Unpublished Study Pegs Cost Of New Drug At $231 Million

BOSTON - An unpublished study that examines the cost of bringing a new drug to market has revived debate over whether data should be released to the public prior to peer review. The study, which cites a figure that is nearly twice as high as previous estimates, is already being used in ongoing discussions over federal policies on drug pricing and the investment in research by the pharmaceutical industry. The just-completed analysis by the Tufts University Center for the Study of Drug Development reported that it takes 12 years and the equivalent of $231 million (in 1987 dollars) to bring a single new prescription drug to market. The cost cited is more than double that of 10 years ago, even after adjusting for inflation, and the length of time is two years longer than previously judged. About half of the total consists of "time costs," or money lost by forgoing other investments, while the rest - $114 million - is actual out-of-pocket expenses. At a time when federal and state legislators are trying to hold the line on reimbursement expenses for prescription drugs, the Tufts study is being embraced by the pharmaceutical industry to help justify current drug prices. For instance, the Pharmaceutical Manufacturers Association (PMA), a trade group representing more than 100 United States firms, sponsored a joint news conference with Tufts on April 19 in Washington, D.C., to publicize the center's findings. A PMA spokeswoman predicts, "We believe this will be the definitive study on drug development." At press time, the Tufts survey had not yet been submitted for publication to any professional peer-review journal. PMA has also incorporated the new figure in an advertisement that appeared on the op-ed pages of the Washington Post and other papers as part of a series entitled "Innovations in Medicine." Overall, the ads convey the message that drug companies are spending large sums to develop new drugs but that costs are rising even more quickly than the potential payoff in sales. Although it does not address those issues, the new study does draw attention to the propriety of releasing unreviewed data via the media. In its simplest form, the issue is whether research that has not been checked by experts in the field should be used in setting broad social policy. After nearly two years of organizing the survey, selecting participants, collecting raw data, and analyzing the results, why would investigators release the study before submitting it for publication? The report's lead author, Tufts economist Joseph A. DiMasi, concedes that the apparent rush to go public is out of the ordinary. But he defends the center's actions as a necessary response to slow leaks of the results. "The review and publication process is a lengthy one," DiMasi explains. "There's a lot of industry and congressional interest at this time. Also, one of the coauthors had mentioned it [the survey] at a meeting, and the numbers set in a couple of people's minds. To set the record straight," he says, "we decided to go public with our figures at this time." Before the recent press conference, drug company officials had been using a figure of $125 million for the cost of a new drug. That figure is based on a 1987 study by Stephen Wiggins of Texas A&M University that was commissioned by PMA. It has never been published in a peer-reviewed journal. Numerous recent examples show the confusion that can occur when investigators deliver data to the public before the findings are reviewed by experts. The foremost case, according to Bruce Lewenstein, a Cornell University science historian, may be the worldwide attention devoted to prepublication announcements by the University of Utah of experiments by B. Stanley Pons and Martin Fleischmann that purported to prove the existence of room-temperature nuclear fusion. "For the average person, the concern is how reliable the information is," says Lewenstein, who is studying the press-science connection in the cold fusion affair. "No piece of scientific information is certain, but the more kinds of review it has gone through, the more the core of the information that is talked about will hold up - that it won't disappear due to trivial error." PMA spokesman Jeffrey C. Warren seems to agree. But he indicates that PMA has clear reasons for wanting the results made public as soon as possible, even before any kind of peer review. "It would have been our preference if the article had been published, and we think it will be," Warren says. "But we think [the information] is important. The cost of these drugs is escalating so rapidly, we thought it important to communicate this data, this snapshot, to the congressional community." But such a strategy is denounced by critics of the drug industry. Sidney Wolfe, director of the Washington-based Public Citizen Health Research Group, says he is skeptical about the results because the Center for the Study of Drug Development gets financial support from the pharmaceutical industry. And its decision to present the results at a press conference only adds to his doubts. "I don't trust the study at all, Wolf says. By ignoring peer review and going directly to the media with the results, he adds, firms are trying to "gain more legislative changes favorable to the drug industry. They're saying, `We won't develop drugs unless you give us this or that.' It's blackmail." As it happens, Congress has already launched its own independent investigation into the cost of developing new drugs. Last fall the congressional Office of Technology Assessment (OTA) began such a project, due to be completed in May 1991. Project director Judith L. Wagner explains that Congress wanted to answer "a larger question of what's happening in the pharmaceutical R&D environment that is driving costs and returns." OTA hopes to develop a model for the best way to quantify the cost of developing new drug products. DiMasi and his coauthors - economist Henry G. Grabowski and pharmacologist Louis Lasagna, who are affiliated with the Tufts Center, and economist Ronald W. Hansen of the University of Rochester - have turned over their study to OTA at its request so the agency can include the report in its general review of methodology. "What we were asked to do was to look at the existing evidence, to comment and critique the methodology and to design a study that would provide more accurate estimates - if we determined that was needed," Wagner says. Until that study is completed, few on Capitol Hill seem willing to comment specifically on the Tufts findings. A spokesman for Rep. John D. Dingell (D-Mich.), chairman of the House committee that has jurisdiction over the Food and Drug Administration, declined comment until the study is published, but he discounted fears that unreviewed data would be accepted blindly just because they have been widely publicized. "Numbers do get used in congressional debates," he points out. "But every side has a different number." Even PMA's Warren says that it's the general trend in expenses, not one number, that's of greatest interest to legislators. "I do think that for the purpose of policy-making, it's the rapid escalation and not a particular figure that's important." Jack Jackson writes about science and business issues from Boston.

Unpublished Study Pegs Cost Of New Drug At $231 Million

Interested in reading more?

Become a Member of