In response to recent questioning of the integrity of the dissemination of results in biomedical literature, three medical researchers from Stanford and Duke University are pointing to the need for increased access to data from clinical research.
In a viewpoint article published online today in JAMA Internal Medicine, the authors, Robert Califf, MD, and Jonathan McCall of Duke, and Robert Harrington, MD, of Stanford, write that it’s time for both industry and academia to "catch up to other areas of society:"
The liberation of information once held in secret has toppled regimes and transformed societal expectations regarding progress and possibilities. Access to data from clinical research should be truly democratized.
The goal of clinical research should be to add to the body of evidence that can guide decisions about personal health and health policies, the authors write - but things like selective omission of important findings, inaccuracies in published studies, and the use of unreliable data systems are all hindering this. Harrington and his colleagues outline the critical issues that need to be addressed - issues that concern "(1) the value of the research question, (2) the quality of the execution of the research, and (3) the complete and balanced presentation of all relevant data in the publication" - and sound a hopeful note:
The good news is that powerful tools exist to address and potentially surmount these issues. These tools are the ClinicalTrials.gov registry and the ongoing movement toward data transparency.
ClinicalTrials.gov was originally created to provide researchers, physicians, and the public with ready access to information on clinical trials. More recently, legal requirements to register studies have expanded to encompass the reporting of results, including adverse events, within 1 year of ascertainment of the last primary end point. These requirements are designed to ensure that findings from almost all trials relevant to US medical practice that involve a drug or device are available in a single, accessible public registry. If the requirements of ClinicalTrials.gov and other international registries are maintained and strengthened in areas where they are currently deficient, the benefits should be substantial...
Previously: Research shows small studies may overestimate the effects of many medical interventions, Outing bias in scientific research, A critical look at the difficulty of publishing “negative” results, Examining the impact of unpublished research on medicine and Testing medical ‘truths’