Book Review

Risk versus Risk: Tradeoffs in protecting health and the environment

N Engl J Med 1996; 335:140-141July 11, 1996

Article

Risk versus Risk: Tradeoffs in protecting health and the environment
Edited by John D. Graham and Jonathan Baert Wiener. 337 pp. Cambridge, Mass., Harvard University Press, 1995. $39.95. ISBN: 0-674-77304-7

Risk tradeoffs are ubiquitous in modern society, and ordinary citizens as well as public officials regularly weigh risks and countervailing risks, either explicitly or implicitly, in making decisions. Sometimes these decisions are informed by careful scientific analysis and assessments, but all too often the risks are not well understood or the tradeoffs are not well analyzed. Yet in many cases careful assessments not only are possible but also can reduce individual and collective risks.

John Graham and Jonathan Wiener have assembled nine case studies to illustrate the complexities of what they refer to as risk-tradeoff analysis, sometimes successful and sometimes not. These cases encompass a variety of actual situations in which risks can be and often are measured and in which the selection of particular treatments or policies can lead to the emergence of countervailing risks. These case studies involve individual choices about personal health, tradeoffs involving highway safety, and the social regulation of dietary and environmental risks. The editors' goal was “to help decisionmakers identify risk tradeoffs, and to structure their decisionmaking so that these tradeoffs are intelligently resolved and overall risk reduction is better achieved.”

The editors outline their approach to risk-tradeoff analysis and explain the risk tradeoffs in terms of whether or not the population experiencing the countervailing risk, and the type of countervailing risk, are the same as the original population and its risk. The medical case studies tend to involve risk substitution for individuals, whereas the environmental or social-regulation examples tend to involve the transfer of risks to others and the transformation from one type of risk to another (e.g., from acute infectious diseases to cancer). The editors have not only organized the volume but also authored individually or together with others all but one of the case studies. Nonetheless, these case studies are highly varied. The medical case studies, on estrogen therapy for menopause and on the use of clozapine therapy for schizophrenia, are the most straightforward and do not really illustrate the complex tradeoffs that the book claims to address. The chapter on the consequences of eating more fish is even less compelling. I would have found a topic such as the detection and treatment of prostate cancer far more interesting, because serious risk tradeoffs make a definitive and universal recommendation for a course of action far from obvious. The environmental case studies, on the other hand, clearly reveal the complexity of decision making in response to highly publicized risks.

The book ends with a 45-page essay by the editors on resolving risk tradeoffs that in part draws lessons from the case studies but focuses on ways of improving government decision making through the wise use of risk-tradeoff analysis. One senses that Graham and Wiener have written much of this before, and the proposals for “constructive reforms” in government processes, although interesting, seem not to follow directly from the material in the rest of the book.

Have Graham and Wiener achieved their goal? That depends on who reads the book. As a reader oriented to quantitative assessment, I did not need much convincing regarding the central thesis about the need to understand risk tradeoffs. My own discipline of statistics includes a formal probabilistic and statistical structure for the evaluation of risk substitution (i.e., the theory of competing risks). But the book contains virtually no technical details, and thus for me the discussion of alternative risks and their measurement in most of the case studies is simplistic at best. Furthermore, the figures and tables offer little of the richness I have come to expect in a carefully argued quantitative presentation. For me the book is a disappointment.

A quote on the jacket states that “the book will be widely read among regulatory decision makers who must sharpen their analysis and judgment — and it should be read by those who have discovered there is no tooth fairy in regulatory decision making.” Perhaps so. I agree that this book might well educate such readers about the need for careful risk-tradeoff analysis, but it does not really present them with a carefully described methodology, either new or old. I doubt that it would enable the uninitiated regulatory decision maker to think systematically about a new problem involving a complex set of risk tradeoffs.

Stephen E. Fienberg, Ph.D.
Carnegie Mellon University, Pittsburgh, PA 15213