Elizabeth Warren Published an Important Editorial in a Top Medical Journal, and Almost No One Noticed

Photo: Win McNamee/Getty Images

In August, Elizabeth Warren — yes, as in Massachusetts senator Elizabeth Warren — penned an editorial published in the New England Journal of Medicine, and nobody noticed. Well, almost nobody. Warren’s article was met mostly with crickets in the mainstream media, but it caused a stir within academia due to the controversial nature of her argument: In her editorial, Warren called for open sharing of clinical-trial data, an important measure ensuring that the public and experts (meaning those outside the officials tasked with reviewing drugs and medical devices for approval) get access to the evidence. Open sharing is essential to sound science, which depends on researchers’ ability to analyze and replicate one another’s work. Without that check, safety and effectiveness data may be exaggerated, and side effects swept under the rug in the rush to market, hurting consumers when drugs are later recalled or retroactively issued warning labels.

Since 2007, the U.S. Food and Drug Administration has required scientists to upload clinical-trial results (except Phase I safety trials) to a database, clinicaltrials.gov, a year after completion of data collection. In practice, however, this rarely happens. One Nature report from 2013 found that only half of all clinical-trial results are published. Pharmaceutical-company-backed studies are even less likely to be reported, at a rate of 17 percent, according to a 2015 NEJM review. The lack of transparency enables researchers and companies to bury negative results, skewing data positive — a widespread problem in scientific publishing that contributes to drug recalls and generally misleads consumers. Although unexpected, Warren’s position is in line with her general stance toward industry, and she wrote that she hoped the proposal would “help to address concerns about conflicts of interest that may arise when clinical trials are funded by industry sponsors that stand to profit from favorable research results.”

Despite echoing the World Health Organization and leading medical ethicists’ recommendations on data-sharing, Warren’s editorial, in which she endorsed a proposal put forth by the International Society of Medical Journal Editors that would make sharing clinical-trial data a mandatory requirement for study publication, was met with sizable pushback. An international consortium of scientists (many of whom reported conflicts of interest and/or funding from pharmaceutical companies) published a competing editorial in NEJM objecting to the proposal, arguing that it would have the opposite of its intended effect, leading researchers to hoard their data and further delay publishing for fear that others not involved in the initial study would piggyback off their results, achieving scientific publishing glory without putting in the work. So-called “research parasites” are a widespread fear among scientists. Even those ultimately supportive of open sharing tend to drag their feet, citing “lack of time or low priority.

But not publishing clinical-trial data in a timely manner can have disastrous consequences, causing major harm to consumers when drugs hit the market only to be recalled after causing unforeseen side effects. One of the most famous and heartbreaking examples of this is antidepressants and adolescent suicide risk: More than 20 years after the drugs were first introduced in the late ‘80s, independent reviews concluded that SSRIs, including Prozac and Paxil, increased teens’ risk for suicide and suicidal ideation. Several tragic deaths, not to mention lawsuits, could potentially have been avoided had the data been made public earlier.

Researchers’ unease over data-stealing is not entirely without merit — Watson and Crick infamously cribbed data from Rosalind Franklin, who was cut out when they were awarded the Nobel Prize for discovering the molecular structure of DNA. Warren notes in her editorial that science has always been a cumulative effort, however, and proposes that resources for data-sharing be built into study grants and budgets from the get-go. Without decisive action, nothing will change, and the safety and efficacy of drugs will continue to be obscured, leaving consumers vulnerable and in the dark.