Oxford Textbook of Public Health
Dean B. Baker and Philip J. Landrigan
Extent of occupational injury and disease
Estimates of cost and economic loss
Causes of under-recognition of occupational disease
Major types of occupational injury and disease
Occupational lung diseases
Occupational skin disorders
Occupational infectious diseases
Occupational reproductive disorders
Severe occupational traumatic injuries
Occupational exposure to noise
Economic globalization and workers’ health
Special populations of workers
Workers in developing nations
Recognition of occupational disease
Surveillance of occupational disease
Occupational hazard surveillance
Occupational disease surveillance
Prevention of occupational disease
Elimination or substitution
Use of personal protective equipment
Right to know
Workers constitute a large and important population. The World Health Organization (WHO) estimated in 1997 that the global labour force was about 2600 million with 75 per cent of these working people in developing countries (WHO 1997). The officially registered working population includes 60 to 70 per cent of the world’s adult males and 30 to 60 per cent of the adult female population. In addition, the International Labour Organization (ILO) estimated that in 1996 there were 250 million children between the ages of 5 and 14 years who were engaged in economic activity—at least 120 million of them on a full-time basis (IPEC 1999). Most people between the ages of 22 and 65 spend approximately 40 per cent of their waking hours at work, so working is a central feature of most people’s lives (Leigh et al. 1997).
Workers suffer a broad range of illness caused by hazards encountered in the workplace (Rom 1998; Stellman 1998). The illnesses include:
lung cancer and mesothelioma in asbestos workers
cancer of the bladder in dye workers
leukaemia in workers exposed to benzene
chronic bronchitis in workers exposed to dusts
disorders of the nervous system in workers using solvents
heart disease in workers exposed to carbon monoxide
impairment of reproductive function in men and women using lead and certain pesticides
chronic diseases of the musculoskeletal system in workers who suffer repetitive trauma.
National and international organizations, policies, and laws concerned with workplace safety and health have developed rapidly during the past three decades in response to concerns about workers’ health. In the United States, the Occupational Safety and Health Act was passed by Congress in 1970. Its goal was ‘to assure as far as possible every working man and woman in the nation safe and healthful working conditions’. In the United Kingdom, the first legislation pertaining to occupational health was the Act for Better Regulations of Chimney Sweeps and Their Apprentices, enacted in 1788. This Act was followed by passage of a series of protective laws through the nineteenth and early twentieth centuries. The Health and Safety at Work Act, enacted in 1974, provides a broad legislative framework for the protection of workers through specific regulations. The Control of Substances Hazardous to Health Regulations of 1988 oblige employers to control hazardous substances through comprehensive assessment and documentation of potential risks from using these substances in the workplace. The European Union adopted a policy in 1989 on the Fundamental Social Rights of Workers, emphasizing the need for safety and health protection in the workplace, improvements in living and working conditions, and provision of social protection for workers. The European Union has encouraged harmonization in occupational safety and health practices among member states through issuance of directives (Wright 1998). These directives provide guidance to nations worldwide on laws needed to protect workers. Laws governing workplace safety and health now exist in virtually all developed countries and in a growing number of developing countries.
Despite the passage of protective legislation, success in reducing work-related illness has remained elusive. For example, silicosis, a disease recognized since antiquity, is still the most widespread occupational lung disease (WHO 1999). Approximately 300 deaths are attributed to silicosis annually in the United States (OSHA 1996). It is found in as many as 25 per cent of Korean and Malaysian miners. Although used less frequently than in the past in many developed countries, global production of asbestos continues to increase, and it remains widely dispersed even in countries where new uses are banned. Sales of asbestos in the 1990s actually increased in a number of developing nations as a consequence of aggressive marketing campaigns which claimed misleadingly that some forms of sbestos are ‘safe’ (Landrigan et al. 1999). Occupational hazards are being relocated from countries with more protective laws to countries with less regulations or less enforcement (LaDou 1995). Finally, there is substantial uncertainty about possible risks to health of new technologies and changes in work organization associated with economic globalization.
Occupational diseases are highly preventable. They arise from man-made conditions and therefore can be prevented through alteration of those conditions. The fundamental public health techniques of surveillance, identifying groups at risk, and intervening to control the spread of disease may all be applied effectively in the control of work-related illness. Moreover, occupational disease is not necessarily restricted to the workplace. Toxic hazards may escape from the workplace or be released into the community environment to pollute the air, drinking water supply, or food chain. Also, occupational toxins have been transported home on the shoes and clothing of contaminated workers to cause such illnesses as lead poisoning, chloracne, and mesothelioma in children and other family members. There is growing recognition of the close relationship between occupational and environmental health.
This chapter will review the impact of workplace hazards on workers’ health, including changes associated with economic globalization and impacts on special populations of workers such as women and children. It will discuss approaches to more efficient identification and better surveillance of the illnesses caused by work. Finally, it will review and evaluate the hierarchy of techniques used for the prevention of occupational injuries and illnesses.
Extent of occupational injury and disease
In most nations there is no completely reliable source of information on the extent of work-related injuries and diseases. Even so, the ILO estimates that work-related injuries and illnesses kill 1.1 million people worldwide each year, including 300 000 worker fatalities from 250 million workplace accidents (WHO 1999). An estimated 160 million new cases of work-related diseases occur each year worldwide.
In the United States, employers reported 6.3 million work injuries and 515 000 cases of occupational illnesses in 1994 (NIOSH 1996). Because of limitations in using only employer reports to measure work-related injuries and illnesses, Leigh et al. (1997) used multiple data sources to develop more accurate estimates of the annual incidence and mortality associated with occupational injuries and illnesses in the United States. They aggregated national and regional datasets collected by the Bureau of Labor Statistics, the National Council on Compensation Insurance, the National Center for Health Statistics, the Health Care Financing Administration, and other government agencies. They applied an attributable risk proportion method to assess incidence of and mortality from occupational injuries and illnesses. For this method, they reviewed the scientific literature to derive conservative estimates for the proportion of specific injuries, diseases, or mortality that are work related. Examples of the proportions of mortality they attributed to occupational causes are cancer (6–10 per cent), cardiovascular and cerebrovascular disease (5–10 per cent among adults up to 65 years of age, or 0.6–1.2 per cent overall), chronic respiratory disease (10 per cent), pneumoconioses (100 per cent), nervous system disorders (1–3 per cent), and renal disorders (1–3 per cent). Based on this analysis, they estimated that there are approximately 6500 job-related deaths from injuries, 13.2 million non-fatal injuries, 60 300 deaths from occupational illness, and 862 200 new occupational illnesses annually in the United States workforce of 120 million persons (Leigh et al. 1997).
Estimates of cost and economic loss
Total economic losses due to occupational injuries and illnesses are large. The ILO estimated that overall economic losses from work-related injuries and illnesses in 1997 were approximately 4 per cent of the world’s gross national product (WHO 1999). In 1992, the direct cost in compensation for work-related injuries and illnesses in European Union countries was 27 000 million ecus. The major categories of work-related injuries and diseases responsible for these costs were musculoskeletal (40 per cent), heart diseases (16 per cent), injuries (14 per cent), respiratory diseases (9 per cent), and central nervous system conditions (8 per cent).
To assess costs associated with work-related conditions in the United States, Leigh et al. (1997) used the human capital method that decomposes costs into direct categories such as medical and insurance administration expenses as well as indirect categories such as lost earnings, lost home production, and lost fringe benefits. They applied these costs to the estimated annual job-related deaths from injury and illnesses, non-fatal injuries, and illnesses cited above. The total annual cost for 1992 was estimated to be US$171 billion, comprising US$65 billion direct costs and US$106 billion indirect costs. Injuries cost US$145 billion and illnesses cost US$26 billion. These estimates are likely to be low because the numbers of occupational injuries and illness are probably undercounted in the national datasets and because they ignore costs associated with pain and suffering as well as within-home care provided by family members.
Causes of under-recognition of occupational disease
Although recording of workplace injuries is reasonably accurate in most developed countries, surveillance systems generally result in substantial underestimates of the actual cases of occupational illness. One explanation for under-recognition of occupational disease is the inherent difficulty in diagnosing occupational diseases and in establishing cause-and-effect relationships. The link between occupation and disease is often elusive, because most occupational diseases are not distinct clinically and pathologically from diseases associated with non-occupational aetiologies. For example, the skin cancer caused by polycyclic aromatic hydrocarbons is similar in appearance to that caused by sunlight. Similarly, solvent-induced encephalopathy may easily be attributed to old age and lead-induced nephropathy to high blood pressure or diabetes. Only in rare instances, such as the associations between asbestos and mesothelioma (Selikoff et al. 1964) and between vinyl chloride monomer and angiosarcoma of the liver (Creech and Johnson 1974), is the causal association between occupational exposure and disease readily established on clinical grounds alone.
A second cause of the under-recognition of occupational disease is that the majority of chemicals in commerce have never been evaluated with regard to their potential toxicity. Only about 12 000 of the estimated 70 000 chemicals commonly used in industry have been tested for toxicity in animals (LaDou 1997). Furthermore, the toxicity testing has concentrated primarily on high-dose, acute effects and on the long-term risk of cancer. Toxicity testing of reproductive, neurological, immunological, and other adverse effects remains quite limited.
The long latency which typically elapses between occupational exposure and onset of illness is a third factor which may obscure the occupational aetiology of chronic disease. For example, few occupational cancers appear within 10 or even 20 years of first exposure. Similarly, the chronic neurotoxic effects of solvents may become evident only after decades of exposure. A worker so affected may well have retired. In such a case, it is unlikely that the worker will be diagnosed as having a disease of occupational origin.
Lack of awareness among health practitioners about the hazards found at work is a fourth cause of underestimation of occupational disease, reflecting the fact that most physicians are not adequately trained to suspect work as a cause of disease (IOM 1988; Goldman et al. 1999). Very little time is devoted in most medical schools to teaching physicians to take a proper occupational history, to recognize the symptoms of common industrial toxins, or to recall the known associations between occupational exposures and disease.
Compounding this lack of medical awareness is the limited ability of many workers to provide an accurate report of their exposures. Workers may have had multiple toxic exposures in a variety of jobs over a working lifetime. In most countries, there are no requirements to inform workers of the nature or hazard of the materials with which they work. Even in the United States, employers’ reporting requirements remain limited under the Hazard Communication Standard and under state right-to-know laws (National Research Council 1987). In many instances, an ill patient will simply not know about his or her past occupational exposures.
Finally, given the potential financial liability associated with the finding that a disease is of occupational origin, employers may be resistant to recognizing the work-relatedness of a disorder, especially in cases where personal habits or non-occupational pursuits are possible contributory factors. Since employers are often in the best position to recognize causal associations between workplace exposures and disease, this conflict of interest represents a major obstacle to obtaining accurate estimates of the burden of occupational illness.
Major types of occupational injury and disease
Occupational illness can affect virtually every organ system (Rom 1998; Stellman 1999). Occupational diseases of the lung and skin are common since these organs have substantial surface areas in direct contact with toxic substances. Noise-induced hearing loss and musculoskeletal disorders are the most common disorders arising from physical factors in the workplace. Occupational cancer is a major concern because of the high mortality associated with many forms of cancer. Increasing attention has been paid in recent years to diseases affecting the neurological, reproductive, and immunological systems as sensitive measures have become available to demonstrate adverse effects of chronic low-level occupational exposures. The following examples illustrate major occupational injuries and diseases. Table 1 shows examples of occupational diseases associated with pertinent industries and toxic agents.
Table 1 Examples of occupational diseases and related hazards by occupation and agent
Occupational lung diseases
Occupational lung diseases are very important in the field of occupational medicine because the lung is an accessible target organ for airborne toxic substances. Major categories of lung disease include the ‘dust diseases’ of the lung or pneumoconioses, lung cancer, occupational asthma, industrial bronchitis and other effects of irritants, and infections. Silicosis is the most common pneumoconiosis worldwide. Exposure to silica occurs in a wide variety of occupations such as sandblaster, miner, miller, pottery worker, foundry worker, and workers using abrasives. The Occupational Safety and Health Administration (OSHA) estimates that there are over 2 million workers potentially exposed to silica dust in the United States, including 100 000 workers in high-risk jobs (OSHA 1996). The International Agency for Research on Cancer has classified crystalline silica as a known human carcinogen (IARC 1997).
Asbestos is also an important cause of pneumoconiosis and other lung diseases. Asbestos has been responsible for over 200 000 deaths in the United States and it will cause millions more deaths worldwide (Collegium Ramazzini 1999). All forms of asbestos cause asbestosis, a progressive fibrosis of the lungs. The preponderance of scientific evidence indicates that all forms of asbestos also cause lung cancer and malignant mesothelioma. Peto et al. (1999) estimate that deaths from malignant mesothelioma among men in Western Europe will increase from over 5000 in 1998 to about 9000 in 2018. They estimate that there will be more than half a million asbestos-related malignant mesothelioma cancer deaths in Western Europe over the next 35 years. Given the long latency, the future burden of mortality resulting from asbestos will be substantial even if all future exposures were to be eliminated completely.
Bronchial asthma affects about 5 to 10 per cent of the population in developed countries, and estimates in the United States suggest that occupational asthma accounts for up to 28 per cent of asthma in adults (NIOSH 1996). In some jurisdictions, occupational asthma has become the most prevalent occupational lung disease, exceeding silicosis and asbestosis. Even so, prevalence studies of occupational asthma usually underestimate the number of affected workers because these workers tend to quit jobs where they suffer such symptoms. Concern about occupational asthma has increased because studies have shown that many workers who develop occupational asthma do not recover completely even several years after removal from exposure (Venables and Chang-Yeung 1997; Perfetti et al. 1998; Ross and McDonald 1998).
Many gases, fumes, and aerosols are directly toxic to the respiratory tract by causing acute inflammation. Examples include soluble irritants such as hydrogen chloride, ammonia, and sulphur dioxide which produce effects in the eyes, nasopharynx, and large airways. Less soluble irritants such as nitrogen dioxide, ozone, or phosgene produce few upper respiratory symptoms, but in high exposure can cause a toxic pneumonitis. Long-term exposure can lead to lung fibrosis.
About 300 to 350 substances have been identified as occupational carcinogens (WHO 1997). They include chemical substances such as benzene and asbestos, physical hazards such as ionizing radiation, and biological hazards such as viruses. It is estimated that approximately 16 million workers in the European Union are exposed to carcinogens at work. The most common cancers due to these workplace exposures are cancers of the lung, bladder, skin, pleura (mesothelioma), liver, haematopoietic tissue, bone, and soft connective tissue.
Estimates of the fraction of cancers caused by occupational exposures vary from 4 to 40 per cent (Pearce et al. 1998). The large variability in the estimates arises from differences in the datasets used and the assumptions applied. The most commonly accepted estimate is 4 per cent with a plausible range, based on the best quality studies, being 2 to 8 per cent. However, if one considers not the whole population, but only the adult population in which exposure to occupational carcinogens almost exclusively occurs, the proportion of cancer attributed to occupation would increase to about 20 per cent among those exposed (Pearce et al. 1998).
Leigh et al. (1997) estimated that 6 to 10 per cent of the cancer mortality in the United States could be attributed to workplace exposures, accounting for 31 025 to 51 708 deaths annually. The work-related attributable proportion varies by cancer type. About 10 per cent of lung cancers, 21 to 25 per cent of bladder cancer, and nearly 100 per cent of mesotheliomas in the general population are related to occupational exposures (NIOSH 1996). For workers with exposure to specific carcinogens, the percentage of cancer attributed to the occupational exposure is even higher, approaching 50 per cent for asbestos in the development of lung cancer and 100 per cent for vinyl chloride in the development of angiosarcoma of the liver (NIOSH 1996).
Occupational skin disorders
Skin disorders are among the most commonly reported occupational diseases (Adams 1999). In 1997 the estimated rate of occupational skin disorders in the United States was 6.7 per 10 000 full-time workers, accounting for about 13 per cent of reported occupational diseases—second only to disorders associated with repeated trauma (BLS 1998). Contact dermatitis accounts for almost 90 per cent of these skin disorders, while about 5 per cent are due to skin infections. Although skin disorders are relatively easily diagnosed, occupational skin diseases are believed to be severely under-reported, so that the actual rate may be many times higher than officially reported (NIOSH 1996). Occupational skin disorders are unevenly distributed among industries. A worker in agriculture, forestry, fishing, or manufacturing has three times the risk of developing a work-related skin disease as a worker in other industries.
Occupational infectious diseases
Naturally much attention about infectious diseases has focused on health care settings, although infections can be transmitted in other workplaces such as research laboratories and animal processing facilities. Within health care settings, awareness has grown about the risk of infection by hepatitis (especially hepatitis B and hepatitis C virus), HIV, and tuberculosis (Mycobacterium tuberculosis). Before the widespread use of hepatitis B virus vaccine, approximately 8700 acute cases of hepatitis B virus infection were reported among health care workers each year in the United States (NIOSH 1996). The risk of hepatitis B virus infection following a needle-stick injury with a contaminated needle varies from 2 to 40 per cent, depending on the antigen status of the source patient. The risk of hepatitis C virus transmission ranges from 3.3 to 10 per cent. The potential for hepatitis B or hepatitis C virus transmission is greater than that for HIV, but the modes of transmission are similar. An increased risk of HIV infection has been shown to exist in settings in which workers may be exposed to blood or body fluids (NIOSH 1996).
Transmission of M. tuberculosis is a recognized risk in health care facilities. After years of declining incidence rates, multidrug-resistant tuberculosis re-emerged as a major occupational health problem during the 1990s in major cities in the United States which serve populations with high rates of multidrug-resistant tuberculosis (CDC 1999). The magnitude of risk for health care workers varies by the type of health care facility, the prevalence of tuberculosis in the community, the patient population served, and the effectiveness of the infection control programme. The rates of multidrug-resistant tuberculosis in health care facilities decreased by the end of the decade following implementation of the OSHA standard (OSHA 1991) designed to prevent exposure to infectious agents and strengthening of workplace tuberculosis control programmes (CDC 1999).
Infectious diseases can be especially prevalent in developing countries, resulting in higher risks for workers in these countries. Some of the infections result directly from the work, while others are indirectly related to work. Examples include vector-borne diseases like malaria, water- and food-borne diseases resulting from poor sanitation and inadequate potable water, and zoonoses among agricultural workers.
Occupational reproductive disorders
The overall contribution of occupational exposures to reproductive disorders is not known because there has been little research in this area until recently. More than 1000 workplace chemical have shown reproductive effects in animal studies, but most have not been studied in humans (NIOSH 1996). Furthermore, virtually no studies have been done on physical and biological agents that may affect fertility and pregnancy outcomes. Epidemiological research on occupational hazards to reproduction has expanded in recent years (Lindbohm 1999).
It has been well documented that lead and the pesticide dibromochloropropane cause testicular injury with resultant reduction in sperm count. Also lead can cross the placenta in a pregnant woman worker to cause neurological impairment in the fetus. Other substances associated with adverse reproductive outcomes for which human evidence is strong include methyl mercury, solvents such as carbon disulphide, oestrogens, anaesthetic gases, ethylene oxide, polychlorinated biphenyls, and physical agents such as ionizing radiation (Frazier and Hage 1998).
Occupational exposures can cause a wide range of reproductive disorders in both males and females. Effects of exposure in males include altered sperm number, shape, or function, altered sperm transfer, and altered hormones or sexual performance—all of which may lead to subfecundity or impaired capability to conceive a viable child (Lemaster 1998). Exposures in females may cause menstrual disorders, infertility, chromosomal aberrations, breast milk alteration, early onset of menopause, and suppressed libido.
Reproductive disorders also include adverse effects on the offspring of the exposed worker. Potential fetal effects from maternal exposure include preterm delivery, fetal loss, prenatal death, low birth weight, altered sex ratio, congenital malformations, childhood malignancies, infant or childhood illness, and developmental disabilities (Lemaster 1998). Less is known about male-mediated exposure effects on the offspring.
Musculoskeletal injuries include both acute and chronic injury to the musculoskeletal system, other than acute trauma. These conditions are one of the leading problems affecting workers. In the United States, back disorders account for 27 per cent of all non-fatal occupational injuries and illnesses involving days away from work (NIOSH 1996). More than half of the working population develop low back injury at some time in their working career. Musculoskeletal injuries are the principal cause of disability of people in their working years.
The incidence of musculoskeletal disorders associated with cumulative or repetitive work has increased dramatically in recent years to become the most commonly reported occupational disease. In 1997, 276 600 musculoskeletal disorders due to repeated trauma were reported in American workplaces (BLS 1998). This figure represents 64 per cent of all reported occupational illness cases in the United States. The most frequently reported upper extremity disorders affect the hand or wrist area, including the most widely recognized condition—carpal tunnel syndrome.
Severe occupational traumatic injuries
These injuries include such events as amputations, fractures, severe lacerations, eye losses, acute poisonings, and burns. The National Institute for Occupational Safety and Health (NIOSH) estimates that at least 10 million persons in the United States suffer traumatic injuries at work each year; the average annual occupational fatality rate for the United States workforce is 7 per 100 000 workers (CDC 1993). On average, 17 workers died each day during 1998 in the United States (BLS 1999). Major causes of deaths were highway motor vehicle crashes (24 per cent), homicides (12 per cent), falls (10 per cent), caught in or compressed by equipment or objects (6 per cent), electrocutions (7 per cent), and being struck by falling objects (5 per cent). The largest number of fatalities occurred among truck drivers, construction trades, farm occupations, and sales occupations. Occupations with the highest rates of fatal injuries were fishermen, timber cutters and loggers, aeroplane pilots and navigators, structural metal workers, taxicab drivers, and construction workers. Homicide and violence in the workplace have received increasing attention as major causes of occupational fatalities. Homicide is the leading cause of work-related death for females. Homicide is the leading cause of occupational fatalities in some of the largest and fastest growing industry sectors—retail trade, services, and finance/insurance/real estate.
Occupational exposure to noise
Noise is a widespread problem that has substantial impact on the prevalence of hearing loss among the working population. Estimates indicate that 30 million people in the United States work at sites where the level of noise, 85 dB or higher, presents an increased risk of noise-induced hearing loss (NIOSH 1996). One worker in four exposed occupationally to 90 dB of noise over a working lifetime will develop a hearing impairment caused by noise.
Economic globalization and workers’ health
Rapid technological innovation and the proliferation of multinational organizations are driving the formation of a global economy that has a substantial impact on workers’ safety and health. Technological change is creating fundamental transformations in the way corporations organize production, trade goods, invest capital, and develop new products (CAETS 1988). Technology allows virtually instantaneous communication among widely dispersed operations. Advanced manufacturing technologies have changed patterns of productivity and employment. Improved air and sea transportation has greatly accelerated the flow of people and goods. These technological developments have created greater interdependence among firms and nations. At the same time, the rapid rate of innovation means that competitive advantages are fleeting and companies must function with ever increasing efficiency to survive in the global economy.
The strategy is for corporations to be agile and rapidly responsive to market demands (Menzies 1998). This strategy has led to concepts such as re-engineering, computer-integrated manufacturing, just-in-time manufacturing, and lean production. Quality circles, total quality management, and other ‘cultural training’ programmes train workers to identify with the competitive goals of management. New technologies have been implemented to increase productivity and make flexible work schedules possible. However, these technologies can also mean loss of control for workers, increased work speed, and more repetitious work—each of which has been associated with increased job stress (Schnall et al. 2000). Employment is both more flexible and less secure as corporations use technology to ensure that individual workers are dispensable and that they conform to the competitive needs of the corporation. Consequently, there has been a dramatic growth in contracted work and non-standard forms of employment such as part-time and home-based work. Shift work and irregular work hours have increased significantly among those who are employed. The contingent workforce, which includes self-employed, temporary, and part-time workers, was estimated to include 34 to 42 million workers in 1996, or more than one in four workers in the United States (Department of Labor 1997). These workers typically have less training about hazards, less access to occupational health services, and less access to other social services such as medical and unemployment insurance or programmes. It is difficult under these circumstances for traditional forms of labour protection, such as government regulations and representation by unions, to function efficiently.
The global economy has also led to shifts in the distribution of occupational hazards among regions of the world. In the industrially developed nations, the principal shift has been from a manufacturing-based economy to an economy that is based on the provision of services and the transfer of information. In consequence, exposure to classic hazards such as silica, asbestos, and heavy chemicals are becoming less important in these nations, while exposure to new synthetic materials and solvents, as well as the ergonomic exposures associated with repetitive work before computer terminals, have become more important (Mustard 1997). In developing nations, by contrast, major hazards have resulted from the export of dangerous industries, materials, and occupations from the industrially developed to the developing nations. In some instances, this export can lead to devastating disasters such as the explosion in the chemical plant at Bhopal, India, that killed several thousand people. Another example is the international boom in the microelectronics industry, which now employs hundreds of thousands of workers worldwide under poorly controlled and highly exploitative conditions, producing products primarily for use in developed nations (LaDou 1995).
The global economy has led to the negotiation of trade agreements, such as the North American Free Trade Agreement, which define conditions of work in the context of trade facilitation and barriers. In some cases, agreements have led to standards that raise the level of protection to workers in countries where previously such protection was minimal; however, in many cases, agreements have encouraged de-unionization and movement away from work protections in order to ‘harmonize’ protections at a low, but common level among trading partners (Armstrong 1998). A major challenge for nations and international organizations is to implement policies that balance the demands of the global market economy with appropriate protections for workers’ health and well being.
Special populations of workers
Recognition has increased that workplace hazards impact disproportionately on some worker populations—such as those in developing nations, as well as child labourers, women workers, and impaired workers (Frumkin and Pransky 1999; IPEC 1999). These populations are especially impacted because of the interaction between their work roles and broader roles in society, as well as by their particular exposures in the workplace. For example, workers in developing nations may be dually affected by hazards in the workplace and low sanitation in their communities. Children who work full time do not have access to education. Low literacy increases the potential that the children will be exposed to dangerous conditions in the workplace; at the same time, it is an obstacle to the children’s future economic security as adults. Women workers in virtually all societies are expected to maintain dual roles as workers and primary caregivers in the home. The full impact of work on the health of these populations must be understood in the broader context of their roles in society and in the workplace.
Workers in developing nations
Approximately eight of 10 workers in the global workforce are from the developing world (Jeyrathnam 1998). Workers’ health should be viewed in the context of national development. Occupational health policy-makers in many nations must consider a balance between adverse impacts on workers’ health and the economic advantages of rapid development by allowing foreign investigators access to low-cost labour and conditions of weak labour protections.
The relationship between workers’ health and development is complex for many reasons (Jeyrathnam 1998; Frumkin 1999). For example, workers in many developing countries may be affected by poor nutrition or endemic diseases, such as malaria, in which work may aggravate the condition, or which make the worker more susceptible to the effects of workplace exposures. Workers in these countries also generally have lower educational backgrounds and are often inadequately trained to handle the new technologies and potential hazards. There may be high turnover with little management investment in worker training. Consequently, workers may not be aware of health risks and safe practices.
Working conditions in developing countries may present special hazards because of tropical climates and poor, if any, building ventilation in production facilities. Much of the production equipment is imported from developed countries so that replacement parts and service may be unavailable. The machinery may be used or considered obsolete for use in the developed countries, while new and safer equipment may be unavailable or too expensive.
The social organization of work in developing countries also affects workers’ health (Frumkin 1999). In addition to the large number of workplaces with a small number of workers, large proportions of the workforce work in the ‘informal’ sector, which consists of small, often home-based, businesses that have no government registration and oversight. For example, estimates of informal sector employment range from 49 to 99 per cent of the approximately 235 million total labour force in Africa.
Finally, countries of the developing world may have access to advanced technologies from the developed world without having developed legal or administrative infrastructure to control their adverse impacts on the workforce (Jeyarathnam 1998). Even if developing countries adopt standards and legislation from more developed nations, there is often a shortage of trained personnel to recognize and manage workplace hazards (Frumkin 1999).
Children are an important population of workers worldwide. The ILO estimated that, in 1996, 250 million children 5 to 14 years old were engaged in economic activities worldwide—at least 120 million of them on a full-time basis (IPEC 1999). Africa has the highest incidence of child labour, with 41 per cent compared with 22 per cent in Asia and 17 per cent in Latin America. Child labour also exists in many industrialized countries. Child labour has become an important issue because the children are often exploited in the workplace and denied basic human rights, such as access to education (IPEC 1999). In addition, many children work in dangerous jobs and they may be more susceptible to workplace hazards (CDC 1996; Warshaw 1998).
Poverty is the primary reason why children work. Poor households need the money, and children commonly contribute around 20 to 25 per cent of family income. Furthermore, families may have a tradition of children following in their parents’ footsteps. If the family has a tradition of engaging in a hazardous occupation, it is likely that the children will continue in the trade.
The most common explanations about why employers hire children are the lower cost and specific skills afforded by children—the ‘nimble fingers’ argument. However, ILO research has concluded that the ‘nimble fingers’ argument is not valid (IPEC 1999). An actual reason is that employers believe that children are easier to manage because they are less aware of their rights, less troublesome, more compliant, more trustworthy, and less likely to be absent from work.
Many children work in hazardous occupations and are at greater risk of suffering ill effects than adult workers. These children may have greater exposure to hazards than adult workers in the same occupation because the children tend to be given the most menial jobs, which may involve higher exposures to toxic substances. Children are more susceptible to the same hazards faced by adult workers because they differ from adults in their anatomical, physiological, and psychological characteristics. Children using hand tools designed for adults run a higher risk of fatigue and injury. Personal protective equipment may also not fit and provide real protection. Furthermore, children may not be aware as adults of workplace dangers nor knowledgeable of the precautions to be taken at work (CDC 1996). Children are also more vulnerable to psychological and physical abuse than are adults, and suffer deeper psychological damage when they are denigrated or oppressed.
A resurgence of child labour is also occurring in developed nations. A Congressional Report documents a rise in the frequency of sweatshops employing children in the United States (General Accounting Office 1988). Each year in New York State, for example, more than 1000 children receive workers’ compensation awards for injuries incurred on the job; over 40 per cent of these awards each year are for permanent disability (Belville et al. 1993). The most important reason for the re-emergence of child labour in the United States is the increase in poverty. The number of American children living in poverty has more than doubled in many areas of the nation since the early 1980s (Landrigan 1993). During the same time, there was a relaxation in federal government enforcement of labour laws, such as the ban on home piecework in the garment and electronic industries. Studies indicate that the risk of injuries is 10 times higher in illegal, exploitative work than in legally permissible employment.
The issue of child labour has received increasing attention (Warshaw 1998; IPEC 1999). This is reflected in the number of organizations involved in the cause of children and child workers. For example, the International Programme on the Elimination of Child labour (IPEC) was launched in 1992 and, as of 1999, developed into a 90-country alliance. The aim of IPEC is the elimination of child labour, giving priority to its worst forms. The ‘worst forms’ comprise all forms of slavery or practices similar to slavery; the use, procurement, or offering of a child for prostitution or production of pornography; the use, procurement, or offering of a child for illicit activities; and work which is inherently likely to harm the health, safety, or morals of children (IPEC 1999). Priorities to end child labour were defined through the Convention on the Worst Forms of Child Labour in 1999 (Convention No. 182). Withdrawing children from the worst forms of child labour requires improved legislation and enforcement, improved methodologies for identifying the children, rehabilitation of the children, provision of viable alternatives to the children, and awareness raising at all levels of society.
Women are a special worker population because of the significant interplay between their roles in society, socio-economic condition, and occupation (Paltiel 1998). Women’s roles in virtually all societies are defined in relation to their reproductive functions and responsibilities as family caregivers. Paid employment of women has increased in most countries, but this employment has increased the conflict between paid work and women’s traditional family responsibilities. In many societies, early marriage, repeated child bearing, low education, and poverty all disproportionately impact on women workers (Loewenson 1999). The dual roles of women as workers and unpaid caregivers is especially challenging for sole-support mothers, who comprise 20 to 30 per cent of households worldwide.
Employment of women in most societies is characterized by occupational segregation, underemployment (doing seasonal and part-time work below their level of education), and barriers to advancement. Occupational segregation means that women tend to be clustered into a small number of occupations while being under-represented in most others (Stellman 1999). For example, professional women tend to be in teaching, nursing, and other health care specialties. In manufacturing, women tend to have jobs in assembly and small machine operations. Women in developing countries tend to be employed in sectors such as agriculture, textiles and clothing, food processing, and social services (Loewenson 1999). Compared with men, women work for smaller industries or organizations, are more often in informal work with little protection, have less opportunity for work control, and face the psychological demands of people-oriented or machine-paced work (Paltiel 1998). While some countries have enacted laws prohibiting gender discrimination, many countries have formal restrictions on women’s employment.
Gender differences are observed in the rates of occupational injuries and illnesses, but these differences are primarily because of differences in the conditions of work or exposures rather than being due to genetic differences (Stellman 1999). As noted above, women tend to work in different occupations than men with a different distribution of hazards. Even when employed in the same industry, women generally do different jobs or different tasks than men so their exposures may be different. Even when doing the same task, women may have different levels of exposure because of variation in the effectiveness of engineering controls and personal protective equipment—which are generally designed for men.
The actual risks of occupational injuries and illnesses to women is not known because a large proportion of women work in the informal sector, and because there has been inadequate research on women workers. Women have higher rates of repetitive strain injuries, especially carpal tunnel syndrome, than men (Stellman 1999). This difference is because women’s jobs typically involve more repetitive motion and more static effort. Women are also concentrated in health care occupations where there is greater risk of infections. In addition to toxic hazards, women also face sexual harassment and gender-based violence in the workplace.
There has been inadequate research on the effects of occupational hazards in women because much of past research has been done in industries for which women were largely excluded. It is possible to consider past research on male workers, but it may not be possible or justified to generalize the findings to women (Blair et al. 1999). For example, research on males cannot address the possibility of gynaecological disorders among women. It is also theoretically possible that there could be gender-specific responses, for example if the effects are hormonally mediated. Gender-specific effects have been seen for some carcinogens in animal studies (Blair et al. 1999). It is unclear whether gender-specific effects occur in humans, and so more attention must be given to studying occupational hazards in women.
Many people can make constructive contributions in the workplace although they have some type of physical impairment. North American employers, generally in response to legal requirements for workplace human rights, are developing positive policies and strategies for management of a diverse workforce, including impaired workers. The United States has developed probably the most comprehensive legislation for disabled workers, including legislation regarding their entitlements in education, employment and all other spheres of living (Paltiel 1998). Reasonable accommodations are changes made to the work environment, job responsibilities, or conditions of work that provide opportunities for workers with special needs to perform essential job functions. Reasonable accommodation can cover the special needs of persons with impairments or those workers with chronic or recurrent disease, including persons with AIDS. Accommodation may include technical assistance devices, customization including personal protective equipment and clothing, and changes to processes, location, or timing for essential job functions.
In the United Kingdom, the Disability Discrimination Act 1995 prohibits employers from discriminating against applicants and employees with disabilities. Employers also should make reasonable accommodations for a known impairment.
Recognition of occupational disease
Recognition is the key initial step in preventing and managing occupational injuries and diseases in workers. Associations between occupational exposures and disease are typically recognized in three ways: clinical observation, epidemiological analysis, and toxicological evaluation of chemical substances.
The alert clinician is the key to clinical recognition of occupational disease. Historical examples of occupational illnesses which have been recognized by alert clinicians include angiosarcoma of the liver in workers exposed to vinyl chloride monomer (Creech and Johnson 1974), lung cancer in workers manufacturing bis(chloromethyl)ether (Figueroa et al. 1973), bladder cancer in aniline dye workers (Rehn 1895), and mesothelioma in asbestos workers (Selikoff et al. 1964).
Keys to the recognition of occupational illness are that the clinician is alert to the possibility that any patient may have an occupational disease and therefore obtains an adequate occupational exposure history on all patients, possesses basic knowledge about the pathogenesis and clinical presentation of the major types of occupational disease, and knows how to report suspected cases of occupational illness to public health authorities so that additional cases caused by the same exposures can be recognized or prevented. Table 1 can be used as a guide to medical conditions or potential occupational exposures for which the clinician should elicit a detailed occupational medical history for evaluation.
The occupational history
The occupational history is the principal clinical instrument for the diagnosis of occupational disease. It may not be possible to obtain a detailed occupational history on every patient. However, the clinician should routinely ask screening questions of every patient that provide an indication as to whether a complete occupational history is warranted. At a minimum, every patient should be asked about his or her current job, and about the longest held previous jobs, by industry and occupation. A general question should be asked about occupational exposures to chemicals, fumes, gases, dust, noise, radiation, and other physical hazards at work. If the patient reports exposure to any of these agents, it may be useful to ask if he or she thinks that there is a health hazard at work. In addition to these screening questions, the clinician should pay attention during the medical history and review of systems to any temporal relationships reported between work and the onset of symptoms.
If information from the routine interview suggests an occupational aetiology, the physician should obtain a more detailed history of exposures. Data on duration and intensity of exposures are particularly important. It is necessary to learn how the patient worked with the suspected toxin and to consider all jobs ever held, places of employment, products manufactured, and materials with which the patient worked.
If toxic exposures are identified or strongly suspected and an occupational cause seems likely, further follow-up enquiries may need to be made through the patient’s labour union, companies where he or she has been employed, company physicians, or state or local health departments. Information on toxic substances used in a workplace may be legally available to patients under governmental ‘right-to-know’ laws.
All epidemiological study designs can be used to study occupational hazards, but some study designs are especially prominent in occupational epidemiology (Checkoway et al. 1989). Major strategies used by epidemiologists to recognize occupational diseases are the cross-sectional medical study, the proportional mortality study, the historical cohort mortality study, and secondary analysis of vital statistics and other population-based health data.
These studies, in which questionnaires are administered or physical examinations performed on a population of workers at a single point in time, are useful for the identification of acute short-latency conditions or stable conditions which do not result in workers leaving employment. A limitation inherent in cross-sectional studies is the difficulty of determining the temporal relationship between exposure and disease; review of exposure and medical records may be helpful in establishing the time sequence. Whenever possible, it is desirable to follow-up an initial cross-sectional evaluation and to continue clinical observations prospectively over time. Serial evaluations of a population of workers can provide extremely useful data about the development and causation of occupational disease.
Proportional mortality studies
These studies compare the pattern of causes of death among a group of workers with that in the general population or in another comparison population. They are relatively quick and inexpensive to perform, because they require information about only those employees who have died. This information is often available from pension or retirement plans. However, the proportional mortality study is susceptible to a variety of biases and should be considered a ‘hypothesis-generating’ approach preliminary to conducting more definitive cohort or case–control studies.
Historical cohort mortality studies
These studies are more common in occupational epidemiology than in other fields of epidemiology. This study design utilizes employment records to identify a cohort of workers at some time in the past. The subsequent mortality experience of the cohort is determined by reviewing death certificates and other sources of information. Cause-specific mortality rates are compared with those in the general population or with non-exposed members of the same cohort. The historical cohort mortality study is an important tool for establishing cause and effect associations between work exposures and fatal occupational disease and also for the quantitative assessment of occupational risks. These studies are particularly useful in worker populations, because these groups can generally be well defined through the use of employment records and seniority lists. Frequently, the limiting factor in historical cohort studies is the poor quality of the data on past exposures. However, nested case–control studies undertaken within a cohort can be used to examine past exposures in greater detail.
Epidemiological analysis of population-based health data
This approach can be useful for the surveillance of large populations of workers (such as all workers in a state) and for the recognition of new exposure–disease associations. An example is the use of population-based tumour registries to identify occupations with elevated risks of cancer, or of state-based vital record systems to assess occupation-specific factors, such as a high risk of death by electrocution in farmers. Data on occupational exposures in a registry may be limited simply to an occupation or industry code and investigators can group subjects with common exposures using the ‘job-exposure matrix’ technique (Hoar 1983). Case–control studies can then be performed on groups identified through the matrix to obtain detailed information from individuals about their past occupational exposures.
Toxicological analysis of chemical substances is an important means of assessing cause-and-effect relationships. The particular strength of toxicological analysis as a tool for disease prevention derives from the fact that it can precede occupational exposure. In contrast, medical and epidemiological studies can be undertaken only after exposure has already occurred. Thus chemicals found in laboratory tests to cause adverse health effects can be banned or strictly controlled to minimize the exposure of workers and community members.
Premarket testing is the most effective means of assessing the toxicity of new chemical compounds. However, until the passage of the Toxic Substances Control Act in 1976, there was no legal requirement in the United States for prospective evaluation of the toxicity of new industrial compounds. Many thousands of compounds whose introduction to commerce antedated passage of the Act remain untested. Also, testing procedures for new compounds have not been standardized; responsibility for deciding whether or not to test a new chemical and for the development of protocols for evaluating toxicity is left almost entirely to the discretion of manufacturers.
Research is an essential activity to further the recognition of occupational hazards. Governments are the single largest source of research funds, which are predominantly organized into national research programmes. In 1995, NIOSH initiated a nationwide planning process to guide occupational health research in the United States (NIOSH 1999). During the next few years, approximately 500 organizations and individuals provided input into the research agenda (Rosenstock et al. 1998). This process resulted in a framework for research and a list of priority research areas, which has been called the National Occupational Research Agenda. Priorities were identified by working groups with broad representation of employers, workers, safety and health professionals, public agencies, industry, and labour organizations. Criteria used to guide the evaluation of priorities included the following: seriousness of the hazard based on death, injury, disease, disability, and economic impact, number of workers exposed or magnitude of risk, potential for risk reduction, expected trend in importance of the research area, sufficiency of existing research, and probability that research will make a difference. The process identified 21 priority research areas which were divided into three categories: diseases and injuries, work environment and workforce, and research tools and approaches (Table 2). This framework has been used by NIOSH, other agencies, and private groups to prioritize funding for research. The planning process is now being used by a number of other countries to establish their research priorities (NIOSH 1999).
Table 2 United States National Occupational Research Agenda priority areas
At the international level, there are, in addition to sections of the ILO and the WHO, research institutions such as the European Joint Safety Institute and the International Agency for Research into Cancer which carry out international programmes of research in occupational safety and health.
Surveillance of occupational disease
Surveillance is the collection, analysis and dissemination of results for the purpose of prevention (Halperin 1996). Hazard surveillance provides a means of assessing toxic occupational exposures to a population and thus of assessing risk (Wegman and Froines 1985; Markowitz 1998). A hazard surveillance system identifies chemicals in use, the industries and occupations where they are used, and the extent and magnitude of worker exposure. It also provides a means of identifying changes in patterns of exposure and of noting emerging toxic hazards. Disease surveillance provides a means of assessing the amount and types of occupational disease, time trends, and distribution according to geography, industry, and occupation. These two types of surveillance complement each other. Each is an integral component of a complete occupational health surveillance system.
Occupational hazard surveillance
Accurate assessment of the extent and resultant risk to populations of exposure to toxic occupational chemicals requires determination of the following parameters in a hazard surveillance system:
identification of chemicals in use by industry
description of each of the industrial processes in which these chemicals are used
assessment of the number of workers exposed to particular substances by process
assessment of current exposure levels in various processes
identification of workplace settings in which there exists potential for increased risk owing to the synergistic effects of simultaneous exposures to several potential hazards
assessment of the toxicity of specific agents (based on animal, human, or short-term test data)
description and assessment of the effectiveness of controls to limit exposure.
Most countries do not have adequate systems for occupational hazard surveillance. The only systematic attempts in the United States to elicit most of the information described above were undertaken by NIOSH in national surveys conducted between 1972 and 1974 and between 1980 and 1983. The first of these surveys, the National Occupational Hazard Survey, was conducted in a sample of nearly 5000 industrial facilities across the United States. The second survey, the National Occupational Exposure Survey, was a geographically stratified probability sample of 4490 facilities covering nearly 2 million employees in over 500 different types of industries. The information provided by these two surveys has been valuable to estimate the number of workers exposed to specific agents in specific industries or occupations, but the information is now quite dated (Markowitz 1998). The fact that these surveys have not been repeated is evidence of the cost and effort involved in performing a large hazard surveillance survey.
Another approach to hazard surveillance in the United States has been use of the OSHA Integrated Management Information System (IMIS) (Markowitz 1998). The IMIS is a database of exposure measurements obtained in OSHA workplace inspections. The database has been used, for example, to identify the distribution and extent of exposure to lead (Froines et al. 1990) and wood dust (Teschke et al. 1999) in American industries. The IMIS data are limited, however, because OSHA workplace investigations are not conducted on a representative sampling basis. Furthermore, OSHA concentrates its limited inspection resources on relatively few agents and industries so not many hazards can be evaluated using IMIS (Markowitz 1998). Clearly more systematic and ongoing programmes for occupational hazard surveillance are needed.
Occupational disease surveillance
Occupational disease surveillance has made significant gains in the United States during the past decade (Halperin 1996; Markowitz 1998). NIOSH has implemented several new programmes including the National Occupational Mortality Surveillance (NOMS), the Adult Blood Lead Epidemiology and Surveillance (ABLES), and the Sentinel Event Notification Systems for Occupational Risks (SENSOR) programmes. In addition, the Bureau of Labor Statistics, United States Department of Labor, compiles data on workers’ compensation claims and has implemented the Census of Fatal Occupational Injuries based on legal reporting of work-related fatalities. These programmes show the various strategies and sources of data used for disease surveillance.
Under the NOMS programme, data from over 500 000 death certificates are collected annually from 23 states that record industry and occupation on the death certificate. These data are use to track occupation-specific conditions, such as pneumoconioses, and to identify occupation–disease associations using the proportional mortality study approach.
The ABLES programme obtains reports from over 25 states that have state-based registries of lead poisoning. These registries obtain most of the data from required reporting by medical laboratories of elevated blood lead levels.
The SENSOR programme is based on the concept of a sentinel health provider, which is a medical care provider or facility that is likely to provide medical care for workers. This subset of providers is then enrolled in an active occupational disease reporting system for a limited number of defined conditions—for example, silicosis (Maxfield et al. 1997), asthma (Jajosky et al. 1999), pesticides (Maizlish et al. 1995b), and carpal tunnel syndrome (Maizlish et al. 1995a). This approach is useful because specialist providers are more likely to recognize an occupational condition and it is not feasible to include all providers in an active reporting system.
Surveillance data reported by the Bureau of Labor Statistics are based on workers’ compensation claims (BLS 1998). However, as discussed above, there are significant limitations in relying on official reporting of work-related diseases for occupational health surveillance. Because of these limitations, the annual Census of Fatal Occupational Injuries reported by the Bureau of Labor Statistics compiles data from multiple federal, state, and local sources, including death certificates, workers’ compensation reports, reports to regulatory agencies, medical examiners reports, police reports, and even news reports (BLS 1999). Surveillance is more effective if the condition is discrete and clearly related to work, such as an on-the-job fatality, multiple sources of data are ascertained, and an active reporting system is implemented.
A limitation in these public health disease surveillance systems is that they rely on reporting mechanisms in which recognition of the problem occurs only after the fact. Primary and secondary prevention of occupational disease requires a more direct strategy in which disease surveillance is conducted in the workplace itself. Disease surveillance in the workplace uses the health history and the results of periodic physical and laboratory examinations of workers to estimate levels of exposure to toxins and to assess early effects of exposure. Direct surveillance in the workplace is valuable because many occupational standards are based on minimal amounts of human or animal data; thus, prior to the introduction of many chemicals to the workplace, it is unknown whether workers will be adequately protected. Also, health effects of workplace exposures may vary among workers depending on individual constitutional characteristics. Surveillance of individual workers in the workplace is therefore useful to identify unforeseen hazards and to protect workers who are at increased risk.
Prevention of occupational disease
Primary prevention of occupational disease requires the elimination or reduction of hazardous exposures. Such elimination of hazard is most efficiently accomplished prior to the release of a new chemical substance through premarket toxicological evaluation. Primary prevention with regard to chemicals requires either the elimination of toxic materials and their replacement by less hazardous substitutes or the use of tight processes and controls, such as complete enclosure or ventilation at the source of aerosol generation. Secondary prevention—the early detection of occupational disease in its presymptomatic stage where it can still be controlled or cured—is also feasible. It depends on the ability to identify work-related illness efficiently and effectively through screening workers at high risk for occupational disease using state-of-the-art biological markers. Tertiary prevention—the prevention of complication and disability of existing illness—depends on the development and wide application of appropriate diagnostic techniques for identification of persons with already established occupational illness. Prevention on all three levels requires solid information on the potential effects of specific occupational exposures, as well as data on the industries, occupations, and geographical areas in which hazardous substances are used. The hierarchy of strategies for preventing occupational diseases is shown in Table 3 (Schulte 1995).
Table 3 Hierarchy of strategies for the prevention of occupational disease
The most important prevention strategy is the primary prevention of exposure to toxic chemical, physical, or biological agents. Reductions in exposure can be accomplished by using the techniques listed below, in descending order of preference.
Elimination or substitution
Elimination of a hazardous material or substitution with a thoroughly evaluated less hazardous material is the most efficacious method of controlling a workplace hazard. In some cases, this method may also be the least expensive. Several examples of effective materials substitution occurred in the late 1970s when outbreaks of neurological disease were documented from exposure to the neurotoxins, methyl-n-butyl ketone, dimethylaminopropionitrile and Lucel-7 (Horan et al. 1985). The aetiology of these episodes of chemicallyinduced neurological illness was recognized in clinical epidemiological studies, later buttressed by results of toxicological investigations. In each case, the manufacturer discontinued use of the product following identification of the neurotoxic agent. A less hazardous product was substituted.
Selection of a less hazardous process or equipment also represents a meaningful control strategy. For example, substitution of a continuous process for an intermittent process almost always results in a decrease of exposure. Where an entire process does not need to be changed to reduce hazards, equipment substitution may achieve the desired reduction in exposure. An example is use of a degreaser with a low-speed hoist rather than dipping parts by hand.
The primary engineering controls used to reduce worker exposure to toxic substances are ventilation and process isolation or enclosure. Ventilation is one of the most effective and widely used control measures. Control of hazards by ventilation is usually further subdivided into two categories: local exhaust ventilation and general exhaust ventilation. The most effective approach for implementing ventilation controls is as follows: conduct an engineering study to evaluate sources of exposure; develop an engineering design; install a system based on the design; evaluate the completed system to ensure that the air contaminant has been effectively controlled.
Isolation is defined as the interposing of a barrier between a hazard and workers who might be injured or made ill by the hazard. Isolation may refer to storage of materials, such as flammable liquids, enclosure or removal of equipment to another area (such as noisy generators), or isolation of processes or of the workers themselves (e.g. by enclosing a sawmill worker in a sound-proof ventilated booth to protect him from noise and wood dust). For example, the petroleum industry uses automated remote processing in plants based on centralized computer control of process equipment. Thus workers are largely isolated from hazards except in maintenance operations and during process upsets.
Alteration of work practices can help to reduce exposure to hazards. A common example is wet-sweeping rather than dry-sweeping dust. Another example is vacuuming cotton lint off spinning machines rather than blowing it off with compressed air, a practice which creates airborne dust particles.
Administrative controls are methods of controlling total worker exposure by job rotation, work assignment, or time periods spent away from the hazard. With administrative controls, the level of exposure to the hazard is not diminished; instead, the duration of exposure is reduced and exposure is spread more widely among the workforce. For example, the current air standard for inorganic lead in the United States is 50 mg/m3 based on an 8-h day. A worker could permissibly be exposed to 100 mg/m3 for a total of 4 hours and then rotated to a job without lead exposure as an administrative control. The most common use in industry of administrative controls is to reduce overall noise exposure through rotation. Given the typical demands of production and the potential for misuse, administrative methods of controls are not an optimal mode of control.
Programmes for encouraging personal hygiene constitute another, although less efficient, approach to reducing exposure. In some instances, management may encourage or even require showers and a change to clean clothes at the end of the working day. Naturally, management should provide these showers, changing facilities, lockers, and work clothes if indicated; in fact, several American OSHA standards, such as the occupational lead standard, require management to provide such facilities.
A subtle but potentially important route of exposure is ingestion of toxic agents by eating, smoking, or applying cosmetics in the workplace. To prevent such exposure, management should provide separate eating facilities outside production areas. Workers should be encouraged to wash their hands before eating or smoking.
Use of personal protective equipment
Respirators, gloves, protective clothing, ear plugs, and muffs are all common forms of personal protective equipment in use throughout industry. They can play an important role, provided that carefully designed personal protective equipment programmes are in place and the equipment itself is frequently and regularly checked. It is important, however, to recognize that programmes of personal protection never constitute as efficient a means of protection as engineering or process controls. Personal protective equipment is intended to reduce exposures to toxic substances which have already been dispersed in the workplace as the result of inadequate ventilation or incomplete enclosure. The principal valid use of personal protective equipment as an approach to the prevention of occupational exposure is, therefore, during certain maintenance operations, in which static controls are not operating, or during process upsets. Unfortunately, programmes for personal protective equipment, such as respirator programmes, are often ill defined, given inadequate attention, and used instead of engineering controls, with poor maintenance of the necessary equipment.
Workers’ compensation is a legal system designed to provide income support, medical payments, and rehabilitation payments to workers injured on the job, as well as to provide benefits to survivors of fatally injured workers (Ison 1998). Essentially all industrialized countries and many others have workers’ compensation programmes (Barth 1995). However, the majority of countries in the developing world do not have such programmes. As of 1995, only 16 per cent of workers in Africa, 43 per cent in Latin America, and 23 per cent in Asia have protection from social security systems that include workers’ compensation (Sekimipi et al. 1995). Each country has a different system with varying approaches to benefits, coverage of workers and medical conditions, eligibility criteria, financing, and administration. With the exceptions of the United States, Canada, Australia, and Germany, the programmes are administered or overseen nationally by the central government. In the United States, each of the 50 states, as well as three federal jurisdictions, has an autonomous workers’ compensation system.
Internationally, workers’ compensation programmes are structured along three lines (Barth 1995; Ison 1998). Perhaps the most common are those programmes that are embedded in a country’s social security system. Since the country’s social insurance programmes are integrated, permanent disability and survivor benefits and medical benefits are paid at levels that do not distinguish significantly between a work-related and non-work-related injury. The benefits are similar and therefore there is little need for controversy about the cause of an injury or illness. Thus these systems have little litigation related to workers’ compensation. Another approach is one in which workers’ compensation is funded and administered separately from that of the social security scheme, but the two programmes are closely related. An example of such links is where workers’ compensation benefit is ended at the normal retirement age and old age benefits begin automatically. Germany has such an intermediate approach. At the other end of the continuum is the United States, where the social security programmes and the workers’ compensation programmes have almost no links.
Although the many national systems are distinct, they have several characteristics in common (Barth 1995; Ison 1998, Plumb and Cowell 1998). Virtually all programmes provide some protection from income loss because of workplace injury or disease. The costs of health care are provided either through the workers’ compensation programme or in conjunction with the country’s social security and health care systems. An important general characteristic is that workers’ compensation is a no-fault system. An injured worker does not need to prove that his or her injury was the result of employer negligence. For a worker to qualify for benefits, only three conditions usually must be met: (a) there must be an injury or illness; (b) the injury or illness must ‘arise out of and in the course of employment’; (c) there must be medical costs, rehabilitation costs, lost wages, or disfigurement. If the claim is accepted, medical care and rehabilitation expenses are fully covered; lost wages are partially reimbursed. In most instances, specific benefit formulas are prescribed by law. Employers are legally responsible for paying most of these benefits to injured workers. Employers pay these costs through insurance premiums, social security payroll taxes, or self-insurance. The programmes in most countries operate through public insurance, but private insurers exist for some types of benefit programmes, especially in the United States.
Workers’ compensation is structured theoretically as a no-fault system in order to minimize the amount of litigation that had developed under the common-law system used previously by workers seeking redress from employers for work-related injuries. Furthermore, workers’ compensation has wider coverage than the common law system in that injuries and illnesses are compensated even if they are only partially work related. Generally, diseases are considered eligible for compensation if occupational exposure is the sole cause of the disease, is one of several causes of the disease, is aggravated by or aggravates a non-occupational exposure, or hastens the onset of disability. However, in exchange for this wider coverage and the introduction of a less litigious system, covered workers generally are not allowed to sue employers through common law. They also are given lower awards than those given through juries in negligence suits and cannot seek compensation for ‘pain and suffering’ beyond their physical injury.
Although intended to minimize litigation, workers’ compensation systems do not actually eliminate the legal process. In some jurisdictions, such as most states in the United States, insurance carriers or employers have the right to contest a claim. The basis for contesting most claims is the question as to whether the injury is work related. Proof of work-relatedness is usually straightforward for acute injuries, and therefore relatively few claims for acute injury are contested. However, the great majority of claims for chronic occupational diseases are contested. The burden of proving that disease is occupational in origin lies with the worker. If a claim is contested, the worker must find a lawyer willing to represent him or her, and then identify physicians who can convince the referee who hears the case that the disease is work related. Although most of these cases are settled in favour of the injured worker, delay until settlement can be quite long. For example, the mean delay is 390 days for pneumoconiosis claims. Moreover, the injured worker may continue to work during this time, aggravating the injury. There is inherent difficulty in diagnosing an occupational disease and establishing a cause-and-effect relation, and therefore it may be difficult, if not impossible, to prove that a disease is work related. The legal process can be time-consuming and disheartening, discouraging the worker from filing a claim in the first place or encouraging an early settlement for substantially less than the defined benefit.
The amount of litigation associated with workers’ compensation is much less in Europe than in the United States. Although the system is distinct in each country, there are generic characteristics that account for the contrast with the American experience. One factor is a heavier reliance on a schedule or list of covered diseases. While the definition of covered conditions varies among the countries, the use of a schedule within each system establishes a presumption of work-relatedness and reduces the proportion of controverted claims (Lesage 1998). Another distinction is that European workers’ compensation systems do not utilize the adversarial system to resolve disputes, such as those involving questions of aetiology or extent of impairment. The common practice is to hold hearings to find the facts without using lawyers and often without the private insurer or employer challenging a claimant’s position. Challenges to claims may arise, but they are handled by a governmental social insurance agency rather than by the private parties. A third distinction from the United States is that in European nations the determination of compensation in technically difficult claims is generally based on the position taken by professionals (such as physicians) who are employees or regular consultants to the compensation agency and not witnesses hired by the plaintiff or defendant. Thus the compensation administrator is not forced to reconcile different technical perspectives that may reflect the differing interests of the contending parties.
The most important difference between the workers’ compensation experience in Europe and the United States is the availability in Europe of relatively generous alternatives to and supplements of workers’ compensation benefits. In most of the world’s industrialized countries, workers are entitled to publicly provided or required health care benefits regardless of the work-relatedness of the condition. Furthermore, income-maintenance programmes because of disability from illness or injury are common. The availability of universal medical care and strong social insurance programmes provides alternative sources of support that reduce the dependence of the injured worker on the workers’ compensation system. Consequently, the determination as to whether an illness is work related is not critical to the worker’s health and social security needs.
A concern expressed often about integrating workers’ compensation programmes into broader social security programmes is that employers may have less incentive to provide a safe workplace. However, it is unclear whether workers’ compensation costs, even in jurisdictions in which the employer is legally accountable, actually function as an effective incentive for employers to prevent the occurrence of occupational diseases. In developed countries, the costs on average to employers of workers’ compensation are only about 2 to 3 per cent of payroll costs (Barth 1995). Consequently, a workers’ compensation system at best can play a small role in discouraging unsafe or hazardous employer practices. The ideal system would provide both benefits for the injured worker and strong incentives for employers to prevent occupational injury and illness, but the first priority of a workers’ compensation system should be to ensure prompt provision of full benefits to the injured worker.
Right to know
The ‘right-to-know’ concept refers to the mandatory sharing of information regarding workplace exposure to toxic substances between employers and workers, regulatory agencies, and in some cases communities near a workplace. The fundamental assumption in the right-to-know concept is that this transfer of information will prompt activity that will improve worker health (Ashford and Caldart 1985). In fact, there have been several instances of workers themselves playing a direct role in the discovery of occupational health problems. Two examples are the discovery of lung cancer in workers exposed to bis(chloromethyl)ether (Figueroa et al. 1973) and sterility in workers exposed to dibromochloropropane (Whorton et al. 1977). Until recently, however, workers have remained largely ignorant of the potential hazards of the chemicals with which they work.
The right of workers to know about potential hazards necessarily implies a corresponding duty on employers to provide that information. Employers’ duties can be considered in three categories. Firstly, the duty to generate or retain information means that an employer would be required to perform environmental or medical monitoring and to retain the records pertaining to that monitoring for a specified period of time. This duty is specified under some of the OSHA comprehensive standards, such as those for asbestos and lead. Secondly, the duty to disclose information on request means that an employer must provide copies of exposure or biological monitoring data to a worker or worker representative if that information is requested. For example, the OSHA Access to Employee Exposure and Medical Records Standard attempts to ensure that exposure, medical, and biological monitoring records are preserved and that workers or their representatives have access to them. Thirdly, the duty to inform refers to an employer’s or manufacturer’s obligation to disclose information about potential toxic substances in the workplace. Under the OSHA Hazard Communication Standard, employers have a duty to inform workers of the identity of the substances with which they work through labelling the product containers and disclosing the source of supply through the use of Material Safety Data Sheets. The standard also requires that workers must be trained in methods to detect the presence of hazardous chemicals, the hazards of the chemicals, and protective measures.
Workplace training programmes must be implemented effectively in order to assure workers of their ‘right to know’. For example, Kahan et al. (1999) evaluated programmes established under workers’ right-to-know regulations in Israel. They interviewed 552 workers and 33 safety officers employed at 50 industrial plants. They found that most of the worker’s knowledge about work hazards was based on informal sources, and not on those stipulated by the regulations. Furthermore, 5 per cent of the workers were unable to read and another 22 per cent had educational levels below that necessary to understand technical material provided by the employer. In more than one-third of the cases, the workers and their safety officers disagreed about the existence of hazards in the workplace. This research demonstrates that employers must be more aware of the need to identify hazards and communicate effectively to their employees in understandable language and terms.
Workers have additional rights to information about toxic hazards through other federal statutes and regulations. For example, the Toxic Substances Control Act in the United States imposes requirements on chemical manufacturers and processors to develop health effects data. The Toxic Substances Control Act requires testing, premarket manufacturing notification, and reporting of information. Unfortunately, as noted above, the great majority of chemicals in commercial use were able to reach the market before requirements for premarket testing were promulgated. The Act does impose a duty to disclose to the Environmental Protection Agency any information which supports the conclusion that a substance or mixture presents a substantial risk of injury to health. Thus medical screening or biological monitoring data obtained by an employer indicating a substantial risk of injury must be reported to the Environmental Protection Agency. The National Labor Relations Act also provides a mechanism by which workers can gain access to information about hazardous working conditions. Since the National Labor Relations Act provides employees with a limited right to refuse hazardous work, it has been interpreted to mean that employees must be informed about those hazards. In addition, access is available to employees who are members of unions through the collective bargaining process. It has been held that unions have a right of access to exposure and medical records, so that they can bargain effectively with the employer regarding conditions of employment. Through legal avenues such as these, workers’ access to information about potential hazardous exposures in the workplace has been expanded.
Workers suffer a broad range of injuries and illnesses caused by hazards encountered in the workplace. Despite the existence of protective legislation in many countries, the burden of injury and illness on workers remains significant. It is essential for medical practitioners and public health programmes to recognize, prevent, and manage work-related injuries and illnesses. There is a need for international co-ordination of occupational health protection for workers, given the increasing globalization of the world economy. Several approaches have been proposed to address this issue. For example, there should be harmonization of health, safety, and environmental standards in a way that does not unfairly impose a competitive disadvantage on the newly industrialized nations. Governments and multinational corporations should share the most advanced technologies and resources. Rather than allowing companies to manufacture products banned for use in their own country, governments in developed nations should provide financial incentives for their industries to develop and export safer products and technologies. At a minimum, international systems should be established to ensure complete notification of potential hazards, including labelling the contents of raw materials and products.
Adams, R.M. (ed.) (1999). Occupational skin disease (3rd edn). W.B. Saunders, Philadelphia, PA.
Armstrong, P. (1998). Transformation in markets and labour. In Encyclopaedia of occupational health and safety (ed. J. Stellman) (4th edn), pp. 24.15–17. International Labour Office, Geneva.
Ashford, N.A. and Caldart, C.C. (1985). The ‘right-to-know’. Toxics information transfer in the workplace. Annual Review of Public Health, 6, 383–401.
Barth, P.S. (1995). Compensating workers for occupational diseases: an international perspective. International Journal of Occupational and Environmental Health, 1, 147–58.
Belville, R., Pollack, S.H, Godbold, J. H., and Landrigan, P.J. (1993). Occupational injuries among working adolescents in New York State. Journal of the American Medical Association, 269, 2754–9.
Blair, A., Hoar Zahm, S., and Silverman, D.T. (1999) Occupational cancer among women: research status and methodologic considerations. American Journal of Industrial Medicine, 36, 6–17.
BLS (Bureau of Labor Statistics) (1998). Workplace injuries and illnesses in 1997. Bureau of Labor Statistics, US Department of Labor, Washington, DC.
BLS (Bureau of Labor Statistics) (1999). National census of fatal occupational injuries, 1998. Bureau of Labor Statistics, US Department of Labor, Washington, DC.
CAETS (Council of Academies of Engineering and Technological Sciences) (1988). Globalization of technology: international perspectives. National Academy Press, Washington, DC.
CDC (Centers for Disease Control and Prevention) (1993). Fatal injuries to workers in the United States, 1980–1989: a decade of surveillance. CDC, NIOSH, Cincinnati, OH.
CDC (Centers for Disease Control and Prevention) (1996). Work-related injuries and illnesses associated with child labor—United States, 1993. Morbidity and Mortality Weekly Report, 45, 464–8.
CDC (Centers for Disease Control and Prevention) (1999). Progress toward the elimination of tuberculosis—United States, 1998. Morbidity and Mortality Weekly Report, 48, 732–6.
Checkoway, H., Pearce, N.E., and Crawford-Brown, D.J. (1989). Research methods in occupational epidemiology. Oxford University Press, New York.
Collegium Ramazzini (1999). Call for an international ban on asbestos. American Journal of Industrial Medicine, 36, 227–9.
Creech, J.L., Jr and Johnson, M.N. (1974). Angiosarcoma of liver in the manufacture of polyvinyl chloride. Journal of Occupational Medicine, 16, 150–1.
P>Department of Labor (1997). Current population survey, 1996. US Department of Labor, Washington, DC.
Figueroa, W.G., Raszkowski, R., and Weiss, W. (1973). Lung cancer of chloromethyl methyl ether workers. New England Journal of Medicine, 228, 1096–7.
Frazier, L.M. and Hage, M.L. (ed.) (1998). Reproductive hazards of the workplace. Wiley, New York.
Froines, J.R., Baron, S., Wegman, D.H., and O’Rourke, S. (1990). Characterization of airborne concentrations of lead in US industry. American Journal of Industrial Medicine, 18, 1–17.
Frumkin, H. (1999). Across the water and down the ladder: occupational health in the global economy. Occupational Medicine, 14, 637–63.
Frumkin, H. and Pransky, G. (ed.) (1999). Special populations. Occupational Medicine, 14.
General Accounting Office (1988). Sweatshops in the US. Opinions on their extent and possible enforcement options. Publication No.GAO/HRD-88–130BR, General Accounting Office, Washington, DC.
Goldman, R.H., Rosenwasser, S., and Armstrong, E. (1999). Incorporating an environmental/occupational medicine theme into the medical school curriculum. Journal of Occupational and Environmental Medicine, 41, 47–52.
Halperin, W.E. (1996). The role of surveillance in the hierarchy of prevention. American Journal of Industrial Medicine, 29, 321–3.
Hoar, S. (1983). Job exposure matrix methodology. Journal of Toxicology. Clinical Toxicology, 21, 9–26.
Horan, J.M., Kurt, T.L., Landrigan, P.J., Melius, J.M., and Singal, M. (1985). Neurologic dysfunction from exposure to 2-t-butylazo-2-hydroxy-5-methylhexane (BHMH): a new occupational neuropathy. American Journal of Public Health, 75, 513–17.
IARC (International Agency for Research on Cancer) (1997). Working group on the evaluation of carcinogenic risks to humans: silica, some silicates, and coal dust and para-aramid fibrils. IARC, Lyon, France.
IOM (Institute of Medicine) (1988). Role of the primary care physician in occupational and environmental medicine. National Academy Press, Washington DC.
IPEC (International Programme on the Elimination of Child Labour) (1999). IPEC action against child labour – achievements, lessons learned and indications for the future (1998–1999). ILO, Geneva.
Ison, T.G. (1998). Workers’ compensation systems: overview. In Encyclopaedia of occupational health and safety (ed. J. Stellman) (4th edn), pp. 25.2–14. ILO, Geneva.
Jajosky, R.A., Harrison, R., Reinisch, F., et al. (1999). Surveillance of work-related asthma in selected U.S. states using surveillance guidelines for state health departments—California, Massachusetts, Michigan, and New Jersey, 1993–1995. Morbidity and Mortality Weekly Report, 48, 1–20.
Jeyarathnam, J. (1998). Occupational health trends in development. In Encyclopaedia of occupational health and safety (ed. J. Stellman) (4th edn), pp. 20.2–4. ILO, Geneva.
Kahan, E., Lemesh, C., Pines, A., Mehoudar, O., Peretz, C., and Ribski, M. (1999). Workers’ right-to-know legislation: does it work? Occupational Medicine, 49, 11–15.
LaDou, J. (ed.) (1995). Special Issue: international issues in occupational health. International Journal of Occupational and Environmental Health, 1, 76–222.
LaDou, J. (1997). The practice of occupational medicine. In Occupational and environmental medicine (ed. J. LaDou) (2nd edn), pp. 1–5. Appleton and Lange, Stamford, CT.
Landrigan, P.J. (1993). Child labor: a re-emergent threat. American Journal of Industrial Medicine, 24, 267–8.
Landrigan, P.J. and Baker, D.B. (1991). The recognition and control of occupational disease. Journal of the American Medical Association, 266, 676–80.
Landrigan, P.J., Nicholson, W.J., Suzuki, Y., and LaDou, J. (1999). The hazards of chrysotile asbestos. A critical review. Industrial Health, 37, 271–80.
Leigh, J.P., Markowitz, S.B., Fahs, M., Shin, C., and Landrigan, P.J. (1997). Occupational injury and illness in the United States. Estimates of costs, mortality, and morbidity. Archives of Internal Medicine, 157, 1557–68.
Lemaster, G.W. (1998). Occupational exposures and effects on male and female reproduction. In Environmental and occupational medicine (ed. W. Rom) (3rd edn), pp. 223–44. Lippincott–Raven, Philadelphia, PA.
Lesage, M. (1998). Work-related diseases and occupational diseases: the ILO international list. In Encyclopaedia of occupational health and safety (ed. J. Stellman) (4th edn), pp. 26.2–6. International Labour Office, Geneva.
Lindbohm, M.-L. (1999). Women’s reproductive health: some recent developments in occupational epidemiology. American Journal of Industrial Medicine, 36, 18–24.
Loewenson, R.H. (1999). Women’s occupational health in globalization and development. American Journal of Industrial Medicine, 36, 34–42.
Maizlish, N., Rudolph, L., Dervin, K., and Sankaranarayan, M. (1995a). Surveillance and prevention of work-related carpal tunnel syndrome: an application of the Sentinel Event Notification System for Occupational Risks. American Journal of Industrial Medicine, 27, 715–29.
Maizlish, N., Rudolph, L., and Dervin, K. (1995b). The surveillance of work-related pesticide illness: an application of the Sentinel Event Notification System for Occupational Risks (SENSOR). American Journal of Public Health, 85, 806–11.
Markowitz, S.B. (1998) The role of surveillance in occupational health. In Environmental and occupational medicine (ed. W. Rom) (3rd edn), pp. 19–29. Lippincott–Raven, Philadelphia, PA.
Maxfield, R., Alo, C., Reilly, M.J., et al. (1997). Surveillance for silicosis, 1993—Illinois, Michigan, New Jersey, North Carolina, Ohio, Texas, and Wisconsin. Morbidity and Mortality Weekly Report, 46, 13–28.
Menzies, H. (1998). Globalizing technologies and the decimation/transformation of work. In Encyclopaedia of occupational health and safety (ed. J. Stellman) (4th edn), pp. 24.17–21. International Labour Office, Geneva.
Mustard, F. (1997). The economy and social equity in a period of major technoeconomic change. Scandinavian Journal of Work, Environment and Health, 23 (Supplement 4), 10–15.
National Research Council (1987). Counting injuries and illnesses in the workplace. Proposals for a better system (ed. E.S. Pollack and D.B. Keimig). National Academy Press, Washington DC.
NIOSH (National Institute for Occupational Safety and Health) (1996). National occupational research agenda update. DHHS (NIOSH) Publication No. 96-115, NIOSH, Cincinnati, OH.
NIOSH (National Institute for Occupational Safety and Health) (1999). National occupational research agenda—May 1999. DHHS (NIOSH) Publication No. 99-124, NIOSH, Cincinnati, OH.
OSHA (Occupational Safety and Health Administration) (1991). Occupational exposure to bloodborne pathogens. Final rule: 29 CFR 1010.1030, 6 December 1991. OSHA, Washington, DC.
OSHA (Occupational Safety and Health Administration) (1996). OSHA fact sheet. Silica dust exposures can cause silicosis. Fact Sheet OSHA 96–54, OSHA, Washington, DC.
Paltiel, F. (1998). Shifting paradigms and policies. In Encyclopaedia of occupational health and safety (ed. J. Stellman) (4th edn), pp. 24.2–5. International Labour Office, Geneva.
Pearce, N., Boffetta, P., and Kogevinas, M. (1998). Cancer—introduction. In Encyclopaedia of occupational health and safety (ed. J. Stellman) (4th edn), pp. 2.1–18. International Labour Office, Geneva.
Perfetti, L., Cartier, A., Ghezzo, H., Gautrin, D., and Malo, J.L. (1998). Follow-up of occupational asthma after removal from or diminution of exposure to the responsible agent: relevance of the length of the interval from cessation of exposure. Chest, 114, 398–403.
Peto, J., Decarli, A., La Vecchia, C., Levi, F., and Negri, E. (1999). The European mesothelioma epidemic. British Journal of Cancer, 79, 666–72.
Plumb, J.M. and Cowell, J.W.F. (1998) An overview of workers’ compensation. Occupational Medicine, 13, 241–72.
Rehn, L. (1895). Blasengeschwuelste bei Fuchsinarbeitern. Archiv für Klinische Chirurgie, 50, 588.
Rom, W. (ed.) (1998). Environmental and occupational medicine (3rd edn). Lippincott–Raven, Philadelphia, PA.
Rosenstock, L., Olenec, C., and Wagner, G.R. (1998). The National Occupational Research Agenda: a model of broad stakeholder input into priority setting. American Journal of Public Health, 88, 353–6.
Ross, D.J. and McDonald, J.C. (1998). Health and employment after a diagnosis of occupational asthma: a descriptive study. Occupational Medicine, 48, 219–25.
Schnall, P.L., Belkic, K., Landsbergis, P., and Baker, D. (ed.) (2000). The workplace and cardiovascular disease. Occupational Medicine, 15, 1–334.
Schulte, P.A. (1995). Introduction: the role of biomarkers in the prevention of occupational disease. In Biomarkers and occupational health—problems and perspectives (ed. M.L. Mendelsohn, J.P. Peeters, and M.J. Normandy), pp. 1–6. Joseph Henry Press, Washington, DC.
Sekimipi, D.K., Tambellini, A.T., and Trung, L.V. (1995). The global economy and occupational health. International Journal of Occupational and Environmental Health, 1, 76–9.
Selikoff, I.J., Churg, J., and Hammond, E.C. (1964). Asbestos exposure and neoplasia. Journal of the American Medical Association, 188, 22.
Stellman, J. (ed.) (1998). Encyclopaedia of occupational health and safety (4th edn). International Labour Office, Geneva.
Stellman, J.M. (1999). Women workers: the social construction of a special population. Occupational Medicine, 14, 559–80.
Teschke, K., Marion, S.A., Vaughn, T.L., Morgan, M.S., and Camp, J. (1999). Exposures to wood dust in US industries and occupations, 1979 to 1997. American Journal of Industrial Medicine, 35, 581–9.
Venables, K.M. and Chan-Yeung, M. (1997). Occupational asthma. Lancet, 349, 1465–9.
Warshaw, L. (1998). Precarious employment and child labour. In Encyclopaedia of occupational health and safety (ed. J. Stellman) (4th edn), pp. 24.9–15. International Labour Office, Geneva.
Wegman, D.H. and Froines, J.R. (1985). Surveillance needs for occupational health. American Journal of Public Health, 75, 1259–61.
Wright, F.B. (1998). Occupational health and safety: the European Union. In Encyclopaedia of occupational health and safety (ed. J. Stellman) (4th edn), pp. 23.31–34. International Labour Office, Geneva.
WHO (World Health Organization) (1997). Executive summary. Health and environment in sustainable development: five years after the Earth Summit. WHO, Geneva.
WHO (World Health Organization) (1999). Occupational health: ethically correct, economically sound. Fact sheet 84, WHO, Geneva.
Whorton, D., Krauss, R.M., Marshall, S., and Milby, T.H. (1977). Infertility in male pesticide workers. Lancet, ii, 1259–61.