Hazards magazine * online briefing * 25 June 2007

Burying the evidence
How the UK is prolonging the
occupational cancer epidemic

Professor Andrew Watterson
Head of the Occupational and Environmental Health Research Group, University of Stirling.

Professor Rory O’Neill
Occupational and Environmental Health Research Group, University of Stirling; Editor, Hazards magazine

Executive summary

The UK authorities are failing to acknowledge or deal effectively with an epidemic of work-related cancers. The government’s Health and Safety Executive is underestimating the exposed population, the risks faced as a result of those exposures and the potential for prevention.

HSE fails to recognise the social inequality in occupational cancer risk, which is concentrated in skilled and unskilled manual workers and lower employment grades, or the greater likelihood these groups will experience multiple exposures to work-related carcinogens at work and in the wider environment. It cannot quantify and continues to neglect the largely uninvestigated and unprioritised risk to women.

HSE’s approach includes neither a requirement nor a strategy for the reduction in the number and volumes of cancer-causing substances, processes and environments at work. Its recommendations for action range from complacent to non-existent.

The result is that the UK faces at least 20,000 and possibly in excess of 40,000 new cases of work-related cancer every year, leading to thousands of deaths and an annual cost to the economy of between £29.5bn and £59bn (1). Preventing just 100 of these cancer deaths would more than offset the entire Health and Safety Executive annual budget.

This report outlines 20 key flaws in the HSE’s approach and makes recommendations to address effectively the UK’s occupational cancer crisis.

1. Under-estimating the epidemic

The UK government’s workplace safety agency, the Health and Safety Executive, greatly under-estimates the number of workers exposed to a workplace cancer risk and the numbers developing occupational cancer. HSE still presents findings of a 1981 US study as “the best overall estimate available” (2). This is the basis of its estimate of just 6,000 occupational cancer deaths each year.

The findings of HSE’s preferred source, the 1981 Doll/Peto report (3), have been disputed and described as “discredited”, with authors criticising both errors in methodology that led to a substantial under-estimate of the true incidence (4) and the pro-industry leanings of the lead author (5). Recent analyses suggest the real number of work-related cancer deaths in the UK each year is at least 12,000 cases and could be as high as 24,000 (6). New figures from the International Labour Organisation support claims that the occupational cancer rate in developed nations is substantially higher than HSE’s estimate (7). Basing official policies on Doll/Peto estimates has resulted in a chronic failure to secure either the resources or the priority required for meaningful preventive action.

Exposure to a workplace cancer risk is not a minor concern affecting few people. The European Union’s CAREX database of occupational exposures to carcinogens concluded: “According to the preliminary estimates, there were circa 5 million workers (22 per cent of the employed) exposed to the agents covered by CAREX in Great Britain in 1990-93. The number of exposures was circa 7 million” (8). Other recent studies have suggested the at risk group may in fact be increasing. Even by the CAREX estimate, over a fifth of the UK workforce has been exposed to possible human carcinogens and for these workers most of the resultant cancers will only emerge in a couple of decades or more.

Where HSE acknowledges there is a risk, its new estimates of the at risk working population seem designed to downplay the problem. In the case of cancer-causing beryllium, for example, HSE in 2003 said 250 workers in the UK were estimated to be continuously exposed and 1,000 workers occasionally exposed to “very low concentrations of beryllium or beryllium oxide.” Its 2007 estimates say there are fewer than 1,000 exposed in under 100 workplaces; in other materials HSE says beryllium use in increasing. HSE also ignores exposures to beryllium in the scrap metal and recycling industries. Even collectors of old radio sets take a more concerned stance, flagging up risks of beryllium exposures when dismantling radio sets.

2. Safety oversight

HSE’s under-resourced inspectorate is not capable of ensuring adequate safety oversight of Britain’s workplaces. This was highlighted by the recent “world’s largest” outbreak of Extrinsic Allergic Alveolitis and occupational asthma, in which over 100 workers at Powertrain in Birmingham developed serious and chronic occupational diseases, despite the risks being well-established over a decade earlier (9). The poorly controlled agent responsible, metalworking fluid, is also an occupational cancer risk.

Under-reporting of conditions including chrome ulcers – a warning sign of exposures that could lead to chromium related lung cancers - indicate HSE’s intelligence on many occupational carcinogen exposures is lacking. The proliferation of small firms means more workers are likely now to be working in firms with inadequate systems to recognise and deal with risks, and will remain almost entirely off HSE’s radar. Workplaces enforced by HSE can expect an inspection only once every 13 years, or three times in a person’s working lifetime. This is about half the inspection frequency seven years ago (10).

HSE initiatives to address occupational cancers are a small component of an HSE disease reduction programme, which is a small part of the HSE ill-health reduction programme, which is small part of the HSE Fit3 programme. This is not enough.

3. Estimates of chemical use

HSE’s estimates of those at risk of and developing occupational cancers fail to take adequate account of the rapid increase in the numbers and volumes of substances used in the workplace. About 100,000 chemicals are in industrial use, an estimated 30,000 used in the EU in high volumes (manufactured or imported in volumes over 1 tonne).

The UK Chemical Industries Association in 2006 reported that over the last decade the chemical industry had grown more than five times faster than the average for all industries. It noted “the chemical industry accounts for 2 per cent of UK GDP and 11 per cent of manufacturing industry’s gross value added. Turnover, which includes the sales of merchanted goods, eg. chemicals imported and then re-sold, was estimated at £50 billion in 2003. Over the same period sales of domestically produced chemicals were £34 billion.”

The industry’s own figures show that in 2004, while almost 50 per cent of chemical industry sites had implemented environmental management systems, under 15 per cent had equivalent health and safety management systems (11). The UK industry employs directly about 230,000 workers.

4. Vulnerable populations

Chemical usage is not just an issue for those in primary manufacturing or processing. Vulnerable workers, for example, hairdressers and cleaners, use highly toxic chemicals routinely and work largely unseen by statutory safety authorities and without occupational health guidance or access to health and safety expertise.

When assessing the impact of work cancers on the working population, it is also important to note that almost all the risk is concentrated in a relatively small section of the workforce. Work-related cancer is far more common in blue-collar workers – there is an undeniable correlation between employment in lower status jobs and an increased risk (12). Studies have found, for example, that 40 per cent of the lung and bladder cancer cases in certain industrial groups can be caused by occupational exposures (13). French statistics office figures published in 2005 found 1 in 8 workers were exposed to carcinogens at work, but that the figure was 25 per cent for manual workers and just 3 per cent for managers (6).

The 1998 CAREX report for Great Britain (8) concluded workplace exposures to carcinogens were restricted to about one-fifth of the working population. If the occupational cancer risk was equal across the population, based on HSE’s figure of 6,000 deaths a year, this would equate to 1 per cent of all deaths being caused by occupational cancers in any given year. However, the responsible exposures are limited to a much smaller group who bear most of the risk, suggesting that 5 per cent or more of deaths in this group could be caused by occupational cancers.

Nor are all the exposed workers in big firms with occupational health facilities, safety officers and sophisticated control systems. Each year members of the UK Chemical Business Association (CBA) distribute more than 2.5 million tonnes of chemicals. CBA estimates that 95 per cent of Europe’s chemical industry is comprised of small and medium sized enterprises.

Cancers in workers in small firms are unlikely to be attributed to work unless they are otherwise rare cancers, and the workers are unlikely to have access to informed occupational safety or medical advice.

5. Complex exposures

The UK strategy fails to take adequate account of complex workplace exposures – multiple exposures at one time, and multiple changing exposures through a working lifetime. The use of more substances in higher volumes in a greatly increased number of products and processes creates the potential for highly complex working environments with complex, combined exposures to workplace carcinogens or substances that could increase vulnerability to carcinogens. A general dusty environment, for example, can overwhelm the body’s mucociliary clearance system, allowing easier passage of airborne substances into the body. Existing exposure standards and control policies do not reflect this total carcinogen dose or the complexity of some mixed workplace exposures, which could create a gross exposure greatly in excess of the exposure limit for any single substance.

The number of jobs per working lifetime has increased markedly in the last 30 years, with most workers now having 10 or more jobs in the period between joining the workforce and retirement. New technologies and processes mean workplace exposures will in many instances change markedly throughout a working lifetime.

As the UK has no occupational cancer registry or systematic measures to ascertain or register exposures, HSE does not know who has been exposed, to what, where or when. This means only a minority of cancers – generally otherwise rare and work specific – stand any chance of being recognised. This has implications for both prevention and for workers’ welfare. Potentially hazardous conditions will not be recognised or addressed and victims of occupational cancer are unlikely to be compensated or receive a timely diagnosis and treatment.

6. Inaction on known risks

Even where the Health and Safety Executive recognises a workplace exposure may cause cancer, this is frequently overlooked in its practical guidance. HSE’s metalworking fluid webpages (14) omit any mention of the occupational cancer risk from the general health risks section and guidance for occupational health advisers, and treat exposure as a general hazard “to be prevented where reasonably practicable” rather than a cancer hazard where far more stringent stipulations should apply. Recent evidence suggests there is even more reason for HSE to issue an explicit cancer warning. A Harvard University report noted that existing studies substantially under-estimate the metalworking fluids cancer risk (15).

When medium density fibreboard (MDF) – a composite creating potential for exposure to two recognised carcinogens during manufacture or machining – became the subject of a recent safety controversy, HSE backed entirely the industry line on safety, failing to acknowledge the cancer risk (16). It did not modify this position when formaldehyde was in 2004 upgraded to the International Agency for Research and Cancer’s top cancer risk category, Group 1 (17). Wood dust was already rated as an IARC group 1 carcinogen.

On other substances, HSE has trailled behind other national regulatory agencies in recognising a workplace cancer risk, in at least one instance actively promoting the use of a cancer-causing substance. In February 2000, chemical manufacturer Dow failed in a bid to stop Australia's chemical standards body NICNAS labelling the common industrial solvent trichloroethylene as a carcinogen and mutagen. It was two years before HSE issued an equivalent warning.

HSE’s 2002 alert said employers should consider using an alternative solvent or cleaning process or, if this was not possible, enclosing the degreasing process as far as possible. Prior to this, HSE had for a decade – including the two years in which Australian workers had been warned of the cancer risks - been explicitly recommending trichloroethylene use as an “ozone friendly” alternative to the more worker-friendly trichloroethane. UK unions in the 1970s had run successful campaigns to get rid of trichloroethylene, in some cases negotiating trichloroethane as a safer alternative. Alternative processes, friendly both to the environment and the workforce, had been available when HSE was recommending trichloroethylene use.

7. Failure to recognise work cancers

Many important workplace cancers are entirely overlooked by HSE. Breast cancers are not treated as a serious work-related risk to be addressed in HSE’s strategy, despite evidence that large numbers could be at risk (18).

HSE also misses entire categories of workers known to have an elevated cancer risk. Its list of targeted work cancer risks does not include “painting”. However, painters comprise a large occupational group classified as facing a Group 1 cancer risk by the International Agency for Research on Cancer, the top risk rating (19).

On wood dust, another Group 1 IARC carcinogen, HSE says this on the cancer risk: “Established for cancer of the nasal cavity or sinuses in cabinet makers and machinists exposed to wood dust.” In fact, wood dust is a nasopharyngeal cancer and possibly lung cancer risk (20), and has been identified in almost all woodworking occupations, not just cabinet making and machinists.

Other occupational groups recognised elsewhere and in the literature as facing an elevated risk for a range of cancers are also ignored. This includes the increased risk for non-Hodgkin’s lymphoma and other cancers in farmworkers and a range of cancers in firefighters, including primary site brain cancer, primary site bladder cancer, primary site kidney cancer, primary non-Hodgkin’s lymphoma, primary site ureter cancer, primary site colorectal cancer and primary leukaemia (6).

Even for substances like asbestos, HSE limits its analysis primarily to the risk of lung cancer and mesothelioma, despite known associations with many other cancers. Cancers including gliomas, head and neck cancers (21), breast and haemopoetic cancers, all linked to work exposures, are among those largely or entirely ignored.

8. Outdated and complacent data

HSE is overly reliant on data from the EU and IARC, much of it out of date. For example, its briefings on benzene, cadmium and diesel exhaust are dangerously outdated, and greatly under-estimate the potential cancer risk. Nor has HSE practical systems to review and act on new information in a timely manner, or to revise assessments.

On a rare occasion HSE did revisit its occupational cancer estimates, it revised them down. It now says the ratio of asbestos related lung cancers to mesotheliomas may be lower than 1 to 1 – a 2005 HSE paper (22) puts the ratio of asbestos lung cancers to mesotheliomas at between 2/3 and 1 to 1 - much lower than many other estimates. The authors acknowledge their figure will miss some cancers because it under-estimates the effects of chrysotile (white asbestos), which has been the dominant exposure since 1970. And their analysis only includes cancer deaths up to the age of 74, whereas many asbestos related lung cancer deaths occur in older workers. While many observers believe the ratio of asbestos-related lung cancers to mesothelioma may be closing as fewer workers are experiencing the very high exposures which were linked to much higher numbers of lung cancers, and a drop in smoking will reduce those caused by the synergy between smoking and asbestos exposure, HSE’s new estimate is significantly lower than generally cited figures.

HSE data sources are also inadequate. Its chemical-by-chemical approach relies on the limited work already done. Yet relatively few chemicals in use in the workplace have been thoroughly assessed for chronic health risks, fewer still providing sufficient satisfactory data to be listed as a human cancer risk by either IARC or the EU. Many workplace and substances likely to present a substantial cancer risk are assumed safe as a result of HSE’s approach, whereas lack of data mean no adequate assessment of the cancer risk has been undertaken. This is not a basis for a protective approach, and leaves workers facing relatively uncontrolled exposures to substances which may as data accumulate be proven to be cancer-causing.

Further, HSE’s exposure standards for known carcinogens fall below the levels expected elsewhere. For example, HSE’s assessment that at an exposure level for chromium of 10 µg/m³ “no further risk reduction measures are needed” would be completely unacceptable in Germany, where this no requirement to act stipulation would only be accepted if the exposure level was a tenth of this at 1 µg/m³, and then only in the short-term. Long-term, no further action would only be required if the airborne chromium levels were 0.1 µg/m³, one-hundredth the UK level.

9. Failure to reduce risks

HSE’s approach is entirely about limited controls on a limited number of carcinogens in a limited number of circumstances. It does nothing to reduce the overall number or volumes of carcinogens at use in workplaces.

A more responsible approach would be to set targets for “sunsetting” the most potent carcinogens, and to introduce a toxics use reduction approach to ensure safer methods and processes are used where they are effective (23). Toxics Use Reduction is an approach which has not only been used effectively, but has also received strong support from industry (24). Such forward thinking strategies are supported elsewhere. The Canadian Strategy for Cancer Control (CSCC), a coalition of cancer prevention, health service and other bodies, has made a public stand in favour of this “primary prevention” of occupational cancer (25).

Numerous reports have warned that failure to act promptly on early warnings has in the past led to entirely avoidable epidemics of occupational disease, including workplace asbestos, benzene and radiation cancers (26).

10. Low level exposures

Large groups running a small risk from exposure to workplace carcinogens still amount to a large number of affected workers, a public health burden entirely missed by HSE. For example, a report this year identified a five fold increase in breast cancer in developed industrial regions was in part due to exposure to industrial chemicals (27).

Equally unlikely to be classified as occupational are the many lung cancers caused by exposure to “nuisance” dusts, for example general building or foundry dusts. Lung cancer is the standout cancer killer, but is overwhelmingly attributed to smoking. In fact, as smoking levels decline in developed nations a much larger than expected number of “never smoked” lung cancers are being seen. Occupation is a clear contributory factor to these cancers.

A February 2007 report in the Journal of Oncology, for example, concluded one-in-five lung cancers in females and almost 1-in-10 in men occur in people who have never smoked (28). That would equate to approaching 3,000 non-smoking lung cancer deaths in UK women each year, and 2,000 deaths in men. Obviously, workplace exposures would be a co-factor in the lung cancers experienced by the smoking group too - smoking does not make you immune to occupational lung carcinogens. The evidence suggests it does in fact greatly increase the likelihood of getting a work-related cancer.

11. Exposed workers unassisted

In the great majority of cases HSE is failing to provide workers and former workers information about the cancer risk posed by their past and current exposures to workplace carcinogens. This has three main detrimental effects.

Firstly, these workers are not in a position to make informed decisions about their working environment and seek improvements. Secondly, they are not in a position to seek the health surveillance necessary to improve the chances of an early cancer diagnosis and therefore cure. Survival rates in the UK for some of the major cancer killers, for example lung cancer, are low compared to other developed nations. This is in a large part due to late diagnosis.

Finally, lack of awareness and support means most workers developing occupational cancers receive no compensation or related benefits, even for extremely well known causes of work cancer like asbestos. For example, in 2002, 1,862 people died from the asbestos cancer mesothelioma, but only 54 per cent of those people received Industrial Injuries Benefit. However, far fewer still receive this benefit for asbestos-related lung cancer. Despite it being accepted that there is at least one lung cancer death for each mesothelioma death - and this is a conservative estimate - only 60 lung cancer payments were made in 2001 in the UK. Contrast this with Germany, where 767 benefit payments were made for asbestos-related lung cancer and 665 benefits for mesothelioma in 2001 (29).

Most occupational cancers occur in older workers, so there is also a compelling case for health agencies other than HSE to greatly increase the resources available to provide advice, screening and support for retired workers at risk of developing work-related cancers.

12. Mistakes on short latencies

HSE makes an assumption that occupational cancers have a long latency period, with today’s cancers the result of exposures a working generation ago. This has two damaging effects. It allows HSE to assume today’s cancers are the result of historic working conditions, much worse than those in workplaces today. It also allows it to downplay the risk facing the current working generation.

For example, on wood dust, HSE says: “There is a latency of 20 years between exposure and tumour development.” However, many cancers including those caused by wood dust can have much shorter latency periods. For wood dust, latency can be just 10 years. HSE also says the mode of action is “uncertain”. In fact the occupational cancer risk from wood dust is as a result of inhalation, and is therefore easily preventable.

13. Lack of employer knowledge

HSE makes unprovable assumptions about improved occupational hygiene standards and risks, and about the willingness and capability of firms to recognise and control risks. HSE’s own studies have shown many chemical companies had, at least until the mid- to late-1990s, little or no knowledge of their duties under the chemical control regulations and most were unaware of relevant occupational exposure limits. At the time, research suggested many products including potentially carcinogenic dyes were being imported without adequate warnings (30).

14. Exposures inside and out of work

Assumptions about exposure levels and risk based solely on exposures in the workplace under-estimate the total toxic load many workers experience. Workers can be exposed at work and in the general environment. Women exposed to cancer-causing endocrine disrupting chemicals at work, for example, will frequently have additional exposures to substances acting in the same way outside of work.

Likewise, farmworkers exposed to pesticides may face household, environmental and dietary exposures to the same or related chemicals. A 2007 study found agricultural workers exposed to high levels of pesticides have a raised risk of brain tumours. All agricultural workers exposed to pesticides had a slightly elevated brain tumour risk, but the paper reported the risk was more than doubled for those exposed to the highest levels. The study, published online in the journal Occupational and Environmental Medicine in May 2007, also found a significant risk among people who used pesticides on houseplants (31).

15. Changing industry

HSE’s chemical by chemical approach fails to take account of the rapid evolution of industry and industrial processes. Evidence of an emerging cancer problem in the microelectronics industry (32) has not elicited prompt, precautionary action from HSE, a failing that has attracted international criticism (33). Similarly, while HSE at least theoretically advocates a precautionary approach in the fast emerging nanotechnology industry, in practice there is little understanding of the hazards posed or how they might be controlled. The industry, meanwhile, largely free from the attention of the resource depleted enforcement agency, is growing at a startling pace (34).

The problem is not limited to those employed in new industries, but also the much greater number affected by the changing nature of existing jobs or job functions. More workers performing routine tasks are exposed to chemicals as a result of the tendency to opt for quick chemical fixes applied by poorly skilled workers as an alternative to labour- and resource-intensive skilled labour. For example, many local authorities are opting to use pesticides for routine weed control as an alternative to employing skilled parks and garden staff.

16. Changing work patterns

In addition new age discrimination regulations and government moves to raise the retirement age mean many workers are likely to have more years of exposure to potential risk. The UK already has one of Europe’s highest proportions of older workers in work. The changing nature of employment – long hours, frequent changes of job, frequent changes of job task, irregular hours and shiftwork all impact on occupational cancer risks. Breast and other cancers have been linked to shiftwork.

17. Risks to women

Women are now better represented in the workforce, are likely to spend longer in the workforce and do a much wider range of jobs. Since 1975 men’s employment has declined from around nine out of ten to eight out of ten (79 per cent) for men of working age (16–64). At the same time women’s employment has increased from around six out of ten to seven out of ten (70 per cent) for women of working age (16–59).

HSE’s approach fails to take adequate account of the increasing participation of women in the labour market, or of the risks to women, or of risks in women-dominated employment areas. For example, HSE’s estimates of risks in the health sector include far fewer jobs, cancers and exposures than the equivalent guidance from the US National Institute for Occupational Safety and Health.

HSE cannot present any credible assessment of the occupational cancer risks faced by women, of the number of cancers in women related to workplace exposures.

For example, breast cancer is the most common cancer in the UK and one of the top cancer killers. However, it does not appear on HSE’s priority action list or in its estimates of the numbers affected by occupational cancer, despite clear evidence associating industrial exposures with elevated cancer risk. An October 2005 report, 'Breast cancer – an environmental disease: the case for primary prevention,' concluded there was “incontrovertible evidence” that many industrial chemicals and radiation are major contributors to overall breast cancer rates (18).

18. Work and lifestyle interactions

The UK approach to occupational cancers fails to take account of the workplace contributions to supposedly “lifestyle” cancers. For example, work stress is associated with poor behaviour patterns, including smoking and other substance abuse behaviours (35).

Workplace exposures can also “potentiate” the effect of tobacco smoke. For example, the synergy between asbestos exposure and tobacco smoke is well reported, the combination creating a massively increased risk compared to exposure to either carcinogen alone. And a 2005 paper concluded exposure to wood dust increased the chances of developing not only nasal cancer but also lung cancer, finding the risk of lung cancer was increased by 57 per cent with wood dust exposure in absence of smoking, by 71 per cent for smoking in the absence of wood dust exposure, and by 187 per cent for individuals who were exposed to both smoking and wood dust (20).

19. Parental exposures and children

The impact of work exposures can also be intergenerational, with the impact of work on workers’ children wholly ignored in HSE’s analysis. For example, a 2003 University of Massachusetts Lowell report noted “evidence increasingly indicates that parental and childhood exposures to certain toxic chemicals including solvents, pesticides, petrochemicals and certain industrial by-products (dioxins and polycyclic aromatic hydrocarbons) can result in childhood cancer” (36). Recent research has reinforced this evidence (37).

And paraoccupational cancers – cancers in those incidentally exposed to carcinogens via exposure to asbestos on the clothing of a parent or spouse, for example – are being seen with increasing frequency (38).

20. Future problem

The increase in overall life expectancy and declining death rates from other causes, means for the current working generation cancer will have longer to develop and less competition as a cause of death. And while mortality from cancer is falling as a result of improved diagnosis and treatment, the incidence of cancer is not, supporting the case for greater preventive efforts.

Key recommendations

• Occupational cancer prevention should be recognised by the government as a major public health priority and should be allocated resources accordingly.

• A national occupational cancer and carcinogens awareness campaign should be launched as a matter of urgency.

• The Health and Safety Executive should convene a tripartite working party, including representatives of unions, health and safety campaign organisations and occupational disease victims’ and advocacy organisations, to review its occupational cancer strategy.

• Wherever possible, IARC Group 1 and Group 2A carcinogens should be targeted for “sunsetting”, a phase out within a designated timeframe, to be replaced by safer alternatives.

• Toxics Use Reduction legislation, already used successful in some US jurisdictions, should be introduced to encourage the use of the safest suitable substances and processes. The precautionary principle should be applied to substances suspected of causing cancer in humans.

• A national system of occupational health records should be developed to ensure adequate recording of workplace exposures and other occupational cancer risk factors. Employers must have a duty to inform any workers of their exposures to known or suspected workplace cancer risks and carcinogens.

• A National Exposure Database should be created.

• The Health and Safety Executive should provide resources for training of union safety reps in “lay epidemiology”, techniques for the early recognition of work-related diseases, including cancer.

• The UK should implement properly the European Union law requiring workers to have access to occupational health services.

• The government Industrial Injuries Benefit Scheme should be revised and extended to include a wider range of occupational cancers in it scope. There should be a consideration of the introduction of a “rebuttal presumption” of work-causation for cancers with an established association with work.

1 REACH Partial Regulatory Impact Assessment after Common Position, Defra, May 2006. Defra costs an occupational cancer death at £2.46m.

2 HSE cancer statistics website, accessed 21 June 2007.

3 Doll R and Peto R. The causes of cancer: Quantitative estimates of avoidable risks of cancer in the United States today. Journal of the National Cancer Institute, volume 66, number 6, pages 1191-1308, 1981.

4 Lin Fritschi and Tim Driscoll. Cancer due to occupation in Australia, Australian and New Zealand Journal of Public Health, volume 30, number 3, pages 213-219, 2006.
Richard Clapp, Genevieve Howe, Molly Jacobs Lefevre. Environmental and cccupational causes of cancer: A review of recent Scientific literature. Lowell Center for Sustainable Production, University of Massachusetts Lowell, September 2005. Executive summary • Full report [pdf]

5 Lennart Hardell and others. Secret ties to industry and conflicting interests in cancer research, American Journal of Industrial Medicine, volume 50, issue 3, pages 227 – 233, 2007 [abstract]

6 Work Cancer Prevention Kit, Hazards magazine online report, March 2007

7 Päivi Hämäläinen, Jukka Takala, Kaija Leena Saarela. Global estimates of fatal work-related diseases, American Journal of Industrial Medicine (AJIM), volume 50, pages 28-41, 2007. [abstract]

8 Kauppinen T and others. Occupational exposure to carcinogens in the European Union. Occupational and Environmental Medicine, volume 57, pages 10-18, 2000. Carex webpages
Carcinogenic exposure information for Great Britain, Carex, 1998.

9 Powertrain Occupational Respiratory Disease Outbreak: Report of Immunological Investigation MU/06/01, HSL, March 2007 [pdf]
Outbreak of respiratory disease at Powertrain Ltd, Longbridge, Birmingham - emerging lessons, HSE, April 2006 [pdf]

10 Come clean: HSE faces a funding and enforcement crisis, Hazards magazine, number 95, July-September 2006.

11 Facts and figures: UK chemical industry, CIA, January 2006 [pdf]

12 Prevention of occupational cancer, World Health Organisation (WHO) GOHNET newsletter, number 11, 2006 [pdf]
Infante P. Cancer and blue-collar workers: Who cares? New Solutions, volume 5, number 2, 1995.

13 Vineis P and Simonato L. Proportion of lung and bladder cancers in males resulting from occupation: a systematic approach. Archives of Environmental Health, volume 46, pages 6-15, 1991.

14 HSE metalworking fluids webpages
Hazards magazine online commentary

15 Elizabeth J Malloy, Katie L Miller and Ellen A Eisen. Rectal cancer and exposure to metalworking fluids in the automobile manufacturing industry, Occupational and Environmental Medicine , volume 64 , pages 244-249, 2007 [abstract] • Related editorial

16 Hazards magazine, November 2005 online cancer report

17 IARC Monographs on the Evaluation of Carcinogenic Risks to Humans. Formaldehyde, 2-Butoxyethanol and 1-tert-Butoxy-2-propanol, volume 88, December 2006. online

Report of IARC meeting to re-evaluate formaldehyde for IARC monograph, volume 88, held 2-9 June 2004.

18 Facts about women & men in Great Britain 2006, EOC [pdf]
Breast cancer – an environmental disease: the case for primary prevention www.nomorebreastcancer.org.uk

19 IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, volume 47, Some organic solvents, resin monomers and related compounds, pigments and occupational Exposures in paint manufacture and painting, 1989 [pdf]

20 Carlos H. Barcenas, George L. Delclos, Randa El-Zein, Guillermo Tortolero-Luna, Lawrence W. Whitehead, Margaret R. Spitz. Wood dust exposure and the association with lung cancer risk. Am J Ind Med. 2005: 47: 349-57 [abstract]

21 James T Brophy, Margaret M Keith and others. Cancer and construction: What occupational histories in a Canadian community reveal, International Journal of Occupational and Environmental Health (IJOEH), volume 13, page 32-38, 2007 [pdf]

22 Darnton, A. J., McElvenny, D. M., Hodgson, J. T. (2005) Estimating the number of asbestos related lung cancer deaths in Great Britain from 1980-2000, Annals of Occupational Hygiene, published online 26 August 2005 [abstract]. doi:10.1093/annhyg/mei038

23 Toxics Use Reduction Institute. www.turi.org

24 Five Chemicals Alternatives Assessment Study, TURI, July 2006 [pdf]

25 Prevention of occupational and environmental cancers in Canada: A best practices review and recommendation, Canadian Strategy for Cancer Control (CSCC), May 2005 [pdf]

26 Late lessons from early warnings: The precautionary principle 1896-2000, European Environment Agency, Environmental issue report number 22, 2001.

27 Environment and breast cancer: Science Review. Silent Spring Institute, 2007.

28 Lung cancer incidence in never smokers, Journal of Clinical Oncology, volume 25, number 5, pages 469-471, 2007 [abstract].

29 A little compensation. Hazards magazine online briefing, May 2005.

30 Industry's perception and use of occupational exposure limits. HSE Contract research report, CRR 144, HSE, 1997.
Europe-wide check finds 40 per cent of new dyestuffs illegally marketed, according to report on enforcement project. HSE news release E191:96, 6 November 1996.

31 Dorothée Provost and others. Brain tumours and exposure to pesticides: a case-control study in, southwestern France, Occupational and Environmental Medicine published online 30 May 2007; doi: 10.1136/oem.2006.028100 [abstract]

32 Richard W Clapp. Mortality among US employees of a large computer manufacturing company: 1969-2001, Environmental Health: A Global Access Science Source, volume 5, number 30, 2006. www.ehjournal.net
Cancer among current and former workers at National Semiconductor (UK) Ltd, Greenock: results of an investigation by the Health and Safety Executive, ISBN 0717621448, HSE, 2001 [pdf]

33 Watterson, A. Regulation of occupational health and safety in the semiconductor industry: enforcement problems and solutions. IJOEH (2006). 12(1):72-80.
Bailar, J; Greenberg, M; Harrison, R; LaDou, J; Richter, E; Watterson, A, Phase 2, SCCOSH, SVTC (2002) Cancer Risk in the Semiconductor Industry: A Call for Action. IJOEH 8(2):163-68

34 Hazards nanotechnology webpages
Nanotechnology information sheet - HSIN1, HSE, June 2004 [pdf]
HSE NanoAlert Service Issue 1. December 2006 [pdf]

35 Hazards magazine work stress, smoking and drug and alcohol webpages
Prevention in the workplace, European Monitoring Centre for drugs and drug addiction, 1997.

36 Toxic chemicals and childhood cancer: A review of the evidence. Tami Gouveia-Vigeant and Joel Tickner, Lowell Center for Sustainable Production, University of Massachusetts Lowell, USA, May 2003 [pdf].

37 Editorial: Role of exposure to environmental chemicals in the developmental basis of disease and dysfunction, Reproductive Toxicology, volume 23, pages 257–259, 2007.

38 Hazards magazine asbestos webpages