The legislative basis of the federal Canadian government's control of toxic chemicals is described and examples are given of the practical application, ranging from recommendations to a ban on the sale of the product. The ordered sequence of risk assessment and the application of risk estimation techniques are considered. It is clear that the ultimate political decision is not amenable to simplistic scientific analysis, although risk analysis is valuable in defining, rather than solving, the problem.
Regulatory agencies require numbers to provide health protection. The manner in which these numbers are derived from animal experiments and human epidemiology is considered together with the limitations and inadequacies of these numbers. Some recent examples of risk assessment in Canada are given including asbestos, drinking water, and indoor air quality. The value of these numbers in providing a measure of the hazard in a wider perspective is stressed, although they can never be the sole determinant of public policy.
The value of risk-benefit analysis for drugs is outlined with emphasis on its special qualities compared with decisions on environmental agents, occupational risks and other societal risks. Consideration is given to the issue of risk perception and examples given of risk assessment in practice in Canada for Bendectin, isotretinoin, bovine somatropin (somatotrophin) and oral contraceptives. The changes in regulatory approaches for AIDS drugs are noted. It is suggested that an understanding of the risk-benefit balance should be incorporated into the drug evaluation process, perhaps eventually by the use of quality-of-life measures.
An account is given of the manner in which regulatory decisions for a number of environmental chemicals have been made in Canada. The sources of evidence for such decisions are reviewed and applied to tris(2,3-dibromopropyl)phosphate, drinking water contaminants, polychlorinated biphenyls, lead, dioxins, and urea-formaldehyde foam insulation. Finally, the societal implications of too mechanistic an approach to risk estimations are considered.
The legislative application of regulatory toxicology in Canada is reviewed, together with the sources of experimental evidence used for action. Examples are given of the critical toxicological information that led to a regulatory decision. Risk numbers have only been used to a limited extent in Canada. Some possibilities for future research are offered.