National Occupational Research Agenda

Research Tools and Approaches

Because workplace safety and health issues are broad in scope and diversity, research aimed at improving worker safety and health demands the application of numerous scientific disciplines. Traditional research approaches have identified much of what is now known about occupational safety and health, but much has gone undetected because of deficiencies in the tools available to date. Further advances in identifying current hazards, controlling recognized hazards, and identifying and preventing the adverse consequences of emerging hazards will rely on the development and application of innovative research methods and approaches. This section of the Agenda addresses and identifies eight priority areas that can be categorized as research tools and approaches needed for meeting the challenges facing the occupational safety and health community.

Cancer Research Methods

Of the approximately 500,000 deaths from cancer in the United States each year, 4% (20,000) are thought to be related to exposures in the workplace. Although exposures to a number of recognized occupational carcinogens have been reduced or eliminated, many workers continue to be exposed to suspected cancer-causing hazards. Prevention of occupational cancer requires the ability to identify exposures that have carcinogenic potential and eliminate or sharply reduce their presence in the workplace. Epidemiologic research must provide the identification of selected new cohorts to enable correlation of specific exposures with human carcinogenicity. New biological markers of exposures and/or cancer-related outcomes need to be identified and integrated into epidemiologic studies. Because epidemiologic data regarding the carcinogenicity of many exposures are not currently available, research methods to evaluate and improve on the predictive value of animal and in vitro systems must be aggressively pursued.

Importance

Of the approximately 500,000 deaths from cancer in the United States each year, 4% (20,000) are thought to be related to exposures in the workplace. Although exposures to a number of well-recognized occupational carcinogens have been reduced or eliminated, many workers continue to be exposed to known or suspected cancer-causing agents. For example, several million U.S. workers are potentially exposed to substances classified by the International Agency for Research on Cancer (IARC) as human carcinogens. It is estimated that about 10% of lung cancers, 21% to 27% of bladder cancers, and nearly 100% of mesotheliomas in the general U.S. population are related to occupational exposure to recognized carcinogens. For workers with documented exposure to specific carcinogens, the percentage of site-specific cancer attributed to the exposure may be even higher, approaching 100% for vinyl chloride in the development of angiosarcoma of the liver, and 50% for asbestos in the development of lung cancer. Although there have been great advances in the treatment of some cancers, methods to prevent cancer are limited. Because of the lack of methods for ready identification of carcinogens among the approximately 4 million chemical mixtures presently in commercial use, and because of limited models for extrapolation of results from animals to humans, it has been necessary to rely on epidemiologic evidence of excess cancer among exposed workers. By the time such evidence is available, thousands to millions of workers may have been exposed to carcinogens. For example, it was estimated in 1982 that before the late 1970s (when convincing epidemiologic evidence of the carcinogenicity of asbestos was already well established), at least 27 million U.S. workers had been exposed to asbestos, resulting in 8,000 asbestos-related cancer deaths per year until well into the next century. Unfortunately, for most chemicals in use today, the epidemiologic evidence that might trigger voluntary or mandatory actions to reduce exposure does not exist or is sufficiently limited or disputed that workers continue to be exposed to potential carcinogens.

Research Opportunities

Progress in occupational cancer research has had an enormous impact not only on the protection of workers, but also on the entire field of environmental health (because best estimates of community risks are derived from data in occupational studies). However, this history is built on alarming tolls of dead workers. Today, the revolution in molecular biology has fortunately opened powerful new research approaches that may lead to information that could be used to take protective measures before workers suffer the consequences of these exposures. Advances in understanding the mechanisms of cancer causation are beginning to improve the ability of scientists to use laboratory research to evaluate the carcinogenic potential of a substance and to describe the hazard to humans with ever-increasing accuracy. More research is needed on comparative mechanisms of toxicity and on the development of rapid and inexpensive test systems to complement or modify traditional animal toxicity tests. Validated biomarkers of exposure and effect should be developed and used as part of the effort to determine the human burden of chemicals and their possible health consequences. Tools of molecular biology should be used to understand interactions of chemicals with critical target genes and to develop more accurate and less expensive methods to estimate worker exposure to chemicals. The characterization of the human genome will lead to a catalogue of human susceptibility genes, raising questions about how genetic differences influence individual response to agents encountered in the workplace.

Control Technology and Personal Protective Equipment

Recognized safety and health hazards can be managed by a variety of engineering, administrative, and worker protection techniques. These may include design changes to equipment, modifications to training efforts, or the use of personal protective equipment. Basic and applied research is needed to identify, evaluate, and develop control strategies for specific hazards and to assure their practicality and usability in workplaces.

Importance

All occupational safety and health research should have as its ultimate goal the reduction of worker exposures to hazards. Exposures to occupational hazards can be managed by the application of engineering controls, administrative policies, and personal protective equipment. Engineering controls include substitution of a safe material for a hazardous one, design changes to equipment, or modification of work methods to eliminate or reduce hazards. Changes in work practices and management policies and training programs are examples of administrative controls. In some cases where it is not otherwise possible to maintain a healthy work environment, personal protective equipment (PPE) such as respirators and protective clothing can be used to isolate workers from the hazard.

Although a great deal of research has been conducted to develop ways to control workplace hazards, the need for research in control technology and protective equipment research continues to be crucial. Limited information exists to predict the effectiveness of existing or proposed engineering controls. For many hazards and hazardous industries, control measures have not been developed because of lack of awareness of the hazard or insufficient technical and financial resources. Also, as new workplace hazards are identified, new control measures must be developed. In some cases, control measures have been proposed, but they have not been evaluated, or may not be commercially available. Often, existing controls may reduce or eliminate exposures to safety and health hazards when they are properly used; but they may not be used because of a lack of acceptance, or they may be perceived to be cost-prohibitive. In many cases, only a few parts of a job contribute most of the actual exposure and identification of the specific hazardous points would focus efforts to control exposures. In jobs where personal protective equipment is the only available control option, it must be not only be effective, but also practical for use in the workplace. It must not introduce a hazard greater than the one it is intended to prevent, a concern that has been raised by health care workers who have developed significant allergic responses to latex as a result of wearing latex gloves to prevent bloodborne infectious exposures. Personal protective equipment must also be designed and made available to properly fit and protect the growing numbers of female and minority workers.

Research Opportunities

Research in control technology and personal protective equipment can have widespread, direct impact on the safety and health of workers. A new low-cost approach to exhausting airborne lead fumes is reducing hazardous exposures in radiator repair shops across the country. The substitution of plastics for glass in bottled goods is helping prevent low back disorders among workers who are handling and transporting beverages. Respirators with improved filters are increasing the safety of workers in workplaces ranging from health care facilities to metal fabrication shops. Rapid advances in technology are dramatically increasing opportunities for improved worker protection. Robotics, computers, and satellite navigational systems might allow dangerous tasks involved in pesticide application and hazardous waste remediation to be carried out without exposure to workers. They might also allow for the elimination of many physically injurious tasks. Microsensing devices might assess workers' exposure to environmental contaminants, notify workers before chemicals break through protective clothing, and identify failures in containment systems for hazardous materials. New materials in clothing would improve the protection of fire fighters from burns, explosions, and hazardous chemicals. Opportunities abound for improving worker safety and health through new efforts in this underinvested field of occupational research.

Exposure Assessment Methods

Exposure assessment is a multidisciplinary field central to deciding whether and how to use resources for reducing workplace exposures, and to defining exposure-response relationships in epidemiologic studies. Rapid, inexpensive measurement tools and improved data analysis methods are needed for the collection of adequate exposure data and for effective intervention. These advancements will lead to (1) better identification of at-risk workers, (2) better identification of the most cost-effective control and intervention strategies, (3) better understanding of exposure-response relationships, and (4) improved baseline data for standard setting and risk assessment.

Importance

Exposure assessment is a rapidly evolving, multidisciplinary research activity. Its purpose is to provide environmental data with which to decide whether and how to reduce workplace exposures, and to define exposure-response relationships in epidemiologic studies. Imprecise estimation of exposure is often cited as the major limitation in epidemiologic research, hampering the ability to detect environmental causes of disease. Improved exposure assessments will lead to more precise characterization of exposure-response relationships for chemical, physical, and biological agents, and to more appropriate exposure limits for hazardous agents. Employers often have insufficient exposure data to guide selection of exposure controls or to justify the necessary financial investments. Moreover, accurate exposure data are equally important in evaluating the effectiveness of those controls after their implementation. The lack of cost-effective methods and measurement tools that can be used by nonspecialists has been a major obstacle to collecting adequate exposure data and instituting effective controls.

In the past 15 to 20 years, the scope of occupational exposure assessment has broadened considerably as a result of changes in technology and increased attention to nonindustrial work settings. At least three major gaps in current methods will drive development of exposure assessment methods in the next decade: (1) the lack of sufficiently precise exposure assessments to support accurate epidemiologic studies in the complex environments of today's workplaces, (2) the lack of practical measurement techniques that can be applied at reasonable cost in many workplaces where hazards may exist, and (3) the lack of validated methods for measuring relevant exposure and total dose data directly from biological samples obtained by relatively noninvasive techniques.

Research Opportunities

Researchers from a variety of fields (including industrial hygiene, chemistry, physics, molecular biology, epidemiology, and medicine) will pursue a variety of research paths to develop exposure assessment methods that are more precise, low-cost and easy to use, and more biologically based. For example, computer models may be developed to extrapolate information from historical data of limited exposure measurements to apply to large study populations, and to incorporate short-duration but high-intensity exposures such as leaks or spills into the models. Easy-to-use, direct-reading instruments and test kits will be developed to measure exposures rapidly and inexpensively in a variety of workplaces for routine monitoring, evaluating the success of control technologies, and providing data for research studies. Technological advances will permit measurement of low concentration of chemicals and biomarkers in biological specimens such as blood, urine, saliva and sweat, and research will link these concentrations to internal dose at the target organs. Laboratory analytical methods will be designed for inexpensively measuring numerous chemicals in a single sample. Research into improved measurement and interpretation of biomarkers will allow a more selective evaluation of the effects of structurally similar chemicals. Finally, research into exposure survey design and exposure data analysis methods will lead to more meaningful data for health risk assessments.

Such research will result in more clearly defined exposure assessment methods and strategies that can be recommended for wider adoption. More consistent use of well-designed exposure assessment methods will promote comparability among exposure data sets and enhance the utility of the data for a broad range of prevention activities. During the next 10 years, improved exposure assessment methods will lead to better identification of at-risk workers, better identification of the most cost-effective control and intervention strategies, better understanding of exposure-response relationships, and improved baseline data for standard setting and risk assessment, all of which are central to improving occupational safety and health.

Health Services Research

Health services research includes assessment of the way in which health care is organized and paid for and the effectiveness of the treatment and prevention of diseases and injuries. This research, which provides much of the data necessary for the formulation of health policy, is largely undeveloped when it comes to occupational safety and health. Diverse approaches are urgently needed to address important concerns about access to care for work-related problems, quality of care (including clinical and preventive practice guidelines), health professional needs and availability, and cost and service utilization patterns.

Importance

According to the Bureau of Labor Statistics, a total of 6.8 million non-fatal injuries and illnesses were reported in private industry workplaces during 1994, a rate of 8.4 cases for every 100 full-time workers. Nearly 3 million of the 6.8 million cases resulted in lost workdays or in restricted work activity. Since 1980, rates have varied within a range of 7.5 to 9.0 cases per 100 workers. It should be noted that these data are criticized for greatly underestimating occupational illnesses.

A portion of the costs of work-related injury and illness are captured in the workers' compensation system. The overall costs, including payments for health care, are huge and have grown tremendously in recent decades. Between 1983 and 1993, the proportion of total workers' compensation costs obligated for medical and hospitalization payments increased from 32% to 41%. In 1994, work-related injuries alone were estimated to cost $121 billion in medical expenses and losses in productivity and wages, a figure that excludes the cost of work-related diseases.

Given the magnitude of the problem, relatively little is known about delivering medical treatment for work-related conditions. For both emergency and nonemergency services, there is only limited information about the extent, quality, outcome, and costs of services provided by employer-based employee health services, private physicians, independent occupational health clinics, and hospital emergency departments.

Far too little is known about the experience of injured workers in the workers' compensation system. In an increasing number of States, employers are permitted to select the injured worker's medical care provider; but there have been few studies comparing the cost and quality of medical care and the extent of disability benefits provided to injured workers in employer-choice States versus employee-choice States. Although managed care is being used increasingly to provide workers' compensation medical services, it is not known how these services compare with those provided under the fee-for-service arrangements that traditionally have been employed in workers' compensation. The well-documented national shortage of occupational medicine physicians has unmeasured health and economic consequences. The increased use of managed care systems (particularly those without in-house occupational health expertise) may further aggravate the limited accessibility to health professionals trained to recognize, treat, and prevent work-related disease.

Long-standing public policy debates have emerged about the advantages and disadvantages of integrating the medical component of workers' compensation into private health insurance; unfortunately there is little research to evaluate this issue effectively. Likewise, there are few empirical data to evaluate whether the financial incentives built into the workers' compensation system are successful in preventing injury and illness.

Research Opportunities

Occupational safety and health will benefit greatly from the concerted application of many of the scientific methods developed by health service researchers. For example, there are wide gaps in the published literature on the social and economic costs of occupational injury and illness, as well as the costs and benefits of regulation and other approaches to hazard prevention. Many treatments widely used in occupational medicine have not been evaluated for efficacy and cost, nor is there information on their frequency, costs, and impact (local, regional, or national). What is the most effective rehabilitation plan to restore the physical capacity of a worker with a low back disorder or carpal tunnel syndrome? What psychologic interventions are most effective in preventing post-traumatic stress syndromes for victims and witnesses of severe or fatal traumatic work injuries or violence at the worksite? How do the costs and benefits of alternative medical procedures for work injuries and illnesses compare, taking into account long-term effects on productivity, health care utilization, and employment? How fully can a worker's capacity be restored and what limitations should be placed on future activities to prevent reinjury? What supply of specialists is needed in occupational medicine, nursing, industrial hygiene, safety, and engineering? What extent of training for primary care providers would most cost-effectively promote prevention and improve medical care? The restructuring of the health care industry offers critical new opportunities for health services research. Increasing data collection and analysis by health care insurance and provider organizations should dramatically improve the capacity of researchers to evaluate health care quality and cost issues in occupational safety and health.

Intervention Effectiveness Research

The goal of occupational safety and health interventions is to prevent disease and injury through combinations of techniques such as control technologies, exposure guidelines and regulations, worker participation programs, and training. The goal of intervention research is to determine the efficacy and effectiveness of these techniques and programs. New intervention research will assure better use of limited resources in workplace applications of prevention and control strategies. This research uses multidisciplinary approaches and focused field studies. Intervention model development, worker participation, cost effectiveness, hazard identification, and control evaluation are some of the key elements of this research.

Importance

The goal of intervention research is to develop practical strategies and techniques that effectively reduce or prevent workplace injuries and illnesses. Workplace safety and health interventions include but are not limited to developing and implementing specific engineering control technologies, process and work organization changes, information dissemination and health communication practices, worker/management participatory safety and health programs, safety and health training, selective use of personal protective equipment, and inspection and enforcement of protective exposure limits.

Intervention research is the testing and evaluation of interventions, programs, and policies. To date, a variety of approaches to intervention has been developed to protect worker safety and health across a broad spectrum of industries. Although there have been measurable improvements in worker safety and health, only a few interventions, alone or in combination, have been systematically evaluated. Consequently, many interventions are undertaken based on faith and expert judgment without convincing evidence that these approaches are effective. However, there are excellent examples of interventions that have been evaluated and shown to be effective. Successful interventions to reduce musculoskeletal disorders of the upper extremities include worker participation programs to identify problems, coupled with the development and implementation of process changes or engineering controls. Interventions to reduce toxic solvent exposure in dry cleaning establishments include retrofitting (adding parts or changing parts of) cleaning equipment, substitution of chemicals, and development of safety and health educational materials to reduce worker exposures. Hearing loss interventions have included regulations requiring auditory testing and noise control programs at the worksite. These programs can be effective when testing is done with care, when workers are educated about results and when the program is carefully maintained, updated, and implemented by a committed team of workers and management.

Although many intervention strategies have been applied to industrial settings, knowledge about what works best is limited. Many questions remain unanswered. What are the best techniques to evaluate the effectiveness of implemented control technologies? What are the barriers to the acceptance of new control technologies and approaches to eliminating or altering these barriers? What factors motivate the voluntary adoption of protective work practices? What roles do researchers, consultants, trainers, worker organizations, and industry trade groups play as partners in intervention efforts? What organizational and economic factors predict success in prevention programs, and how can programs be tailored to take account of these factors? How can intervention efforts target areas of greatest need? Why do managers and workers in some organizations implement occupational safety and health programs when others do not?

Intervention research is a new and multidisciplinary field that requires skills and disciplines not traditionally applied to occupational safety and health research. Behavioral scientists, economists, organizational theorists, and engineers, among others, should be included in interdisciplinary efforts to identify, develop, and evaluate practical prevention and control strategies. Employers, public decision-makers, and workplace safety and health teams need this information to assure better use of limited resources by making informed decisions about which prevention strategies work best.

Research Opportunities

There has been little research evaluating the impact of interventions on safety and health outcomes. The Office of Technology Assessment evaluated selected OSHA standards. Several States have assessed the impact on injury rates of State requirements that companies establish safety plans and safety and health committees, and OSHA is evaluating new inspection programs such as Maine 200. Intervention research includes the development, implementation, and evaluation of control technologies and other methods to reduce worker exposures. The ultimate questions to be answered are what works best at enhancing worker safety and health and why it does or does not work. The lack of answers hampers the introduction and maintenance of public and private-sector occupational safety and health programs; these programs face increasing demands that they document cost-effectiveness and impacts on health. Corporate safety and health programs, regulatory requirements and voluntary consensus standards, workers' compensation policies and loss-control programs, engineering controls, and educational campaigns are among the types of interventions that need to be developed, implemented, and evaluated. Data collected from such research will direct effective strategies to improve the safety and health of workers.

Risk Assessment Methods

Risk assessment is essential for setting occupational safety and health priorities and for demonstrating health impairment when promulgating occupational standards. Risk assessment has been most often applied in assessing the risk of carcinogens, often with animal bioassay data. However, evaluation of these procedures has been limited, and questions abound as to whether the resulting risk estimates are reasonable. Risk assessment for noncarcinogens, particularly quantitative approaches, is even less well developed. Improved methods are needed for using animal bioassay data and human health effects data to generate risk estimates for cancer and noncancer effects and injury.

Importance

Risk assessment is a process in which hazard, exposure, and dose-response information are evaluated. These evaluations determine whether an exposed population is at greater-than-expected risk of disease (cancer or noncancer endpoints) or injury. Once this is established, the magnitude and nature of the increased risk can be explored further, using either qualitative or quantitative approaches. Qualitative risk assessments are generally descriptive and indicate that disease or injury is likely or unlikely under specified conditions of exposure. On the other hand, quantitative risk assessments provide a numerical estimation of risk based on mathematical modeling. For example, under given specific exposure conditions, it is expected that one person per 1,000 would develop a disease or injury.

Quantitative risk assessments require (1) data providing as much detail as possible on exposures relevant to the adverse health outcomes of interest, and (2) development of a mathematical model describing that exposure-response relationship. Risk assessments based on experimental animal and molecular biologic data provide detailed information on the exposure-response relationships. However, there is often substantial concern about the validity of using risk assessments based on susceptible animal species tested at high constant doses to estimate the risks to workers who may have much lower and more variable workplace exposures. Risk assessments based on epidemiologic, population-based studies may have real-world relevance to workers, but they generally suffer from a number of limitations. These include potential confounding by risk factors for exposures other than the exposure of interest, variability in workplace exposures for any particular substance or mixture of exposures, individual variability in health response, and detection of statistically significant changes in adverse health outcomes. The integration of mechanistic data, human data, toxicity testing data, and biomathematics can be useful for developing methods that strengthen the scientific foundation on which risk assessments are based.

The risk assessment process has become increasingly formal and sophisticated over the past decade. There are many who support a greatly expanded and even more formal role for risk assessment in establishing national priorities and providing a justification for regulatory actions by Federal agencies. In occupational safety and health regulation, that process began when the U.S. Supreme Court ruled in the "benzene decision" [Industrial Union Department v. American Petroleum Institute, 448 U.S. 607 (1980)] that the Occupational Safety and Health Administration (OSHA) could not issue a standard without demonstrating a significant risk of material health impairment. The ruling allowed (but did not demand) that numerical criteria could be used to determine whether a risk is "significant." As a result of that Supreme Court ruling, risk assessment became standard practice in OSHA rulemaking for health standards, and quantitative risk assessments are preferred whenever data, modeling techniques, and biological understanding are adequate to support their development.

Research Opportunities

Research to improve risk assessment methods is needed from a wide range of scientific disciplines to provide more reliable methods for estimating the risk of adverse effects related to work. Substantial controversy surrounds currently available cancer risk assessment models, and models for noncancer effects are even less well developed. Lagging even more are methods for assessment of safety risks. Innovative and practical new approaches to modeling are needed. In addition, research needs to be directed to the following areas: designing epidemiologic and toxicologic studies that provide detailed and accurate exposure-response relationship data for specific hazards; generating more data on which to base models that include intake distribution, metabolism, and elimination; developing biologic markers for exposures and effects; and utilizing existing occupational safety and health data to ensure that human observations complement and validate risk estimates derived from animal data. Research efforts should also evaluate how risk assessment estimates are used in risk management, communicated to the public, and perceived by workers and employers.

Social and Economic Consequences of Workplace Illness and Injury

The 1993 workers' compensation cost of $57 billion reflects only a small portion of the social and economic consequences of occupational injuries and illnesses. Understanding the total human and economic impacts of occupational injuries and illnesses is crucial to setting priorities and shaping other components of the occupational safety and health research agenda. Social scientists have developed tools to describe and measure both the human and economic impacts of workplace injuries and illnesses and to evaluate the quality and effectiveness of health care. Research is needed to examine the impact of occupational injuries and illnesses on workers, their families, employers, communities, and the nation; describe and measure the effects of medical care on these costs; and target and evaluate the economic benefit of prevention efforts.

Importance

Each year, millions of occupational illnesses and injuries occur in the United States. Individuals affected by these health problems often become unable to work, or their ability to work is limited by physical impairment. The costs of work-related illness and disability (both in human and economic terms) justify devoting substantial resources to the control of workplace hazards; yet surprisingly little attention has been paid to describing and measuring these costs. Between 1972 and 1993, employer costs for providing workers' compensation rose from $6 billion to $57 billion, an annual growth rate of 12.5%. A 1991 study found that only 60% of persons reimbursed for work injuries received workers' compensation. Thus, it appears that only a fraction of health care costs and earnings lost through work injuries and illnesses is covered by workers' compensation. In addition to lost earnings and health care costs, the U.S. economy sustains other substantial costs that are hidden (i.e., unrecognized).

In addition to the direct costs of lost earnings and health care costs related to occupational injury and disease, there are numerous indirect economic costs. Employers sustain some of these, including additional hiring and training costs, disruption of work processes by workplace mishaps, and the effects of workplace injuries or exposures on the productivity of coworkers who feel at heightened risk. Other indirect costs are borne by the injured workers and their families--for example, reduced income, depletion of savings, and loss of homes; increased expenditures for professional counseling and purchased caregiver services in the home; home modifications and equipment related to disability; and deferral or loss of education for family members. Other costs may fall on the community in the form of increased use of social service programs.

There are also substantial noneconomic consequences of workplace injuries and illnesses on quality of life. Physical and psychological functioning in everyday activities may be affected, self-esteem and self-confidence may be reduced, and an individual's role in the family and community may change. Even less research has been focused on these nonmonetary costs. Studies of unemployed workers and their families and of people with chronic illnesses and disabling injuries show that income and employment losses, illness, and physical impairment can have profound human consequences on both workers and their families. Better measures of both economic impacts (direct and indirect) and noneconomic impacts will help improve targeting of resources for research, prevention, and compensation.

The consequences of work-related injury and illness on the quality and length of life are mediated by health care; therefore, medical management and treatment of occupational conditions should include consideration of the impact on the worker's post-injury wages, overall quality of life, and ability to use valued skills and knowledge. Although the amount of health care provided to workers with occupational injury or illness is substantial, relatively little is known about whether and to what extent this care succeeds in improving functioning and quality of life.

Research Opportunities

What happens to the spouse and children of a farmer, migrant worker, or construction laborer who is seriously disabled or killed on the job? What are the health care costs to Medicare for retirees with work-related cancers and lung diseases? How much do social and rehabilitative services for treatment of work-related disabilities cost State and local governments? What portions of State taxes and private health insurance premiums are providing welfare or health care to injured and disabled workers who are not being reimbursed by workers' compensation? Until recently, there has been virtually no research on the economic or social impact of work injuries and illnesses in the United States. Furthermore, there has been little research that compares the costs (or benefits) of safety and health programs with the total economic and social costs of workplace injuries or fatalities. Industry and government often lack the analytic tools and economic information to assess the effect of safety and health investments on the bottom line. Furthermore, State and national policy-makers may rely on inadequate information to target the most damaging and costly occupational safety and health problems.

Developing and conducting research to fill this tremendous information gap will require the collaboration of industry, labor, management, government, academia and others. Research will provide comprehensive national estimates for the economic burden of all occupational injuries and illnesses and specific estimates for the burden on targeted groups (e.g., specific industries, income groups, minorities, and teenage workers). Research will also expand understanding of the impact of work injuries and diseases beyond workers themselves to include their families, particularly the welfare of their children. Research will quantify cost-shifting between State workers' compensation systems and other public and private health insurance systems, and it will assess the effectiveness of the implementation of NORA. Finally, research will provide a reliable new basis for targeting and evaluating the effectiveness of investments in prevention.

Surveillance Research Methods

Surveillance systems describe where occupational injuries or illnesses are occurring, how frequent they are, whether they are increasing or decreasing, and whether our prevention efforts have been effective. The public health community relies on surveillance information to set research and prevention priorities, but critical gaps in current systems limit their usefulness. These systems need to be updated and expanded, and new systems and methodologies need to be developed. Data from these systems will then effectively contribute to the recognition and elimination of work-related morbidity and mortality.

Importance

The prevention of occupational disease and injury depends on the implementation of a variety of activities including testing chemicals and tools before they are introduced into commerce, using engineering controls and personal protective equipment to limit exposures, and providing early diagnosis and effective therapy of injured or ill workers to minimize disability when preventive measures have failed. Surveillance is the key to this system. Occupational safety and health surveillance systems collect, analyze, and disseminate relevant information about hazards found in the workplace as well as about work-related diseases and injuries. Surveillance systems identify where the problems are and are not, how frequent the problems are, whether they are increasing or decreasing, and whether prevention efforts have been effective. The public health and occupational health communities rely on surveillance information to set priorities for prevention. Although there has been substantial progress in the last decade in development and field testing of new data collection systems for occupational disease and injury surveillance, much remains to be done. Methods and systems for hazard surveillance are much less well developed.

A number of ongoing national and State-based disease and injury surveillance systems yield data useful for targeting occupational injury and illness prevention activities. For example, the NIOSH National Traumatic Occupational Fatalities Surveillance System (NTOF) identifies occupational injury fatalities based on death certificates and allows description of causes of death and comparison of rates among industries and occupations as well as trends over time. The NIOSH Fatality Assessment and Control Evaluation Program (FACE) provides in-depth field investigations of individual occupational fatalities and is effective in identifying and disseminating prevention information. The quilt work of Federal and State programs for injury and disease surveillance is further extended by the National Health Interview Survey and the National Health and Nutrition Examination Survey of the National Center for Health Statistics, the Annual Survey of Occupational Injuries and Illnesses of the Bureau of Labor Statistics, the National Electronic Injury Surveillance System of the Consumer Product Safety Commission, and many other effective and newly developed surveillance systems. The State-based Sentinel Events Notification for Occupational Risks (SENSOR) Program utilizes a diversity of sources to collect data about illness and injury among workers, including laboratory reports, hospital discharge information, workers' compensation reports, and physician reports. However, similar hazard surveillance systems do not exist.

Hazard surveillance could serve as the basis for the primary prevention of work-related morbidity and mortality because it is directed at earlier recognition of risks than are systems that simply tabulate injuries and illnesses once they have occurred. Hazard surveillance systems could help improve worker safety and health by: (1) identifying and quantifying exposure to occupational safety and health hazards associated with chemical, physical, and biological agents, biomechanical stress, unguarded machinery, elevated work surfaces, electrical energy, and psychosocial factors or job stressors; (2) targeting high-risk groups for interventions; (3) evaluating the effect of engineering technologies on the mitigation of exposures; (4) anticipating morbidity and mortality; and (5) disseminating important safety and health information. The lack of hazard surveillance systems creates a serious gap in the type of data necessary to prevent occupational disease, injury, and death.

Research Opportunities

Despite significant progress in developing and improving surveillance systems for work-related injuries, illnesses, and hazards, much remains to be done. For example, States participating in the Adult Blood Lead Epidemiology and Surveillance (ABLES) Program collect information about adult blood lead levels. This information permits intervention efforts to be targeted at high risk groups by identifying adults exposed to toxic levels of lead, such as workers involved in bridge painting.

Assorted data sources and models of surveillance exist in the public and private sectors, but most still await implementation as comprehensive, integrated national systems. This is an important research need, because NIOSH and its partners in the private and public sectors have limited data to assess nationally or locally the impact of intervention efforts on worker safety and health. Targeting high-risk populations for interventions using existing surveillance systems is also difficult.

The current restructuring of health care delivery systems throughout the United States provides a new opportunity to address these needs. Small research investments could link comprehensive health data systems to identify, track, and target occupational safety and health problems and provide information for decisions to develop interventions or to improve related medical care. Hazard surveillance remains the most compelling, least investigated approach. It promises to identify risks and exposures at worksites and industries, and risks accompanying prototypes of new technologies before injuries and illnesses occur.