A new report from the Institute of Medicine urges companies, government agencies and other research institutions to share data from clinical trials in an effort to make it easier for other researchers to independently verify and reproduce results. Clinical trials are studies that test the safety and effectiveness of new health interventions, such as drugs and therapies, including mental health therapies. The 249-page report,opens in new window released in January, calls on stakeholders in clinical trials to "foster a culture in which data sharing is the expected norm."
In particular, the report calls for researchers to have a plan, prior to beginning a study, for when and where they will release their results. That plan should include releasing summary-level results, including information on adverse events, no later than one year after a trial's end. It also calls for researchers to release complete data packages within 18 months of a trial's end, or, in the case of products subject to regulatory approval, within 30 days after obtaining that approval.
"This is a much-needed step," says APA Publisher Gary VandenBos, PhD. "There currently is a major crisis in science, across all disciplines, about the reproducibility of findings. … You want the data related to clinical trials to be the most scientifically valid it can be, and open data is one way of providing checks on that."
The Institute of Medicine does not have the authority to enforce its recommendations. But the 23 study sponsors include a wide variety of government agencies, including the National Institutes of Health and the Food and Drug Administration, and industry heavyweights such as AstraZeneca Pharmaceuticals, Bristol-Myers Squibb, GlaxoSmithKline and Johnson & Johnson. The report's authors recommended that those sponsors take the lead to convene "a multi-stakeholder body with global reach and broad representation to address, in an ongoing process, the key challenges associated with sharing clinical trial data."
In addition to enforcement, those challenges include technical hurdles, such as developing large and secure "data warehouses" to house and distribute the data to qualified researchers, according to an editorialopens in new window by Jeffrey Drazen, MD, editor of the New England Journal of Medicine and a member of the IOM report committee.
"Although there are technical hurdles," he writes, "we need to change how we think about data. We need to view it as a community resource, much like a shared park, rather than as personal property."
— Lea Winerman