Follow Us

The Heart of the Matter

How Congress and Special Interests Kept Crucial Clinical Trial Data Secret

Nancy Watzman
Bill Allison
Additional research
Sarah Dorsey and Luke Rosiak
Video by
Noah Kunin
Music by
Noah Kunin, The Family Simpson

Meet the Players

  • The Patient

    Bray Patrick-Lake, mother of two and director of a Colorado nonprofit serving the homeless, joined a clinical trial testing a medical device designed to close a hole in her heart as a way to cure her debilitating migraines.

    Read more
  • The Company

    St. Jude Medical Inc. hoped to make a bundle off the Premere PFO closure device as a treatment for migraine headaches, but canceled clinical trials of the device after having difficulty recruiting patients.

    Read more
  • The Lobbyists

    The pharmaceutical and medical device industry spent millions on campaign contributions for key lawmakers and hired nearly 200 lobbyists, including many who had worked for key congressional offices or committees.

    Read more
  • The Congress

    Under heavy lobbying by the drug and medical device industries, with less than an hour to review final text, the House voted to drop a legislative provision that would have required companies to make publicly available results of clinical trials for products that do not go to market.

    Read more
  • The Approval Process

    A 2007 law requires companies to post clinical trial results for products that receive FDA approval at, but not for unapproved products. At recent hearings held by the National Institutes of Health, industry trade groups testified that trial data for unapproved products should remain secret.

    Read more


In June 2008, Bray Patrick-Lake, a mother of two and director of a Colorado nonprofit serving the homeless, volunteered for a clinical trial, regulated by the Food and Drug Administration, to test a medical device designed to close a hole in her heart and, in doing so, put an end the migraine headaches that were ruining her life.

Three months after she joined the study—which required passing through a surgical procedure to receive the device (though like all participants, she would be unsure whether or not she actually had received it), the device manufacturer, St. Jude Medical Inc., terminated the study. The company stated it had been unable to find enough volunteers for the trial.

Because trade associations including AdvaMed, which represents firms like—and including—St. Jude Medical, lobbied Congress to oppose a provision their members didn’t like, and because the House had only one hour to read the final version of the 422-page bill before floor debate and a vote was scheduled, Patrick-Lake—and the rest of us—are unable to find out what the company found out when conducting research on the implant.

Did the device alleviate migraines in test subjects? Did any test subjects report side effects? Is the device that was implanted in Patrick-Lake’s heart a boon or an extraneous foreign body? Right now, she has no way of knowing. While she considers herself cured of her migraines, further tests showed that the device (which she did receive) had not completely closed the hole in heart, leaving her vulnerable to strokes. To remove the device would require open heart surgery.

For the lobbyists and members of Congress, it was business as usual in Washington. The lobbyists got a few phrases changed in a lengthy bill—phrases that would have required public access to results of clinical trials that did not lead to an approved medical device or drug. Members of the House and Senate rushed to vote on the bill with little time to note the change. For Patrick-Lake, Washington’s standard operating procedures have left her—and the public--in the dark about the device in her heart.

What is true for Patrick-Lake is true for the thousands of people who volunteer for studies of drugs or medical devices every year that companies for one reason or another do not bring to market. What industry claims is proprietary information could be useful to doctors and patients as they decide what sort of treatment is best for any number of conditions. As Steven Nissen, chairman of the cardiology department at the Cleveland Clinic, says, “If you expose human beings to an experimental treatment, the public has a fundamental right to see the results of those experiments.”

Watch the video and read the profiles of the players, and see what happens to an ordinary American when K Street and Congress conspire to keep Americans in the dark. And bear in mind that this struggle continues: AdvaMed and other industry interests continue to push to keep some clinical trial data secret.

Return to top of page

The Patient

It wasn’t until she was six months pregnant at age 34 that Bray Patrick-Lake discovered that she was suffering from severe migraines. It was Christmas time, and she was at the shopping mall with her 18-month-old son in his stroller. Suddenly, the whole right side of her body became paralyzed and she was unable to speak. That lasted about ten minutes—and then she got a really bad headache.

She’d been sick off and on during the whole pregnancy. But she wasn’t sure what it was—pregnancy can bring on all sorts of symptoms. After the incident at the mall, however, her obstetrician sent her to a neurologist. She went through test after test, and then, finally, she got the diagnosis. And suddenly, all those headaches that she’d suffered over the previous two decades—headaches she’d written off to her sinuses or the occasional hangover—made a lot more sense.

While she was pregnant, there wasn’t much the doctors could do for her. She couldn’t go on heavy medications for fear it would harm her developing baby. After the baby arrived, she nursed her son for 13 months, so she still couldn’t go on medication because it might seep into her breast milk.

Her symptoms continued. She would slur her words, or one side of her body would get heavy. She kept getting migraines. Finally she reached the point where she was suffering from about 14 attacks a month. “A day didn’t go by where I didn’t have some kind of symptom,” she said. But Patrick-Lake, an athlete who rock climbed competitively and had lettered in four sports in school, just powered through it.

When she was finally able to start medications, she quickly discovered she couldn’t tolerate them. “They worked on my migraine, but they turned my life upside down,” she said. She’d get fits of insomnia, crying jags. She lost 20 pounds in 12 weeks. Her hair fell out. When she drove her kids to school, she’d find herself taking the wrong road. It affected her work, as director of a nonprofit serving the homeless in Longmont, Colo.

Pretty soon she was seeing a full roster of medical professionals, because nobody knew quite how to classify her. She saw a cardiologist, a pulmonologist, a primary care doctor.

She got passed around. An echocardiogram showed that she had a common heart abnormality called a patent foramen ovale (PFO)—a small hole in the heart. In utero, all fetuses have this condition, which diverts blood away from undeveloped lungs, but the hole usually closes after birth. It’s estimated that as many as one in four people may have PFOs, but in most people the condition is considered to be benign. However, recent research that shows that PFOs may be linked to migraines for at least some people.

Exasperated, she took to the Internet and started researching. That’s how she discovered that St. Jude Medical was conducting a clinical trial for its Premere PFO closure device, which was already on the market in Europe but had yet to be approved by the U.S. Food and Drug Administration.

“I could have spent $10,000 and gotten one of these devices outside the country, or I could participate in a clinical trial,” said Patrick-Lake. “Part of me felt that this should be studied, that if I went through this that other people wouldn’t have to suffer.” By participating in the trial, she would also receive the device free of charge.

Patrick-Lake officially enrolled in the study on May 26, 2008. The qualification process had been extremely stringent. She had to prove that she’d failed all other treatment for migraines. She had to keep a diary of her migraines for 30 days. She had to repeat cardiac testing to show that she, indeed, suffered from a PFO.

On June 24, 2008, she underwent a heart catheterization, in which a small plastic tube was inserted into one of her arteries and was pushed up into her heart. She was participating in a “double-blind” study, designed so that neither her doctors nor she would know whether she actually received the device. She did know that two-thirds of the study participants would receive the device, one-third would not—and that she wouldn’t find out which group she was in for a full year. Getting this implant was serious business. If a patient were ever to need it removed, she would need to undergo open heart surgery.

In the months that followed, Patrick-Lake did get some relief from her migraines. She also suffered some side effects. She developed an enormous bruise from the catheterization procedure, suffered from heart palpitations and chest pains, and developed drug-induced hepatitis presumably from the clot preventing drug Plavix that she was required to take. But she was still happy. “I would take chest pain and palpitations any day over my migraines,” she says.

Just three months after Patrick-Lake had the Premere device implanted, in September 2008, St. Jude Medical announced it was canceling the trial, because, the company said, it could not recruit enough patients. (The entry on, the government Web site that informs the public of ongoing research into medical devices and drugs, has yet to be updated to note this fact.)

“I found out from the Internet,” she said, rather than through any official notification. “It sent me on such a loop—there were so many emotions. Nobody told me, and I have this device in my body for the rest of my life. It was scary because I knew it was an investigational device.” Patrick-Lake said when she followed up with her trial contacts, they said they were going to notify her at her next regular follow up appointment. She never did receive any official letter from St. Jude about the trial cancellation.

She always strongly suspected she had the device—she says she can actually feel it in her body—but she didn’t find out officially until six months after her heart catheterization procedure, in November 2008. (She was “unblinded” after six months instead of one year because of the complications she’d been suffering.)

While Patrick-Lake believes the Premere implant cured her migraines, she also found out that it did not close the hole in her heart completely. This means she’s still considered a high risk for strokes. And among the trial participants she found through Internet chat rooms and sites, some had problems after being implanted with the device.

She wonders if St. Jude knows of problems with the device that the company is not revealing. But she has no way of finding out. “I think I have the right to know that,” she says. It upset her to find that because the trial was cancelled, St. Jude Medical was not required to release trial results to her or to anybody else. “To provide for my own best care in the future, I should know everything about this device and be able to share that with my doctor.

“Science needs patients and patients need science,” says Patrick-Lake, who wants to make clear that she’s a strong proponent of clinical trials. “But patient care and safety must come first.”

Patrick-Lake was shocked to find out later that in 2007, the year before she joined the trial, Congress had considered—and then, under the influence of lobbyists—rejected—a requirement that would have required companies to disclose data for products that don’t make it to market. When she found out the text of the final bill was available for just one hour before Congress debated it, she was outraged.

“Patients didn’t have a fighting chance,” she says. “This bill was being debated and there was only one hour to respond? What could I possibly do about that? I live in Colorado. It’s stunning that this could happen.”

Now Patrick-Lake has become an advocate for patient rights. She flew on her own dime in June 2009 to testify before the FDA’s committee on transparency, urging agency officials to do a better job of treating patients and that the FDA be more transparent. She has also joined the circulatory system devices panel of the medical devices advisory committee as a patient representative.

A spokesperson for St. Jude Medical would not comment on the situation, stating that “Because patients are still in the follow up period of the study, it is premature to comment to publications at this time.”

Return to top of page

The Company

In March 2006, St. Jude Medical Inc., a publicly traded company that makes implants to treat cardiac and neurological disorders, began testing a device it hoped would bolster the company’s bottom line in years to come and enrolled their first participant for the trial.

The device in question, the Premere PFO was designed to treat a common heart defect, called patent foramen ovale. Another company, Velocimed LLC, actually developed the Premere implant; St. Jude Medical acquired that firm in February 2005 for $82.5 million.

Though the Premere still needed the approval of the Food and Drug Administration before it could be sold in the United States when St. Jude Medical acquired Velocimed, the company believed that the device would be a money maker. In Sept. 2005, John Heinmiller, the company’s CFO and treasurer, told potential investors at a health care conference hosted by Bear Stearns that the implant was an important part of its long-term strategy.

The prospects for Premere looked good. The previous May, the company launched it and two other new cardiology devices in Europe. And in August, about a month before Heinmiller’s remarks, the FDA approved St. Jude Medical’s request to begin enrolling patients in its ESCAPE trial for suffers of extreme migraines, which would determine whether the device could help sufferers of migraine headaches.

The company initially announced that the ESCAPE trial would be complete by 2007. However, in September 2008, St. Jude Medical stopped enrollment in the ESCAPE study, a company representative said, after concluding that it would not recruit enough participants to complete the trial until 2016.

The following month, in October 2008, chief executive officer Dan Starks told investors that the ESCAPE migraine trial was closed because of difficulty in enrolling enough patients—only 50 had signed up. While “the space” of exploring PFO closure for migraine treatment “remains of interest,” he said, the regulatory requirements for enrolling patients in the study proved too great. Apparently the company found it difficult to recruit patients under the stringent study design, which required patients to undergo extensive testing and agree to a trial where they had a two-out-of-three chance of receiving the device.

“It’s not something that we’ve referred to for some time as one of our growth drivers and it would be appropriate to take it off the list of our growth drivers that we're particularly excited about,” he told investors on a third quarter earnings conference call in 2008.

A company representative later stated that the company has no intention of re-opening the study. Meanwhile, the information collected in that study from participants such as Bray Patrick-Lake remain the property of the company.

Although the Premere device never won FDA approval for treatment of migraines, St. Jude Medical continues to sell and profit from the device in Europe. The company’s Web site advertises the implant’s virtues on a Web site that cautions its information is “not intended for users in the United States or those affiliated with U.S. institutions, as this website contains information about product and therapies that are not available or authorized for use in U.S.”

Return to top of page

The Lobbyists

In 2007, when Congress was debating legislation on transparency in clinical trials, the pharmaceutical and medical device industries were well positioned to make their voices heard.

According to the Center for Responsive Politics, employees, their family members and PACs of firms in the industry contributed $3.4 million to members of the House Energy and Commerce and the Senate Health Education and Labor Committees. Those two committees had primary responsibility for drafting and debating the reauthorization bill that sought to bring greater transparency to clinical trial data, called the Medical Device User Fee Amendments Act.

The drug and device manufacturers also had a few hundred lobbyists in their employ. The three main trade associations for these interests—the Pharmaceutical Manufacturers Association of America, the Advanced Medical Technology Association (AdvaMed), and the Medical Device Manufacturers Association—had by themselves hired nearly 200 lobbyists.

A provision in the House version of the bill that would have made results from terminated clinical trials public was eliminated. But sadly, there is no federal lobbying disclosure that reveals whether a particular lobbyist or group of lobbyists had an impact on the language on clinical trial data that ended up in the final bill. According to sources who helped draft the House version of the bill, key negotiators agreed to drop the provision concerning the unapproved device's data.

No disclosure tells us whether lobbyists for AdvaMed, PhRMA or the Medical Device Manufacturers Association contacted members of the conference committee who settled on the bills final language. Lobbyists for all three trade groups did disclose that they lobbied Congress on the bill but there is no way to know who they contacted, what they discussed and what impact they actually had.

St. Jude Medical, the manufacturer of the Premere PFO Closure System, itself spent $400,000 on lobbying that year. The firm is also a member of AdvaMed, which spent $3.1 million; one of the company’s board members, Stuart M. Essig, serves on the Advamed’s board.

Among the staff and lobbyists AdvaMed hired were a few who had been co-workers and colleagues of the key congressional players they were seeking to influence:

  • David Nexon worked for more than two decades as the Democratic health policy staff director and as senior health policy advisor to Sen. Ted Kennedy before joining Advamed as senior executive vice president. Kennedy sponsored the Senate version of the Medical Device User Fee Amendments Act.
  • Julie Cohen was a legislative assistant for Sen. Herb Kohl, D-Wis., worked for the Senate Health, Education and Labor Committee (Kohl is a member) and later served as staff director for the Senate Aging Committee. She joined AdvaMed in May 2007 as vice president of government affairs.
  • Amy Jensen Cunniffe served as a legislative assistant to then-House Speaker Dennis Hastert from 1995 to 2002. She serves as director of the trade group’s government affairs staff.
  • Prior to joining the Alpine Group, Courtney Johnson worked for the House Energy and Commerce Committee, which had jurisdiction over the legislation in the House. AdvaMed hired the firm to lobby on the legislation.
  • AdvaMed also hired John McManus, whose former employers include the House Ways & Means Subcommittee on Health, Rep. Bill Thomas, and Eli Lilly.

In 2007, 11 firms reported that AdvaMed was a client; in addition, the trade group employed five in-house lobbyists. Overall, it reported spending more than $3.1 million to influence the federal government.

Return to top of page

The Congress

In April 2007, Sen. Edward Kennedy, D-Mass., introduced S. 1082, the Medical Device User Fee Amendments Act, directed toward a prosaic end: the bill would reauthorize fees that drug and medical device manufacturers pay that fund the U.S. Food and Drug Administration’s work. Without the reauthorization, the fees would expire, and Congress would have to find some other source of funding for the FDA.

Reauthorizations are typically not controversial measures, but by 2007, growing public and congressional concern stemming from a series of high profile drugs that proved dangerous despite winning FDA approval made the reauthorization bill the subject of greater scrutiny. Vioxx, developed by Merck as the next new wonder pain relief drug, was found to increase the incidence of heart attacks among users and removed from the market in 2004. That same year, the first of several studies found a link between youth suicides and common antidepressants like Paxil, made by GlaxoSmithKline, and Zoloft, sold by Pfizer.

In 2004, Congress held hearings on these approved drugs that had deadly consequences, excoriating the companies involved and the FDA for ignoring the danger signals revealed from clinical trials and other evidence. And in 2006, the General Accounting Office, Congress’ investigative arm, published a report highlighting systemic problems in the FDA’s ability to respond to dangers caused by approved drugs.

The reauthorization legislation in 2007 became a vehicle for addressing some of these wrongs. Among the remedies proposed was more transparency, which Kennedy highlighted on the Senate floor:

“In many cases, companies have hidden evidence of safety problems. Our bill addresses this abuse by including a public database of all clinical trials and their results. Listen to that: all clinical trials and their results. We protect the trademark aspects of the particular item but require the publication of all clinical trials and their results. A company will no longer be able to hide the results if they do not show what the company wanted.”

A government Web site called had been posting ongoing clinical trials by companies seeking FDA approval for drugs or medical devices meant to treat “serious or life-threatening diseases or conditions” since 1997. However, that site only provided lists of ongoing trials.

The Senate legislation required that companies report the results of their clinical trials when they were finished, provided that the drugs or devices received FDA approval.

In the House, however, lawmakers took a different approach to the clinical trials database. Like its Senate counterpart, the House version of the bill, H.R. 2900, included language that made more information available on clinical trials, including the publication of results from these trials. But the House bill specifically stated that the results must be made available to the public whether or not the drug or device being tested had been approved by the FDA.

Consumer advocates argued that it was important to inform patients of results of trials even for products that are not approved. “If you expose human beings to an experimental treatment, the public has a fundamental right to see the results of those experiments,” said Dr. Steven Nissen, chairman of the cardiology department at the Cleveland Clinic, who helped develop the House proposal.

Medical device and drug manufacturers far preferred the Senate version, arguing that publishing information about products that had not received approval could hurt their bottom line. “The current language…would harm device innovation without any benefit to patients,” testified Stephen J. Ubl, president and CEO of the Advanced Medical Technology Association (AdvaMed), “When there is no FDA-approved product, information related to the device design and to the design of the trial and its endpoints is the only intellectual property a company may have. Such disclosures could have the unintended consequence of eliminating many small device companies from the marketplace." [PDF]

On May 9, the Senate passed its version of the legislation, S. 1082, by a vote of 93 to 1. On July 11, the House passed its version of the bill—including the broader transparency requirement—by a vote of 403 to 16. Then the bill sat. And sat. Meanwhile, lobbyists for interests opposed to the transparency provision, including a trade group that included St. Jude Medical Inc. among its members, pushed their point of view. [See “The Lobbyists”]

The bill sat in limbo until September, when Congress faced a crucial deadline. The user fee provisions in the law were about to run out, and unless Congress reauthorized them—that is, passed the new version of the bill—by the end of the month, the FDA couldn’t collect from drug and device makers the funding it needed to continue its safety reviews of drugs and medical devices.

On September 19, a new version of the bill emerged, some 422 pages long. The bill was available for just one hour before House members began debate. The House voted 405 to 7 to approve it, and the Senate followed suit the next day, passing it by unanimous consent.

The House language, which would have required that results data for unapproved products be made available, was gone. The new language looked a lot more like the Senate version—except now there was explicit language about “unapproved products.” According to the bill, it would be up to federal agencies to determine through regulations whether any data would be made available for products that are not approved by the FDA.

According to sources who helped draft the House version of the bill, key negotiators agreed to drop the provision concerning the unapproved devices data. When the final language went to the full House, there was only an hour to review it before the debate; no member objected to the change.

But the question of who should have access to data collected from patients participating in terminated clinical trials is still being raised. In April 2009, the NIH held a public hearing for comments as it develops new regulations. The same interests that weighed in on the reauthorization legislation have commented—AdvaMed and PhRMA among them.

“Medical device trials may be stopped for reasons unrelated to safety but rather due to an inability to obtain continued financing, or due to changed company priorities,” wrote Tara Federici, vice president of technology and regulatory affairs, in submitted testimony. “[Requring disclosure] may also disclose confidential or proprietary information to a company’s competitors preventing the original company from successfully pursuing the affected technology later. Because the product was never FDA approved or cleared, the device cannot be marketed and thus cannot present risks to patients or the public.”

However, as the case of Bray Patrick-Lake shows, a device may well present risks to a patient even if the device never goes to market.

“Information gained from all trials involving human subjects should be part of the world’s scientific database,” William Vaughn, a health policy analyst for Consumers Union, testified at the hearing. “Knowing what doesn’t work is in many ways as important as knowing what does.”

“Information learned in this study will be useful to all doctors who take care of patients with migraine.”-- Quote from the consent forms that Bray Patrick-Lake signed when enrolling in St. Jude Medical’s clinical trial of Premere PFO implant.

Return to top of page

The Approval Process

The Food and Drug Administration’s drug and medical device approval process is designed to protect the public from cures that do more harm than good. However, sometimes crucial data collected along the way that could save or improve lives are not made public. The thousands of people who volunteer their bodies for clinical trials are literally left in the dark, as are the doctors who must make decisions about how to treat their patients.

People choose to enroll in clinical trials in the hope of finding some relief from severe pain or a bit of hope in warding off a deadly illness. At such a vulnerable moment, they may not always check all the fine print when they sign the papers to join a study, and often are not aware of all the ramifications involved in the byzantine agency drug and medical device approval process.

While researchers are supposed to notify patients if there are any dangers that emerge in studying a device or drug, and are obliged to follow patients who participate all the way through a trial, companies do not always live up to this.

For example, in 2007, the FDA sent a warning to Boston Scientific after investigators discovered that the company stopped following up with patients implanted with an experimental stent meant to prevent rupturing in aneurysms. The company, which stopped studying the product after finding out it frequently fractured, did not tell patients about the full scope of the problem, the New York Times reported. For patients who participate in medical device trials, lacking this information is particularly disturbing—after all, they have these devices implanted in their bodies indefinitely.

The agency suffers from severe shortcomings in following up on medical devices after they reach the market place, according to June 2009 testimony by Marcia Crosse of the U.S. Government Accountability Office (GAO) before a Congressional committee. While the number of “adverse events” associated with medical devices has increased substantially in recent years, the FDA lacks the resources to review all the submitted reports in a timely manner, “ she said. “FDA officials told us that while they have a number of strategies to prioritize their reviews of adverse event reports, they still cannot review all the reports they receive.”

Often clinical data collected for devices or drugs that don’t make to the market could provide crucial information—however, this information is not made generally available to the public.

For example, ciglatazone, a drug that was the first in a particular class of drugs developed to treat diabetes, was shown to cause liver damage in clinical trials, and never reached the market. Another drug from the same class, troglitazone, received approval. Later it was also found to cause liver damage. The data from these trials was not made public.

“Knowing the adverse event profile of ciglitazone and others in the class, might have cued clinicians and drug safety experts to look earlier and more closely for these adverse events,” wrote Allan Coukell a pharmacist and director of the Pew Prescription Project, in recent comments to the NIH.

There is no requirement that a company or the FDA release information about clinical trials if the drug or medical device in question is not approved for the market place. Congress considered such a requirement in 2007, when passing a law reauthorizing industry user fees that help fund the agency’s work, but dropped it under heavy lobbying by the drug and medical device industries.

Medical device process

  • How risky is the medical device? The riskiest are given a class III rating—those that “support or sustain human life, are of substantial importance in preventing impairment of human health, or which present a potential, unreasonable risk of illness or injury.” (Recently the agency has come under harsh criticism for approving devices needing more scrutiny under the fast track “510(k)” procedure designed for less risky devices.)
  • Class III medical devices must receive “premarket approval” from the FDA. This means that companies must prove that a device is safe and effective, typically by conducting clinical trials.
  • Companies testing the riskiest medical devices often must apply for an “investigational device exemption” that allows them to give devices to people enrolled in clinical trials.
  • Companies are required to register a clinical trial on the government database soon after the first patient is enrolled in the study.
  • Companies seeking approval for devices must submit supporting information to the FDA for analysis and, in most cases, review by and advisory panel of outside experts.
  • Within 180 days of an application, FDA issues a decision about whether to approve or deny application to bring product to the market. These approval documents are generally available in PDF format on the FDA Web site.
  • A 2007 law requires companies to report the results from clinical trials at for all approved products. It is under the discretion of the agencies whether they must also report the results for products that never make it to the market place.
Return to top of page