Article type
Year
Abstract
Background: Little is known about the epidemiology and publication history of randomized controlled trials (RCTs) discontinued due to futility, harm, or insufficient recruitment. We hypothesized that the data already collected are often not made available to the scientific community. The lack of reporting potentially enhances the problem of wasted research efforts and raises ethical concerns, because trial participants consent on the premise of contributing to the advancement of medical knowledge. It may also lead to bias in meta-analyses if relevant data from discontinued trials are not considered.
Objectives: To describe the epidemiology and publication history of discontinued RCTs.
Methods: We included protocols of RCTs approved by the research ethics committee of the University of Freiburg (Germany) between 2000 and 2002. We searched electronic databases and surveyed trialists for information about project/publication status. In case of discontinuation, further information about the reasons was collected.
Results: Of 418 approved RCTs, 194 (46%) were published in full, and 224 (54%) were not. Overall, 300 (72%) RCTs were either completed according to protocol, ongoing, or not started at all; for 54 (13%) we received no survey response, and 64 (15%) were definitely discontinued. Of the latter, 8 (13%) were published and 56 (87%) were not. The reasons for discontinuation were slow recruitment (41%), futility (12%), harm (8%), funding problems (14%), and others (25%) (Table1).
Conclusions: About half of all approved RCTs were not published in full and at least 15% were discontinued. Most of the data collected until discontinuation were not published and thus not easily available for meta-analyses. Slow recruitment was the most frequent reason for discontinuation. Our findings emphasize the necessity for a screening tool to identify trials at risk of being discontinued because of recruitment problems. This could help reduce the waste of scarce resources in clinical research.
Objectives: To describe the epidemiology and publication history of discontinued RCTs.
Methods: We included protocols of RCTs approved by the research ethics committee of the University of Freiburg (Germany) between 2000 and 2002. We searched electronic databases and surveyed trialists for information about project/publication status. In case of discontinuation, further information about the reasons was collected.
Results: Of 418 approved RCTs, 194 (46%) were published in full, and 224 (54%) were not. Overall, 300 (72%) RCTs were either completed according to protocol, ongoing, or not started at all; for 54 (13%) we received no survey response, and 64 (15%) were definitely discontinued. Of the latter, 8 (13%) were published and 56 (87%) were not. The reasons for discontinuation were slow recruitment (41%), futility (12%), harm (8%), funding problems (14%), and others (25%) (Table1).
Conclusions: About half of all approved RCTs were not published in full and at least 15% were discontinued. Most of the data collected until discontinuation were not published and thus not easily available for meta-analyses. Slow recruitment was the most frequent reason for discontinuation. Our findings emphasize the necessity for a screening tool to identify trials at risk of being discontinued because of recruitment problems. This could help reduce the waste of scarce resources in clinical research.