Sunday, October 05, 2008

Media coverage of drug studies lacking

by Elizabeth Cooney
05 oct 2008--The news media isn't doing a very good job when it comes to reporting on drug studies published in leading scientific journals, Cambridge researchers report. Journalists too often let readers down by leaving out information that might reflect bias in the results, and use brand names that might implicitly promote the drug in question, the researchers conclude.
Almost half of the stories in newspapers and on online news sites neglected to mention pharmaceutical company funding for drug trials whose results were published in high-impact medical journals, according to a review of more than 300 news articles conducted by Dr. Michael Hochman and colleagues from Cambridge Health Alliance and Harvard Medical School. Their study was paid for by Cambridge Hospital. Hochman, a third-year resident in internal medicine, is a contributor to The Boston Globe.
More than a third of the news stories used only the brand name of the drug, the review also found.
"As a doctor I've become increasingly worried in recent years that company-funded research, to put it simply, can't be trusted. The rofecoxib scandal -- better known as the Vioxx scandal -- is one example of this," Hochman said in an interview. "I am worried that lay readers and patients don't always look at it this way and I think it's very important when patients or readers read a medical story that they are at least aware of who sponsored it and what the motives are of who sponsored it."
All the trials in question were paid for by drug companies and all had their funding sources included in the journal articles. A growing number of scientific journals require authors to make disclosures in order to reveal potential conflicts of interest.
Despite this showing, when editors at the top 100 newspapers were surveyed about their coverage, 82 of the 93 who responded said their stories always or often indicated drug-company funding, and 71 editors said they used generic names rather than brand names. Only three newspapers had written policies requiring reporters to include company funding of drug studies in their stories. Only two newspapers had written policies mandating the use of generic names rather than brand names for drugs.
The Boston Globe's written policy states that health/science reporters must always ask the doctors and scientists they interview whether they have any financial stake in a study, even if they have already made a disclosure to the journal in which their work is published. Conflicts should be included in the news story, the policy says, as well as whether a researcher or physician declined to answer questions about it. Reporters must also always determine who paid for a study and what control they had over how it was conducted or described in a paper.
The Globe's policy does not address using generic vs. brand names in its stories.
Hochman's article appears in the Journal of the American Medical Association, one of the five journals the Cambridge researchers reviewed from 2004 through 2008. The others are the New England Journal of Medicine, The Lancet, Annals of Internal Medicine and Archives of Internal Medicine. Studies have shown that company-funded research is more likely to have favorable results than non company-funded research, Hochman said.
"If a company sponsors a study, they are doing it so that it promotes the product," he said. "When I read a newspaper, I see a lot of times they don't do a very good job of indicating how the study was funded."
Journals could make funding disclosures more prominent, Hochman said, including them in scientific paper abstracts, the short descriptions that may be all a reader sees. Press releases sent to news outlets should --- but almost never do -- include information on who funded a study, Hochman's team also found.
Reporters aren't alone in using brand names, Hochman said.
"I think all of us -- doctors, patients, and the news media -- have gotten into what I would consider to be bad habit of referring to medications by brand name rather than generic name," he said. "The commercial interest should be as separate as possible from medical decision making."

No comments: