The Yale University Open Data Access Project made clinical trial data for medical devices and diagnostics publicly available for the first time on Jan. 14.
YODA’s partnership with pharmaceutical giant Johnson & Johnson, which began with an agreement to share pharmaceutical trial data on Jan. 6, will allow medical researchers to access the company’s clinical trial results for medical devices and diagnostics, providing those researchers with troves of medical data they had not been able to access until now. YODA, acting as an independent intermediary, will review proposals from investigators to view anonymous patient data and grant or deny access.
Although this new step is meant to address the fact that only around a third of experiments involving human subjects are openly published, it faced significant opposition during its development.
“People told us no one would ever want to work with us,” said Director of the Center for Outcomes Research & Evaluation at Yale-New Haven Hospital Harlan Krumholz ’80, who directs YODA. “Now we have Johnson & Johnson.”
The YODA team was developed out of a concern for what they viewed to be a prevailing culture of exclusivity among medical industry researchers.
Joseph Ross MED ’06 professor of medicine and public health and YODA co-principal investigator, said there is an expectation that when researchers conduct a trial, the results belong to them and not the public.
Both researchers said they first got directly involved in YODA after reviewing data for the Vioxx case, a suit bringing charges against Merck & Company for distributing a painkiller that was found to be harmful. Krumholz said that as he analyzed the data for the Vioxx case, he noticed that the risks associated with the product could have been identified sooner if unbiased investigators had been able to review Merck’s data without depending on the company for access.
“I learned about how a lot of science is openly conducted but not shared. In many cases, it gets buried,” Krumholz said, adding that people at Merck knew Vioxx was harmful before it was taken off the market, but were able to withhold that information due to the opaque nature of medical industry research.
Nihar Desai, assistant professor of cardiology and member of YODA leadership, said concerns about the efficacy and safety of drugs and devices can only be understood if relevant data are made public.
But Ross said that despite the clear advantages of transparent clinical data, opposition to the YODA project came from many angles. He noted that regulators, clinical trialists and pharmaceutical companies have not always been supportive of their mission.
According to Krumholz, those who were more hesitant were academic researchers, who have incentives to keep the data they uncover private. In a study of National Institute of Health funded trials, Krumholz saw that investigators are not even required to share their findings with the NIH if they are not published.
While a junior faculty member, Krumholz approached a number of investigators who owned data he wanted to work with. He either got no response or a resounding “No.”
“The culture is that it’s rude to ask,” Krumholz said.
Desai agreed that it must be frustrating to know that the data one needs to answer a question exists but is unavailable. He noted that overcoming the prevailing culture of secrecy has been one of the most significant hurdles in forming YODA.
Desai noted that the YODA project has focused on striking the right balance between keeping science transparent and being respectful of the diligence required by investigators to design and execute large clinical trials.
Ross emphasized that sharing company data will improve future science research. Only around a third of research on human subjects is published — the other two thirds, he said, are not being used to inform science. Publishing negative results is as important as publishing positive ones, as it prevents other researchers from duplicating studies that have already been conducted, he said.
For Desai, the service YODA does for patients who participate in trials is especially important. Patients take part in trials despite the risks because they want to help future patients, he said. Making trial data publicly available does justice to them — they have a right to know the outcome of a study in which they participated, he said.
“These patients are doing society’s work by participating [in clinical trials],” he said. “We should celebrate the sacrifices that patients make.”
Currently, the YODA project only discloses results to investigators, but Desai said that it would be good if, eventually, a mechanism was created for patients to also be able to see such data.
While the current landscape of “territoriality in medicine” has made the YODA project an uphill battle, Desai said it is important to remember that scientific research is a public good.
Currently, there are 90 clinical trials listed for request on the YODA Project website, based on collaborations with Medtronic and Janssen, a subsidiary of Johnson & Johnson.