From Medscape General Medicine > Departments > Book Reviews

It Ain't Necessarily So: How Media Make and Unmake the Scientific Picture of Reality

Reviewed by: Cathy Tokarski

Authors and Disclosures

Posted: 11/06/2001

Print This

Introduction

By David Murray, Joel Schwartz, S. Robert Lichter
Rowman & Littlefield
Copyright 2001
249 pages
ISBN: 0-7425-1095-6
$24.95 hardcover

At a time when the nation is gripped by the fear of bioterrorism and seeks definitive answers from the federal government, a new book that examines the media's less-than-rigorous reporting on scientific findings and its influence on policy-making and public opinion is especially timely.

David Murray, director of the Statistical Assessment Service in Washington, D.C., Joel Schwartz, senior adjunct fellow at the Hudson Institute, an Indianapolis-based think tank, and S. Robert Lichter, president of the Center for Media and Public Affairs in Washington, D.C., draw a troubling picture of the gullibility of media coverage of studies in the natural and social sciences. They look at the agenda-driven mission of advocacy groups that sponsor these works, and the dependence of politicians on research-supported "solutions," and offer up a wealth of examples from print and television coverage to illustrate this dilemma.

Their insights into the way scientific studies are formulated, conducted, and reported by the press illustrate how biased or misleading -- or even grossly inaccurate -- some news reports can be. Since consumers of this information can be subject to unwitting manipulation, the authors assert, they must become more skeptical about how to better assess research findings, determine their value, and communicate that knowledge.

Extensively referenced and well written, It Ain't Necessarily So is divided into 3 sections. The first describes the media's troubling tendency to ignore findings that suggest a positive trend, such as the declining number of AIDS diagnoses in the mid-1990s, or to slant the findings in a more negative, but attention-getting, manner. The authors also skewer the media's tendency to elevate preliminary but compelling research findings, such as a nonpeer-reviewed study on the effects of day care on the mother-infant relationship.

The second section, examining the ambiguity of measuring scientific and social phenomena, could serve as an introductory (though more entertaining) course in statistics. Witty commentary, vivid examples, and sharp insight make what could be a tedious exercise in dissecting research methodology an entertaining and instructive guide. In 5 chapters, the authors examine the perils of "tomato statistics," or cases where news reports draw attention to what appears to be a high number of incidents such as date rape or domestic assault but in fact are fueled by faulty research assumptions and definitions.

Readers also learn the vulnerabilities of research findings that rely on proxy instead of direct measurement; the tendency of data to be used to support only 1 conclusion, not several plausible conclusions; and the tendency of the media to exaggerate health and environmental risks. Examples used to illustrate this section, such as misleading statistics about AIDS among women and a rise in mortality from infectious diseases, may serve as a helpful model to the physician or healthcare professional asked by patients to explain the significance of the latest health news report.

While the book is replete with examples of incomplete reporting and biased research, the authors frequently point out the steps that readers can take to become better consumers of scientific information. In some cases, such as reporting on the incidence rates and risk of breast cancer, reporting has grown more sophisticated, the authors state. A widespread perception that 1 in 8 women will develop breast cancer has been refined in recent years by the media to reflect the fact that lifetime risk is not the same as the risk a woman faces at any one point in her life. "Activist contentions have repeatedly been questioned, and alternative interpretations of the statistics have been readily offered," the authors state. "In this notable instance, the print media have often excelled at explaining the realities of risk" (123).

The ongoing debate over to what extent research is influenced by industry sponsorship will make the final section of particular interest to many readers.

In one example, the authors relate the uproar that ensued in 1996 following publication of a study concluding that women who had abortions faced a slightly increased risk of breast cancer than women who did not. Instead of closely examining the study results or questioning its methodology, the Philadelphia Inquirer focused on the coauthor's personal views, which included speaking out in Christian media outlets and serving as an expert witness for antiabortion groups in Philadelphia federal court.

"All that may be true," the authors point out, but it is unclear why it is relevant. "Should only pro-choice researchers be allowed to study the medical effects of abortion?" (153). The Inquirer also failed to point out that one of the study's coauthors was pro-choice, and that the study was published in a peer-reviewed journal.

A second example showed reporting by the print media, but not network television, capable of more discriminating coverage. A 1996 study published in the Journal of the American Medical Association concluded that women with breast implants face a small increased risk of developing connective tissue disease. The lion's share of the $18.3 million study came from the National Institutes of Health ($17 million); the remaining $1.3 million came from Dow Corning, the company that manufactured silicone breast implants until they were banned in the United States in 1992. Dow did not learn of the study results until shortly before they were published.

In covering the JAMA study's findings, articles published in the Boston Globe, Chicago Tribune, and The New York Times mentioned Dow's financial contribution to the research without suggesting that it tainted the validity of the results. Articles included quotes by the leading researcher who underscored the importance of the "arm's length" relationship between Dow's participation and the study's findings. However, a report on CBS Evening News highlighted Dow's funding without stating the amount, and declared that future studies from Harvard and the NIH "will try to assess the real risk" of breast implants and connective tissue disease.

In its concluding chapter, the authors acknowledge that their effort to bring a more rigorous examination to the proliferation of scientific studies by a harried press and results-driven advocacy organizations "may seem like the labors of Sisyphus." That may be true, but the information provided in It Ain't Necessary So gives readers the necessary tools to begin an ascent.

• Print This

Medscape General Medicine. 2001;3(4) © 2001 Medscape

• Radiology Benefit Management Programs: Who Benefits?
• Did Changing Primary Care Delivery Models Change Performance?