By Mark D. Uehling
August 18, 2004 | Apple pie. Motherhood. The flag. All three are somewhat popular. But in the life sciences, a far more eagerly embraced topic is a mammoth database for clinical trial data. Elite medical journals are on board, as is the august American Medical Association. The Congress believes it already required exactly such a database. Problem solved?
No. Two leading companies, Merck and GlaxoSmithKline, will use their own sanitized databases. Other big sponsors of research have not announced their plans. A third approach, a federal Web site called ClinicalTrials.gov, is designed to help patients find trials -- not to help scientists understand the science in trials. Repurposing it to aid science would be like tinkering with a canoe to make it highway-worthy.
On the political front, Washington lawmakers have consistently chosen not to provide adequate money for FDA programs related to public safety; those are now funded by the industry user fees instead. Congress will not fund a database for clinical trials during a Republican administration. In the event of a Democratic administration in 2005, a year or more may be needed to find an FDA chief satisfactory to the industry, at which point the discussion will begin anew.
Finally, there are technical issues that make an industrywide database all but impossible. Agreeing on a data model and standard terminology is rare within individual companies in the industry, where chiefs of individual therapeutic areas such as cardiovascular disease or central nervous systems still wield significant power. Getting the entire industry and the government to agree on the structure of such a database is something that has never occurred before.
The trigger for calls for a new database was fresh reports of well-managed psychiatric data -- for both adults and children. Shocking many in Washington, pharmaceutical companies were charged with putting a few scientific facts in the best scientific light. The Lancet weighed in with the following: "By providing only partial information about safety and efficacy, GSK caused doctors to have a biased and misleading picture of the drug. Doctors were thus unable to assess the balance of its risks and benefits and could not discharge their professional duties to their patients."
Strong words, to be sure. And there's no doubt that the pharmaceutical industry controls the vast majority of biomedical research. A study at Yale University indicated that 62 percent of research was funded by corporations and that such studies are 3.6 times more likely to support the sponsor's product than studies by scientists funded by government. Scientific journals do their best to sort out authors on the payrolls of one company or another, but the deck is stacked against them -- negative results seldom are submitted for publication, and only pharma-friendly scientists are picked to run trials in the first place.
But the economics of clinical trials cannot be changed with a stern editorial. No matter how many journal editors wring their hands, the fact remains that the industry funds the research and thereby controls the people who conduct it -- and any data that emerge. The only possible role a large clinical database might play is in public relations. There it might well serve to distract the industry’s critics -- however briefly -- from drug prices charged to U.S. consumers.