Sunlight Foundation Reporting Group

FDA Holds Back Information on 9 of the 25 Most Prescribed Drugs

When your doctor makes a decision about which popular prescription drug to give youwhether you're suffering from high cholesterol, asthma, ulcers, or some other common ailmentit's quite likely that he or she lacks easy access to basic research data about the drug.

A review by the Sunlight Foundation found that safety and efficacy information analyzed as part of the federal drug approval process is not available online for nine out of the top 25 prescribed brand name pharmaceuticals in the United Statespopular drugs such as Lipitor, Plavix, and Synthroid. Of those reviews that are made available on-line, the great majority were approved after 1997.

The U.S. Food and Drug Administration (FDA) Web site makes drug approval background documents available online only for drugs that were approved from 1998 on, according to spokesperson Sandy Walsh. Some information for drugs approved in 1997 or earlier is online, but only if someonea researcher, a physician, a citizen, a journalistmade a formal Freedom of Information Act (FOIA) request for the information. Even then, in some cases the agency does not plan to post older information on-line because of the poor quality of some of the preserved documents.

That still leaves approval information on many popular drugs that treat common conditions behind closed doors. Researchers or consumers interested in obtaining this information must file Freedom of Information Act (FOIA) requests and wait for weeks or months to receive the information.

It's unclear whether such information would be affected by the Obama Administration's new Open Government Directive, which asks agencies to identify and publish at least three high value datasets not yet available on-line at Data.gov and also develop an open government planessentially, a plan to release other public datain about six months. But proponents of more transparency in the pharmaceutical industry argue that such data should be a prime candidate for release.

"Absolutely, they should be included," said Dr. Steven Nissen, chairman of the department of cardiovascular medicine at the Cleveland Clinic Foundation, when asked if basic drug review information for older drugs should be on the FDA's list for inclusion under the new directive. At the time of this posting, a request for comment from the FDA had not been answered.

As of now, the FDA posts approval information on drugs online in an awkward PDF image format that makes it difficult for researchers or the public to use it. The reason for that, according to testimony by Dr. Erick H. Turner, who formerly worked as a drug reviewer for the agency, is that FDA staff typically receive the documents in an electronic format, print out the documents and physically redact certain information, then turn the documents back in to PDFs. The published PDFs are not hyper-linked or text searchable, and therefore are hard to navigate.

"When a drug is approved and a review reaches the FOIA office, rather than redact it electronically, the staff apparently prints out a hard copy, redacts it "manually" using White-Out, then scans it in," wrote Turner. "The result is a very long image PDF document in which all the textual information has been lost. The only way to find anything in such a document is the low-tech method of visually scanning each page."

In addition, posted documents often have sections that are redacted by agency reviewers, usually because such information is deemed proprietary. For example, information for one drug, the antibiotic Levaquin, included more than 58 pages that were blacked out by agency reviewers. There is no way to know why this information was removed.

The background documents, which analyze studies submitted by companies as part of the drug approval process, can be quite useful to researchers and the public. Doctors base their decisions on how to treat patients by reading the available medical literature, typically in peer-reviewed journals such as Journal of the American Medical Association and the New England Journal of Medicine. But drug companies lack incentive to publish negative information in such journals about their products, say experts.

The studies analyzed by the FDA, however, which are not necessarily published in these journals, can provide crucial insights for medical professionals. In a much noted study published last year in the New England Journal of Medicine, researchers, including Dr. Turner, analyzed data contained in FDA documents for 12 popular antidepressantseight of which they had to obtain via FOIAand found that several commonly prescribed anti-depressants dont work as well as the literature published in major medical journals suggests. Indeed, of 71 FDA-registered studies the authors surveyed, 31 percent were never published, including 22 studies that showed results that the FDA had deemed negative or questionable.

Below are the top 25 prescribed drugs in the U.S. in 2008 (as reported on Drugs.com, a free privately-run Web site that provides drug information to consumers and health care professionals) with those for which FDA review documents are not available online indicated with colored rows.