Reviews

The operation was a success (but the patients died)

How media spin distorted the outcomes of a study comparing radical prostatectomy with watchful waiting

   Last week, one simple health message dominated the US media: radical prostate surgery for prostate cancer saves lives. The media were reporting the results of a Swedish trial (see news p 613) published in the New England Journal of Medicine (2002;347:781-9)[Abstract/Free Full Text]yet the trial showed no such thing.

The trial concluded that although radical surgery did reduce disease specific mortality, there was "no significant difference between surgery and watchful waiting in terms of overall survival." A companion study published in the same issue of the New England Journal found that the surgery also failed to improve men's quality of life.

Despite these conclusions, headlines across the United States were unequivocally positive about the benefits of surgery. The New York Times of 12 September said: "Prostate cancer surgery found to cut death risk"; the Chicago Tribune (12 September) read: "Surgery benefit cited for prostate cancer"; and CBS News announced in broadcast and website leads on 12 September: "Study: prostate removal saves lives."

Similar headlines appeared over reports from ABC News, Associated Press, Reuters, and the Canadian Broadcasting Corporation.

Some went even further. Although the trial had nothing to do with screening of healthy men, Dr Patrick Walsh, director of the department of urology at Johns Hopkins University School of Medicine, was quoted as saying that the study results would finally "quiet" critics of prostate cancer screening and treatment.

Dr Otis Brawley, associate director for cancer control at the Winship Cancer Institute of Emory University and a noted critic of prostate cancer screening, was quoted as saying: "It's really the first [study] that shows that radical prostatectomy does something positive."

However, a few editors and journalists remained circumspect in the face of such enthusiasm. The Washington Post's David Brown wrote on 12 September: "Men with prostate cancer who decide not to undergo surgery and instead opt to treat only the symptoms of their disease do just about as well as men who are operated on." National Public Radio's Joanne Silberner made similarly sober comments in her 11 September broadcast.

Dr Brawley, who was quoted in the New York Times article, told me that he thought the article gave "too positive a spin." What the study does show, he said, is that "men who have prostatectomies just exchange one form of death for another within six years." What Dr Brawley meant by "positive" was that there was a possible avenue for further study.

Journalists may overlook nuance and important distinctions, such as the difference between disease specific mortality and all-cause mortality, said Dr Ivan Oransky, who teaches medical journalism at New York University. "If you read the New York Times headline you would be left with the impression that having a radical prostatectomy would lead to a decreased risk of death, but the study showed no such thing. It reduces your risk of death from prostate cancer but that's not necessarily a positive finding. For example, if I were to say, `Listen we have a great new procedure for people with inoperable brain cancerwe're going to cut off every one's head and then I can absolutely guarantee that you won't die of brain cancer,' that wouldn't prove very much, would it?"

Positive reporting bias may occur for several reasons. Beyond the seemingly universal desire to believe in cures, which may underlie some of the bias seen in media reports, however, are corporate interests. Journalists who seek out experts often turn to professional associations, non-profit health groups, and patient organisations as seemingly credible and disinterested. However, what many journalists don't realise, or fail to report, is that many of these groups receive substantial funding from drug companies, or have other vested interests.

Hospitals facing brutal competition to stay afloat have started buying contractsfor up to $200 000 (£140 000 or 208 000)with television stations to feature them in "news" reports, said an article in the 28 February 2000 issue of Business Week. Do such payments create pressures to exert improper influence over news reports? A poll of 300 news directors found that 43% felt such improper influence existed.

Dr Oransky believes that the sort of investigative reporting that is standard in other areas, such as political coverage, is lacking in medical and science reporting; "How many medical reporters have ever filed a FOIA [Freedom of Information Act] request? Too many science and medical journalists report researchers' assertions uncritically."

Ray Moynihan, a journalist who specialises in how the media report medical issues, believes medical journalists should report whether the experts they interview have financial conflicts of interest. "If you don't, you are arguably misleading the public and doing them a huge disserviceand that is still a very live problem."

In an interesting postscript, New York Times science writer Gina Kolata, author of the 12 September story, but not its headline, wrote a second article on 17 September, which had an entirely differentand balancedspin and tone. This article talked about the "dilemma" that men face, since surgery for prostate cancer doesn't reduce all-cause mortality.

Jeanne Lenzer, freelance journalist. 

New Paltz, New York, USA jeanne.lenzer{at}verizon.net