Three days ago, in a media blitz reminiscent of the Women’s Health Initiative, the National Cancer Institute released preliminary findings of the STAR trial in the most prestigious, rigourously peer-reviewed scientific (not) journal in existance – The American Media. For those of you who haven’t heard, the STAR Trial is a head-to-head comparison of Tamoxifen and Raloxifene (Evista) in preventing breast cancers in high risk post-menopausal women. Everyone has been waiting to see if Evista would prove to be as effective as Tamoxifen in preventing breast cancers without the same risks of uterine cancer and blood clotting.
I had intially intended this post to be a summary of the STAR Trial findings, and why I am thrilled to hear them. But when I went to find the journal article where the results were published, I discovered that there is no paper. There is not even a meeting abstract. Just a press release that tells us the “results are being submitted for publication”.
This appears to be yet another in a new and, in my opinion, disturbing trend among medical scientists – the announcement of study data by press release rather than by publication in a peer review journal.
Look, you want to hold a press conference on your cloned sheep? Be my guest. If your data sucks, you’ll get caught at publication, and no one gets hurt (except the sheep). But when you study involves humans, and expecially when it also deals with the incredibly sensitive and inflammatory topic of breast cancer, your data should be released in the forum best-designed to allow physicians and patients to responsibly and reliably assess the study findings – the peer reviewed medical journal. To do otherwise is unfair to the American public and their physicians.
Don’t tell me your data is too important to wait for the peer review process. Most journals have a fast track for important timely studies. But the study results are always embargoed from the press until publication.
Don’t get me wrong. If the STAR trial results stand up to peer review, I’ll be the first to cheer. But you tell me – How am I supposed to counsel my patients about this data when it is being presented without the oversight of peer review, and with no discussion of study methods, statistical analysis or data interpretation?
Trust me, my patients want to talk to me about this study. I’ve had five phone calls abut it so far, and more, I know, in the coming weeks.
What am I to say? “Sorry, Mrs. Daughter of a Mother with Breast Cancer who lies awake at night worried that she’s next in the cancer line. Call me back in June. That ‘s when the abstract is being presented at the American Society for Clinical Oncology in Atlanta. Better yet, let’s wait for the paper. An abstract really is not something on which to base medical care.”
When will the paper be published? Gee, I don’t know because it hasn’t even been submitted yet!
I pity the poor person who has to review that paper. Judge the data to be wrong, and you set off shockwaves among the public, who already think they know the trial results. Criticize the statistical methods, and you’ll just confuse everyone. Rubber stamp it, and you lose your credibility. I’m telling you, the researchers had better be right on this study. Because the American female public, still staggering from the roller coaster ride of the Women’s Health Initiative, can’t take another one.
The Roller Coaster at Coney Island
And one more thing. I find it fascinating that the STAR data were released just 3 days prior to Evista manufacturer Eli Lily’s quarterly results, and 1 week prior to the stockholder’s annual meeting.
Oh, I’m sorry. I’ll stop…
Last updated 03/7/10
