Article Text


Occupational medicine: at a turning point or an expansion
  1. A Blair
  1. Correspondence to:
 Dr A Blair
 Occupational and Environmental Epidemiology Branch, Division of Cancer Cause and Prevention, National Cancer Institute, Executive Plaza South, Room 8118, Bethesda, MD 20892, USA;

Statistics from

Commentary on the paper by Coggon (see page 281)

David Coggon, in his editorial on “Occupational medicine at a turning point”,1 makes a number of insightful observations and thoughtful suggestions regarding health and hazards in the workplace. He notes that “As many of the most serious occupational hazards have been successfully addressed, attention has shifted increasingly to other work related disorders that are rarely fatal”. He points out that the human and economic costs of these less fatal conditions are considerable and that the current research and intervention models may require some rethinking to deal effectively with them. I found his discussion and recommendations on this topic timely, and they will help the expansion of occupational research into this new area.

Considerable progress has been made on reducing and eliminating many serious occupational hazards, particularly in developed countries. It may, however, be premature to assume that chemical and physical hazards are under control. The understanding of occupational exposures in the development of chronic diseases and conditions is far from complete. Cancer has received much more attention than neurological, immunological, reproductive, and developmental outcomes, but even for cancer, there are areas where the literature is quite limited. However, for cancer, I think a compelling case can still be made for the need for further work on occupational risk factors. First, occupational exposures and cancer among women, minorities, and workers in small businesses and in developing countries have not been studied nearly as intensively as among white men in developed countries. Although the impact of some, maybe most, occupational exposures may not differ among population subgroups, we need a considerably firmer database before we conclude that findings for one group can stand for all. Second, the changing nature of the workforce, manufacturing processes, and the geographical location of many industries further underscores the need for future research among these under-investigated groups. Third, although some 30 different substances are well recognised as occupational carcinogens, there are hundreds more under suspicion where the available evidence is inconclusive. Fourth, there are many occupations where cancer appears to be excessive, but where the hazardous agent has not been identified. Finally, we have barely started to evaluate risks from multiple exposures. These limitations apply even more strongly to non-malignant diseases.

Occupational diseases are sometimes portrayed as only a minor contributor to the disease burden of the population. The 1981 estimate by Doll and Peto2 that 4% of cancer deaths in the United States may be due to occupational causes is still widely accepted and is probably a reasonable estimate for other developed countries. This 4% estimate for cancer is sometimes used to argue that occupational exposures are too unimportant to receive much research attention given finite resources. There are a number of arguments against this assertion. First, only tobacco and diet account for a radically different proportion of cancer deaths at 30% and 35%, respectively. According to Doll and Peto, attributable proportions from other established risk factors are 3% for alcohol, 7% from reproductive behaviours, 2% from pollution, 1% from medicines and medicinal procedures, 3% from geophysical factors, and perhaps 10% from infections. Yet these factors are seldom dismissed as unworthy of further investigation. Second, as pointed out by Doll and Peto, occupational exposures are unevenly distributed across the workforce. According to the 1950 census, about 18% of the employed population held blue-collar jobs; this has fallen to 13% in 2000. Since blue-collar occupations are where most of the hazardous exposures are likely to occur, it stands to reason that this is also where the bulk of occupationally related cancer arises. Thus, the proportion of cancer in blue-collar workers due to occupational exposures is probably 5–8 times that of the overall population estimate. Thus, the proportion due to cancer in blue-collar workers could be 20–30%. Finally, occupational exposures are largely involuntary and can be controlled. Availability of calculations of proportion of cancer caused by occupational exposures in the literature underscores the amount of scientific attention cancer has received, and the more limited effort and clear need for further research on occupational causes of other chronic diseases.

Dr Coggon suggests that the involvement of societal/cultural beliefs in the development of illnesses (conditions that are not a pathological state but an “absence of wellbeing”) indicates that if control of exposures is not effective in reducing the occurrence of the illness, modification of cultural beliefs and expectations offers another approach. The dual approach may also have relevance to chronic diseases. There are a number of examples where such psychosocial or cultural factors interact with traditional aetiological agents to modify risk of infectious and chronic diseases.3 The recognition that both factors may be involved enhances preventive opportunities because control over illnesses could be achieved by effective manipulation of either impetus. As indicated by Dr Coggon, epidemiological investigations can, and should, evaluate this possibility.

Commentary on the paper by Coggon (see page 281)


View Abstract


  • Competing interests: none declared

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Linked Articles