The Changing Organization of Work and the Safety and Health of Working People

Knowledge Gaps and Research Directions

Chapter 2: Surveillance

• Gaps in Surveillance of the Changing Organization of Work
• Surveillance Needs
  • Type of Information to be Collected
  • Approaches to Data Collection

Gaps in Surveillance of the Changing Organization of Work

Our capacity to track or describe changing patterns of work is very limited. Thus, we lack the means to determine whether organization of work factors that present known threats to worker safety and health are becoming more or less prevalent in the workplace; we are unable to identify emergent trends in the organization of work that may pose risk; and the distribution of organizational hazards across industry, occupation, worker demographic, and other relevant sectors cannot be known. These knowledge gaps stand as primary obstacles to interventions to protect workers from known organizational hazards and to the conduct of studies to better understand the safety and health effects of emergent and suspected organizational hazards.

Unlike the European Union, which has undertaken cross-national surveys of working conditions (including the organization of work) at 5-year intervals, few mechanisms exist in the United States for recurrent or systematic investigation of the organizational aspects of working conditions. It is notable, however, that this type of information was once collected in the context of the Quality of Employment Survey (QES) that was administered on three occasions during the period 1969–1977.

The Bureau of Labor Statistics (BLS) provides periodic updates on labor market conditions and other topics relevant to organization of work through mechanisms such as the Current Population Survey (CPS), a monthly sample survey of 50,000 households. Examples of information that is relevant to the organization of work and collected by BLS through the CPS and other surveys include data on occupational growth in different sectors of the economy, job displacement and layoffs, alternative employment arrangements (e.g., temporary help agency employment), multiple job holding, earnings and benefits, and hours of work and work schedules. Information about related topics is also provided by other Federal agencies (e.g., Census Bureau and Department of Transportation), private institutes (e.g., Economic Policy Institute, Families and Work Institute), trade associations (e.g., American Management Association), consulting firms (e.g., International Survey Research Corporation), and national and international human and labor rights organizations (e.g., International Labour Office).

Reports from these sources have yielded insights on changing patterns of work that may have implications for worker safety and health. For example, a sampling of findings by these groups suggests the following: (1) hours of work seem to be growing steadily across many occupations and worker populations [Bluestone and Rose 1998; DOL 1999a; Rones et al. 1997]; (2) telecommuting and work at home are increasing steadily [Bureau of the Census 1998; International Telework Association and Council 2000]; (3) the rate of job growth in the temporary help industry appears to far exceed the rate of overall job growth [CRS 1999]; (4) job displacement due to organizational restructuring continues to grow [Hipple 1999]; and (5) job tenure has declined for many workers [BLS 2000; Neumark et al. 1997].

However, from an occupational health perspective, these data and data collection mechanisms leave much to be desired. Concerns include the fragmentary and discontinuous nature of surveys and variation in definition and measurement of working conditions in these surveys (resulting, for example, in widely discrepant assessments of the number of workers in the contingent workforce). Further, systematic data collection is lacking for major innovations in process management and associated human resource practices (e.g., lean production methods) that are spreading rapidly throughout the economy (see Mavrinac et al. [1995] for examples and discussion of these innovations).

In addition, at present no Federal or other systematic efforts exist to capture information about changes in specific job characteristics that are known risks for stress, illness, and injury. For example, since the demise of QES, there has been no way of determining whether job tasks are becoming increasingly or less repetitive, whether workloads are increasing or decreasing, whether workers have reduced or increased control in their jobs, etc., and how these trends vary from one industry/occupation or working population to another.

As a further limitation, present surveys addressing organization of work factors generally collect little or no collateral data on exposure to other workplace hazards or on health outcomes, nor do they allow organization of work data to be linked readily to other exposure data or health data. Thus, these survey efforts cannot be exploited for health effects studies on the organization of work. On the other hand, health surveillance efforts (using workplace-centered data such as workers’ compensation files and injury logs or general population surveys such as the National Health Interview Survey) usually lack sufficient job data to link safety and health outcomes to organization of work factors.

Surveillance Needs

An urgent need exists to implement data collection efforts to better understand worker exposure to organizational risk factors for illness and injury, and how these exposures may be changing. Specifically, these data collection efforts would be designed to

• describe changing exposure to organization of work factors that present known risks for illness and injury,
• detect emerging trends in the organization of work that pose uncertain or suspected risk, and
• describe the distribution of these exposures and trends within industry, occupation, demographic, and other relevant sectors.

In turn, this information would be used to justify and target interventions to reverse hazardous trends in the organization of work, and to identify and prioritize organization of work factors for further study of their safety and health effects.

The design, content, and operation of a national surveillance system for the organization of work is beyond the scope of the present exercise and should be relegated to a stakeholder panel. However, it is possible in the present context to outline important features of such a system.

Type of Information to be Collected

An organization of work surveillance strategy must at least assess organizational factors that have recognized associations with illness and injury or with psychological and physical stress at work. Good examples of these factors include job characteristics (i.e., Work Context factors in Figure 1), such as work roles and workload demands, degree of support and control afforded workers, job security, etc., that have been the subject of extensive study in the job stress and psychosocial epidemiology fields (see Kasl [1992] for a more expansive listing and discussion of these factors).

In addition, a surveillance strategy for the organization of work should include measuring broader organizational structures and practices (i.e., Organizational Context factors in Figure 1) that are presumed to influence job characteristics and risk of illness and injury, but whose effects may not be well understood. Alternative employment arrangements, organizational restructuring, and elements of high performance/lean production work systems (e.g., participative management strategies, just-in-time inventory control, multiskilling, job combination, and team work) are examples of organizational practices that would merit close attention in a contemporary organization of work surveillance system (see Mavrinac et al. [1995] for a more expansive listing and discussion of these practices).

Consistent with the causal model in Figure 3, surveillance of organization of work should extend also to collection of data on workplace safety and health programs that may offset adverse effects of organizational practices. Prime examples of such programs include occupational medicine services, employee assistance programs, safety training, and work-life programs.

Injury and illness information can also be useful in the context of organization of work surveillance. Although data on changing job characteristics and organizational practices alone can yield insights regarding risks of changing organization of work, such inferences would be strengthened if these data could be merged with health surveillance data (although merger of these different data sets often proves difficult). Alternately, some surveys of changing work experiences (e.g., the BLS National Longitudinal Survey) include injury and illness inventories and measures of psychological well-being, thereby enabling direct study of the relationship between organization of work exposures and health. Similarly, the former QES contained measures of perceived stress, job satisfaction, and work-family conflict.

From the standpoint of inferring illness and injury risk, collecting data on physical hazards (e.g., noise, ergonomic hazards) may also prove useful since exposure to these hazards can be influenced by the organization of work and may pose imminent threat to safety and health.

Approaches to Data Collection

Several strategies for improving surveillance of the organization of work can be proposed. The most desirable (and ambitious) approach would be to develop a stand-alone nationally representative survey of the organization of work. Such a survey might be modeled after and expand upon the former QES. The QES captured much of the key surveillance data described previously and was administered on three separate occasions at 4-year intervals during the period 1969–1977. Preferably, followup administration of a core survey of this nature would occur at 3- to 5-year intervals, with supplements to assess emergent conditions of interest, such as the recent trend toward telecommuting and growth of work-life programs. Logically, primary responsibility for such a survey would reside with the National Institute for Occupational Safety and Health (NIOSH) and sister agencies, such as BLS, that have the charge and technical capacity for data collection of this nature.

Second, organization of work surveillance could benefit from efforts to promote access to existing public domain and proprietary data sources that pertain to the changing nature of work. These sources include diverse data sets on the conditions of work that are developed and maintained by government agencies such as BLS, national polling organizations such as the National Opinion Research Center, and trade and industry associations such as the American Management Association or the Society for Human Resource Management. The University of Michigan Panel Study of Income Dynamics, for example, has proven to be useful for analysis of trends in hours of work. Nontraditional data sources such as organizational climate data that are collected periodically within specific organizations and industries may also hold value for surveillance purposes. Researchers also need to be vigilant to opportunities to influence these data sets by annexing items or suggesting content changes to surveys to improve their utility for surveillance purposes.

Finally, attention needs to be given to ways to improve methods and metrics for surveying the organization of work. A need exists for surveys and measures of job characteristics that economize on administration time, yet have acceptable psychometric qualities. A need also exists for development of improved methods for assessing organizational conditions (such as working hours) that have presented difficult measurement problems to researchers. Additionally, a need exists for improved standardization of content across surveys to enable comparison and aggregation of data. Resolution of these methodological and psychometric problems would serve to benefit health effects research on organization of work as well as surveillance.