2

Building a “Smarter” National Surveillance System

To move forward in improving the health and safety of workers through a national occupational safety and health (OSH) surveillance system, the committee developed guiding principles, objectives, and a narrative overview of an ideal surveillance system. As discussed in this chapter, the committee recognizes that an ideal system will take the commitment and resources of numerous stakeholders including workers, employers, health care providers (primary care, occupational health, public health), and regulatory, public health, policy, and research agencies at the local, state, and national levels.

A starting point in this discussion is the definition of public health surveillance as the “ongoing, systematic collection, analysis, and interpretation of health data, essential to the planning, implementation and evaluation of public health practice, closely integrated with the dissemination of these data to those who need to know” (Thacker et al., 2012). To characterize surveillance further: Systematic refers to using consistent methods over time and ongoing refers to continuous or periodic collection of data useful to identify patterns and trends through analysis, which entails routine and targeted application of statistical methods to data, producing information, which public health practitioners interpret to understand the possible need for public health action. The final links in the surveillance chain are the actual communication of important information to those responsible for taking action and the application of data to prevention (Thacker et al., 2012).

GUIDING PRINCIPLES

In setting forth the vision for a national OSH surveillance system to improve worker safety and health, the committee established a set of guiding principles:

OBJECTIVES

The overarching aim of an OSH surveillance system is to protect the health of working people by providing the information needed to target, design, and evaluate efforts to control workplace hazards, prevent occupa-

tional injuries and disease, and reduce associated societal costs. To serve this aim, an ideal national system would have the following objectives:

Objective 1: Guide immediate action to control threats to occupational health and safety.

Capacity is needed to identify health-related events that require immediate control efforts and to provide data and analysis in service of those efforts. Further, capacity is needed to prevent continued exposure to the identified risk. The speed of response is conditioned on the nature of the health event and on the need to identify the cause, the vulnerability of the population at risk, and the length of the latency judged to exist between exposure and health outcome. National surveillance can facilitate the identification of situations requiring immediate response, which will then require interpretation and action not only by national agencies but also by state and local agencies, employers, workers, and others.

Objective 2: Measure the burden of work-related injuries or illnesses and monitor trends over time and space.

The many measures of the burden of work-related injury and illness include the type of injuries and illnesses; the number and rate of injuries and illnesses; indicators of severity, such as requirements for hospitalization, surgery, or number of lost workdays; the short- and long-term economic and social costs; and estimates of the burden of preventable disability, such as disability-adjusted life years, and years of life lost. Monitoring trends in these measures consistently at a national scale allows for assessment of the progress toward national occupational health and safety goals and to the identification of unfavorable trends that require further attention. Further characterizing the burden by geographic region, by time period, by the nature and cause of injury or illness, and by the presence of known risk factors will allow intervention efforts to be targeted more efficiently.

Objective 3: Identify industries, occupations, and worksites as well as populations, defined by sociodemographic characteristics or work arrangements, at high risk for work-related injury, illness, or hazardous exposures.

Hazardous workplace exposures, work-related injuries, and illness are not distributed equally across the workforce or workplaces. Characterizing the impact of work-related injury and illness across working populations permits attention to be directed to populations at higher risk. Working populations are usually described through the demographics of the workforce—age, gender, race and ethnicity, language preference, nativity, or by work

and exposure characteristics defined by industry and occupation of employment, job task or specific chemical, and physical or biological workplace exposures. Other characteristics of work directly related to occupational health include work patterns such as shift work, seasonal work, as well as employer and workplace characteristics such as self-employment, employee turnover rates, and fixed- and non-fixed-location workplaces. Newer working arrangements add important complexity and will require creative efforts for surveillance systems to adapt collection procedures to describe these work settings, particularly multiemployer worksites and work arrangements that include temporary work either through temp agency hires or direct-hire temps and “gig” economy workers.

Objective 4: Detect and respond to new or emerging workplace hazards or facilitate the investigation of new diseases linked to occupational exposures.

The capacity for identifying new and emerging hazards lies largely in case-based surveillance systems involving reporting by employers, employees, and medical entities such as hospitals, laboratories, or health care providers. Ongoing examination of patterns evident in population-based surveillance efforts also can point to emerging problems. The reporting systems and infrastructure to collect and analyze these data require collaborations between state and federal OSH surveillance programs as well as the capacity to rapidly and effectively disseminate information on suspected new causal relationships that impact worker health. Most obviously, this is a concern for new industrial applications but experience suggests that known hazards continue to appear in new settings that can remain hidden if surveillance is not effectively implemented and the data generated and analyzed.

Objective 5: Guide the planning, implementation, and evaluation of programs and policies intended to prevent and control work-related injuries, illnesses, and hazardous workplace exposures.

Surveillance activities and data at the establishment level and at local, state, or national levels provide essential information upon which to build and implement effective occupational health and safety prevention programs. Once implemented, these same surveillance activities allow ongoing assessment of success as well as highlighting where additional efforts are needed. There are a wide variety of programs, each of which requires thoughtful and innovative use of surveillance data to track impact. Examples include those targeting workers (e.g., worker training programs), management administrative efforts (e.g., health and safety management programs), technology impacts (introduction and effective functioning of

TABLE 2-1 Users and Uses of OSH Surveillance Information

specific and company-wide engineering controls), health care personnel and systems (e.g., medical monitoring for consequences of work exposures), employer workplace prevention initiatives (e.g., employer confidentially benchmarking individual establishment performance against industry overall performance levels), and surveillance programs (e.g., environmental monitoring and patterns of illness or injury at work, effectiveness of regulations, success of industry, or geographic-wide risk control efforts).

Objective 6: Generate hypotheses and make data available for research.

In addition to immediate public health action, surveillance data offer great value in enabling new understandings of OSH risks and prevention opportunities. Surveillance can identify unusual patterns of disease or injury or unanticipated associations that indicate the need for further investigation. Follow-up research that takes full advantage of surveillance data can yield additional information about potential risk factors for the health outcomes or exposures under surveillance and generating hypotheses stimulating further research. An overview of the range of different users and associated uses for surveillance findings arrayed across the surveillance objectives is provided in Table 2-1.

AN IDEAL SURVEILLANCE SYSTEM

The guiding principles and surveillance system objectives outlined above serve as the basis for the report’s recommendations. The following overview also provides a description of the committee’s vision for an ideal national occupational health surveillance system for the United States—a model to consider as the proposed surveillance system evolves.

The ideal “smart” national OSH surveillance system is best thought of as not a single system but a system of systems that incorporates and coordinates the collection and analysis of a variety of data sources for capturing workplace exposures and hazards, work-related injuries and illnesses, and their determinants and then provides the platforms and coordinating tools to use the information to improve worker safety and health. The ideal system:

• Relies on dedicated occupational injury and illness surveillance systems as well as inclusion of occupational information into other public health surveillance systems;
• Maximizes the use of new information technologies and changes in health care systems to enhance the efficiency of data collection; provide the flexibility to change the data parameters collected and ensure consistency in data coding; and ensure data quality and timeliness;

• Ensures that an individual workers’ health data are kept confidential while maximizing the utility of these data for prevention; and
• Meets the core surveillance objectives at the national, state, and local levels, is periodically reviewed, and, as appropriate, is modified. Oversight would include prioritizing and coordinating surveillance programs and surveillance research and support activities across the federal agencies. The oversight would provide direction to surveillance activity located within federal agencies charged with conducting surveillance.

The envisioned ideal system is responsive and flexible. It is constructed to take advantage of existing data systems where possible, while initiating efforts to establish and support new data systems to fill identified gaps. To enable linkages across systems and to provide a snapshot of the population’s occupational safety and health status, data-collection and data-management efforts are standardized. It is also coordinated with surveillance activities across all of the Centers for Disease Control and Prevention (CDC) and their respective state and local partners, to enable and promote consideration of occupational health concerns (i.e., the impact of work on health) within the broader public health context.

The populations under surveillance include all currently employed, disabled, and retired individuals in the United States, regardless of the characteristics of their employment arrangements. Characteristics of the population to be collected include age, gender, race and ethnicity, socioeconomic status, shift status, duration of employment, work status, employment arrangement, and security. It would include the collection of information that would allow for the identification of vulnerable populations and occupational health inequities across racial and ethnic groups. Characteristics of workplaces to be collected include size, multiemployer site, public or private status, and union status to identify the extent to which these factors can direct attention to the risks for injury and illness that need attention.

The system involves collaboration and information sharing between labor, public health, and multiple agencies at all levels of government, including the Occupational Safety and Health Administration (OSHA), the Mine Safety and Health Administration (MSHA), the National Institute for Occupational Safety and Health (NIOSH) and other components of CDC, the Department of Transportation, the Federal Aviation Administration, the Department of Energy, the Federal Railroad Administration, and the Nuclear Regulatory Commission. It fosters collaboration across public health domains, allowing for leveraging both surveillance and intervention resources across all relevant domains (e.g., transportation safety, violence prevention, and infectious disease control). The system likewise involves collaboration with a wide range of partners—health care professionals and providers, insurers, industry, and labor—responsible for generating

health data and using surveillance findings to improve worker safety and health. It allows for ongoing input from data users and provides timely and sometimes “real-time” data in formats that meet their needs. Periodic comprehensive reports would be compiled and disseminated that combine information from multiple sources (i.e., employer and employee surveys for injuries and surveys and medical records for illnesses) to ensure the most comprehensive information that minimizes the possibility of providing only a partial picture of the burden of work-related injuries and illnesses.

TOOLS AND SOURCES FOR DATA COLLECTION

Very broadly, the data needed for a national occupational injury, illness, and hazardous exposure surveillance system falls into linked content areas—worker health data linked to the worker’s employer and type of work data. In an ideal system, worker health data span from birth through death linking socioeconomic, environmental, and behavioral influences on health with the worker’s personal health records. Worker employment data characterize (and where possible quantify) exposures to chemical, physical, biological, and psychosocial hazards and employment security throughout the worker’s employment history.

In an ideal system, the national OSH surveillance structure has multiple data-collection processes, which can operate separately or together to build surveillance systems that are needed by the nation. The system has multiple data-collection mechanisms (e.g., data from emergency departments, death certificates, hospital medical records, poison control centers, population-based surveys, hazard data collection, medical laboratories, and specialized radiograph interpretations) such that the nation’s surveillance system can adapt to emerging workplace safety and health issues, rely on data at established periodic intervals, and be assured of data quality and consistency across data-collection systems. Reflecting the advancing knowledge of how work impacts overall health, all health data-collection systems outside the purview of the occupational safety and health surveillance community would include information characterizing work.

Integrated data-collection systems for OHS surveillance consist of deploying worker and employer surveys, using medical and biological testing results, accessing and using administrative data, mandating reporting of specific diseases or exposures, and using vital statistics databases. Often these systems are supplemented with less traditional data sources to assign the outcome as being work related or to incorporate other content that may be valuable for prevention. For example, newspaper articles, police or coroner reports, and other data sources are used in the determination of the work-relatedness of a fatality supplementing higher value data on the death certificate maintained by the state vital statistics program. Likewise,

data collected from equipment failures or manufacturing deficiencies may supplement information about injury causation. The ideal system has real-time data submitted electronically to minimize the cost and time involved in reporting and cost and time to compile the data and make the data available to stakeholders. Data collected include elements identifying an injury as associated with work independent of payer information (i.e., workers’ compensation). Data on the medical and surgical treatment of work injuries assist in describing injury sequelae and evaluating best practices for treatment, predicting return to work and disability outcomes. Using the medical record and/or workers’ compensation payment records, the injury and current treating diagnoses are periodically described (e.g., at 1 month and at recovery) and coupled with employment records to determine predictors for poor outcomes including return to work. Severity data, as measured through hospitalization, surgery, medication use, medical costs, time loss duration, and measures describing the injury linked to the worker’s economic and social well-being, are collected from access to the medical record and from injured worker surveys.

DATA CURATION

National surveillance data will rest within federal agencies that have the responsibility not only to initiate public health actions in a timely manner but to act in a role as data curator—acquiring record-level data, maintaining data quality, and sharing data in a usable form over the lifespan of these data—thus maximizing the use of these data for public benefit. For this to happen, all the involved agencies need to have staff with the education and training to perform this data curation.

With the adoption and execution of policies for the federal government to procure sponsored research data and create public and researcher access to these data, federal agencies sponsoring data collection for surveillance, research, or other activities will acquire significant quantities of digital data. As such, data-collection systems designed for the surveillance of specific occupational conditions or exposures will live beyond their initial and intended usage.

Digital curation is “the active management and enhancement of digital information assets for current and future use” (NRC, 2015). For data to be useful, they need to be accompanied by documentation of their production, processing and analysis workflows, and their technical structure and format. Our ideal national surveillance system relies on efforts to capture occupational safety and health data in a standardized manner across state, federal, and all other data-collection systems. Standardized approaches to collecting information about work and employment characteristics (e.g., industry and occupation), using tools to transform unstructured data to

more structured data (e.g., autocoding mechanism of injury), and methods to impute data of value (e.g., race and ethnicity) are used consistently in surveillance programs. Given the importance of work as a determinant of health, all deliberations regarding the development and evolution of data-collection systems (e.g., the electronic health record or redesign of the National Health Interview Survey) will include members of the occupational safety and health research community.

Occupational safety and health digital data curation is sponsored by NIOSH with allocated staff and information technology resources with expertise in the field (NRC, 2015). Data from internal NIOSH projects, from NIOSH cooperative projects, and from other data holders such as OSHA, BLS, and state workers’ compensation programs would be submitted to a digital data repository. Data curation is not universally applied to all data but considers the long-term utility of the data for research and prevention purposes. Surveillance systems capturing unique content, or those requiring large expenditures of resources, such that duplication is unlikely, are stored. All data, metadata, and analysis methods are documented and provided to the data depository. Legal and ethical constraints regarding making the data publicly available are adhered to yet the agencies balance confidentiality with the interests of protecting workers from worker injury or illness. Systematic efforts to build data repositories which provide record-level individually identifiable data to qualified researchers are enabled through legislation and policy adoption.

BARRIERS TO A SMARTER SURVEILLANCE SYSTEM

There are numerous impediments to improving the current occupational health surveillance system, let alone to developing an ideal system. The committee has grouped these impediments into four categories: (1) confidentiality/privacy; (2) cost; (3) expertise and workforce structure; and (4) culture and mission of organizations. Box 2-1 illustrates the key barriers

within each of these four categories. These barriers lead to multiple gaps in what data are collected and effective use of the data in the current occupational health surveillance system. These gaps are discussed in Chapters 3 and 4. The committee has provided recommendations that are considered feasible to address many of the major gaps identified in Chapters 3 and 4 but understands that to overcome the key barriers identified, particularly related to cost and agency culture, will require a commitment and prioritization by stakeholders and multiple agencies at the federal and state level.

REFERENCES