At HealthNewsReview.org, Gary Schwitzer has a roundup of coverage of a study in the Annals of Oncology entitled, "Bias in reporting of end points of efficacy and toxicity in randomized, clinical trials for women with breast cancer." Schwitzer headlines his post, "Positive spin on cancer studies–warnings to researchers, journalists, and the public."
The article's authors, from the University of Toronto, reviewed 164 Phase III randomized clinical trials of efficacy and toxicity for women with breast cancer. They found that 33% of the studies were biased in their reporting of the outcome, and 67% in the reporting of toxicity of the treatment in question. Researchers are, in other words, making their results look better than they are. The paper's authors describe this as "spin," which, being academics, they define for us: "Spin, a type of bias, is defined as use of reporting strategies to highlight that the experimental treatment is beneficial, despite a statistically non-significant difference in the primary outcome, or to distract the reader from statistically non-significant results." And such bias is prevalent, they conclude.
Schwitzer recaps reporting by Reuters and by Canadian Press, which quotes the study's principal author, Ian Tannock, an oncologist, as saying that many researchers are "like the politicians. Trying to make things look better than they are." Harsh, but backed up by his findings.
What should be done about this?
Schwitzer quotes a journalist who tweeted, "The real answer is fewer study-based stories. We need to break the cycle." Schwitzer's response: "I agree."
I get the point; stories on individual studies can create a misleading impression, particularly if they fail to provide appropriate context, noting whether a particular study, say, conforms with many others or parts ways with the conventional wisdom. Such stories should also stress the limitations of the studies and disclose subterfuge of the kind Tannock is reporting whenever they see it. And many do not.
But stop doing such stories, or "breaking the cycle," as Schwitzer's tweeter puts it? I disagree. I spent many years at the AP and Business Week covering spot news and writing countless stories on new studies. I tried my best to report them accurately. But if I had decided not to do them at all, I would have felt I were substituting my judgment for that of my readers. Who am I to decide what they should and shouldn't read?
The information is public. Readers will find it if they look. A reporter covering spot news can try to help them interpret new findings and assess their significance.
Writing bad stories helps no one; nor do bad studies of the kind described here. But writing no story at all is not the answer, in my view.
-Paul Raeburn
Leave a Reply