(Reuters) - U.S. health officials on Wednesday proposed significantly expanding what researchers are required to report about clinical trials of drugs, devices, and other interventions, addressing concerns that data crucial to patients and physicians is kept secret.
The rules, which are open for public comment until Feb. 19, apply to the database ClinicalTrials.gov, operated by the National Institutes of Health (NIH).
Researchers are documenting ever-more cases in which unfavorable study results are withheld from the public.
"The proposals take important steps toward greater transparency, but there are also disappointments," said Kay Dickersin, director of the Center for Clinical Trials at the Johns Hopkins Bloomberg School of Public Health.
One change would require that human studies funded even partly by the NIH, including surgical procedures and behavioral therapies such as for obesity, not only be registered at ClinicalTrials.gov but also report summary results. Previously, there was no such requirement.
Absent NIH funding, clinical trials conducted on experimental or approved drugs regulated by the U.S. Food and Drug Administration would not have to report safety data.
But researchers would be required to report other results, a significant expansion of the current requirement, which allows data to remain secret if, for instance, they find a drug or other therapy to be ineffective.
"This proposed rule would close an important gap" by making additional information about clinical trials publicly available, FDA commissioner Dr. Margaret Hamburg said in a statement.
In a 2013 study, scientists found that of hundreds of trials they examined half never published results. An analysis of scores of studies of the influenza drug Tamiflu reported in April that published studies painted the drug as safer and more effective than confidential clinical-trial reports submitted to regulators.
Johns Hopkins' Dickersin, who has criticized the lack of transparency in clinical trials, praised the requirement that studies of additional uses of approved drugs, such as whether a neuralgia drug also works against migraines, report results. Currently, doctors are permitted to prescribe such "off-label" use of approved drugs, never knowing whether a study showed they do not work for the additional disorder.
One disappointment, Dickersin said, is that the proposals do not require researchers to reveal details of study protocols, such as what endpoints - heart attacks, for example - it aims for and how. That might let a manufacturer conceal unfavorable results.
Of the 178,000 studies registered on ClinicalTrials.gov, summary results have been posted for 15,000.
(Reporting by Sharon Begley; Editing by Jonathan Oatis)