Most researchers fail to report clinical trial results as required by law

Anyone looking for the outcomes of thousands of closed or completed clinical trials isn't likely to find them, despite government mandates requiring the data be made available.

More than 80% of trials that should have posted results had not done so within one year of completion, and government funded research was the least likely to share findings, according research looking at clinical trial data between 2008 and 2012.

The inability to access trial data has been a bane of researchers, especially those trying to compare the effectiveness and optimal use of newer, more expensive therapies.

In an effort to address such concerns earlier this year, the Institute of Medicine proposed vast changes to how data is shared.

But critics remain dubious that the IOM's efforts will translate into actual practice. Movement toward transparency has remained markedly slow. In the meantime, the existing reporting mandates remain essentially unenforced, and reporting levels are low.

“It's just ethically pathetic,” said Art Caplan, a bioethicist and founding director at the in NYU Langone Medical Center's Department of Population Health. Researchers studying human subjects have a duty to share data as a way to let the scientific community know what's going on, he said. “If you don't do that, you're breaching your covenant.”

In the new study, researchers from the Duke Clinical Research Institute identified 13,327 trials that were “highly likely” to be subject to FDA mandates that require the researchers to report the results.

Overall, only 1,790 studies, about 13%, reported summary results within 12 months after the trial's completion. Only 38% reported results at any time up to Sept. 27, 2013, according to the findings posted Wednesday in the New England Journal of Medicine.

Dr. Monique Anderson, the lead researcher on the study, said the results were surprising given that the Food and Drug Administration Amendments Act of 2007 requires that outcomes of applicable clinical trials be reported within specific time frames on clinicaltrials.gov.

“I think of it the way I think of my taxes,” said Anderson, a cardiologist and assistant professor of medicine at Duke. “I may not like doing it, I may not know how to do it… but it's the law, so I do it.”

Whether the results were reported also varied by who funded the research. Though all groups had relatively low reporting rates, industry-sponsored studies reported their results more frequently than those from the National Institutes of Health and academic institutions.

Of the 8,736 industry-funded clinical trials, 17% had shared results within 12 months as required by the federal law. Only 8.1% of the 1,899 NIH-funded trials and only 5.7% of trials funded by other government and academic institutions had done so.

Anderson said rate of reporting might be dampened by confusion over which institutions are subject to the rule, the labor required to input all of the data into the website and a lack of enforcement by the FDA.

An FDA spokeswoman countered that not all clinical trial results are due within a year of completion and said the agency follows the enforcement provisions of the law when researchers fail to report results as required. .

"Depending on the circumstances, results submission for an applicable clinical trial of an approved, licensed, or cleared product, potentially may not be due for up to three years after the completion date of the applicable clinical trial," the FDA spokeswoman said.

According to the text of the FDAAA legislation, noncompliant institutions are “subject to a civil monetary penalty of not more than $10,000 for each day of the violation after such period until the violation is corrected.”

To date, however, the agency said there have been no formal agency determinations regarding noncompliance of any specific, applicable clinical trial, which it says is a prerequisite to seeking civil money penalties.

The FDA also noted in a letter sent last year to Rep. Leonard Lance (R-N.J.) that the agency had achieved voluntary compliance when it was brought to the attention of the responsible parties that data was missing.

Notably, many industry groups have already been voluntarily participating in clinical trial data-sharing efforts through programs such as the Yale open-data project; clinicalstudydatarequest.com, a project of GlaxoSmithKline; and alltrials.net, an initiative based in Europe.

The IOM is urging more of those types of efforts, as well as timetables for the sharing of adverse-event summaries and analyzable data sets. The FDA said that it is reviewing those recommendations.

Follow Sabriya Rice on Twitter: @sabriyarice