Many Americans are worried about making sure their health data remains absolutely private.
But a newly published paper suggests major benefits could result when pharmaceutical companies share data about patients who undergo clinical trials.
This information, , which can’t be linked with individual patients, might lead to development of new, unforeseen therapies, that were not initially envisioned, thereby making the safety risks these patients took more valuable in the long run.
Rutgers Biomedical and Health Sciences Chancellor Dr. Brian Strom coauthored, which found several benefits to making patient data from clinical trials available to the public.
Strom, who leads the institution within Rutgers that includes much of the former University of Medicine and Dentistry of New Jersey, serves on the independent review panel that has been determining which research projects can have access to the patient data.
Traditionally, pharmaceutical companies have kept patient data from clinical trials private. But starting in May 2013, GlaxoSmithKline agreed to share data from roughly 200 clinical trials dating to 2007. Since then, GSK’s list of available studies has grown to 950, and 10 more companies have joined in the data-sharing system.
“I think this is the way companies should be going,” said Strom. “This is an exciting first-mover approach on the part of GSK.”
The data, which is “de-identified” so that individual patients cannot be determined, is available through.
In the paper published by the New England Journal of Medicine last week, Strom and his three coauthors found that there was broad interest in analyzing the data from outside researchers, who submitted study proposals and agreed to use the data only for the purpose described in a proposal approved by the panel.
In the first year, 36 proposals were approved, 23 of which led to data-sharing agreements. These agreements covered an average of two studies each. The proposals included studies of risk factors for diseases and different treatments for epilepsy.
There’s early evidence that reanalyzing previously published studies “can lead to conclusions about the types of patients who should receive a given treatment that different from the original investigators,” the authors wrote.
“It behooves us as a society to make maximum use of the data,” Strom said, adding that he agreed to join the panel in part due to his philosophical belief that the project could benefit medical science.
The panel accepts or rejects proposals based on their scientific rationale and relevance to medical science or patient care. It also considered the investigators’ experience, whether they had conflicts of interest. and whether they summarized their proposal using clear language
While the Food and Drug Administration analyzes clinical trial data to make its regulatory decisions, the data itself has never been available to the public until now, Strom said.
“It should help industry from a credibility point of view,” Strom said, adding that it would ultimately help public health by taking full advantage of clinical-trial data.
For example, investigators will be able to compare how different drugs affected different types of patients, potentially allowing pharmaceutical companies to determine whether patients who share certain characteristics would benefit from a specific drug, Strom said.
“It gets much more useful as additional companies add data,” he said.
While serving as the leader of a major research institution has caused Strom to cut back some activities -- such as giving up his clinical practice -- he plans to continue to participate in research projects such as taking part in the panel.
“It’s been a big part of my identity my whole life,” Strom said of research.
His interest in the latest research illustrated by where he was when he was interviewed by phone yesterday: Strom was in Asia, preparing for a meeting of the International Society of Pharmacoepidemiology to be held later this week in Taiwan. The focus of that gathering will be on studying the uses and effects of drugs on the health of defined populations.