Article Text


A Citizen’s Guide to Air Pollution
  1. R L Maynard

    Statistics from

    Edited by D V Bates and R B Caton (2nd edition; pp 452 pages; $20.00). David Suzuki Foundation

    This is the second edition of a book that sets out to tell the intelligent layman, or citizen, about air pollution: its sources, effects and how it may be controlled. It succeeds. Anybody who reads this book will gain a useful and remarkably up to date grasp of the subject matter and should be able to take a confident part in discussions of and decision making in this difficult area.

    The book comprises 11 chapters, arranged fairly conventionally, beginning with the history of air pollution and proceeding via effects on health and the statistical issues raised by studies in that area to effects on vegetation, decision making, and air quality management. A chapter on indoor air pollution is added and a useful linking chapter closes the book. Twelve authors contribute, though in many chapters the original author, David Bates, is a co-author. This has ensured a consistently lucid style.

    What, then, of the details? This is not a book that presents all the evidence: it is selective and avowedly so. In these days of systematic review, such a selective approach is often frowned upon. In the section on health effects the selection seems to me very fairly balanced, though the authors do, perhaps, make little of negative studies and do not discuss how such should be interpreted. The thorny issue of publication bias is not touched on. More importantly, the authors do not tell us how important they think the findings reported are or whether they agree with the original investigators’ interpretations. This would be useful to the lay reader; such a reader needs assistance from an experienced author.

    The chapter “Statistical Issues and Causality” is important and elegant. The primary author, James Zidek, is a professor of statistics: he is ably aided by David Bates. The chapter discusses errors, paradoxes, and misinterpretations of data: the discussion could be useful to any research worker and I have not come across such a clear exposition elsewhere. This is not to say that all the chapter is easy reading. How familiar are you with Popper’s notion of intersubjectivity? The authors assert that this, superficially, means that if given all available knowledge, a conclusion is consensually reached by a community possessing that knowledge, then that conclusion would have the status of a fact relative to the knowledge available and the community formulating the conclusion. A reference to the work by Zidek is given. I turned to “The Logic of Scientific Discovery”. Zidek and Bates may well be correct but I was left worried about the difference between objectively verifiable propositions and widely held views. In the air pollution field it is certainly true (and verifiable) that widely held views are taken as facts: this needs further thought and discussion. The use of words to describe ideas is a field of study in itself and I cannot resist dealing with what I think is a persistent error of interpretation. The authors draw attention to a statement made in the UK (referring to the question of causality in the context of time-series studies) “that it would be imprudent not to conclude that the association was causal” and describe this as “an interesting use of a double negative”. The authors seem to think that the wording implied doubt. They are correct but fail to see that the wording implies less doubt than the alternative: it would be prudent to conclude that the association was causal. The wording was carefully chosen for its force.

    In the section dealing with standard setting the authors extol the approach used by the US EPA. This is beyond all doubt exhaustive and open, but is also expensive and few countries other than the USA would wish to adopt such an approach. The climate of litigation in the USA demands such an approach; that climate is regrettable. Valuable insights into cost-benefit analysis are provided, though this is an area in which US policy development is not as clear as it might be and difficult issues remain unresolved. One area that could have stood a longer discussion is that of whether standards should protect individuals or whether they should apply to cities. This is especially important now that time-series studies are being so widely used.

    In conclusion, this is an outstanding book that provides a great deal of well assimilated information and raises issues of great importance. The authors and publishers should be congratulated for producing such a useful book at such low cost. It should be read by all interested in air pollution: laymen and those with a professional interest.

    View Abstract

    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.