Duke Researchers Claim Drug Industry Leads in Trial Results Reporting

The drug industry does a better job than academic researchers of reporting clinical trial results on clinicaltrials.gov.

That’s the conclusion reached by a group of Duke University researchers that includes incoming FDA Deputy Commissioner Robert Califf. The researchers looked at 13,237 trials that they projected should have reported at least one primary outcome to the government website within one and five years after completion.

At one year after the end of a trial, 17 percent of industry-funded studies had reported results, compared with 8.1 percent of NIH-funded trials. Trials funded by other government or academic sources were even less likely to have reported results — just 5.7 percent.

Earlier-phase trials and trials without FDA oversight were also less likely to be reported. Of the trials that did not report results, 6.1 percent had some kind of legal certification or exemption request.

The gap closes among older trials. Looking at trials that were five years post-completion, 41.5 percent of industry-funded trials had reported results, while 38.9 percent of NIH-funded trials and 27.7 percent of trials paid for by other sources had reported. For trials that reported results, the median time that it took sponsors to do so was 14 months for industry, 23 months for the NIH and 21 months for others.

Later-stage trials were also more likely to report results on time, the researchers concluded, possibly because earlier-phase trials often focus on demonstrating proof-of-concept and tend to be considered valuable, closely guarded intellectual property.

In attempting to explain the industry/academia discrepancy in reporting times, the authors note that preparing results summaries can take up to 60 hours of staff time, which may not be paid for by government research grants.

Overall, about 80 percent of industry-sponsored studies either reported results or had a legal reason for delay. This number decreases to 50 percent for NIH-sponsored studies and 45 percent for other studies.

The research was published in Thursday’s New England Journal of Medicine. Read the article at www.fdanews.com/03-12-15-resultsreporting.pdf. — Lena Freund