Protecting Contract Workers: Case Study of the US Department of Energy’s Nuclear and Chemical Waste Management

Abstract

Increased reliance on subcontractors in all economic sectors is a serious occupational health and safety challenge. Short-term cost savings are offset by long-term liability. Hiring subcontractors brings specialized knowledge but also young, inexperienced, inadequately trained workers onto industrial and hazardous waste sites, which leads to increased rates of accidents and injuries.

Reliable data on subcontractor occupational health and safety programs and performance are sparse. The US Department of Energy has an excellent safety culture on paper, but procurement practices and contract language deliver a mixed message—including some safety disincentives.

Its biphasic safety outcome data are consistent with underreporting by some subcontractors and underachievement by others. These observations are relevant to the private and public sectors. Occupational health and safety should be viewed as an asset, not merely a cost.

The increasing reliance on contractors and subcontractors in all economic sectors makes it increasingly difficult to protect them. We examine the challenges of protecting workers, using the US Department of Energy’s (DOE’s) environmental management of its nuclear and chemical waste as a basis for the analysis. These observations apply more generally to all types of hazardous waste remediation (the total or partial removal, treatment, stabilization, or containment of hazardous waste material), to construction, and potentially to all fields that rely on contractors.

The remediation of hazardous nuclear and chemical wastes involves a broad range of activities, conducted by people with varied levels of skills, education, and training, under a variety of supervisory and contractual structures. Protecting these workers from general construction hazards or specific toxic substances is a complex task that involves planning, implementing, training, overseeing, and evaluating health and safety programs and activities. Since 1989, DOE has conducted an extensive and expensive hazardous waste remediation program at many of its sites across the United States. These are sites where radioactive materials were mined, stored, or processed; where nuclear weapons were designed, fabricated or tested; and where radioactive waste was stored, reprocessed, or disposed of. Increasingly, this remediation work has been done by subcontractors, which imposes a serious worker health and safety challenge.

There is a strong national and even international trend for employers of all sizes to rely on contract labor, temporary hires, or “labor-ready” workers (e.g., day laborers) for tasks that have traditionally been performed by direct hires. Historically, labor contractors have played a major role in the construction industry and in American agriculture, but “contracting out” now pervades most sectors of commerce. The Occupational Safety and Health Administration (OSHA) clearly requires site owners and prime contractors to be responsible for ensuring that subcontractors provide for worker health and safety.1 However, labor contractors often skimp on providing workers’ compensation coverage,2 and on the basis of our observations over 25 years of working with the hazardous waste remediation industry, they skimp on safety training as well. The relation between increasing reliance on contract labor or “outsourcing” and health and safety was examined extensively in a report from the John Gray Institute, which identified an appalling lack of data.3 The United States is not alone in these trends.4^,5 Contracting usually saves money and sometimes takes advantage of specialized expertise.

DOE owns or is responsible for more than 100 sites in 34 states (ranging in size from a city block to thousands of square kilometers), many of which contain residual radioactive and chemical waste, a legacy of nuclear weapons production. Since the mid-1990s, the Consortium for Risk Evaluation with Stakeholder Participation (CRESP), a multiuniversity, multidisciplinary consortium (see www.cresp.org), has worked at several DOE sites to evaluate and improve worker protection for DOE and its contractor employees. CRESP’s studies included technical reviews of occupational health and safety programs at 10 sites.6 As part of CRESP, we also performed comprehensive analyses at 2 sites: the very large Savannah River Site7 and the much smaller Paducah Gaseous Diffusion Plant.8 The analyses of the Savannah River Site and the Paducah site consisted of site visits, interviews, examinations of clinical logs and records, reviews of bid specifications, reviews of health and safety plans, and observations of some of the remediation activities. Interviewees included health and safety employees of DOE, prime contractors, and subcontractors as well as procurement and technical representative personnel. In addition to the analyses at the Savannah River Site and the Paducah site, we draw on more than 25 years experience providing medical surveillance services to a variety of hazardous waste management companies.

DEPARTMENT OF ENERGY

Nuclear Weapons Legacy

Historically, under the Atomic Energy Commission and later DOE, the task of protecting workers on the US nuclear weapons complex during World War II and the cold war fell to the large corporations such as DuPont and Westinghouse, which were contracted by the federal government to build its nuclear weapons of mass destruction. To varied extents, these large industrial corporations already had elaborate occupational health and safety programs for protecting the nuclear weapons workforce. Subcontractors played a relatively minor role.

In 1989, DOE’s mission shifted from nuclear weapons production to nuclear and chemical waste site remediation (“cleanup”),9 and reliance on subcontractors increased. In the mid-1990s, faced on the one hand with a multibillion dollar estimate for cleanup, and a new Republican majority in Congress waving an elusive “corporitization” banner, the new Congress required federal agencies to act more like private corporations, without defining how this would work and without demonstrating that it would actually accelerate cleanup while reducing costs. DOE shifted its environmental management strategy. Formerly, each of its sites had been operated by a prime contractor, usually a single large corporation or university, that conducted a broad range of environmental management activities under a Maintenance and Operation Contract. Suddenly DOE required these prime contractors to play solely an integrating role and relied on many levels or tiers of subcontractors to perform the remediation work. It was assumed that such a “corporate” model would reduce costs and accelerate work. Privatization did not (and probably could not) achieve the objectives of improving efficiency and reducing cost; what privatization did achieve was making safety secondary to cost. The US General Accounting Office reported that DOE’s privatization initiative had failed to achieve its goals of “cost savings, keeping projects moving forward . . . or getting improvements in contractor performance.”10

Since its inception, DOE has evolved an elaborate occupational safety and health program that was initiated and overseen by trained staff in its Environment, Safety and Health office. Although the numbers of this staff have shrunk in the last few years, the principles embodied in DOE Order 440.A1, Worker Protection Management for DOE Federal and Contractor Employees,11 and other documents reflect and encourage a strong safety culture and establish the conditions under which DOE or its contractors would be able to recognize hazards, train workers, and prevent hazardous exposures. DOE’s injury metrics have regularly been better than those of the private sector: DOE documents impose responsibility on prime contractors for the health and safety programs and outcomes of subcontractors (lower-tier contractors). However, they also make it possible for individual contracts to specify the extent to which oversight and reporting exist (thereby nullifying to some extent DOE’s oversight). Although the principles and rhetoric are sound and were codified in Worker Safety and Health Program Final Rule (10 CFR 851),12 they have limited applicability in the procurement process. By reducing staff and relying on nonspecialists to supervise safety performance, DOE has lost opportunities to optimize worker protection. Moreover, whereas the health and safety program is excellent on paper, the fiscal exigencies at DOE and DOE’s contract incentives13 encourage prime contractors to choose subcontractors mainly on the basis of cost.

Subcontractor Health and Safety Data

The paucity of data on subcontractor health and safety performance and injury rates was brought to the attention of DOE in the mid-1990s,6^,7 and since then, DOE has made some good progress in obtaining documentation. DOE data are obtained through various data bases and through the prime contractor’s occupational medicine clinics,14 which do not generally service subcontractors. Even detailed DOE epidemiological studies15 cannot access subcontractor data. When subcontractor data are available (e.g., Suruda et al. showed that 59% of workers who died from trench collapses were subcontractor employees),16 denominator data are lacking. Injury data from neither the National Institute for Occupational Safety and Health17 nor the OSHA Fatality Assessment and Control Evaluation Program routinely distinguishes contractors from direct hires. And even in the case of construction data, injuries to employees of temporary agencies that are not identified as construction contractors would not be reported as construction injuries. Recently, DOE has improved the collection of subcontractor injury data, stating, “DOE contractors must record, maintain and post records related to occupational fatalities, injuries, and illnesses occurring among their employees (and subcontractors).”18 The data appear to show both underreporting (many firms report zero recordable and lost workday cases), as well as very high rates for some firms that do report.18

Cost Index Metric for Subcontractors

DOE has developed a cost index metric19 (Table 1 ▶) for ranking its health and safety performance. This incorporates common safety metrics with weighting factors, estimated from studies of direct and indirect costs of injuries. The cost index is the approximate cents lost per hour worked.20 It is apparent that more than half the contractors reported no safety or health events (Table 2 ▶), whereas others reported a cost index more than 10 times higher than the DOE average. This strikingly biphasic distribution is consistent with both the underreporting of accidents and the underperformance of safety.

1.

Cost Index Calculations for Ranking Health and Safety Performance, US Department of Energy

Coefficient	Outcome per Hours Worked
1 000 000	Number of deaths
500 000	Permanent transfers or terminations because of illness or injury
2 000	Lost or restricted workday cases
2 000	Nonfatal cases (not lost workday)
1 000	Number of days lost because of illness or injury
400	Number of restricted workdays

Source: Department of Energy.19

Note. The cost index = .01 × the sum of each row in the table, where the number of outcomes is multiplied by the respective coefficient in that row. The results are summed to achieve a cost index (money lost per 100 hours worked).

2.

Comparison of Injury and Cost Index (CI) Among Prime, Construction, Service, and Security Contractors: Department of Energy, 2003

	Number of Contractors Responded	Total Recordable Cases per 100 Workers	Lost Workday Cases per 100 Workers	CI ($ lost/ 100 Hours Worked)	% with CI = 0	% with CI > 10
Type of contractor
Productiona	8	1.4	0.7	4.66	0%	12%
Constructionb	47	1.6	0.7	4.56	60%	10%
Service	78	1.8	0.6	13.04	62%	20%
Security	29	3.0	1.6	14.13	59%	27%
BLS national		6.4	3.0	NA

Source. Department of Energy.20

Note: BLS = Bureau of Labor Statistics; NA = not available. Prime contractor is the highest level contractor or general contractor reporting directly to the Department of Energy.

^aProduction included operation of the weapons production facilities. These were prime contractors.

^bConstruction includes some prime as well as subcontractors.

The cost index is not widely used in industry, so comparisons are limited.20 DOE production contractors averaged 1.1 lost workday case and 2.9 OSHA recordable cases per 200 000 work hours (1990–1994 incidence rates). The average cost index was 11.8. The prime construction contractors averaged 3.0 and 6.6 on the cost index compared with the 1994 national average of 5.5 and 11.8 for construction contractors with more than 500 employees. Subcontractor data, however, were not included in these calculations. In the first quarter of 2003, DOE provided subcontractor data as well (Table 2 ▶). About 60% of the subcontractors reported zero total recordable or lost-time events, a consistent rate across construction, service, and security contracts, and one that suggests underreporting.20 This cost index, already in use by DOE, is one metric that could be used to evaluate subcontractor health and safety records, to incorporate performance experience into the procurement process, and to stimulate a safety culture at all contracting levels.

Subcontractor Safety

Although nuclear material storage, transport, stabilization, and disposal have a high hazard potential, the ability to quantify radiation facilitates greater safety controls than hazardous chemicals remediation. Moreover, by 1989, when the DOE mission switched to environmental management, the agency and its contractors had decades of experience refining its production health and safety program, yet was clearly a novice in hazardous waste remediation. This is reflected in a nearly 10-times increase in the cost index (Table 3 ▶), which jumped from an average of 0.72 to 6.7 (DOE, unpublished data).

3.

Injury Trends at the Department of Energy Savannah River Site: Consortium for Risk Evaluation with Stakeholder Participation Analysis, 1986–1995

Year	Cents Lost per Hour Worked
1986	0.2
1987	0.7
1988	0.8
1989	1.1
1990	0.8
1991	8.1
1992	4.0
1993	5.7
1994	7.5
1995	8.0

Source. Data were extracted from the Savannah River Site Occupational Medical Department.

Note. The cost index = .01 × the sum of each row in the table, where the number of outcomes is multiplied by the respective coefficient in that row. The results are summed to achieve a cost index (money lost per 100 hours worked).

Note. CRESP=Consortium for Risk Evaluation with Stakeholder Participation.

Consequences of Subcontracting

Terminology is important. DOE’s environmental management mission relies heavily on its prime contractors (also called “general contractors”) which puts legal as well as fiscal distance between the agency and the subcontractors who perform the work. The prime contractors hire subcontractors, called “lower-tier” contractors, who in turn may hire sub-subcontractors.

Subcontracting has several pervasive negative effects on health and safety.

1. Unlike those who had been previously employed by the site owner or operator, many of the workers will be new to the site and unfamiliar with hazards or safety procedures.

2. Tracking safety and documenting injuries requiring honest and accurate reporting will be more difficult for the prime contractor to ensure that all tiers of subcontractors have the required training, medical services, and protection, despite their contractual commitment to compliance.

3. There is an increased likelihood that local building trades workers, employed sequentially by several subcontractors but “owned” by no one, will have deficiencies in training, medical surveillance, and protection.

4. Tracking safety and documenting injuries will be more difficult because each subcontractor is responsible for maintaining their OSHA logs, and sites have no incentives to aggressively capture these statistics.

Mixed Message

DOE’s approach to accidents, near misses, and other safety-related events is well documented in its program “Lessons Learned.” It is also reflected in the Occurrence Reporting and Processing System, Computerized Accident/Incident Reporting System, and the Illness and Injury Surveillance Program.14 However, this excellent health and safety infrastructure has been undermined by budget cuts and personnel losses. These cuts were well underway in the mid-1990s when CRESP was asked to bring a “corporate perspective on occupational medicine” to the Savannah River Site and Paducah, and they continue today with a 25% budget cut for the Illness and Injury Surveillance Program.14

DOE delivers mixed messages to procurement officers, contractors, and subcontractors. Indeed, until the Price-Anderson Amendments Act of 1988, DOE was required to indemnify contractors for liabilities (e.g., accidents and lawsuits) associated with handling nuclear materials, without imposing penalties and holding “its contractors accountable for meeting nuclear safety.”21 A General Accounting Office review concluded that DOE was slow to promulgate a range of safety rules and relied on individual contracts to incorporate safety provisions.21

On paper, DOE makes each contractor responsible for ascertaining that lower-tier contractors (subcontractors) have adequate health and safety programs and procedures in place. This is embodied in documents such as Order 440.1A,11 which guided contractor health and safety for a decade, in the recent Worker Safety and Health Program Final Rule (10 CFR 851).12 In addition, some prime contractor documents21 required procurement officers to include “due consideration of their safety experience” and require the Subcontract Technical Representative to “monitor the work for compliance with all safety and health requirements.”22 But in practice, the laudable goals are often not realized: cost considerations come first, many small subcontractors do not have the expertise to provide complex worker health and safety programs, and few Subcontractor Technical Representatives have the training needed for enforcing safety. Moreover, Rule 10 CFR 851 allows variances if contractors cannot meet all requirements.

Although 10 CFR 851 and its predecessor (Order 440.1A11) imposed far-reaching protection against injury and exposure, the implementation of these regulations is mixed. Contractors are required to “coordinate with the other contractors . . . to ensure that there are clear roles, responsibilities and procedures to ensure the safety and health of workers at multi-contractor work-places”(10 CFR 851.11). However, in our analysis at Paducah, although bidders were required to submit workers’ compensation records, they did not have to submit a hazard analysis or a safety and health plan until after a bid was awarded. Here too, CRESP interviews revealed that some bidders had no idea how to develop such a plan and that the DOE procurement office provided examples from successful bids. These were then simply rewritten without becoming part of the subcontractor’s culture.

When a job is underway, the subcontractors are required (depending on the scope of the project) to provide a health and safety coordinator, and 10 CFR 851 specifies an “industrial hygienist.”12 One bid proposal at Paducah described the proposed health and safety officer’s credentials glowingly, including graduate training in industrial hygiene. But once the bid was secured and the work was underway, the actual onsite industrial hygienist was a technician without any specialized training (authors’ observations).

The main incentives for improved health and safety appeared to be the incorporation of workers compensation experience rating in the awarding of contracts. We could not determine how bad a safety record or a workers’ compensation experience would have to be to contravene a low bid.

CONTRACT LABOR

DOE and its large prime contractors are hardly alone in increasing reliance on subcontractors rather than direct hires. Outsourcing is reported to have many short-term fiscal consequences: reducing payroll and benefit costs and allowing temporal optimization of the workforce. It also facilitates the use of specialized technical expertise for short-term projects. Yet the long-term consequences, particularly liability for illness and injury, are rarely examined. The literature on subcontractor health and safety programs and outcomes is scant, probably because the data on fatalities, injuries, absenteeism, and utilization are not collected in a manner that distinguishes direct hires from different contractor levels. Azarirad et al. have presented statistical models that show a clear increase in injury rates associated with increased subcontracting, which “swamped” any benefits of the increased specialization a subcontractor might bring to a job.23

The largest contractor firms may employ personnel in occupational medicine, industrial hygiene, radiation protection, industrial safety, and training, whereas many small companies employ only 1 safety officer or none at all. Even by the mid-1980s, it became obvious that subcontracting in hazardous waste work created gaps in worker protection. With regard to implementation of the then fresh OSHA Hazardous Waste Operations and Emergency Response (29 CFR 1910.120),24 Gochfeld et al.25 reported a new “side of the hazardous waste industry, notably cleanup contractors that employed untrained, high-turnover, unprotected personnel to perform the most hazardous tasks often through the use of labor-ready workers, rather than their own employees.” The “extensive reliance on subcontractors to perform many of the components of hazardous waste site areas makes it difficult to track responsibility for enforcing the provisions of the OSHA standards.”25 At the same time, we found that more than half of the hazardous waste workers enrolled in our medical surveillance program had been with their company less than 2 years, and usually their previous medical surveillance and training records did not follow them from one company to the next.25

Glazner et al.26 found 3 characteristics of contractors that had below-average lost-worktime injury rates: (1) management always established safety goals for supervisors, (2) the contractors enforced drug testing, and (3) work was completed on budget rather than over budget. However, they cautioned that safety managers were likely to report good safety practices even if they were lacking, and their results reflect a “pattern of counterintuitive results” that require further study. Hinze and Gambatene27 found low turnover, drug testing, and professional training, as well as larger contractor work-force, favored better safety performance.

John Gray Institute Report

The most comprehensive study of safety related to contract labor is detailed in the John Gray Institute report Managing Workplace Safety and Health: the Case of Contract Labor in the US Petrochemical Industry.3 The study was initiated by OSHA after the 1989 catastrophic, subcontractor-caused explosion at the Phillips Oil Refinery in Pasadena, Tex. The study surveyed full-time workers, part-time workers, and managers regarding the extent of subcontracting in the petrochemical industry. The John Gray study results were integrated into OSHA’s Process Safety Management of Highly Hazardous Chemicals standard (29 CFR 1910.119).28 The chief conclusions of the study were that there was widespread and increasing reliance on contract workers who are younger and less experienced than direct hires and serious underreporting of accidents, illnesses, and injuries (data on injury rates for contract workers were either missing or unreliable). The report stated that “The majority of firms . . . advise their plant managers to avoid responsibilities for training and supervising the contract labor force . . . in order to avoid whatever legal and financial responsibilities would be incurred if the plant was found to be a co-employer of the contract workers.”3 At the Savannah River and Paducah sites, DOE likewise expected lower-tier contractors to train their workers on performance and safety.7^,8

On the basis of limited data, the John Gray study found that contract workers were more likely to be involved in accidents than were direct-hire workers. Whether this reflects differences in intrinsic job hazards or lesser experience remains unclear. Subcontractors were more likely to have shorter and less-effective training programs and were less likely to embody a safety culture.

Cause of Accidents

Safety programs have well-established procedures for investigating and preventing accidents. One of these is root-cause analysis, an important component of DOE’s industrial safety program.29 Root causes of most accidents include inadequate training; deficient safety enforcement; unsafe equipment, site conditions, work methods, or sequencing; and a poor attitude towards safety. The contribution of so-called worker error therefore represents primarily a failure of training and supervision. Youth, inexperience, production speed-up, long work hours, and fatigue are risk factors for accidents and injury.

Age and Experience

Accidents and injuries occur disproportionately among inexperienced workers and young workers. Root and Hoeffer30 noted that as early as 1918 the Bureau of Labor Statistics compared the rate of 111 accidents per 1 000 workers who had less than 6 months experience, with 8.5 accidents for workers who had 10 or more years experience. A disproportionate number of injuries occur in the first month of work.20 Subcontract workers are more likely than direct hires to work more than 60 hours a week, and they are more likely to be younger, less educated, and less experienced than are direct hires.3 They are also less likely to have English as a first language, thereby missing out on training and warning information. Habeck31 found that the amount of safety training provided to subcontractor workers was positively correlated with safety performance, although direct hires benefited more from training.32

Occupational Safety and Health Administration

For a long time, the Occupational Safety and Health Administration has been cognizant of problems related to contracting. The Rules of Construction (29 CFR 1926.16)1 impose safety responsibility on all levels of contractors: “the prime contractor assumes the entire responsibility under the contract and the subcontractor assumes responsibility with respect to his portion of the work. With respect to subcontracted work, the prime contractor and any subcontractor or subcontractors shall be deemed to have joint responsibility.” OSHA increasingly relies on voluntary compliance and encourages participation in its voluntary protection programs (http://www.osha.gov/dcsp/vpp/index.html), but unfortunately, OSHA “does not collect complete comparable data” that would facilitate an assessment of the efficacy of voluntary compliance in the protection of subcontract workers.33

Collecting Subcontractor Data

Data needs are simple and reflect little change from the John Gray report of 1991. Subcontractor data are sparse, and reliability of existing subcontractor data uncertain. Even when injuries are broken down by employer, denominator data are lacking. The opportunity for a prospective subcontractor database exists in a ready-made form. Under the Davis-Bacon Act, 40 USCA §§ 276a–276a-5 (1931), construction contractors at federal facilities are required to submit weekly Payroll Certification Forms. This weekly form is now used only to document compliance with the prevailing wage requirement. It includes the name and address of the subcontractor; the project; the name, address, and social security number of each employee assigned to the job; the work classification; the number of hours worked each day and total hours worked each week; as well as rate of pay.

This form provides the ready-made elements of a registry that could have great value for future epidemiological study and risk notification; this could be completed at negligible additional cost. These data, which contain a project identifier and date, could easily be linked to project files that describe the job and identify the potential hazards or exposure. Somewhat less easily, this information could also be linked to worker training documentation, which should be included in all contract arrangements.

Projects that can be identified as “service” do not fall under the Davis-Bacon Act. On paper, Rule 10 CFR 85112 does not distinguish construction contracts from service contracts, and DOE could contractually require a similar weekly worker certification from service contractors to establish its registry. However, there is a substantial incentive for prime contractors to define projects as “service” whenever possible. We learned, for example, that a well drilled for a groundwater pump-and-treat is clearly a construction project. But the same well, drilled for environmental sampling, could be classified as service.

One example concerned the abrasive blasting and painting of about 30000 uranium hexafluoride cylinders at Paducah, which was performed by a building trades company, but under a service rather than construction contract. This removed it from the prevailing wage and payroll certification requirements and also resulted in less oversight. The DOE industrial hygienist assigned to oversight of the abrasive blasting performance and safety could not directly observe or measure in the abrasive blasting enclosure because, as he explained, he did not have respirator clearance under OSHA’s Respirator Clearance standard (29CFR1910.134)34 to enter the area.

CONCLUSIONS AND RECOMMENDATIONS

The reliance on contractors, subcontractors, and labor-ready workers is widespread and increasing35 in both the public and the private sectors of the US economy. Both DOE and OSHA recognize that on the basis of a “hierarchy of responsibility” and DOE rule 10 CFR 851, each level of contractor is responsible for ensuring the adequacy of the health and safety programs of all lower-tier contractors. The opportunity to implement and evaluate health and safety performance is spotty, particularly compared with the ability to monitor production performance. A vertical reporting requirement would improve data gathering: each lower-tier contractor would report safety and health events, including near misses, to their immediate superior contractor; each contractor would, in turn, forward such reports, along with their own records, up the chain all the way to DOE and, when appropriate, to OSHA.

Led by its Office of Health, Safety, and Security, DOE is making strides toward enhancing subcontractor protection and capturing subcontractor safety data for workers involved in remediating hazardous nuclear and chemical waste and related activities. Yet, lucrative financial incentives for heightened productivity exist and are not offset by disincentives for unsafe work practices.12 The suggestion that higher injury rates among subcontractors reflect intrinsically hazardous work cannot be adequately tested, but lack of data is no excuse for failing to provide worker training and protection commensurate with hazards. Subcontractors should have safety procedures and personnel training appropriate to the hazards they face, but training alone is not adequate without oversight and enforcement.29

Site “hosts” such as DOE (or any agency or facility owner) must maintain vigorous oversight of worker health and safety and retain authority as well as responsibility. Host supervision reduces subcontractor accident rates.29 Oversight requires top-down authority as well as bottom-up implementation. It is essential that each contractor is responsible for the health and safety of all lower-tier contractors and that they actively require, seek, and analyze subcontractor safety and health metrics to implement an integrated program and ensure that a safety culture is assimilated throughout the contracting tree.

Our observations and analysis of the DOE contractor program is applicable throughout industry and governmental agencies. As a national policy, data on recordable events and occupational disease and injury should be reported separately for each tier of subcontractor and prime contractor as a first step toward holding contractors at all levels responsible for illness and injury statistics. An important incentive would be to incorporate all subcontractor recordable events into the statistics of each higher tier of contractors. Further, procurement activities should follow existing guidelines, including emphasis on vendor safety performance history. Contracting should not be viewed as a mechanism of distancing site owners, such as DOE or private industry, from the risks and liabilities of the work at its sites. Safety must be viewed as an asset, not just as a cost.