• Publication Date:
  • Publication Type:
    Final Rule
  • Fed Register #:
  • Standard Number:
  • Title:
    Reporting Occupational Injury and Illness Data to OSHA; Final Rule


Occupational Safety and Health Administration

29 CFR Part 1904

[Docket No. R-02]

RIN 1218-AB24

Reporting Occupational Injury and Illness Data to OSHA; Final Rule

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

ACTION: Final rule.

SUMMARY: This final rule amends 29 CFR Part 1904 by adding section 1904.17. Section 1904.17 requires employers to report information to OSHA contained in records that employers are required to create and maintain pursuant to Part 1904, and the number of workers they employed and hours their employees worked during designated periods.

Section 1904.17 will clarify OSHA's authority to collect establishment-specific data by mail for use in agency self-evaluation, deployment of agency resources, periodic reassessment of existing regulations and standards, and rulemaking.

Section 1904.17 was proposed (as section 1904.13) as part of a comprehensive proposal to revise Part 1904. 61 FR 4030 (Feb. 2, 1996). OSHA has determined, however, to take final agency action with respect to section 1904.17 at this time, and to take final action on the remaining Part 1904 issues, including other records access issues, at a later date.

DATES: This final regulation will become effective on March 13, 1997. However, affected parties do not have to comply with the information collection requirements until the Department publishes in the Federal Register the control numbers assigned by the Office of Management and Budget (OMB) to these information collection requirements. Publication of the control numbers notifies the public that OMB has approved these information collection requirements under the Paperwork Reduction Act of 1995.

FOR FURTHER INFORMATION CONTACT: Bonne Friedman, U.S. Department of Labor, Occupational Safety and Health Administration, Office of Information and Consumer Affairs, Room N-3647, 200 Constitution Avenue, NW., Washington, DC 20210, phone (202) 219-8148. For electronic copies of documents, contact the Labor News Bulletin Board at (202) 219-4784, or OSHA's WebPage on the Internet at http://www.osha.gov/. For news releases, fact sheets, and other short documents, contact OSHA FAX at (900) 555-3400 at $1.50 per minute.


I. Background

In 1971, OSHA issued the occupational injury and illness recording and reporting regulation, 29 CFR Part 1904. Part 1904 includes regulations pertaining to criteria for determining whether an occupational injury or illness should be recorded, and provisions that require employers to give employees and OSHA access to such records. It also provides for collection by the Bureau of Labor Statistics (BLS) of data to be used in an occupational injury and illness statistical program administered by BLS. 1904.20, 1904.21, and 1904.22.

In 1990, the Secretary of Labor transferred some of BLS's statistic-gathering functions to OSHA. 55 FR 9033 (Mar. 9, 1990). BLS retains responsibility for conducting its Annual Survey of Occupational Injuries and Illnesses and will continue to issue data that is aggregated by SIC group. But OSHA will also be responsible for administering a national recordkeeping system for occupational injuries and illnesses whose data will be site-specific.

OSHA's February 1996 proposal to revise Part 1904 sought, among other things, to reflect OSHA's new statistics-gathering responsibilities. OSHA proposed to replace sections 1904.20, 1904.21, and 1904.22 with a single reporting provision at 1904.13, which would apply to both BLS and OSHA collections of information by mail or other remote transmittal.

OSHA received 449 written comments and held six days of public meetings. Approximately 124 comments and two oral presentations specifically addressed proposed section 1904.13.

On further consideration, OSHA determined that BLS and OSHA need separate provisions for collection of data by mail. Thus, a single provision applicable to both agencies would not be appropriate, and a new provision specifically addressed to OSHA reporting requirements and procedures should be developed. OSHA further determined to take final action on proposed 1904.13 at this time, and to take final action with respect to the remainder of the proposed revisions of Part 1904 at a later date.

This final rule revises the proposed section 1904.13 and renumbers it as section 1904.17, the next available number in Part 1904. This final rule does not modify or delete the existing regulations at 1904.13, 1904.20, 1904.21, or 1904.22.

II. Explanation of the Final Rule

OSHA has long had in effect rules pertaining to OSHA access to certain information. Section 1904.7 requires employers "to provide, upon request, records provided for in Secs. 1904.2, 1904.4, and 1904.5 [OSHA-required injury and illness logs and forms] for inspection and copying by any representative of the Secretary of Labor. * * *" Section 1910.1020 requires employers to give OSHA and employees the right and opportunity to examine and copy exposure and medical records. Some standards contain requirements for OSHA and employee access to exposure and monitoring data required to be created and maintained by those particular standards. E.g., 29 CFR 1910.1001(m)(5)(I) and (ii) (requiring that OSHA and employee be given access to asbestos exposure monitoring and medical surveillance records).

Section 1904.17 establishes a procedural mechanism for conduct of an annual survey of ten or more employers by mail or other remote transmittal. Information covered by section 1904.17 is information contained in records required to be created and maintained pursuant to Part 1904, the number of workers the respondent employed and the number of hours worked by its employees during designated periods. The rule also specifies that both the request and the response will be made by mail or other remote transmittal. Thus, it is more limited than existing records-access provisions that use terms such as "permit access to" or "make available" and therefore permit OSHA to collect information by on-site record reviews as well as via mail response. The mail-in provision also permits OSHA to coordinate its annual survey with the BLS annual survey. In conducting its 1995 and 1996 annual surveys (1995 data was collected in 1996, 1996 data will be collected in 1997) OSHA provided employers with a carbon-pack form that the employer could complete, separate, and return -- one copy to BLS and another to OSHA. OSHA intends to continue this practice or an equivalent means of avoiding duplicate reporting burdens for employers.

The requests for data reports may be made directly by OSHA, or may be sent to employers by a designee of the Agency, such as a state governmental agency, a government contractor, or another Federal agency such as the National Institute for Occupational Safety and Health (NIOSH). Designating others to exercise this authority will permit a variety of collection methods to be used, depending on which method is the most effective, efficient, and cost effective for the government.

Employers who are normally exempt from keeping injury and illness records under 29 CFR 1904.15 and 29 CFR 1904.16 may be notified by OSHA that they will be required to participate in a particular information collection under 1904.17(a). OSHA will notify these employers in writing in advance of the year for which injury and illness records will be required. OSHA does not expect, in the near term, to take action against 1904.15 and 16 exempt employers based on survey non-response under 1904.17.

III. Issues

1. Use of Data

As explained above and in the proposal, site-specific data reported pursuant to section 1904.13 (now section 1904.17) will be used for a variety of purposes: injury/illness surveillance; development of information for promulgating, revising or evaluating OSHA's safety and health standards; evaluating the effectiveness of OSHA's enforcement, training and voluntary programs; public information; and for directing OSHA's program activities, including scheduled workplace inspections and non-enforcement programs, such as targeted mailings of safety and health information to employers.

Many commenters acknowledged OSHA's need for a reporting requirement or affirmatively stated they had no objections to it. (Ex. 15: 80, 184, 239, 313, 341, 359, 384, 418, 449) However, some commenters who had no objection to the principle of a reporting requirement, expressed concern about the uses to which the data would be put. (Ex. 15: 117, 181, 304) The National Federation of Independent Business argued, for example, that the data should be used for compliance efforts only:

NFIB strongly objects to this provision unless it is expanded to provide adequate safeguards to prevent abuses of written requests, especially for reasons other than OSHA compliance -- i.e., research, surveillance, or public information. In fact, NFIB questions the need for OSHA to have access to data for non-compliance reasons at all. This is another instance where it appears as if OSHA has overstepped its legislative bounds and is attempting to transform a recordkeeping/compliance system into a comprehensive research system of occupational safety and health statistics.

(Ex. 15: 304, p. 25)

Others contended that the data should be used for statistical purposes only. See e.g., Heat Transfer Equipment Company (Ex. 15: 117)("rules must be in place that the information will be used for statistical purposes only and not as a method for determining individual audits and retribution").

The OSH Act directs OSHA to operate a broad program to assure safe and healthy workplace conditions in the majority of America's workplaces, nearly 6,000,000 individual workplace establishments employing approximately 100,000,000 workers. A vital component of this broad program involves the effective use of information to provide for the purposes discussed in the introduction to the OSH Act: for workplace safety and health enforcement, research, information, education, and training. 29 U.S.C. 651.

Section 24 of the Act, 29 U.S.C. 673, directs the Secretary of Labor, in consultation with the Secretary of Health and Human Services, to develop and maintain a program of collection, compilation, and analysis of occupational safety and health statistics. Section 8(c) also directs the Secretary of Labor, in cooperation with the Secretary of Health and Human Services, to prescribe regulations requiring employers to maintain accurate records of, and to make periodic reports on, work-related deaths, injuries, and illnesses.

Additionally, the Government Performance and Results Act of 1993(GPRA)(31 U.S.C. 1101) requires Federal agencies to implement a program of strategic planning, develop systematic measures of performance to assess the impact of individual government programs, and produce annual performance reports.

OSHA believes that collecting injury, illness and employment data from employers to meet these responsibilities represents the most appropriate policy. OSHA also needs establishment-specific data to better target its program activities, including workplace inspections and non-enforcement information and incentive programs, to the more hazardous workplaces. Given budget and personnel constraints, OSHA and the 23 states with OSHA-approved workplace safety and health plans are unable to work directly with all of these workplaces. In fiscal year 1996, OSHA and the States conducted enforcement inspections at approximately 80,000 workplaces (unpublished OSHA analysis of FY 1996 inspection data). At this rate, 75 years would be needed to inspect all of America's workplaces.

Several independent reports concerning occupational injury and illness recordkeeping and occupational safety and health policy have documented and supported OSHA's need for establishment-specific data. In a 1987 report, Counting Injuries and Illnesses in the Workplace: Proposals for a Better System, published by the National Research Council (NRC), the Panel on Occupational Safety and Health Statistics recognized OSHA's need for access to individual establishment data:

The Occupational Safety and Health Administration should be able to obtain individual establishment data and that this might be achieved through the development of an administrative data system, such as that maintained, for example, by the Internal Revenue Service.

(Ex. 4, p. 10)

The panel believed that this data could be used to improve OSHA's enforcement program:

It could provide systematic detailed data that the current program does not now provide; it could give OSHA more effective ways of using its inspection resources to reduce workplace injuries; and it could provide a more systematic bases for monitoring the quality of recordkeeping and reporting.

(Ex. 4, p. 113)

The NRC Panel further suggested that an administrative data system based on the OSHA 200 logs could provide a valuable database for other uses as well, including standard setting, enforcement, program evaluation, and research. (Ex. 4, p. 113) In a 1989 report, the Keystone National Policy Dialogue on Work-Related Illness and Injury Recordkeeping, a group of industry, labor, government and academic representatives with an interest in occupational injury and illness data stated:

The Dialogue group agreed that injury and illness statistics from recordkeeping can and should be used to target (prioritize) enforcement/compliance activity at OSHA.

* * * * *

The data should be usable for macro purposes by SIC codes (high risk -- low risk) as well as in a performance oriented micro targeting of workplace visits. OSHA needs to conserve its resources and should be able to decide upon which industries and workplaces should receive the most attention. However, statistics alone should not be used to exempt any site from inspection. The records and rates at the site level should be used in decision making in conjunction with a review of site programs and spot check inspections.

(Ex. 5, p. 35)

In a 1990 report, Options for Improving Safety and Health in the Workplace, the General Accounting Office (GAO) discussed an option for improving the use of inspection resources by targeting inspection activity with the use of establishment-specific injury and illness data:

OSHA could focus its enforcement, as well as education and training efforts, on employers with high injury and illness rates in industries known to be hazardous.

(Ex. 36, p. 32)

OSHA believes that it can improve the effectiveness and efficiency of its programs by focusing its resources on employers and workplaces that are experiencing serious, ongoing workplace safety and health problems reflected by high rates of workplace injuries and illnesses. At the same time, data that shows workplaces with good safety and health records reflected by low injury and illness rates would allow OSHA to have greater flexibility in working cooperatively and in partnership with safer workplaces. These programs include enforcement programs as well as non-enforcement programs that encourage employers to voluntarily implement effective safety and health programs that protect workers from death, injury and illness.

2. The Use of Alternative Data Sources

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 collecting information from employers. (Ex. 15: 2, 28, 58, 63, 97, 184, 195, 289, 327, 341, 374, 444) For example, Mr. Alex F. Gimble, CSP 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 workers' compensation systems. The comments of the American Health Care Association (AHCA) are representative:

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 data from treatment facilities:

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)

OSHA believes that injury and illness information compiled pursuant to Part 1904, plus employment figures, will be much more reliable and suited to OSHA's needs than any available alternative. While many State workers' compensation programs voluntarily provide injury and illness data to OSHA for various purposes, others do not. And the data vary widely from state to state. Differing workers' compensation laws and administrative systems result in large variations in content, format, accessibility and computerization. Often, workers' compensation databases 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, and incidence rates of occupational injury and illness cannot be computed. Workers' compensation data are also based on insurance accounts, and not on the safety and health experience of individual workplaces. As a result, an individual account often reflects the experience of several workplaces involved in differing business activities.

Only a survey of every member of a selected set of employers about a selected set of data gathered in a relatively short time can tell OSHA which members of the group have the highest or lowest illness and injury rates, how the injury and illness rates are distributed over the field, and the types of injuries and illnesses being experienced in that field, etc. As more surveys are conducted over time, a reliable historical record will emerge.

While OSHA does not believe that alternate source data are satisfactory substitutes for the information covered by 1904.17, the agency does recognize they have value. To the extent information from workers' compensation programs, BLS, 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 assess the quality of its own data. However, consistent with the Congressional mandate of the OSH Act, OSHA needs to maintain its own recordkeeping system and to gather the data for it through a reporting requirement.

3. Scope Issues

Many commenters objected to the breadth of the proposed regulatory text, arguing that it would give the Secretary unfettered discretion to demand any information related to the Act's purposes, at any time, for virtually any reason. (Ex. 25, 58X, 15: 55, 80, 102, 124, 135, 144, 158, 162, 165, 193, 206, 207, 209, 211, 212, 220, 228, 239, 240, 243, 252, 255, 257, 258, 261, 264, 267, 274, 275, 276, 286, 293, 305, 306, 309, 313, 341, 348, 351, 368, 375, 389, 397, 406, 420, 427) A comment by the National Association of Manufacturers sums up the point of view expressed by many others:

It is one thing to have an objectively identified set of employers that must make an annual filing of a census-type survey on a non-discriminatory basis; it is another to give an enforcement agency the authority -- at its sole whim or discretion -- to selectively require one or more employers to file reports that an entire class of employers is required to maintain. It is one thing to have an objectively identified set of information or records that must be included in an annual filing; it is another to give an enforcement agency the authority -- at its sole whim or discretion -- to selectively require one or more employers to generate and file reports containing whatever information the agency identifies so long as it can be described as "regarding [the employer's] activities relating to this [OSH] Act."

(Ex. 25, 15: 305)

It was not OSHA's intention to exercise unfettered discretion to collect any data related to the Act. It was, however, OSHA's intention to create a reliable mechanism for routinized collections, by mail or other remote transmittal, of a limited class of information without unduly burdening employers. Consistent with that goal, and in light of the comments of record, the final reporting rule is carefully circumscribed. The rule authorizes an annual survey -- which, because it will go to more than ten employers, will be subject to the Paperwork Reduction Act (PRA) (See 42 U.S.C. 3502 et seq. and 5 CFR part 1320) -- concerning information contained in records required to be created and maintained by Part 1904 plus employment figures. The rule specifies the time within which responses are to be provided to OSHA. Employers will be able to determine which employers are within the survey group and what information will be collected each year before the survey begins because that information will be made available to the public under a Federal Register notice pursuant to the PRA. Once a survey has received an OMB control number under the PRA, any substantive or material modification would require a new PRA clearance. As indicated in Section IX of this preamble entitled "Paperwork Reduction Act of 1995" the OMB control number for the current annual survey form is 1218-0209. (Section 1904.17 defines the class of information and respondents subject to survey under the rule. The set of employers and information (from within the covered class) to be targeted in each year is fixed as each survey is designed.) One commenter was concerned that the proposed rule could apply to information dating back "decades," creating substantial burdens for employers. (Ex: 15:395, p. 67) Since the final rule establishes an annual survey of information in Part 1904 records, which are required to be kept no more than five years, plus employment information, it presents no issues about "decades-long" records.

A number of commenters argued that as proposed, section 1904.13 violated Fourth Amendment guarantees against unreasonable searches. (Ex. 15:154, 174, 193, 215, 258, 305, 318, 346, 375, 390, 395, 397) Most of these commenters referred to Marshall v. Barlow's, Inc., 436 U.S. 305 (1978), McLaughlin v. Kings Island, 849 F.2d 990 (6th Cir. 1988), and Brock v. Emerson Electric Co, 834 F.2d 994 (11th Cir. 1987).

Barlow's concerned the question whether OSHA must have a warrant to inspect a work site if the employer does not give consent. Kings Island and Emerson Electric concerned on-site records inspections by compliance officers. Section 1904.17 is a reporting requirement; no entry of premises or compliance officer decision making is involved. Thus, these decisions provide little if any support to the commenter's sweeping Fourth Amendment objections. See, Donovan v. Lone Steer, Inc., 464 U.S. 408, 414 (1984) (reasonableness of a subpoena is not to be determined on the basis of physical entry law, because subpoena requests for information involve no entry into nonpublic areas).

Moreover, in its final form the rule is extremely narrow in scope and leaves the agency with limited discretion. Section 1904.17 is restricted to a limited class of information. This information is highly relevant to accomplishment of OSHA's mission. The reporting is done by mail or other remote transmittal, without any intrusion into the employer's premises by OSHA, and is not unduly burdensome. Much of the injury and illness information to be reported is taken from records employers are already required to create, maintain, post, and provide to workers and government officials on request, which means that the employer has a reduced expectation of privacy in the information. Employment figures are critical to OSHA's ability to evaluate the injury and illness data, whereas they are not information that employers may expect to keep secret from the government. In addition, as explained earlier, there is no substitute for a large body of site-specific information gathered by the survey method. The results of the surveys will be uniquely useful to OSHA in meeting Congress' mandate to use reporting requirements and build an effective statistical program around them.

Some commenters argued that the Fourth Amendment requires OSHA to use a subpoena or warrant to get information from employers who do not provide it voluntarily. Since the proposed reporting rule made no explicit provision for enforcement via subpoena or warrant, they contended that the rule was constitutionally deficient. "Production may not be compelled without a search warrant, administrative subpoena or other appropriate vehicle." (National Beer Wholesalers Association. Ex. 15:215.) "The Fourth Amendment * * * requires OSHA to obtain a subpoena or warrant prior to obtaining access to any of the information identified in proposed * * * 1904.13." (The Fertilizer Institute. Ex. 15: 154.) "The proposed rules make no provision for a subpoena or warrant and appear to contemplate that OSHA will use neither. * * * These provisions, to the extent they purport to authorize inspections of records without a warrant or subpoena, violate the Fourth Amendment." (American Iron and Steel Institute. Ex. 15:395.) Certainly, under many circumstances employers can force OSHA to secure a warrant or subpoena enforcement order before giving OSHA access to workplace injury and illness data. These commenters, however, appear to be arguing that including a subpoena or warrant enforcement mechanism in the text of the rule is necessary to adequately protect their Fourth Amendment right to privacy. This is not so. The Fourth Amendment protects against "unreasonable" intrusions by the government into private places and things. Reporting rules that do not incorporate subpoena or warrant procedures are not "unreasonable" per se. See e.g., California Bankers Ass'n v. Shultz, 416 U.S. 21, 67 (1974) (upholding reporting regulation issued under the Bank Secrecy Act of 1970 that did not provide for subpoenas or warrants where the "information was sufficiently described and limited in nature and sufficiently related to a tenable Congressional determination" that the information would have a high degree of usefulness in criminal, tax, or regulatory investigations or proceedings). For example, OSHA has long required employers to report promptly all fatal workplace accidents.

The totality of circumstances surrounding a warrantless or "subpoena-less" reporting requirement or administrative investigation determines its reasonableness. For example, in McLaughlin v. A.B. Chance, 842 F.2d at 727 (4th Cir. 1988), the Fourth Circuit upheld a records access citation against an employer who refused an OSHA inspector access to its OSHA Logs and Forms on the ground that it had a right to insist on a warrant or subpoena. The court upheld the citation because a summary of the information was posted annually on the employee bulletin board, thus diminishing the employer's argument that it has a reasonable expectation of privacy in the information, and the inspector was lawfully on the premises to investigate a safety complaint. In New York v. Burger, 482 U.S. 691, 702-703 (1987), the Supreme Court noted that agencies may gather information without a warrant, subpoena, or consent if the information would serve a substantial governmental interest, a warrantless (or subpoena-less) inspection is necessary to further the regulatory scheme, and the agency acts pursuant to an inspection program that is limited in time, place, and scope. The Burger court went on to uphold a warrantless inspection of records during an administrative inspection of business premises. Consider also the Kings Island and Emerson Electric decisions' concern about the inspector's broad field discretion. Kings Island (noting that under Burger a warrantless or subpoena-less inspection of records might be reasonable, but concluding that the facts of the case did not satisfy Burger analysis); Emerson Electric (noting that under California Bankers an agency may gain access to information without a subpoena or warrant but concluding that facts of that case were not comparable to those reviewed in California Bankers).

It is not OSHA's intention to resolve, in this rulemaking, the question of the procedures the Fourth Amendment may require to enforce the regulatory obligation. Not only are Fourth Ammendment issues ultimately for courts, not agencies to resolve, such issues are rarely suitable for judgement in the abstract. If for example, OSHA were at some future time to issue a citation for nonresponse to a survey questionaire, the Fourth Amendment evaluation would depend on all the particulars of the case. (While the participation in the OSHA Data Collection Initiative is mandatory, OSHA has made a policy decision that it will not issue citations for the failure to respond to the first survey conducted under authority of this rule, which will collect data for calendar year 1996; nor does OSHA intend to issue citations for the 1995 survey already conducted. OSHA will take into consideration its experience with the Data Collection Initiatives when developing policy for future years. However, the nonrespondents to the 1995 and 1996 survey instrument may be subject to an on-site records inspection by an OSHA compliance officer or issued an administrative subpoena.) Further analysis under the principles set forth in the Burger decision must await a specific application of 1904.17 when the particulars of the information request are known. OSHA has, however, structured the final rule to respond to concerns expressed in the case law and to limit its own discretion and eliminate discretion of officials in the field. Section 1904.17 surveys are constrained first by the regulatory text -- the surveys occur no more than once per year, they involve ten or more employers covered by the Act, they are limited to injury and illness information contained in records created and maintained pursuant to Part 1904 and to employment and hours worked, they are accomplished by mail or other remote transmittal, and respondents have at least thirty days to respond. The data from within the covered field and the set of employers or establishments to be canvassed for each survey are definitively fixed during the Paperwork Reduction Act clearance process and are available to the public in connection with Federal Register notices published during the clearance process.

Employers will have ample opportunity to test the Fourth Amendment reasonableness of any survey with which they are faced. Under any follow-up scenario -- warrant records inspection, subpoena demand or notice of a 1904.17 violation -- employers would have advance notice that a response was required, and would have an opportunity to provide the survey data in order to avoid legal process. Employers faced with a survey that they consider an infringement of Fourth Amendment rights of privacy may refuse to respond and raise objections in a warrant enforcement or subpoena proceeding or as a defense if they are issued citations by OSHA. Under the Act, employers are entitled to contest citations and receive an administrative hearing, administrative review of the hearing officer's decision, and federal court of appeals review. 29 U.S.C. 659(c), 660(a).

Some commenters asserted that using reported information for enforcement targeting would violate their privilege against self-incrimination. (Ex. 15:203, 397) These commenters did not explain how the privilege against self-incrimination would be implicated in the reporting requirement or cite any supporting authorities. OSHA would point out, that the privilege against self-incrimination derives from the Fifth Amendment and pertains to criminal proceedings. It has long been settled that the privilege cannot be invoked to resist the disclosure needed for a regulatory purpose unrelated to the enforcement of criminal laws even if a criminal proceeding is a possible consequence of an administrative investigation. See, for example, Shapiro v. United States, 335 U.S. 1, 32-33 (1948) (Fifth Amendment not violated by regulation requiring individuals to keep and produce records "of transactions which are the appropriate subjects of governmental regulation').

4. OSHA's Statutory Authority To Collect Data With a Reporting Rule

Some commenters argued that the proposed reporting rule was not consistent with Sections 8(c) and 24(e) of the Act. Sections 8(c)(2) directs that "the Secretary of Labor * * * 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 * * *." 29 U.S.C. 657(c)(2). Section 24(e) provides that "[o]n the basis of the records made and kept pursuant to section 8(c) of this Act, employers shall file such reports with the Secretary as he shall prescribe by regulation * * *." 29 U.S.C. 673(e).

These commenters argued that the proposed rule merely reiterated the Secretary's entire range of statutory authority to collect information and did not itself prescribe anything, much less limit itself to the injury and illness records mentioned in section 8(c)(2). Moreover, some claimed, it left the compliance officer in the field with unfettered discretion to decide what information to demand. (Ex. 15: 154, 313, 352, 353, 358, 375, 397.) There are several responses to be made on this point. First, OSHA has had the ability to access injury and illness records for many years and is simply clarifying its authority to collect the information through the mail. Second is the fact that the final rule is extremely narrow and specific about the information it covers and how that information is to be gathered. Third, compliance officers do not implement the rule; the agency implements it by conducting large annual surveys, by mail, requesting information within the scope of the rule from employer or establishment groups whose responses the agency judges to be necessary in meeting its multiple responsibilities. Finally, the final rule fits within the terms of Section 8(c).

5. Time Allowed for Employers To File Reports

The proposed rule would have required employers to submit data to OSHA, when OSHA sends them a written request for records, within 21 calendar days of receiving the request. Several commenters provided remarks on the 21 calendar day limitation. (Ex. 15: 65, 127, 347, 405) Some comments supported the 21 day time frame as a reasonable time for employers to comply with a request for information. (Ex. 15: 347, 405) For example, the Westinghouse Company (Ex. 15: 405, P. 4) stated: "This change is acceptable and the time limitations appear reasonable."

OSHA also received comments stating that 21 calendar days is too short a time frame for reporting, and that longer times should be adopted in the final rule. (Ex. 15: 65, 127) For example, the Aluminum Company of America (Alcoa) remarked:

Alcoa believes this is too short and restrictive a time frame given current staff levels and resource demands on employers and their health and safety professionals. * * * OSHA should provide 30 days advanced notification (for planning purposes) and 21 days for response following the advanced notification to the specific employers to be surveyed.

(Ex. 15: 65)

The Laboratory Corporation of America stated:

Reports to be required of employers mentioned in 29 CFR 1904.13 should be handled in one of two ways. The content of the reports needs to be established in advance and a specific date for a deadline for submission provided. Alternatively, if the report content has not yet been established, then a period of time longer than 21 days is needed for response. A period of 45 to 60 days is suggested. Unless the information requested is known in advance to employers, it will take time to communicate and collect this data in a multi-state, multi-location operation. Either of these two options would give more appropriate time for more accurate information to be compiled for these types of employers.

(Ex. 15:127 P. 2)

Other comments supported the 21 day requirement, but suggested that the Secretary maintain some flexibility and discretion to provide more than 21 days for a specific request.

The American Petroleum Institute (API), for example, observed:

Twenty-one days should be the minimum time allowed for employers to respond to such requests.

Recommended language: The employer shall file the requested reports with the Secretary within 21 calendar days of receipt of the request, unless the Secretary allows more than 21 days.

(Ex. 15:375 P. b25)

In light of these comments, OSHA has increased the reporting time to 30 calendar days in this final rule. OSHA believes that the 21 day time frame may be too short for some employers to comply with the request, but believes that 45 or 60 days is too long a time frame for a relatively simple request for summary information contained in existing records. A longer deadline would make it more difficult for OSHA to collect data in a timely fashion, or to conduct quality control measures such as follow-up mailings and phone calls to verify questionable or erroneous data.

Additionally, OSHA agrees that the time frame in the rule should be a minimum time that can be lengthened at the discretion of OSHA. In other words, the final rule requires employers to file reports within 30 calendar days of receipt of the request, unless the written instructions contained in the request specifically allow more than 30 calendar days.

6. Reporting With Computers

OSHA received several comments on the potential role of computers in reporting data to OSHA. (Ex. 15: 011, 163, 184, 390, 402) The OSHA Data Company (Ex. 15: 011) suggested that computer reporting should be a mandatory feature of the data collection system, remarking: "We suggest that recordkeeping in computer readable format should be mandatory and data should be submitted to OSHA in that format."

Other commenters suggested that computer reporting be allowed and encouraged (Ex. 15: 163, 184, 390, 402). The comments of US West Inc. are representative of these comments:

US West requests that OSHA move to implement systems that will allow employers to electronically provide data, such as the data requested in the BLS Survey of Occupational Injuries and Illnesses. Such a method will be more effective, in terms of receiving consistently formatted data, and will be more cost efficient for both employers and the Department of Labor.

(Ex. 15-184)

OSHA believes that there is enormous potential for reducing collection burden on both employers and the government, while improving data quality and consistency, by allowing employers to submit data through computerized reporting systems. However, OSHA does not believe that computerized reporting systems should be mandatory for all employers. Mandatory computer systems could actually increase the burden on those employers who do not have computer systems and on those employers who have computer systems that do not provide simple electronic communications options.

OSHA intends to implement, as soon as possible, options for individual data collection projects that will allow employers to submit data either electronically or through paper forms. For those data collections where computerized submission of data is an option, OSHA will include instructions for computerized submissions in the instructions accompanying the request for information.

7. Miscellaneous Issues

OSHA also received comments on a variety of issues that the Agency believes are worthy of discussion, as follows.

A. The Ability of OSHA To Designate its Collection Authority to Another Entity. The Proposed Rule Did Not Indicate That a Designee Could Collect Information for the Agency Often, OSHA and the Bureau of Labor Statistics have used grants to the states and independent government contractors to collect data on behalf of the Department of Labor. These arrangements allow the Department to collect information using a variety of administrative options that are advantageous to the Federal government and do not increase the burden on respondents. One commenter suggested: "Data should continue to be collected through state agencies." (Ex. 15: 41) In order to maintain the Agency's flexibility to collect data via grants to the states, or to use government contractors, and to be able to collect data through cooperative interagency efforts with the Department of Health and Human Services, OSHA has modified the final rule to require employers to submit information to either OSHA or OSHA's designee.

B. Unfair Effect on Specific Industry Sectors

Several commenters raised concerns over what they regarded as potentially unfair effects of the data collection on smaller employers, small establishments, and employers who rely heavily on part time employees (Ex. 15: 304, 384, 424, 449). Another commenter was concerned that OSHA would attempt to compare data from the longshoring industry to that of other industries and argued that such comparisons would be invalid because longshoring is subject to a different workers' compensation insurance system than other industry sectors (Ex. 15: 95).

Several commenters expressed concern over a perceived and potentially unfair effect of data collections on smaller employers, arguing that the same small number of cases would result in a higher incidence rate for a smaller employer than for a larger employer, or that a small employer may have a high rate for only one year and may have had no cases for many years before and after the year for which the information is collected. (Ex. 15: 304, 384, 449) For example, the Akzo Nobel Corporation observed:

We support this concept, but caution OSHA about using data from only one year, especially for small sites where a single medical case in a plant of 20 employees will give a total recordable rate of about 5. We would consider that a "high" rate, possibly targetable by OSHA, but it might be the first OSHA recordable incident in 3 or 5 years. Caution is advised.

(Ex. 15: 384)

United Parcel Service (UPS) (Ex. 15: 424, p. 9) expressed a concern about the possible effect on firms who rely heavily on part-time labor, stating:

The agency's current practice of determining injury rates as a ratio to hours worked, rather than to employees, has the consequence of inflating injury and illness rates for companies with more workers per hour worked: at least when an outside limit of an 8-hour workday is established, the likelihood, per hour, of injury decreases when more hours are worked. To put it another way, the more workers who work per 8-hour day, the more likely those hours will generate discrete employee complaints. Therefore, OSHA's current practices already distort the apparent safety of workplaces relying heavily on part-time labor.

The Pacific Maritime Association (Ex. 15: 95, p. 10) expressed a concern that injury and illness reports would not provide an accurate comparison with other industries because the longshoring industry is covered by a separate workers' compensation system, stating:

Another very important recommendation concerns the inequities of comparing an industry covered by the Long Shore and Harbor Workers Act compensation program with those covered by Workers' Compensation. Compensation provided by the Long shore program is much more generous than Workers' Compensation and may encourage individuals to remain on compensation longer. This disparity between the two systems is not often acknowledged particularly when injury incident and severity rates are used to identify high hazard industries. It is recommended that OSHA recognize the impact of the Long shore compensation by establishing a specific category for employees who are covered by the Long shore Act. For an example, SIC 4491, Long shoring, may be used as a specific category where employer incident and severity rates may be compared.

These objections are premature, as they relate to certain possible uses of data, not to usefulness for all purposes, and not to the Agency's authority to collect the data in the first instance. Moreover, as the comments themselves made clear, when the time comes for using survey data, it will be possible to factor in special circumstances for subgroups of employers. For example, small employer data could be adjusted to omit smaller employers with only one injury from any analysis of the data.

In regards to the longshoring industry, OSHA has traditionally performed separate analyses of broader databases to prepare employer lists specific to the longshoring industry. OSHA recognizes the unique qualities of this industry, has developed separate standards for maritime industries, including longshoring, and normally performs specialized investigations for longshoring facilities. The problems with data from the longshoring industry can be solved by continuing to look at this industry in a way that does not compare these employers to employers in other industries.

In general, OSHA believes that different approaches to the use of data can effectively deal with differences among different subpopulations of employers, depending on the unique qualities of those subpopulations. OSHA will continue to tailor its analysis of data when these unique situations are encountered.

C. Data Quality Issues

Several commenters discussed the possible adverse impacts on the quality of the data if reporting is required. (Ex. 15: 50, 122, 176, 273, 301, 310, 374, 401, 414). Mr. George R. Cook, CCC-A (Ex. 15: 50) remarked:

If the OSHA Form 300 is to be used to prioritize compliance visits, it is felt this policy will add undue pressure for companies to keep entries off the Form.

The Laborers' Health & Safety Fund of North America (Ex. 15: 310) observed:

The premise of employers self-reporting injuries and illnesses to an agency which may inspect them based on that data is a prescription for mis-reporting.

The Chemical Manufacturers Association (CMA) remarked:

CMA supports targeting of inspections in order for OSHA to better use its resources, but cautions OSHA to carefully consider its approach. CMA is concerned that OSHA carefully consider the relationship between targeting and OSHA's ability to collect accurate and credible data. Valid data collection and analysis are the cornerstone of effective targeting.

CMA recognizes that currently OSHA is not collecting adequate data to target effectively. It is important that OSHA review existing data sources, examine existing targeting programs (e.g. Maine 200) and revise its data collection mechanisms. However, the Administration must carefully evaluate the context in which that data has been collected, as well as identify characteristic flaws in such programs.

(Ex. 15: 301, p. 16)

The quality of any data collected from employers is an ongoing concern for the Agency. OSHA agrees that misreporting, whether intentional or unintentional, can affect the value of the collected data and any conclusions drawn from that data. Misreporting is not, however, an insoluble problem. Controls are available for assuring a reasonable quality of data for use by OSHA, as well as employers and workers. For example, OSHA is implementing a quality control initiative for the current collection of injury and illness records data required by Part 1904 that will include three components; outreach and training for the regulated community to reduce unintentional errors, error screening and follow-back procedures to correct or verify questionable data reported to the agency, and, under certain circumstances, on-site records inspections. OSHA is also planning to use other sources of data, e.g., workers' compensation records and inspection histories, when available, for comparison purposes as an external check on records validity.

D. Effect on Existing Authority

Nothing in Section 1904.17 affects the Secretary's general investigatory authority under Section 8 of the Act or his broad rulemaking authority under Section 8(g)(2).

IV. Economic Analysis

Section 1904.17 applies to all employers within OSHA jurisdiction, including those in general industry, construction, shipyard employment, long shoring, marine terminals, and agriculture. OSHA has determined that the Section 1904.17 regulation does not require the Agency to develop a Final Economic Analysis because it is not a "significant regulatory action" as defined by section 3(f)(1) of Executive Order (E.O.) 12866. This provision of the E.O. covers a regulatory action that is likely to result in a rule that may:

(1) Have an annual effect on the economy of $100 million or more or adversely affect in a material way the economy, a sector of the economy, productivity, competition, jobs, the environment, public health or safety, or State, local, or tribal governments or communities.

Pursuant to this section 1904.17 individual data collections conducted under this regulation will require employers to assemble data and file reports to OSHA. To provide employers with examples illustrative of the kinds of costs and paperwork burdens potentially associated with such data collections, the following paragraphs describe the costs and burden hours associated with two recent Agency data collection efforts. The examples chosen include the two recent data collection initiatives undertaken by OSHA in 1995 and 1996.

The impact analyses developed for the 1995 and 1996 data collections initiatives were published in the Federal Register (60 FR 35231; 61 FR 38227, respectively). OSHA estimated that employers responding to those data collection efforts would be required to spend an estimated $6.95 per response, based on 30 minutes of clerical time at $13.90 per hour. OSHA believes that most firms will assign the survey form to a personnel or payroll clerk with an average wage of $13.90 per hour. This figure is based on a wage rate with benefits for a secretary-typist from Employment and Earnings, January 1996, U.S. Department of Labor, Bureau of Labor Statistics (OSHA has recently updated its wage rate data with more current statistics). The information collected from employers in the 1995 and 1996 data collection initiatives was summary information from the establishment's OSHA Log and Form 200, in addition to information on the number of workers employed and the number of hours worked by these employees in the applicable calendar year. Approximately 70,000 employers were targeted in each of these data collection initiatives, for a total burden estimate of 35,000 hours, or $486,500. OSHA anticipates that future data collection initiatives conducted under section 1904.17 will impose similar burdens -- approximately 30 minutes of clerical time per respondent -- and will therefore not impose a substantial burden on any employer.

The record contains many comments about the burden of recording employment and hours worked information on the OSHA Log -- some favorable but more unfavorable. However, the negative commenters provided no empirical basis by which their burden claims could be quantified. In the absence of such data, OSHA turned to the long experience BLS has accumulated while collecting these same types of data for statistical purposes. For over 25 years, until the BLS injury and illness survey was revised to collect additional data from employers, the BLS collected data identical to the data collected by OSHA in 1996. BLS estimated that completion of its pre-1992 surveys required one half hour of time. A 1992 BLS test conducted on 92 respondents completing only part 1 of the BLS survey form (equivalent to the OSHA form) measured the average respondents completion time at 30.55 minutes.

The occupational injury and illness information from the OSHA records is required by regulation and is easily transferred to the OSHA survey form. The information on employment and hours worked by employees is generally easy to obtain from payroll systems for employees who are paid on an hourly basis, and can be estimated for salaried employees. The survey forms used by OSHA provide the employer with instructions and worksheets to make the calculations as easy as possible. In many cases, the employment and hours worked data are already being reported to unemployment insurance and workers' compensation agencies and can easily be transferred to the OSHA survey form.

As discussed above, OSHA has concluded that promulgation of this regulation, in and of itself, imposes few if any economic costs on potentially affected firms. Individual data collections conducted under this regulation will be subject to OMB review under the procedures specified by the Paperwork Reduction Act of 1995. Employers will thus have an opportunity to comment on any burdens imposed by such data collections when they are carried out in the future.

OSHA has determined that this rule is a significant regulatory action as defined by 3(f)(4) of E.O. 12866. This provision of the E.O. covers a regulatory action that is likely to result in a rule that may:

(4) Raise novel legal or policy issues arising out of legal mandates, the President's priorities, or the principles set forth in the Executive Order.

V. Regulatory Flexibility Act

OSHA is required by the Regulatory Flexibility Act, as amended in 1996, to assess whether its regulations will have a significant impact on a substantial number of small entities. As explained in the Economic Analysis section of this preamble, above, this regulation (section 1904.17, Annual OSHA Injury and Illness Survey of Ten or More Employers) imposes few, if any costs on affected employers, although future data collection efforts conducted under this regulation may impose minimal cost and paperwork burdens on those employers affected by a given data collection effort. OSHA will carefully assess the impacts of individual data collections on employers, including small employers, at the time such efforts are initiated. Pursuant to the Regulatory Flexibility Act, OSHA thus certifies that section 1904.17 will not have a significant impact on a substantial number of small entities.

VI. Environmental Impacts

The provisions of this final regulation have been reviewed in accordance with the requirements of the National Environmental Policy Act (NEPA) of 1969 (42 U.S.C. 432, et seq.), the Council on Environmental Quality (CEQ) NEPA regulations [40 CFR part 1500], and OSHA's DOL Procedures [29 CFR part 11]. As a result of this review, OSHA has determined that this final rule will have no significant effect on air, water, or soil quality, plant or animal life, use of land, or other aspects of the environment.

VII. Federalism

This rule has been reviewed in accordance with Executive Order 12612 (52 FR 41685), regarding Federalism. Because this rulemaking action involves a "regulation" issued under § 8 of the OSH Act, and not a "standard" issued under § 6 of the Act, the rule does not preempt State law, see 29 U.S.C. 667 (a).

VIII. State Plans

The 25 States and territories with their own OSHA approved occupational safety and health plans are: Alaska, Arizona, California, Hawaii, Indiana, Iowa, Kentucky, Maryland, Michigan, Minnesota, Nevada, New Mexico, North Carolina, Oregon, Puerto Rico, South Carolina, Tennessee, Utah, Vermont, Virginia, Virgin Islands, Washington, and Wyoming; Connecticut and New York have state plans covering state and local Government employees only.

Section 18(c)(7) of the OSH Act requires employers in state plan states to "make reports to the Secretary in the same manner and to the same extent as if the plan were not in effect." Today's amendment to 29 CFR part 1904 relates to periodic data surveys which federal OSHA will conduct in all states, including those which administer approved state plans; accordingly, states with state plans are not required to adopt a comparable regulation. In state plan states, the data collected by the federal OSHA survey will be shared with the states for use in administering their plans, and also provide relevant information for OSHA's use in monitoring the state plan as required by section 18(f). Because OSHA's nationwide data survey is not an issue currently addressed by any of the state plans, OSHA's authority to implement the survey is not affected either by operational agreements with state plan states or by the granting of final approval under section 18(e). OSHA's authority under the Act, to take appropriate enforcement action when necessary to compel responses to the survey and to assure the accuracy of the data submitted by employers, will be exercised in consultation with the state in state plan states. The states may also exercise such authority under state law or regulation.

IX. Paperwork Reduction Act of 1995

This final regulation contains information collection requirements. As required by the Paperwork Reduction Act of 1995, the U.S. Department of Labor has submitted a copy of these sections to OMB for its review. (44 U.S.C. 3501 et seq., and 5 CFR part 1320.

Separately, the Department of Labor has received renewed approval for the Annual Survey Form under the Paperwork Reduction Act (OMB number 1218-0209)

List of Subjects in 29 CFR Part 1904

Reports by employers, occupational injuries and illnesses, Occupational Safety and Health, Occupational Safety and Health Administration, Recordkeeping, Reporting.


This document was prepared under the direction of Greg Watchman, Acting Assistant Secretary of Labor for Occupational Safety and Health, U.S. Department of Labor, 200 Constitution Avenue, NW., Washington, DC 20210.

Accordingly, pursuant to sections 8 and 24 of the Occupational Safety and Health Act of 1970 (29 U.S.C. 657, 673), Secretary of Labor's Order No. 1-90 (55 FR 9033), and 5 U.S.C. 553, 29 CFR part 1904 is hereby amended by adding 1904.17 as set forth below.

Signed in Washington, D.C., this 7th day of 1997.

Greg Watchman,
Acting Assistant Secretary of Labor.

PART 1904 -- [AMENDED]

1. The authority citation for Part 1904 is revised to read as follows:

Authority: Secs. 8, 24, Occupational Safety and Health Act of 1970 (29 U.S.C. 657, 673), Secretary of Labor's Order No. 12-71 (36 FR 8754), 8-76 (41 FR 25059), 9-83 (48 FR 35736), 1-90 (55 FR 9033) or 6-96 (62 FR 111), as applicable.

Section 1904.7, 1904.8 and 1904.17 are also issued under 5 U.S.C. 553.

2. Section 1904.17 immediately following 1904.16 is added to read as follows:

1904.17 Annual OSHA Injury and Illness Survey of Ten or More Employers.

(a) Each employer shall, upon receipt of OSHA's Annual Survey Form, report to OSHA or OSHA's designee the number of workers it employed and number of hours worked by its employees for periods designated in the Survey Form and such information as OSHA may request from records required to be created and maintained pursuant to 29 CFR part 1904.

(b) Survey reports shall be sent to OSHA by mail or other means described in the Survey Form within 30 calendar days, or the time stated in the Survey Form, whichever is longer.

(c) Employers exempted from keeping injury and illness records under Secs. 1904.15 and 1904.16 shall maintain injury and illness records required by Secs. 1904.2 and 1904.4, and make Survey Reports pursuant to this Section, upon being notified in writing by OSHA, in advance of the year for which injury and illness records will be required, that the employer has been selected to participate in an information collection.

(d) Nothing in any State plan approved under Section 18 of the Act shall affect the duties of employers to comply with this section.

(e) Nothing in this section shall affect OSHA's exercise of its statutory authorities to investigate conditions related to occupational safety and health.

[FR Doc. 97-3495 Filed 2-10-97; 8:45 am)