Every day, the numbers tick upward. It began with $70,000 last Monday. That Wednesday, it was $170,000. Today, it's $624,726. That's how much the United States government could levy in fines against scientists who run clinical trials and recently failed to comply with a law that says they must make their results public, according to an independent team of researchers based in the United Kingdom. But U.S. government agencies have never penalized anyone for breaking that rule, which was intended to make it easier for doctors and regulators to spot when drugs don't work or have dangerous side effects. The U.S. government's inaction is why the U.K. team created their own website intended to shame the universities, pharmaceutical companies, and government institutes that ignore the law.
"Public accountability is a vitally important tool in public policy, and in improving standards. Public accountability is the reason why many states and countries require restaurants to publish their 'hygiene ratings' on the front door," Ben Goldacre, a physician and health-data researcher at the University of Oxford and the website's lead creator, writes in an email. In addition to embarrassing institutions who are late in posting their results to ClinicalTrials.gov as required, Goldacre hopes charities and other groups that fund trials will look at the site—called TrialTracker—and put pressure on the teams they support. Lastly, an open-data advocacy group that Goldacre founded, AllTrials, will be sending the U.S. Food and Drug Administration a weekly list of late reporters. "We also hope our public tool will help encourage the FDA to enforce the law," he writes.
"What's nice about this tracker is that it allows responsible individuals within an institution to monitor what's happening," says Joe Ross, a professor of medicine at Yale University who was not involved in making the site. A vocal supporter of data sharing, Ross says that, when he has visited universities to give talks, sometimes officials don't even know how many clinical trials their faculty are running at the moment. Imperial College London apparently learned about one overdue trial through a BBC reporter writing a story about Goldacre's work.
Making the results of clinical trials public as soon as possible is vital to ensuring that America's drug supply is safe and effective, proponents say. In fact, the data-reporting law grew out of a specific safety controversy: In 2004, New York State's attorney general filed a lawsuit against the drug company GlaxoSmithKline, alleging that the company had hidden data showing that an antidepressant it made, Paxil, increased the risk of suicidal thoughts and even attempts in children and teens. GlaxoSmithKline eventually settled out of court, but the case led Congress to pass data-reporting rules in the hope of bringing safety problems like Paxil's to light sooner. (STAT News traces this history in a 2015 investigation.)
In fact, certain clinical trials—that is, studies of experimental drugs and devices conducted on human volunteers—have technically had to report their results since 2008. The country's first clinical-trial data-reporting rules were written into the FDA's Amendments Act of 2007. And experts have long known that researchers flout the law without consequence. In 2015, STAT found that most universities, companies, and even government scientists fail to post results on ClinicalTrials.gov for more than half of their trials. Between 2008 and STAT's 2015 story, the feds could have collected $25 billion in fines, STAT reporter Charles Piller calculated. To date, they have collected $0.
Goldacre's TrialTracker fingers more recent offenders than STAT did, and also offers clinical-trial leaders a more generous benefit of the doubt. The way the FDA's Amendments Act was written left many questions about who exactly the law applied to, Goldacre says. So the government wrote a Final Rule in 2016, with clearer details, to apply to clinical trials that wrapped up after the rule went into effect on January 18th, 2017.* Goldacre and his team used the Final Rule to write an algorithm that combs trials that are registered to ClinicalTrials.gov for ones that should fall under the law.
"Our methods are rigorous, and conservative, and very clearly documented in our paper," Goldacre writes. That paper has been posted online, but hasn't yet been peer reviewed.
So, is TrialsTracker an accurate accounting of who's on the hook for reporting their results? Pacific Standard asked the agencies that would know best. The National Institutes of Health—which runs ClinicalTrials.gov—offered a statement that said its spokespeople are "unable to comment on the accuracy of the data being made available by AllTrials," while Lauren Smith Dyer, a spokeswoman for the FDA, writes in an email: "The FDA has not yet had opportunity to fully review the tracker being provided by AllTrials; however, it is often not possible to determine which parties may be noncompliant based solely on the information in the record that is publicly posted on ClinicalTrials.gov." Goldacre is aware that ClinicalTrials.gov doesn't publicly post some information that would allow outsiders to know for sure whether a clinical trial falls under the reporting law, so he and his team devised a workaround for TrialsTracker.**
As for whether the FDA plans to enforce its 2007 Amendments Act and Final Rule, Smith Dyer didn't answer directly. She confirmed that the government has never fined anyone under the data-reporting law. She noted they must alert institutions first and give them a chance to post their late data. And she wrote that, in April of 2017, the agency issued guidance to inspectors, saying they need to "collect certain information related to compliance with ClinicalTrials.gov."
*UPDATE—February 26, 2018: A previous version of this story attributed the Final Rule for clinical-trial reporting to the incorrect government agency. For the curious, the Department of Health and Human Services was responsible for the Final Rule.
**UPDATE—February 26, 2018: A previous version of this story suggested that the FDA posts information to ClinicalTrials.gov. It does not; the National Institutes of Health does.