Health effects of occupational exposure to hazardous substances in Australia - putting criticisms of the Kerr Report in perspective
The National Occupational Health and Safety Commission (NOHSC) has issued the following release in response to recent criticism of the 1996 report "Best estimate of the magnitude of health effects of occupational exposure to hazardous substances". The study was funded by a NOHSC Development Grant and conducted by Professor Charles Kerr and colleagues at the University of Sydney's Department of Public Health and Community Medicine, and is commonly referred to as the Kerr Report.
NOHSC responds to criticism of OHS statistics on health effects from exposure to hazardous substances
The Kerr Report was a key event in occupational health and safety because all previous estimates of work-related death - 500 - had been primarily based on traumatic deaths and did not include deaths related to disease - 2,200. The previous figure is clearly an underestimate of the total picture as we know there are some 500 cases of mesothelioma each year.
"The findings of the Kerr Report are not hard data, which was clearly acknowledged in the report itself," said Dr Tim Driscoll, NOHSC's Kerr Report project manager.
"The results serve as a warning against complacency in the occupational use of hazardous substances and provide a basis for gauging the extent of the problem of ill heath arising insidiously from such use in the Australian working community.
"This in turn enables health policy concerning occupational exposure to hazardous substances to begin to be based on evidence rather than anecdote, and highlights the need for better exposure and risk data relevant to Australian workplaces to be produced and made available to the occupational health and safety community.
"Australia is improving its measurement of exposure to hazardous substances through better use of statistics and health surveillance."
In December 1996 a critique of the Kerr Report was presented in a paper by Mr Christophers and Mr Zammit at the Australian Institute of Occupational Hygiene's annual conference in Perth. The paper concentrates on a consideration of cardiovascular disease, with a smaller section on acute poisoning. Responses to specific criticisms are provided on the final page of this release.
The Kerr Report describes in detail the methodology and findings of a project commissioned by NOHSC to try to assess the impact on the working community of occupational exposure to hazardous substances. The Kerr Report provided the first national estimate of the magnitude of health effects from exposure to hazardous substances.
The lack of sound exposure and risk data in Australia and overseas necessitated the use of an attributable risk methodology - assuming that a percentage of deaths from a given condition are attributable to workplace exposures.
The shortcomings of that methodology are considered at length in the Kerr Report, but this is not acknowledged by Mr Christophers and Mr Zammit, even though their criticism of the Kerr Report is based largely on the methodology. Recognising the limitations of the methodology, the Kerr Report states (p.viii): "Accordingly, all conclusions on resulting estimates must remain guarded."
The estimates produced by the Kerr Report comprise current episodes of ill health that are primarily due to past occupational exposures.
Except for a small proportion of medical conditions - mainly acute poisoning - the estimates were not designed or intended to reflect the effect of current exposures, nor to be used as a census of work-related death.
Australia has a very high and increasing incidence rate of malignant mesothelioma - which is nearly always associated with past exposure to asbestos. A further 10,000 cases are expected by 2020.
NOHSC responses to key objections by Mr Christophers and Mr Zammit (in bold) follow:
Attributable risk is an inherently unscientific approach and should not be used.
The use of attributable risk is a well known methodology in occupational and public health epidemiology. Despite its known limitations, it has been used in a number of published studies. Unfortunately the necessary data and resources that would allow the comprehensive use of exposure and risk information were not - and are still not - available. Without these data, the attributable risk approach was the only feasible option for the conduct of the Kerr Report project.
In the case of cardiovascular disease, the estimate made in the Kerr Report appears speculative.
There is a variety of opinion on the relationship between cardiovascular disease and work. The issue of cardiovascular disease was always seen as a difficult one for the project, but one which was dealt with specifically in the Kerr Report which states (p. 62, stage 1, vol 1): "There is no simple way to deal with this problem. The attribution rates are only estimates, but they remain the best estimates available." The Kerr Report then provides a range of estimates based on a range of attributable proportions.
Mr Christophers and Mr Zammit argue only two chemicals cause cardiovascular disease. However, there is good evidence that there are a number of other chemicals which cause cardiovascular disease which they have not mentioned.
The estimates in the Kerr Report are highly exaggerated.
The estimates for acute poisonings are probably too high, and this is discussed in the Kerr Report. There was no easily accessible data source that would have allowed this to be done more comprehensively or more validly. However, since acute poisonings only make up about two percent of the total mortality estimates in the study, the effect of any error would be small.
Overall, there is no good information to show the estimates are too high. In fact, the Finnish review team considered that the estimates may be too low. In addition, consideration of more recent Australian data on lung cancer deaths and mesothelioma deaths due to asbestos suggest that the Kerr Report may have underestimated these quite significantly.
The Kerr Report lacks the scientific rigour needed to form the basis for Health policy-making
This report was peer reviewed by two international occupational health and safety agencies and a major Australian health agency. All the reviewers' comments and the responses by the authors are published with the main body of the Report, providing complete transparency to the process and the study's strengths and weaknesses.
Media contacts: Nick Miller (02) 9577 9424; Sylvie Marimon (02) 9577 9249
11 March 1997
Page last updated: 15/07/2008