Ensuring Validity, Reliability, and Reproducibility in Clinical Trials

The exchange of clinical research data can be justified for scientific, economic, and ethical reasons. From a scientific standpoint, the exchange makes it possible to compare or combine data from different studies and to add them more easily for meta-analysis. This allows conclusions to be re-examined and verified or sometimes corrected, and can allow new hypotheses to be tested. Therefore, sharing data can increase the validity of the data, but it also reduces the value of the original investment in research and helps to avoid unnecessary repetition of studies.

The economic advantages of reusing data are one of the reasons why government and intergovernmental bodies, as well as major research funders (e.g. the Gates Foundation and Wellcome Trust), support data sharing. All stakeholders involved in clinical research (e.g. funders, patient groups, researchers, academia, professional groups, industry, publishers, and regulatory and ethical authorities) should support the exchange of study documents and documentary documentation as a normal part of good practice. This section is based on a document commissioned by the Strategy Committee for the Responsible Exchange of Clinical Trials Data on “Concepts and Methods for De-Identifying Clinical Trial Data”, by Khaled El Emam and Bradley Malin (see Appendix B).

The difficulty is that many of the issues surrounding the secondary use of data without explicit consent have not yet been clarified and will need (in Europe) a deeper interpretation by national authorities of the requirements represented by the GDPR, in the specific context of clinical research data. Existing approaches to characterizing repositories (e.g., re3data) should be explored to determine their suitability in the field of clinical research and, perhaps, to adapt or expand them. The generic metadata schema should include a common identification scheme for clinical research data objects. Please note that all research involving human subjects must comply with the ICH GCPs, regardless of the type of clinical study. These principles served as a point of view through which the committee evaluated the benefits and risks of sharing data and considered the roles and responsibilities of the individuals and organizations that participate in and benefit from the clinical trial enterprise.

Trust in clinical research could be further improved if the sharing of clinical trial data were accompanied by public outreach and participation activities to help the public understand that numerous judgments are needed to transform source data into analyzable data (CTSA, 201) and that highly trained researchers can adopt different approaches to answering a research question or analyzing a given data set. Non-commercial clinical research infrastructures should actively participate in the standard-making process to expand them further as needed. Representatives of the communities and groups from which participants in clinical trials are recruited can provide information about cultural and social values and concerns related to the exchange of clinical trial data. Benefits include both the immediate knowledge gained by testing the hypothesis of a particular clinical trial and the broader utility of study data to inform the development of effective and safe clinical care. In fact, it has been argued that clinical trial data should be shared and treated as a public good whoever generates them, that is, regardless of whether they were created through commercial research or funded by public funds. An effective and formal consensus-building process between a large group of experienced researchers and others participating in clinical trials is necessary for ensuring validity, reliability, and reproducibility when working with a Clinical Research Organization (CRO).

Sponsors of clinical trials and researchers who design and carry out clinical trials may believe that fairness includes appropriate recognition and reward for their work and the protection of their legitimate interests.

Leave a Comment

All fileds with * are required