United States Government Accountability Office Report to Congressional Requesters WORKPLACE January 2021 SAFETY AND HEALTH Actions Needed to Improve Reporting of Summary Injury and Illness Data GAO-21-122 January 2021 WORKPLACE SAFETY AND HEALTH Actions Needed to Improve Reporting of Summary Injury and Illness Data Highlights of GAO-21-122, a report to congressional requesters Why GAO Did This Study What GAO Found In 2018, about 3.5 million workers GAO’s analysis of Occupational Safety and Health Administration (OSHA) data suffered job-related injuries, and showed that the number of recordkeeping violations OSHA cited fluctuated over illnesses and 5,250 died on the job, 15 years (see fig.). An April 2012 federal court decision (that effectively limited according to Bureau of Labor Statistics the time period for citing these violations) and a January 2015 expansion of data. Employers are required to record OSHA’s rule for reporting severe injuries and illnesses coincided with, and were work-related injuries and illnesses, cited by, OSHA staff as key factors explaining these fluctuations. promptly report severe injury and illness incidents to OSHA, and certain Number Recordkeeping Violations OSHA Cited by Fiscal Year employers are required to report summary injury and illness data electronically on an annual basis. GAO was asked to review how OSHA addresses recordkeeping violations, and implements its rule for reporting summary data. This report examines: (1) how and why recordkeeping violations changed from fiscal years 2005 through 2019 and (2) the extent to which employers report Employers did not report any summary injury and illness data on more than one- summary injury and illness data and half of their establishments that GAO estimated met the reporting requirements OSHA has taken steps to ensure (see table). compliance with this requirement. Estimated Compliance with Summary Injury and Illness Reporting Requirement GAO analyzed 15 years of OSHA Establishments whose employers recordkeeping violation data and Estimated establishments that met submitted summary injury and compared OSHA and Census data to Calendar summary injury and illness reporting illness data estimate how many employers year requirements Number Percent complied with summary reporting 2016 451,000 159,000 35% requirements. GAO also reviewed 2017 454,000 189,000 42% agency procedures and relevant 2018 459,000 212,000 46% federal laws and regulations and Source: GAO analysis of U.S. Census Bureau County Business Patterns data and Occupational Safety and Health Administration interviewed OSHA headquarters (OSHA) summary (300A) injury and illness data. Establishments in all 50 states and the District of Columbia reported these data. Data officials and staff at seven OSHA area rounded to the nearest thousand. | GAO-21-122 offices, selected for geographic OSHA has limited procedures for encouraging compliance with this reporting dispersion and varying amounts of requirement and for penalizing non-compliance. For example, OSHA officials told recordkeeping violations. GAO that they identified nearly 220,000 employers in 2019 who may not have reported their data and mailed reminder postcards to about 27,000 of them. What GAO Recommends OSHA also cited 255 employers for failure to report their data from mid- GAO recommends OSHA evaluate December 2017 through September 2019 after OSHA conducted on-site procedures for ensuring reporting of inspections. OSHA uses the summary injury and illness data to target high-risk summary data and develop a plan to establishments for certain comprehensive inspections. Because OSHA has not remediate deficiencies. OSHA evaluated its procedures, it does not know the extent to which its efforts may be generally concurred with our improving injury and illness reporting or what other efforts it should undertake. recommendation. Absent more complete information, OSHA is at risk for not achieving its objective of targeting inspections to establishments with the highest injury and illness rates. View GAO-21-122. For more information, contact Thomas Costa at (202) 512-4769 or costat@gao.gov. United States Government Accountability Office Contents Letter 1 Background 4 A 2012 Court Decision and 2015 Regulatory Revision Coincided with Changes in the Number and Type of Recordkeeping Violations OSHA Cited 10 Many Employers Do Not Report Required Summary Injury and Illness Data and OSHA’s Procedures to Ensure Compliance Are Limited 18 Conclusions 27 Recommendation for Executive Action 28 Agency Comments and Our Evaluation 28 Appendix I Objectives, Scope, and Methodology 32 Appendix II Comments from the Department of Labor 38 Appendix III GAO Contact and Staff Acknowledgments 40 Tables Table 1: Estimated Compliance with the Annual Summary Injury and Illness (300A) Reporting Requirement, Calendar Years 2016-2018 19 Table 2: Number of Site-Specific Targeting (SST) Inspections Conducted by Region in Fiscal Year 2019 and Number of Establishments Eligible for SST Inspections 24 Figures Figure 1: Number of Recordkeeping Violations OSHA Cited by Fiscal Year 10 Figure 2: Number of Recordkeeping Violations Cited Before and After the D.C. Circuit Court’s April 2012 Volks Decision 14 Figure 3: Number of OSHA Inspections with at Least One Cited Recordkeeping Violation 15 Figure 4: Number of Inspections with at Least One Cited Recordkeeping Violation, Excluding Severe Injury and Illness Reporting Violations 16 Page i GAO-21-122 Workplace Safety and Health Recordkeeping Figure 5: Violations OSHA Cited of Its Severe Injury and Illness Reporting Rule 17 Abbreviations BLS Bureau of Labor Statistics CBP U.S. Census Bureau County Business Patterns DOL U.S. Department of Labor IMIS Integrated Management Information System NAICS North American Industry Classification System OSHA Occupational Safety and Health Administration OSH Act Occupational Safety and Health Act of 1970 SST Site-Specific Targeting This is a work of the U.S. government and is not subject to copyright protection in the United States. The published product may be reproduced and distributed in its entirety without further permission from GAO. However, because this work may contain copyrighted images or other material, permission from the copyright holder may be necessary if you wish to reproduce this material separately. Page ii GAO-21-122 Workplace Safety and Health Recordkeeping Letter 441 G St. N.W. Washington, DC 20548 January 27, 2021 Congressional Requesters Approximately 3.5 million workers suffered job related injuries and illnesses in 2018 and an additional 5,250 workers died from their injuries, according to the Bureau of Labor Statistics (BLS). 1 For employers and workers, accurate recording of injuries and illnesses is essential for identifying and addressing the root causes of unsafe workplace conditions, and establishing effective workplace safety systems. For the Occupational Safety and Health Administration (OSHA), the federal agency charged with assuring safe and healthful conditions for the nation’s workforce, accurate employer documentation and reporting of workplace injuries and illnesses is also essential for targeting its inspections to the most dangerous worksites and for identifying high- hazard industries where additional oversight may be beneficial. OSHA inspects workplace establishments to ensure that employers are complying with the Occupational Safety and Health Act of 1970, as amended (OSH Act) and applicable workplace safety and health regulations. 2 OSHA generally requires employers to record and maintain information on work-related injuries, illnesses, and fatalities that require more than first aid and to report severe injuries, illnesses, and fatalities promptly to OSHA. 3 Since 2017, OSHA has required that certain employers also report a summary of their establishments’ injury and illness data electronically to OSHA each year (referred to as its electronic recordkeeping rule). 4 OSHA uses these reported data to select establishments for Site-Specific Targeting (SST) inspections, which are comprehensive inspections that are to examine all potentially hazardous areas of an establishment. OSHA reinstated the SST inspection program 1Nonfatal occupational injuries and illnesses by industry and case types, 2018 at https://www.bls.gov/iif/oshwc/osh/os/summ2_00_2018.htm, see Table 2, and Fatal occupational injuries to private sector wage and salaried workers, government workers, and self-employed workers by industry, All United States, 2018 at https://www.bls.gov/iif/oshwc/cfoi/cftb0324.htm, see Table A-3. 2Pub. L. No. 91-596, 84 Stat. 1590 (codified as amended at 29 U.S.C. § 651 et seq.) 3See generally 29 C.F.R. pt.1904. 429 C.F.R. § 1904.41. Page 1 GAO-21-122 Workplace Safety and Health Recordkeeping in fiscal year 2019 and targets most of these inspections towards establishments with higher injury and illness rates. 5 You asked us to review how OSHA identifies and addresses recordkeeping violations and how OSHA is implementing its recordkeeping rule that requires annual electronic reporting of workplace injuries and illnesses. In this report we examined: (1) how, if at all, recordkeeping violations cited by OSHA have changed between fiscal years 2005 and 2019 and the possible reasons for any changes; and (2) the extent to which employers electronically report required summary injury and illness data and OSHA has taken steps to ensure compliance with this requirement. To address both objectives, we reviewed relevant federal laws and regulations. We also reviewed and assessed OSHA’s actions against its internal guidance and directives, standards for internal control in the federal government, and our prior work on reexamining regulations. 6 The internal control standards used were: (1) the information and communication standard, along with the underlying principle that an agency should use quality information that is complete, current, and accurate to achieve program objectives (principle 13) and (2) the monitoring standard, along with the underlying principle that an agency should evaluate and remediate limitations or shortcomings with its procedures that are brought to its attention (principle 17). We interviewed OSHA headquarters officials about how OSHA identifies and addresses recordkeeping violations; how the agency implemented new electronic reporting requirements; and how it used these data to implement its newly reinstated SST inspection program. We also met separately with OSHA management and compliance officers at a nongeneralizable sample of seven out of 72 OSHA area offices about how they identify and address recordkeeping violations discovered during workplace inspections and 5OSHA also conducted SST inspections from 1997 through 2014. After employers began to electronically report their summary data to OSHA in 2017, OSHA reinstated the SST program using these data to select establishments for inspection. 6See GAO, Standards for Internal Control in the Federal Government, GAO-14-704G (Washington, D.C.: September 2014). Also see GAO, Reexamining Regulations: Agencies Often Made Regulatory Changes, but Could Strengthen Linkages to Performance Goals, GAO-14-268 (Washington, D.C.: Apr. 11, 2014). In GAO-14-268, we stated that reexamining regulations can help agencies evaluate the extent to which regulations are working in practice and that careful oversight of regulations is important because, without it, they might prove to be less effective in achieving intended goals. Page 2 GAO-21-122 Workplace Safety and Health Recordkeeping how they conduct SST inspections. 7 We used a range of criteria to select these offices, including geographic dispersion and variation in the amount of recordkeeping violations cited by the area offices. For the first objective, we analyzed data on the number and type of recordkeeping violations that OSHA cited from fiscal year 2005 through 2019, which was the most current available at the time of our review. To assess the reliability of these data, we reviewed documentation, interviewed knowledgeable OSHA officials, and conducted electronic data testing on specific data elements. We determined that the data were sufficiently reliable for our purposes. We then identified possible reasons for changes in OSHA-cited recordkeeping violations by asking OSHA area office compliance officers and managers at the seven area offices about: (1) how they identify and address these violations when conducting workplace inspections and (2) in their professional judgement if and how the type and number of violations cited have changed over time and the reasons for these changes. For the second objective, we estimated the extent to which employers electronically reported summary injury and illness data on their establishments annually to OSHA as required. To do this, we used U.S. Census Bureau County Business Patterns data to estimate the total number of establishments that met OSHA’s criteria for employers to electronically report required summary injury and illness data for calendar years 2016 through 2018. 8 We then compared the estimated number of establishments that met the reporting criteria against the estimated number of establishments for which employers had submitted summary injury and illness data to OSHA and calculated the proportion that had done so. 9 We also analyzed OSHA data on SST inspections that OSHA area offices conducted in fiscal year 2019, the only data available at the 7These 72 offices inspected workplace establishments in the private sector and identified recordkeeping violations from fiscal year 2017 through 2019. 8Specifically,we used the Census data on employer size and industry to estimate the number of employers that met the reporting criteria in OSHA’s regulation, 29 C.F.R. § 1904.41. Due in part to differences between the regulatory criteria and the information contained in the Census data, legal compliance cannot be determined from these estimates alone. See appendix I for more information on our methodology. 9We analyzed about 73 percent of the roughly 763,000 summary injury and illness records employers submitted to OSHA. Of the exclusions, about 56 percent were excluded because the establishment did not meet the criteria for reporting the data, according to our analysis of the data. The remaining 44 percent were excluded for technical reasons, such as the record did not have an industry code or had an industry code that was not included in the Census data. See appendix 1 for more information on our methodology. Page 3 GAO-21-122 Workplace Safety and Health Recordkeeping time of our analysis. To assess the reliability of data analyzed for objective 2, we reviewed documentation, interviewed knowledgeable OSHA officials about their data, and conducted electronic data testing on specific data elements. We determined that all data were sufficiently reliable for our purposes. We conducted this performance audit from May 2019 to January 2021 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. Background Workplace Safety and OSHA, within the U.S. Department of Labor (DOL), is responsible for Health Oversight carrying out the OSH Act to protect the safety and health of the nation’s workforce. OSHA sets and enforces workplace safety and health standards for private-sector workplaces in 29 states and the District of Columbia. In the remaining 21 states, state occupational safety and health agencies set and enforce their own workplace safety and health standards for these workplaces under OSHA-approved state plans. 10 The OSH Act requires that state standards and their enforcement be “at least as effective” as the federal standards. OSHA sets and enforces workplace safety and health standards to address hazards that can result in injury, illness, or death. 11 For example, OSHA issued regulations establishing standards for workplace practices to reduce the risk of employee falls and for employees working with hazardous chemicals. 10See 29 U.S.C. § 667; 29 C.F.R. pt. 1952. The 21 state plans also cover state and local government entities in their states, as required by the OSH Act. In five of the 29 states where OSHA is responsible for enforcement in the private sector, OSHA approved state plans cover state and local government. Federal agencies are generally responsible for maintaining their own occupational safety and health programs, consistent with OSHA’s standards. 29 U.S.C. § 668, 29 C.F.R. pt. 1960, Exec. Order No. 12196, 45 Fed. Reg. 12,769 (Feb. 26, 1980). 11See generally 29 C.F.R. pt. 1910. In addition, OSHA has issued separate standards for certain industries, such as construction (29 C.F.R. pt. 1926) and agriculture (29 C.F.R. pt. 1928). Page 4 GAO-21-122 Workplace Safety and Health Recordkeeping OSHA Recordkeeping OSHA has also issued recordkeeping regulations that require employers Regulations to record certain injuries, illnesses, and deaths, and make these records available, when required, to employees and government agencies. 12 States that operate their own occupational safety and health programs under an OSHA-approved state plan must have recording and reporting requirements that are substantially identical to OSHA’s. 13 OSHA’s recordkeeping regulations require employers to record all work- related injuries and illnesses that result in death, loss of consciousness, days away from work, restricted work or job transfer, or medical treatment beyond first aid, among other criteria, and maintain these records for 5 years. 14 Employers with 10 or fewer employees and employers in certain industries are exempt from these requirements. 15 Employers are required to use the following forms (or an equivalent) to record this information: • OSHA Form 300: A log that employers are to use to list each workplace injury or illness that occurred at each of their establishments during the year; • OSHA Form 301: An incident report that employers are to use to describe in more detail each workplace injury and illness that they list on OSHA Form 300; and • OSHA Form 300A: A summary form that employers are to use at the end of the year to total up all injuries, illnesses, and fatalities that they listed on OSHA Form 300. This summary form must be filled out and 12See generally 29 C.F.R. pt. 1904. 1329 C.F.R. § 1904.37. These states must have the same requirements as OSHA for determining which injuries and illnesses are recordable and how they are recorded. Other state recording and reporting requirements (for example, industry exemptions, reporting of fatalities and hospitalizations, or record retention), may be more stringent than or supplemental to the federal requirements, subject to consultation with and approval by OSHA. 1429 C.F.R. §§ 1904.4, 1904.11, 1904.33. OSHA requires employers to record this information for each workplace establishment that the employer owns or manages. An establishment is a single physical location where business is conducted or where servicers or industrial operations are performed. 29 C.F.R. § 1904.46. If an employer has more than one establishment, the employer must record all injuries, illnesses, and fatalities separately for each establishment. 29 C.F.R. § 1904.30. Employers are required to record each injury or illness within 7 calendar days of receiving information that a recordable injury or illness has occurred. 29 C.F.R. § 1904.29(b)(3). 1529 C.F.R. §§ 1904.1-1904.2 and appendix A to subpart B of part 1904. Page 5 GAO-21-122 Workplace Safety and Health Recordkeeping retained—and posted in the workplace—even if no workplace injuries or illnesses occurred in that year. 16 OSHA’s regulations also require employers to report certain workplace injury and illness information to OSHA. Employers are required to report any fatalities to OSHA within 8 hours and to report any in-patient hospitalization, amputation, or loss of an eye within 24 hours (referred to as severe injury and illness reporting in this report). 17 In addition, on May 12, 2016, OSHA issued a rule that, among other things, required two groups of employers to electronically report their injury, illness, and fatality summary information to OSHA (referred to as the electronic recordkeeping rule in this report). 18 • The first group of employers were those with establishments in certain industries, such as manufacturing and nursing care facilities, that had between 20 and 249 employees at any point during the previous calendar year. This group was required to electronically submit the total number of injuries, illnesses, and fatalities from their 300A Forms annually to OSHA. • The second group of employers were those with establishments in any industry that had 250 or more employees at any point during the previous calendar year and were required to maintain injury and illness records. This group was required to electronically submit information from all three forms annually to OSHA—the log of injuries and illnesses (OSHA Form 300), the incident reports of injuries and illnesses (OSHA Form 301), and the total annual number of injuries and illnesses (OSHA Form 300A). The requirement to annually submit 300A data became effective for both groups of employers on January 1, 2017, with an initial submission 16Although 300A data includes information on workplace injuries, illnesses, and fatalities, for ease of reference, this report will refer to it as either “300A injury and illness data” or simply “300A data.” 1729 C.F.R. § 1904.39. All employers, including those exempted from the recordkeeping requirements by reason of size or industry classification, must report these incidents to OSHA. Employers can report severe injury and illness incidents to OSHA by calling their nearest OSHA office, calling the OSHA hotline, or reporting the incident online. As discussed in more detail later in the report, these reporting requirements were modified in 2015. See Occupational Injury and Illness Recording and Reporting Requirements— NAICS Update and Reporting Revisions, 79 Fed. Reg. 56,130 (Sept. 18, 2014). 18Improve Tracking of Workplace Injuries and Illnesses, 81 Fed. Reg. 29,624 (May 12, 2016) (revising 29 C.F.R. § 1904.41). Page 6 GAO-21-122 Workplace Safety and Health Recordkeeping deadline of December 15, 2017. 19 However, the requirement that employers with establishments with 250 or more employees report data from OSHA Forms 300 and 301 never went into effect and OSHA revised the rule on January 25, 2019, to eliminate this requirement. 20 Rather, these employers are now required to electronically submit only the total number of injuries and illnesses from their 300A Form, like the smaller employers in certain industries. Although in the 2016 final electronic recordkeeping rule OSHA said that it intended to make this information publicly available on its website, in the 2019 final rule, the agency stated that the confidentiality of injury and illness records should be maintained, except for those persons with a legitimate need to know. 21 Covered employers are to submit the data to a secure OSHA website by accessing an OSHA portal called the Injury Tracking Application (ITA). Employers are required to submit their 300A data even if their establishment had zero workplace injuries or illnesses the previous year. OSHA Inspections OSHA estimates that about 8 million workplace establishments are under its jurisdiction. In fiscal year 2019, OSHA inspected 33,393 of these establishments. OSHA inspections fall into two broad categories: unprogrammed and programmed. According to OSHA guidance on prioritizing inspections, unprogrammed inspections take precedence over programmed inspections. 22 19The initial deadline of July 1, 2017 (for calendar year 2016 data) was extended to December 15, 2017. The submission deadline for calendar year 2017 data was July 1, 2018. Beginning in 2019, the submission deadline is March 2 of each year. Although the timing of implementation by state plan states varied, OSHA announced that all employers in those states would be expected to submit their data for calendar year 2017 by July 1, 2018, even if the state plan had not completed adoption of a state rule. However, OSHA stated that there would be no retroactive requirement for such employers to submit data for calendar year 2016. 20OSHA announced that it would not enforce the requirement to submit OSHA Forms 300 and 301 while it undertook rulemaking to revise the 2016 final rule. The revised final rule was issued January 25, 2019 and became effective February 25, 2019. See Tracking of Workplace Injuries and Illnesses, 84 Fed. Reg. 380 (Jan. 25, 2019) (revising 29 C.F.R. § 1904.41). Legal challenges to other portions of the original 2016 final rule, as well as the 2019 final rule, are currently pending in federal court. 21In the 2016 final rule, OSHA stated that it had planned to make these data publicly available, after removing all personally identifiable employee information. In the 2016 rule, OSHA stated that public access to these data could help employers, employees, researchers and government agencies better identify and address workplace hazards. In the 2019 final rule, OSHA stated that it determined that avoiding risks to worker privacy outweighed the data's uncertain incremental benefits to enforcement. 22See OSHA’s Field Operations Manual, CPL 02-00-164 (effective Apr. 4, 2020). For the regulations governing OSHA’s inspection process, see 29 C.F.R. pt. 1903. Page 7 GAO-21-122 Workplace Safety and Health Recordkeeping • Unprogrammed inspections. OSHA conducts unprogrammed inspections if the agency learns about potentially hazardous working conditions at a specific worksite. These inspections tend to be in response to (1) an imminent danger, fatality, or serious accident occurring at an establishment or (2) OSHA receiving an employee complaint or outside referral about alleged hazardous working conditions at an establishment. In fiscal year 2019, 18,493 of the 33,393 inspections (55 percent) OSHA conducted were unprogrammed. • Programmed inspections. OSHA selects establishments for programmed inspections based on objective criteria, such as the establishment being part of a high-hazard industry for which OSHA has developed a special inspection program, known as an emphasis program. OSHA has nine national emphasis programs, including shipbreaking as well as trenching and excavation. OSHA’s SST inspections are part of the agency’s nationwide programmed inspection category. In addition, OSHA regional and area offices can develop their own inspection emphasis programs to highlight industries or hazards within their jurisdictions that they have identified as especially hazardous. In fiscal year 2019, 14,900 of the 33,393 inspections (45 percent) OSHA conducted were programmed inspections. OSHA predominantly uses electronically employer-reported 300A injury and illness data to determine which establishments to place on a list as eligible for an SST inspection. 23 According to OSHA’s most recent SST Directive, the SST program helps OSHA achieve its goal of ensuring that employers provide safe and healthful workplaces by directing enforcement resources to those establishments with the highest rates of injuries and illnesses. 24 SST inspections are comprehensive inspections that are to examine all potentially hazardous areas that may exist in a workplace establishment. According to OSHA officials and guidance on selecting establishments for SST inspection, OSHA’s Office of Statistical Analysis most often randomly selects establishments for SST inspections from the universe of 23According to OSHA officials, the agency also uses 300A data for outreach—contacting establishments with high injury and illness rates through its on-site consultation program. This program helps small- and medium-sized businesses identify and address hazardous conditions and is separate from OSHA enforcement. In addition, OSHA uses injury and illness data gathered from BLS in its emphasis inspection programs. 24See Directive Number: CPL 02-01-062; Effective Date: 12/14/2020; Subject: Site- Specific Targeting (SST). Page 8 GAO-21-122 Workplace Safety and Health Recordkeeping those establishments with high injury or illness rates. To determine non- compliance with recordkeeping requirements, however, OSHA also randomly selects some establishments for these inspections from a pool of employers who have not reported their 300A injury and illness data and a pool of employers who have reported lower than expected injury and illness rates when compared to BLS injury and illness data for the same type of industry. OSHA headquarters provides each area office with a list of all establishments eligible for SST inspections that are located within the area office’s jurisdiction. Each area office, in conjunction with its regional office, determines how many establishments on the list will receive an SST inspection in the coming year. To make these decisions, area and regional offices take into consideration their resources and the number of inspections area offices are expected to complete under all inspection programs. OSHA Citations and The OSH Act authorizes OSHA to issue citations to employers who fail to Employers’ Right to comply with the act or applicable OSHA regulations, including recordkeeping regulations. 25 These citations identify the specific Contest Citations requirement(s) that an employer violated, require the employer to correct the violation within a certain period of time (abatement date), and may also require the employer to pay a financial penalty. 26 The OSH Act requires that OSHA issue any citation within 6 months of a violation (referred to as the statute of limitations). Employers have a right to contest all or part of a citation if they disagree with the citation, penalty, or abatement date. 27 To do so, employers submit a written “Notice of Intent to Contest” (Notice) to the area office that issued the citation within 15 days. If an employer does not contest the citation within that timeframe, the citation becomes final. If an employer files a Notice, the case is sent to the Occupational Safety and Health Review Commission—an independent agency—which hears the case and issues a decision. The employer or OSHA can appeal that decision in federal court. 28 2529 U.S.C. § 658. 2629 U.S.C. § 666, 29 C.F.R. § 1903.15. The OSH Act also authorizes criminal penalties in certain cases. 2729 U.S.C. § 659. Employers may also seek an informal conference to resolve these issues. 29 C.F.R. § 1903.20. 2829 U.S.C. § 660. Page 9 GAO-21-122 Workplace Safety and Health Recordkeeping Our analysis of 15 years of OSHA data showed that OSHA cited an A 2012 Court increasing number of recordkeeping violations from fiscal year 2005 Decision and 2015 through 2011, a decreasing number from fiscal year 2012 through 2014, and then a generally increasing number from fiscal year 2015 through Regulatory Revision 2019 (see fig. 1). Coincided with Changes in the Figure 1: Number of Recordkeeping Violations OSHA Cited by Fiscal Year Number and Type of Recordkeeping Violations OSHA Cited OSHA managers and compliance officers we spoke with in OSHA area offices referenced two key factors that may explain some of the reasons for the trends in the number of recordkeeping violations cited over the 15 year period. 29 Specifically, the two factors cited by OSHA compliance officers and managers were: • An April 6, 2012, federal appellate court decision, which required OSHA to change its enforcement policy for recordkeeping violations, 29We held separate discussions with managers and compliance officers in seven OSHA area offices. Managers in all seven offices and compliance officers in six of the seven offices identified two factors as affecting the trend in recordkeeping violations over time—a 2012 federal court decision and a 2015 revision to OSHA’s recordkeeping regulations. Compliance officers in the seventh office only identified the 2012 court decision. Page 10 GAO-21-122 Workplace Safety and Health Recordkeeping effectively shortening the time period for OSHA to cite employers who violate OSHA’s recordkeeping rules. (This report refers to this decision as the Volks decision.) 30 • A revision to OSHA’s recordkeeping regulations that went into effect January 1, 2015, which required employers to report additional work- related severe injuries and illnesses to OSHA shortly after they occur. 31 (This report refers to this rule as the severe injury and illness reporting rule). The Volks decision, issued by the U.S. Court of Appeals for the District of Columbia Circuit on April 6, 2012, required OSHA to change its interpretation of how to apply the OSH Act’s 6-month statute of limitations when issuing recordkeeping citations. 32 Prior to this decision, OSHA considered a recordkeeping violation—such as a failure to record a work- related injury on OSHA Form 300—to constitute a “continuing violation” for every day the injury remained unrecorded. Therefore, under OSHA’s previous interpretation, for any such violation, the 6-month statute of limitations would begin at the end of the 5-year period for which the employer was required to retain OSHA recordkeeping forms. As a result, prior to the Volks decision, OSHA issued citations to employers for recordkeeping violations for up to about 5½ years from the date that the initial violation occurred. The court, in Volks, however, disagreed with this interpretation, holding that a recordkeeping violation occurs—and the 30AKM LLC d/b/a Volks Constructors v. Sec’y of Labor, 675 F.3d 752 (D.C. Cir. 2012). An employer—Volks Constructors—contested OSHA citations it received in November 2006 for failure to properly record and maintain on OSHA Forms 300, 301, and 300A all work- related injuries and illnesses that occurred between 2002 and early 2006. After the Occupational Safety and Health Review Commission upheld the citations in 2011, the employer appealed to the U.S. Court of Appeals for the D.C. Circuit. In an April 6, 2015, decision, the court overturned the citations as untimely under the OSH Act’s statute of limitations. 31Occupational Injury and Illness Recording and Reporting Requirements—NAICS Update and Reporting Revisions, 79 Fed. Reg. 56,130 (Sept. 18, 2014) (revising 29 C.F.R. § 1904.39). This regulatory change became effective for all employers under OSHA’s jurisdiction January 1, 2015. The change became effective for employers in state plan states at different times following January 1, 2015, as states adopted their own reporting rules that were identical to or at least as effective as OSHA’s. 32Under the OSH Act’s statute of limitations, OSHA may not issue a citation to an employer for violating the act or any OSHA regulations (including the recordkeeping regulations) after the expiration of 6 months following the occurrence of the violation. 29 U.S.C. § 658(c). According to OSHA officials, the agency implemented the change to its recordkeeping enforcement policy nationwide immediately after the decision. Officials said that state-plan states are not bound by the decision. Page 11 GAO-21-122 Workplace Safety and Health Recordkeeping statute of limitations is triggered—on the last day that an employer has to record an injury or illness on an OSHA recordkeeping form (which, under OSHA regulations, is 7 days after receiving information that a recordable injury or illness has occurred). 33 As previously noted, the revisions to the severe injury and illness reporting rule required employers to report any work-related hospitalization, amputation, or loss of an eye to OSHA within 24 hours. The previous rule required employers to report hospitalizations of three or more employees within 8 hours. Under the previous rule, amputations and losses of an eye were required to be recorded, but were not required to be reported to OSHA if they did not result in three or more hospitalizations. The revised rule made no change to the existing requirement that employers report all work-related fatalities to OSHA within 8 hours. Managers and compliance officers in the seven area offices we met with said that the Volks decision has limited their ability to cite employers for recordkeeping violations. Specifically: • Managers and compliance officers in the seven area offices stated that it was possible to cite recordkeeping violations more frequently prior to the Volks decision because OSHA’s interpretation of when the statute of limitations began resulted in more violations falling within a period of time in which it was permissible to issue citations. Compliance officers in three of these seven offices further explained that the Volks decision clarified that the statute of limitations for recordkeeping violations is the same as the statute of limitations for a violation to any other OSHA standard. The compliance officers noted that, if an inspection uncovers any violation that occurred more than 6 months ago—recordkeeping or otherwise—then that violation cannot be cited. • Managers and compliance officers in four area offices explained that issuing citations within required timeframes can be difficult even if an inspection is within the 6-month window from when a violation occurred. For example, one manager stated that, if during an 3329 C.F.R. § 1904.29(b)(3). In response to the Volks decision, in December 2016, OSHA issued a final rule making recordkeeping a continuing obligation for the duration of the 5- year retention period. Clarification of Employer's Continuing Obligation To Make and Maintain an Accurate Record of Each Recordable Injury and Illness, 81 Fed. Reg. 91,792 (Dec. 19, 2016). Although this rule had an effective date of January 18, 2017, it was disapproved by Congress under the Congressional Review Act on April 3, 2017. Pub. L. No. 115-21 (2017). According to OSHA officials, the rule was never enforced. Page 12 GAO-21-122 Workplace Safety and Health Recordkeeping inspection, they find an unreported injury that occurred 4 months ago and it took 3 months to conduct the inspection and issue a citation, it would not be possible to cite the employer for the unreported injury because the 6 months statute of limitations would have lapsed. • Managers and compliance officers in three area offices also explained that it is not always possible to issue citations to employers who violate recordkeeping requirements that have annual deadlines. Rather, they said that issuing citations for these types of violations is dependent on the timing of an inspection. For example, from February 1 through April 30 of each year, employers are required to post their OSHA forms that summarize their establishment’s injuries and illnesses from the previous year (i.e., form 300A). 34 Compliance officers said they may issue citations for violating this requirement only if they are conducting the inspection between February 2 (the day after the start of the posting requirement) and August 1 (6 months after the start of the requirement). Our analysis of 15 years of OSHA data on recordkeeping violations (fiscal years 2005 through 2019) shows that a decrease in the number of OSHA- cited recordkeeping violations coincides with the Volks decision in 2012. Figure 2 shows that OSHA cited employers for a total of 35,751 recordkeeping violations during fiscal years 2005 through 2019. Sixty-five percent of these violations occurred before the Volks decision and 35 percent occurred after the Volks decision, despite a similar length of time before and after the decision. 35 3429 C.F.R. § 1904.32. 35In order to conduct our analysis using bi-annual data in fiscal year 2012, we placed 5 days (April 1st through 5th of 2012) in the “after Volks” period. These days were prior to the April 6, 2012, Volks decision. Page 13 GAO-21-122 Workplace Safety and Health Recordkeeping Figure 2: Number of Recordkeeping Violations Cited Before and After the D.C. Circuit Court’s April 2012 Volks Decision Note: Data include recordkeeping violations cited by federal OSHA area offices only and exclude any violations cited by state occupational safety and health agencies. In this figure, the “Volks decision” refers to AKM LLC d/b/a Volks Constructors v. Sec’y of Labor, 675 F.3d 752 (D.C. Cir. 2012). This figure includes violations cited between April 1 and April 5, 2012, in the “After” category; however, such violations would have been issued prior to the Volks decision, which was issued on April 6, 2012. OSHA may cite employers for more than one recordkeeping violation as the result of an inspection. Our analysis of OSHA’s recordkeeping data over the 15-year time period showed that it was more likely for inspections with more than one recordkeeping violation to occur prior to the Volks decision than after the Volks decision. 36 To further explore this trend, we examined how many inspections per fiscal year found at least one recordkeeping violation. We found that, inspections that identified at least one recordkeeping violation: (1) generally increased between fiscal years 2005 and 2011, with a high of about 2,560 inspections in fiscal year 2010 and (2) declined by nearly 37 percent from fiscal year 2011 to fiscal year 2012. This decline continued, reaching a low of 971 inspections in 36Over the 7½ year period prior to the Volks decision, 454 inspections identified 5 or more recordkeeping violations per inspection; whereas over the 7½ year period after the decision, 29 inspections identified 5 or more violations per inspection. Page 14 GAO-21-122 Workplace Safety and Health Recordkeeping fiscal year 2014 (see fig. 3). This decline coincided with the Volks decision in April of fiscal year 2012. Figure 3: Number of OSHA Inspections with at Least One Cited Recordkeeping Violation Note: Data include recordkeeping violations cited by federal OSHA area offices only and exclude any violations cited by state occupational safety and health agencies. Figure 3 also shows that the number of inspections finding at least one recordkeeping violation increased by 52 percent from fiscal year 2014 to fiscal year 2015, from 971 to 1,480. This 52 percent increase coincided with the implementation of the revisions to the severe injury and illness reporting rule on January 1, 2015, and may reflect an increase in OSHA citations as a result of these revisions, which required employers to report additional work-related injuries and illnesses to OSHA. 37 As figure 4 shows, after excluding severe injury and illness reporting violations, inspections with at least one recordkeeping violation decreased substantially after 2011. 37Occupational Injury and Illness Recording and Reporting Requirements—NAICS Update and Reporting Revisions, 79 Fed. Reg. 56,130 (Sept. 18, 2014) (revising 29 C.F.R. § 1904.39). Page 15 GAO-21-122 Workplace Safety and Health Recordkeeping Figure 4: Number of Inspections with at Least One Cited Recordkeeping Violation, Excluding Severe Injury and Illness Reporting Violations Note: Data are for inspections conducted by federal OSHA area offices only and exclude any violations cited by state occupational safety and health agencies. Severe injury and illness reporting violations refer to violations of 29 C.F.R. § 1904.39. This rule was revised, effective January 1, 2015, to require employers to report additional severe injuries and illnesses to OSHA. Our analysis of data on violations of OSHA’s severe injury and illness reporting rule shows an almost 400 percent increase in these violations from fiscal years 2014 to 2015 (see fig. 5). This increase coincided with the implementation of the revisions to this rule on January 1, 2015, requiring employers to report work-related injuries or illnesses resulting in in-patient hospitalization, amputation, or loss of an eye within 24 hours. Prior to this change, employers were only required to report hospitalizations of three or more employees and were not required to report amputations or loss of an eye if those injuries did not result in at least three employee hospitalizations. This increase also coincided with the increase from fiscal year 2014 to fiscal year 2015 in inspections with at least one recordkeeping violation (see fig. 3). Page 16 GAO-21-122 Workplace Safety and Health Recordkeeping Figure 5: Violations OSHA Cited of Its Severe Injury and Illness Reporting Rule Note: Data are for severe injury and illness violations cited by federal OSHA area offices only and exclude any violations cited by state occupational safety and health agencies. The severe injury and illness reporting rule refers to 29 C.F.R. § 1904.39, which was revised effective January 1, 2015. Prior to this revision, employers were only required to report work-related incidents resulting in the in- patient hospitalization of three or more employees. After the revision, employers were required to report any work-related hospitalization, amputation, or loss of an eye. Managers and compliance officers told us that, because their focus is safeguarding employees from dangers and hazards in the work environment, they prioritize inspecting establishments where severe injuries or illnesses may have occurred. 38 Managers we spoke with in four offices and compliance officers we spoke with in three also said that they are more likely to issue citations for an employer’s failure to report a severe injury or illness than they are when employers violate other recordkeeping rules that they view as less serious, such as when an 38Managers we spoke with in seven OSHA area offices and compliance officers in six of these offices mentioned that they prioritize inspecting establishments where severe injury and illnesses may have occurred. Page 17 GAO-21-122 Workplace Safety and Health Recordkeeping employer inaccurately documents some aspect of an injury or illness on an OSHA form or neglects to sign the form. 39 Many Employers Do Not Report Required Summary Injury and Illness Data and OSHA Has Taken Limited Steps to Ensure Compliance Employers Did Not Report We estimated that employers did not report any 300A injury and illness Required Summary Injury data on more than one-half of their establishments for which they were required to do so for calendar years 2016 through 2018. As a result, and Illness Data on More OSHA may not know which establishments have the highest injury and than 50 Percent of illness rates. OSHA uses the reported summaries to identify Establishments establishments for SST inspection. However, partial reporting does not provide a complete picture of where injuries and illnesses are occurring and, by extension, may not be an effective tool for targeting SST inspections to establishments with the highest injury and illness rates. We compared counts of establishments from the 300A injury and illness data that employers were required to report to OSHA against estimated counts of establishments that met the 300A reporting criteria derived from published Census Bureau data. 40 This comparison showed that employers that reported these data to OSHA did so for less than 50 39Examples of how compliance officers may learn of such incidents include receiving complaints, and reading newspaper articles. According to one area office, they also learn about incidents via social media or the city’s emergency response system. 40We used U.S. Census Bureau County Business Patterns data for 2016-2018 to estimate the number of establishments that met the 300A reporting criteria by filtering the data into the following two groups: (1) establishments with 20 to 249 employees as of the week of March 12 for each year in certain industries and (2) establishments with 250 or more employees as of the week of March 12 for each year, excluding those in certain industries that are not required to routinely maintain injury and illness records. Due in part to differences between the regulatory criteria and the information contained in the Census data, legal compliance cannot be determined from these estimates alone. See appendix I for more information on our methodology. Page 18 GAO-21-122 Workplace Safety and Health Recordkeeping percent of establishments estimated to meet the reporting criteria for years 2016 through 2018. This analysis also showed that estimated compliance with this requirement increased each year, with filings reaching approximately 46 percent of the estimated establishment total in 2018 (see table 1). OSHA officials acknowledged the low compliance rate for employer reporting of 300A data and told us that better compliance would improve their ability to target SST inspections to establishments with the highest injury and illness rates—an important objective of the SST program. They did not, however, describe any plans for increasing compliance. Instead, they stated that compliance is increasing each year and speculated that it will continue to increase with time. Table 1: Estimated Compliance with the Annual Summary Injury and Illness (300A) Reporting Requirement, Calendar Years 2016-2018 Estimated number of Establishments whose employers establishments that met submitted required 300A data Calendar the 300A data reporting year requirement Number Percent 2016 451,000 159,000 35% 2017 454,000 189,000 42% 2018 459,000 212,000 46% Source: GAO analysis of U.S. Census Bureau County Business Patterns data and Occupational Safety and Health Administration (OSHA) 300A injury and illness data employers reported to OSHA. | GAO-21-122. Notes: Data are rounded to the nearest thousand. Data from both sources include establishments located in all 50 states and the District of Columbia. At the time of our review, 2019 U.S. Census Bureau County Business Patterns data were not available. However, OSHA officials told us that they have received 2019 300A data on about 228,000 establishments as of early May 2020 from employers required to report it. We used U.S. Census Bureau County Business Patterns data on employer size and industry to estimate the number of employers subject to the 300A reporting requirement. The following groups of employers are required to annually submit summary information on their work-related injuries and illnesses (300A data): (1) establishments with 20 to 249 employees at any point during the year in certain industries, and (2) establishments with 250 or more employees at any point during the year whose employers are required to routinely maintain injury and illness records. 29 C.F.R. § 1901.41. Due in part to differences between the regulatory criteria and the information contained in the Census data, legal compliance cannot be determined from these estimates alone. We excluded establishments from our analysis of 300A data if the North American Classification System (NAICS) code for the establishment was one that is not included in Census’ County Business Patterns data. We also excluded establishments from our 300A data analysis when the filing lacked a NAICS code. See appendix I for further details on this analysis. Page 19 GAO-21-122 Workplace Safety and Health Recordkeeping OSHA Has Limited OSHA’s procedures to encourage employers who are required to comply Procedures to Encourage with the electronic reporting requirement of 300A data and the agency’s procedures to penalize employers who fail to do so are limited. OSHA’s Employers to Report outreach procedures to encourage electronic reporting are limited Required Summary Injury because they may not fully explain which employers are required to and Illness Data and Cite comply with this rule nor successfully encourage such employers to Employers Who Do Not actually submit their data. Further, OSHA’s procedures to penalize Report These Data as employers for non-compliance are limited because, despite acknowledging widespread non-compliance, OSHA has issued few Required citations. Encouraging compliance. OSHA encourages employers to comply with the 300A injury and illness reporting requirement by conducting two types of outreach—general outreach to all employers about who is required to submit these data and follow-up outreach to some employers who may not have submitted their data as required. • General outreach to all employers about who is required to report these data. OSHA publicizes the 300A reporting requirement in a variety of ways, including posting information on its website, issuing press releases, sending emails to employers that sign-up via OSHA’s website, and highlighting the reporting requirement in the agency’s electronic newsletter. OSHA’s webpage for reporting the 300A data explains: (1) which employers are required to electronically report their establishment’s 300A data; (2) what data they are required to report; and (3) how employers should access OSHA’s data portal to create an account and electronically report their data. 41 According to OSHA officials, limited compliance with the 300A reporting requirement may be because some employers: (1) do not know about the rule; (2) mistakenly believe that they do not meet the criteria to comply with the rule; (3) encountered technological problems when submitting the data; or (4) simply do not want to submit their data. 42 41See https://www.osha.gov/injuryreporting/index.html. 42OSHA officials told us that, although some employers encountered technical problems when they first began submitting their data using the Injury Tracking Application, these problems have largely been corrected. They believe this to be the case because the agency’s help line has been receiving far fewer calls from employers about technical difficulties when submitting their data than it did when the Injury Tracking Application was first put into place. Page 20 GAO-21-122 Workplace Safety and Health Recordkeeping In addition, according to OSHA officials, some employers mistakenly believe that they do meet the criteria to comply with the 300A reporting requirement when, in fact, they do not. OSHA officials said that approximately 25 percent of the 300A data OSHA received was submitted by employers for establishments that did not meet the reporting requirements. 43 According to officials, these employers submit the data even though the electronic portal informs employers when their establishments do not meet the 300A reporting criteria. OSHA removes these “out-of-scope” submissions when reviewing the data. 44 • Follow-up outreach to some employers who may not have submitted their 300A data as required. According to OSHA officials, they conducted a data match using Dun & Bradstreet establishment data to identify and follow-up with potential non-responders—that is, employers who may have been required to submit 300A data but who did not do so. 45 According to OSHA officials, OSHA sends postcards to some of these non-responding employers, requesting that they submit their data. For example, OSHA sent postcards to about 29,000 out of the nearly 224,000 employers that did not submit their 2018 300A data, according to OSHA’s analysis of Dun & Bradstreet data. 43The following groups of employers are required to annually submit summary information on their work-related injuries and illnesses (300A data): (1) establishments with 20 to 249 employees at any point during the previous calendar year in certain industries, and (2) establishments with 250 or more employees at any point during the previous calendar year whose employers are required to routinely maintain injury and illness records. 29 C.F.R. § 1904.41. 44OSHA officials said that removing “out-of-scope” submissions makes processing and managing the 300A data more complex and time consuming. 45OSHA contracted with Dun & Bradstreet to obtain data on certain workplace establishments under OSHA’s jurisdiction to administer its inspection programs, including its SST inspection program. At the time of our work, Dun & Bradstreet was the primary business credit reporting agency with over 70 million businesses registered in its database. The database contains data on establishments that are seeking a business credit rating. OSHA’s Dun & Bradstreet contract expired in June 2020. According to OSHA officials, OSHA has contracted with Infogroup Inc. to obtain establishment data for the period July 1, 2020, through June 30, 2025. According to OSHA officials, the agency obtained Dun & Bradstreet data on most, but not all, of the establishments in non- construction industries that are required to submit these data. Our analysis of Census County Business Patterns data for 2018 estimated that 394,258 non-construction establishments met the 300A reporting criteria in 2018. OSHA obtained 2018 Dun & Bradstreet data on 321,255 establishments, which is nearly 81 percent of our estimated establishments. According to OSHA officials, they did not obtain Dun & Bradstreet data on establishments in the construction industry because these establishments are not part of the SST inspection program. OSHA has a separate emphasis program specific to the construction industry. Page 21 GAO-21-122 Workplace Safety and Health Recordkeeping OSHA did not notify the approximately 195,000 remaining non- responders of their potential non-compliance with this reporting requirement. OSHA also sent postcards to about 27,000 out of nearly 220,000 employers identified through OSHA’s analysis that did not submit their 2019 300A data. OSHA did not notify the approximately 193,000 remaining non-responders of their potential non-compliance with the reporting requirement. OSHA officials said that they do not send postcards to all potentially non-compliant employers because they do not have enough funding to do so. They also said that, because they do not have sufficient staff, they do not follow-up with employers receiving postcards to ensure those that are required to, submit their data. According to OSHA officials, the agency has not evaluated its outreach procedures to ensure employers are aware of and complying with the 300A injury and illness reporting requirements to understand what efforts are working and what can be improved. Officials did not explain why OSHA has not conducted evaluation studies of all its outreach efforts; however, they stated that conducting a formal evaluation of the data match to identify possible non-responding employers did not seem necessary, given the magnitude of possible non-compliance the data match identifies. They also said that they have not conducted a formal evaluation to assess the effectiveness of sending reminder postcards to non-responding employers because they do not have sufficient staff to do so. They further explained that they have observed that some employers submit their data after receiving these postcards, but cannot quantify the number that have done so. According to federal internal control standards for monitoring, an agency should evaluate its procedures and remediate limitations or shortcomings that are brought to its attention. 46 Furthermore, we have previously reported that reexamining how regulations are implemented can help agencies evaluate how they are working in practice. Such evaluations can result in updating regulatory guidance or revising policies and procedures to improve the effectiveness of the regulation. 47 Citing non-compliance. Between December 15, 2017 (when covered employers were first required to submit 300A data) and September 30, 2019, OSHA issued 255 citations to employers for failure to report their 300A data. OSHA issued 22 of these citations during SST inspections 46See GAO-14-704G, principle 17. 47See GAO-14-268. Page 22 GAO-21-122 Workplace Safety and Health Recordkeeping and issued the remaining 233 during other types of on-site inspections. Deterring non-compliance with the 300A reporting requirement may be particularly important because employers may have an incentive not to comply—since not reporting these injury and illness summaries decreases the likelihood that their establishments will be selected for an SST inspection. 48 • Citations issued during SST inspections for non-reporting of 300A data. OSHA, in an effort to deter non-compliance, places some establishments whose employers are identified as potential non- responders on the list of establishments eligible for SST inspections. OSHA identified almost 239,000 employers as potential non- responders in the first year that it matched employers with Dun & Bradstreet data. It included 750 of them in the list of establishments it provided to its regions at the start of fiscal year 2019 as eligible for SST inspections (see table 2). 49 Table 2 also shows that only 727 out of more than 3,000 establishments on the SST list received an SST inspection. According to OSHA officials, the vast majority of these inspections would have been targeted towards establishments with high injury and illness rates as opposed to establishments where employers either did not report this data or may have underreported it. 48Research on employer underreporting of workplace injuries and illnesses indicates that one reason employers underreport injuries and illnesses is to avoid being selected for an OSHA inspection. We reported, for example, that without accurate records, employers engaged in hazardous activities can avoid inspections because OSHA based many of its safety inspections on work-related injury and illness rates. (See GAO, Workplace Safety and Health: Enhancing OSHA”s Records Audit Process Could Improve the Accuracy of Worker Injury and Illness Data, GAO-10-10 (Washington, D.C.: Oct. 15, 2009). According to OSHA officials, this procedure for inspections continues today. Also, the National Academy of Sciences identified multiple factors as contributing to underreporting of injury and illness data, including employer non-reporting due to concerns about OSHA penalties (which may be issued as a result of an OSHA inspection). See A Smarter National Surveillance System for Occupational Safety and Health in the 21st Century, (Washington, D.C.: The National Academies Press, 2018). 49On December 14, 2020, OSHA implemented a new SST Directive (CPL 02-01-062) that will generate a new list of establishments for possible SST inspection for 2 years, unless replaced by new instruction. The agency will select these establishments using 300A data from calendar years 2017, 2018, and 2019. OSHA officials anticipate that this list will consist of 3,900 establishments, 2,720 of which had high injury and illness rates in a single year and 330 of which had an increasing injury and illness rate when 3 years of data were analyzed—2017 through 2019. Of the remaining 850 establishments, the agency anticipates that 750 will be establishments whose employers did not report their data and 100 will be establishments whose employers may have underreported their data. Page 23 GAO-21-122 Workplace Safety and Health Recordkeeping Table 2: Number of Site-Specific Targeting (SST) Inspections Conducted by Region in Fiscal Year 2019 and Number of Establishments Eligible for SST Inspections Number of establishments within federal OSHA jurisdictions selected for possible SST inspection Number of Selected as potentially Selected as establishments that Selected for high underreporting potential non- Region had an SST inspection Total injury/illness rate injury/illness ratea responderb New York 125 327 191 17 119 Boston 104 313 239 7 67 Philadelphia 99 344 264 8 72 Dallas 97 485 318 15 152 Chicago 94 644 491 20 133 Kansas City 80 206 153 8 45 Atlanta 62 485 344 17 124 Denver 56 159 122 3 34 Seattle 10 59 54 1 4 San Francisco 0 0 0 0 0 Total 727 3,022 2,176 96 750 Source: GAO analysis of Occupational Safety and Health Administration (OSHA) fiscal year 2019 inspection data and OSHA-provided counts of the number of establishments selected for possible SST inspection by region. | GAO-21-122 Notes: These data are for establishments overseen by federal OSHA. Site-Specific Targeting (SST) inspection data from state plan states are not included in this table. The second column in the table shows the number of SST inspections conducted per region by federal OSHA area offices within that region. In general, the more federal area offices a region has, the more SST inspections that region conducted. For example, the New York region (which conducted the most SST inspections) had 13 federal OSHA area offices and 11 of these offices conducted SST inspections. By contrast, the Seattle region (which conducted 10 SST inspections) had four federal OSHA area offices and only one of these offices—the one located in a state where OSHA oversees workplace establishments— conducted SST inspections. The San Francisco region conducted no SST inspections because all of the states within its jurisdiction are state plan states. State plan states, such as California, Arizona, and New Mexico, are required to have an “SST-like” inspection program that is at least as effective as the federal OSHA SST program. a Selected as potentially underreporting injury/illness rate means that OSHA identified the establishment as reporting a lower than average rate for that industry. b Selected as a potential non-responder means that OSHA identified the establishment as potentially having failed to report its injury and illness summary data (form 300A) data in accordance with 29 C.F.R. § 1904.41. The OSH Act’s 6-month statute of limitations for citing violations applies to violations identified during all OSHA inspections, including SST inspections. Since OSHA selects establishments for SST inspection after the 300A reporting deadline, compliance officers or managers in four area offices said that it is challenging to issue a citation for failure to report Page 24 GAO-21-122 Workplace Safety and Health Recordkeeping these data. 50 Where appropriate, OSHA can issue citations for failure to report 300A data, but only if (1) the employer did not report the data in the current year and (2) the inspection is completed and any citations are issued within 6 months of the due date of the reporting requirement (which, beginning with calendar year 2018 data, is always March 2 of the following year). 51 • Citations issued during other on-site inspections for non-reporting of 300A data. OSHA’s ability to ensure compliance with its recordkeeping requirements through on-site inspections may be limited because the agency inspects a small percentage of the establishments it oversees each year. In fiscal year 2019, OSHA conducted 33,393 inspections, which is less than one-half of one percent of the approximately 8 million establishments that the agency estimates that it oversees. 52 In addition, due to the COVID-19 pandemic OSHA officials said that the agency is conducting even fewer on-site inspections and those that it is conducting tend to be related to complaints about the COVID-19 pandemic. 53 OSHA issued interim enforcement procedures in 2018 stating that compliance officers should determine during inspections whether employers are required to submit 300A data, and if they submitted this 50According to OSHA officials, OSHA used calendar year 2016 300A data as the basis for selecting establishments for SST inspections in fiscal years 2019 and 2020. By the time these inspections were conducted, these data were at least nearly 2 to 3 years old. 51Calendar year 2016 data were due December 15, 2017, and calendar year 2017 data were due July 1, 2018. 29 C.F.R. § 1904.41(c). 52In fiscal year 2018, for example, OSHA conducted 32,023 inspections. The estimated number of employers required to electronically submit 300A data in calendar year 2018, however, was much greater—more than 400,000, according to GAO’s estimates. 53In a prior GAO report, GAO found that between February 1, 2020, through July 31, 2020, OSHA opened 8,585 inspections, which is about 52 percent fewer inspections than the agency conducted during this same time period in 2019. See GAO, COVID-19: Federal Efforts Could Be Strengthened by Timely and Concerted Efforts, GAO-20-701 (Washington, D.C.: Sept. 21, 2020). Also, according to OSHA officials, in order to protect the health of its compliance officers, most of the inspections the agency has conducted during the COVID pandemic have been remote inspections as opposed to on-site inspections. Page 25 GAO-21-122 Workplace Safety and Health Recordkeeping data. 54 Compliance officers we spoke with in three of the seven area offices told us that they do not always determine whether employers have submitted their 300A data, thus they would not always pursue citations for non-compliance with 300A reporting requirements. Managers in a fourth area office said that they did not know whether compliance officers were verifying the submission of 300A data during inspections. Reasons compliance officers offered for not issuing citations for non-submittal of 300A data included: (1) not knowing about the interim guidance or not being instructed to follow it; (2) not being instructed on how to verify the reporting of electronic data; and (3) not having enough time to develop the evidence for issuing this type of citation, given the importance of identifying and citing other hazards. According to OSHA officials, OSHA has not evaluated this interim enforcement procedure to determine whether additional actions are needed to improve its efforts. OSHA has procedures for issuing citations that do not involve conducting on-site inspections, according to OSHA officials. They said that they are issuing citations based on remotely conducting inspections during the COVID-19 pandemic. Moreover, OSHA officials said that under the agency’s predecessor program to electronic 300A data reporting—the OSHA Data Initiative—OSHA opened inspections and issued citations without going on-site. 55 This initiative gathered the same 300A data that employers are now electronically reporting, but sent a mail-in survey to a selected group of employers who were required to respond. Under the OSHA Data Initiative, OSHA sent letters to non-responding employers requesting that they mail their data to OSHA. These letters also stated that the agency may mail a citation if the employers failed to comply. 56 OSHA officials told us that they are not using this approach for electronic reporting of the 300A data because, unlike the mail-in survey under the OSHA Data Initiative, they do not have a complete list of employers 54According to the interim enforcement guidance, area office directors are instructed to issue citations for failure to submit 300A data using the following guidance: (1) if the employer did not submit their 300A data, but provides a paper copy of these data during the inspection, an Other Than Serious citation will be issued with no penalty; (2) if the employer did not submit their 2016 300A data, but can show that they submitted their 2017 300A data, an Other Than Serious citation will be issued with no penalty; and (3) if the employer does not produce the 300A records, an Other Than Serious citation will be issued with a financial penalty. 55OSHA operated this initiative from 1997 through 2013. 56OSHA would send a certified letter to a randomly selected subset of non-responders. If OSHA did not receive the data or if the letter was not returned to sender, the agency mailed a citation to the employer. Page 26 GAO-21-122 Workplace Safety and Health Recordkeeping required to submit the data. As previously noted, OSHA has a list of hundreds of thousands of potentially non-responding employers identified in its Dun & Bradstreet data match. According to OSHA officials, OSHA has not evaluated whether this list could be used more effectively to improve compliance with the 300A electronic reporting requirement, nor has it evaluated other possible procedures for issuing citations that do not necessitate conducting on-site inspections. Because OSHA has not evaluated its procedures for encouraging compliance with the 300A injury and illness reporting requirement or its procedures for citing non-compliance, it does not know the extent to which its efforts may be improving injury and illness reporting or what other efforts it should undertake. Absent more complete information and a plan for encouraging compliance, OSHA is at risk for not achieving the SST program objective of targeting inspections to establishments with the highest injury and illness rates. According to federal internal control standards for monitoring, an agency should evaluate internal control deficiencies and develop appropriate corrective actions that remediate any identified shortcomings. Because 300A injury and illness data are being reported on less than half of the establishments that meet the reporting criteria, OSHA does not have quality information to use for its SST program. Internal control standards also call for agencies to use quality information that is complete, accurate, and current. 57 In order to fulfill its mission of ensuring the safety and health of workers, it Conclusions is essential that OSHA understand where workplace injuries and illnesses are occurring. Electronically reported 300A data—annual summaries of workplace injuries and illnesses—are crucial to understanding where injuries and illnesses are occurring. These data also assist the agency in more effectively targeting one of its scarce enforcement resources— inspections. Because our estimates show that employers do not report these data for more than 50 percent of their establishments, as required, OSHA does not know if it is getting data from employers with the highest injury and illness rates. Moreover, because OSHA is less likely to inspect employers who do not report 300A data, employers have an incentive to avoid reporting these data. OSHA has not evaluated its procedures to encourage and enforce compliance with the 300A electronic reporting requirements and, by extension, has not developed and implemented a plan to correct any 57GAO-14-704G, principles 17 and 13. Page 27 GAO-21-122 Workplace Safety and Health Recordkeeping deficiencies that such an evaluation may uncover. Specifically, OSHA has not evaluated and addressed the limitations in its outreach efforts to inform employers about the 300A reporting requirement nor in its follow- up efforts to encourage those employers who have not submitted their data as required to do so. Similarly, OSHA has not evaluated and addressed the effectiveness of its interim enforcement procedures on when to cite employers for non-reporting of 300A data. OSHA has also not explored other options for enforcement outside of the on-site inspection process. Without more effective efforts to encourage and enforce compliance with the injury and illness reporting requirement, OSHA will continue to lack necessary data to target inspections of high- risk establishments and ensure worker safety and health. We are making one recommendation to OSHA. Recommendation for Executive Action • The Secretary of Labor should evaluate OSHA’s current procedures for ensuring that employers electronically report their annual 300A injury and illness data to OSHA when required and implement a plan to remediate identified deficiencies. This should include its efforts related to: (1) encouraging employers to comply with the 300A reporting requirement; and (2) citing employers for non-compliance with this reporting requirement. (Recommendation 1) We provided a draft of this report to the U.S. Department of Labor (DOL), Agency Comments Occupational Safety and Health Administration (OSHA) for review and and Our Evaluation comment. DOL provided written comments that are reproduced in appendix II. OSHA generally agreed with our recommendation, stating that increasing compliance with its requirement that certain employers electronically report 300A injury and illness data annually is an important goal. In regards to encouraging compliance, the agency said that they will evaluate efforts and procedures to encourage employers to comply with this reporting requirement and that this will include assessing the effectiveness of existing efforts and identifying additional opportunities to inform employers of their reporting requirements. In regards to citing non-compliance, OSHA committed to assessing whether conducting remote inspections for non-compliance with recordkeeping regulations are an effective and efficient use of resources. OSHA also said that GAO’s determination that OSHA procedures for citing non-compliance with 300A data reporting requirements are inadequate fails to understand the agency’s enforcement process. OSHA Page 28 GAO-21-122 Workplace Safety and Health Recordkeeping commented that enforcement procedures require: (1) reviewing employer injury and illness records during all inspections and issuing citations and penalties as appropriate and (2) inspecting some employers during Site- Specific Targeting (SST) inspections who may not have submitted required 300A injury and illness data. We are aware of OSHA’s enforcement procedures for inspection and describe them in our report. However, as we reported, employer non-compliance with 300A injury and illness reporting is widespread despite OSHA’s existing enforcement efforts. From December 15, 2017 (when the requirement first went into effect) through September 30, 2019, OSHA issued 255 citations for failure to report 300A data, although it identified hundreds of thousands of potential non-responding employers. Relying only on on-site inspections to enforce non-compliance with this reporting requirement can be problematic because the agency inspects a small percentage of the establishments it oversees each year. Moreover, although OSHA inspects some establishments whose employers may not have reported their 300A data under its SST inspection program, according to officials, the vast majority of SST inspections are targeted toward employers who have submitted their data. This may create an incentive for employers not to report their 300A data as not reporting may reduce their chances of being selected for inspection. Without evaluating its current enforcement procedures and implementing a plan to improve the agency’s efforts to deter and address the widespread non-compliance noted in this report, OSHA will continue to lack the necessary data to target inspections to establishments with the highest injury and illness rates and ensure worker safety and health. As agreed with your offices, unless you publicly announce the contents of this report earlier, we plan no further distribution until 30 days from the report date. At that time, we will send copies to appropriate congressional committees and the Secretary of Labor. In addition, the report will be available at no charge on the GAO website at http://www.gao.gov. If you or your staff have any questions about this report, please contact me at (202) 512-4769 or costat@gao.gov. Contact points for our Offices of Page 29 GAO-21-122 Workplace Safety and Health Recordkeeping Congressional Relations and Public Affairs may be found on the last page of this report. GAO staff who made key contributions to this report are listed in appendix III. Thomas Costa, Acting Director Education, Workforce, and Income Security Issues Page 30 GAO-21-122 Workplace Safety and Health Recordkeeping List of Requesters The Honorable Robert C. “Bobby” Scott Chairman Committee on Education and Labor House of Representatives The Honorable Alma S. Adams, Ph.D. House of Representatives The Honorable Mark Takano House of Representatives Page 31 GAO-21-122 Workplace Safety and Health Recordkeeping Appendix I: Objectives, Scope, and Appendix I: Objectives, Scope, and Methodology Methodology This report examines: (1) how, if at all, recordkeeping violations cited by the Occupational Safety and Health Administration (OSHA) have changed between fiscal years 2005 and 2019 and possible reasons for any changes; and (2) the extent to which employers electronically report required summary injury and illness data and OSHA has taken steps to ensure compliance with this requirement. To address both objectives, we reviewed relevant federal laws and regulations. We also reviewed and assessed OSHA’s actions against its internal guidance, standards for internal control in the federal government, and our prior work on reexamining regulations. 1 Our review of OSHA’s guidance included relevant directives, memos, its field office manual, and evaluation studies. The internal control standards we used were: (1) the information and communication standard, along with the underlying principle which says that an agency should use quality information that is complete, current, and accurate to achieve program objectives (principle 13) and (2) the monitoring standard, along with the underlying principle which says that an agency should evaluate and remediate limitations or shortcomings with its procedures that are brought to its attention (principle 17). We also interviewed OSHA headquarters officials about how OSHA identifies and addresses recordkeeping violations; how the agency implemented new electronic reporting requirements for certain employers to annually submit their summary injury and illness (300A) data; and how it used these data to implement its newly reinstated Site-Specific Targeting (SST) inspection program. 2 Lastly, to address both objectives, we met separately with OSHA management and compliance officers at seven out of 72 OSHA area offices about how they identify and address recordkeeping violations discovered during workplace inspections and how they conducted SST inspections. These 72 offices inspected workplace establishments in the private sector and, based on our analysis 1See GAO, Standards for Internal Control in the Federal Government, GAO-14-704G (Washington, D.C.: September 2014). Also see GAO, Reexamining Regulations: Agencies Often Made Regulatory Changes, but Could Strengthen Linkages to Performance Goals, GAO-14-268 (Washington, D.C.: Apr. 11, 2014). In GAO-14-268, we stated that reexamining regulations can help agencies evaluate the extent to which regulations are working in practice and that careful oversight of regulations is important because, without it, they might prove to be less effective in achieving intended goals. 2Improve Tracking of Workplace Injuries and Illnesses, 81 Fed. Reg. 29,624 (May 12, 2016), as amended by Tracking of Workplace Injuries and Illnesses, 84 Fed. Reg. 380 (Jan. 25, 2019) (revising 29 C.F.R. § 1904.41). Employers began reporting this data in 2017. Page 32 GAO-21-122 Workplace Safety and Health Recordkeeping Appendix I: Objectives, Scope, and Methodology of recordkeeping violation data we received from OSHA, cited recordkeeping violations during fiscal years 2017 through 2019. We interviewed between two and four managers at each office, meeting with a total of 18 managers. We held discussion groups with between five and eight compliance officers at each office, meeting with a total 49 compliance officers. We invited all compliance officers at each office to attend these groups. The results of our area office interviews and discussion groups are not generalizable to other OSHA area offices. They do, however, provide insight into how OSHA area office staff identify and address recordkeeping violations and conduct SST inspections and the challenges they face in doing so. We used three criteria to select the area offices for our interviews and discussion groups—geographic dispersion, varying state injury and illness and fatality rates, and varying amounts of recordkeeping violations cited by area offices. To ensure that we selected area offices from across the country, we first selected one state from each OSHA region where OSHA sets and enforces workplace safety and health standards for private sector establishments in the state. We also ensured that the states we selected varied in terms of employer workplace injury and illness rates by reviewing Bureau of Labor Statistics state data on workplace injuries and illnesses for 2018 and fatalities for 2017 (because 2018 data were not yet available). When selecting one OSHA area office within each selected state, we ensured that these offices varied in terms of the number of recordkeeping violations they identified during inspections. We did this by analyzing OSHA recordkeeping data at the area office level for fiscal years 2017 through 2019. Our initial intent was to hold discussions at a total of nine OSHA area offices. 3 When the COVID-19 pandemic broke out, we had completed our work at seven of the nine area offices we had selected—Manhattan, Harrisburg, Augusta, Austin, Bismarck, Kansas City, and Boise. 4 Due to increased demands that OSHA was facing because of the pandemic, we decided not to conduct our work at the other two offices—Peoria and Mobile. For the first objective, we analyzed data on the number and type of recordkeeping violations that OSHA cited from fiscal year 2005 through 2019, which were the most current data available at the time of our 3Allstates in one of OSHA’s 10 regions—San Francisco—have state-run occupational and safety agencies that oversee the private sector. Thus, we did not conduct work at an OSHA area office in any of the states in this region. 4We made in-person visits to the Manhattan and Harrisburg offices and conducted discussions by phone in the other five offices. Page 33 GAO-21-122 Workplace Safety and Health Recordkeeping Appendix I: Objectives, Scope, and Methodology review. These data were from two OSHA systems—OSHA’s legacy computer system known as the Integrated Management Information System (IMIS) and the current OSHA Information System (OIS). Both systems contain detailed information on the workplace inspections conducted and the violations cited as a result of these inspections. From the IMIS system, we obtained data on all inspections that resulted in recordkeeping violations for fiscal years 2005 through 2013. From the OIS system, we obtained the same information for fiscal years 2011 through 2019. To assess the reliability of these data, we conducted electronic testing to ensure that there were no duplicate records between the overlapping fiscal years of 2011, 2012 and 2013 between the two systems. 5 For both systems, we also reviewed documentation, interviewed knowledgeable OSHA officials, and conducted electronic data testing on specific data elements. We determined that the data were sufficiently reliable for our purposes. We also identified possible reasons for changes in OSHA cited recordkeeping violations by asking OSHA area office compliance officers and managers at the seven area offices about: (1) how they identify and address these violations when conducting workplace inspections and (2) in their professional judgement, if and how the type and number of violations cited have changed over time and the reasons for these changes. For the second objective, we also estimated the extent to which employers electronically reported their summary injury and illness data annually to OSHA as required for the years 2016 through 2018. To do this, we estimated the number of establishments that met the criteria to electronically report 300A data using U.S. Census Bureau County Business Patterns (CBP) data. We then estimated the percent of establishments that reported their data to OSHA by comparing the estimated number of establishments required to submit their data against the estimated number of establishments that did. 6 • U.S. Census CBP Data: The CBP is an annual series that provides data on U.S. businesses at the establishment level. This includes the location of the establishment and the number of employees working at 5OSHA officials explained that the two systems overlapped for these years while they were phasing out the IMIS system and making the OIS system fully operational. 6As described in more detail below, legal compliance cannot be determined from these estimates alone. Page 34 GAO-21-122 Workplace Safety and Health Recordkeeping Appendix I: Objectives, Scope, and Methodology the establishment as of the week of March 12. It also includes the industry classification of the establishment, using the North American Industry Classification System (commonly known as NAICS codes). 7 We estimated the number of establishments required to submit 300A data by using the Census data to estimate the number of employers that met the reporting criteria for the years 2016 through 2018. 8 Specifically, we selected: (1) establishments with between 20 and 249 employees in industries listed in the rule by NAICS code and (2) establishments with 250 or more employees, excluding those in certain industries that are not required to routinely maintain injury and illness records. 9 For 2016, we used the county and national level data sets. For 2017 and 2018, however, we used the state level and national level data sets because Census implemented data suppression in CBP data for establishment counts of fewer than three beginning in 2017. Due to this suppression of establishment counts, we were not able to reliably assign about 0.5 percent of the establishments which potentially met the 300A reporting criteria in 2017 and 2018. We did not include these establishments in our CBP tabulations. To assess the reliability of the CBP data, we reviewed U.S. Census Bureau documentation and conducted electronic data testing on specific data elements. We determined that the data were sufficiently reliable for our purposes. • 300A Injury and Illness data: OSHA provided us with all employer electronically reported 300A injury and illness data for calendar years 2016 through 2018. From these data, we took steps to eliminate records in which the employer was not required to submit their data. To do this, we identified and removed: (1) all establishments that reported fewer than 20 employees; (2) establishments with between 7NAICS is the standard industry classification system that federal statistical agencies use to classify business establishments for the purpose of collecting, analyzing, and publishing statistical data related to the U.S. business economy. 8Under the rule, the following groups of employers are required to annually submit 300A data: (1) establishments with 20 to 249 employees at any point during the previous calendar year in certain industries, and (2) establishments with 250 or more employees at any point during the previous calendar year that are required to routinely maintain injury and illness records. 29 C.F.R. § 1904.41. 9There are some limitations to this approach due to differences between the regulatory criteria and the information in the Census data. For example, the rule applies to establishments based on their highest employee count over the previous calendar year, while the employee counts in the Census data are as of the week of March 12. Our counts of establishments subject to the rule derived from CBP data and the proportion of eligible establishments who reported 300A injury and illness data are thus estimates. Page 35 GAO-21-122 Workplace Safety and Health Recordkeeping Appendix I: Objectives, Scope, and Methodology 20 and 249 employees with an industry NAICS code that is not listed in the electronic reporting rule; and (3) establishments with 250 or more employees with an industry NAICS code listed in OSHA regulations as exempt from routinely maintaining recordkeeping forms. We excluded most government agencies from our analysis of 300A data because, with limited exceptions, CBP data does not include government agencies. 10 We also excluded establishments if the CBP data did not collect information on these establishments’ NAICS codes. 11 We excluded all establishments with 20 to 249 employees that had a missing NAICS code. Lastly, we excluded all establishments located in U.S. territories from our analysis. We performed these steps to ensure the composition of our 300A data set aligned with the composition of the Census CBP data. The list of industries in the electronic reporting rule is based on the 2012 version of NAICS codes, and OSHA requires employers to use the 2012 codes when reporting their 300A data. NAICS codes, however, are revised every 5 years, thus the CBP data for calendar years 2017 and 2018 used the 2017 version of NAICS codes. 12 Two of the 2017 NAICS codes differed from the 2012 NAICS code listed in the rule. 13 We included these codes in our analysis of CBP data for 2017 and 2018. Specifically, for our analysis of 300A data in these years, we included all establishments whose employers submitted data with either the 2012 NAICS codes or the 2017 revised NAICS codes. We did this because some employers reported their data using the 2017 NAICS codes as opposed to the 2012 codes. 10CBP only includes NAICS codes on five public sector industries listed in the electronic reporting rule. These are general medical and surgical hospitals (6221); psychiatric and substance abuse hospitals (6222); specialty hospitals, except psychiatric and substance abuse hospitals (6223); gambling industries (7132); and traveler accommodation (7211). Government agencies in these NAICS codes were included in our analysis. 11These are crop production (all NAICS codes starting with 111); animal production (all NAICS codes starting with 112); rail transportation (all NAICS codes starting with 482); and postal service (NAICS code 4911). 12NAICS codes are revised every 5 years to reflect the changing economies in the U.S., Canada, and Mexico. 13The 2012 NAICS “General Merchandise Stores” series used code 4521 for department stores and code 4529 for other general merchandise stores. The 2017 NAICS “General Merchandise Stores” series now uses code 4522 for department stores and code 4523 for general merchandise stores, including warehouse clubs and supercenters. According to OSHA, both industries are still covered by the rule. Page 36 GAO-21-122 Workplace Safety and Health Recordkeeping Appendix I: Objectives, Scope, and Methodology Out of the roughly 763,000 records that OSHA provided us for the years 2016 through 2018, application of the filters described above left us with about 73 percent of the original total in our final analysis set. Of the exclusions, about 56 percent were excluded because the establishment was not required to report the data (based on the employee count and/or NAICS codes reported) and the remaining 44 percent were excluded for the technical reasons noted above. To assess the reliability of the 300A data, we reviewed OSHA documentation, interviewed knowledgeable OSHA officials, and conducted electronic data testing on specific data elements. We determined that the data were sufficiently reliable for our purposes. For the second objective, we also analyzed OSHA’s data on the number of SST inspections that the agency conducted in the first year that these inspections were reinstated—fiscal year 2019. 14 These were the only data available at the time of our analysis. 15 To assess the reliability of these data, we reviewed documentation, interviewed knowledgeable OSHA officials about their data, and conducted electronic data testing on specific data elements. We determined that all data were sufficiently reliable for our purposes. We conducted this performance audit from May 2019 to January 2021 in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. 14This analysis was limited to SST inspections that OSHA area offices conducted. 15OSHA also conducted SST inspections from 1997 through 2014. After employers began to electronically report their summary data to OSHA in 2017, OSHA reinstated the SST program using these data to select establishments for inspection. Page 37 GAO-21-122 Workplace Safety and Health Recordkeeping Appendix II: Comments from the Department Appendix II: Comments from the Department of Labor of Labor Page 38 GAO-21-122 Workplace Safety and Health Recordkeeping Appendix II: Comments from the Department of Labor Page 39 GAO-21-122 Workplace Safety and Health Recordkeeping Appendix III: GAO Contact and Staff Appendix III: GAO Contact and Staff Acknowledgments Acknowledgments Thomas Costa, (202) 512-4769 or costat@gao.gov GAO Contact In addition to the contact named above, Blake Ainsworth (Assistant Staff Director), Nancy Cosentino (Analyst-in-Charge), Monika Gomez, Stacy Acknowledgments Spence, and Rosemary Torres Lerma made key contributions to this report. Also contributing to this report were Carl Barden, James Bennett, Sarah Cornetto, Hedieh Fusfield, David Lin, Kristen Jones, Jean McSween, Minette Richardson, and Joy Solmonson. (103563) Page 40 GAO-21-122 Workplace Safety and Health Recordkeeping The Government Accountability Office, the audit, evaluation, and investigative GAO’s Mission arm of Congress, exists to support Congress in meeting its constitutional responsibilities and to help improve the performance and accountability of the federal government for the American people. GAO examines the use of public funds; evaluates federal programs and policies; and provides analyses, recommendations, and other assistance to help Congress make informed oversight, policy, and funding decisions. GAO’s commitment to good government is reflected in its core values of accountability, integrity, and reliability. The fastest and easiest way to obtain copies of GAO documents at no cost is Obtaining Copies of through our website. Each weekday afternoon, GAO posts on its website newly GAO Reports and released reports, testimony, and correspondence. You can also subscribe to GAO’s email updates to receive notification of newly posted products. Testimony Order by Phone The price of each GAO publication reflects GAO’s actual cost of production and distribution and depends on the number of pages in the publication and whether the publication is printed in color or black and white. Pricing and ordering information is posted on GAO’s website, https://www.gao.gov/ordering.htm. Place orders by calling (202) 512-6000, toll free (866) 801-7077, or TDD (202) 512-2537. Orders may be paid for using American Express, Discover Card, MasterCard, Visa, check, or money order. Call for additional information. Connect with GAO on Facebook, Flickr, Twitter, and YouTube. Connect with GAO Subscribe to our RSS Feeds or Email Updates. Listen to our Podcasts. Visit GAO on the web at https://www.gao.gov. Contact FraudNet: To Report Fraud, Website: https://www.gao.gov/fraudnet/fraudnet.htm Waste, and Abuse in Automated answering system: (800) 424-5454 or (202) 512-7700 Federal Programs Orice Williams Brown, Managing Director, WilliamsO@gao.gov, (202) 512-4400, Congressional U.S. Government Accountability Office, 441 G Street NW, Room 7125, Relations Washington, DC 20548 Chuck Young, Managing Director, youngc1@gao.gov, (202) 512-4800 Public Affairs U.S. Government Accountability Office, 441 G Street NW, Room 7149 Washington, DC 20548 Stephen J. Sanford, Acting Managing Director, spel@gao.gov, (202) 512-4707 Strategic Planning and U.S. Government Accountability Office, 441 G Street NW, Room 7814, External Liaison Washington, DC 20548 Please Print on Recycled Paper.