Image: PD
1. Poor participant recruitment was found to be the most common cause of randomized controlled trial (RCT) discontinuation.
2. Factors associated with not publishing results of RCTs include small target sample size, single center design, industry sponsorship, RCT discontinuation, and studies of healthy volunteers as opposed to patients.
Evidence Rating Level: 2 (Good)
Study Rundown: Despite the fact that randomized controlled trials (RCTs) require a great deal of resources, a proportion of them are never completed. There are several potential reasons for this including inability to recruit an adequate number of participants, clear benefit from the intervention, and clear harm from the intervention. This study retrospectively reviewed RCT protocols from six institutions in Germany, Switzerland, and Canada approved between 2000-2003 to determine the reasons for the discontinuations that occurred. Of the 1017 included RCT protocols, 253 trials were discontinued (24.9%, 95%). The most common cause of discontinuation was poor recruitment (101 of 1017 trials). When the study period ended, 567 (55.8%) of the RCTs were published in peer-reviewed journals. Overall, uncompleted RCTs were less likely to be published than completed RCTs (55.1% vs. 33.6%, p<0.001). A strength of this study is the thorough analysis done by experts of each RCT protocol. A weakness of the study is the fact that the trials come from a relatively small number of sites.
Click to read the study in JAMA
Relevant Reading: Learning from failure–rationale and design for a study about discontinuation of randomized trials (DISCO study)
In-Depth [retrospective cohort]: This study retrospectively reviewed 1017 RCT protocols approved between 2000 and 2003 by six research ethics committees in Germany, Switzerland, and Canada. In the cohort, 253 trials were discontinued (24.9%, 95% CI 22.3%-27.6%), most commonly because of inadequate participant recruitment (101/1017, 9.9%, 95% CI 8.2%-12.0%). The factors associated with discontinuation due to poor recruitment were small planned sample size (p=0.04) and investigator as opposed to industry sponsorship (p<0.001). Administrative reasons for discontinuation were identified in 3.8% of trials. Discontinuation due to early intervention benefit occurred in 0.9% of trials and discontinuation due to harm of the intervention occurred in 2.4% of trials. In all, 567 (55.8%, 95% CI 52.6%-58.8%) of the RCTs were published as full articles in journals. The factors associated with non-publication were small sample size target (p<0.001), single center RCTs (p=0.001), investigator sponsorship (p=0.002), discontinued RCTs (p<0.001), and RCTs with patients as opposed to healthy volunteers (p<0.001).
More from this author: Risk-reduction counseling with HIV testing may not decrease rates of sexually transmitted infections, Autoantibodies may precede symptom onset in Sjögren’s Syndrome, Varenicline and buproprion more effective than varenicline alone for tobacco abstinence, Peer based health education program linked to reduced waist circumference, No overall change in the prevalence of obesity in the United States between 2003 and 2012
©2012-2014 2minutemedicine.com. All rights reserved. No works may be reproduced without expressed written consent from 2minutemedicine.com. Disclaimer: We present factual information directly from peer reviewed medical journals. No post should be construed as medical advice and is not intended as such by the authors, editors, staff or by 2minutemedicine.com. PLEASE SEE A HEALTHCARE PROVIDER IN YOUR AREA IF YOU SEEK MEDICAL ADVICE OF ANY SORT.
