Scientists draw line with study sponsors Medical journals won't let drug firms write conclusions

BOSTON--The world's top medical journals will set a new policythat gives final say on the conclusions of studies to researchers whoconduct the work, not the drug companies that pay for them.

The editor of one journal said the rules, which will be publishednext month, will give scientists "a lot of clout they don't have now"when dealing with drug company sponsors.

The new rules will require that authors of studies have controlover the content of reports submitted for journal publication andthat they have access to all the data gathered.

The stakes in such research are enormous, because drug studiesoften cost tens of millions of dollars to conduct. Their conclusionscan determine whether new medicines get approved by the Food and DrugAdministration.

Sponsors' influence on studies may be a matter of emphasis, saidDr. Jeffrey Drazen, editor of the New England Journal of Medicine.

"I don't think people are blatantly lying," he said. "But are yougetting the whole story? People try to put a spin on things. We wouldprefer to have the spin applied by an academic investigator whoseonly interest is human health."

A spokesman for a pharmaceutical trade group in Washington saidthe journals are overreacting.

"In the vast majority of cases, these studies are conductedsoundly and with scientific credibility," said Jeff Trewhitt of thePharmaceutical Research and Manufacturers of America. "If you playfast and loose with the data, it will catch up with you eventually.If you lose credibility with the FDA, doctors and patients, you havea major problem."

Typically drug firms contract with physicians and medicalinstitutions to carry out big studies of medicines. In theory, thephysicians are largely independent of their sponsors and reachconclusions they believe are accurate.

However, the drug company sponsors often compile the research incomputer databases and help write about the findings for submissionto journals.

In a few reported cases, companies have attempted to blockpublication of unfavorable results.

"That is most obviously of concern," said Dr. Frank Davidoff,former editor of the Annals of Internal Medicine. "Nobody knows howoften agreements get written that way. The number of egregiousexamples is small, but the sense is a lot more of that goes on thananyone is able to document."