to require employers to submit reports on any or all injuries and illnesses occurring to their employees, there are currently only three situations where OSHA requires an employer to report occupational injury and illness records to the government. First, an employer must report to OSHA within eight hours any case involving a work-related fatality or the in-patient hospitalization of three or more employees as the result of a work-related incident (former 29 CFR 1904.8, final rule 1904.39). These provisions were revised in 1994 to reduce the reporting time for these incidents from 48 hours to 8 hours and reduce the number of hospitalized employees triggering a report from five workers to three workers (59 FR 15594 (April 1, 1994)). Changes made to this section in 1994 have largely been carried forward in the final rule being published today.

Second, an employer who receives an annual survey form from the Bureau of Labor Statistics must submit its annual injury and illness data to the BLS. The BLS conducts an annual survey of occupational injuries and illnesses under 29 CFR 1904.20-22 of the former rule (1904.41 of the final rule). Using a stratified sample, the BLS sends survey forms to randomly selected employers, including employers who, under Part 1904, would otherwise be exempt from the duty to keep the OSHA Log and Summary. These otherwise exempt employers are required to keep an annual record of the injuries and illnesses occurring among their employees that are recordable under Part 1904 if the BLS contacts them as part of the annual survey. At the end of the year, these employers must send the results of recordkeeping to the BLS. The BLS then tabulates the data and uses them to prepare national statistics on occupational injuries and illnesses. The BLS survey thus ensures that the injury and illness experience of employers otherwise exempted from the requirement to keep OSHA records -- such as employers with 10 or fewer employees in the previous year and employers in certain Standard Industrial Classification (SIC) codes -- is reflected in the national statistics. In accordance with its statistical confidentiality policy, the BLS does not make public the identities of individual employers.

Finally, OSHA may require employers to send occupational injury and illness data directly to OSHA under a regulation issued in 1997. That section of this regulation is entitled Annual OSHA Injury and Illness Survey of Ten or More Employers. It allows OSHA or the National Institute for Occupational Safety and Health (NIOSH) to collect data directly from employers. This section was published in the Federal Register on February 11, 1997 (62 FR 6434) and became effective on March 13, 1997. It has been included in this final rule as section 1904.41 without substantive change; however, this section has been rewritten in plain language for consistency with the remainder of Part 1904.

OSHA had several interrelated reasons for revising its recordkeeping rule. The overarching goal of this rulemaking has been to improve the quality of workplace injury and illness records. The records have several important purposes, and higher quality records will better serve those purposes. OSHA also believes that an improved recordkeeping system will raise employer awareness of workplace hazards and help employers and employees use and analyze these records more effectively. In revising its recordkeeping rule, the Agency also hopes to reduce underreporting and to remove obstacles to complete and accurate reporting by employers and employees.

A major goal of the revision has been to make the system simpler and easier to use and understand and to update the data on which the system is based. For example, OSHA has updated the list of partially exempt industries to reflect the most recent data available. The revisions to the final rule will also create more consistent statistics from employer to employer. Further, by providing more details about the system in the regulation itself and writing the rule in plain language, fewer unintentional errors will be made and the records will be more consistent. More consistent records will improve the quality of analyses comparing the injury and illness experience of establishments and companies with industry and national averages and of analyses looking for trends over several years.

Another objective of the rulemaking has been to lessen the recordkeeping burden on employers, reduce unnecessary paperwork, and enhance the cost-effectiveness of the rule. The final rule achieves this objective in several ways. It updates the partially exempt industry list, reduces the requirement to keep track of lengthy employee absences and work restrictions caused by work- related injuries and illnesses and, above all, greatly simplifies the forms, regulatory requirements, and instructions to make the system easier for employers and employees to manage and use.

In this rulemaking, OSHA has also addressed some of the objections employers have raised in the years since OSHA first implemented the injury and illness recordkeeping system. For example, the final rule includes a number of changes that will allow employers to exclude certain cases, eliminate the recording of minor illness cases, and allow employers maximum flexibility to use computer equipment to meet their OSHA recordkeeping obligations.

OSHA is also complying with the President's Executive Memorandum on plain language (issued June 1, 1998) by writing the rule's requirements in plain language and using the question-and-answer format to speak directly to the user. OSHA believes that employers, employees and others who compile and maintain OSHA records will find that the plain language of the final rule helps compliance and understanding.

Many of OSHA's goals and objectives in developing this final rule work together and reinforce each other. For example, writing the regulation in plain language makes the rule easier for employers and employees to use and improves the quality of the records by reducing the number of errors caused by ambiguity. In some cases, however, one objective had to be balanced against another. For example, the enhanced certification requirements in the final rule will improve the quality of the records, but they also slightly increase employer burden. Nevertheless, OSHA is confident that the final rule generally achieves the Agency's goals and objectives for this rulemaking and will result in a substantially strengthened and simplified recordkeeping and reporting system.

The quality of the records OSHA requires employers to keep is of crucial importance for anyone who uses the resulting data. Problems with completeness, accuracy, or consistency can compromise the data and reduce the quality of the decisions made on the basis of those data. Several government studies, as well as OSHA's own enforcement history, have revealed problems with employers' injury and illness recordkeeping practices and with the validity of the data based on those records.

A study conducted by the National Institute for Occupational Safety and Health (NIOSH) between 1981 and 1983 revealed that 25 percent of the 4,185 employers surveyed did not keep OSHA injury and illness records at all, although they were required by regulation to do so (Ex. 15:407-P).

A study of 192 employers in Massachusetts and Missouri conducted by the BLS in 1987 reported that an estimated 10 percent of covered employers did not maintain OSHA records at all, total injuries were underrecorded by approximately 10 percent (even though both overrecording and underrecording were discovered), lost workday injuries were undercounted by 25 percent, and lost workdays were undercounted by nearly 25 percent. Approximately half of the uncounted lost workdays were days of restricted work activity, and the other half were days away from work. Some of the underrecording was due to employers entering lost time cases on their records as no-lost-time cases (Exs. 72-1, 72-2).

Through its inspections of workplaces, OSHA has also discovered that some employers seriously underrecord injuries and illnesses. In cases where the inspector has found evidence that the employer willfully understated the establishment's injury and illness experience, OSHA has levied large penalties and fines under its special citation policy for egregious violations. OSHA has issued 48 egregious injury and illness recordkeeping citations since 1986 (Ex. 74).

As part of the OSHA Data Initiative (ODI), a survey allowing OSHA to collect injury and illness data from employers to direct OSHA's program activities, the Agency conducts Part 1904 records audits of 250 establishments each year. The following table shows the results of the audits conducted to date.

1996 Through 1998 OSHA Recordkeeping Audit Results *
Error type	Data reference year (percent)
Error type	1996	1997	1998
Cases not entered on employers Log	13.56	10.49	12.91
Lost workday cases recorded as non-lost workday cases	8.39	6.53	6.21
Non-lost workday cases recorded as lost workday cases	(**)	2.10	1.94
Total major recording errors	21.95	19.11	21.07
Total cases recorded without major errors	78.05	80.89	78.93

* The results were tabulated using unweighted data and should not be used to draw broad conclusions about the recordkeeping universe.
** Data not calculated for 1996.
Source: OSHA Data Initiative Collection Quality Control: Analysis of Audits on 1996-1998 Employer Injury and Illness Recordkeeping.

When OSHA's recordkeeping regulation was first promulgated in 1971, many industry safety experts were concerned that the regulations and the instructions on the forms did not provide adequate guidance for employers. They requested that the Department of Labor provide additional instructions on employers' recordkeeping obligations and clarify several recordkeeping issues. The BLS responded in 1972 by publishing supplemental instructions to the recordkeeping forms, BLS Report 412, What Every Employer Needs To Know About OSHA Recordkeeping (Ex. 1). These supplemental instructions were designed to help employers by providing detailed information on when and how to record injury and illness cases on the recordkeeping forms. The supplemental instructions clarified numerous aspects of the rule, including the important recordability criteria that outline which injuries and illnesses are work-related and thus recordable. This BLS Report was revised and reissued in 1973, 1975, and 1978.

In response to requests from labor and industry, and after publication in the Federal Register and a public comment period, the BLS 412 report series was replaced in April of 1986 by the Recordkeeping Guidelines For Occupational Injuries And Illnesses (the Guidelines) (Ex. 2). The Guidelines contained an expanded question-and-answer format similar to that of the BLS 412 report and provided additional information on the legal basis for the requirements for recordkeeping under Part 1904. The Guidelines provided clearer definitions of the types of cases to be recorded and discussed employer recordkeeping obligations in greater detail. The Guidelines also introduced a number of exceptions to the general geographic presumption that injuries and illnesses that occurred "on-premises" were work-related to cover situations where the application of the geographic presumption was considered inappropriate. Further, the Guidelines updated the lists that distinguished medical treatment from first aid and addressed new recordkeeping issues. The BLS also published a shortened version of the Guidelines, entitled A Brief Guide to Recordkeeping Requirements for Occupational Injuries and Illnesses (Ex. 7).

Although the 1986 edition of the Guidelines clarified many aspects of the recordkeeping regulation, concerns persisted about the quality and utility of the injury and illness data. In response to inquiries from employers, unions, employees, BLS, and OSHA staff, the Agency issued many letters of interpretation. These letters restated the former rule's regulatory requirements, interpreted the rules as they applied to specific injury and illness cases, and clarified the application of those requirements. A number of these letters of interpretation have been compiled and entered into the docket of this rulemaking (Ex. 70). OSHA has incorporated many of the prior interpretations directly into the implementation questions and answers in the regulatory text of the final rule, so that all affected employers will be aware of these provisions.

Because of concern about the injury and illness records and the statistics derived from them, several organizations outside OSHA have studied the recordkeeping system. The National Research Council (NRC), the Keystone Center, and the General Accounting Office (GAO) each published reports that evaluated the recordkeeping system and made recommendations for improvements. OSHA has relied on these studies extensively in developing this final rule.

In response to concern over the underreporting of occupational injuries and illnesses and inconsistencies in the national data collected by the BLS, Congress appropriated funds in 1984 for the BLS to conduct a quality assurance study of its Annual Survey of Occupational Injuries and Illnesses. The BLS asked the National Research Council (NRC) to convene an expert panel to analyze the validity of employer records and the BLS annual survey, to address any problems related to determining and reporting occupational diseases, and to consider other issues related to the collection and use of data on health and safety in the workplace.

In 1987, NRC issued its report, Counting Injuries and Illnesses in the Workplace: Proposals for a Better System (Ex. 4). The report contained 24 specific recommendations (Ex. 4, Ch. 8). In sum, the NRC panel recommended that BLS take the following steps to improve the recordkeeping system: (1) Modify the BLS Annual Survey to provide more information about the injuries and illnesses recorded; (2) discontinue the Supplementary Data System, replace it with a grant program for States and individual researchers, and develop criteria for the detail and quality of the data collected by the replacement system; (3) conduct an ongoing quality assurance program for the Annual Survey to identify underreporting by comparing the information on employers' logs with data from independent sources; (4) implement a system of surveillance for occupational disease, including the collection of data on exposure to workplace hazards; (5) improve the collection of national occupational fatality data; (6) implement an administrative data system that would allow OSHA to obtain individual establishment data to conduct an "effective program for the prevention of workplace injuries and illnesses * * *"; and (7) thoroughly evaluate recordkeeping practices in individual establishments, using additional resources requested from Congress for that purpose to avoid diverting resources from OSHA inspections of workplace hazards (Ex. 4, p. 10).

In 1987, The Keystone Center convened 46 representatives from labor unions, corporations, the health professions, government agencies, Congressional staff, and academia for a year-long dialogue to discuss occupational injury and illness recordkeeping. Two years later, Keystone issued its final report, Keystone National Policy Dialogue on Work-related Illness and Injury Recordkeeping, 1989 (Ex. 5). The report focused on four major topics: (1) Recordkeeping criteria; (2) OSHA enforcement procedures; (3) injury and illness data systems; and (4) occupational illnesses. The Keystone report recommended that: (1) OSHA and the BLS should revise various aspects of the recording criteria; (2) OSHA should use injury and illness data to target enforcement efforts; (3) the BLS should revise the Guidelines to make them easily and uniformly understood; (4) the BLS should develop a national system to collect and disseminate occupational injury and illness information; and (5) OSHA and the BLS should broaden the type of information collected concerning occupational illness and make the information available to employees and government agencies for appropriate purposes such as research and study.

An August 1990 report by the GAO, Options for Improving Safety and Health in the Workplace (Ex. 3), discussed the importance of employer injury and illness records. The GAO noted that these records have several major uses. They help employers, employees and others understand the nature and extent of occupational safety and health problems. They help employers and employees identify safety and health problems in their workplaces so that they can correct the problems. They also enable OSHA to conduct research, evaluate programs, allocate resources, and set and enforce standards. The report focused on the use of the records in OSHA enforcement, particularly in targeting industries and worksites for inspections and determining the scope of inspections.

The GAO report found that there was "possibly significant injury and illness underrecording and subsequent underreporting" (Ex. 3, p. 3). The GAO report gave three main reasons for inaccurate recording and reporting: (1) Employers intentionally underrecord injuries and illnesses in response to OSHA inspection policies or management safety competitions; (2) employers unintentionally underrecord injuries and illnesses because they do not understand the recording and reporting system; and (3) employers record injuries and illnesses inaccurately because they do not place a high priority on recordkeeping and do not supervise their recordkeepers properly. The GAO report noted that OSHA's revised enforcement procedures, which included increasing its fines for recordkeeping violations and modifying its records-review procedures, would likely help to improve the accuracy of recordkeeping. The GAO recommended that the Department of Labor study the accuracy of employers' records using independent data sources, evaluate how well employers understand the revised Guidelines, and audit employers' records in selected enforcement activities.

OSHA has developed a four-part strategy to improve the quality of the injury and illness records maintained by employers. The first component is to provide information, outreach and training to employers to make them more aware of the recordkeeping requirements, thereby improving their compliance with these requirements. For example, information on injury and illness recordkeeping is included in many of OSHA's publications and pamphlets, on the OSHA CD-ROM, and on OSHA's Internet site. OSHA personnel answer thousands of recordkeeping questions each year in response to phone calls and letters. OSHA also trains employers at the OSHA Training Institute in recordkeeping procedures and provides speakers on this topic for numerous safety and health events.

The second component is improved enforcement of the recordkeeping requirements. OSHA continues to review employer records during many of its workplace inspections. OSHA also audits the records of some employers who submit data to OSHA under former section 1904.17 (recodified as section 1904.41 Requests from OSHA for Data in the final rule). Although OSHA does not issue citations for minor reporting and recording violations, the Agency does cite and fine employers when it encounters serious or willful injury and illness recordkeeping problems.

The third component of OSHA's overall plan is this revision of the injury and illness recordkeeping rule. The revised final rule will streamline the recordkeeping system by simplifying the forms and the logic used to record an individual case. It will also consolidate the instructions that were formerly contained in the rule itself, in the Guidelines, and in many interpretative letters and memoranda. In addition, the final rule will improve the quality of the injury and illness records by changing several requirements to ensure that data are entered correctly. OSHA has simplified and streamlined the recordkeeping forms and processes to reduce errors. Other changes include: (1) Simplifying and clarifying the definitions of terms such as "medical treatment," "first aid," and "restricted work" to reduce recording errors; (2) providing specific recordkeeping guidance for specific types of injuries and illnesses; (3) including a detailed discussion of the process of determining whether an injury or illness is work-related; (4) giving employees greater involvement by improving their access to records and providing a longer posting period for the annual summary; (5) requiring higher level management officials to certify the records; (6) adding a falsification/penalty statement to the Summary; (7) adding a disclaimer to the Log to clarify that an employer who records an injury or illness is not admitting fault, negligence or liability for workers' compensation or insurance purposes; and (8) requiring the employer to establish a process for employees to report injuries and illnesses and to tell employees about it, and explicitly prohibiting the employer from discriminating against employees who report injuries and illnesses.

In 1995, the Keystone Center reassembled a group of business, labor, and government representatives to discuss draft proposed changes to the recordkeeping rule. OSHA shared its draft proposed revision of the rule with the participants and the public. The draft was also reprinted in several national safety and health publications. Written comments generated by the on-going dialogue were used to help develop the proposal and the final rule, and they are in the rulemaking record (Ex. 12).

OSHA consulted with the Advisory Committee on Construction Safety and Health (ACCSH) before issuing the proposed rule. ACCSH made specific recommendations to OSHA for improving the recordkeeping system as it applied to the construction industry. OSHA gave the ACCSH recommendations careful consideration and responded by modifying the proposal in several areas. The ACCSH recommendations, OSHA's written briefing, and the relevant portions of the transcripts of the October and December 1994 ACCSH meetings are also part of the public record (Ex. 10).

OSHA published a Notice of Proposed Rulemaking (NPRM) on February 2, 1996 (61 FR 23), giving formal notice that the Agency proposed to revise the injury and illness recording and reporting regulations, forms, and supplemental instructions (Ex. 14). The proposed rule reflected a number of suggestions made by the Keystone participants and ACCSH.

The NPRM invited all interested parties to submit comments on the proposal to the docket by May 2, 1996. In response to requests from members of the public, OSHA held two public meetings during the comment period and extended the comment period to July 1, 1996.

OSHA received 449 written comments in response to the NPRM and compiled 1200 pages of transcripts from 60 presentations made at the public meeting. Comments and testimony were received from a broad range of interested parties, including corporations, small business entities, trade associations, unions, state and local governments, professional associations, citizens groups, and safety and health organizations. OSHA has carefully reviewed all of the comments and testimony in its preparation of the final rule.

As described in greater detail below, the final rule revises OSHA's regulation for the recording and reporting of work-related deaths, injuries and illnesses. The rule is part of a comprehensive revision of the OSHA injury and illness recordkeeping system.

The final rule becomes effective, on January 1, 2002. At that time, the following recordkeeping actions will occur:

(1) 29 CFR Part 1904, entitled Recording and Reporting Occupational Injuries and Illnesses, will be in effect.

(2) The State plan provisions in 29 CFR Part 1952, Section 1952.4, entitled Injury and Illness Recording and Reporting Requirements will be in effect.

(A) OSHA Form 300, OSHA Injury and Illness Log, and OSHA Form 300 A Summary, which will replace the former OSHA Form 200, Log and Summary of Occupational Injuries and Illnesses; and

(B) OSHA Form 301, OSHA Injury and Illness Incident Record, which will replace the former OSHA Form 101, Supplementary Record of Occupational Injuries and Illnesses.

(B) A Brief Guide to Recordkeeping Requirements for Occupational Injuries and Illnesses, 1986.

(5) All letters of interpretation regarding the former rule's injury and illness recordkeeping requirements will be withdrawn and removed from the OSHA CD-ROM and the OSHA Internet site.

Two proposed regulatory sections in OSHA's 1996 Notice of Proposed Rulemaking (NPRM) have not been carried forward in this rulemaking. They are: (1) Falsification of, or failure to keep records or provide reports (Proposed section 1904.16), and (2) Subcontractor records for major construction projects (Proposed section 1904.17).

Paragraphs (a) and (b) of proposed section 1904.16, "Falsification of, or failure to keep records or provide reports," were included in the proposal because they had been included in the former rule. The proposed section included a provision stating that employers may be subject to criminal fines under section 17(g) of the Act for falsifying injury and illness logs and may be cited and fined under sections 9, 10, and 17 of the Act for failure to comply with the recordkeeping rule. Several commenters favored retention of this proposed provision in the final rule because, in their view, OSHA needs strong enforcement of the recordkeeping rule to make sure that employers keep accurate records (see, e.g., Exs. 15: 11, 289). Others, however, objected to the proposed provision (see, e.g., Exs. 15: 22, 335, 375). The views of this latter group were reflected in a comment from the American Petroleum Institute (Ex. 15: 375), which urged OSHA to delete this section from the rule in its entirety because nothing like it is found in any other OSHA regulation or standard. In the final rule, OSHA has decided that this section is not needed to enforce the final rule, and when need be, to issue citations and levy penalties.

The Keystone report recommended, and OSHA proposed, to require construction employers to maintain "site logs," or comprehensive injury and illness records, for major construction projects. The Keystone report noted that construction sites are normally composed of multiple contractors and subcontractors, each of whom may be present at the site for a relatively short period of time, and that no records of the safety and health experience of the site are readily available, either to OSHA or to employers and employees.

In an attempt to address this problem, the proposed provision would have required site-controlling employers in the construction industry to maintain a separate record reflecting the overall injury and illness experience of employees working for sub-contract construction firms for any construction site having an initial construction contract value exceeding $1,000,000. The site-controlling employer would thus have been required to record the injuries and illnesses of subcontractor employees who were employed by construction employers with 11 or more employees working at the site at any time during the previous calendar year.

Many commenters strongly favored the addition of a construction site log provision to the final rule (see, e.g., Exs. 20; 29; 35; 36; 45; 15: 48, 110, 113, 129, 136, 137, 141, 181, 224, 266, 278, 310, 350, 359, 369, 375, 394, 407, 413, 415, 418, 425, 438, 440). Several of these commenters urged OSHA to expand this "multi-employer" log concept to employers in other industries (see, e.g., Exs. 35; 15: 48, 113, 129, 369, 415, 418, 438). For example, the AFL-CIO (Ex. 15: 418) encouraged OSHA to "[e]xpand this recommendation to all industries. As the Agency is well aware, safety and health problems related to multi-employer worksites and contract work are a major concern in many industries beyond construction. Many of the major chemical explosions and fatalities at steel mills, power plants and paper mills have been related to contract work. With more and more businesses contracting out services for on-site activities, the safety and health concern associated with these practices is growing."

Other commenters argued that the proposed site log provisions should be expanded to include injuries and illnesses to construction employees working for employers who would otherwise be exempt from OSHA recordkeeping requirements because they employ fewer than 11 workers (see, e.g., Exs. 20; 15: 350, 359, 369, 407, 425). Two of these commenters recommended adding a requirement to the final rule requiring the site-controlling employer to assist smaller employers with their records (Exs. 15: 350, 359).

Several commenters recommended adding provisions to the final rule that would provide greater access to the construction site log by employees (see, e.g., Exs. 15: 129, 310, 394) and by other employers (see, e.g., Ex. 15: 310). Others recommended that OSHA include in the final rule a requirement for the site-controlling employer to collect the number of hours worked by each subcontractor to make it easier to calculate each subcontractor's injury and illness rates (see, e.g., Exs. 15: 310, 369, 394), and some commenters recommended that the final rule contain a requirement for subcontractors to report work-related injuries and illnesses to the site-controlling employer (see, e.g., Exs. 15: 359, 369, 440).

The Building and Construction Trades Department (BCTD), AFL-CIO discussed many of these issues while commenting in favor of site logs:

On the project level, the fragmentation of employers on construction sites makes it impossible to assess fully safety and health on a particular project. Since the origins of OSHA, injury and illness recordkeeping has been the responsibility of each individual employer. Nevertheless, the hazards of construction activity are shared by employees across the site, and are not specific to a single employer. Employees are often injured or made ill by circumstances that are not under their own employer's full control. The balkanization of recordkeeping contributes to the failure of full and complete communication in construction.

What is needed, at a national and the project level, is a way to record and count the injuries and illnesses that occur on specific projects. We need to know about illnesses and injuries that are associated with distinct types of construction activity, with the various phases of construction, and with the methods, materials, and hazards that are common to those types of work. Furthermore, we need to develop a measure of injury and illness that spans employers, to get a picture of the aggregate outcomes affecting all actors on a common site. Only with such a tool can the construction industry establish and meet performance benchmarks for safety and health.

Site logs would be useful to all of the actors in the occupational safety and health arena. First, employers would benefit from the collection of this data. General contractors increasingly use safety and health information in selecting their subcontractors, and in evaluating projects. Site logs will give them a new tool for both self-evaluation and the evaluation of other contractors. Similarly, subcontractors are often ignorant of the safety and health performance of other contractors and the general contractor. Site logs will lead to better information for all contractors on the project.

Second, employees will benefit from site logs. The site log will focus employers' attentions upon the risks and hazards that are encountered across the worksite. By concretely illustrating that hazards are everyone's problems, the site log will prompt employers and employees to minimize those hazards and to maximize site safety and health.

Third, owners will benefit from site logs. Today, many owners are selecting contractors on the basis of the contractors' rates for lost work days and total recordables. In many cases, these rates are a poor measure for the owner's purpose. An owner's typical concern is with how well a general contractor manages safety and health on the entire site, not with how many injuries and illnesses occurred within that contractor's own workforce. Site logs can be used to measure the management performance of the general contractor, and will greatly assist the owners in their quest for construction safety.

Finally, OSHA will find the site logs to be enormously useful in its efforts to become a "data-driven" agency. First, a project-centric focus will allow OSHA to focus its enforcement and consultation resources. Site logs will be useful to OSHA in scheduling inspections during the phases of construction which appear, through this data, to present the most risks, and in focusing its inspections at construction sites, since the recent illness and injury history of the entire site can be assayed by examining a single document. By the same token, the information revealed by the logs will assist OSHA in reaching out to employers to provide consultative services. Site specific data will also aid OSHA in developing safety and health standards that are appropriately tailored to the risks and hazards of specific types of construction.

The BCTD is convinced that private actors will use site logs to improve safety and health performance. If OSHA establishes a requirement that site logs be kept, the private marketplace will use this new tool to the betterment of employee safety and health (Ex. 15: 394).

Other commenters opposed the addition of a site log provision to the final rule (see, e.g., Exs. 43; 51; 15: 9, 17, 21, 38, 40, 43, 61, 67, 74, 77, 97, 111, 116, 119, 121, 126, 151, 155, 163, 170, 194, 195, 204, 213, 235, 242, 256, 260, 262, 263, 265, 269, 270, 281, 294, 298, 304, 305, 312, 314, 341, 342, 351, 356, 364, 377, 389, 395, 397, 401, 406, 412, 423, 433, 437, 443, 441). The most common argument presented by these commenters was that records should only be kept by the employer, and that one employer should not keep records for another employer's employees (see, e.g., Exs. 15: 9, 116, 126, 163, 195, 204, 260, 262, 265, 281, 294, 304, 312, 314, 341, 342, 351, 364, 389, 395, 396, 397, 401, 406, 423, 433). The Jewell Coal and Coke Company (Ex. 15: 281) stated that:

[t]he sub-contractor should be responsible for keeping up with their own employee injury/illness records as they are the ultimate responsible party for their own employees under worker's compensation regulations and in all other legal issues. This proposal would appear to be trying to switch total responsibility to the site controlling employer for that record keeping purpose and taking the responsibility off the subcontractor with whom the responsibility should lie. It is, we feel, unfairly discriminatory against the site-controlling employer in this case and we are strongly opposed to the wording of this proposal. Even the alternative proposal in this section places the ultimate responsibility upon the project owner for collection of accident and illness information and send it to OSHA. Again we are strongly opposed to the wording of this proposal because it takes the responsibility for record keeping off the sub-contractor and places the ultimate responsibility on the project owner, a responsibility that we feel belongs to the sub-contractor irregardless of their size.

Brown & Root, Inc. (Ex. 15: 423) added "A site controlling employer cannot be held responsible for determining which injuries and illnesses of a subcontractor's employees are recordable. A contractor cannot become involved in the medical records of employees who do not work for him or her. The subcontractor employer has to be held accountable and responsible for his own employees, this responsibility cannot be delegated to another contractor. The number of employees or the value of the construction project is irrelevant."

Some of the commenters who generally opposed this provision agreed that site-specific data would be useful if it could be collected by a method that allowed each employer to keep its own records (see, e.g., Exs. 15: 9, 116, 195, 260, 262, 265, 304, 364, 401). Other commenters pointed out that there would be problems in getting accurate data from subcontractors (see, e.g., Exs. 15: 242, 263, 269, 270, 310, 314, 377, 395, 397, 406) or suggested that the site-controlling employer should not be held responsible for the quality of the records received from subcontractors (see, e.g., Exs. 33; 15: 176, 195, 231, 273, 294, 301, 305, 312, 351).

The Alabama Branch of the Associated General Contractors of America, Inc. (AGC) cited difficulties associated with other regulatory requirements that could result from the proposed OSHA site log requirement:

This could place an undue hardship on the site controlling employer far beyond his ability to appoint and manage independent contractors and subcontractors without there being other entangling both federal and state obligations, which would lead to the subcontractor's employees being declared employees of the controlling contractor. Many states use the common law to make a determination of the employer/employee relationship, as well as the Internal Revenue Service. This employee/employer relationship under the common law usually says if a controlling contractor exercises any control as to time, place, method or result of a person's work that they are in fact defacto employees of the controlling contractor, for social security purposes and other state purposes. Therefore, I think it is shallow thinking to believe that the general contractor with 100 subcontractors should have all 5,500 employees under their control and avoid other legal entanglements, without the ability to actually control the subcontractor.

The National Federation of Independent Business (NFIB) expressed concern about the proposed site log provision as it would relate to OSHA's multi-employer citation policy (Ex. 15: 304), and the Small Business Administration (Exs. 51: 67, 437) argued that the proposed requirement would require competing employers to share sensitive business information.

A number of commenters objected to the requirement because of the additional burden it would place on employers (see, e.g., Exs. 51; 15: 40, 43, 67, 77, 97, 119, 121, 163, 194, 204, 235, 242, 256, 263, 269, 270, 294, 298, 304, 312, 314, 356, 377, 389, 395, 397, 406, 412, 437, 441), arguing that the proposed requirement would result in duplication (see, e.g., Exs. 51; 15: 9, 38, 67, 77, 119, 155, 204, 304, 312, 351, 356, 364, 377, 395, 397, 437). For example, the American Iron and Steel Institute (Ex. 15: 395) stated that the proposed requirement would place a "near impossible burden on the 'site controlling employer' " to determine the size of each subcontractor to decide which subcontractors would be required to keep records.

A number of commenters also questioned the value of the statistical data that would be produced by a site log requirement (see, e.g., Exs. 51; 15: 61, 62, 67, 74, 77, 97, 121, 151, 194, 312, 314, 351, 389, 395, 433, 437, 433), and several participants were concerned that the records would not be useful for accident prevention purposes (see, e.g., Exs. 15: 121, 151, 312, 351, 389, 433) .

OSHA received many comments addressing miscellaneous points related to the proposed construction site log requirement. For example, some commenters suggested limiting the scope of the project records required to be maintained (see, e.g., Exs. 15: 17, 21, 111, 116, 213, 155), while others argued that the proposed dollar threshold ($1 million) for a covered construction project was too low and should be raised (see, e.g., Exs. 15: 17, 111, 116, 441). Others suggested that the site log requirement should be triggered by the time duration of the project (Ex. 15: 116); the number of construction workers at the site (Ex. 15: 111); or include only construction employers with more than 11 employees (see, e.g., Exs. 15: 170, 213, 405). Some commenters urged the Agency not to expand the site log concept beyond the construction industry (see, e.g., Exs. 33; 15: 176, 231, 273, 301, 397). Finally, several commenters urged OSHA to make any site log provision in the final rule compatible with the corresponding provisions of the Process Safety Management Standard (29 CFR 1910.119), especially if the site log requirement in the recordkeeping rule was expanded beyond construction (see, e.g., Exs. 33; 15: 159, 176, 231, 273, 301, 335).

Based on a thorough review of the comments received, OSHA has decided not to include provisions in the final that require the site-controlling employer to keep a site log for all recordable injuries and illnesses occurring among employees on the site. OSHA has made this decision for several reasons. First, such a provision would not truly capture the site's injury and illness experience because many subcontractors employ 10 or fewer employees and are therefore exempt from keeping an OSHA Log. To require these very small employers to keep records under Part 1904 for the periods of time they worked on a construction site meeting the dollar threshold for this provision would be a new recordkeeping burden. This would create considerable complexity for these employers and for the site-controlling employer. Second, under the Data Initiative (section 1904.41 of the final rule), OSHA now has a means of targeting data requests for records of the safety and health experience of categories of employers and can therefore obtain the data it needs to establish inspection priorities in a less administratively complex and less burdensome way when the Agency needs such data. Third, OSHA was concerned with the utility of the data that would have been collected under the proposed site log approach, because of the time lag between collection of the data and its use in selecting employers for inspections or other interventions. In many cases work at the site would be complete before the data was collected and analyzed. Finally, a site log requirement is not necessary to enable general contractors to compare the safety records of potential subcontractors since they can require such information as a condition of their contractual arrangements without OSHA requirements. For these reasons, the final rule does not contain a site log provision.

Several commenters suggested that the Agency use data from existing data sources, such as state workers' compensation agencies, insurance companies, hospitals, or OSHA inspection files, instead of requiring separate data for OSHA recordkeeping purposes (see, e.g., Exs. 15: 2, 28, 58, 63, 97, 184, 195, 289, 327, 341, 374, 444). For example, Alex F. Gimble observed:

Since similar data are readily available from other sources, such as the National Safety Council, insurance carriers, etc., why not use these statistics, rather than go through this duplication of effort at taxpayer expense? Another approach would be to utilize data collected by OSHA and State Plan compliance officers during site visits over the past 25 years (Ex. 15: 28).

Several commenters suggested that OSHA use injury and illness data from the workers' compensation systems in lieu of employer records. The comments of the American Health Care Association (AHCA) are representative of the views of these commenters:

AHCA encourages OSHA to consider the use of workers' compensation data in lieu of proposed OSHA 300 and 301 forms. Pursuing the enactment of legislation that would allow OSHA access to every state's workers' compensation data would eliminate the need for employers to maintain two sets of records, provide OSHA with necessary safety and health data, and ease administrative and cost burdens now associated with recordkeeping for employers in every industry across the country (Ex. 15: 341).

Ms. Diantha M. Goo recommended the use of injury and illness data obtained from treatment facilities rather than the OSHA records:

The accuracy and usefulness of OSHA's reporting system would be vastly improved if it were to shift responsibility from employers (who have a vested interest in concealment) to the emergency rooms of hospitals and clinics. Hospitals are accustomed to reporting requirements, use the correct terminology in describing the accident and its subsequent treatment and are computerized (Ex. 15: 327).

In response to these comments, OSHA notes that the injury and illness information compiled pursuant to Part 1904 is much more reliable, consistent and comprehensive than data from any available alternative data source, including those recommended by commenters. This is the case because, although some State workers' compensation programs voluntarily provide injury and illness data to OSHA for various purposes, others do not. Further, workers' compensation data vary widely from state to state. Differing state workers' compensation laws and administrative systems have resulted in large variations in the content, format, accessibility, and computerization of that system's data. In addition, workers' compensation databases often do not include injury and illness data from employers who elect to self-insure.

Additionally, most workers' compensation databases do not include information on the number of workers employed or the number of hours worked by employees, which means that injury and illness incidence rates cannot be computed from the data. Workers' compensation data are also based on insurance accounts (i.e., filed claims), and not on the safety and health experience of individual workplaces. As a result, an individual account often reflects the experience of several corporate workplaces involved in differing business activities. Finally, as discussed below in the Legal Authority section of the preamble, the OSH Act specifically sets out the recordability criteria that must be included in the OSHA recordkeeping system envisioned by the Congress when the Act was passed. The Congress intended that all non-minor work-related injuries and illnesses be captured by the OSHA recordkeeping system, both so that individual establishments could evaluate their injury and illness experience and so that national statistics accurately reflecting the magnitude of the problem of occupational injury and illness would be available.

Although OSHA disagrees that any of the alternate sources of data are satisfactory substitutes for the information gathered under Part 1904, the Agency recognizes that data from these sources have value. To the extent that information from workers' compensation programs, the BLS statistics, insurance companies, trade associations, etc., are available and appropriate for OSHA's purposes, OSHA intends to continue to use them to supplement its own data systems and to assess the quality of its own data. However, consistent with the Congressional mandate of the OSH Act, OSHA must continue to maintain its own recordkeeping system and to gather data for this system through recording and reporting requirements applicable to covered employers.

The requirements of 29 CFR 1952.4 describe the duties of State-Plan states to implement the 29 CFR 1904 regulations. These requirements are discussed in Section IX of the preamble, State Plans, and in the preamble discussion for section 1904.37, State recordkeeping regulations.

In addition to the issues discussed above, three issues concerning recordkeeping warrant discussion: analysis of the data, training and qualifications of recordkeepers, and recordkeeping software.

During OSHA's public meetings, Eric Frumin of the Union of Needletrades, Industrial and Textile Employees, AFL-CIO (UNITE) urged OSHA to include a requirement for employers to analyze the OSHA 1904 data in depth to discover patterns and trends of occupational injury and illness, stating that:

[y]ou're telling the employers to evaluate information that's coming to them, and I say that to stress the point that's a very logical, common sense requirement and you're not generally speaking asking them to do that once they compile a log. You stop short of asking employers to evaluate the log in toto, to look for the kinds of trends and comparisons and so forth that we've been discussing here. I think it's important for OSHA to consider some -- making such a requirement, particularly in light of a fairly consistent pattern of testimony in this proceeding, wherein employers now do not analyze what's on the log in much depth. * * * But what has emerged at the end of the day is not a whole lot of use of the information on the log for -- in terms of analyzing it for trends and various associations or conclusions about how to protect people, how to stop the injuries and illness (Ex. 58X, pp. 372 -- 375).

In the final rule, OSHA has not included any requirement for employers to analyze the data to identify patterns or trends of occupational injury and illness. OSHA agrees with Mr. Frumin that analysis of the data is a logical outgrowth of maintaining records. Employers and employees can use such analyses to identify patterns and trends in occupational injuries and illnesses, and use that information to correct safety and health problems in the workplace. OSHA encourages both employers and employees to use the data for these purposes. However, a requirement of this type would go beyond the scope of the recording and reporting rule, which simply requires employers to keep records of work-related injuries and illnesses, and report the data under certain circumstances. OSHA believes that requirements of this type are better addressed through an OSHA standard, rather than the 1904 recordkeeping regulation.

The American Federation of Labor and Congress of Industrial Organizations (AFL-CIO) suggested that OSHA add requirements for the training of the individual who maintains the 1904 records for the employer, stating that:

[a]nother important issue relates to the qualifications and responsibilities of the individual filling out the 300 log and Form 301. Most workplaces generally have a non-safety and health professional entering this information in the 300 log after the decision of a recordable injury or illness has been made. In our view it is important that these individuals have proper training about the recordkeeping rule and the employer's recordkeeping system. In order to assure the most accurate and complete recording of work-related injuries and illnesses, we encourage the Agency to consider developing guidelines for the qualifications and training of these individuals (Ex. 15: 418).

OSHA has not included a training requirement for the person entering the information on the Part 1904 records in this final rule. The Agency believes that the Section 1904.32 provisions of the final rule calling for annual review of the records and certification of the annual summary by a company executive will ensure that employers assign qualified personnel to maintain the records and to see that they are trained in that task. Further, because OSHA did not include training requirements in its 1996 proposal, the Agency has not gathered sufficient information in the rulemaking docket about whether specific training provisions would have utility, as well as the appropriate qualifications and training levels that would assist in writing such provisions at this time.

As part of its outreach and training program accompanying this rule, OSHA will be providing speeches and seminars for employers to help them train their recordkeeping staff. OSHA will also be producing materials employers can use to help train their recordkeeping staff, including free software employers can use to keep records, training programs, presentations, course outlines, and a training video. All of these materials will be available through OSHA's Internet home page at http://www.osha.gov/.

In its proposal (61 FR 4048), OSHA asked the public to comment on whether or not OSHA should develop computer software to make injury and illness recordkeeping easier for employers, and discussed the features that would be desirable for such software. Those features were:

-- the ability to assist the employer in evaluating the entered data through several preset analytical tools (e.g., tables, charts, etc.); and

-- computer based training tools to assist employers in training employees in proper recordkeeping procedures.

OSHA also suggested that any such software should be in the public domain and/or be available at cost to the public and asked the following questions: What percentage of employers have computers to assist them in their business? What percentage of employers currently use computers for tracking employee-related information (payroll, timekeeping, etc.)? Should the distribution be through the Government, public domain share-ware distribution, or other channels? Should OSHA develop the software or only provide specifications for its requirements?

Several commenters said that most business establishments had computers (see, e.g., Exs. 15: 9, 95, 163, 281, 288, 375). The American Health Care Association (AHCA) estimated that 50% to 70% of their members used computers (Ex. 15: 341), and Raytheon Constructors, Inc. estimated that 60% of employers are using computers. OSHA agrees that computers are available in most businesses, although certainly not all of them. The agency also notes that these comments were made in 1996, and that businesses' computer usage has grown since that time.

A number of commenters urged OSHA to produce and distribute software to help employers keep the Part 1904 records (see, e.g., Exs. 35; 36; 51; 15: 9, 26, 32, 34, 67, 68, 76, 87, 95, 105, 109, 111, 129, 154, 157, 170, 181, 182, 197, 225, 235, 239, 247, 272, 277, 281, 283, 288, 303, 313, 327, 341, 347, 350, 352, 353, 356, 394, 405, 406, 409, 418, 426, 437, 438). The commenters gave various reasons for favoring the provision of OSHA-provided software, including reducing the burden and cost of the rule for employers (see, e.g., Exs. 15: 87, 95, 111, 170, 182, 197, 350), saving businesses programming costs (Ex. 15: 277), helping small businesses (Ex. 51; 15: 67), resulting in more uniform data (see, e.g., Exs. 36; 15: 32, 153, 170, 181, 347, 409, 418), and facilitating analysis of the data (see, e.g., Exs. 35; 15: 153, 418). For example, the Ford Motor Company stated that "Ford feels that the development of recordkeeping software by OSHA, which will employ a decision-making logic, automatic form generation, the ability to assist the employer in evaluating the entered data, and a tutorial section to assist employers in training is necessary. This will enhance the uniformity of data collection amongst all users, which is currently lacking" (Ex. 15: 347). The Muscatine Iowa Chamber of Commerce Safety Committee (Ex. 15: 87) added that:

"[e]very feature identified as a minimum requirement would be a great benefit to employers attempting to comply with the OSHA recordkeeping requirements. Prompts which would in any way aid in the determination of recordability would be appreciated by any person without a great deal of experience in filing OSHA reports. We feel these features are especially important now with the changes in forms and information to be collected."

Several of the commenters who urged OSHA to provide computer software tempered their support by asking that the use of such software should be optional and not mandatory (see, e.g., Exs. 15: 60, 109, 154, 198, 225, 247, 272, 303, 394), and several other commenters recommended that OSHA provide both software and specifications so employers could use the OSHA product to build their own data systems (see, e.g., Exs. 15: 170, 247, 283).

A number of commenters told OSHA that the Agency should not produce software to help employers with their 1904 recordkeeping obligations (see, e.g., Exs. 15: 78, 82, 85, 156, 163, 324, 348, 359, 363, 374, 375, 378, 402, 414). Several of these commenters suggested OSHA produce software performance specifications for the industry (see, e.g., Exs. 15: 156, 163, 357, 387). The commenters had various reasons for opposing the production of software. Several stated that each employer wants different data in its own unique form (see, e.g., Exs. 15: 78, 85, 375, 414). For example, the Central Vermont Public Service Corporation (Ex. 15: 85) stated that "[b]usinesses using safety related software use programs that can perform OSHA recordkeeping and workers' compensation functions in one package. It is unlikely that software developed by OSHA will perform workers' compensation functions and therefore it will not be well received or utilized by business." Other commenters stated that OSHA should focus elsewhere, that the private sector could produce software more economically (see, e.g., Exs. 15: 357, 375, 387), and that OSHA software is not needed (see, e.g., Exs. 15: 363, 378). For example, the Synthetic Organic Chemical Manufacturers Association, Inc. (SOCMA) stated that "[a]n outside organization with software development expertise should develop the software. OSHA's limited resources should go directly toward improving safety and health in the workplace" (Ex. 15: 357). The Air Transport Association added: "[m]ost major companies have developed their own software to support required OSHA recordkeeping, and others have taken advantage of commercially available programs. We see no need for OSHA to enter this market" (Ex. 15: 378).

OSHA has decided that the Agency will produce software for employers to use for keeping their OSHA 1904 records. There is obviously a need for the Agency to provide outreach and assistance materials for employers, particularly small employers, to help them meet their obligations in the least burdensome way possible, and software will clearly help achieve this goal. In addition, computer software will improve the consistency of the records kept by employers, and will assist them with analysis of the data. At this time, OSHA has not developed the software or its specifications, but will make every effort to produce and distribute software to assist employers by the time this final rule becomes effective. Use of the OSHA produced software will be optional; employers are not required to use this software and may keep records using paper systems. Employers are also free to produce their own software, or to purchase software.

A. The Final Recordkeeping Rule Is a Regulation Authorized by Sections 8 and 24 of the Act

The Occupational Safety and Health Act authorizes the Secretary to issue two types of final rules, "standards" and "regulations." Occupational safety and health standards, issued pursuant to section 6 of the Act, specify the measures to be taken to remedy known occupational hazards. 29 U.S.C. 652(8), 655. Regulations, issued pursuant to general rulemaking authority found, inter alia, in section 8 of the Act, are the means to effectuate other statutory purposes, including the collection and dissemination of records on occupational injuries and illnesses. 29 U.S.C. 657(c)(2).

OSHA is issuing this final recordkeeping rule as a regulation pursuant to the authority expressly granted by sections 8 and 24 of the Occupational Safety and Health Act, 29 U.S.C. 657, 673. Section 8 authorizes the Secretary to issue regulations she determines to be necessary to carry out her statutory functions, including regulations requiring employers to record and report work-related deaths and non-minor injuries and illnesses.⁽¹⁾ Section 8(c)(1) of the Act requires each employer to "make, keep and preserve, and make available to the Secretary [of Labor] or the Secretary of Health [and Human Services], such records regarding his activities relating to this Act as the Secretary, in cooperation with the Secretary of Health and Human Services, may prescribe by regulation as necessary or appropriate for the enforcement of this Act or for developing information regarding the causes and prevention of occupational accidents and illnesses." Section 8(c)(2) further provides that the "Secretary, in cooperation with the Secretary of Health and Human Services, shall prescribe regulations requiring employers to maintain accurate records of, and to make periodic reports on, work-related deaths, injuries and illnesses other than minor injuries requiring only first aid treatment and which do not involve medical treatment, loss of consciousness, restriction of work or motion, or transfer to another job." Section 8(c)(3) empowers the Secretary to require employers to "maintain accurate records of employee exposures to potentially toxic materials or harmful physical agents which are required to be monitored or measured under Section 6."

Section 8(g)(1) authorizes the Secretary "to compile, analyze, and publish, whether in summary or detailed form, all reports or information obtained under this section." Section 8(g)(2) of the Act empowers the Secretary "to prescribe such rules and regulations as he may deem necessary to carry out his responsibilities under the Act."

Section 24 contains a similar grant of regulatory authority. It requires the Secretary to "develop and maintain an effective program of collection, compilation, and analysis of occupational safety and health statistics * * * The Secretary shall compile accurate statistics on work injuries and illnesses which shall include all disabling, serious, or significant injuries and illnesses, whether or not involving loss of time from work, other than minor injuries requiring only first aid treatment and which do not involve medical treatment, loss of consciousness, restriction of work or motion, or transfer to another job." Section 24 also empowers the Secretary to "promote, encourage, or directly engage in programs of studies, information and communication concerning occupational safety and health statistics." Finally, Section 24 requires employers to "file such reports with the Secretary as he shall prescribe by regulation, as necessary to carry out his functions under this chapter."

Section 20 of the Act, 29 U.S.C. 669, contains additional implicit authority for collecting and disseminating data on occupational injuries and illnesses. Section 20(a) empowers the Secretaries of Labor and Health and Human Services to consult on research concerning occupational safety and health problems, and provides for the use of such research, "and other information available," in developing criteria on toxic materials and harmful physical agents. Section 20(d) states that "[i]nformation obtained by the Secretary and the Secretary of [HHS] under this section shall be disseminated by the Secretary to employers and employees and organizations thereof."

Two federal circuit Courts of Appeals have held that rules imposing recordkeeping requirements are regulations and not standards, and are thus reviewable initially in the district courts, rather than the Courts of Appeals. Louisiana Chemical Assn. v. Bingham, 657 F.2d 777, 782-785 (5th Cir. 1981) (OSHA rule on Access to Employee Exposure and Medical Records); Workplace Health & Safety Council v. Reich, 56 F.3d 1465, 1467-1469 (D.C. Cir. 1995) (OSHA rule on Reporting of Fatality or Multiple Hospitalization Incidents). These courts applied a functional test to differentiate between standards and regulations: standards aim toward correction of identified hazards, while regulations serve general enforcement and detection purposes, including those outlined in section 8. E.g., Workplace Health & Safety Council, 56 F.3d at 1468. See also United Steelworkers of America v. Reich, 763 F.2d 728, 735 (3d Cir. 1985) (Hazard Communication rule is a standard because it aims to ameliorate the significant risk of inadequate communication about hazardous chemicals). Clearly, the recordkeeping requirements in this final rule serve general administrative functions: They are intended to "aid OSHA's effort to identify the scope of occupational safety and health problems," to "serve as the foundation for national statistics on the number and rate of workplace injuries and illnesses" and "to raise employers" awareness of the kinds of injuries and illnesses occurring in their workplaces." See Functions of the Recordkeeping System, supra. Therefore, the final rule falls squarely within the mandate of sections 8 and 24 of the Act and is properly characterized as a regulation.

B. The Legal Standard: The Regulation Must Be Reasonably Related to the Purposes of the Enabling Legislation

Under section 8, the Secretary is empowered to issue "such * * * regulations as [s]he may deem necessary to carry out [her] responsibilities under this Act[,]" including regulations requiring employers to record and to make reports on "work-related deaths, injuries and illnesses other than minor injuries requiring only first aid treatment and which do not involve medical treatment, loss of consciousness, restriction of work or motion or transfer to another job." 29 U.S.C. 657(g)(2), (c)(2). Similarly, section 24 directs the Secretary to compile accurate statistics on "all disabling serious, or significant injuries and illnesses, whether or not involving loss of time from work, other than minor injuries. * * *" 29 U.S.C. 673(a). Where an agency is authorized to prescribe regulations "necessary" to implement a statutory provision or purpose, a regulation promulgated under such authority is valid "so long as it is reasonably related to the enabling legislation." Mourning v. Family Publications Service, Inc., 411 U.S. 356, 369 (1973).

Section 8(g)(2) is functionally equivalent to the enabling legislation at issue in Mourning; therefore a reviewing court must examine the final recordkeeping rule's relationship to the purposes of section 8. Cf. Louisiana Chemical Assn. v. Bingham, 550 F. Supp. 1136, 1138-1140 (W.D. La. 1982), aff'd, 731 F.2d 280 (5th Cir. 1984) (records access rule is directly related to the goals stated in the Act and supported by the language of section 8).

C. The Final Recordkeeping Rule's Key Provisions Are Reasonably Related to the Purposes of the OSH Act

The goal of this final rule, as stated in the Summary, is to improve the quality and consistency of injury and illness data while simplifying the recordkeeping system to the extent consistent with the statutory mandate. To achieve this purpose, the final rule carries forward the key elements of the existing recordkeeping scheme, with changes designed to improve efficiency, equity, and flexibility while reducing, to the extent practicable, the economic burden on individual establishments. The central requirements in the final rule may be summarized as follows: All non-exempt employers must record all work-related, significant injuries and illnesses. As discussed below, OSHA's approach to each of these elements -- the scope of the exemptions from recording requirements, the meaning of "work-relationship," and the criteria for determining whether an injury or illness is "significant" -- is reasonable and directly related to the statutory language and purpose.

The final rule contains two categories of exemptions that, together, relieve most employers of the obligation routinely to record injuries and illnesses sustained by their employees. Section 1904.1 contains a "very small-employer" exemption: Employers need not record injuries or illnesses in the current year if they had 10 or fewer employees at all times during the previous year, unless required to do so pursuant to Sections 1904.41 or 1904.42. Section 1904.2 contains a "low-hazard industry" exemption: Individual business establishments are not required to keep records if they are classified in specific low-hazard retail, service, finance, insurance, or real estate industries.

a. The size-based exemption. Section 8(d) of the Act expresses Congress' intent to minimize, where feasible, the burden of recordkeeping requirements on employers, particularly small businesses: "Any information obtained by the Secretary, the Secretary of [HHS], or a State agency under this Act shall be obtained with a minimum burden upon employers, especially those operating small businesses. Unnecessary duplication of efforts in obtaining information shall be reduced to the maximum extent feasible." 29 U.S.C. 657(d).

Since 1972, the Secretary has exempted very small businesses from most recordkeeping requirements. On October 4, 1972, OSHA issued a provision, codified at 29 CFR 1904.15(a), exempting employers from routine injury and illness reporting requirements for the current year if they had no more than seven employees during the previous year. The exemption did not relieve these businesses from the obligation to report fatality and multiple hospitalization incidents to OSHA and to participate in the BLS annual survey when selected to do so. 37 FR 20823 (October 4, 1972). In 1977, the Secretary amended section 1904.15 to make it applicable to businesses having ten or fewer employees during the year preceding the current reporting year. 42 FR 38568 (July 29, 1977). As support, the amendment cited the Department of Labor appropriations acts for fiscal years 1975 and 1976, which exempted employers having ten or fewer employees from most routine recordkeeping requirements, and Section 8(d) of the Act. Id. The Secretary determined that the amendment appropriately balanced the interest of very small businesses while preserving the essential purposes of the recordkeeping scheme:

The [exemption] has been carefully designed to carry out the mandate of section 8(d) without impairing the Act's basic purpose. Thus, the [exemption] will not diminish the protections afforded employees under the Act because all employers * * * remain subject to the enforcement provisions of the Act. The [exemption] will continue to require * * * small employers * * * to report fatalities and multiple hospitalizations and to participate in the BLS annual survey when selected to do so.

In the present rulemaking, the Secretary proposed to enlarge the scope of the exemption to include employers, in industries other than construction, having 19 or fewer employees during the entire previous calendar year. 61 FR 4057 (February 2, 1996). At the same time, the proposal asked for public comment on whether "the small employer partial exemption [should] remain the same, be eliminated, or be expanded?" 61 FR 4043. In reaching a final decision on this matter, the Secretary resolved two interrelated questions. First, she determined that there is no sound basis for departing from OSHA's prior interpretation that the Act permits a carefully crafted exemption for very small employers. Second, she determined that limiting the exemption to employers with ten or fewer employees effectuates Congress' intent with the minimum degree of impairment to the overall recordkeeping scheme. The first question is essentially one of statutory construction, and is therefore considered below. The second question calls for an analysis of the record and is addressed in the preamble explanation for section 1904.1 of the final rule.

It is a fundamental principle of administrative law that an agency which chooses to reverse a previously held position must supply a "reasoned analysis" of its decision. Motor Vehicle Mfgrs Assn. v. State Farm Mutual Automobile Insurance Co., 463 U.S. 27, 42 (1983). After careful consideration, the Secretary finds no persuasive basis for eliminating the small-employer exemption in this rule. As a threshold matter, nothing has changed the agency's long-held view that section 8(d) permits a carefully tailored exemption from recordkeeping requirements for very small businesses. 42 FR 10016 (February 18, 1977). This interpretation is consistent with the literal wording of the statute and is further confirmed by the provisions in the Department's appropriations acts for FY 1975 and 1976, exempting employers with ten or fewer employees from routine recordkeeping and reporting requirements. See 42 FR 5356 (January 28, 1977) (noting restriction in FY 1975 and 1976 appropriations acts and stating OSHA would continue to treat firms of up to 10 employees as exempt pending permanent change in the regulations to expand the small-employer exemption).

OSHA also concludes that a very small business exemption limited to the routine recording and reporting of non-fatal injuries and illnesses will not seriously undermine the recordkeeping system. OSHA explained in Section I. of the preamble that there are three primary purposes for recordkeeping and reporting requirements. First, the records are the foundation for national statistics published by the BLS on the number and rate of workplace injuries and illnesses, as well as their source, nature and type. Second, the records provide information useful to employers and employees in their efforts voluntarily to locate and eliminate workplace safety and health hazards. Finally, the records are useful to OSHA in targeting its enforcement efforts and in efficiently conducting its safety and health inspections.

Exempting very small businesses from routine recordkeeping will not significantly compromise these goals. The exemption has no effect upon the obligation of these businesses to participate in the national statistical survey administered by the BLS. See the discussion of § 1904.42 in Section V. Summary and Explanation. If a small business is selected for participation in the survey, it must keep a log of injuries and illnesses and make reports as required by the BLS. Id. Thus, even the smallest firms continue to be represented in the national injury and illness statistics.

The second purpose is not seriously compromised by the exemption because injury and illness records are less necessary as an aid to voluntary compliance efforts by very small employers and their employees than they are for larger employers. OSHA's experience is that, in establishments with only a few employees, management and production personnel typically work in close concert. Because of their size, such establishments also tend to record fewer occupational injuries and illnesses. Accordingly, in very small firms, managers are likely to have first-hand knowledge of those occupational injuries and illnesses that occur in their workplaces. By the same token, it is reasonable to believe that employees in very small firms are generally aware of the injuries that occur in their workplaces and do not rely heavily upon access to employer records to inform themselves about occupational hazards. In short, review and analysis of injury and illness records by very small business employers, or by their employees, may not be required for awareness of workplace conditions.

Finally, routine injury and illness records are of limited usefulness to OSHA in targeting and conducting inspections. Many OSHA inspections are conducted in response to a specific complaint or referral alleging unsafe conditions, or in response to a workplace catastrophe or fatality. A large number of inspections are also conducted under special emphasis programs at the national and local level. The remaining inspections are conducted at specific worksites in the construction industry and in other non-construction industries selected under a planned schedule. Construction inspections are selected using an econometric model that predicts the best time to conduct an inspection at a specific construction project. The general industry scheduled inspections are targeted primarily toward employers with extremely high rates of occupational injury and illness, using data supplied by employers to the OSHA Data Initiative (ODI) under the requirements of former section 1904.17, Annual OSHA Injury and Illness Survey of Ten or More Employers (now section 1904.41). Due to budget, paperwork burden and logistical constraints, OSHA collects data only from employers in high hazard industries, and has generally not collected data from employers with fewer than 40 workers.

OSHA is also prohibited from conducting scheduled inspections of employers with 10 or fewer employees in low hazard industries by an annual rider on OSHA's appropriations bills which has been renewed annually for many years. Thus, OSHA does not collect data from very small employers, and they are excluded from the general industry scheduled inspection program. Because very small firms have been wholly excluded from the general schedule inspection program, the routine injury and illness records of very small businesses have been of little use to OSHA in targeting inspections. Should OSHA wish to include very smaller employers in a special emphasis inspection program or other initiative, the agency may require any business, regardless of its size, to keep records and make reports as necessary. See 29 CFR 1904.41.

OSHA also finds that access to the Log and Incident Report would be of little value to compliance officers in conducting inspections of very small businesses initiated by a complaint or report of a fatality or an accident resulting in multiple hospitalizations. OSHA has long acknowledged that while injury and illness records are frequently useful in identifying hazardous areas or operations within larger establishments subject to programmed inspections, they are significantly less important in the conduct of inspections in the smallest businesses. As OSHA has stated, "experience has shown that when dealing with small employers, the injury and illness records * * * are normally not needed by the CSHO to locate hazards during an inspection. In those cases where log information may be needed, the CSHO can easily obtain the information by interviewing the employees." 42 FR 10016 (February 18, 1977). See also 47 FR 57699, 5700 (December 28, 1982) (in conducting complaint or fatality inspections, the hazard information is usually provided by the complaint itself, or through prompt investigation.) For these reasons, the Secretary believes that an exemption for very small employers, reasonably tailored to the purposes served by recordkeeping requirements, is appropriate.

b. The hazard-based exemption. Since 1982, OSHA has exempted from routine recordkeeping requirements certain industries classified in OMB's Standard Industrial Classification (SIC) Manual. The 1982 exemption was limited to establishments in SIC Industry Groups that (1) were not subject to general schedule inspections, and (2) had average lost workday case injury rates, as published by the BLS, at or below 75% of the national average. In 1982, the industry groups that met these criteria were those classified as retail trade, finance, insurance, real estate, and services -- SIC codes 52-89, excluding 52-54, 70, 75, 76, 79, and 80. 47 FR 57699-57,700 (December 28, 1982).

The purpose of the exemption "was to further OSHA's continuing effort under section 8(d) of the Act to reduce the paperwork burden on employers without compromising worker safety and health." 47 FR 57700. Exempting low-hazard industries from routine record-keeping was justified, OSHA explained, for the same reasons that warranted exempting very small businesses. Injury and illness records from establishments in the affected SIC codes were not of significant benefit to OSHA because these industry groups were not then targeted for general schedule inspections. Id. The records were not a significant source of information for employers and employees because BLS data showed that approximately 94% of all establishments in the affected industry groups could be expected to have fewer than two injuries per establishment on an annual basis. Id. Finally, the exemption would not affect the reliability of safety and health statistics because the affected establishments would continue to participate in the BLS annual survey of occupational injuries and illnesses. Id.

OSHA continues to believe that a properly tailored exemption for low-hazard industries is appropriate. Congress intended in section 8(d) to minimize the recordkeeping burden on all employers, not only small businesses. Exempting from routine injury and illness reporting requirements those employers whose records are unlikely to be of significant benefit to OSHA, or to the employers and their employees, serves this important interest. However, OSHA recognizes that the balance between the interest of minimizing recordkeeping burdens and that of ensuring accurate, reliable and useful information is a delicate one. In the final rule, OSHA has substantially revised the list of exempt low-hazard industries based upon more reliable three-digit industry classification data. See the discussion of § 1904.1, in the following Summary and Explanation. With these changes, OSHA believes that the rule strikes the appropriate balance.

Section 8 of the Act directs the Secretary to prescribe regulations requiring employers to "maintain accurate records of * * * work-related deaths injuries and illnesses [of a non-minor nature]. 29 U.S.C. 657(c)(2). The definition of work-relationship in section 1904.5 of the final rule is consistent, in all but one respect, with the definition in the Guidelines to the former rule. The final rule states that an injury or illness is work-related "if an event or exposure in the work environment either caused or contributed to [it] or significantly aggravated a pre-existing injury or illness. Work-relatedness is presumed for injuries and illnesses resulting from events or exposures occurring in the work environment, unless an exception listed in section 1904.5(b)(2) specifically applies" (emphasis added).

The Guidelines state that, "[i]f an event * * * occurred in the work environment that caused or contributed to the injury", the case would be recordable, assuming it meets the other requirements for recordability. Ex. 2 at p. 32 (original emphasis). Further instructions in the Guidelines provided that:

The general rule is that all injuries and illnesses which result from events or exposures occurring to employees on the employer's premises are presumed to be work related. This presumption is rebuttable. * * * However, the nature of the activity which the employee is engaged in at the time of the event or exposure, the degree of employer control over the employee's activity, the preventability of the incident, or the concept of fault do not affect the determination.

Ex. 2 at p. 34 (original emphasis). The only significant difference between the final rule and the former rule is that the final rule requires that work "significantly" aggravate a pre-existing injury or illness before the case is recordable.

OSHA's approach to work-relationship in both the former and the final recordkeeping rules reflects two important principles. The first is that work need only be a causal factor for an injury or illness to be work-related. The rule requires neither precise quantification of the occupational cause, nor an assessment of the relative weight of occupational and non-occupational causal factors. If work is a tangible, discernible causal factor, the injury or illness is work-related. The second principle is that a "geographic presumption" applies for injuries and illnesses caused by events or exposures that occur in the work environment. These injuries and illnesses must be considered work-related unless an exception to the presumption specifically applies.

The final rule's geographic presumption reflects a theory of causation similar to that applied by courts in some workers' compensation cases. Under the "positional-risk" test, an injury may be found to "arise out of" employment for compensation purposes if it would not have occurred but for the fact that the conditions and obligations of employment placed the claimant in the position where he or she was injured. See 1 Larson's Workers' Compensation Law section 6.50 (1977). Accord, Odyssey/Americare of Oklahoma v. Worden, 948 P.2d 309, 311 (Okla. 1997). Under this "but for" approach to work-relationship, it is not necessary that the injury or illness result from conditions, activities or hazards that are uniquely occupational in nature. Accordingly, the presumption encompasses cases in which an injury or illness results from an event at work that is outside the employer's control, such as a lightning strike, or involves activities that occur at work but that are not directly productive, such as horseplay.

The proposed rule asked for comment on whether OSHA should abandon its historic approach and adopt a new test for determining work-relationship. 61 FR 4044, 4045. The proposal outlined three alternative tests in which the determination of work-relationship turned on the degree to which the injury or illness was linked to occupational causes, as compared with personal factors such as off-the job activities, aging, or pre-existing medical conditions. Two of these alternative tests required evidence of a high degree of work causation to establish work-relationship. Alternative 1 required that occupational factors be the "sole cause" of the injury or illness; any evidence of non-work related causal factors was sufficient to exclude the case. Alternative 2 required that occupational factors be the "predominant cause" before the case could be considered work-related. See 61 FR 4044. Some commenters suggested a modification to Alternative 2 that would have involved substitution of the word "substantial" or "significant" for "predominant."

The third alternative test was significantly more expansive than that adopted in the final rule. Under Alternative 3, an injury or illness would be considered work-related if the work environment had any possibility of playing a causal role. 61 FR 4044.

Some commenters favored a somewhat different test for work-relationship that focused on the nature of the injury-causing event in the workplace. This test would include in the OSHA records only those cases resulting from uniquely occupational or job-related activities or processes. Supporters of this approach argued that it would exclude injuries and illnesses caused by factors at work that are unrelated to production tasks, or that are unpreventable by the employer's safety and health program.

After careful consideration of the record, OSHA believes that the final rule's test for work-relationship is both more consistent with the Act's purpose and more practical than the "quantified occupational cause" tests or the "unique occupational conditions" test. The language of the statute itself indicates that Congress did not intend to give "work-related" a narrow or technical meaning, but rather sought to cover a variety of causal relationships that might exist in workplaces. Section 2 of the Act addresses injuries and illnesses arising out of "work situations." Sections 2(b)(1), 2(b)(2), and 2(b)(4) refer to "places of employment," and to the achievement of safe and healthful "working conditions." Section 2(b)(7) seeks to assure that no employee will suffer diminished health or life expectancy as a result of his "work experience." Section 2(b)(12) states that one of the Act's purposes is to provide for reporting procedures which "accurately describe the nature of the occupational safety and health problem." Section 2(b)(13) encourages joint labor-management efforts to reduce injuries and disease "arising out of employment."

This conclusion is further supported by the Act's stated purpose to promote research into the causes and prevention of occupational injuries and illnesses. Section 2 of the Act establishes Congress' intent to improve occupational safety and health, inter alia, by:

Providing for research in the field of occupational safety and health, including the psychological factors involved, and by developing innovative methods, techniques and approaches for dealing with occupational safety and health problems. 29 U.S.C. § 651(b)(5)

[E]xploring ways to discover latent diseases, establishing causal connections between diseases and work in environmental conditions, and conducting other research relating to health problems. * * * 29 U.S.C. § 651(b)(6).

Providing for appropriate reporting procedures with respect to occupational safety and health which will help achieve the objectives of this Act and accurately describe the nature of the occupational safety and health problems. 29 U.S.C. § 651(b)(12).

The legislative history of the Act demonstrates Congress' awareness of the importance of developing information for future scientific use. The Committee Report accompanying the Senate bill reported to the floor noted that,

[i]n the field of occupational health, the view is particularly bleak, and due to the lack of information and records, may well be considerably worse than we currently know. * * * Recent scientific knowledge points to hitherto unsuspected cause-and-effect relationships between occupational exposures and many of the so-called chronic diseases -- cancer, respiratory ailments, allergies, heart disease, and others. In some instances, the relationship appears to be direct: asbestos, ionizing radiation, chromates, and certain dye intermediaries, among others, are directly involved in the genesis of cancer. In other cases, occupational exposures are implicated as contributory factors. The distinction between occupational and non-occupational illnesses is growing increasingly difficult to define.

S. Rep. No. 1282, 91st Cong., 2d Sess. 2 (1970), reprinted in Subcommittee on Labor of the Senate Committee on Labor and Public Welfare, Legislative History of the Occupational Safety and Health Act of 1970 (Committee Print 1971) at 142 (Leg. Hist.). With this background in mind, the committee stated that it "expects the Secretary of Labor and the Secretary of [HHS] will make every effort through the authority to issue regulations and other means, to obtain complete data regarding the occurrence of illnesses, including those resulting from occupational exposure which may not be manifested until after the termination of such exposure." Leg. Hist. at 157.

Both the Senate and the House Committees expressed concern that the statute not be interpreted in a way that would result in under-reporting of injuries and illnesses. The Senate report states:

The committee recognizes that some work-related injuries or ailments may involve only a minimal loss of work time or perhaps none at all, and may not be of sufficient significance to the Government to require their being recorded or reported. However, the committee was also unwilling to adopt statutory language which, in practice might result in under-reporting. The committee believes that records and reports prescribed by the Secretary should include such occurrences as work-related injuries and illnesses requiring medical treatment or restriction or reassignment of work activity, as well as work-related loss of consciousness.

Leg. Hist. at 157. The House Report similarly noted that while some injuries and illnesses might not be of enough value to require recordation, "the greater peril" lay in allowing under reporting. Leg. Hist. at 860. Therefore, the report added, "[the] language 'all work-related injuries, [and illnesses]' should be treated as a minimum floor. * * *"

In light of these purposes, it is apparent that Congress did not, in Section 8, mean to limit recordable "work-related" injuries and illnesses only to those caused primarily or substantially by work. It is evident from the statute that Congress wanted employers to keep accurate records of non-minor injuries and illnesses, in part, to serve as a basis for research on the causes and prevention of industrial accidents and diseases. This research is needed, among other reasons, to further examine and understand those occupational factors implicated as contributory causes in injuries and diseases. To serve this purpose, the records should include cases in which there is a tangible connection between work and an injury or illness, even if the causal effect cannot be precisely quantified, or weighed against non-occupational factors.

The first two alternative quantification theories outlined in the preamble would exclude important information from the records. These theories would eliminate cases in which the work environment is believed to have played a definite role in the accident or the onset of disease, but not enough is known to quantify the effect of work factors or to assess the relative contribution of work and non-work factors. However, the information provided by cases having a tangible, yet unquantifiable, connection with the work environment is useful to employers, employees and researchers and thus serves the recordkeeping purposes envisioned by Congress.

On the other hand, the third alternative theory in the proposal would sweep too broadly. A work-relationship test that is met if work has "any possibility of playing a role in the case" would include virtually every injury or illness occurring in the work environment. 61 Fed. Reg. 4044. Recording cases in which the causal connection to work is so vague and indefinite as to exist only in theory would not meaningfully advance research, or serve the other purposes for requiring recordkeeping. For these reasons, OSHA has rejected the three alternative theories outlined in the proposal.

The "unique occupational activity" test, which some commenters favored instead of the geographic presumption, would limit recorded injuries and illnesses to those caused by an activity or process peculiarly occupational in nature. Supporters of this approach identified several types of cases that would be work-related under the geographic presumption, but not recordable under an activities-based approach. These include cases in which the injury or illness was not caused by the physical forces or hazards unique to industrial processes, cases in which the employee was not injured while performing an activity or task directly related to production, and cases in which the injury or illness was not preventable by the employer.

The "unique occupational activity" test is unsuitable for essentially the same reasons that militate against the first two alternatives described in the proposal. The statutory language and purpose do not reflect a Congressional intent to limit recording only to those cases resulting from uniquely occupational hazards or activities. Rather, the statute shows that Congress knew that employees were being injured and made ill in a variety of ways and under a variety of circumstances, and wanted employers to record all cases causally related to the work environment. The "but-for" theory underlying the geographic presumption is a widely accepted legal test for causation and is consistent with the statutory language and purpose.

The "unique occupational activities" test, like the "quantification" tests, would likely result in exclusion of important information from the records. An activity-based test for work-relationship could obscure the role of factors in the work environment not directly linked to production, such as violence perpetrated by employees and others or tuberculosis outbreaks. In addition, the precise causal mechanism by which an employee has been injured or made ill at work may not be known at the time of the accident, or may be misunderstood. To serve the statute's research purposes, the records must reflect not only those injuries and illnesses for which the precise causal mechanism is apparent at the time of recordation, but also those for which the mechanism is imperfectly understood. The alternative approaches to work-relationship would severely limit the usefulness of injury and illness data for research purposes, particularly research to uncover latent patterns of health impairment and disease and to establish causal connections between diseases and exposure to particular hazards.

The Occupational Safety and Health Review Commission has affirmed the approach to work-relationship taken in the former rule. General Motors Corp., Inland Div., 8 O.S.H. Cas. (BNA) 2036, 2039-2040 (August 29, 1980). The issue in General Motors was whether the employer was required to record respiratory ailments of three employees, based on notations from the employees' treating physicians that their ailments were probably related to exposure to a chemical substance at work. The Commission rejected the employer's argument that the recordkeeping rule required recording only of illnesses directly caused by work activities, stating:

To accept Respondent's interpretation would impose a static view of scientific knowledge. Only illnesses in which the known cause was the occupational environment would be recorded. Unknown medical correlations between disease and the workplace would be obscured by this inadequate recording obligation. Under this interpretation of the statute and regulations, OSHA and NIOSH would be significantly restrained from fulfilling their statutory obligation of making the workplace healthier. * * * [T]he primary purpose of the recording obligation is to develop information for future scientific use.

8 O.S.H. Cas. at 2040. Accordingly, OSHA believes that there is a sound legal basis for the definition of work-relationship in the final rule.

There are also sound policy justifications. The approach to "work-relationship" adopted in the final rule is more cost-effective than the alternative approaches and will result in more accurate injury and illness data. OSHA expects that for each reported injury or illness, employers generally will be able to apply the geographic presumption more easily and quickly than a test requiring an assessment of the relative contribution of employment and personal causes. The incremental reduction in the time necessary to complete each entry, when multiplied by the total number of entries per year, will result in a substantial cumulative saving in paperwork burden in comparison to the burden that would be imposed by the alternatives.

The geographic presumption will also produce more consistent and accurate reporting. OSHA believes that it would be difficult to measure the precise degree to which personal and occupational factors cause accidents or illnesses. Accordingly, any test requiring that job duties or tasks be "significant" or "predominant" causative factors would necessarily involve a high degree of subjective judgment. There is likely to be substantial inconsistency, both in the treatment of successive, similar cases by the same employer, and in the treatment of such cases among different employers. Moreover, such a test would fail to capture cases in which the workplace contribution to an injury or illness was imperfectly known or misunderstood at the time the case was reported. Recording all cases caused by events or exposures at work, with only limited exceptions, produces data that enables OSHA, employers and others to better understand the causal relationships present in the work environment. Although OSHA has not adopted a test for determining significant contribution by work, the final rule does include provisions to make sure that workplace aggravation of a pre-existing injury must be significant before work relationship is established (see discussion of 1904.5(b)(4)).

A number of commenters argued that because OSHA's mission is to eliminate preventable occupational injuries and illnesses, the determination of work-relatedness must turn upon whether the case could have been prevented by the employer's safety and health program. Dow expressed this view as follows:

[T]he goal of this recordkeeping system should be to accurately measure the effectiveness of safety and health programs in the workplace. Activities where safety and health programs could have no impact on preventing or mitigating the condition should not be logged and included in the Log and Summary nor used by OSHA to determine its inspection schedule. If the event was caused by something beyond the employer's control, it should not be considered a recordable event that calls into question a facility's safety and health program. * * * Credibility in this regulation rests on whether the recorded data accurately reflects the safety and health of the workplace. Including events where the workplace had virtually no involvement undermines the credibility of the system and results in continued resistance to this regulation.

Ex. 15-335B. The law firm of Constangy, Brooks and Smith, LLC, urged OSHA to adopt the second alternative definition in the proposal because cases that are "predominantly caused by workplace conditions" are the ones most likely to be preventable by workplace controls. They stated, "[s]ince OSHA's ultimate mission is the prevention of workplace injuries and illnesses, it is reasonably necessary to require recording only when the injury or illness can be prevented by the employer." Ex. 15-345.

OSHA believes that these comments reflect too narrow a reading of the purposes served by injury and illness records. Certainly one important purpose for recordkeeping requirements is to enable employers, employees and OSHA to identify hazards that can be prevented by compliance with existing standards or recognized safety practices. However, the records serve other purposes as well, including facilitating the research necessary to support new occupational safety and health standards and to better understand causal connections between the work environment and the injuries and illnesses sustained by employees. As discussed above, these purposes militate in favor of a general presumption of work-relationship for injuries and illnesses that result from events or exposures at the worksite, with exceptions for specific types of cases that can be safely excluded without significantly impairing the usefulness of the database.

Section 1904.7 of the final rule sets forth the criteria to be used by employers in determining whether work-related occupational injuries and illnesses are significant, and therefore recordable. Under § 1904.7, a work-related injury or illness is significant for recordkeeping purposes if it results in any of the following: death, days away from work, restricted work or transfer to another job, medical treatment beyond first aid, or loss of consciousness. Employers must also record any significant injury or illness diagnosed by a physician or other licensed health care professional even if it does not does not result in the one of the listed outcomes. OSHA's definition of a "significant" injury or illness in this context is based on two key principles discussed below. The first is that the requirement for recording only significant cases applies equally to "injuries" and "illnesses" for recordkeeping purposes. The second principle is that the criteria expressly mentioned in the Act, such as death, loss of consciousness or restriction of work, are mandatory but not exclusive indicia of significance; any significant injury or illness diagnosed by a physician or other licensed health care professional must also be recorded. These two principles are addressed below, while the definitions applicable to the specific criteria themselves, and related evidentiary issues, are discussed in the preamble explanation for section 1904.7.

a. The significant case requirement applies equally to injuries and illnesses; employers are no longer to report insignificant illnesses. OSHA distinguishes between injuries and illnesses based on the nature of the precipitating event or exposure. Cases which result from instantaneous events are generally considered injuries, while cases which result from non-instantaneous events, such as a latent disease or cumulative trauma disorder, are considered illnesses. Id.

Under the former recordkeeping regulations, occupational injuries had to be recorded if they were non-minor in nature; that is, if they resulted in loss of consciousness, or required medical treatment, time off work, restriction of work, lost time, or transfer to another job. 61 FR 4036. However, all occupational illnesses had to be reported, regardless of severity. Id. This difference in the severity threshold for recording injuries and illnesses had, in the past, been based upon the particular phrasing of section 8(c)(2) of the Act:

The Secretary * * * shall prescribe regulations requiring employers to maintain accurate records of, and to make periodic reports on, work-related deaths, injuries and illnesses, other than minor injuries requiring only first aid treatment and which do not involve medical treatment, loss of consciousness, restriction of work or motion, or transfer to another job."

29 U.S.C. 657(c)(2). Because the severity criteria appear in the clause defining "minor injuries," OSHA had construed the section to require recordation of all work-related illnesses, even those that do not meet the severity characteristics expressly applicable to "injuries."

OSHA has reconsidered its position in this rulemaking, and has concluded that the former rule was inappropriate in several respects. First, although the severity characteristics listed in section 8(c)(2) of the Act apply expressly to "injuries," the Act contains persuasive indications that Congress also meant to require recordation only of "significant" illnesses, as determined by reasonable criteria. Section 24(a) states that "[t]he Secretary shall compile accurate statistics on work injuries and illnesses which shall include all disabling, serious, or significant injuries and illnesses * * * other than minor injuries requiring only first aid treatment and which do not involve medical treatment * * * ." 29 U.S.C. 673 (a). The legislative history also supports this view. The statement of the House managers on the resolution of conflicting House and Senate bills states that:

A Senate bill provision without a counterpart in the House amendment permitted the Secretary to require an employer to keep records and make reports on "all work-related deaths, injuries and illnesses." The House receded with an amendment limiting the reporting requirement to injuries and illnesses other than of a minor nature, with a specific definition of what is not of a minor nature.

Leg. Hist. at 1190 (emphasis supplied). The former rule did not appropriately implement this intent. In the first place, OSHA's prior interpretation that section 8(c)(2) limits the applicability of the listed severity criteria only to injuries does not necessarily mean that illnesses must be recorded without regard to their significance. As a textual matter, such a reading simply leaves open the question of what, if any, severity criteria apply to illnesses.

OSHA believes that the Act does not support a different severity threshold for injuries than for illnesses. OSHA is now persuaded that its prior reading of section 8(c)(2) placed too much emphasis on the fact that the severity criteria modify the word "injuries" in the clause, "other than minor injuries requiring only first aid treatment and which do not involve medical treatment, loss of consciousness, restriction of work or motion or transfer to another job." 29 U.S.C. 657(c)(2). Congress' failure to list specific severity criteria for illnesses, as it did for injuries, does not, in itself, compel the inference that two different sets of criteria must apply. Congress meant to limit recordation to significant injuries and illnesses alike, and absent strong indications to the contrary, it is reasonable to presume that Congress meant the same severity threshold to apply to both conditions.

In addition, there are strong policy reasons for avoiding a distinction between injuries and illnesses based on severity. OSHA explained in the proposal that the current distinction between injuries and illnesses based on the nature of the precipitating event has caused some degree of confusion and uncertainty. Using one set of criteria for severity means that employers will not have to decide whether a case is an injury or an illness in determining its recordability. This simplifies the recordkeeping system, resulting in more accurate injury and illness data while reducing the recordkeeping burden for employers who are required to maintain records (61 FR 4036). Employers will continue to classify each recordable case as either an injury or an illness on the OSHA 300 Log, but the decision no longer has any effect on whether or not the case must be recorded.

b. The criteria listed in the Act are mandatory but not exclusive indicia of significance. A final issue relating to significance is the effect to be given a finding that an injury or illness results in, or does not result in, one of the outcomes listed in the statute: death, days away from work, restricted work or transfer to another job, medical treatment beyond first aid, or loss of consciousness. The implication arising from the wording of section 8(c)(2) and section 24 is that if an injury or illness results in one of the listed outcomes, it must be deemed significant for recordkeeping purposes. This position, which reflects OSHA's longstanding, consistent interpretation of the statute, was not seriously questioned in the rulemaking. Accordingly, the final rule requires that a work-related injury or illness be recorded if it results in one of the outcomes mentioned in the statute.

The final rule also requires that a case be recorded, whether or not it results in one of the listed outcomes, if it involves a significant injury or illness diagnosed by a physician or other licensed health care professional. 29 CFR 1904.10(b). Nothing in the statute compels the conclusion that the criteria mentioned in sections 8 and 24 are the exclusive indicia of severity for recordkeeping purposes. Congress directed the Secretary to collect data on "all disabling, serious, or significant injuries and illnesses, whether or not involving loss of time from work," other than minor injuries * * * which [do not result in one of the listed outcomes]. 29 U.S.C. 673(a). A reasonable reading of this language is that while an injury that meets one of the listed criteria is non-minor and must be recorded, the converse does not necessarily follow. An injury or illness may reasonably be viewed as significant, and therefore recordable, even if it is not immediately followed by death, loss of consciousness, or job-related disability. For example, an employee diagnosed with an unquestionably serious work-related disease, such as asbestosis or mesothelioma, may forego or postpone medical treatment and continue temporarily to perform his or her normal job duties. Focusing exclusively on the basic criteria listed in the statute in cases such as these could result in underrecording of serious cases. Accordingly, the final rule requires employers to record any significant injury or illness that is diagnosed. A thorough discussion of this requirement, including a definition of what constitutes a "significant" injury or illness for this purpose, is contained in the preamble discussion of section 1904.7.

Because the provisions of the final recordkeeping rule, as explained above and in the subsequent sections of this preamble, are reasonably related to the statutory purposes, the Secretary finds that the rule is necessary to carry out her responsibilities under the Act. The rule is therefore a valid exercise of the Secretary's general rulemaking authority under Section 8. Cf. Mourning v. Family Publications Services, 411 U.S. 356.

The following sections discuss the contents of the final 29 CFR Part 1904 and section 1952.4 regulations. OSHA has written these regulations using the plain language guidance set out in a Presidential Memo to the heads of executive departments and agencies on June 1, 1998. The Agency also used guidance from the Plain Language Action Network (PLAN), which is a government-wide group working to improve communications from the Federal government to the public, with the goals of increasing trust in government, reducing government costs, and reducing the burden on the public. For more information on PLAN, see their Internet site at http://www.plainlanguage.gov/.

The plain language concepts encourage government agencies to adopt a first person question and answer format, which OSHA used for the Part 1904 rule. The rule contains several types of provisions. Requirements are described using the "you must * * *" construction, prohibitions are described using "you may not * * *", and optional actions that are not requirements or prohibitions are preceded by "you may * * *." OSHA has also included provisions to provide information to the public in the rule.

The Purpose section of the final rule explains why OSHA is promulgating this rule. The Purpose section contains no regulatory requirements and is intended merely to provide information. A Note to this section informs employers and employees that recording a case on the OSHA recordkeeping forms does not indicate either that the employer or the employee was at fault in the incident or that an OSHA rule has been violated. Recording an injury or illness on the Log also does not, in and of itself, indicate that the case qualifies for workers' compensation or other benefits. Although any specific work-related injury or illness may involve some or all of these factors, the record made of that injury or illness on the OSHA recordkeeping forms only shows three things: (1) that an injury or illness has occurred; (2) that the employer has determined that the case is work-related (using OSHA's definition of that term); and (3) that the case is non-minor, i.e., that it meets one or more of the OSHA injury and illness recording criteria. OSHA has added the Note to this first subpart of the rule because employers and employees have frequently requested clarification on these points.

The following paragraphs describe the changes OSHA has made to the Purpose provisions in Subpart A of the final rule, and discusses the Agency's reasons for these changes. Proposed section 1904.1 of Subpart A contained three separate paragraphs. Proposed paragraph (a) stated that the purpose of the recordkeeping rule (Part 1904) was "to require employers to record and report work-related injuries, illness and fatalities." It also described several ways in which such records were useful to employers, employees, OSHA officials, and researchers evaluating and identifying occupational safety and health issues.

Proposed paragraph (b) noted that the recording of a job-related injury, illness or fatality did not necessarily impute fault to the employer or the employee, did not necessarily mean that an OSHA rule had been violated when the incident occurred, and did not mean that the case was one for which workers' compensation or any other insurance-related benefit was appropriate. The third paragraph in proposed section 1904.1, proposed paragraph (c), stated that the regulations in Part 1904 had been developed "in consultation with the Secretary of Health and Human Services" (HHS), as required by Section 24(a) of the Act.

In the final rule, OSHA has moved much of this material, which was explanatory in nature, from the regulatory text to the preamble. This move has simplified and clarified the regulatory text. The final rule's Purpose paragraph simply states that: "The purpose of this rule (Part 1904) is to require employers to record and report work-related fatalities, injuries and illnesses." This final rule statement is essentially identical to the first sentence of the proposed Purpose section. It clearly and succinctly states OSHA's reasons for issuing the final rule.

A number of commenters (see, e.g., Exs. 25; 15: 199, 305, 313, 346, 348, 352, 353, 375, 418, 420) specifically addressed proposed section 1904.1. The principal points raised by these commenters concerned: (1) Statements in proposed paragraph (a) about the quality of the data captured by the records; (2) proposed paragraph (b)'s discussion of the relationship between OSHA recordkeeping and employer/employee fault, violations of OSHA rules, and the workers' compensation system, and (3) the statement in proposed paragraph (c) that discussed OSHA's consultation with the Secretary of Health and Human Services in developing this rule. Each of these issues is discussed in detail below.

Most comments on proposed paragraph (a) took issue with the language that OSHA used to describe the statistical use of the records (see, e.g., Exs. 25, 15: 305, 346, 348, 375, 420). Typical of these comments is one from the National Association of Manufacturers: "We urge OSHA to remove the following unverified and conclusory statement from § 1904.1(a): "The records: * * * accurately describe the nature of occupational safety and health problems for the Nation, State or establishment" (Exs. 25, 15: 305). OSHA did not intend this statement to attest with certainty to the validity of national occupational statistics. Proposed section 1904.1(a) merely paraphrased section 2(b) of the Act, which states that such records "will help achieve the objectives of this Act and accurately describe the nature of the occupational safety and health problem." In response to commenters, OSHA has simplified the final rule by deleting the proposed listing of the functions of the records required by this rule.

As discussed earlier, proposed paragraph (b) stated that the recording of a case did not "necessarily mean that the employer or employee was at fault, that an OSHA standard was violated, or that the employee is eligible for workers' compensation or other insurance benefits." The last sentence of proposed paragraph (b) described the various types of workplace events or exposures that may lead to a recordable injury or illness.

A number of commenters agreed with the proposed statements on fault, compliance, and the relationship between the recording of a case and workers' compensation or other insurance (see, e.g., Exs. 25, 15: 305, 346, 420). Employers have frequently asked OSHA to explain the relationship between workers' compensation reporting systems and the OSHA injury and illness recording and reporting requirements. As NYNEX (Ex. 15: 199) noted,

[t]he issue of confusion between OSHA recordkeeping and workers' compensation/insurance requirements cannot be totally eliminated as the workers' compensation criteria vary somewhat from state to state. There will always be some differences between OSHA recordability and compensable injuries and illnesses. The potential consequences of these differences can be minimized, however, if all stakeholders in the recordkeeping process (i.e., employers, employees, labor unions, OSHA compliance officials) are well informed that OSHA recordability does not equate to compensation eligibility. This can be facilitated by printed reminders on all of the OSHA recordkeeping documents (e.g., forms, instructions, pamphlets, compliance directives, etc.).

As NYNEX observed, employers must document work-related injuries and illnesses for both OSHA recordkeeping and workers' compensation purposes. Many cases that are recorded in the OSHA system are also compensable under the State workers' compensation system, but many others are not. However, the two systems have different purposes and scopes. The OSHA recordkeeping system is intended to collect, compile and analyze uniform and consistent nationwide data on occupational injuries and illnesses. The workers' compensation system, in contrast, is not designed primarily to generate and collect data but is intended primarily to provide medical coverage and compensation for workers who are killed, injured or made ill at work, and varies in coverage from one State to another.

Although the cases captured by the OSHA system and workers' compensation sometimes overlap, they often do not. For example, many injuries and illnesses covered by workers' compensation are not required to be recorded in the OSHA records. Such a situation would arise, for example, if an employee were injured on the job, sent to a hospital emergency room, and was examined and x-rayed by a physician, but was then told that the injury was minor and required no treatment. In this case, the employee's medical bills would be covered by workers' compensation insurance, but the case would not be recordable under Part 1904.

Conversely, an injury may be recordable for OSHA's purposes but not be covered by workers' compensation. For example, in some states, workers' compensation does not cover certain types of injuries (e.g., certain musculoskeletal disorders) and certain classes of workers (e.g., farm workers, contingent workers). However, if the injury meets OSHA recordability criteria it must be recorded even if the particular injury would not be compensable or the worker not be covered. Similarly, some injuries, although technically compensable under the state compensation system, do not result in the payment of workers' compensation benefits. For example, a worker who is injured on the job, receives treatment from the company physician, and returns to work without loss of wages would generally not receive workers' compensation because the company would usually absorb the costs. However, if the case meets the OSHA recording criteria, the employer would nevertheless be required to record the injury on the OSHA forms.

As a result of these differences between the two systems, recording a case does not mean that the case is compensable, or vice versa. When an injury or illness occurs to an employee, the employer must independently analyze the case in light of both the OSHA recording criteria and the requirements of the State workers' compensation system to determine whether the case is recordable or compensable, or both.

The American Federation of Labor and Congress of Industrial Organizations (AFL-CIO) urged OSHA to emphasize the no-fault philosophy of the Agency's recordkeeping system, stating:

The AFL-CIO is encouraged by some provisions currently in the proposed rulemaking which indirectly address underreporting. But, we believe the Agency must take it one step further. To adequately address this problem, the Agency must encourage employers to adopt a "no fault system" philosophy in the workplace and remove barriers which discourage the reporting of injuries and illnesses by employees. This philosophy will not only encourage workers to report injuries and illnesses, but also encourage those individuals (e.g., supervisors, safety personnel) responsible for recording this data to report all recordable incidents (Ex. 15: 418).

OSHA believes that the note to the Purpose paragraph of the final rule will allay any fears employers and employees may have about recording injuries and illnesses, and thus will encourage more accurate reporting. Both the Note to Subpart A of the final rule and the new OSHA Form 300 expressly state that recording a case does not indicate fault, negligence, or compensability.

The Workplace Health and Safety Council, the American Coke and Coal Chemicals Institute, and the National Oilseed Processors Association (Exs. 15: 313, 352, 353) all urged OSHA to improve on this paragraph of the proposed rule in two ways. First, these commenters asked OSHA to remove the word "necessarily" from the language of proposed paragraph (b), which stated that recording did not "necessarily mean" that anyone was at fault, that a standard had been violated, or that the case was compensable:

The qualification "necessarily" robs the [proposed] sentences of their meaning and makes them inaccurate. Using the word erroneously implies that merely listing an injury sometimes does mean that the employer or employee was at fault, that an OSHA standard was violated, or that the employee is eligible for workers' compensation. Clearly, this is not what OSHA intended to convey. Indeed, the word "necessarily" may actually worsen the problem OSHA seeks to solve, for attorneys and consultants reading the proposed provision might well advise employers that the provision actually endorses some uses of a listing against an employer.

OSHA should, therefore, delete the word "necessarily. * * *" Alternatively, the sentence in the regulation should read: "That an injury or illness is recordable has no bearing on whether the employer or employee was at fault, an OSHA standard violated, or the employee is eligible for workers' compensation. * * *" The legend in the form would be similarly changed (Exs. 15: 313, 352, 353).

(a) much preferred additional solution, would be for OSHA to promulgate in the final version a provision that makes inadmissible in all proceedings, both those under the OSH Act and those under any state or federal law, the entries in Form OSHA 300 and 301 as evidence of fault or culpability. Such a regulation would give employers the necessary assurance that their recordkeeping forms would not be used against them. Injured employees would lose nothing by this, for they could still be permitted to prove the fact of injury, its work-relatedness, and its consequence, with normal proof. They would simply not be permitted to introduce the forms as evidence of culpability. Such a rule would implement, be consistent with, and be authorized by Section 4(b)(4) of the Act, which prohibits the Act from affecting workers' compensation and tort schemes.

OSHA agrees with the point made by these commenters about the proposed rule's use of the word "necessarily." Accordingly, the word necessarily has been deleted from the Note to the Purpose paragraph of the final rule. However, OSHA has rejected the suggestion made by these commenters to limit the admissibility of the forms as evidence in a court proceeding. Such action is beyond the statutory authority of the agency, because OSHA has no authority over the courts, either Federal or State.

In the proposal, the no-fault statement was followed by a listing of the various causes of recordable injuries and illnesses: "Recordable workplace injuries and illnesses result from a variety of workplace events or exposures, including but not limited to: accidents, exposure to toxic materials or harmful physical agents, intentional acts of violence, or naturally occurring events such as a tornado or earthquake." The American Petroleum Institute (API) (Ex. 15: 375) objected to this proposed sentence describing the various examples of injury and illness causality, stating:

To help the system have much-needed credibility, "regardless of fault or preventability" should not be applied beyond reasonable limits. Specifically, it shouldn't mean "tornado or earthquake" or other sudden, unforeseen catastrophic events over which the employer clearly could not have any control. Employers can, however, exercise control to prevent injury from some types of naturally occurring events. The terms "tornado or earthquake" should be replaced with more reasonable examples.

In the final rule, OSHA has decided to eliminate the sentence of examples to make the regulatory text clearer and more concise. However, OSHA notes that many circumstances that lead to a recordable work-related injury or illness are "beyond the employer's control," at least as that phrase is commonly interpreted. Nevertheless, because such an injury or illness was caused, contributed to, or significantly aggravated by an event or exposure at work, it must be recorded on the OSHA form (assuming that it meets one or more of the recording criteria and does not qualify for an exemption to the geographic presumption). This approach is consistent with the no-fault recordkeeping system OSHA has adopted, which includes work-related injuries and illnesses, regardless of the level of employer control or non-control involved. The issue of whether different types of cases are deemed work-related under the OSHA recordkeeping rule is discussed in the Legal Authority section, above, and in the work-relationship section (section 1904.5) of this preamble.

In a comment on proposed paragraph (a), the National Association of Manufacturers (NAM) (Exs. 25, 15: 305) argued that the OSHA recordkeeping system should only collect information on

"the most significant hazards, those that lead to the most significant injuries and illnesses * * *" and that the purpose paragraph of the final rule be revised to state "The purpose of this Part is to require employers to record and report disabling, serious and significant work-related injuries and illnesses, and work-related fatalities."

OSHA does not agree with this interpretation of the OSH Act. As discussed in the Legal Authority section, above, Congress stated clearly that the OSHA recordkeeping system was intended to capture "work-related deaths, injuries and illnesses, other than minor injuries requiring only first aid treatment and which do not involve medical treatment, loss of consciousness, restriction of work or motion, or transfer to another job" (§ 8(c)(2)) (emphasis added). The words "disabling, serious, and significant," suggested by NAM, are at variance with Congress' clear intent. OSHA concludes that the guidance given by Congress -- that employers should record and report on work-related deaths, and on injuries and illnesses other than minor injuries, establishes the appropriate recording threshold for cases entered into the OSHA recordkeeping system.

A few commenters recommended that OSHA delete paragraph (c) of the proposed Purpose section (see, e.g., Exs. 25, 15: 305, 346, 348, 420), and in the final rule, OSHA has done so because the paragraph merely attested to OSHA's cooperation with other agencies on this rule. Although the rule has, in fact, been developed in cooperation with the Department of Health and Human Services, and specifically with the National Institute for Occupational Safety and Health (NIOSH), there is no need to include this information in the regulatory text itself.

The coverage and partial exemption provisions in Subpart B of the final rule establish which employers must keep OSHA injury and illness records at all times, and which employers are generally exempt but must keep records under specific circumstances. This subpart contains sections 1904.1 through 1904.3 of the final rule.

OSHA's recordkeeping rule covers many employers in OSHA's jurisdiction but continues to exempt many employers from the need to keep occupational injury and illness records routinely. This approach to the scope of the rule is consistent with that taken in the former recordkeeping rule. Whether a particular employer must keep these records routinely depends on the number of employees in the firm and on the Standard Industrial Classification, or SIC code, of each of the employer's establishments. Employers with 10 or fewer employees are not required to keep OSHA records routinely. In addition, employers whose establishments are classified in certain industries are not required to keep OSHA records under most circumstances. OSHA refers to establishments exempted by reason of size or industry classification as "partially exempt," for reasons explained below.

The final rule's size exemption and the industry exemptions listed in non-mandatory Appendix A to Subpart B of the final rule do not relieve employers with 10 or fewer employees or employers in these industries from all of their recordkeeping obligations under 29 CFR Part 1904. Employers qualifying for either the industry exemption or the employment size exemption are not routinely required to record work-related injuries and illnesses occurring to their employees, that is, they are not normally required to keep the OSHA Log or OSHA Form 301. However, as sections 1904.1(a)(1) and 1904.2 of this final recordkeeping rule make clear, these employers must still comply with three discrete provisions of Part 1904. First, all employers covered by the Act must report work-related fatalities or multiple hospitalizations to OSHA under § 1904.39. Second, under § 1904.41, any employer may be required to provide occupational injury and illness reports to OSHA or OSHA's designee upon written request. Finally, under § 1904.42, any employer may be required to respond to the Survey of Occupational Injuries and Illnesses conducted by the Bureau of Labor Statistics (BLS) if asked to do so. Each of these requirements is discussed in greater detail in the relevant portion of this summary and explanation.

In § 1904.1 of the final rule, OSHA has retained the former rule's size-based exemption, which exempts employers with 10 or fewer employees in all industries covered by OSHA from most recordkeeping requirements. Section 1904.1, "Partial exemption for employers with 10 or fewer employees," states that:

(1) If your company had ten (10) or fewer employees at all times during the last calendar year, you do not need to keep OSHA injury and illness records unless OSHA or the BLS informs you in writing that you must keep records under § 1904.41 or § 1904.42. However, as required by § 1904.39, all employers covered by the OSH Act must report to OSHA any workplace incident that results in a fatality or the hospitalization of three or more employees.

(2) If your company had more than ten (10) employees at any time during the last calendar year, you must keep OSHA injury and illness records unless your establishment is classified as a partially exempt industry under § 1904.2.

(1) Is the partial exemption for size based on the size of my entire company or on the size of an individual business establishment?

The partial exemption for size is based on the number of employees in the entire company.

(2) How do I determine the size of my company to find out if I qualify for the partial exemption for size?

To determine if you are exempt because of size, you need to determine your company's peak employment during the last calendar year. If you had no more than 10 employees at any time in the last calendar year, your company qualifies for the partial exemption for size.

The original OSHA injury and illness recording and reporting rule issued in July 1971 required all employers covered by the OSH Act to maintain injury and illness records. In October 1972, an exemption from most of the recordkeeping requirements was put in place for employers with seven or fewer employees. In 1977, OSHA amended the rule to exempt employers with 10 or fewer employees, and that exemption has continued in effect to this day. All employers, however, have always been required to report fatalities and catastrophes to OSHA and to participate in the BLS survey, if requested to do so.

As discussed in the Legal Authority section of this preamble, the 10 or fewer employee threshold is consistent with Congressional intent: the 1977 Federal Register notice announcing the new exemption cited the Department of Labor appropriations acts for fiscal years 1975 and 1976, which exempted employers having 10 or fewer employees from most routine recordkeeping requirements, and Section 8(d) of the Act, as the major reasons for raising the exemption size threshold from seven to 10 employees. The 1977 Notice stated that the new size threshold appropriately balanced the interest of small businesses while preserving the essential purposes of the recordkeeping scheme:

The final rule published today maintains the former rule's partial exemption for employers in all covered industries who have 10 or fewer employees. Under the final rule (and the former rule), an employer in any industry who employed no more than 10 employees at any time during the preceding calendar year is not required to maintain OSHA records of occupational illnesses and injuries during the current year unless requested to do so in writing by OSHA (under § 1904.41) or the BLS (under § 1904.42). If an employer employed 11 or more people at a given time during the year, however, that employer is not eligible for the size-based partial exemption.

In the 1996 proposal, OSHA contemplated raising the threshold for the size-based exemption to 19 employees for all employers except those in the construction industry. In proposing this more extensive exemption, OSHA stated that BLS Annual Survey data appeared to indicate that small businesses in this size category had proportionately fewer injuries and illnesses and were thus safer places to work. However, since the proposal, OSHA has analyzed the record evidence on this point and now believes that small businesses are not generally likely to be less hazardous than larger businesses and, in fact, are likely, as a general matter, to be more hazardous than large businesses. OSHA's reasoning is described below.

Comments to the record make clear that the recording of fewer injuries and illnesses by very small firms could have many causes other than a lower level of hazards. For example, the National Institute for Occupational Safety and Health (NIOSH) submitted a comment to the record that described numerous studies based on fatality and workers' compensation data that suggest that smaller businesses are at least as hazardous as larger businesses (Ex. 15: 407). NIOSH also argued that the BLS estimated injury and illness incidence rates for small employers may be erroneously low, i.e., may be the result of underreporting rather than a lower injury rate. The following comment from NIOSH explains these concerns:

From a public standpoint, NIOSH does not support a partial exemption from recordkeeping requirements for employers in the construction industry with 10 or fewer employees, and non-construction employers with 19 or fewer employees. Research indicates significant safety and health problems in "small" establishments which employ a substantial proportion of the workforce. One-quarter of the civilian, full-time workforce is employed in establishments with fewer than 25 employees (Oleinick et al. 1995).

The Occupational Safety and Health Administration (OSHA) notes [in the proposal to the recordkeeping rule] that "the Annual Survey data show that small employers generally experience much lower patterns of injuries and illnesses than medium size firms." However, recent literature comparing Annual Survey data and workers compensation data questions the validity of the estimated rates for small employers obtained through the Bureau of Labor Statistics (BLS) Annual Survey. Moreover, fatal and nonfatal work injuries are a significant risk among small businesses in hazardous industries and many industries with high fatal and nonfatal injury rates are comprised primarily of small companies. In addition, NIOSH research indicates that small companies have less access to safety and health programs that might reduce injuries and illnesses than larger companies [NIOSH 1988a].

Though the Annual Survey of Occupational Injuries and Illnesses has consistently reported that employers with fewer than 20 employees have significantly lower rates of injuries and illnesses, there is concern that these low incidence rates are an artifact of the reporting system. Analysis of compensable injuries with >7 missed workdays in Michigan indicates that the pattern of lower injury rates among small employers is not consistent across industry divisions. Though the services and trade industry divisions show a marked decline in compensable injury rate for small size firms, the higher risk industries of construction and transportation/utilities show relatively little decline in the compensable injury rate for employers with fewer than 25 employees. Comparison of the demographic characteristics of the Michigan work force with the demographic characteristics of injured workers suggest that high risk groups (e.g., males, younger workers [<35 years of age], construction, manufacturing, transportation, and blue collar workers) are over-represented among workers injured in small size firms (<25 workers). Using cumulative lost work time as a surrogate for severity of injury, the Michigan study also found that with one exception (construction), compensable injuries to workers in small firms were at least as serious as compensable injuries in larger firms [Oleinick et al. 1995] (Ex. 15: 407).

Since publication of the recordkeeping proposal, OSHA has done considerable research into the issue of fatality, injury, and illness rates in small companies. The results of this research also point to underreporting, rather than safer workplaces, as a likely reason for the lower-than-average injury and illness numbers reported by small employers. The most telling evidence that injury and illness underreporting is prevalent among small firms is the substantial discrepancy between the fatality rates in these firms and their injury and illness rates.

Most professionals agree that occupational fatality data are more reliable than occupational injury and illness data, primarily because fatalities are more likely to be reported than injuries. The work-related BLS fatality data appear to confirm this belief, showing that although businesses with fewer than 10 employees account for only 4% of the total workforce, they account for 28% of occupational fatalities. Furthermore, although businesses with fewer than 20 employees comprise only 26% of the total workforce, they account for 36% of all occupational fatalities (see Mendeloff, "Using OSHA Accident Investigations to Study Patterns in Work Fatalities," J. Occup. Med 32: 1117, 1119 (1990) (Ex. 15: 407 F)). These data strongly suggest that very small businesses are disproportionately hazardous places to work.

Many safety and health professionals also believe that injuries and illnesses are substantially underreported by small employers (see, e.g., Exs. 4, 5, 15: 407). However, the occupational injury and illness data reported by employers to the BLS in connection with its Annual Survey of Occupational Injuries and Illnesses show lower rates of injuries and illnesses for firms in the smallest size classes than for those in larger classes. In an effort to understand why smaller firms might have lower injury and illness incidence rates, the authors of one study found that: (1) occupational fatality rates were highest in businesses with fewer than 50 employees; (2) businesses with fewer than 50 employees were least likely to have occupational health services available; and (3) lost workday injury rates in several major industry categories are highest (i.e., the injuries are most severe) in these facilities. From these findings, the authors concluded:

It is difficult to imagine a set of workplace conditions in small establishments that would lead simultaneously to lower injury rates, higher fatality rates, and equal, or greater, injury severity measured by missed work time, especially since these establishments were less likely to provide injury prevention and safety services (Oleinick et al., "Establishment Size and Risk of Occupational Injury," Am. J. Med. 28(1): 2-3 (1995) (Ex. 15: 407 N)).

After considering a number of explanations that might explain this apparent incongruity, these authors rejected all explanations except one -- underreporting by small firms:

With the rejection of alternative explanations, there is a strong likelihood of underreporting as the explanation, and we estimate that the annual [BLS] survey substantially undercounts injuries in small establishments (Oleinick et al., 1995 (Ex. 15: 407 N)).

NIOSH agrees, noting that "recent literature comparing Annual Survey data and workers compensation data questions the validity of the estimated rates for small employers obtained through the BLS Annual Survey" (Ex.15: 407). Thus, the apparent discrepancy between the high fatality rate in the smallest firms (i.e., those with fewer than 20 employees) and the low rates of injuries and illnesses reported by those same firms is likely to be the result of underreporting rather than lower relative hazards.

A Wall Street Journal (Feb. 3, 1994) computer analysis of more than 500,000 Federal and State safety-inspection records came to the same conclusions, i.e., that employees of small businesses are at greater risk of exposure to workplace hazards than employees of larger businesses, and that BLS data for small firms seriously understate injuries and illnesses in such firms. From 1988 through 1992, the analysis found an incidence of 1.97 deaths per 1,000 workers at workplaces with fewer than 20 employees, compared with an incidence of just 0.004 deaths per 1,000 workers at workplaces with more than 2,500 workers. Thus, an employee's risk of death was approximately 500 times higher at the smallest businesses compared with the risk at the largest businesses. Similarly, while one in six employees at small businesses worked in an area cited for a serious safety violation, only one in 600 did so at the largest businesses. This means that employees in small businesses are 100 times more likely to be exposed to a serious hazard at work than those in the largest businesses, a finding that is consistent with the higher fatality rates in very small workplaces (Wall Street Journal, February 3, 1994).

In the final rule, OSHA has decided to continue the Agency's longstanding practice of partially exempting employers with 10 or fewer employees from most recordkeeping requirements, but not to extend the exemption to non-construction businesses with 19 or fewer employees, as was proposed. OSHA has determined that increasing the number of employers partially exempted is not in the best interests of the safety and health of their employees. First, as NIOSH's comments (Ex. 15: 407), the Oleinick et al. study (1995), the Mendeloff article (1990), and the Wall Street Journal study (1994) all indicate, businesses with 20 or fewer employees tend to be relatively hazardous places to work, and their employees have a disproportionately high risk of work-related death. Second, as NIOSH and others point out, there is reason to believe that these very small workplaces also experience disproportionately high numbers of injuries and illnesses, and that the BLS statistics for these workplaces substantially underreport the extent of job-related incidents at these establishments (Ex. 15: 407, Oleinick et al. 1995, Wall Street Journal 1994 (Ex. 15: 407 N). Finally, under the 10 or fewer employee partial exemption threshold, more than 80% of employers in OSHA's jurisdiction are exempted from routinely keeping records. Increasing the threshold for the size exemption would deprive even more employers and employees of the benefits of the information provided by these injury and illness records and reduce the number of establishments where the records can be of use to the government during an on-site visit. OSHA also believes that keeping the OSHA Log and Incident Report is important for national statistical purposes.

The final rule also retains the former rule's size exemption threshold (10 or fewer employees) for construction employers. OSHA proposed separate size thresholds for construction and nonconstruction firms, i.e., the Agency proposed to exempt firms in construction with 10 or fewer employees and non-construction firms with 19 or fewer employees from routine recordkeeping requirements. Comments on this aspect of the proposal were mixed. Some commenters agreed that OSHA should continue the exemption for construction employers with ten or fewer employees (see, e.g., Exs. 15: 145, 170, 197, 288). Other commenters urged that employers in the construction industry not be exempted from recordkeeping at all (see, e.g., Exs. 15: 62, 74, 414). For example, Robert L. Rowan, Jr. stated that:

[s]mall contractors often lack adequate safety knowledge, programs and safeguards to prevent injuries and illnesses. I believe that data obtained from these small contractors will point to a trend that these employees have a relatively high frequency of injuries that are related to tasks involving construction work such as excavations and fall hazards. I suggest that there be no exemptions for recordkeeping for any construction employer (Ex. 15: 62).

Other commenters asked OSHA to use a single size threshold for employees in all industries and to raise the size exemption threshold to more than 19 employees across the board (see, e.g., Exs. 15: 67, 304, 312, 344, 437). For example, the Sheet Metal and Air Conditioning Contractors' National Association (SMACNA) remarked:

The recordkeeping standard is considered to be a horizontal standard, which by definition, means that it covers all industries. SMACNA members own and operate sheet metal fabrication shops where they design and create the products which are then installed in the construction process, including duct work and all types of specialty and architectural sheet metal. Sheet metal fabrication shops fall under the manufacturing classification and are therefore subject to general industry standards. SMACNA contractors also construct with the components that they fabricate. Therefore, as contractors they must also comply with the OSHA standards for construction.

OSHA's arbitrary two tier record keeping requirement will cause confusion among SMACNA contractors as to which classification they are under and when they have to maintain records. With the volumes of regulations that contractors already must comply with, it is only logical that if OSHA truly wishes to simplify its recordkeeping requirements it would create a uniform standard for all industries. * * *

SMACNA urges OSHA to create a uniform horizontal standard and increase the exemption for the construction industry to cover employers with 19 or fewer employees (Ex. 15: 116).

After a review of the record and reconsideration of this issue, OSHA agrees that there should be only one size exemption threshold across all industries and finds that the threshold should be 10 or fewer employees. This threshold comports both with longstanding Agency practice and Congressional intent. Further, as discussed above, OSHA finds that extending this threshold to include firms with 11 to 19 employees is not warranted by the evidence. Firms in this size range have a disproportionately large number of fatalities, and their lower reported injury and illness rates are likely to be the result of underreporting rather than fewer hazards. Thus, companies in this size class need the information their OSHA records provide to improve conditions in their workplaces and to protect their employees from job-related injuries, illnesses, and deaths. Likewise, OSHA does not believe that it would be appropriate to remove the partial exemption for construction employers with 10 or fewer employees, as some commenters suggested (see, e.g., Exs. 15: 67, 304, 312, 344, 437). Using the same size threshold for all OSHA-covered industries also makes the rule simpler and is more equitable from industry to industry.

Many commenters supported raising the size-based exemption threshold (see, e.g., Exs. 27, 15: 26, 27, 67, 102, 123, 145, 170, 173, 182, 198, 247, 288, 304, 359, 375, 378, 392, 401, 437). For example, the American Society of Safety Engineers (ASSE) remarked:

ASSE supports exempting businesses under twenty (20) employees from the standard with some specific industry exemptions. Enforcing this regulation for businesses of less than twenty (20) employees would be detrimental to small business from the recordkeeping/ bureaucracy perspective, and may not generate any significant data. ASSE wishes to clarify, however, that this position should not be interpreted to mean that small businesses should be exempted from safety and health laws. We believe that all employees are entitled to an equal level of safety and health regardless of the size of their place of employment. Exempting a paperwork requirement does not change this level of commitment (Ex. 15: 182).

Two commenters suggested that OSHA use an even higher threshold for determining the size-based exemption (Exs. 15: 357, 408). The Synthetic Organic Chemical Manufacturers Association (SOCMA) stated "* * * SOCMA believes that OSHA should modify the small employer exemption by increasing it to 40 employees. This alternative approach would reduce the employer paperwork burden while improving the accuracy of injury and illness information" (Ex. 15:357). Similarly, the American Dental Association (ADA) commented "The ADA suggests that OSHA expand the proposed exemption from 'fewer than 20 employees' to 'fewer than 25 employees.' This would bring the small-employer exception into conformity with many federal and state employment laws. It would also serve as a more reasonable dividing line between small employers and others" (Ex. 15:408).

Some commenters, however, objected to OSHA's proposed exemption of employers in the 11 to 20 employee size range (see, e.g., Exs. 15:62, 369, 379, 407, 415, 418). Among these was the International Brotherhood of Teamsters (IBT), which stated:

IBT maintains the importance of recording of all occupational injuries and illnesses. For that same reason, International Brotherhood of Teamsters does not support increasing the trigger for non-construction employers from ten to nineteen employees. Although injuries due to preventable causes occur in all types and sizes [of businesses], a disproportionately high number of fatalities occur in the smallest businesses. According to an analysis of BLS and OSHA data, then assistant secretary of labor, Joe Dear, told the House of Representative's Small Business Committee, "Businesses with fewer than eleven workers account for 33 percent of all fatalities even though they account for less than 20 percent of employees." According to a study by the National Federation of Independent Businesses, "generally businesses with fewer employees do less to improve safety than those with more." Large corporations can afford the full-time services of a safety engineer and industrial hygienist, whereas often small firms cannot. IBT contends that it is up to OSHA to protect the workers and institute prevention measures. The use of required recordkeeping of data helps to reach that aim by providing hard data. If the data is going to be used as a prevention tool, it must be collected from the entire workforce not just a subgroup (Ex. 15:369).

Reliance on a single size exemption threshold also addresses the point made by SMACNA: that many small employers perform construction work and also manufacture products and would therefore be uncertain, if the rule contained two size exemption thresholds, as to whether they are required to keep records or not.

OSHA's proposed rule stated that the size exemption would apply to employers based on the number of employees employed by the employer "for the entire previous calendar year." The Office of Advocacy of the Small Business Administration (SBA) observed (Ex. 15:67, p. 4) that this statement could be interpreted in various ways, and expressed concern that it could be taken to refer to the total number of employees who had been employed at one time or another during the year rather than the total employed at any one time of the year. The SBA office recommended that OSHA provide clearer guidance. OSHA agrees with the SBA that the proposed regulatory language was ambiguous. Accordingly, the final rule clarifies that the 10 or fewer size exemption is applicable only if the employer had fewer than 11 employees at all times during the previous calendar year. Thus, if an employer employs 11 or more people at any given time during that year, the employer is not eligible for the small employer exemption in the following year. This total includes all workers employed by the business. All individuals who are "employees" under the OSH Act are counted in the total; the count includes all full time, part time, temporary, and seasonal employees. For businesses that are sole proprietorships or partnerships, the owners and partners would not be considered employees and would not be counted. Similarly, for family farms, family members are not counted as employees. However, in a corporation, corporate officers who receive payment for their services are considered employees.

Consistent with the former rule, the final rule applies the size exemption based on the total number of employees in the firm, rather than the number of employees at any particular location or establishment. Some commenters suggested that the size exemption should be based on the number of employees in each separate establishment rather than the entire firm (see, e.g., Exs. 15: 67, 201, 437). For example, Caterpillar Inc. (Ex. 15: 201) noted:

We do object to the note to [proposed] paragraph 1904.2(b)(2) which bases size exemptions on the total number of employees in a firm rather than the establishment size. Size exemptions must be based upon individual establishment size. The factors that make recordkeeping difficult and unproductive for small facilities are not eliminated by adding small facilities together. Small facilities are usually unique and adding together the injury and illness experience of different small facilities will not produce a valid database for accident analysis or accident prevention planning. Injury and illness data collection is difficult because of small facility size and lack of recordkeeping expertise and resources. The benefits of collecting information in small facilities does not justify the costs. It is illogical to base the size exemption on anything other than the size of each separate establishment.

OSHA does not agree with this comment because the resources available in a given business depend on the size of the firm as a whole, not on the size of individual establishments owned by the firm. In addition, the analysis of injury records should be of value to the firm as a whole, regardless of the size of individual establishments. Further, an exemption based on individual establishments would be difficult to administer, especially in cases where an individual employee, such as a maintenance worker, regularly reports to work at several establishments.

Section 1904.2 of the final rule partially exempts employers with establishments classified in certain lower-hazard industries. The final rule updates the former rule's listing of partially exempted lower-hazard industries. Lower-hazard industries are those Standard Industrial Classification (SIC) code industries within SICs 52-89 that have an average Days Away, Restricted, or Transferred (DART) rate at or below 75% of the national average DART rate. The former rule also contained such a list based on data from 1978-1980. The final rule's list differs from that of the former rule in two respects: (1) the hazard information supporting the final rule's lower-hazard industry exemptions is based on the most recent three years of BLS statistics (1996, 1997, 1998), and (2) the exception is calculated at the 3-digit rather than 2-digit level.

The changes in the final rule's industry exemptions are designed to require more employers in higher-hazard industries to keep records all of the time and to exempt employers in certain lower-hazard industries from keeping OSHA injury and illness records routinely. For example, compared with the former rule, the final rule requires many employers in the 3-digit industries within retail and service sector industries that have higher rates of occupational injuries and illnesses to keep these records but exempts employers in 3-digit industries within those industries that report a lower rate of occupational injury and illness. Section 1904.2 of the final rule, "Partial exemption for establishments in certain industries," states:

(1) If your business establishment is classified in a specific low hazard retail, service, finance, insurance or real estate industry listed in Appendix A to this Subpart B, you do not need to keep OSHA injury and illness records unless the government asks you to keep the records under § 1904.41 or § 1904.42. However, all employers must report to OSHA any workplace incident that results in a fatality or the hospitalization of three or more employees (see § 1904.39).

(2) If one or more of your company's establishments are classified in a non-exempt industry, you must keep OSHA injury and illness records for all of such establishments unless your company is partially exempted because of size under § 1904.1.

(1) Does the partial industry classification exemption apply only to business establishments in the retail, services, finance, insurance or real estate industries (SICs 52-89)?

Yes. Business establishments classified in agriculture; mining; construction; manufacturing; transportation; communication, electric, gas and sanitary services; or wholesale trade are not eligible for the partial industry classification exemption.

(2) Is the partial industry classification exemption based on the industry classification of my entire company or on the classification of individual business establishments operated by my company?

The partial industry classification exemption applies to individual business establishments. If a company has several business establishments engaged in different classes of business activities, some of the company's establishments may be required to keep records, while others may be exempt.

(3) How do I determine the Standard Industrial Classification code for my company or for individual establishments?

You determine your Standard Industrial Classification (SIC) code by using the Standard Industrial Classification Manual, Executive Office of the President, Office of Management and Budget. You may contact your nearest OSHA office or State agency for help in determining your SIC.

Employers with establishments in those industry sectors shown in Appendix A are not required routinely to keep OSHA records for their establishments. They must, however, keep records if requested to do so by the Bureau of Labor Statistics in connection with its Annual Survey (section 1904.42) or by OSHA in connection with its Data Initiative (section 1904.41). In addition, all employers covered by the OSH Act must report a work-related fatality, or an accident that results in the hospitalization of three or more employees, to OSHA within 8 hours (section 1904.39).

In 1982, OSHA exempted establishments in a number of service, finance and retail industries from the duty to regularly maintain the OSHA Log and Incident Report (47 FR 57699 (Dec. 28, 1982)). This industry exemption to the Part 1904 rule was intended to "reduce paperwork burden on employers without compromising worker safety and health."

The 1982 list of partially exempt industries was established by identifying lower hazard major industry groups in the SIC Divisions encompassing retail trade, finance, insurance and real estate, and the service industries (SICs 52-89). Major industry groups were defined as the 2-digit level industries from the SIC manual published by the U.S. Office of Management and Budget (OMB). Industries in these major industry groups were partially exempted from coverage by Part 1904 if their average lost workday injury rate (LWDI) for 1978-80 was at or below 75% of the overall private sector LWDI average rate for that year. Industries traditionally targeted for OSHA enforcement (those in SICs 01 through 51, comprising the industry divisions of agriculture, construction, manufacturing, transportation and public utilities, mining, and wholesale trade) remained subject to the full recordkeeping requirements. Although the 1982 Federal Register notice discussed the possibility of revising the exempt industry list on a routine basis, the list of partially exempt industries compiled in 1982 has remained unchanged until this revision of the Part 1904 rule.

The proposed rule would have updated the industry exemption based on more current data, and would have relied on 3-digit SIC code data to do so. The only change from the former rule taken in the proposal would have been reliance on LWDI rates for industries at the 3-digit, rather than 2-digit, level.

Evaluating industries at the 3-digit level allows OSHA to identify 3-digit industries with high LWDI rates (DART rates in the terminology of the final rule) that are located within 2-digit industries with relatively low rates. Conversely, use of this approach allows OSHA to identify lower-hazard 3-digit industries within a 2-digit industry that have relatively high LWDI (DART) rates. Use of LWDI (DART) rates at the more detailed level of SIC coding increases the specificity of the targeting of the exemptions and makes the rule more equitable by exempting workplaces in lower-hazard industries and requiring employers in more hazardous industries to keep records.

Under the proposal, based on their LWDI (DART) rates, the following industries would have been required to keep records for the first time since 1982:

In the Issues section of the preamble to the proposed rule, OSHA asked the public to comment on the appropriateness of the proposed exemption procedure, and on whether or not OSHA should expand this approach to industries in SICs 01 through 51. The Agency also asked for alternative approaches that would reduce employer paperwork burden while retaining needed injury and illness information, and for estimates of the costs and benefits associated with these alternatives. OSHA notes that the final rule is based on the most recent data available (1996-1998). Although it has relied on the methodologies proposed (3-digit SIC codes, industries below 75% of the national average LWDI rate), there have been a few shifts in the industries proposed to be covered and those actually covered by the final rule. Thus this final rule will continue to exempt eating and drinking places (SICs 581 and 582) but will not exempt automobile parking (SIC 752).

A number of commenters supported OSHA's proposal to apply the 1982 exemption criteria to the service and retail industries at the three-digit SIC level (see, e.g., Exs. 27; 15: 26, 199, 229, 247, 272, 299, 359, 375, 378, 392). However, a number of commenters opposed any exemptions from the Part 1904 requirements on the basis of industry classification (see, e.g., Exs. 15: 9, 13, 31, 62, 78, 83, 129, 153, 154, 163, 186, 197, 204, 234, 350, 379, 399, 414). The International Paper Company explained its reasons for opposing industry exemptions as follows:

Exempting employers with low incidence rates is inconsistent with a major objective of the recordkeeping rules; specifically, measuring the magnitude of work-related injuries and illnesses. Exemption of specific industrial classifications or small employers may bias statistics which are used by OSHA for identifying industries for inspections. These exemptions may also impact statistics related to less traditional, but increasingly more frequent exposures such as bloodborne pathogens, tuberculosis, motor vehicle incidents or workplace violence.

Exempting employers with low incidence rates does not provide any measurable relief from paperwork requirements. Time spent on recordkeeping is primarily dedicated to decision making regarding work relationship and recordability, not actual Log entries or completing supplemental reports. Simplifying the decision making process is the best way to reduce the burden of recordkeeping, not exempting employers (Ex. 15:399).

Injury and illness recordkeeping is the most basic step an employer must take in order to begin to address workplace hazards. Responsible employers recognize that injury and illness records are a useful tool for development of sound company safety and health programs. This information is also critical to the workers themselves, by raising awareness about how and where people are getting hurt, they in turn use this information to work to eliminate the causes of such injuries and illnesses. Therefore it is disturbing that in the proposed revised standard, there still exist industry exemptions for recordkeeping and reporting. Prior to 1983, all employers covered by OSHA with more than ten employees were required to maintain injury and illness records.

* * * SEIU believes that such exemptions are unwarranted and violate the specific language of the Occupational Safety and Health Act. * * * The Act does not provide for excluding entire classes of occupationally injured and sick workers. Furthermore, little recordkeeping will be required for industries that are safe and experience low rates of injuries and illnesses. It is critical that OSHA require recordkeeping for all industries, especially since many previously exempt sectors now experience increasing rates of injury and illness. Many of these industry sectors are also dramatically expanding -- therefore, continued recordkeeping is even more critical (Ex. 15:379).

From the point of view of injury and illness prevention. * * * an establishment that does not track its injury and illness experience cannot effectively administer a prevention program. * * *

Although OSHA encourages employers to track the occupational injuries and illnesses occurring among their employees and agrees that doing so is important for safety and health prevention efforts, OSHA has decided in the final rule to continue the long-established practice of exempting employers in industries with lower average lost workday incidence rates from most OSHA recordkeeping requirements but to tie the exemption as closely as possible to specific 3-digit SIC code data.

Accordingly, non-mandatory Appendix A of the final rule identifies industries for exemption at the 3-digit SIC code level. Although this approach does make the list of exempt industries longer and more detailed, it also targets the exemption more effectively than did the former rule's list. For example, the final rule does not exempt firms in many of the more hazardous 3-digit SIC industries that are embedded within lower rate 2-digit SIC industries. It does, however, exempt firms in relatively low-hazard 3-digit SIC industries, even though they are classified in higher hazard 2-digit SIC industries. Where Days Away, Restricted, or Transferred (DART, formerly LWDI) rate calculations exempt all of the 3-digit SIC industries within a given 2-digit industry, the exempt industry list in Appendix A displays only the 2-digit SIC classification. This approach merely provides a shorter, simpler list.

For multi-establishment firms, the industry exemption is based on the SIC code of each establishment, rather than the industrial classification of a firm as a whole. For example, some larger corporations have establishments that engage in different business activities. Where this is the case, each establishment could fall into a different SIC code, based on its business activity. The Standard Industrial Classification manual states that the establishment, rather than the firm, is the appropriate unit for determining the SIC code. Thus, depending on the SIC code of the establishment, one establishment of a firm may be exempt from routine recordkeeping under Part 1904, while another establishment in the same company may not be exempt.

Several commenters suggested that OSHA use an alternate method for determining exemptions (see, e.g., Exs. 15: 97, 201, 359). The National Safety Council (Ex. 15: 359), for example, urged OSHA to "evaluate other exemption procedures before incorporating one into proposed section 1904.2."

OSHA has evaluated other approaches but has decided that the 3-digit DART rate method is both simpler and more equitable than the former 2-digit method. By exempting lower-hazard industry sectors within SICs 52-89, OSHA hopes both to concentrate its recordkeeping requirements in sectors that will provide the most useful data and to minimize paperwork burden. No exemption method is perfect: any method that exempts broad classes of employers from recordkeeping obligations will exempt some more hazardous workplaces and cover some less hazardous workplaces. OSHA has attempted to minimize both of these problems by using the most current injury and illness statistics available, and by applying them to a more detailed industry level within the retail, financial and service sectors than was formerly the case. OSHA has also limited the scope of the exemptions by using an exemption threshold that is well below the national average, including only those industries that have average DART rates that are at or below 75% of the national average DART rate. The rule also limits the exempt industries to the retail, financial and service sectors, which are generally less hazardous than the manufacturing industry sector.

The Orlando Occupational Safety and Health Customer Council asked: "What is the criteria for exemptions? For example, large auto dealers who also perform auto repair work are exempt, while smaller auto repair shops are not exempt. Why not classify the organization by the most hazardous occupation [within that organization]?" (Ex. 15: 97).

In response to this query, OSHA notes that the exemption procedure is reasonably straightforward, as the following example illustrates: the automobile dealer industry is exempt because its DART rate, as indicated by its average over three years of BLS data, is below 75% of the national average rate. Automobile repair shops are not exempted, however, because their rate is higher than the 75% cutoff. If OSHA were to base its recordkeeping requirements on the most hazardous occupation within a given industry, assuming that occupation-specific within-industry injury and illness data were available, as this commenter suggests, the number of establishments in individual industries that would have to keep records would greatly increase. This is because even relatively safe industries have some number of employees who engage in relatively hazardous occupations. For example, workers who transport currency, coins, and documents for banks and other financial institutions are engaged in a fairly hazardous occupation. They may be injured in many different ways, ranging from highway accidents, to lifting of heavy parcels, to robberies. However, the experience of these few employees within the industry does not accurately reflect the relative degree of hazard confronting the vast majority of employees in the financial industries. Although it is certainly not perfect, OSHA believes that the BLS lost workday injury rate (DART rate) is a better comparative statistic than the injury rate for a particular occupation because it reflects the risk to the average worker within the particular industry. Moreover, while it is relatively easy to classify employees according to occupation, it is unclear how to classify individual employers with regard to detailed occupation, and OSHA is also not aware of data that would permit such classification.

The Caterpillar Corporation (Ex. 15: 201) suggested that OSHA adjust the formula used to determine which industries are exempted:

You propose to base your exemption on achieving less than 75% of the average private sector lost workday injury rate; however, we would recommend expanding the size of the exemption to include all industries below the private sector average. We have no objection to your proposal to eliminate the "nesting" problem within 2-digit SIC code groups, as long as the exemption size is maximized. The recordkeeping paperwork burden for small and relatively safe industries is significant and not justified based upon the benefits received.

OSHA has decided in the final rule to continue to use a formula that will exempt retail, finance and services industries from most recordkeeping requirements if they have a Days Away, Restricted, or Transferred (DART) rate that is at or below 75% of the national average rate. OSHA believes that the 75% threshold will ensure that only industries with relatively low injury and illness rates are exempted from these requirements. Using the national average DART rate, rather than 75% of the national DART rate, as the threshold for exemption purposes would exempt employers whose industries were merely average in terms of their DART rate.

OSHA received many comments from firms in industries that have been exempt from most OSHA recordkeeping requirements since 1982 but that would have been required by the proposed rule to keep records. Most of these commenters opposed their industry's inclusion within the scope of the proposed rule. For example, several commenters from the restaurant industry objected to the fact that SICs 581 and 582, eating and drinking places, would have been covered (see, e.g., Exs. 15: 3, 4, 5, 6, 7, 8, 12, 20, 22, 55, 96, 125, 202, 311). The National Restaurant Association remarked:

-- the proposal, if promulgated, will cost eating and drinking establishments an estimated $17 million in the first year alone;

-- the additional recordkeeping obligations under the proposed rule duplicate data already available to OSHA from other sources; and

-- the current data does not justify removal of the partial recordkeeping exemption for eating and drinking establishments (Ex. 15: 96).

In the final rule, the exemption for eating and drinking places is retained, because the recent data indicate that these industries have DART rates that are below 75% of the national rate.

Two commenters addressed the proposed removal of the exemption for SIC 553, auto and home supply stores (Ex. 15: 367, 402). For example, the Automotive Parts and Accessories Association (APAA) stated:

The vast majority of auto parts stores are similar to other retailers which would still be exempt under this proposal. * * * [m]ore than three quarters of the automotive parts retailers which are proposed to be saddled with the full Log requirements would have little or no potential injury or illness experience to justify the added mandate (Ex. 15: 367).

Several commenters discussed the proposed removal of the exemption for SIC 721, laundry, dry cleaning and textile rental services (see, e.g., Exs. 15: 183, 244, 326). Typical of the views expressed by these commenters was the comment of the Textile Rental Services Association of America (TRSA):

TRSA is strongly opposed to OSHA's proposal to eliminate the partial exemption from recordkeeping and reporting requirements for laundry, cleaning, and garments services for Standard Industrial Classification (SIC) 721. TRSA believes that the proposed inclusion of the textile rental industry is unjustified. Because the textile rental industry has historically been proactive when it comes to workplace safety and has been 75% below the industry average for lost work days, we contend that OSHA's plan to eliminate the partial exemption from injury/illness recordkeeping requirements is unwarranted (Ex. 15: 183).

The National Association of Home Builders (NAHB) commented on the proposed inclusion in the recordkeeping system of a variety of industries closely associated with the home building industry:

As a result of using a 3 digit Standard Industrial Classification (SIC), "Real Estate Offices" (SIC 651) will now be required to report and record injury and illness data if they have more than 19 workers during the year. A cursory analysis of the hazards associated with real estate offices seems to indicate limited exposure to high hazards (Ex. 15: 323).

The primary arguments put forth by these commenters are as follows: (1) The occupational injury and illness data collected under Part 1904 are available to OSHA from other sources; (2) OSHA's data requirements are burdensome; (3) the use of even more current data would change the list of exempted industries; and (4) some of the individual industries that would be covered are relatively safe.

In response, OSHA notes that, although statistical information on average work-related injury and illness rates in industries is available from the BLS and other sources, information about the hazards present at specific workplaces is not available to OSHA from those same sources. OSHA recognizes that the maintenance of these records imposes some burden on businesses in the form of paperwork. However, the benefits of keeping records are also clearly substantial: informed employers can use the data to provide greater protection for their employees and to receive the benefits that accrue from prevention efforts in the form of fewer injuries and illnesses. In addition, the records are useful to OSHA in the inspection process. OSHA also believes that the process for selecting exempt industries must be as objective as possible, and that exemptions must rely upon timely and objective information about the safety and health experience of a given industry. The lost workday injury rates published by the Bureau of Labor Statistics provide the most consistent and reliable nationwide statistics available for this purpose, and OSHA is therefore relying on these data. The 75% of the national rate cutoff strikes a reasonable balance between collecting data likely to be useful and avoiding unnecessary burden. OSHA has used the most recent data available at this time in establishing the final list of partially exempt industries. OSHA also has used data from a three-year period (1996-1998) rather than a one-year period to reduce year-to-year variation in the data.

Other commenters argued that their industry should not be exempt because their workplaces continue to pose risk to the workers in them. For example, the American Nurses Association (ANA) opposed the partial exemption of doctor's offices and health services:

ANA urges OSHA to remember the purpose of the Act, to protect the health and safety of ALL workers, when deliberating on exempting employers from this standard. As stated before, health care workers risk of exposure to injury and illness is not limited to one setting. Therefore, the Standard Industrial Classifications (SICs) 801 Offices and Clinics of Doctors of Medicine and SIC 809 Miscellaneous Health and Allied Services should not be exempt from this standard (Ex. 15: 376).

The International Brotherhood of Teamsters (IBT) also argued against excluding certain health care service industries:

IBT has concerns when the use of this analysis will grant partial exemptions to SIC codes 801 (offices and clinics of doctors), 807 (medical and dental offices), and 809 (miscellaneous health and allied services). All three of these SIC codes are covered under other OSHA rules (such as the bloodborne pathogen standard and ethylene oxide standard) and have medical surveillance requirements to detect adverse health effects. OSHA should require that these workplaces keep records of work related illnesses or injuries that occur. Especially, since OSHA has already determined that there is a significant risk of harm from exposures in these workplaces (Ex. 15: 369).

OSHA recognizes that workers in establishments that are exempt under the 75% DART rate criterion will continue to be exposed to job-related hazards and to experience workplace injuries and illnesses. However, because these industries' overall injury rate is below the 75% cutoff, they qualify for exemption, along with other financial, service and retail industries that fall below that injury rate threshold. Exemption of an industry on the basis of its lower-than-average DART rate does not mean that all establishments within that industry have such rates or that workers in that industry will not experience injuries and illnesses. The 1904 partial exemption does not exempt employers from any other OSHA regulation or standard, so employees in these industries will continue to benefit from the protection offered by the OSHA standards. For example, while doctors' and dentists' offices are partially exempt under the 1904 regulation, they are still required to comply with the OSHA Bloodborne Pathogens Standard (29 CFR 1910.1030). Use of the 75% criterion merely provides a cutoff point, based on BLS injury and illness rates, for different industry sectors. OSHA believes that it is appropriate to use the 75% cutoff point because, in general, it is an appropriate overall indicator of the relative hazard rank of an industry. OSHA recognizes that no average across-establishment statistic can capture the injury and illness experience of all occupations or establishments within that industry.

For some SIC codes, the BLS Annual Survey does not publish data at the three-digit level. The survey is designed to provide data at the four-digit level in the manufacturing industries and at the three-digit level in all other industries, primarily because of budget constraints that limit the amount of data the BLS can collect and process. However, the survey has other publication criteria that make some of the data at this detailed level unpublishable. Under the proposal, coverage would have been based on the industry's LWDI rate. If a 3-digit sector did not have published data, OSHA proposed to use the data for the two-digit industry group for that sector.

One 3-digit sector affected by this approach was dental offices (SIC 802), which the proposal would have covered because the entire 2-digit health care sector has a relatively high injury and illness rate. The American Dental Association (ADA) suggested that OSHA use an alternative approach to exempt dentists from coverage rather than rely on a strict data protocol for making the decision:

[d]ental offices are very much like physicians' offices in terms of size, scope of activity, and degree of occupational health risk. For purposes of this rulemaking, however, physicians' offices have been granted a categorical exemption while dentists' offices (SIC Code 802) have not. Even dental laboratories (SIC Code 807) have been granted a categorical exemption from this rule, although it is unlikely that anyone would assert that dental laboratories are safer and more healthful places to work than dental offices. The ADA is unaware of any data suggesting that dental offices should be treated differently than either physicians' offices or dental laboratories (Ex. 15: 408).

The more recent data published by the BLS for the years 1996, 1997, and 1998 include specific estimates of the injury and illness experience for SIC 802 (dental offices) in that period. The dental office industry experienced a 3-year average rate of days away, restricted, or transferred injuries of 0.2 per 100 workers in those years, a rate well below 75% of the national average. Therefore, the final rule exempts employers classified in SIC 802 from routine recordkeeping requirements.

The proposed rule would have removed SIC 736 (personnel supply services) from the list of exempted industry sectors; however, because this industry's more recent average DART (formerly LWDI) rate (for the years 1996, 1997, and 1998, the base years OSHA is using to determine lower-hazard industry exemptions) is above 75% of the national average cutoff, SIC 736 is not exempted under the final rule. The final rule (see section 1904.31(b)(2)) requires the "using firm" to record the injuries and illnesses of temporary workers that are "leased" from a personnel supply service, providing that the using firm supervises these workers on a day-to-day basis.

The National Association of Temporary and Staffing Services commented on the proposed removal of the exemption for SIC 736:

The proposed rules also would lift the partial exemption for employers classified under SIC Code 7363 (help supply services). Those employers, among others, were exempted from injury and illness record keeping requirements in 1982 because they had low work place injury rates. The proposal to lift the exemption is based on reported increased injury rates for these employers. However, since records for the vast majority of staffing firm employees are maintained by the worksite employer as explained above, the practical effect of lifting the exemption for staffing firms would be to require them to maintain records for their home office clerical and administrative workers -- for whom there is no evidence of increased work place illnesses or injuries. Hence, we urge OSHA to retain the partial exemption for SIC 7363.

If the exemption is not retained in the case of SIC 7363 employers, it would be especially important for the final rules to expressly provide, as set forth above, that there is no intent to impose a dual reporting requirement. At least one state OSH office already has construed the proposed lifting of the partial exemption as creating an obligation on the part of staffing firms to maintain records for all of its employees, including temporary employees supervised by the worksite employer. This is clearly inconsistent with the intent of the proposed rule and should be clarified (Ex. 15: 333).

The final rule makes clear that, when a "leased" or "temporary" employee is supervised on a day-to-day basis by the using firm, the using firm must enter that employee's injuries and illnesses on the using firm's establishment Log and other records. Injuries and illnesses occurring to a given employee should only be recorded once, either by the temporary staffing firm or the using firm, depending on which firm actually supervises the temporary employees on a day-to-day basis. (see the discussion for § 1904.31, Covered employees, for an in-depth explanation of these requirements.)

Some commenters suggested that OSHA should grant partial exemptions to specific industries within SICs 01 through 51 (agriculture, forestry and fishing; mining; construction; manufacturing; transportation, communications, electric, gas and sanitary services; and wholesale trade) that had lost workday incidence rates that were below 75% of the average rate for all industries instead of limiting such exemptions to industries in SICs 52-89 (see, e.g., Exs. 15: 77, 95, 184, 201, 357, 359, 374, 375). Typical of these comments was one from the Synthetic Organic Chemical Manufacturers Association (SOCMA):

SOCMA believes that the partial exemption from recordkeeping requirements should be consistent for all standard industrial classifications. SOCMA supports the use of injury rates, rather than SIC Codes, as a criterion for partial exemption from recordkeeping requirements, provided the same criterion is applied to all work sites. For example, if the performance measure was 75 percent of the private sector average, then all industries with injury rates below this average should be exempt.

There is sound basis for this shift in OSHA's approach. It has been found in the past that some industries in partially exempt SIC Codes 52 -- 89 have had high injury rates while some in the "manufacturing" SIC Codes 01-51 have had low injury rates. This has resulted in insufficient or unavailable injury and illness information for some facilities in SIC Codes 52-89 with high injury rates. Inspection resources are wasted if injury and illness information is not available during the inspection of high injury rate facilities. Conversely, requiring full recordkeeping for facilities with low injury rates results in a facility wasting resources on unnecessary recordkeeping. All businesses, regardless of SIC Code, should be treated equally and should have the opportunity to be exempt based on injury rates (Ex. 15: 357).

The National Automobile Dealers Association (NADA) urged OSHA to exempt truck dealerships [classified in SIC 50], even though they are considered wholesale rather than retail establishments, because of their similarity to automobile dealerships [SIC 551], which are exempted:

NADA strongly urges OSHA to exempt truck dealerships (SIC 5012), the overwhelming majority of whom are small businesses as recognized by the Small Business Administration (SBA).* * * A limited exemption for truck dealerships is justified under the same criteria used for automobile dealerships (Ex. 15: 280).

On the other hand, some commenters agreed with OSHA's proposal to require all businesses in SICs 01-51 to keep injury and illness records (see, e.g., Exs. 15: 170, 199, 369). The International Brotherhood of Teamsters (IBT) remarked: "IBT does not support using the same analysis of data at the three digit level of those industries in SIC 01 through 51 (industries historically not exempted from recordkeeping requirements). IBT maintains the importance of recording of all occupational injuries and illnesses" (Ex. 15: 369). A major utility, New England Power, agreed: "We believe that the existing exemption criteria for SICs 52-89 should remain the same. Although many industries would fall within the exemption criteria in SICs 01-51, they are still higher hazard industries producing valuable data on injury/ illness experience" (Ex. 15: 170). The NYNEX Corporation also agreed with OSHA's proposed approach:

We are not in favor of extending the concept of industry-wide recordkeeping exemptions to the list of three digit codes in the group 01-51 that were identified in the proposal. Even though these groups have average injury and illness case rates that are less than 75% of the private sector average, the nature of the work operations performed within these industries suggests that the variation above and below average for individual establishments could be much greater than with SIC Codes 52-89. An exemption for this group of establishments could mask the existence of some very high case rates within this group (Ex. 15: 199).

After a review of the recent BLS data, OSHA's own experience, and the record of this rulemaking, OSHA has decided that it is appropriate to require firms in industries within the SIC 01 through 51 codes to comply with OSHA's requirements to keep records. Thus, the final rule, like the proposed rule and the rule published in 1982, does not exempt firms with more than 10 employees in the industry divisions of agriculture, mining, construction, manufacturing, wholesale trade, transportation and public utilities (SICs 01 -- 52) from routine recordkeeping.

Although OSHA no longer restricts its inspection targeting schemes to employers in these SICs, these industries have traditionally been, and continue to be, the focus of many of the Agency's enforcement programs. OSHA believes that it is important for larger employers (i.e., those with more than 10 employees) in these industries to continue to collect and maintain injury and illness records for use by the employer, employees and the government. As noted in the comments there is a wide variation in injury/illness rates among establishments classified in these industries. Further, as a whole, these industries continue to have injury and illness rates that are generally higher than the private sector average and will thus benefit from the information that OSHA-mandated records can provide about safety and health conditions in the workplace. In 1998, the lost workday injury and illness rate for the entire private sector was 3.1. As can be seen in the following table of lost workday injury and illness rates by industry division, all of the covered divisions exceeded 75% of the national average LWDI rate (2.325) for the private sector as a whole, while the exempted industry divisions had substantially lower rates.

Industry sector	1998 lost workday injury and illness rate
Agriculture, forestry and fishing (SIC 01-09)	3.9
Mining (SIC 10-14)	2.9
Construction (SIC 15-17)	4.0
Manufacturing (SIC 20-39)	4.7
Transportation, communications, electric, gas and sanitary services (SIC 40-49)	4.3
Wholesale trade (SIC 50 & 51)	3.3
Retail trade (SIC 52-59)	2.7
Finance, Insurance & Real Estate (SIC 60-67)	0.7
Services (SIC 70-87)	2.4

The problems that may be encountered by exempting additional industries are exemplified by an analysis of the petrochemical industry and the manufacturers of chemicals and petroleum products, classified in SICs 28 and 29. If the industry exemption were applied to these industries, injury and illness records would not be required for highly specialized plants that make industrial inorganic chemicals, plastics materials and synthetic resins, pharmaceuticals, industrial organic chemicals, and petroleum refineries. These industries have relatively low occupational injury and illness rates, but they are not truly low-hazard industries. All of these facilities make, use and handle highly toxic chemicals and consequently have the potential for both acute overexposure and chronic exposures of their employees to these substances. These industries, for example, are the industries to which OSHA health standards, such as the benzene, ethylene oxide, and methylene chloride standards, apply. Because occupational illnesses, particularly chronic illnesses, are notoriously underreported (see, e.g., Exs. 15: 407, 4, 5), the LWDI rates for these industries do not accurately reflect the level of hazard present in these facilities. In addition, these types of facilities are prone to major safety and health problems, including explosions, toxic releases and other events that often lead to fatalities and serious injuries. The safety and health problems of these facilities are not limited to workers, but extend to hazards posed to the general public. In addition, OSHA frequently inspects these facilities because of their potential for catastrophic releases, fires, and explosions, and the Part 1904 injury and illness records have been extremely useful for this purpose.

The Agency finds that continuing, and improving on, the Agency's longstanding approach of partially exempting those industries in SIC codes 52-89 that have DART rates, based on 3 years of BLS data, below 75% of the private-sector average strikes the appropriate balance between the need for injury and illness information on the one hand, and the paperwork burdens created by recording obligations, on the other. The BLS Annual Survey will, of course, continue to provide national job-related statistics for all industries and all sizes of businesses. As it has done in the past, the BLS will sample employers in the partially exempt industries and ask each sampled employer to keep OSHA records for one year. In the following year, BLS will collect the records to generate estimates of occupational injury and illness for firms in the partially exempt industries and size classes, and combine those data with data for other industries to generate estimates for the entire U.S. private sector. These procedures ensure the integrity of the national statistics on occupational safety and health.

The list of partially exempted industry sectors in this rule is based on the current (1987) revision of the SIC manual. The Office of Management and Budget (OMB) is charged with maintaining and revising the system of industrial classification that will replace the SIC. The new system is used by U.S. statistical agencies (including the BLS). Under the direction of OMB, the U.S. government has adopted a new, comprehensive system of industrial classification that will replace the SIC. The new system is called the North American Industrial Classification System (NAICS). NAICS will harmonize the U.S. classification system with those of Canada and Mexico and make it easier to compare various economic and labor statistics among the three countries. Several commenters expressed concern about this change in industrial classification systems (see, e.g., Exs. 15: 70, 182, 183, 379). For example, the American Society of Safety Engineers (ASSE) stated:

The Society is concerned with the recent Office of Management Budget (OMB), proposal to change the Economic Classification Policy from the Standard Industrial Classification System to the North American Industry Classification System. We recommend that OSHA study what the effect would be of promulgating a new regulation partially based on SIC codes when these codes could be potentially replaced/revised with a new classification system (Ex. 15: 182).

Although the NAIC industry classification system has been formally adopted by the United States, the individual U.S. statistical agencies (including the BLS) are still converting their statistical systems to reflect the new codes and have not begun to publish statistics using the new industry classifications. The new system will be phased into the nation's various statistical systems over the next several years. The BLS does not expect to publish the first occupational injury and illness rates under the new system until the reference year 2003. Given the lag time between the end of the year and the publication of the statistics, data for a full three-year period will not be available before December of 2006.

Because data to revise the Part 1904 industry exemption based on the NAIC system will not be available for another five years, OSHA has decided to update the industry exemption list now based on the most recent SIC-based information available from BLS for the years 1996, 1997 and 1998. OSHA will conduct a future rulemaking to update the industry classifications to the NAIC system when BLS publishes injury and illness data that can be used to make appropriate industry-by-industry decisions.

The proposal inquired whether OSHA should adopt a procedure for adjusting the industry exemption lists as the injury and illness rates of various industries change over time. A number of commenters urged OSHA to update the exemption list periodically (see, e.g., Exs. 15: 27, 87, 170, 181, 199, 272, 280, 359, 374, 375, 392, 407). Some commenters suggested various time periods, such as annually (Ex. 15: 374), every 3 years (see, e.g., Exs. 15: 87, 181, 199, 407), every 5 years (see, e.g., Exs. 15: 170, 181, 262, 272, 359, 375), or every 5 to 10 years (Ex. 15: 392). Southwestern Bell Telephone suggested that the list should be modified whenever changes in the injury and illness rates warrant a change (Ex. 15: 27). In the opinion of the National Safety Council, "How often the SIC exemption should be updated depends on how well and how quickly OSHA can communicate changes in the exempt industry list to those affected. The Council recommends updating the list every 3 to 5 years" (Ex. 15: 280).

Several commenters, however, opposed frequent updating of the SIC exemption list. For example, the Orlando Safety and Health Customer Council stated: "Changes to SIC exemptions should be limited to a minimum of every 5 years. This would reduce confusion" (Ex. 15: 97). The National Institute for Occupational Safety and Health (NIOSH) generally opposed industry exemptions but recommended that, if they were continued, they be updated as follows:

If OSHA continues to provide this exemption for low injury rate SICs, NIOSH recommends that the list of partially exempt SICs be placed in an Appendix. Because the injury and illness experience of an industry can change over time (e.g., SIC 58 and SIC 84 had injury rates at or below 75% of the private sector average in 1983, but above 75% of the private sector average in 1990 and 1992), OSHA should periodically review and modify the list of partially exempt industries. NIOSH recommends that the criteria for partial exemptions be placed in the regulatory text, while placing the list of partially exempt industries in an Appendix as noted so that the list could be updated periodically by administrative means rather than by changing the regulation. In addition to the partial exemption criteria, the regulatory text should specify the interval (in years) for reviewing and revising the list of those industries that qualify. NIOSH recommends an interval of 3 years for the review and revision process (Ex. 15: 407).

OSHA agrees with those commenters who favored regular updating of the SIC code exemption list. For the list to focus Agency resources most effectively on the most hazardous industries, it must be up-to-date. Industries that are successful in lowering their rates to levels below the exemption threshold should be exempted, while those whose rates rise sufficiently to exceed the criterion should receive additional attention. Unfortunately, the change in industry coding systems from the Standard Industrial Classification (SIC) system to the North American Industry Classification (NAIC) system will require a future rulemaking to shift to that system. Therefore, there is no value in adding an updating mechanism at this time. The automatic updating issue will be addressed in the same future rulemaking that addresses the NAIC system conversion.

Partial Exemptions for Employers Under the Jurisdiction of OSHA-Approved State Occupational Safety and Health Plans

Robert L. Rowan, Jr. expressed a concern that the OSHA State-Plan States could have differing industry exemptions from those applying to federal OSHA states, commenting:

In regard to the note in OSHA's Coverage and Exemption Table that "some states with their own occupational safety and health programs do not recognize the federal record keeping exemptions". I am deeply concerned. I would prefer that all jurisdictions enforce the same requirements. This will be confusing and create needless problems for businesses with sites in numerous states if requirements are not enforced equally (Ex. 15: 62).

For those States with OSHA-approved State plans, the state is generally required to adopt Federal OSHA rules, or a State rule that is at least as effective as the Federal OSHA rule. States with approved plans do not need to exempt employers from recordkeeping, either by employer size or by industry classification, as the final Federal OSHA rule does, although they may choose to do so. For example, States with approved plans may require records from a wider universe of employers than Federal OSHA does. These States cannot exempt more industries or employers than Federal OSHA does, however, because doing so would result in a State rule that is not as effective as the Federal rule. A larger discussion of the effect on the State plans can be found in Section VIII of this preamble, State Plans.

Section 1904.3 of the final rule provides guidance for employers who are subject to the occupational injury and illness recording and reporting requirements of other Federal agencies. Several other Federal agencies have similar requirements, such as the Mine Safety and Health Administration (MSHA), the Department of Energy (DOE), and the Federal Railroad Administration (FRA). The final rule at section 1904.3 tells the employer that OSHA will accept these records in place of the employer's Part 1904 records under two circumstances: (1) if OSHA has entered into a memorandum of understanding (MOU) with that agency that specifically accepts the other agency's records, the employer may use them in place of the OSHA records, or (2) if the other agency's records include the same information required by Part 1904, OSHA would consider them an acceptable substitute.

OSHA received very few comments on the issue of duplicate recordkeeping under different agency rules. The Fertilizer Institute (TFI) recommended that OSHA make the data mandated by OSHA and MSHA more consistent (Ex. 15:154). However, MSHA and OSHA have different recordkeeping requirements because the agencies' mandate and uses of the data differ. The approach OSHA takes in the final rule, which is to continue to accept data kept by employers under other Federal requirements if the two federal agencies have made an agreement to do so, or if the data are equivalent to the data required to be kept by Part 1904, appears to be the best way to handle the problem raised by the TFI.

Subpart C of the final rule sets out the requirements of the rule for recording cases in the recordkeeping system. It contains provisions directing employers to keep records of the recordable occupational injuries and illnesses experienced by their employees, describes the forms the employer must use, and establishes the criteria that employers must follow to determine which work-related injury and illness cases must be entered onto the forms. Subpart C contains sections 1904.4 through 1904.29.

Section 1904.4 provides an overview of the requirements in Subpart C and contains a flowchart describing the recording process. How employers are to determine whether a given injury or illness is work-related is set out in section 1904.5. Section 1904.6 provides the requirements employers must follow to determine whether or not a work-related injury or illness is a new case or the continuation of a previously recorded injury or illness. Sections 1904.7 through 1904.12 contain the recording criteria for determining which new work-related injuries and illnesses must be recorded on the OSHA forms. Section 1904.29 explains which forms must be used and indicates the circumstances under which the employer may use substitute forms.

Section 1904.4 of the final rule contains provisions mandating the recording of work-related injuries and illnesses that must be entered on the OSHA 300 (Log) and 301 (Incident Report) forms. It sets out the recording requirements that employers are required to follow in recording cases.

Paragraph 1904.4(a) of the final rule mandates that each employer who is required by OSHA to keep records must record each fatality, injury or illness that is work-related, is a new case and not a continuation of an old case, and meets one or more of the general recording criteria in section 1904.7 or the additional criteria for specific cases found in sections 1904.8 through 1904.12. Paragraph (b) contains provisions implementing this basic requirement.

Paragraph 1904.4(b)(1) contains a table that points employers and their recordkeepers to the various sections of the rule that determine which work-related injuries and illnesses are to be recorded. These sections lay out the requirements for determining whether an injury or illness is work-related, if it is a new case, and if it meets one or more of the general recording criteria. In addition, the table contains a row addressing the application of these and additional criteria to specific kinds of cases (needlestick and sharps injury cases, tuberculosis cases, hearing loss cases, medical removal cases, and musculoskeletal disorder cases). The table in paragraph 1904.4(b)(1) is intended to guide employers through the recording process and to act as a table of contents to the sections of Subpart C.

Paragraph (b)(2) is a decision tree, or flowchart, that shows the steps involved in determining whether or not a particular injury or illness case must be recorded on the OSHA forms. It essentially reflects the same information as is in the table in paragraph 1904.4(b)(1), except that it presents this information graphically.

The former rule had no tables or flowcharts that served this purpose. However, the former Recordkeeping Guidelines (Ex. 2) contained several flowcharts to help employers make decisions and understand the overall recording process. The proposed rule included a flowchart as Appendix C to Part 1904 -- Decision Tree for Recording Occupational Injuries and Illnesses. OSHA received very few comments in response to proposed Appendix C, and no commenters objected to the decision tree concept. The commenters who discussed the decision tree supported it, and many suggested that it should be incorporated into the computer software OSHA develops to assist employers with keeping the records (see, e.g., Exs. 51, 15: 38, 67, 335, 407, 438).

In the final rule, OSHA has decided to include the flowchart because of its usefulness in depicting the overall recording process. OSHA has not labeled the flowchart non-mandatory, as some commenters (see, e.g., Ex. 15: 335) suggested, because the recording of injuries and illnesses is a mandatory requirement and labeling the flowchart as non-mandatory could be confusing.

This section of the final rule sets out the requirements employers must follow in determining whether a given injury or illness is work-related. Paragraph 1904.5(a) states that an injury or illness must be considered work-related if an event or exposure in the work environment caused or contributed to the injury or illness or significantly aggravated a pre-existing injury or illness. It stipulates that, for OSHA recordkeeping purposes, work relationship is presumed for such injuries and illnesses unless an exception listed in paragraph 1904.5(b)(2) specifically applies.

Implementation requirements are set forth in paragraph (b) of the final rule. Paragraph (b)(1) defines "work environment" for recordkeeping purposes and makes clear that the work environment includes the physical locations where employees are working as well as the equipment and materials used by the employee to perform work.

Paragraph (b)(2) lists the exceptions to the presumption of work-relatedness permitted by the final rule; cases meeting the conditions of any of the listed exceptions are not considered work-related and are therefore not recordable in the OSHA recordkeeping system.

This section of the preamble first explains OSHA's reasoning on the issue of work relationship, then discusses the exceptions to the general presumption and the comments received on the exceptions proposed, and then presents OSHA's rationale for including paragraphs (b)(3) through (b)(7) of the final rule, and the record evidence pertaining to each.

Section 8(c)(2) of the OSH Act directs the Secretary to issue regulations requiring employers to record "work-related" injuries and illnesses. It is implicit in this wording that there must be a causal connection between the employment and the injury or illness before the case is recordable. For most types of industrial accidents involving traumatic injuries, such as amputations, fractures, burns and electrocutions, a causal connection is easily determined because the injury arises from forces, equipment, activities, or conditions inherent in the employment environment. Thus, there is general agreement that when an employee is struck by or caught in moving machinery, or is crushed in a construction cave-in, the case is work-related. It is also accepted that a variety of illnesses are associated with exposure to toxic substances, such as lead and cadmium, used in industrial processes. Accordingly, there is little question that cases of lead or cadmium poisoning are work-related if the employee is exposed to these substances at work.

On the other hand, a number of injuries and illnesses that occur, or manifest themselves, at work are caused by a combination of occupational factors, such as performing job-related bending and lifting motions, and factors personal to the employee, such as the effects of a pre-existing medical condition. In many such cases, it is likely that occupational factors have played a tangible role in causing the injury or illness, but one that cannot be readily quantified as "significant" or "predominant" in comparison with the personal factors involved.

Injuries and illnesses also occur at work that do not have a clear connection to a specific work activity, condition, or substance that is peculiar to the employment environment. For example, an employee may trip for no apparent reason while walking across a level factory floor; be sexually assaulted by a co-worker; or be injured accidentally as a result of an act of violence perpetrated by one co-worker against a third party. In these and similar cases, the employee's job-related tasks or exposures did not create or contribute to the risk that such an injury would occur. Instead, a causal connection is established by the fact that the injury would not have occurred but for the conditions and obligations of employment that placed the employee in the position in which he or she was injured or made ill.

The theory of causation OSHA should require employers to use in determining the work-relationship of injuries and illnesses was perhaps the most important issue raised in this rulemaking. Put simply, the issue is essentially whether OSHA should view cases as being work-related under a "geographic" or "positional" theory of causation, or should adopt a more restrictive test requiring that the occupational cause be quantified as "predominant," or "significant," or that the injury or illness result from activities uniquely occupational in nature. This issue generated substantial comment during this rulemaking, and the Agency's evaluation of the various alternative tests, and its decision to continue its historic test, are discussed below.

The final rule's test for work-relationship and its similarity to the former and proposed rules. -- The final rule requires that employers consider an injury or illness to be "work-related" if an event or exposure in the work environment either caused or contributed to the resulting condition or significantly aggravated a pre-existing injury or illness. Work relatedness is presumed for injuries and illnesses resulting from events or exposures occurring in the work environment, unless an exception in § 1904.5(b)(2) specifically applies.

Under paragraph 1904.5(b)(1), the "work environment" means "the establishment and other locations where one or more employees are working or are present as a condition of their employment. The work environment includes not only physical locations, but also equipment or materials used by the employee during the course of his or her work."

The final rule's definition of work-relationship is essentially the same as that in both the former and proposed rules except for the final rule's requirement that the work event or exposure "significantly" aggravate a pre-existing injury or illness. The Guidelines interpreting the former rule stated that

Work-relationship is established under the OSHA recordkeeping system when the injury or illness results from an event or exposure in the work environment. The work environment is primarily composed of: (1) The employer's premises, and (2) other locations where employees are engaged in work-related activities or are present as a condition of their employment. (Ex. 2 at p. 32).

The proposed rule also contained a similar definition of "work-related" and "work environment." The only significant difference between the proposed and the final rule definitions is that the proposed rule also would not have required a "significant" aggravation of a pre-existing condition before it became recordable; under the proposal, any aggravation would have been sufficient (see 61 FR 4059).

Although OSHA proposed to continue its existing definition of work-relationship, it sought comment on the following three alternative tests:

1. Exclude cases with any evidence of non-work etiology. Only cases where the work event or exposure was the sole causative factor would be recorded;

3. Record all cases where the work event or exposure had any possibility of contributing to the case (emphasis added). (61 FR 4045)

The first two alternative tests described in the proposal would have required the employer to quantify the contribution of occupational factors as compared to that of personal factors. These tests are referred to in the Legal Authority section, and in this preamble, as the "quantified occupational cause" tests. Of these tests, Alternative 2 -- record only injuries and illnesses predominantly caused by occupational factors -- received the most comment. Typical of these comments were those of the Dow Chemical Company, which expressed the view of many in industry that "[a] system that labels an injury or illness attributable to the workplace even though the workplace contribution may be insignificant does not lead to an effective, credible or accurate program" (Ex. 15: 335). Other commenters stated that recording only those cases where work was the predominant cause would improve the system by focusing attention on cases that are amenable to employer abatement (see, e.g., Exs. 22, 15: 13, 27, 34, 38, 52, 60, 69, 71, 72, 82, 97, 102, 108, 109, 122, 136, 137, 141, 146, 147, 149, 152, 154, 159, 163, 169, 171, 174, 176, 181, 197, 198, 199, 200, 201, 214, 218, 224, 230, 231, 238, 239, 260, 262, 265, 266, 272, 273, 277, 278, 287, 288, 290, 297, 301, 302, 303, 307, 313, 317, 318, 330, 335, 346, 352, 353, 370, 375, 382, 378, 383, 384, 386, 388, 396, 401, 402, 404, 405, 425, 426, 430).

Some commenters (see, e.g., Exs. 15: 185, 199, 205, 332, 338, 349, 354, 358, 375, 421, 440) offered a slight modification on Alternative 2. They suggested that using a term other than predominant, such as "substantial" or "significant," would avoid the need to define "predominant" as a percentage. For example, United Technologies (Ex. 15: 440) opposed "placing a percentage on the degree of contribution" because doing so would not be practical. Further, according to this commenter, "work relationship should be established in cases where the workplace contributed substantially to the injury or illness, as determined by an occupational physician." Arguing along the same lines, the American Petroleum Institute (API) (Ex. 15: 375) stated that it supported "in principle the work-relatedness concept presented by OSHA as Alternative 2, but feels "predominant" might be too difficult to administer as a fundamental criterion. API proposes that work-relatedness should exist when an event or exposure in the workplace is a significant factor resulting in an injury or illness. * * *" Organization Resource Counselors, Inc. (Ex. 15: 358) added: "[T]he Congressional intent in drafting these sections was to require the collection of work-related information about significant work-related injuries and illnesses." The General Electric Company (Ex. 15: 349) said that "OSHA needs to allow the facility the flexibility to record only those cases that are "more likely than not" related to workplace exposure or tasks. This determination can be made during the incident investigation. A good test of work-relatedness is whether the injury would have been prevented by full compliance with the applicable OSHA standard."

Proposed Alternative 1, which would have required the recording only of cases where work was the sole cause, was also supported by a large number of commenters (see, e.g., Exs. 15: 9, 39, 87, 95, 119, 123, 145, 151, 152, 179, 180, 183, 185, 204, 205, 225, 229, 234, 242, 259, 263, 269, 270, 304, 341, 363, 377, 389, 393, 414, 433, 443). Typical of this view was the comment of the American Health Care Association (Ex. 15: 341):

If OSHA's primary concern is to address those workplace hazards or risks that cause or may cause employee injury/illness then the agency should confine recordability to those injuries and illnesses that are directly caused by a workplace event or exposure. This approach, in turn, will focus the employer's attention on those unsafe workplace conditions that need to be corrected to protect all workers exposed to or at risk from the unsafe conditions.

The National Federation of Independent Business (Ex. 15: 304) supported Alternative 1 "because under such a system evidence of non-work-related factors is excluded thus the decision-making process is dramatically simplified and the tally is very credible." The Painting and Decorator Contractors of America (Ex. 15: 433) added: "[T]his approach is also consistent with OSHA's intent (and the Congressional mandate in the Paperwork Reduction Act of 1995) to reduce compliance burdens as this would be the simplest method for employers to apply."

Some commenters favored a closely related test for work relationship that would place primary emphasis on the nature of the activity that the employee was engaged in when injured or made ill. This test is referred to the Legal Authority section and in this preamble section as the "unique occupational activities" test. Its supporters argued that whether an injury or illness occurs or manifests itself at work is less important than whether or not the harm has been caused by activities or processes peculiar to the workplace. The AISI argued that:

[I]t is clear that Congress intended OSHA's authority to regulate to be limited to "occupational hazards" and conceived of such hazards as "processes and materials" peculiar to the workplace. * * * Congress did not give OSHA the authority to regulate hazards if they "grow out of economic and social factors which operate primarily outside the workplace. The employer neither controls nor creates these factors as he controls or creates work processes and materials." Congress was concerned with dangerous conditions peculiar to the workplace; it did not have in mind the recording of illnesses simply because they appear at work (internal citations omitted) (Ex. 15: 395).

Dow Chemical made a similar point in arguing that the criteria for determining work-relationship should include whether the activity the employee was engaged in at the time of the injury or onset of illness was for the direct benefit of the employer or was a required part of the job (Ex. 15: 335B). According to Dow, the activity-based test would be more accurate than the geographic presumption (OSHA's historic test) because it would omit injuries due to hazards beyond the employer's control:

Examples to illustrate this point include the employee who during his break attempts to remove a plastic insert in a condiment container with a knife and ends up cutting himself which requires three stitches. This activity, while it happened on company grounds, was not for the direct benefit of the company nor a requirement of his job, and there was no way for the employer to prevent it (Ex. 15: 335B).

A significant number of commenters supported OSHA's long-standing test in which work factors must be a cause, but not necessarily a "significant" or "predominant" cause, and a geographic presumption applies if "events or exposures" in the work environment either caused or contributed to the resulting condition, or aggravated a pre-existing condition (see, e.g., Exs. 15: 74, 153, 362, 369, 394, 407, 418, 429). For example, NIOSH (Ex. 15: 407) favored this approach because "[o]verreported cases can be identified and accounted for in data analysis, in contrast to the other alternatives which stress specificity at the expense of sensitivity and would result in unreported cases." The AFL-CIO argued that:

* * * [c]apturing all workplace illnesses and injuries, even those for which the predominant cause cannot be proven to be work-related, can lead to early recognition of problems and abatement of hazardous conditions. Our experience has shown us that when comprehensive records of all possible cases are kept, patterns of injury and illness emerge, enabling us to target problem areas/ factors that previously may not have been associated with that specific work environment. The inclusion of all cases will lead to prevention strategies that can reduce the risk of serious illness and injury to workers. Inclusion of all cases that have a workplace link will also assist in the recognition of diseases that are caused by synergistic effects. (Ex. 15: 418)

The American Industrial Hygiene Association (AIHA) argued that continuing OSHA's historic approach to work-relationship is particularly important in the case of occupational illnesses because:

Occupational illnesses differ from injuries in that minor or early symptoms of illness are often an important indicator of a more serious disease state, while a minor injury usually goes away without further developments. By the time serious disabling symptoms have surfaced. a disease may be very far progressed and irreversible. Training courses such as Hazard Communication are geared toward educating the workforce to recognize and report symptoms of overexposure, presumably for disease prevention. AIHA does not want this information to be de-emphasized or lost (Ex. 15: 153).

Alternative 3 described in the proposal would have required that an injury or illness be considered work related "if the worker ever experienced a workplace event or exposure that had any possibility of playing a role in the case." This "mere possibility" test is substantially different than OSHA's historical definition of work-relationship, which required that the injury or illness have a tangible connection with the work environment. Although some commenters supported Alternative 3, apparently on the assumption that it was in fact OSHA's proposed definition, analysis of these comments suggests that the parties involved recognized that an injury must have a real, not merely theoretical, link to work to be work-related. No commenter suggested a rationale for recording cases having only a theoretical or speculative link to work.

OSHA has given careful consideration to all of the comments and testimony received in this rulemaking and has decided to continue to rely in the final rule on the Agency's longstanding definition of work-relationship, with one modification. That modification is the addition of the word "significantly" before "aggravation" in the definition of work-relatedness set forth in final rule section 1904.5. The relevant portion of the section now states "an injury or illness is to be considered work-related if an event or exposure in the work environment either caused or contributed to the injury or illness or significantly aggravated a pre-existing injury or illness" (emphasis added).

In the final rule, OSHA has restated the presumption of work-relationship to clarify that it includes any non-minor injury or illness occurring as a result of an event or exposure in the work environment, unless an exception in paragraph 1904.5(b)(2) specifically applies. OSHA believes that the final rule's approach of relying on the geographic presumption, with a limited number of exceptions, is more appropriate than the alternative approaches, for the following reasons.

One important distinction between the geographic test for causation and the alternative causation tests is that the geographic test treats a case as work-related if it results in whole or in part from an event or exposure occurring in the work environment, while the alternative tests would only cover cases in which the employer can determine the degree to which work factors played a causal role. Reliance on the geographic presumption thus covers cases in which an event in the work environment is believed likely to be a causal factor in an injury or illness but the effect of work cannot be quantified. It also covers cases in which the injury or illness is not caused by uniquely occupational activities or processes. These cases may arise, for example, when: (a) an accident at work results in an injury, but the cause of the accident cannot be determined; (b) an injury or illness results from an event that occurs at work but is not caused by an activity peculiar to work, such as a random assault or an instance of horseplay; (c) an injury or illness results from a number of factors, including both occupational and personal causes, and the relative contribution of the occupational factor cannot be readily measured; or (d) a pre-existing injury or illness is significantly aggravated by an event or exposure at work.

As discussed in the Legal Authority section, the statute's language and the Legislative History support a definition of work-relationship that encompasses all injuries and illnesses resulting from harmful events and exposures in the work environment, not only those caused by uniquely occupational activities or processes. A number of commenters acknowledged the broad purposes served by OSHA's recordkeeping requirements and urged continued reliance on the former rule's definition of "work-related" (see, e.g., Exs. 15: 65, 198, 350, 369, 418). For example, the AFL-CIO noted, "[o]ur experience has shown us that when comprehensive records of all possible cases are kept, patterns of injury and illness emerge, enabling us to target problem areas/factors that previously may not have been associated with that specific work environment" (Ex. 15: 418) (emphasis added).

On the other hand, those commenters favoring the "quantified occupational cause" test or the "unique occupational activity" test maintained that injury and illness records have more limited functions. Some commenters argued that because OSHA's mission is to eliminate preventable occupational injuries and illnesses, the determination of work-relationship must turn on whether the case could have been prevented by the employer's safety and health program. The Dow Chemical Company expressed this view as follows:

. . . Credibility in this regulation rests on whether the recorded data accurately reflects the safety and health of the workplace. Including events where the workplace had virtually no involvement undermines the credibility of the system and results in continued resistance to this regulation (Ex. 15: 335B).

The law firm of Constangy, Brooks and Smith, LLC, urged OSHA to adopt the proposal's second alternative ("predominant cause") because cases that are "predominantly caused by workplace conditions" are the ones most likely to be preventable by workplace controls. Their comment stated, "[s]ince OSHA's ultimate mission is the prevention of workplace injuries and illnesses, it is reasonably necessary to require recording only when the injury or illness can be prevented by the employer" (Ex. 15-345). Other commenters opposed the recording of cases in which the injury or illness arises while the employee is on break, in the rest room, or in storage areas located on the employer's premises. These commenters claimed that use of the geographic presumption results in recording many injuries and illnesses that have little or no relationship to the work environment (see, e.g., Exs. 15: 231, 423, 424G).

OSHA believes that the views of Dow Chemical and others in support of the proposal's alternative tests for work-relationship reflect too narrow a reading of the purposes served by the OSHA injury and illness records. Certainly, one important purpose for recordkeeping requirements is to enable employers, employees, and OSHA to identify hazards that can be prevented by compliance with existing standards or recognized safety practices. However, the records serve other purposes as well, including providing information for future scientific research on the nature of causal connections between the work environment and the injuries and illnesses sustained by employees. For example, the records kept by employers under Part 1904 produced useful data on workplace assaults and murders, which has permitted OSHA, employers, and others to focus on the issue of violence in the workplace. This has led, in turn, to efforts to reduce the number of such cases by implementing preventive measures. Although this issue was not anticipated by the 1904 system, the broad collection of injury, illness and fatality data allowed useful information to be extracted from the 1904 data. As discussed in the Legal Authority section, these purposes militate in favor of a general presumption of work-relationship for injuries and illnesses that result from events or exposures occurring in the work environment, with exceptions for specific types of cases that may safely be excluded without significantly impairing the usefulness of the national job-related injury and illness database.

At the same time, OSHA is sensitive to the concerns of some commenters that the injury and illness records are perceived as a measure of the effectiveness of the employer's compliance with the Act and OSHA standards. OSHA emphasizes that the recording of an injury or illness on the Log does not mean that a violation has occurred. The explanatory materials accompanying the revised OSHA Forms 300 and 301 contain the following statement emphasizing this point: "Cases listed on the Log of Work-Related Injuries and Illnesses are not necessarily eligible for Workers Compensation or other insurance benefits. Listing a case on the Log does not mean that the employer or worker was at fault or that an OSHA standard was violated."

The Alternative Tests for Work-Relationship Will Likely Lead Both to Inconsistent Determinations and to Underreporting of Cases

Under the first two alternative tests for work-relationship described in the proposal, the decision on work-relationship would depend upon the degree to which the injury or illness resulted from distinctly occupational causes. Whether labeled "sole cause," "predominant cause," or "significant cause," these alternative tests would require the employer, in each case, to distinguish between the occupational and non-occupational causal factors involved, and to weigh the contribution of the occupational factor or factors. Requiring the occupational cause to be quantified in this way creates practical problems militating against the use of these alternative tests in the final recordkeeping rule.

The most serious problem is that there is no reliable, objective method of measuring the degree of contribution of occupational factors. The absence of a uniform methodology for assessing the extent of work contribution caused several industry commenters to endorse the former rule's position on work-relationship. For example, the American Automobile Manufacturers Association (AAMA) noted that an ideal system would focus on cases in which the work environment was a major contributor to the injury or illness. Nevertheless, the AAMA argued against adopting the predominant cause test, stating: "until a system is developed in which employers can measure objectively and consistently whether or not the work environment is a major contributor to a workplace injury or illness, we favor continuing the definition of work-relationship as it currently exists" (Ex. 15: 409). The Ford Motor Co. also argued in favor of continuing the existing definition:

Ford feels that the work environment should be a major contributor to an injury or illness for the case to be considered work-related. However, we are unsure how employers can measure objectively, consistently and equally whether the work environment is a major contributor. The use of a checklist by a health care provider to determine whether the work environment was a major contributor for a case to be considered work-related would be overly burdensome and subjective. Until a system is developed by which employers can measure objectively, consistently and equally whether or not the work environment is a major contributor to a workplace injury or illness, we favor continuing the definition of work relationship as it currently exists (Ex. 15: 347).

Based on a review of the record, OSHA agrees with those commenters who supported a continuation of the Agency's prior practice with regard to reliance on the geographic presumption for determinations of work-relatedness. OSHA finds that this approach, which includes all cases with a tangible connection with work, better serves the purposes of recordkeeping. Accordingly, the final rule relies on the geographic presumption, with a few limited exceptions, as the recordkeeping system's test for work-relationship.

In addition to the definition of work-relatedness, commenters addressed the issue of who should make the determination of work-relatedness in a given case (see, e.g., Exs. 15: 27, 35, 102, 105, 127, 193, 221, 281, 305, 308, 324, 325, 341, 345, 347, 385, 387, 390, 392, 397, 420). Some commenters believed that a trained medical professional should make this determination, while others argued that the employer should make the ultimate decision about the work-relatedness of occupational injuries and illnesses. Some supported the use of the work-relatedness checklist for specific disorders included by OSHA in the proposal. For example, the American Public Health Association (Ex. 15: 341) commented:

We also believe that work-relatedness should only be established by the documented determination of a qualified health care provider with specific training related to the type of case reported. OSHA's checklist for determining work-relatedness. . . .should be used and expanded to include potentially recordable cases, i.e., excluding first aid treatment.

The Dow Corning Corporation (Ex. 15: 374) argued that the employer should make the determination, albeit with the assistance of a health care professional:

This assessment process should include interviews with knowledgeable people regarding the duties and hazards of the employee's job tasks in addition to the employee interview. If inaccurate or misleading information is given to the health care provider improper or inaccurate conclusions may be reached with regard to the incident cause. A health care provider's assessment of work-relationship is typically viewed as difficult to overcome, even if it is made with incomplete information. We recommend that the health care provider's checklist be used as only one input in the work-relationship decision and that the final decision should still rest with the employer.

Deere and Company (Ex. 15: 253) opposed leaving the determination of work-relatedness to a health care professional:

We strongly disagree with any provision that would allow a physician to make a final determination of work-relatedness. The only time a physician should have any input into the actual determination of work-relatedness is if they are knowledgeable of the employer's workplace environment and the specific job tasks performed by employees. Frequently, physicians will state that a condition was caused by an employee's job without having any knowledge of the specific tasks being performed by the employee. This is an unacceptable usurpation of employers' rights and we oppose any attempt to codify it in a federal regulation.

However, several participants opposed making any work-relatedness checklist mandatory (such as the one OSHA proposed) (see, e.g., Exs. 15: 68, 170, 201, 283, 434). The American Trucking Association's comment (Ex. 15: 397) was typical of this view:

We do not, however, support a requirement that employers must use a mandatory checklist to determine work-relatedness. . . . Because the checklist asks for medical information, the employer would find itself in conflict with the confidentiality requirements imposed under the Americans With Disabilities Act. 29 C.F.R. § 1630.14. Moreover, a mandatory checklist would be unnecessarily time-consuming and subjective. Finally, we note that inclusion of item 5(b), "possible work contribution," biases the checklist in favor of work-relatedness. In the absence of a clear indication of whether or not the workplace caused or substantially caused the condition, asking a provider or employee if it were "possible" that the workplace contributed to or aggravated the injury/illness invites an affirmative response.

OSHA has concluded that requiring employers to rely on a health care professional for the determination of the work-relatedness of occupational injuries and illnesses would be burdensome, impractical, and unnecessary. Small employers, in particular, would be burdened by such a provision. Further, if the professional is not familiar with the injured worker's job duties and work environment, he or she will not have sufficient information to make a decision about the work-relatedness of the case. OSHA also does not agree that health care professional involvement is necessary in the overwhelming majority of cases. Employers have been making work-relatedness determinations for more than 20 years and have performed this responsibility well in that time. This does not mean that employers may not, if they choose, seek the advice of a physician or other licensed health care professional to help them understand the link between workplace factors and injuries and illnesses in particular cases; it simply means that OSHA does not believe that most employers will need to avail themselves of the services of such a professional in most cases.

Accordingly, OSHA has concluded that the determination of work-relatedness is best made by the employer, as it has been in the past. Employers are in the best position to obtain the information, both from the employee and the workplace, that is necessary to make this determination. Although expert advice may occasionally be sought by employers in particularly complex cases, the final rule provides that the determination of work-relatedness ultimately rests with the employer.

Paragraph 1904.5(b)(2) of the final rule contains eight exceptions to the work environment presumption that are intended to exclude from the recordkeeping system those injuries and illnesses that occur or manifest in the work environment, but have been identified by OSHA, based on its years of experience with recordkeeping, as cases that do not provide information useful to the identification of occupational injuries and illnesses and would thus tend to skew national injury and illness statistics. These eight exceptions are the only exceptions to the presumption permitted by the final rule.

(i) Injuries or illnesses will not be considered work-related if, at the time of the injury or illness, the employee was present in the work environment as a member of the general public rather than as an employee. This exception, which is codified at paragraph 1904.5(b)(2)(i), is based on the fact that no employment relationship is in place at the time an injury or illness of this type occurs. A case exemplifying this exception would occur if an employee of a retail store patronized that store as a customer on a non-work day and was injured in a fall. This exception allows the employer not to record cases that occur outside of the employment relationship when his or her establishment is also a public place and a worker happens to be using the facility as a member of the general public. In these situations, the injury or illness has nothing to do with the employee's work or the employee's status as an employee, and it would therefore be inappropriate for the recordkeeping system to capture the case. This exception was included in the proposal, and OSHA received no comments opposing its adoption.

(ii) Injuries or illnesses will not be considered work-related if they involve symptoms that surface at work but result solely from a non-work-related event or exposure that occurs outside the work environment. OSHA's recordkeeping system is intended only to capture cases that are caused by conditions or exposures arising in the work environment. It is not designed to capture cases that have no relationship with the work environment. For this exception to apply, the work environment cannot have caused, contributed to, or significantly aggravated the injury or illness. This exception is consistent with the position followed by OSHA for many years and reiterated in the final rule: that any job-related contribution to the injury or illness makes the incident work-related, and its corollary -- that any injury or illness to which work makes no actual contribution is not work-related. An example of this type of injury would be a diabetic incident that occurs while an employee is working. Because no event or exposure at work contributed in any way to the diabetic incident, the case is not recordable. This exception allows the employer to exclude cases where an employee's non-work activities are the sole cause of the injury or illness. The exception was included in the proposal, and OSHA received no comments opposing its adoption.

(iii) Injuries and illnesses will not be considered work-related if they result solely from voluntary participation in a wellness program or in a medical, fitness, or recreational activity such as blood Alternative Treatments, physical, flu shot, exercise classes, racquetball, or baseball. This exception allows the employer to exclude certain injury or illness cases that are related to personal medical care, physical fitness activities and voluntary blood donations. The key words here are "solely" and "voluntary." The work environment cannot have contributed to the injury or illness in any way for this exception to apply, and participation in the wellness, fitness or recreational activities must be voluntary and not a condition of employment.

This exception allows the employer to exclude cases that are related to personal matters of exercise, recreation, medical examinations or participation in blood Alternative Treatments programs when they are voluntary and are not being undertaken as a condition of work. For example, if a clerical worker was injured while performing aerobics in the company gymnasium during his or her lunch hour, the case would not be work-related. On the other hand, if an employee who was assigned to manage the gymnasium was injured while teaching an aerobics class, the injury would be work-related because the employee was working at the time of the injury and the activity was not voluntary. Similarly, if an employee suffered a severe reaction to a flu shot that was administered as part of a voluntary inoculation program, the case would not be considered work-related; however, if an employee suffered a reaction to medications administered to enable the employee to travel overseas on business, or the employee had an illness reaction to a medication administered to treat a work-related injury, the case would be considered work-related.

This exception was included in the proposal, and received support from a number of commenters (see, e.g., Exs. 15: 147, 181, 188, 226, 281, 304, 341, 345, 363, 348, 373). Other commenters supported this proposal but suggested consolidating it with the proposed exception for voluntary activities away from the employer's establishment (see, e.g., Exs. 15-176, 231, 248, 249, 250, 273, 301). OSHA has decided not to combine this exception with another exception because questions are often asked about injuries and illnesses that arise at the employer's establishment and the Agency believes that a separate exception addressing voluntary wellness programs and other activities will provide clearer direction to employers.

(iv) Injuries and illnesses will not be considered work-related if they are solely the result of an employee eating, drinking, or preparing food or drink for personal consumption (whether bought on the premises or brought in). This exception responds to a situation that has given rise to many letters of interpretation and caused employer concern over the years. An example of the application of this exception would be a case where the employee injured himself or herself by choking on a sandwich brought from home but eaten in the employer's establishment; such a case would not be considered work-related under this exception. On the other hand, if the employee was injured by a trip or fall hazard present in the employer's lunchroom, the case would be considered work-related. In addition, a note to the exception makes clear that if an employee becomes ill as a result of ingesting food contaminated by workplace contaminants such as lead, or contracts food poisoning from food items provided by the employer, the case would be considered work-related. As a result, if an employee contracts food poisoning from a sandwich brought from home or purchased in the company cafeteria and must take time off to recover, the case is not considered work related. On the other hand, if an employee contracts food poisoning from a meal provided by the employer at a business meeting or company function and takes time off to recover, the case would be considered work related. Food provided or supplied by the employer does not include food purchased by the employee from the company cafeteria, but does include food purchased by the employer from the company cafeteria for business meetings or other company functions. OSHA believes that the number of cases to which this exception applies will be few. This exception was included in the proposal and received generally favorable comments (see, e.g., Exs. 15: 31, 78, 105, 159, 176, 181, 184, 188, 345, 359, 428).

(v) Injuries and illnesses will not be considered work-related if they are solely the result of employees doing personal tasks (unrelated to their employment) at the establishment outside of their assigned working hours. This exception, which responds to inquiries received over the years, allows employers limited flexibility to exclude from the recordkeeping system situations where the employee is using the employer's establishment for purely personal reasons during his or her off-shift time. For example, if an employee were using a meeting room at the employer's establishment outside of his or her assigned working hours to hold a meeting for a civic group to which he or she belonged, and slipped and fell in the hallway, the injury would not be considered work-related. On the other hand, if the employee were at the employer's establishment outside his or her assigned working hours to attend a company business meeting or a company training session, such a slip or fall would be work-related. OSHA also expects the number of cases affected by this exception to be small. The comments on this exception are discussed in more detail in the section concerning proposed Exception B-5, Personal Tasks Unrelated To Employment Outside of Normal Working Hours, found later in this document.

(vi) Injuries and illnesses will not be considered work-related if they are solely the result of personal grooming, self-medication for a non-work-related condition, or are intentionally self-inflicted. This exception allows the employer to exclude from the Log cases related to personal hygiene, self-administered medications and intentional self-inflicted injuries, such as attempted suicide. For example, a burn injury from a hair dryer used at work to dry the employee's hair would not be work-related. Similarly, a negative reaction to a medication brought from home to treat a non-work condition would not be considered a work-related illness, even though it first manifested at work. OSHA also expects that few cases will be affected by this exception.

(vii) Injuries will not be considered work-related if they are caused by motor vehicle accidents occurring in company parking lots or on company access roads while employees are commuting to or from work. This exception allows the employer to exclude cases where an employee is injured in a motor vehicle accident while commuting from work to home or from home to work or while on a personal errand. For example, if an employee was injured in a car accident while arriving at work or while leaving the company's property at the end of the day, or while driving on his or her lunch hour to run an errand, the case would not be considered work-related. On the other hand, if an employee was injured in a car accident while leaving the property to purchase supplies for the employer, the case would be work-related. This exception represents a change from the position taken under the former rule, which was that no injury or illness occurring in a company parking lot was considered work-related. As explained further below, OSHA has concluded, based on the evidence in the record, that some injuries and illnesses that occur in company parking lots are clearly caused by work conditions or activities -- e.g., being struck by a car while painting parking space indicators on the pavement of the lot, slipping on ice permitted to accumulate in the lot by the employer -- and by their nature point to conditions that could be corrected to improve workplace safety and health.

Paragraph 1904.5(b)(2)(viii) allows the employer to exclude cases of common cold or flu, even if contracted while the employee was at work. However, in the case of other infectious diseases such as tuberculosis, brucellosis, and hepatitis C, employers must evaluate reports of such illnesses for work relationship, just as they would any other type of injury or illness.

(ix) Mental illness will not be considered work-related unless the employee voluntarily provides the employer with an opinion from a physician or other licensed health care professional with appropriate training and experience (psychiatrist, psychologist, psychiatric nurse practitioner, etc.) stating that the employee has a mental illness that is work-related.

Exception (ix) is an outgrowth of proposed Exception B-11 -- Mental illness, unless associated with post-traumatic stress. There were more than 70 comments that addressed the issue of mental illness recordkeeping. Two commenters suggested that OSHA postpone any decision on the issue: the National Safety Council (Ex. 15: 359) recommended further study, and the AFL-CIO (Ex. 15: 418) stated that the problem of mental illness in the workplace was so prevalent and so important that it should be handled in a separate rulemaking devoted to this issue.

A few commenters, including NIOSH (Ex. 15: 407), the American Psychological Association (Ex. 15: 411), the AFL-CIO (Ex. 14: 418), the United Steelworkers of America (Ex. 15: 429), and the United Brotherhood of Carpenters Health and Safety Fund of North America (Ex. 15: 350) argued that recording should not be limited to post-traumatic stress as OSHA had proposed but should instead include a broader range of mental disorders. The primary arguments of this group of comments were:

§ Workers are afflicted with a number of mental disorders caused or exacerbated by work, and the statistics should include those disorders just as they include physical disorders;

§ If the records include only post-traumatic stress as a mental disorder, many work-related cases of mental illness will go unreported (6,000 mental illness cases are reported to the BLS and involve days away from work, but less than 10% of these are post-traumatic stress cases), and the statistics will be skewed and misinterpreted;

§ Workers' compensation does not restrict compensable mental illnesses to post-traumatic stress cases;

§ Employers are recording and reporting all mental disorders now and thus would not be burdened by continuing the practice.

Arguments in support of treating mental illnesses no differently from any other injury or illness were made by the American Psychological Association (Ex. 15: 411):

The American Psychological Association strongly opposes OSHA's proposal to consider a mental illness to be work related only if it is "associated with post-traumatic stress." We feel that this proposal disregards an accumulating body of research showing the relationship between mental health/illness and workplace stressors. Mental illness associated with post traumatic stress is only one form of mental illness and use of this singular definition would exclude much of the mental illness affecting our nation's workforce.

Job stress is perhaps the most pervasive occupational health problem in the workplace today. There are a number of emotional and behavioral results and manifestations of job stress, including depression and anxiety. These mental disorders have usually been captured under the "mental illness" category but would no longer be recognized if the proposed reporting guidelines were enacted.

The 1985 National Health Interview Survey (Shilling & Brackbill, 1987) indicated that approximately 11 million workers reported health-endangering levels of "mental stress" at work. A large and growing body of literature on occupational stress has identified certain job and organizational characteristics as having deleterious effects on the psychological and physical health of workers, including their mental health. These include high workload demands coupled with low job control, role ambiguity and conflict, lack of job security, poor relationships with coworkers and supervisors, and repetitive, narrow tasks (American Psychological Association, 1996). These include role stressors and demands in excess of control. More precise analyses reveal that specific occupations and job factors present particular risks. For example, machine-paced workers (involving limited worker control of job demands) have one of the highest levels of anxiety, depression, and irritation of 24 occupations studied (Caplan et al., 1975). Health professionals (e.g., physicians, dentists, nurses, and health technologists) have higher than expected rates of suicide which is most often related to depression (Milham, 1983) and of alcohol and drug abuse (Hoiberg, 1982). Nurses and other health care workers have increased rates of hospitalizations for mental disorders (Gundersson & Colcord, 1982; Hoiberg, 1982). This information about specific risks within different occupations provides important information for possible intervention and training to improve conditions while at the same time, indicating the possibility of specific stressors that need to be addressed within the job. This type information would be lost with the proposed reporting guidelines.

Fourteen commenters opposed having to record mental illness cases of any kind (Exs. 15: 78, 133, 184, 248, 249, 250, 304, 348, 378, 395, 406, 409, 412, 424). Their primary arguments were:

§ It is often impossible, even for a health care professional, to determine objectively which mental disorders are work-related and which are not;

§ Workers have a right to privacy about mental conditions that should not be violated; employers fear the risk of invasion of privacy lawsuits if they record these cases on "public records"; because of confidentiality concerns, workers are unlikely to disclose mental illnesses, and employers will therefore be unable to obtain sufficient information to make recordability determinations;

§ Mental illnesses are beyond the scope of the OSHA Act; Congress intended to include only "recognized injuries or illnesses";

§ Recording mental disorders opens the door to abuse; workers may "fake" mental illnesses, and unions may encourage workers to report mental problems as a harassment tactic; and

The American Iron and Steel Institute (AISI) (Ex.15: 395) expressed the concerns of the group of employers opposed to any recording of mental conditions:

OSHA should eliminate its proposed recording requirements for mental illness. OSHA's proposed rule includes changes in an employee's psychological condition as an "injury or illness," and [proposed] Appendix A presumes that mental illness "associated with post-traumatic stress" is work related. Employers, employees, and OSHA have been wrestling for 25 years with the proper recording of fairly simple injuries like back injuries, sprains, and illnesses caused by chemical exposures. Requiring employers to record something as vague as psychological conditions will impose impossible burdens on employers (and compliance officers) and thus will create an unworkable recordkeeping scheme.

Moreover, too little is known about the etiology of most mental conditions to justify any presumption or conclusion that a condition that surfaces at work was "caused" by something in the work environment. It is hard to imagine a mental illness appearing at work that is not a manifestation of a preexisting condition or predisposition. Thus, the only sensible approach is to exclude all mental illnesses from recording requirements.

Many commenters from business and trade associations either agreed with OSHA's proposal or recommended an even stricter limitation on recordable mental disorders (see, e.g., Exs. 33, 15: 27, 31, 38, 46, 79, 122, 127, 132, 153, 170, 176, 181, 199, 203, 226, 230, 231, 273, 277, 289, 301, 305, 307, 308, 313, 325, 332, 352, 353, 368, 384, 387, 389, 392, 410, 427, 430, 434). Points raised by these commenters included recommendations that OSHA should require:

§ Recording only of those mental illnesses that arise from a single, work-related traumatic or catastrophic event, such as a workplace explosion or an armed robbery;

§ Recording only of those mental illnesses that are directly and substantially caused by a workplace incident;

§ Recording only of diagnosed mental illnesses resulting from a single workplace event that is recognized as having the potential to cause a significant and severe emotional response;

§ Recognition only of post-traumatic stress cases or related disorders that include physical manifestations of illness and that are directly related to specific, objectively documented, catastrophic work-related events; and

§ Recording only of diagnosed conditions directly attributable to a traumatic event in the workplace, involving either death or severe physical injury to the individual or a co-worker.

Several commenters suggested the use of a medical evaluation to determine diagnosis and/or work-relationship in cases of mental illness (see, e.g., Exs. 15: 65, 78, 105, 127, 170, 181, 184, 226, 230). For example, the Aluminum Company of America (Ex. 15: 65) stated that:

OSHA should define mental health conditions for recordkeeping purposes as conditions diagnosed by a licensed physician or advanced health care practitioner with specialized psychiatric training (i.e., psychiatric nurse practitioner). Work-relatedness of the mental health condition should be determined by a psychiatric independent medical evaluation.

A comment from the Department of Energy (Ex. 15: 163) stated that any diagnosis of mental illness should be made by at least two qualified physicians, and CONSOL Inc. (Ex. 15: 332) and Akzo Nobel (Ex. 15: 387) wanted the rule to require that any such diagnosis meet the criteria of the Diagnostic and Statistical Manual, Version IV (DSM-IV). Commenters had different opinions about the minimum qualifications necessary for a health care professional to make decisions about mental health conditions; specifically, some commenters urged OSHA to exclude "counselors" (Ex. 15: 226) or to include "only psychiatrists and Ph.D. psychologists" (Ex. 15: 184).

A number of commenters suggested excluding from the requirement to record any mental illness related to personnel actions such as termination, job transfer, demotions, or disciplinary actions (see, e.g., Exs. 15: 68, 127, 136, 137, 141, 176, 184, 224, 231, 266, 273, 278, 301, 395, 424). The New York Compensation Board (Ex. 15: 68) noted that New York's workers' compensation law excludes such cases by specifying that mental injuries are compensable with the exception of injuries that are the "direct consequence of a lawful personnel decision involving a disciplinary action, work evaluation, job transfer, demotion, or termination taken in good faith by the employer."

Finally, several employers raised the issues of the privacy of an employee with a mental disorder, the need to protect doctor-patient confidentiality, and the potential legal repercussions of employers breaching confidentiality in an effort to obtain injury and illness information and in recording that information (see, e.g., Exs. 15: 78, 153, 170, 195, 260, 262, 265, 277, 348, 392, 401, 406, 409). Some of these commenters suggested that an employer should only have the obligation to record after the employee has brought the condition to the attention of the employer, either directly or through medical or workers' compensation claims, and in no case should doctor-patient confidentiality be breached. (Issues related to confidentiality of the Log are discussed in detail in the summary and explanation of § 1904.35, Employee Involvement.)

After a review of the comments and the record on this issue, OSHA has decided that the proposed exception, which would have limited the work-relatedness (and thus recordability) of mental illness cases to those involving post-traumatic stress, is not consistent with the statute or the objectives of the recordkeeping system, and is not in the best interest of employee health. The OSH Act is concerned with both physical and mental injuries and illnesses, and in fact refers to "psychological factors" in the statement of Congressional purpose in section 2 of the Act (29 U.S.C. 651(b)(5)).

In addition, discontinuing the recording of mental illnesses would deprive OSHA, employers and employees, and safety and health professionals of valuable information with which to assess occupational hazards and would additionally skew the statistics that have been kept for many years. Therefore, the final rule does not limit recordable mental disorders to post traumatic stress syndrome or any other specific list of mental disorders. OSHA also does not agree that recording mental illnesses will lead to abuse by employees or others. OSHA has required the recording of these illnesses since the inception of the OSH Act, and there is no evidence that such abuse has occurred.

However, OSHA agrees that recording work-related mental illnesses involves several unique issues, including the difficulty of detecting, diagnosing and verifying mental illnesses; and the sensitivity and privacy concerns raised by mental illnesses. Therefore, the final rule requires employers to record only those mental illnesses verified by a health care professional with appropriate training and experience in the treatment of mental illness, such as a psychiatrist, psychologist, or psychiatric nurse practitioner. The employer is under no obligation to seek out information on mental illnesses from its employees, and employers are required to consider mental illness cases only when an employee voluntarily presents the employer with an opinion from the health care professional that the employee has a mental illness and that it is work related. In the event that the employer does not believe the reported mental illness is work-related, the employer may refer the case to a physician or other licensed health care professional for a second opinion.

OSHA also emphasizes that work-related mental illnesses, like other illnesses, must be recorded only when they meet the severity criteria outlined in § 1904.7. In addition, for mental illnesses, the employee's identity must be protected by omitting the employee's name from the OSHA 300 Log and instead entering "privacy concern case" as required by § 1904.29.

The proposed rule contained eleven exceptions to the geographic presumption. Some of these exceptions are included in the final rule, and therefore are discussed above, while others were rejected for various reasons. The following discussion addresses those proposed exemptions not adopted in the final rule, or not adopted in their entirety.

Proposed Exception B-5. Personal Tasks Unrelated To Employment Outside of Normal Working Hours. The proposed rule included an exception for injuries and illnesses caused solely by employees performing personal tasks at the establishment outside of their normal working hours. Some aspects of this proposed exception have been adopted in the final, but others have not. Almost all the comments on this proposed exception supported it (see, e.g., Exs. 15: 31, 78, 105, 121, 159, 281, 297, 336, 341, 350), and many suggested that the exception be expanded to include personal tasks conducted during work hours (see, e.g., Exs. 15: 176, 184, 201, 231, 248, 249, 250, 273, 301, 335, 348, 374). Caterpillar, Inc. (Ex. 15: 201) offered an opinion representative of the views of these commenters: "We agree with this exception but it should be expanded to include any personal tasks performed during work hours if the work environment did not cause the injury or illness. Expanding this exemption will be consistent with the exemptions for voluntary wellness program participation and eating, drinking, and preparing one's own food."

One commenter disagreed with the proposed exception (the Laborers Safety and Health Fund of North America (Ex. 15: 310)) and cited as a reason the difficulty of determining the extent to which, for example, a case involving an employee misusing a hazardous chemical after hours because he or she did not receive the necessary Right-to-Know training from the employer would qualify for this exception.

Several commenters suggested that OSHA clarify what it meant by the terms "personal tasks" and "normal working hours" (see, e.g., Exs. 15: 102, 304, 345). For example, a representative of Constangy, Brooks & Smith recommended that:

More explanation be provided regarding the further limitation on this exclusion. For example, does this section of the proposal envision the exclusion of injuries and illnesses resulting from personal tasks performed during overtime (i.e., outside of normal working hours)? If I am injured while talking to my spouse on the phone during regular business hours, must the case be recorded, while if the same injury occurs during overtime, the case is non-recordable? Also, how are injuries to salaried employees (who are exempt from overtime) treated under this aspect of the proposal? I submit that if these issues are not fully "fleshed out" in the proposal or its preamble, this subparagraph will result in the creation of more questions than it resolves.

The National Federation of Independent Business (NFIB) (Ex. 15: 304) asked OSHA "to specify that the 'normal working hours' refers to the work schedule of the employee not the employer. If this distinction is not made clear, this proposal arguably could deny this exemption to establishments which operate during non-standard operating hours (e.g., 24 hours a day, weekends, after 5 PM, etc.) -- and we assume this is not OSHA's intent."

OSHA believes that injuries and illnesses sustained by employees engaged in purely personal tasks at the workplace, outside of their assigned working hours, are not relevant for statistical purposes and that information about such injuries and illnesses would not be useful for research or other purposes underlying the recordkeeping requirements. OSHA has therefore decided to include some parts of the proposed exception in the final rule. Additional language has been added to the exception since the proposal to clarify that the exception also applies when the employee is on the premises outside of his or her assigned working hours, as the NFIB pointed out.

OSHA does not agree, however, with those commenters who suggested that the exception be expanded to include personal tasks performed by employees during work hours. As discussed in preceding sections of this summary and explanation and in the Legal Authority discussion, there are strong legal and policy reasons for treating an injury or illness as work-related if an event or exposure in the work environment caused or contributed to the condition or significantly aggravated a pre-existing condition. Under this "but-for" approach, the nature of the activity the employee was engaged in at the time of the incident is not relevant, except in certain limited circumstances. Moreover, OSHA believes that it would be difficult in many cases for employers to distinguish between work activities and personal activities that occur while the employee is on-shift. Accordingly, the final rule codifies parts of this proposed exception in paragraph 1904.5(b)(v) in the following form: "The injury or illness is solely the result of an employee doing personal tasks (unrelated to their employment) at the establishment outside of the employee's assigned working hours."

Proposed Exception B-6. Cases Resulting From Acts of Violence by Family Members or Ex-spouses When Unrelated to Employment, Including Self-inflicted Injuries. The final rule does not exempt workplace violence cases from the Log, although it does allow employers to exclude cases that involve intentionally self-inflicted injuries. The final rule thus departs substantially from the proposal in this respect. The proposed exception, which would have exempted domestic violence and self-inflicted cases from the Log, drew many comments. The comments generally fell into four categories: (1) those urging OSHA to require the recording of all cases of violence occurring at the establishment; (2) those recommending that no violence cases at the establishment be recorded; (3) those recommending recordation only of violence cases perpetrated by certain classes of individuals; and (4) those urging OSHA to require the recording of cases involving violence related to employment without regard to the perpetrator. The comments on the proposed exception are discussed below.

No exemption/record all injuries and illnesses arising from violent acts. A number of commenters objected to OSHA's proposed exemption of domestic violence cases from the list of recordable injuries, arguing that all acts of violence occurring at the workplace should be recorded (see, e.g., Exs. 15: 31, 54, 56, 88, 90, 91, 93, 94, 99, 101, 103, 104, 106, 111, 114, 115, 144, 186, 187, 238, 345, 362, 407, 418, 439). For example, the North Carolina Department of Labor stated that "if an employer must log the injuries sustained as a result of workplace violence then the employer may also institute needed security measures to protect the employees at the establishment. An employer should be required to log any 'preventable' injury (above first aid) that an employee sustains at the establishment" (Ex. 15: 186). The Miller Brewing Company also supported recording all acts of workplace violence, based on the following rationale: "I envision a scenario involving an angry husband attempting to kill his wife but, because he is a "bad shot," another employee is killed. Why should killing an innocent bystander be a reportable event, whereas a fatality involving a spouse is excluded?" (Ex. 15: 442).

Exception for all violent acts. There were commenters who thought injuries and illnesses resulting from violence were outside of OSHA's purview and should not be recorded at all (see, e.g., Exs. 15: 28, 75, 96, 107, 203, 254, 289). For example, the Quaker Oats Company (Ex. 15: 289) stated that "[w]orkplace violence in any form is a personal criminal act, and in no way, shape or form should violence be labeled under hazards in the workplace or even [be] monitored by OSHA. A person who may turn to violent behavior from family, personal, or job dispute is a matter of NLRB [National Labor Relations Board], law enforcement or state employment statutes, not industrial safety." The National Restaurant Association (Ex. 15: 96) agreed:

Congress passed the Occupational Safety and Health Act to regulate workplace hazards dealing with the workplace environment or processes that employers could identify and possibly protect. The Congress did not contemplate that this statute would be used to redress incidents over which the employer has no ability to control, such as the unpredictability of workers or nonworkers committing violent, tortuous acts towards others. This issue was litigated unsuccessfully by OSHA in Secretary of Labor v. Megawest Financial, Inc., OSHRC Doc. No. 93-2879 (June 19, 1995). OSHA apparently is attempting in this NPR to obtain by regulatory fiat what was rejected by case law and to displace state tort law actions by using the OSH Act to police social behavior.

Recording work-related violence except acts of certain classes of individuals. There were many commenters who supported the proposed exception, which would only have excluded acts of violence on employees committed by family members and ex-spouses and self-inflicted injuries and illnesses. The proposed exception as drafted was supported by some commenters (see, e.g., Exs. 15: 78, 198, 350, 359). Others thought the exception should be expanded to include not only family members and ex-spouses, but also live-in partners, friends, and other intimates (see, e.g., Exs. 15: 80, 122, 153, 181, 213, 325, 363, 401), while others argued that the exemption should apply to the general public, i.e., to all people (see, e.g., Exs. 15: 9, 111, 119, 151, 152, 179, 180, 239, 260, 262, 265, 272, 303, 304, 341, 356, 375, 401, 430).

Typical of comments in support of a broader exception were the remarks of the National Oilseed Processors Association (Ex. 15: 119):

The only time violence in the work place should be considered work-related is when it is associated with a work issue and committed by an employee or other person linked to the business, e.g., a customer. Any other act of violence is not under the control of the employer and should not be considered work-related.

Exempting acts of violence based strictly on acts committed by family members, a spouse, or when self-inflicted is too limited. Instead, the exemption should be based on the relationship of the perpetrator to the employer. The employer should be no more responsible for some random act of violence by a crazy individual walking in off the street who is in no way associated with the employer than it should be for an act of violence by a family member.

Southern California Edison (Ex. 15: 111) stated that "violence is another example that should be excluded from being work-related if the employee personally knows the attacker. This would include family members or coworkers. Only those acts of violence that result from random criminal activity should be included (i.e., robbery, murder, etc.)." TU Services (Ex. 15: 262) recommended "that only cases that involve acts of violence that are the result of random criminal activity should be recorded. Cases that involve anyone with a personal relationship with the employee should be excluded." The American Feed Industry Association (Ex. 15: 204) and United Parcel Service (Ex. 15: 424), on the other hand, argued that cases involving workplace violence should only be recorded if the perpetrator was a fellow employee.

Record all violent acts directly related to employment regardless of who commits the act. Commenters favoring this approach suggested that violence by family members or others should be recorded if linked to work, but that all personal disputes should be exempt (see, e.g., Exs. 15: 105, 146, 176, 184, 231, 273, 297, 301, 313, 336, 348, 352, 353, 374, 389, 392). The Workplace Health and Safety Council (Ex. 15: 313) proposed the following exception:

Cases will not be considered work-related if they result solely from acts of violence committed by one's family, or ex-spouse, or other persons when unrelated to the worker's employment, including intentionally self-inflicted injuries. Violence by persons on the premises in connection with the employer's business (including thieves and former employees) is considered work related even if committed by one's family or ex-spouse.

The American Ambulance Association (Ex. 15: 226) stated simply: "AAA believes that OSHA should define what is work-related violence and assume that all other acts are not work-related, and eliminate the family and non-family distinction." The United Auto Workers (Ex. 15: 438) agreed:

Incidents of intentional violence should be recorded only if they arise from employment activities. Incidents between employees, or between employees and non-employees which rise from personal disputes should not be recorded. Existing data show that the number of incidents of interpersonal violence between coworkers or workers and intimates is small, although these incidents do get high visibility. Therefore, exclusion of these small number of cases will have little effect on statistical measures.

Some commenters urged OSHA to place some restrictions on the proposed exception. For example, two commenters argued that cases involving violence should only be recorded for occupations where there is a reasonable potential of encountering violence (Exs. 15: 335, 409). The American Automobile Manufacturers Association (AAMA) stated that:

Workplace violence as a reasonable function of an employee's employment should be recorded, for example: a cashier injured in a robbery attempt at a 24-hour retail establishment. An example of "unreasonable" recordable workplace violence that should not be recordable (i.e., where an employee was simply "in the wrong place at the wrong time") would be a flight crew that perishes mid-flight from a terrorist's bomb. These cases have nothing to do with the individual's employer, only that they happened to be victims at the employer's place of employment. It is AAMA's understanding that the purpose of the subject standard is to collect information pertaining to injuries and illnesses that arise out of conditions in the workplace, with the end objective being to use that information to correct or mitigate these conditions so as to prevent additional injuries or illnesses.

Caterpillar Inc. (Ex. 15: 201) suggested that "a predominant contributor concept, similar to that being proposed to help establish work-relatedness, could be utilized in cases where the clear cause of violence is not readily apparent."

In the final rule, OSHA has decided not to exclude from recording those injury and illness cases involving acts of violence against employees by family members or ex-spouses that occur in the work environment or cases involving other types of violence-related injuries and illnesses. The final rule does exempt from recording those cases resulting from intentionally self-inflicted injuries and illnesses; these cases represent only a small fraction of the total number of workplace fatalities (three percent of all 1997 workplace violence fatalities) (BLS press release USDL 98-336, August 12, 1998). OSHA believes that injuries and illnesses resulting from acts of violence against employees at work are work-related under the positional theory of causation. The causal connection is usually established by the fact that the assault or other harmful event would not have occurred had the employee not, as a condition of his or her employment, been in the position where he or she was victimized. Moreover, occupational factors are directly involved in many types of workplace violence, such as assaults engendered by disputes about working conditions or practices, or assaults on security guards or cashiers and other employees, who face a heightened risk of violence at work. Accordingly, OSHA does not accept the premise, advanced by some commenters, that workplace violence is outside the purview of the statute.

In some cases, acts of violence committed by a family member or ex-spouse at the workplace may be prevented by appropriate security measures enforced by employers. Moreover, information about workplace injuries due to assaults by family members or ex-spouses is relevant and should be included in the overall injury and illness data for statistical and research purposes. Omitting the proposed exception also obviates the need for employers to make distinctions among various degrees of personal relationships. Accordingly, the final rule does not allow employers to exclude injuries and illnesses resulting from violence occurring in the workplace from their Logs. However, some cases of violence will be excluded under § 1904.5(b)(2)(v), which exempts an injury or ilness that is solely the result of an employee doing personal tasks (unrelated to their employment) at the establishment outside of the employee's assigned working hours. For example, if an employee arrives at work early to use a company conference room for a civic club meeting, and is injured by some violent act, the case would not be considered work related.

OSHA has decided to maintain the exclusion for intentionally self-inflicted injuries that occur in the work environment in the final rule. The Agency believes that when a self-inflicted injury occurs in the work environment, the case is analogous to one in which the signs or symptoms of a pre-existing, non-occupational injury or illness happen to arise at work, and that such cases should be excluded for the same reasons. (see paragraph 1904.5(b)(2)(ii)). The final rule at paragraph 1904.5(b)(2)(vi) therefore includes that the part of exception proposed that applied to injuries and illnesses that are intentionally self-inflicted.

Proposed Exception B-7. Parking Lots and Access Roads. This proposed exception, which in effect would have narrowed the definition of "establishment" to exclude company parking lots, had approximately equal numbers of commenters in favor and opposed. The final rule includes some aspects of the proposed exemption. In favor of recording injuries in parking lots and on access roads were the commenters represented by Exs. 24, 15: 41, 72, 310, 362. Typical of the views of this group was that of the Association of Operating Room Nurses (AORN) (Ex. 15: 72), which noted that:

[e]mployee parking lots should be included in defining "work-related." Perioperative nurses and other surgical service providers may be required on a "call" basis during the night hours. Consequently they enter and leave parking lots at unusual times when traffic in the lots is minimal. These providers may be at increased risk for random violence. Absent the "call" requirement, the employee would not be in the parking lot at the time of the injury. Further, if the employee is paid for travel time to and/or from the facility, injuries occurring during that period should be considered "work-related."

The AFL-CIO (Ex. 15: 362) added that employers may be less likely to provide lighting, security and other controls that could prevent violent assaults in parking lots and access roads if injuries occurring there are not recordable.

The opposite view, in support of the proposed exception for parking lots, was expressed by several employers (see, e.g., Exs. 15: 27, 45, 176, 185, 195, 231, 248, 249, 250, 273, 289, 301, 304, 341, 363). The National Wholesale Druggists Association (NWDA) (Ex. 15: 185) supported the proposed exclusion:

[i]nevitably, activities that take place in the company parking lot or on the company access road are not only outside of the employer's dominion and control but also are most often not related in any way to the employee's work. Including injuries that occur in these locations as part of the OSHA log would lead to an inaccurate reflection of injury data as a whole. OSHA should retain this exemption. An employer has no control over an employee's commute to and from the workplace, with the exception of arrival and departure times for the work day. If OSHA requires the reporting of injuries that occur during the employee's commute, the number of injuries reported would increase dramatically.

The National Federation of Independent Business (Ex. 15: 304) stated that the proposed exception would be consistent with workers' compensation rules.

OSHA has concluded that a limited exception for cases occurring on parking lots is appropriate but that the broader exception proposed is not. The final rule thus provides an exception for motor vehicle injury cases occurring when employees are commuting to and from work. As discussed in the preamble that accompanies the definition of "establishment" (see Subpart G of the final rule), OSHA has decided to rely on activity-based rather than location-based exemptions in the final rule. The parking lot exception in the final rule applies to cases in which employees are injured in motor vehicle accidents commuting to and from work and running personal errands (and thus such cases are not recordable), but does not apply to cases in which an employee slips in the parking lot or is injured in a motor vehicle accident while conducting company business (and thus such cases are recordable). This exception is codified at paragraph 1904.5(b)(2)(vii) of the final rule.

Proposed Exception B-8. Never Engaged in an Activity That Could Have Placed Stress On the Affected Body Part. This proposed exception would have allowed employers not to record cases if no aspect of the worker's job placed stress on the affected body part or exposed the worker to any chemical or physical agent at work that could be associated with the observed injury or illness. This proposed exception received support from a number of employers (see, e.g., Exs. 15: 176, 185, 231, 273, 301, 341, 359, 406). For example, the National Wholesale Druggists' Association stated that "Such injuries or illnesses are obviously not caused by any work-related activities and should therefore be excluded from any reporting and recording requirements' (Ex. 15: 185).

Deleting the word "never" from the proposed exception was also supported by many respondents (see, e.g., Exs. 15: 146, 279, 304, 335, 374, 392, 395, 430, 431, 442). Representative of the latter group is the following comment by the BF Goodrich Company (Ex. 15: 146):

The use of the term "never" in this exemption requires too harsh a test for case evaluation. A back injury should not be recordable because the employee lifted a box 10 years previous to the injury. A more reasonable evaluation criteria meeting the same intent could be stated as below: The injury or illness is not work-related if it cannot be associated with the employee's duties or exposures at work.

Taking an opposing view to the proposed exception were the AFL-CIO (Ex. 15: 418), the United Steelworkers of America (Ex. 15: 429), and the United Brotherhood of Carpenters Health and Safety Fund of North America (Ex. 15: 350). The AFL-CIO stated that:

We believe when evaluating injuries this approach could logically work in most cases, but in cases of chemical exposures and musculoskeletal disorders this logic does not hold merit. If the Agency attempts to apply this approach to the aforementioned types of cases, the employer will have to become an epidemiologist, ergonomist or toxicologist to determine if these cases meet the recordability criteria set forth in this proposal . . . . We encourage the Agency to omit this provision from the final standard. Because of the increasing numbers of workers being medically diagnosed for multiple chemical sensitivity and the exposures some workers receive without any knowledge until years after the incident, the Agency must carefully think about the inclusion of this provision to the final standard.

[T]his [exception] would exclude those cases where symptoms arise at work, but are caused by accidents or exposures away from work. The UBC H&SF agrees with the theory of this provision, but emphasizes that the task placed on employers to determine causation by exposures away from work would in many cases be impossible. Also the apportionment of causation is not discussed in this analysis and would allow some to record cases .01 percent caused by work and others to not record cases 99 percent caused by work. For the foregoing reasons, that this requirement is unworkable, we urge it be dropped from the final rule.

Based on a review of the record on this issue, OSHA has decided not to include this proposed exception in the final rule. On reflection, the proposed language is confusing and would be difficult to apply. The underlying concept, to the extent it has merit, is better covered in the exemption paragraph 1904.5(b)(2)(ii). As discussed in preceding sections of this summary and explanation for section 1904.5, there are sound legal and policy justifications for defining work-relationship broadly to include injuries and illnesses that result from events or exposures in the work environment. The proposed exception would effectively "swallow" the geographic presumption theory of causation underpinning the rule by shifting the focus of enquiry in every case to the employee's specific job duties. As OSHA has noted, the geographic presumption includes some cases in which the illness or injury cannot be directly linked to the stresses imposed by job duties. For example, if an employee trips while walking on a level factory floor and breaks his arm, the injury should be recordable. The comments supporting the proposed exemption do not, in OSHA's view, provide a basis for excluding these types of cases from recording on the Log.

Proposed Exception B-9. Voluntary Community Activities Away From The Employer's Establishment. This proposed exemption drew two comments supporting it as written (Exs. 15: 78, 304), and several other participants recommended that it be expanded to exclude injuries and illnesses that arise from voluntary community activities wherever they occur (see, e.g., Exs. 15: 146, 184, 272, 303, 359). Typical of these comments is one from U.S. West (Ex. 15: 184), which stated that "[e]mphasis should be on the activity that occurred, not the location of the activity."

The United Brotherhood of Carpenters, Health & Safety Fund of North America (Ex. 15: 350) agreed with the proposed exception, except for cases where the employee is present as a condition of employment or in the employer's interest. It commented:

[A]t the surface this exception seems to make perfect sense. However, real employment relationships and real employer-community relationships do not fit such clean characterizations. Many times employees are forced to become "team players" and volunteer for unpaid off-establishment activities. Many employers engage in community "good will" generating activities by having their employees volunteer. For the above reasons we urge that cases occurring away from the employer's establishment be considered work-related if the employee is engaged in any activity in the interest of the employer or is there as a condition of employment.

OSHA has decided not to include this proposed exception in the final rule because the final rule's overall definition of work-environment addresses this situation in a simple and straightforward way. If the employee is taking part in the activity and is either working or present as a condition of employment, he or she is in the work environment and any injury or illness that arises is presumed to be work-related and must then be evaluated for its recordability under the general recording criteria. Thus, if the employee is engaged in an activity at a location away from the establishment, any injury or illness occurring during that activity is considered work-related if the worker is present as a condition of employment (for example, the worker is assigned to represent the company at a local charity event). For those situations where the employee is engaged in volunteer work away from the establishment and is not working or present as a condition of employment, the case is not considered work-related under the general definition of work-relationship. There is thus no need for a special exception.

Proposed Exception B-10. The Case Results Solely From Normal Body Movements, not Job-Related Motions or Contribution from the Work Environment. This proposed exception generated some support (see, e.g., Exs. 15: 107, 147, 173, 185, 341, 348, 373, 392) but also caused much confusion about the meaning of the phrases "normal body movement" and "job-related" (see, e.g., Exs. 15: 80, 83, 89, 98, 146, 176, 225, 226, 231, 239, 273, 301, 304, 313, 352, 353, 355, 359, 406, 424). The following comment by the American Gas Association (Ex. 15: 225) is representative of those in this group:

'[N]ormal body movements' needs clarification since OSHA has not set forth any reasons for excluding it. OSHA's language states that there is an exclusion "* * * provided that activity does not involve a job related motion and the work environment does not contribute to the injury or illness". OSHA goes on to elaborate that illnesses or injuries should not be recorded if they are not related to an identifiable work activity. However, OSHA also states the exclusion would not apply if it involved repetitive motion or if the work environment either caused or contributed to the injury or illness. This language is ambiguous and redundant. Repetitive motion injury/illness conditions should be treated in the same way as any other condition. There should be a work-related exclusion if the work environment did not cause or contribute to the injury/illness.

[T]he definition of work-related resulting from normal body movements is too broad. The definition excludes walking, talking, etc. 'provided the activity does not involve a job-related motion.' Does that mean that if an employee is walking to the rest room and becomes ill, the illness is not work-related, but, if he/she is walking from the rest room back to his/her work station, it is work-related? If the employee is engaged in social talk, the illness is not work-related, but, if he/she is engaged in a conversation regarding some aspect of work, the illness is work-related?

Other commenters objected to the concept of excluding cases resulting from normal body movements from the Log (Ex. 56X, pp. 51, 52; Ex. 15: 418). Walter Jones of the International Brotherhood of Teamsters used the following example:

We do take opposition to some of the exceptions. For cases that result in normal body movement, I'd like to just bring another example up. We have a member who after spending most of his morning sorting about 700 different boxes, on break in a normal, unencumbered motion, dropped his pencil and picked it up, had a back spasm and his back went out. And I know that according to the way the standard is written, or the regulation is written, that this can be attributed to work activity. But the reason we bring it up is we need to be careful in trying to be that exact because an employer will take an uninformed employee and may take liberties (Ex. 56X, pp. 51, 52).

OSHA has decided not to include a recordkeeping exception for injuries or illnesses associated with normal body movements in the final rule. The proposed provision was intended to exclude the recording of cases that happened to occur in the work environment without any real work contribution. However, the comments on this issue have convinced OSHA that the proposed provision is unnecessary, would be unworkable, and would result in incomplete and inconsistent data. The case cited by the Teamsters is but one example of a legitimate work-related injury that could go unrecorded if OSHA were to adopt this provision in the final rule. Further, the final rule already makes clear that injuries and illnesses that result solely from non-work causes are not considered work-related and therefore are excluded from the Log, and establishes the requirements employers must follow to determine work-relationship for an injury or illness when it is unclear whether the precipitating event occurred in the workplace or elsewhere (see paragraph 1904.5(b)(3)). According to the requirements in that section, the employer must evaluate the employee's work duties and the work environment to decide whether it is more likely than not that events or exposures in the work environment either caused or contributed to the condition or significantly aggravated a pre-existing condition. If so, the case is work-related.

In addition to commenting on the eleven proposed exceptions, interested parties suggested adding some exceptions to the final rule. This section contains a discussion of those additional exemptions suggested by commenters but not adopted in the final rule.

Acts of God: The International Dairy Foods Association (IDFA) suggested that OSHA exclude any injury or illness that was "the result of an "Act of God," such as, but not limited to, an earthquake or a tornado" (Ex. 15: 203). OSHA has not adopted such an exception because doing so would not be in keeping with the geographic presumption underpinning this final rule, and would exclude cases that are in fact work-related. For example, if a worker was injured in a flood while at work, the case would be work-related, even though the flood could be considered an act of God. Accordingly, if workplace injuries and illnesses result from these events, they must be entered into the records (for a more detailed discussion of this point, see the Legal Authority section, above).

Phobias: The American Crystal Sugar Company (Ex. 15: 363) suggested that OSHA add an exception from recording for cases involving phobias:

I would also like to suggest exempting an employee's loss of consciousness based on a fear-based phobia, i.e., fainting at the sight of blood. Occasionally an OSHA regulation may require blood tests, such as checking lead levels in blood. There are a few employees that will lose consciousness at the sight of a needle. These phobias are not limited to medical procedures, but may include spiders, snakes, etc. In several of our factories, the occupational health nurse will administer tetanus boosters as a service to our employees. Employees that have a phobia about injections can (and do) lose consciousness, which now makes what was intended as a service an OSHA recordable accident.

OSHA has not included an exception from recording in the final recordkeeping regulation for phobias or any other type of mental illness. The scenario described by the American Crystal Sugar Company, which involved fainting from fear of an injection offered as a service to employees, might be considered non-work-related under the exception codified at paragraph 1904.5(b)(2)(iii), Voluntary participation in a medical activity. OSHA also believes that it would be unreasonable to omit a case of loss of consciousness resulting from the administration of a blood test for lead exposure at work. These tests are necessitated by the employee's exposure to lead at work and are required by OSHA's lead standard (29 CFR 1910.1025). The other scenarios presented by these commenters, involving spiders, snakes, etc., would also be work-related under the geographic presumption.

Illegal activities and horseplay: Several commenters suggested an exception for an employee engaging in illegal activities, horseplay, or failing to follow established work rules or procedures (see, e.g., Exs. 15: 49, 69, 117, 151, 152, 179, 180, 203, 368, 393). The comment of the American Network of Community Options and Resources (ANCOR) (Ex. 15: 393) is representative of those on this issue:

Employees who fail to follow employer training and best practices or violate established policy present a threat not only to other employees and consumers/customers, but also to employers held responsible for the consequences of their actions. For example, ANCOR does not believe that employers should have to use these recording and reporting procedures when illnesses and injuries are a result of an employee engaged in illegal activities or fails/ violates established procedures.

OSHA has not adopted any of these recommended exceptions in the final recordkeeping rule because excluding these injuries and illnesses would be inconsistent with OSHA's longstanding reliance on the geographic presumption to establish work-relatedness. Furthermore, the Agency believes that many of the working conditions pointed to in these comments involve occupational factors, such the effectiveness of disciplinary policies and supervision. Thus, recording such incidents may serve to alert both the employer and employees to workplace safety and health issues.

Non-occupational degenerative conditions: Two commenters also asked OSHA to include in the final rule a recording exception for non-occupational degenerative conditions (Exs. 15: 176, 248) such as high blood pressure, arthritis, coronary artery disease, heart attacks, and cancer that can develop regardless of workplace exposure. OSHA has not added such an exception to the rule, but the Agency believes that the fact that the rule expects employers confronted with such cases to make a determination about the extent to which, if at all, work contributed to the observed condition will provide direction about how to determine the work-relatedness of such cases. For example, if work contributes to the illness in some way, then it is work-related and must be evaluated for its recordability. On the other hand, if the case is wholly caused by non-work factors, then it is not work-related and will not be recorded in the OSHA records.

Determining Whether the Precipitating Event or Exposure Occurred in the Work Environment or Elsewhere

Paragraph 1904.5(b)(3) of the final rule provides guidance on applying the geographic presumption when it is not clear whether the event or exposure that precipitated the injury or illness occurred in the work environment or elsewhere. If an employee reports pain and swelling in a joint but cannot say whether the symptoms first arose during work or during recreational activities at home, it may be difficult for the employer to decide whether the case is work-related. The same problem arises when an employee reports symptoms of a contagious disease that affects the public at large, such as a staphylococcus infection ("staph" infection) or Lyme disease, and the workplace is only one possible source of the infection. In these situations, the employer must examine the employee's work duties and environment to determine whether it is more likely than not that one or more events or exposures at work caused or contributed to the condition. If the employer determines that it is unlikely that the precipitating event or exposure occurred in the work environment, the employer would not record the case. In the staph infection example given above, the employer would consider the case work-related, for example, if another employee with whom the newly infected employee had contact at work had been out with a staph infection. In the Lyme disease example, the employer would determine the case to be work-related if, for example, the employee was a groundskeeper with regular exposure to outdoor conditions likely to result in contact with deer ticks.

In applying paragraph 1904.5(b)(3), the question employers must answer is whether the precipitating event or exposure occurred in the work environment. If an event, such as a fall, an awkward motion or lift, an assault, or an instance of horseplay, occurs at work, the geographic presumption applies and the case is work-related unless it otherwise falls within an exception. Thus, if an employee trips while walking across a level factory floor, the resulting injury is considered work-related under the geographic presumption because the precipitating event -- the tripping accident -- occurred in the workplace. The case is work-related even if the employer cannot determine why the employee tripped, or whether any particular workplace hazard caused the accident to occur. However, if the employee reports an injury at work but cannot say whether it resulted from an event that occurred at work or at home, as in the example of the swollen joint, the employer might determine that the case is not work-related because the employee's work duties were unlikely to have caused, contributed to, or significantly aggravated such an injury.

In paragraph 1904.5(b)(4), the final rule makes an important change to the former rule's position on the extent of the workplace aggravation of a preexisting injury or illness that must occur before the case is considered work-related. In the past, any amount of aggravation of such an injury or illness was considered sufficient for this purpose. The final rule, however, requires that the amount of aggravation of the injury or illness that work contributes must be "significant," i.e., non-minor, before work-relatedness is established. The preexisting injury or illness must be one caused entirely by non-occupational factors.

A number of commenters on OSHA's proposed rule raised the issue of recording injuries that were incurred off the job and then were aggravated on the job (see, e.g., Exs. 15: 60, 80, 95, 107, 176, 201, 204, 213, 281, 308, 313, 338, 368, 375, 395, 396, 406, 424, 427, 428, 441). The National Roofing Contractors Association (NRCA) commented that "[t]his definition [includes] aggravating a pre-existing condition. While NRCA believes that the exemptions provided [in the proposed rule] are a step in the right direction, this provision could require that an employer record an injury that originally occurred outside the employer's workplace. The motion or activity that aggravated the injury may not represent any substantial hazard, yet would still be recorded" (Ex. 15: 441). The United Parcel Service (Ex. 15: 424) objected to the inclusion of the concept of aggravation in the definition of work-relatedness:

[a]nother flaw in the proposal arises from its proposed recording requirement in the case of "aggravation" of prior conditions. As drafted, the rule would require reporting as an occupational injury or illness a musculoskeletal condition arising away from work which becomes aggravated by performing job duties (i.e., the job increases discomfort), when accompanied by swelling or inflammation. Thus, an employee who hurts his wrist playing tennis on the weekend and who returns to his word processing job Monday would have a reportable MSD under the rule. With such criteria for recordation, reported occupational injuries and illnesses would skyrocket, and yet most often these reports would reflect conditions arising away from work.

[i]t is very important that injuries that are not truly work-related not be the subject of mandatory recording. For example, if an employee were injured off the job and came to work to "try it out" (i.e., to see if he or she was capable of performing the normal job functions), resulting pain might be seen as "aggravation" and become recordable on that basis. The true source of injury, however, would be outside the workplace, and recording would produce an artificially inflated rate of injuries and illnesses, and a profile that was inaccurate.

Several commenters were concerned about the aggravation of preexisting injuries in the context of recurrences or new cases (see, e.g., Exs. 15: 210, 204, 338) . For example, Caterpillar Inc. (Ex. 15: 201) stated that:

[b]ack injuries, repetitive motion injuries, and other chronic conditions which have degenerative or aging causal factors often recur without a new work accident and further without a new work accident capable of causing the underlying condition. Even if a new work accident occurs, the accident should be serious enough to cause the underlying condition before the new case presumption is applicable. The effect of this would be to eliminate minor aggravation of preexisting conditions from consideration as new injuries.

LeRoy E. Euvard, Jr., of the Safety and Environmental Staff Company (Ex. 15: 80), suggested that:

[a]ggravation of a pre-existing condition should not be recordable if normal body movements or events cause the aggravation. For example, a smoker with asthma or other obstructive airway disease may experience shortness of breath while climbing a flight of stairs. A person with degenerative disk disease may experience pain while lifting a normal bag of groceries. If performing similar activities at work likewise aggravates the condition, it should not be recordable.

As discussed above, OSHA agrees that non-work-related injuries and illnesses should not be recorded on the OSHA Log. To ensure that non-work-related cases are not entered on the Log, paragraph 1904.5(b)(2)(ii) requires employers to consider as non-work-related any injury or illness that "involves signs or symptoms that surface at work but result solely from a non-work-related event or exposure that occurs outside the work environment."

The Agency also believes that preexisting injury or illness cases that have been aggravated by events or exposures in the work environment represent cases that should be recorded on the Log, because work has clearly worsened the injury or illness. OSHA is concerned, however, that there are some cases where work-related aggravation affects the preexisting case only in a minor way, i.e., in a way that does not appreciably worsen the preexisting condition, alter its nature, change the extent of the medical treatment, trigger lost time, or require job transfer. Accordingly, the final rule requires that workplace events or exposures must "significantly" aggravate a pre-existing injury or illness case before the case is presumed to be work-related. Paragraph 1904.5(a) states that an injury or illness is considered work-related if "an event or exposure in the work environment either caused or contributed to the resulting condition or significantly aggravated a pre-existing injury or illness."

Paragraph 1904.5(b)(4) of the final rule defines aggravation as significant if the contribution of the aggravation at work is such that it results in tangible consequences that go beyond those that the worker would have experienced as a result of the preexisting injury or illness alone, absent the aggravating effects of the workplace. Under the final rule, a preexisting injury or illness will be considered to have been significantly aggravated, for the purposes of OSHA injury and illness recordkeeping, when an event or exposure in the work environment results in: (i) Death, providing that the preexisting injury or illness would likely not have resulted in death but for the occupational event or exposure; (ii) Loss of consciousness, providing that the preexisting injury or illness would likely not have resulted in loss of consciousness but for the occupational event or exposure; (iii) A day or days away from work or of restricted work, or a job transfer that otherwise would not have occurred but for the occupational event or exposure; or (iv) Medical treatment where no medical treatment was needed for the injury or illness before the workplace event or exposure, or a change in the course of medical treatment that was being provided before the workplace event or exposure. OSHA's decision not to require the recording of cases involving only minor aggravation of preexisting conditions is consistent with the Agency's efforts in this rulemaking to require the recording only of non-minor injuries and illnesses; for example, the final rule also no longer requires employers to record minor illnesses on the Log.

Paragraph 1904.5(b)(5) stipulates that pre-existing conditions, for recordkeeping purposes, are conditions that resulted solely from a non-work-related event or exposure that occurs outside the employer's work environment. Pre-existing conditions also include any injury or illness that the employee experienced while working for another employer.

Employees may be injured or become ill as a result of events or exposures away from the employer's establishment. In these cases, OSHA proposed to consider the case work-related only if the employee was engaged in a work activity or was present as a condition of employment (61 FR 4063). In the final rule, (paragraph 1904.5(b)(1)) the same concept is carried forward in the definition of the work environment, which defines the environment as including the establishment and any other location where one or more employees are working or are present as a condition of their employment.

Thus, when employees are working or conducting other tasks in the interest of their employer but at a location away from the employer's establishment, the work-relatedness of an injury or illness that arises is subject to the same decision making process that would occur if the case had occurred at the establishment itself. The case is work-related if one or more events or exposures in the work environment either caused or contributed to the resulting condition or significantly aggravated a pre-existing condition, as stated in paragraph 1904.5(a). In addition, the exceptions for determining work relationship at paragraph 1904.5(b)(2) and the requirements at paragraph 1904.5(b)(3) apply equally to cases that occur at or away from the establishment.

As an example, the work-environment presumption clearly applies to the case of a delivery driver who experiences an injury to his or her back while loading boxes and transporting them into a building. The worker is engaged in a work activity and the injury resulted from an event -- loading/unloading -- occurring in the work environment. Similarly, if an employee is injured in an automobile accident while running errands for the company or traveling to make a speech on behalf of the company, the employee is present at the scene as a condition of employment, and any resulting injury would be work-related.

The final rule continues (at § 1904.5(b)(6)) OSHA's longstanding practice of treating injuries and illnesses that occur to an employee on travel status as work-related if, at the time of the injury or illness, the employee was engaged in work activities "in the interest of the employer." Examples of such activities include travel to and from customer contacts, conducting job tasks, and entertaining or being entertained if the activity is conducted at the direction of the employer.

The final rule contains three exceptions for travel-status situations. The rule describes situations in which injuries or illnesses sustained by traveling employees are not considered work-related for OSHA recordkeeping purposes and therefore do not have to be recorded on the OSHA 300 Log. First, when a traveling employee checks into a hotel, motel, or other temporary residence, he or she is considered to have established a "home away from home." At this time, the status of the employee is the same as that of an employee working at an establishment who leaves work and is essentially "at home". Injuries and illnesses that occur at home are generally not considered work related. However, just as an employer may sometimes be required to record an injury or illness occurring to an employee working in his or her home, the employer is required to record an injury or illness occurring to an employee who is working in his or her hotel room (see the discussion of working at home, below).

Second, if an employee has established a "home away from home" and is reporting to a fixed worksite each day, the employer does not consider injuries or illnesses work-related if they occur while the employee is commuting between the temporary residence and the job location. These cases are parallel to those involving employees commuting to and from work when they are at their home location, and do not have to be recorded, just as injuries and illnesses that occur during normal commuting are not required to be recorded.

Third, the employer is not required to consider an injury or illness to be work-related if it occurs while the employee is on a personal detour from the route of business travel. This exception allows the employer to exclude injuries and illnesses that occur when the worker has taken a side trip for personal reasons while on a business trip, such as a vacation or sight-seeing excursion, to visit relatives, or for some other personal purpose.

The final rule's travel-related provisions (at paragraph 1904.5(b)(6)) are essentially identical to those proposed (63 FR 4063), with only minor editorial changes, and are also parallel to those for determining the work-relationship of traveling employees under the former recordkeeping system (Ex. 2, pp. 36, 37). OSHA received various comments and suggestions about how best to determine work relationship for traveling employees. A few commenters endorsed OSHA's proposed approach (see, e.g., Exs. 15: 199, 396, 406). Other commenters believe, however, that employer control of, or the authority to control, the work environment should be determinative because activities outside the employer's control fall outside the scope of the employer's safety and health program (see, e.g., Exs. 15: 335, 396, 409, 424). The comments of the Dow Chemical Company (Ex. 15: 335) are typical of these views:

[t]ravel on public carriers such as commercial airlines, trains, and taxi services or pre-existing conditions that are aggravated during normal unencumbered body motions, or injuries that occur off-the-job but do not impair someone until they arrive at work are all beyond the control of the employer and the scope of any safety and health program. The commercial plane that crashes while the employee was flying on company business or the taxi accident while the employee was trying to get to the airport to fly on company business are events which, while tragic, are beyond the scope of an employer's control and beyond the reasonable reach of that employer's safety and health program.

However, as discussed in the Legal Authority section and the introduction to the work-relationship section of the preamble, OSHA has decided not to limit the recording of occupational injuries and illnesses to those cases that are preventable, fall within the employer's control, or are covered by the employer's safety and health program. The issue is not whether the conditions could have, or should have, been prevented or whether they were controllable, but simply whether they are occupational, i.e., are related to work. This is true regardless of whether the employee is injured while on travel or while present at the employer's workplace. An employee who is injured in an automobile accident or killed in an airline crash while traveling for the company has clearly experienced a work-related injury that is rightfully included in the OSHA injury and illness records and the Nation's occupational injury and illness statistics. As the American Industrial Hygiene Association (Ex. 15: 153) remarked:

The workforce is increasingly made up of service sector jobs. Computers, materials movement, travel, violence are all emerging and increasing sources of occupational injury and illness. Many of these newer trends in cases may not involve lost workdays, but are recordable and significant to the workforce none the less. Many of the clean, non-manufacturing employers who were traditionally exempt from recordkeeping have risk in these and other emerging areas about which OSHA should be collecting data.

Two commenters specifically objected to the inclusion of cases involving client entertainment (Ex. 15: 409, 424). The American Association of Automobile Manufacturers (AAMA) remarked:

AAMA agrees with OSHA that injuries/illnesses to employees during travel status are work-related and recordable. However, AAMA takes strong exception to the inclusion of 'entertaining or being entertained for the purpose of transacting, discussing, or promoting business.' We find the notion of recording an illness for an employee, while he/she was engaged in a business related dinner, and subsequently suffering acute onset of diarrhea leading to hospitalization for gastroenteritis, to be inappropriate. OSHA needs to remove this obligation from the final rule. (Ex. 15: 409)

OSHA does not agree with this comment, because the Agency believes that employees who are engaged in management, sales, customer service and similar jobs must often entertain clients, and that doing so is a business activity that requires the employee to work at the direction of the employer while conducting such tasks. If the employee is injured or becomes ill while engaged in such work, the injury or illness is work-related and should be recorded if it meets one or more of the other criteria (death, medical treatment, etc.). The gastroenteritis example provided by the AAMA is one type of injury or illness that may occur in this situation, but employees are also injured in accidents while transporting clients to business-related events at the direction of the employer or by other events or exposures arising in the work environment.

On the other hand, not all injuries and illnesses sustained in the course of business-related entertainment are reportable. To be recordable, the entertainment activity must be one that the employee engages in at the direction of the employer. Business-related entertainment activities that are undertaken voluntarily by an employee in the exercise of his or her discretion are not covered by the rule. For example, if an employee attending a professional conference at the direction of the employer goes out for an evening of entertainment with friends, some of whom happen to be clients or customers, any injury or illness resulting from the entertainment activities would not be recordable. In this case, the employee was socializing after work, not entertaining at the direction of the employer. Similarly, the fact that an employee joins a private club or organization, perhaps to "network" or make business contacts, does not make any injury that occurs there work-related.

Two commenters recommended that OSHA eliminate the exceptions for determining work-relationship while employees are on travel and simply require all injuries and illnesses occurring while an employee is on travel status to be considered work-related (Exs. 15: 350, 418). For example, the AFL-CIO (Ex. 15: 418) suggested:

We would also strongly encourage the Agency to re-evaluate [proposed] Appendix A Section C "Travel Status". The AFL-CIO believes that employees in "travel status" (e.g., traveling on company business) should be considered engaged in work-related activities during ALL of their time spent on the trip. This includes all travel, job tasks, entertaining and other activities occurring during "travel status."

OSHA believes that expanding the concept of work-related travel to include all of the time the worker spends on a trip would be inconsistent with the tests of work-relationship governing the recording of other injuries and illnesses and would therefore skew the statistics and confuse employers. As the Dow Chemical Company (Ex. 15: 335) stated:

While the employee is traveling for the benefit of the company, it cannot be said that 100% of their time is engaged in work-related activities. Employees engage in personal and social activities while traveling on company business that is not for the direct benefit of the company nor a condition of employment and which cannot be impacted by an employer's safety or health program. Often there is "free time" while traveling and employees engage in a myriad of activities such as shopping, sightseeing, dining out with friends or family that may be in the area, and the like. These are activities that do not benefit the company and are outside the company's control or reasonable reach of its safety and health programs. These are activities which, if the employee were engaged in them at their normal work location, would not be recordable; but just by the fact that they happen to be traveling for business purposes raises these otherwise non-recordable cases into those subject to the recordkeeping rule.

OSHA agrees with Dow that there are situations where an injury or illness case involving an employee who is on travel status should be excluded from the records. There is no value in recording injuries and illnesses that would not be recorded under non-travel circumstances. For example, there is no value to including in the statistics an injury sustained by an employee who slips and falls in a motel room shower or who is injured in an automobile accident while on personal business, or becomes the victim of random street violence while doing personal shopping on a business trip. OSHA is therefore continuing the Agency's practice of excluding certain cases while employees are in travel status and applying the exceptions to the geographic presumption in the final rule to those occurring while the worker is traveling.

The Department of Energy (Ex. 15: 163) expressed a concern about overseas travel, remarking "For employees who travel in the U.S., the standard makes sense. For employees who travel out of the country, additional burdens to them are generally incurred. Travelers to tropical locations or other areas with different fauna and microbes may incur diseases that are not indigenous to the U.S." In response, OSHA notes that the recordkeeping regulation does not apply to travel outside the United States because the OSH Act applies only to the confines of the United States (29 U.S.C. § 652(4)) and not to foreign operations. Therefore, the OSHA recordkeeping regulation does not apply to non-U.S. operations, and injuries or illnesses that may occur to a worker traveling outside the United States need not be recorded on the OSHA 300 Log.

The final rule also includes provisions at § 1904.5(b)(7) for determining the work-relatedness of injuries and illnesses that may arise when employees are working at home. When an employee is working on company business in his or her home and reports an injury or illness to his or her employer, and the employee's work activities caused or contributed to the injury or illness, or significantly aggravated a pre-existing injury, the case is considered work-related and must be further evaluated to determine whether it meets the recording criteria. If the injury or illness is related to non-work activities or to the general home environment, the case is not considered work-related.

The final rule includes examples to illustrate how employers are required to record injuries and illnesses occurring at home. If an employee drops a box of work documents and injures his or her foot, the case would be considered work-related. If an employee's fingernail was punctured and became infected by a needle from a sewing machine used to perform garment work at home, the injury would be considered work-related . If an employee was injured because he or she tripped on the family dog while rushing to answer a work phone call, the case would not be considered work-related. If an employee working at home is electrocuted because of faulty home wiring, the injury would not be considered work-related.

This provision is consistent with longstanding Agency practice under the former recordkeeping system. It was also included in the proposed rule (63 FR 4063), which read "An injury or illness will be considered work-related if it occurs while the employee is performing work for pay or compensation in the home, if the injury or illness is directly related to the performance of work rather than the general home environment or setting."

A number of commenters supported OSHA's proposed approach to recording the injuries and illnesses of employees who work at home (see, e.g., Exs. 15: 31, 146, 176, 231, 273, 301, 336, 348, 375, 406, 409, 413, 427, 429). The comments of the Council of Community Blood Centers (CCBC) (Ex. 15: 336) are typical of the views of these participants:

CCBC believes this is a good rule and should stay on the books. Accident or illness should be work-related if it occurs at home and is related to performance of the work, not the general home environment or setting. Workers often are off the premises in a variety of situations, such as travel, providing repair services, or consultation. Just as injuries in these situations are reportable, so should those during work at home, if authorized by the employer.

A large number of commenters objected to the proposed approach, however (see, e.g., Exs. 65, 66, 78, 89, 105, 111, 123, 194, 200, 225, 239, 260, 262, 265, 277, 288, 330, 335, 341, 345, 360, 387, 393, 401, 406, 409, 430, 434, 440). Most of these commenters objected because of the employer's perceived inability to control working conditions in the home environment (see, e.g., Exs. 15: 89, 163, 194, 239, 262, 288, 330, 345, 360). For example, the Fort Howard Corporation commented:

Fort Howard strongly opposes OSHA's proposal to consider any injuries and illnesses as "work-related" if it occurs while the employee is performing work for pay or compensation in the home if the injury or illness is directly related to the performance of the work. Employers have absolutely no control over employees' homes. They cannot oversee employees who are doing the work nor can they effectively monitor the manner the work is conducted or the environment in which it is conducted. OSHA's proposal could place employers in the role of insuring the home as a safe work environment. (Ex. 15: 194)

Again, as discussed above, OSHA is concerned that all non-minor work-related cases be recorded on the Log and become part of the national statistics, both because these injuries and illnesses provide information about the safety and health of the work environment to employers, employees, and safety and health professionals and because collecting them may allow previously obscured safety and health issues to be identified. Injuries and illnesses occurring while the employee is working for pay or compensation at home should be treated like injuries and illnesses sustained by employees while traveling on business. The relevant question is whether or not the injury or illness is work-related, not whether there is some element of employer control. The mere recording of these injuries and illnesses as work-related cases does not place the employer in the role of insuring the safety of the home environment.

The law firm of Leonard, Ralston, Stanton & Remington, Chartered (Ex. 15: 430) raised questions about OSHA's role when employees perform office work activities in a home office:

The increasing incidence of home work (or "telecommuting") raises some interesting issues. For example, does OSHA assume that its right of inspection extends to an employee's private home? If so, has the Agency examined the constitutionality of this position? What control does the Agency assume an employer has over working conditions in a private home? Does the Agency expect the employer to inspect its employees' homes to identify unsafe conditions? Must the employer require an employee to correct unsafe conditions in the home (e.g., frayed carpet which presents a tripping hazard; overloaded electrical wiring or use of extension cords; etc.) as a condition of employment? If so, who must pay the cost of necessary home improvements?

OSHA has recently issued a compliance directive (CPL 2-0.125) containing the Agency's response to many of the questions raised by this commenter. That document clarifies that OSHA will not conduct inspections of home offices and does not hold employers liable for employees' home offices. The compliance directive also notes that employers required by the recordkeeping rule to keep records "will continue to be responsible for keeping such records, regardless of whether the injuries occur in the factory, in a home office, or elsewhere, as long as they are work-related, and meet the recordability criteria of 29 CFR Part 1904."

With more employees working at home under various telecommuting and flexible workplace arrangements, OSHA believes that it is important to record injuries and illnesses attributable to work tasks performed at home. If these cases are not recorded, the Nation's injury and illness statistics could be skewed. For example, placing such an exclusion in the final rule would make it difficult to determine if a decline in the overall number or rate of occupational injuries and illnesses is attributable to a trend toward working at home or to a change in the Nation's actual injury and illness experience. Further, excluding these work-related injuries and illnesses from the recordkeeping system could potentially obscure previously unidentified causal connections between events or exposures in the work environment and these incidents. OSHA is unwilling to adopt an exception that would have these potential effects. As the BF Goodrich Company (Ex. 15: 146) said, "[s]pecific criteria to address employee work-at-home situations is appropriate to assure consistent reporting in our changing work environment."

Employers may occasionally have difficulty in determining whether new signs or symptoms are due to a new event or exposure in the workplace or whether they are the continuation of an existing work-related injury or illness. Most occupational injury and illness cases are fairly discrete events, i.e., events in which an injury or acute illness occurs, is treated, and then resolves completely. For example, a worker may suffer a cut, bruise, or rash from a clearly recognized event in the workplace, receive treatment, and recover fully within a few weeks. At some future time, the worker may suffer another cut, bruise or rash from another workplace event. In such cases, it is clear that the two injuries or illnesses are unrelated events, and that each represents an injury or illness that must be separately evaluated for its recordability.

However, it is sometimes difficult to determine whether signs or symptoms are due to a new event or exposure, or are a continuance of an injury or illness that has already been recorded. This is an important distinction, because a new injury or illness requires the employer to make a new entry on the OSHA 300 Log, while a continuation of an old recorded case requires, at most, an updating of the original entry. Section 1904.6 of the final rule being published today explains what employers must do to determine whether or not an injury or illness is a new case for recordkeeping purposes.

The basic requirement at § 1904.6(a) states that the employer must consider an injury or illness a new case to be evaluated for recordability if (1) the employee has not previously experienced a recorded injury or illness of the same type that affects the same part of the body, or (2) the employee previously experienced a recorded injury or illness of the same type that affected the same part of the body but had recovered completely (all signs and symptoms of the previous injury or illness had disappeared) and an event or exposure in the work environment caused the injury or illness, or its signs or symptoms, to reappear.

The implementation question at § 1904.6(b)(1) addresses chronic work-related cases that have already been recorded once and distinguishes between those conditions that will progress even in the absence of workplace exposure and those that are triggered by events in the workplace. There are some conditions that will progress even in the absence of further exposure, such as some occupational cancers, advanced asbestosis, tuberculosis disease, advanced byssinosis, advanced silicosis, etc. These conditions are chronic; once the disease is contracted it may never be cured or completely resolved, and therefore the case is never "closed" under the OSHA recordkeeping system, even though the signs and symptoms of the condition may alternate between remission and active disease.

However, there are other chronic work-related illness conditions, such as occupational asthma, reactive airways dysfunction syndrome (RADs), and sensitization (contact) dermatitis, that recur if the ill individual is exposed to the agent (or agents, in the case of cross-reactivities or RADs) that triggers the illness again. It is typical, but not always the case, for individuals with these conditions to be symptom-free if exposure to the sensitizing or precipitating agent does not occur.

The final rule provides, at paragraph (b)(1), that the employer is not required to record as a new case a previously recorded case of chronic work-related illness where the signs or symptoms have recurred or continued in the absence of exposure in the workplace. This paragraph recognizes that there are occupational illnesses that may be diagnosed at some stage of the disease and may then progress without regard to workplace events or exposures. Such diseases, in other words, will progress without further workplace exposure to the toxic substance(s) that caused the disease. Examples of such chronic work-related diseases are silicosis, tuberculosis, and asbestosis. With these conditions, the ill worker will show signs (such as a positive TB skin test, a positive chest roentgenogram, etc.) at every medical examination, and may experience symptomatic bouts as the disease progresses.

Paragraph 1904.6(b)(2) recognizes that many chronic occupational illnesses, however, such as occupational asthma, RADs, and contact dermatitis, are triggered by exposures in the workplace. The difference between these conditions and those addressed in paragraph 1904.6(b)(1) is that in these cases exposure triggers the recurrence of symptoms and signs, while in the chronic cases covered in the previous paragraph, the symptoms and signs recur even in the absence of exposure in the workplace. This distinction is consistent with the position taken by OSHA interpretations issued under the former recordkeeping rule (see the Guidelines discussion below). The Agency has included provisions related to new cases/continuations of old cases in the final rule to clarify its position and ensure consistent reporting.

Paragraph 1904.6(b)(3) addresses how to record a case for which the employer requests a physician or other licensed health care professional (HCP) to make a new case/continuation of an old case determination. Paragraph (b)(3) makes clear that employers are to follow the guidance provided by the HCP for OSHA recordkeeping purposes. In cases where two or more HCPs make conflicting or differing recommendations, the employer is required to base his or her decision about recordation based on the most authoritative (best documented, best reasoned, or most persuasive) evidence or recommendation.

The final rule's provisions on the recording of new cases are nearly identical to interpretations of new case recordability under the former rule. OSHA has historically recognized that it is generally an easier matter to differentiate between old and new cases that involve injuries than those involving illnesses: the Guidelines stated that "the aggravation of a previous injury almost always results from some new incident involving the employee * * * [w]hen work-related, these new incidents should be recorded as new cases on the OSHA forms, assuming they meet the criteria for recordability * * *" (Ex. 2, p. 31). However, the Guidelines also stated that "certain illnesses, such as silicosis, may have prolonged effects which recur over time. The recurrence of these symptoms should not be recorded as a new case on the OSHA forms. * * * Some occupational illnesses, such as certain dermatitis or respiratory conditions, may recur as the result of new exposures to sensitizing agents, and should be recorded as new cases."

OSHA developed and included specific guidance for evaluating when cumulative trauma disorders (CTDs) (ergonomic injuries and illnesses, now known as musculoskeletal disorders, or MSDs) should be recorded as new cases in the Ergonomics Program Management Guidelines For Meatpacking Plants (Ex. 11, p. 15) which were published in 1990. These Guidelines provided:

If and when an employee who has experienced a recordable CTD becomes symptom free (including both subjective symptoms and physical findings), any recurrence of symptoms establishes a new case. Furthermore, if the worker fails to return for medical care within 30 days, the case is presumed to be resolved. Any visit to a health care provider for similar complaints after the 30-day interval "implies reinjury or reexposure to a workplace hazard and would represent a new case."

Thus, the former rule had different "new case" criteria for musculoskeletal disorders than for other injuries and illnesses. (For the final rule's recording criteria for musculoskeletal disorders, see Section 1904.12.)

OSHA's recordkeeping NPRM proposed a single approach to the identification of new cases for all injuries and illnesses, including musculoskeletal disorders. The proposal would have required the recurrence of a pre-existing injury or illness to be considered a new case to evaluate for recordability if (1) it resulted from a new work event or exposure, or (2) 45 days had elapsed since medical treatment, work restriction, or days away from work had ceased, and the last sign or symptom had been experienced. The proposed approach would, in effect, have extended the recurrence criteria for musculoskeletal disorders to all injury and illness cases, but would have increased the no-medical-intervention interval from 30 to 45 days. A recurrence of a previous work-related injury or illness would have been presumed, under the proposed approach, to be a new case if (1) it resulted from a new work accident or exposure, or (2) 45 days had elapsed since medical treatment had been administered or restricted work activity or days away had occurred and since the last sign or symptom had been experienced. This proposed presumption would have been rebuttable if there was medical evidence indicating that the prior case had not been resolved. In the proposal, OSHA also asked for input on the following questions related to new case recording:

OSHA solicits comment on the appropriateness of the 45-day interval. Is 45 days too short or long of a period? If so, should the period be 30 days? 60 days? 90 days? or some other time period? Should different conditions (e.g. back cases, asthma cases etc.) have different time intervals for evaluating new cases?

OSHA is also seeking input for an improved way to evaluate new cases. Should a new category of cases be created to capture information on recurring injuries and illnesses? One option is to add an additional "check box" column to the proposed OSHA Form 300 for identifying those cases that are recurrences of previously recorded injuries and illnesses. This would allow employers, employees and OSHA inspectors to differentiate between one time cases and those that are recurrent, chronic conditions. This approach may help to remove some of the stigma of recording these types of disorders and lead to more complete records. OSHA solicits input on this approach. Will a recurrence column reduce the stigma of recording these types of cases? Should recurrences be included in the annual summaries? Should a time limit be used to limit the use of a recurrence column?

In response to the views and evidence presented by commenters to the record, OSHA has decided not to adopt the proposed approach to the recording of new/recurring cases in the final rule. Commenters expressed a wide variety of views about the recording of recurring injury and illness cases. Some commenters favored the proposed approach as drafted. Others, however, objected to it on many grounds: (1) the time limit should be longer or shorter than the 45 days proposed; (2) the proposed approach would result in under- or overreporting; (3) it would conflict with workers' compensation requirements; (4) it was too restrictive (5) it would encourage excessive use of the health care system; and (6) it should be replaced by a physician or other licensed health care professional's opinion.

A number of commenters supported OSHA's proposed approach (see, e.g., Exs. 15: 27, 65, 70, 151, 152, 154, 179, 180, 181, 185, 186, 188, 214, 331, 332, 336, 359, 387, 396, 424, 428). Representative of these comments was one from The Fertilizer Institute (TFI):

TFI agrees with OSHA's proposed 45 day criterion for the recording of new cases. Concerning OSHA's solicitation of comments on whether different conditions should have different evaluation periods, TFI encourages OSHA to adopt a single time period for all conditions. Different evaluation periods for different conditions will lead to complexity and confusion without any resulting benefit to recordkeeping (Ex. 15: 154).

Other commenters supported the concept of using a time limit for determining new cases, but thought the number of days should be higher (see, e.g., Exs. 15: 45, 49, 61, 82, 89, 131, 147, 184, 235, 331, 389). Some commenters generally opposed the time limit concept but made recommendations for longer time periods if OSHA decided in the final rule to adopt a time limit (see, e.g., Exs. 15: 38, 79, 89, 111, 136, 137, 141, 194, 224, 246, 266, 278, 288, 299, 313, 335, 352, 353, 430). The longer intervals suggested by commenters included 60 days (see, e.g., Exs. 15: 82, 389); 90 days (see, e.g., Exs. 15: 38, 49, 79, 147, 184, 246, 299, 313, 331, 335, 352, 353, 430); 120 days (Ex. 15: 194); 180 days (see, e.g., Exs. 15: 61, 111, 136, 137, 141, 224, 266, 278, 288); one year (Ex. 15: 131); and five years (Ex. 15: 89).

A large number of commenters opposed the proposed approach for identifying new cases that would then be tested for their recordability (see, e.g., Exs. 15: 33, 38, 39, 41, 78, 79, 89, 95, 102, 107, 111, 119, 127, 133, 136, 137, 141, 153, 171, 176, 194, 199, 203, 224, 225, 231, 246, 266, 273, 278, 281, 288, 289, 299, 301, 305, 307, 308, 313, 335, 337, 341, 346, 348, 352, 353, 375, 395, 405, 410, 413, 424, 425, 428, 430, 440). Some commenters argued that the proposed 45-day interval was arbitrary (see, e.g., Exs. 15: 119, 203, 289, 313, 352, 353, 395), that it conflicted with workers' compensation new case determinations (see, e.g., Exs. 15: 38, 119, 136, 137, 141, 224, 266, 278), that the approach would not work in the case of chronic injury (see, e.g., Exs. 33; 15: 176, 199, 231, 273, 299, 301, 305, 308, 337, 346, 348, 375), or that the proposed 45-day rule would result in over-reporting of occupational injuries and illnesses (see, e.g., Exs. 15: 119, 127, 136, 137, 141, 171, 199, 224, 266, 278, 305, 337, 424, 425). The comments of the NYNEX Corporation (Ex. 15: 199) illustrate the general concerns of these commenters:

We do not agree, however, with the second criterion of a symptom free 45 day period following medical treatment, restriction, or days away from work. This criterion fails to take into account the persistent nature of many chronic or recurring conditions, i.e., back strains, musculoskeletal disorders, where the symptoms may disappear for a period of time, but the underlying conditions are still present. If adopted, this criterion could cause injury and illness data to be artificially inflated with the onset of "new" cases, which in fact are recurrences of existing conditions. This in turn could lead to false epidemics and a diversion of resources from more legitimate workplace concerns.

On the other hand, William K. Principe of Constangy, Brooks & Smith, LLC (Ex. 15: 428) was concerned that the proposed method would result in fewer recordable cases:

Since many employees will report that they continued to experience symptoms or that they continue to have good days and bad days, the new rule will result in many fewer recordable CTD [cumulative trauma disorder] cases. In fact, at some hand-intensive manual operations, the number of CTD cases should be drastically reduced under the proposal that 45 days must elapse since the last symptom. There is something fundamentally wrong with a recordkeeping system that one year shows a high incidence of CTDs and the next shows a dramatic decline, when the underlying conditions remain virtually identical.

United Parcel Service (Ex. 15: 424) stated that there should be no time limit to determining whether or not a case is a recurrence:

In UPS's experience, however, it is a simple process to determine, by medical referral or by examining prior medical history, whether a condition is a recurrence. This has long been the practice, and indeed the [proposal] contemplates it will remain the practice through the first 44 days. It does not become any more complex on the 45th, 50th, or 100th day; and if in an individual employer's judgment it does, then the employer may of course report the condition as a new injury.

Three commenters disapproved of OSHA's approach because it would have been applicable to all recurrences and they believe that each case must be evaluated on its own merits (Exs. 15: 78, 184, 203). The International Dairy Foods Association (IDFA) described this concern succinctly: "Each injury has its own resolution based on the injury, illness, degree, and numerous other factors that are characteristic of the individual. As such, it is impossible for OSHA or anyone else to set a valid number of days even if the resolution period is set on the basis of the type of illness/injury" (Ex. 15: 203).

In addition, the proposed 45-day approach was interpreted differently by different commenters. For example, David E. Jones of the law firm Ogletree, Deakins, Nash, Smoak & Stewart (ODNSS) suggested:

The words "either" and "or" * * * should be deleted because an aggravation of the previously recorded injury or illness brought about within the 45-day period would require the entry of a new case at that time, thus negating the 45-day rule, leading to the adverse result that the 45-day rule otherwise would rectify. Accordingly, ODNSS recommends * * * "A recurrence of a previous work-related injury or illness is a new case when it (1) results from a new work event or exposure and (2) 45 days have elapsed since medical treatment, restricted work activity, or days away from work (as applicable) were discontinued and the employee has been symptom-free (including both subjective symptoms and physical findings) (emphasis added) (Ex. 15: 406).

In the final rule, OSHA has decided against the proposed approach of determining case resolution based on a certain number of days during which the injured or ill employee did not lose time, receive treatment, have signs or symptoms, or be restricted to light duty. OSHA agrees with those commenters who argued that the proposed approach was too prescriptive and did not allow for the variations that naturally exist from one injury and illness case to the next. Further, the record contains no convincing evidence to support a set number of days as appropriate. OSHA thus agrees with those commenters who pointed out that adoption of a fixed time interval would result in the overrecording of some injury and illness cases and the underrecording of others, and thus would impair the quality of the records.

Further, OSHA did not intend to create an "injury free" time zone during which an injury or illness would not be considered a new case, regardless of cause, as ODNSS suggested. Instead, OSHA proposed that a case be considered a new case if either condition applied: the case resulted from a new event or exposure or 45 days had elapsed without signs, symptoms, or medical treatment, restricted work, or days away from work. There are clearly cases where an event or exposure in the workplace would be cause for recording a new case. A new injury may manifest the same signs and symptoms as the previous injury but still be a new injury and not a continuation of the old case if, for example, an employee sustains a fall and fractures his or her wrist, and four months later falls again and fractures the wrist in the same place. This occurrence is not a continuation of the fracture but rather a new injury whose recordability must be evaluated. The final rule's approach to recurrence/new case determinations avoids this and other recording problems because it includes no day count limit and relies on one of the basic principles of the recordkeeping system, i.e., that injuries or illnesses arising from events or exposures in the workplace must be evaluated for recordability.

In response to those commenters who raised issues about inconsistency between the OSHA system and workers' compensation, OSHA notes that there is no reason for the two systems, which serve different purposes (recording injuries and illnesses for national statistical purposes and indemnifying workers for job-related injuries and illnesses) to use the same definitions. Accordingly, the final rule does not rely on workers' compensation determinations to identify injuries or illness cases that are to be considered new cases for recordkeeping purposes.

Another group of commenters argued that the 45-day recording requirement would lead employers to spend money on unnecessary and costly health care (see, e.g., Exs. 15: 136, 137, 141, 224, 266, 278, 305, 346, 348, 375). The views of the American Petroleum Institute (API) are representative: "OSHA's proposal would also add substantially to employers' costs since it could require employees to make frequent trips to a health care professional, even if symptom free, just to avoid being recorded repeatedly on the OSHA log as new cases" (Ex. 15: 375). Union Carbide Corporation (Ex. 15: 396) also remarked on the proposed approach's potential incentive for medical follow-up, but viewed such an incentive as a positive phenomenon, stating "One benefit [of the proposed approach] is that it encourages medical follow-up for the employee." Although the proposed approach would not have "required" an employer to send a worker to a physician or other licensed health care professional, and OSHA is not persuaded that employers would choose to spend money in this way merely to avoid recording an occasional case as a new case, elimination of any set day-count interval from the final rule will also have made the concerns of these commenters moot.

OSHA also received a number of suggestions about the role of physicians and other licensed health care professionals (HCP) in new case determinations. A number of commenters recommended that the decision to record should be based solely on the opinions of a physician or other licensed health care professional (see, e.g., Exs. 33: 15: 39, 95, 107, 119, 127, 133, 225, 289, 332, 335, 341, 387, 424, 440). The National Grain and Feed Association, the National Oilseed Processors Association, and the Grain Elevator and Processing Society (Ex. 15: 119) commented as a group and recommended that "[r]elying on a physician's opinion rather than an arbitrary timeframe would simplify recordkeeping and help ensure that the records are consistent with existing and accepted workers' compensation plans."

Other commenters recommended that, if OSHA adopted a day count time limit, the rule should specifically allow a physician's opinion to be used to refute a new case determination (see, e.g., Exs. 15: 65, 181, 184, 203). Several others simply asked OSHA to provide more guidance on what type of medical evidence could be used in new case determinations (see, e.g., Exs. 15: 176, 231, 273, 301, 430). The National Wholesale Druggists' Association (NWDA) suggested that "OSHA should also include a provision that the employee obtain written approval from a doctor that the employee's condition has been resolved before going back to work. Determining the end of treatment should be left in the hands of a medical professional and OSHA should require some type of documentation to that effect" (Ex. 15: 185).

OSHA has not included any provisions in the final rule that require an employer to rely on a physician or other licensed health care professional or that tell a physician or other licensed health care professional how to treat an injured or ill worker, or when to begin or end such treatment. In the final rule OSHA does require the employer to follow any determination a physician or other licensed health care professional has made about the status of a new case. That is, if such a professional has determined that a case is a new case, the employer must record it as such. If the professional determines that the case is a recurrence, rather than a new case, the employer is not to record it a second time. In addition, the rule does not require the employee, or the employer, to obtain permission from the physician or other licensed health care professional before the employee can return to work. OSHA believes that the employer is capable of, and often in the best position to, make return-to-work decisions.

Southern California Edison (Ex. 15: 111) expressed concern that imposing a day limit would not take differences between types of injuries and illnesses into account, stating "A recurrence of a previous work-related injury or illness should only be considered a new case when the injury or illness has completely healed. Severe muscle and nerve damage can take many weeks or months to properly heal." The final rule takes such differences into account, as follows. If the previous injury or illness has not healed (signs and symptoms have not resolved), then the case cannot be considered resolved. The employer may make this determination or may rely on the recommendation of a physician or other licensed health care professional when doing so. Clearly, if the injured or ill employee is still exhibiting signs or symptoms of the previous injury or illness, the malady has not healed, and a new case does not have to be recorded. Similarly, if work activities aggravate a previously recorded case, there is no need to consider recording it again (although there may be a need to update the case information if the aggravation causes a more severe outcome than the original case, such as days away from work).

The Quaker Oats Company (Ex. 15: 289) suggested that employers should be permitted by the rule to decide whether a given case was a new case or not, without requirements in the rule:

The 45 day interval on determining if a case is a new one or should be counted under a previous injury should be left to the discretion of the employer. They have the most intimate knowledge of the work environment, medical treatment of the affected employee and the status of their work-related injury or illness. I will agree that it is a difficult matter to decide and to assure consistency throughout industry * * * I believe that any number of days would simply be an arbitrary attempt at quantifying something that is best left to the medical judgment of a healthcare professional.

Under the OSHA recordkeeping system, the employer is always the responsible party when it comes to making the determination of the recordability of a given case. However, if OSHA did not establish consistent new case determination criteria, a substantial amount of variability would be introduced into the system, which would undermine the Agency's goals of improving the accuracy and consistency of the Nation's occupational injury and illness data. Accordingly, OSHA has not adopted this suggested approach in the final rule.

A number of commenters argued that the occurrence of a new event, exposure, or incident should be required to trigger the recording of a new case (see, e.g., Exs. 33, 15: 102, 171, 176, 231, 273, 301, 307, 308, 405, 410, 413, 425). Representative of these comments was one from the Voluntary Protection Programs Participants' Association (VPPPA), which recommended that OSHA "adopt a definition for new case that requires the occurrence of a new work-related event to trigger a new case. In the absence of this, the case would be considered recurring" (Ex. 15: 425). OSHA agrees with the VPPPA that if no further event or exposure occurs in the workplace to aggravate a previous injury or illness, a new case need not be recorded. However, if events or exposures at work cause the same symptoms or signs to recur, the final rule requires employers to evaluate the injury or illness to see if it is a new case and is thus recordable.

The OSHA statistical system is designed to measure the incidence, rather than prevalence, of occupational injury and illness. Incidence measures capture the number of new occupational injuries and illnesses occurring in a given year, while prevalence measures capture the number of such cases existing in a given year (prevalence measures thus capture cases without regard to the year in which they onset). Prevalence measures would therefore capture all injuries and illnesses that occurred in a given year as well as those unresolved injuries and illnesses that persist from previous years. The difference is illustrated by the following cases: (1) A worker experiences a cut that requires sutures and heals completely before the year ends; this injury would be captured both by an incidence or prevalence measure for that particular year. (2) Another worker retired last year but continues to receive medical treatment for a work-related respiratory illness that was first recognized two years ago. This case would be captured in the year of onset and each year thereafter until it resolves if a prevalence measure is used, but would be counted only once (in the year of onset) if an incidence measure is used.

Because the OSHA system is intended to measure the incidence of occupational injury and illness, each individual injury or illness should be recorded only once in the system. However, an employee can experience the same type of injury or illness more than once. For example, if a worker cuts a finger on a machine in March, and is then unfortunate enough to cut the same finger again in October, this worker has clearly experienced two separate occupational injuries, each of which must be evaluated for its recordability. In other cases, this evaluation is not as simple. For example, a worker who performs forceful manual handling injures his or her back in 1998, resulting in days away from work, and the case is entered into the records. In 1999 this worker has another episode of severe work-related back pain and must once again take time off for treatment and recuperation. The question is whether or not the new symptoms, back pain, are continuing symptoms of the old injury, or whether they represent a new injury that should be evaluated for its recordability as a new case. The answer in this case lies in an analysis of whether or not the injured or ill worker has recovered fully between episodes, and whether or not the back pain is the result of a second event or exposure in the workplace, e.g., continued manual handling. If the worker has not fully recovered and no new event or exposure has occurred in the workplace, the case is considered a continuation of the previous injury or illness and is not recordable.

One reason for the confusion that is apparent in some of the comments on the proposal's approach to the recording of recurrences may be the custom that developed over the years of referring to recordable recurrences of work-related injuries and illnesses as "new cases." See for example, 61 FR 4037/1 ("employers may be dealing with a re-injury or recurrence of a previous case and must decide whether the recurrence is a "new case" or a continuation of the original case.") The term "new case" tends to suggest to some that the case is totally original, when in fact new cases for OSHA recordkeeping purposes include three categories of cases; (1) totally new cases where the employee has never suffered similar signs or symptoms while in the employ of that employer, (2) cases where the employee has a preexisting condition that is significantly aggravated by activities at work and the significant aggravation reaches the level requiring recordation, and (3) previously recorded conditions that have healed (all symptoms and signs have resolved) and then have subsequently been triggered by events or exposures at work.

Under the former rule and the final rule, both new injuries and recurrences must be evaluated for their work-relatedness and then for whether they meet one or more of the recording criteria; when these criteria are met, the case must be recorded. If the case is a continuation of a previously recorded case but does not meet the "new case" criteria, the employer may have to update the OSHA 300 Log entry if the original case continues to progress, i.e., if the status of the case worsens. For example, consider a case where an employee has injured his or her back lifting a heavy object, the injury resulted in medical treatment, and the case was recorded as a case without restricted work or days away. If the injury does not heal and the employer subsequently decides to assign the worker to restricted work activity, the employer is required by the final rule to change the case classification and to track the number of days of restricted work. If the case is a previous work-related injury that did not meet the recording criteria and thus was not recorded, future developments in the case may require it to be recorded. For example, an employee may suffer an ankle sprain tripping on a step. The employee is sent to a health care professional, who does not recommend medical treatment or restrictions, so the case is not recorded at that time. If the injury does not heal, however, and a subsequent visit to a physician results in medical treatment, the case must then be recorded.

OSHA and employers and employees need data on recurring cases because recurrence is an important indicator of severity over the long term. Just as the number of days away is a useful indicator of health and safety risk at a particular establishment, so is the total number of injury and illness events and of exposures resulting in health consequences that occur in an establishment or industry. Further, any realistic assessment of occupational safety and health conditions should reflect the fact that some but not all injuries and illnesses have long-term consequences. In other words, a safety and health analysis should give less weight to an injury or illness that has a clear and relatively quick recovery without impairment of any kind and an injury or illness that is chronic in nature or one that involves recurring episodes that are retriggered by workplace events or exposures.

Ignoring the fact that an occupational injury or illness is a recurrence occasioned by an event or exposure in the workplace would result in an underestimate of the true extent of occupational injury and illness and deprive employers, employees, and safety and health professionals of essential information of use in illness prevention. The other extreme, requiring employers to record on-going signs or symptoms repeatedly, even in the absence of an event or exposure in the workplace, would result in overstating the extent of illness. In terms of the recordkeeping system, deciding how most appropriately to handle new cases requires a balanced approach that minimizes both overrecording and underrecording. OSHA has dealt with this problem in the final rule by carefully defining the circumstances under which a chronic and previously recorded injury or illness must be considered closed and defining the circumstances under which a recurrence is to be considered a new case and then evaluated to determine whether it meets one or more of the recordability criteria.

OSHA's proposal to apply a single criterion to the determination of the recordability of all recurrences of previously recorded injuries and illnesses received support from several commenters (see, e.g., Exs. 15: 31, 61, 70, 154, 203, 396). The final rule uses one set of criteria for determining whether any injury or illness, including a musculoskeletal disorder, is to be treated as a new case or as the continuation of an "old" injury or illness. First, if the employee has never had a recorded injury or illness of the same type and affecting the same part of the body, the case is automatically considered a new case and must be evaluated for recordability. This provision will handle the vast majority of injury and illness cases, which are new cases rather than recurrences or case continuations. Second, if the employee has previously had a recorded injury or illness of the same type and affecting the same body part, but the employee has completely recovered from the previous injury or illness, and a new workplace event or exposure causes the injury or illness (or its signs or symptoms) to reappear, the case is a recurrence that the employer must evaluate for recordability.

The implementation section of § 1904.6 describes these requirements and includes explanations applying to two special circumstances. In the first case, paragraph 1904.6(b)(1) the employee has experienced a chronic injury or illness of a type that will progress regardless of further workplace exposure. Cases to which this provision applies are serious, chronic illness conditions such as occupational cancer, asbestosis, silicosis, chronic beryllium disease, etc. These occupational conditions generally continue to progress even though the worker is removed from further exposure. These conditions may change over time and be associated with recurrences of symptoms, or remissions, but the signs (e.g., positive chest roentgenogram, positive blood test) generally continue to be present throughout the course of the disease.

The second kind of case, addressed in paragraph 1904.6(b)(b)(2), requires employers to record chronic illness cases that recur as a result of exposures in the workplace. These conditions might include episodes of occupational asthma, reactive airways dysfunction syndrome (RADS), or contact allergic dermatitis, for example.

Paragraph 1904.6(b)(3) recognizes the role of physicians and other licensed health care professionals that the employer may choose to rely on when tracking a "new case" or making a continuation of an old case determination. If a physician or other licensed health care professional determines that an injury or illness has been resolved, the employer must consider the case to be resolved and record as a new case any episode that causes the signs and symptoms to recur as a result of exposure in the workplace. On the other hand, if the HCP consulted by the employer determines that the case is a chronic illness of the type addressed by paragraph 1904.6(b)(1), the employer would not record the case again. In either case, the employer would evaluate it for work-relatedness and then determine whether the original entry requires updating or the case meets the recording criteria. Paragraph (b)(3) also recognizes that the employer may ask for input from more than one HCP, or the employer and employee may each do so, and in such cases, the rule requires the employer to rely on the one judged by the employer to be most authoritative.

In the proposal, OSHA asked commenters whether the Log should include a column with a check-box that could be marked if a case was a recurrence of a pre-existing condition (61 FR 4037). Some commenters supported the proposed approach (see, e.g., Exs. 15: 27, 39, 61, 65, 89, 154, 186, 214, 235, 277, 299, 305, 332, 336). For example, the National Association of Manufacturers (NAM) suggested that, in lieu of adopting a 45-day time limit, OSHA should add a column to the Log: "If the Agency believes there is a need to track the number of recurring cases, we believe the better approach would be to add a column to the log which would permit the original entry for each injury or illness to be updated in the event of a recurrence" (Ex. 15: 305). The American Association of Homes and Services for the Aging (AAHSA) agreed:

[t]here should be a column on the injury and illness log for employers to check for reoccurring injuries. This addition would help the employer to identify possible patterns or problems associated with a specific job and find solutions. Recommendation: Add a column to the injury and illness log allowing the employer to check when an employee is having a repetitive injury or illness (Ex. 15: 214).

Other commenters did not support the proposal's approach to tracking recurrences (see, e.g., Exs. 15: 70, 78, 136, 137, 141, 151, 152, 179, 180, 194, 224, 266, 278). The comments of Kathy Lehrman, RN, Occupational Health Nurse (Ex. 15: 136) are representative of these comments:

The addition of a column to record recurrent conditions would not reduce the stigma and would lead to increased health care provider visits to avoid having an ongoing case labeled as a new case. * * * I do not see the value of including a new category of case designation. This runs counter to the simplification objective.

After a review of the comments on this issue, OSHA has decided not to include such a check-box on the Log. The final rule adds several columns to the OSHA 300 form to collect data on the number of restricted workdays and on various types of occupational injuries and illnesses. The addition of these columns, and the decision to provide more space on the Log to add information on the case, has used up the available space on the form. Requiring employers to record recurrences would also be burdensome and make the rule more complex. Further, OSHA did not propose such a requirement, and this issue raises questions not adequately aired in the record. For example, if an employee has recurring episodes of low back pain, should the employer be required to record each day the employee experiences such pain as a recurring injury? OSHA is also unsure how recurrence data should be captured and used in the Nation's injury and illness statistics. For example, would a separate data set on recurrences, similar to data on injuries and illnesses, be produced by the BLS?

OSHA has therefore decided that it is not appropriate to add a column to the Log to capture data on recurring injuries and illnesses. However, OSHA recognizes that data on injury and illness recurrence may be useful to employers and employees at individual worksites and encourages employers who wish to collect this additional information to do so; however, the final rule does not require employers to provide recurrence data on the Log.

Section 1904.7 contains the general recording criteria for recording work-related injuries and illnesses. This section describes the recording of cases that meet one or more of the following six criteria: death, days away from work, restricted work or transfer to another job, medical treatment beyond first aid, loss of consciousness, or diagnosis as a significant injury or illness by a physician or other licensed health care professional.

Paragraph 1904.7(a) describes the basic requirement for recording an injury or illness in the OSHA recordkeeping system. It states that employers must record any work-related injury or illness that meets one or more of the final rule's general recording criteria. There are six such criteria: death, days away from work, days on restricted work or on job transfer, medical treatment beyond first aid, loss of consciousness, or diagnosis by a physician or other licensed heath care professional as a significant injury or illness. Although most cases are recorded because they meet one of these criteria, some cases may meet more than one criterion as the case continues. For example, an injured worker may initially be sent home to recuperate (making the case recordable as a "days away" case) and then subsequently return to work on a restricted ("light duty") basis (meeting a second criterion, that for restricted work). (see the discussion in Section 1904.29 for information on how to record such cases.)

Paragraph 1904.7(b) tells employers how to record cases meeting each of the six general recording criteria and states how each case is to be entered on the OSHA 300 Log. Paragraph 1904.7(b)(1) provides a simple decision table listing the six general recording criteria and the paragraph number of each in the final rule. It is included to aid employers and recordkeepers in recording these cases.

Paragraph 1904.7(b)(2) requires the employer to record an injury or illness that results in death by entering a check mark on the OSHA 300 Log in the space for fatal cases. This paragraph also directs employers to report work-related fatalities to OSHA within 8 hours and cross references the fatality and catastrophe reporting requirements in § 1904.39 of the final rule, Reporting fatalities and multiple hospitalizations to OSHA.

Paragraph 1904.7(b)(2) implements the OSH Act's requirements to record all cases resulting in work-related deaths. There were no comments opposing the recording of cases resulting in death. However, there were several comments questioning the determination of work-relatedness for certain fatality cases and the appropriateness of reporting certain kinds of fatalities to OSHA. These comments are addressed in the sections of this preamble devoted to work-relationship and fatality reporting (sections 1904.5 and 1904.39, respectively).

Paragraph 1904.7(b)(3) contains the requirements for recording work-related injuries and illnesses that result in days away from work and for counting the total number of days away associated with a given case. Paragraph 1904.7(b)(3) requires the employer to record an injury or illness that involves one or more days away from work by placing a check mark on the OSHA 300 Log in the space reserved for day(s) away cases and entering the number of calendar days away from work in the column reserved for that purpose. This paragraph also states that, if the employee is away from work for an extended time, the employer must update the day count when the actual number of days away becomes known. This requirement continues the day counting requirements of the former rule and revises the days away requirements in response to comments in the record.

Paragraphs 1904.7(b)(3)(i) through (vi) implement the basic requirements. Paragraph 1904.7(b)(3)(i) states that the employer is not to count the day of the injury or illness as a day away, but is to begin counting days away on the following day. Thus, even though an injury or illness may result in some loss of time on the day of the injurious event or exposure because, for example, the employee seeks treatment or is sent home, the case is not considered a days-away-from-work case unless the employee does not work on at least one subsequent day because of the injury or illness. The employer is to begin counting days away on the day following the injury or onset of illness. This policy is a continuation of OSHA's practice under the former rule, which also excluded the day of injury or onset of illness from the day counts.

Paragraphs 1904.7(b)(3)(ii) and (iii) direct employers how to record days-away cases when a physician or other licensed health care professional (HCP) recommends that the injured or ill worker stay at home or that he or she return to work but the employee chooses not to do so. As these paragraphs make clear, OSHA requires employers to follow the physician's or HCP's recommendation when recording the case. Further, whether the employee works or not is in the control of the employer, not the employee. That is, if an HCP recommends that the employee remain away from work for one or more days, the employer is required to record the injury or illness as a case involving days away from work and to keep track of the days; the employee's wishes in this case are not relevant, since it is the employer who controls the conditions of work. Similarly, if the HCP tells the employee that he or she can return to work, the employer is required by the rule to stop counting the days away from work, even if the employee chooses not to return to work. These policies are a continuation of OSHA's previous policy of requiring employees to follow the recommendations of health care professionals when recording cases in the OSHA system. OSHA is aware that there may be situations where the employer obtains an opinion from a physician or other health care professional and a subsequent HCP's opinion differs from the first. (The subsequent opinion could be that of an HCP retained by the employer or the employee.) In this case, the employer is the ultimate recordkeeping decision-maker and must resolve the differences in opinion; he or she may turn to a third HCP for this purpose, or may make the recordability decision himself or herself.

Paragraph 1904.7(b)(3)(iv) specifies how the employer is to account for weekends, holidays, and other days during which the employee was unable to work because of a work-related injury or illness during a period in which the employee was not scheduled to work. The rule requires the employer to count the number of calendar days the employee was unable to work because of the work-related injury or illness, regardless of whether or not the employee would have been scheduled to work on those calendar days. This provision will ensure that a measure of the length of disability is available, regardless of the employee's work schedule. This requirement is a change from the former policy, which focused on scheduled workdays missed due to injury or illness and excluded from the days away count any normal days off, holidays, and other days the employee would not have worked.

Paragraph 1904.7(b)(3)(v) tells the employer how to count days away for a case where the employee is injured or becomes ill on the last day of work before some scheduled time off, such as on the Friday before the weekend or the day before a scheduled vacation, and returns to work on the next day that he or she was scheduled to work. In this situation, the employer must decide if the worker would have been able to work on the days when he or she was not at work. In other words, the employer is not required to count as days away any of the days on which the employee would have been able to work but did not because the facility was closed, the employee was not scheduled to work, or for other reasons unrelated to the injury or illness. However, if the employer determines that the employee's injury or illness would have kept the employee from being able to work for part or all of time the employee was away, those days must be counted toward the days away total.

Paragraph 1904.7(b)(3)(vi) allows the employer to stop counting the days away from work when the injury or illness has resulted in 180 calendar days away from work. When the injury or illness results in an absence of more than 180 days, the employer may enter 180 (or 180+) on the Log. This is a new provision of the final rule; it is included because OSHA believes that the "180" notation indicates a case of exceptional severity and that counting days away beyond that point would provide little if any additional information.

Paragraph 1904.7(b)(3)(vii) specifies that employers whose employees are away from work because of a work-related injury or illness and who then decide to leave the company's employ or to retire must determine whether the employee is leaving or retiring because of the injury or illness and record the case accordingly. If the employee's decision to leave or retire is a result of the injury or illness, this paragraph requires the employer to estimate and record the number of calendar days away or on restricted work/job transfer the worker would have experienced if he or she had remained on the employer's payroll. This provision also states that, if the employee's decision was unrelated to the injury or illness, the employer is not required to continue to count and record days away or on restricted work/job transfer.

Paragraph 1904.(b)(3)(viii) directs employers how to handle a case that carries over from one year to the next. Some cases occur in one calendar year and then result in days away from work in the next year. For example, a worker may be injured on December 20th and be away from work until January 10th. The final rule directs the employer only to record this type of case once, in the year that it occurred. If the employee is still away from work when the annual summary is prepared (before February 1), the employer must either count the number of days the employee was away or estimate the total days away that are expected to occur, use this estimate to calculate the total days away during the year for the annual summary, and then update the Log entry later when the actual number of days is known or the case reaches the 180-day cap allowed in § 1904.7(b)(3)(v).

OSHA received a large number of comments on how days away should be counted. The issues addressed by commenters included (1) whether to count scheduled workdays or calendar days, (2) whether the day counts should be "capped," and, if so, at what level, (3) how to count days away or restricted when employees are terminated or become permanently disabled, and (4) how to handle cases that continue to have days away/ restricted from one year to the next.