Drug Evaluation Committee Study on Methodology of Data De-identification for CTDS (Clinical Trial Data Sharing)
February 2021
Regarding the sharing of Individual Patient Data (IPD) from clinical trials (CTDS), the Data Science Subcommittee has previously released a report introducing the overview of CTDS and de-identification techniques, a newsletter summarizing the latest developments, and a report on relevant regulations and key considerations for compliance.
Although CTDS originated in Europe and the United States, an increasing number of companies in Japan are now adopting CTDS, leading to progress in data utilization.On the other hand, it is difficult to visualize specifically how the processing (de-identification) intended to reduce the risk of identifying individuals linked to the data is carried out and to what extent it affects the data’s usefulness. This poses an obstacle for companies in establishing de-identification processes and can also hinder researchers using de-identified data from understanding the limitations of that data.
In light of this situation, this report aims to provide a concrete understanding of how de-identification is performed and to what extent it affects the usefulness of the data. To this end, it introduces not only the de-identification process itself but also the results of de-identification using actual data.While this document is limited to the scope of CTDS, we believe that as data utilization advances beyond clinical trial data, the underlying principles of data de-identification are not limited to clinical trial data. We hope this report will be of assistance to those involved in CTDS and clinical trial operations as they advance CTDS activities.
Study of Data De-identification Methods Related to CTDS (Clinical Trial Data Sharing) (663KB)