A new analysis of 585 large clinical trials registered with the government website ClinicalTrials.gov finds that more than a quarter have never been published in scientific journals, while slightly more than three-quarters did not post results online.
Dr. Christopher W. Jones, an attending physician at Cooper Medical School of Rowan University and the study’s lead author, said that about 250,000 clinical trial participants “were exposed to the risks of trial participation without the societal benefits which accompany the dissemination of trial results,” MedicalDaily.com reports. The study was published online by the British Medical Journal (BMJ).
Not publishing clinical trial results has generated legal and ethical controversies. Critics assert that unfavorable results of some industry-funded clinical trials do not get published, wrongfully protecting bad products or dangerous drugs. Non-publication may also violate a law requiring trials with human participants to not only be registered but also have results posted on the clinical trial website, according to MedicalDaily.com.
Dr. Jones and co-author Dr. Timothy Platts-Mills investigated 585 trials with at least 500 participants registered on ClinicalTrials.gov and completed before January 2009. Results of 171 trials (29 percent) had not been published by November 2012, the end date for the literature search. Seventy-eight percent of the unpublished trials (133 trials) had no results available on the NIH website. Non-publication was more common among industry-funded trials (32 percent) compared to an 18 percent non-publication rate for government, university, or grant-funded trials.
Dr. Platts-Mills, an assistant professor of emergency medicine at the University of North Carolina, said clinical trials provide “an essential source of information for how to care for patients,” according to MedicalDaily.com. New policies are needed to ensure that trial results “are made publicly available in a timely manner.”