14 Dec 1 in 5 Surgical Trials Discontinued Before Completion
Medical Research: What is the background for this study? What are the main findings?
Dr. Bhangu: Randomised clinical trials are widely regarded as the type of evidence in medical research most likely to change practice and improve patient care. However, they are challenging to perform, expensive to deliver and rely on patients’ willingness to participate for the benefit of their wider community. Results of these studies should be disseminated widely in order to promote new medical knowledge. Unfortunately, some clinical trials are terminated early or fail to reach publication after completion. This leads to lost data and wastage of finite resources. Clinical trials within surgical disciplines present unique challenges, which may further impact on dissemination of evidence. We aimed to investigate the fate of surgical trials.
Disappointingly, we found that 1 in 5 surgical trials are terminated early before completion, most commonly due to poor recruitment of research participants. Of trials which do reach completion, 1 in 3 are not published, indicating a significant waste of resources. A systematic approach to contact trial investigators during the study proved largely unsuccessful, implicating further hidden barriers to identifying trial data.
Medical Research: What should clinicians and patients take away from your report?
Dr. Bhangu: A clear understanding of the problems faced is essential for future improvement to take place. All stakeholders, including clinicians, journal editors, funders and patient-ethics liaison groups must work together to promote efficient use of research resources.
Medical Research: What recommendations do you have for future research as a result of this study?
Dr. Bhangu: Firstly, resources should be targeted towards teams who have the resources and infrastructure to deliver larger multicentre trials. Within this framework, new investigators should be nurtured.
Secondly, researchers should be able to work with industry funding, whilst maintaining full academic independence. Data storage should again be independently held, removing a potential barrier to publication
Finally, a permanent contact should be made publicly available from the point of trial registration, available to clinicians, researchers and patients.