April 14, 2006 | In the past few years, the pharmaceutical industry has been forced by medical journal editors and Congress to make public more data about clinical trials. Long-secret data — lots of it — is now available online. Registries are here, most people are used to them, and they’re not going away. But can the industry do registries and databases better?

This month, the Center For Business Intelligence convenes a meeting on the topic in Arlington, Virginia, which will try to help attendees grasp the nuances of both public registries and results databases. Crucially, the conference has representatives from every major constituency in the clinical trial debate: sponsors, medical journals, patient advocacy groups, and regulatory agencies from the United States and abroad.

The perspective of Hoffman-La Roche is especially interesting. More than two-dozen people at the Swiss company put their heads together and decided that making the information available to the public was paramount. The company worked out an arrangement with the publishing and consulting firm Centerwatch. In addition to its paid services, Centerwatch has a wealth of free public information on its site, and Roche decided that would ensure its data had the highest possible visibility.

In less than a year, after spending a million francs just to tweak its IT systems, Roche has put data from 300 trials on its Centerwatch site, with half a dozen full-time employees working on just registries and results databases. Roche’s Richard Barr, director of strategic communication, admits that the needs of registries and databases are in flux. He sees the conference as an opportunity to look at the raison d’être for these new data reservoirs. The initial reason for a registry, he notes, was to help patients find trials. Then medical journal editors had concerns.

Now there are additional reasons. “It’s morphed into this general discussion about transparency,” he says. “What kind of transparency are we looking for? What needs to be transparent to whom, when, and how?” It will be impossible to determine success, he suggests, until there is some industrywide agreement on what the right objective is.

Signifying Nothing
Barr has concerns about many registries — including, it would seem, his own company’s. “We risk being more opaque than transparent,” he says. “By putting everything out there, we may communicate nothing. We’re putting so much data and so many kinds of data out there that the ultimate goal of trying to help patients remains a challenge.”

He says there is evidence that ordinary patients find navigating all registries difficult: “People are having a hard time finding the information they want and knowing what to do based on that information.” Citing research of ordinary visitors to the ClinicalTrials.gov Web site, he says many patients have many more questions than that site can answer. “All parties contributing to registries need to ensure that users are at the table when creating, and refining, purportedly user-friendly systems,” Barr says. “If we better understand what users are looking for, we have a better chance of providing it.”

Barr says that one lesson of Roche’s registry has been that flexibility needs to be built into any design. Politics, from any direction, may require changes on the fly. “The ground rules have changed and changed again,” he says. “It is such a mobile situation. The original goal has morphed. The one thing we learned was that you must design a flexible solution; otherwise, your solution will be a problem six months later.”

Cheryl Karol is global director of Roche’s department of ethics in clinical research. She candidly concedes that the spate of news reports about clinical trials has left the reputation of the entire industry in tatters. The trust of the public, clearly, is vital for the industry’s ability to recruit patients. “We just don’t have it,” she says. “It’s a shame. It’s important to have that restored. We need to regain that public trust.”

She and most readers are familiar with sizable efforts by industry to comply with the requests of medical journal editors. Karol says that it may be premature to suggest there is a thaw in the relations between the industry and the academic editors. But she expresses the hope that they will understand how extensive, and sincere, the industry’s effort to publish details about trials really is.

Karol notes that registries and databases and portals are proliferating. Some of the consequences of that are good: There are more data for patients and physicians to find. But there is the potential for confusion, even bewilderment, at such a profusion of data. One worthy and laudable site, Karol says, is run by the International Federation of Pharmaceutical Manufacturers & Associations. But Karol notes the public still has some lessons to learn about the clinical trial process. “Most people don’t understand the clinical trial process at all,” she says.