This article is more than 5 years old. Information may no longer be current.
Randomized controlled trial discontinuation common
Click Here to Manage Email Alerts
A recent retrospective cohort study found that a large number of randomized controlled trials are terminated before completion — most frequently due to poor recruitment — and the results of these trials are often unpublished.
Researchers assessed 1,017 randomized controlled trials approved between 2000 and 2003 by six research ethics committees in Canada, Germany and Switzerland. In total, 894 trials involved patients and 123 involved healthy volunteers. The researchers evaluated factors such as whether the trial was completed, reasons provided for discontinuation if it occurred and publication status for each trial.
During a median follow-up of 11.6 years through April 27, 2013, 253 (24.9%) of the included trials were discontinued.
The most common reason provided for discontinuation was insufficient recruitment (9.9% of all evaluated trials), followed by administrative issues (3.8% of trials), futility (3.6%) and harm (2.4%). Study discontinuation was reported to ethics committees in 37.9% of cases, including 23.8% of those discontinued due to poor recruitment, 38.5% discontinued due to administrative issues, 43.2% discontinued due to futility and 54.2% discontinued due to harm.
Multivariate analysis indicated inverse associations between discontinuation due to poor recruitment and industry vs. investigator sponsorship (8.4% of trials vs. 26.5%; OR=0.25; 95% CI, 0.15-0.43) and larger sample size (–0.7% per 100-person increment; OR=0.96; 95% CI, 0.92-1). Sensitivity analyses did not significantly alter results, and discontinuation rates did not differ according to country of origin (P=.63).
Upon completion of follow-up, 55.8% of all evaluated studies were published as journal articles, including 39.6% of those discontinued due to poor recruitment. Trial discontinuation was associated with less chance for publication compared with trial completion, according to multivariate analysis (OR=3.19; 95% CI, 2.29-4.43).
“For investigator-sponsored [randomized controlled trials], stakeholders including trial investigators, funding agencies and [research ethics committees] need to develop strategies to prevent trial discontinuation due to poor recruitment,” the researchers wrote. “… Further research is necessary to determine the optimal length of pilot studies and to develop reliable prediction models for recruitment performance.”
Further, additional efforts are necessary to ensure that discontinued trials are reported to ethics committees, and that results from discontinued trials are published, the researchers said. “The nonpublication of results from discontinued — or from completed — [randomized controlled trials] represents a waste of valid data that could contribute to systematic reviews and meta-analyses.”
Disclosure: See the full study for a list of relevant financial disclosures.
Click Here to Manage Email Alerts