MacArthur SES & Health Network
MacArthur SES & Health Network

printable version

Indicators of Work Environment: Physical Work

Summary prepared by Mark Cullen in collaboration with the Social Environment working group. Last revised November, 2000.

Chapter Contents

  1. Exposures to Physical Hazards in the Workplace
  2. Measurement of Workplace Exposures to Physical Hazards
  3. Citations and Further Reading

Exposures to Physical Hazards in the Workplace Environment

There is an extensive literature documenting the contributions of chemical, physical and biologic hazards in the workplace to the development of chronic disease and premature mortality. Many forms of cancer, chronic heart, lung, renal and hepatic insufficiency, neuropsychiatric disorders, asthma and other allergic disorders and major infectious diseases such as tuberculosis have all been strongly associated with one or more workplace exposures in epidemiologic studies of exposed workers, compared to otherwise similar populations not exposed. Invariably, the extent of severity and risk is a function of length and intensity of exposure. In some workplaces the level of risk for a particular condition or outcome may be so high as to dwarf other biologic determinants of health, as in asbestos insulators for whom cancer, not cardiovascular disease is the leading cause of death.

Having said that, it is important to point out that there exist over 100,000 different hazards in the workplaces of developed countries, of which hundreds, if not thousands have been associated with increase risk for adverse health, at least in heavily exposed groups. The effects of these agents, however, are very specific to the agent. In other words, asbestos causes pulmonary fibrosis and several forms of cancer, but contributes in no way to other chronic diseases. Benzene cause aplastic anemia and leukemia, but has no measurable effect on other major organ systems except in massive overexposure. Moreover, with the exception of physical hazards such as noise or biomechanical stressors, there is no workplace hazard to which any appreciable portion of the general population is substantially exposed; for most the fraction exposed in the general population will be less than 1%. Putting these two facts together, it is evident that the relationships between adverse exposure and altered patterns of health and disease are highly specific for exposure and outcome. In general, almost all studies of worker health which have included assessment of such risk have been based in specific workplaces where substantial numbers have the exposure of interest, and where the focus is on cause-specific, rather than general health outcomes.

Measurement of Workplace Exposures to Physical Hazards

For the purpose of both epidemiologic study as well as control of risk there is an extensive science of measurement of exposures, including an entire academic discipline—industrial hygiene—dedicated to this task. Both technological strategies—i.e., sampling devices and laboratory methods—and various surrogate strategies for reconstructing historic exposures have been extensively documented in numerous texts and about a dozen specialty journals devoted to these topics which are beyond the scope of this discussion (see Citations and Further Reading). The main point here is that only when a meaningful fraction of a population under study has a single or group of closely related exposures, and, when the outcome of interest includes the possible effects of such exposures, are any of these methods relevant. The corollary is, of course, that if any substantial fraction is exposed to such an agent or related agents, and if the endpoints do include possible effects or effects which may compete with them, then the appropriate application of such measurement tools is essential to avoid confounding or other bias.

back to top

Several investigators of health effects in larger populations have attempted to develop more generic measures or exposures for inclusion in studies of larger populations with workers from diverse backgrounds. For example, the Health and Retirement Survey included a panel of nine generic questions regarding dusts, chemicals, organic solvents, metal fumes, pesticides and electromagnetic fields in an attempt to determine whether such factors predict disability or ill-health in older working age men and women. The Harvard Six Cities Study used a single question regarding occupational exposure in an attempt to discriminate the respiratory effects of "chemicals, dusts and fumes" from other effects of regional air pollution. Likewise, several recent studies of adult asthma have used response to a single question regarding exposure to such materials in an attempt to distinguish occupational from other, non-occupational contributions to risk and outcome.

Although these investigators are to be congratulated for attempting to consider workplace exposure in assessing health risks which, in theory, such exposures could modify, the effects noted have been generally modest, as would be predicted since the system of classification lumps together disparate toxicants with differing effects, and does not allow for any meaningful categorization of dose. It would be somewhat akin to asking whether subjects took "medications, not otherwise specified." As a consequence, the measured effects are, as likely as not, a function of unmeasured confounding by SES itself (i.e., job status, since high exposed jobs are generally low prestige jobs), rather than serving to quantify the contribution of physical work environment per se.

The bottom line is that detailed and specific information regarding exposure to occupational hazards is crucial if a sizable portion of any study population is exposed, and if the possible health end-points include or compete with those of interest. This would include study of any population defined by work status, as well as one defined by a narrow geographic area in which one or two large employers may sufficiently dominate that common work exposures are likely. For larger or more heterogenous populations, the inclusion of information as in the Health and Retirement Study is valuable, but it should be appreciated more as a marker of job from an SES perspective than as well classified exposure information in its own right.

back to top

Citations and Further Reading

General Texts of Occupational Health

  1. Rosenstock L, Cullen MR, eds. Textbook of Clinical Occupational and Environmental Medicine. Saunders. Philadelphia. 1994 (Second Edition pending 2001)
  2. Rom WN, ed. Environmental and Occupational Medicine. Third Edition. Lippincott-Raven. Philadelphia. 1998.

Texts Detailing Methods for Exposure Assessment for Epidemiologic Studies

  1. Checkoway H, Pearce NE, Crawford-Brown DJ. Research methods in Occupational Epidemiology. Oxford University Press. Oxford. 1989 (especially chapters 2 and 9).
  2. Armstrong BK, White E, Saracci R. Principles of Exposure Measurement in Epidemiology. Oxford University Press. Oxford. 1992

Research Reports which have used Generic Exposure Surrogates

  1. Ferris BG Jr, Speizer FE, Spengler JD et al. Effects of sulfur oxides and respirable particles on human health: methodology and demography of populations in study. Am Rev Resp Dis 1979; 120:767-79.
    (Basic methods of the Harvard six cities study, using broad surrogate of occupational exposure to respiratory hazards)
  2. Blanc PD. Toren K. How much adult asthma can be attributed to occupational factors?. American Journal of Medicine. 1999;107:580-7.
  3. Toren K. Brisman J. Olin AC. Blanc PD. Asthma on the job: work-related factors in new-onset asthma and in exacerbations of pre-existing asthma. Respiratory Medicine. 2000; 94(6):529-35.
    (Strategies for use of generic exposure data for study of adult asthma)
  4. Juster FT, Suzman R. An overview of the Health and Retirement Study. J Human resources. 1995; 30:S7-S56
    (Overview of the survey design for the Health and Retirement study including several surrogate measures for self-reported work exposure)
  5. Zwerling C. Sprince NL. Davis CS. Whitten PS. Wallace RR. Heeringa SG. Occupational injuries among older workers with disabilities: a prospective cohort study of the Health and Retirement Survey, 1992 to 1994. American Journal of Public Health. 1998; 88(11):1691-5.
    (Example of successful use of surrogate measures for physical work factors as predictors of injury and musculoskeletal disorders)

back to top


UCSF Home About UCSF Search UCSF UCSF Medical Center