One of the golden rules of Trial Master File (TMF) management is to avoid duplicate filing of documents. We all know that the primary objectives of the TMF is to provide step-by-step documentation of the clinical trial and to protect the rights of patients. It is also vitally important for the TMF to facilitate, and not hinder, the audit process during inspections. Clinical research professionals who are responsible for creating TMFs should be mindful of this always. Imagine the frustration of an inspector who must contend with repeated instances of identical documents and is then required to spend time and energy determining whether there are minute differences between these documents. However, just how do we define and treat duplicate documents? This may seem incredibly obvious, but … what if:
A record possesses corresponding classification and metadata? This scenario could indicate that the document may be a duplicate — even if it has the same date. For example, it is possible there will be several IRB approvals on the same date; however, the approvals may be linked to unique submissions and should be filed accordingly.
Multiple documents have identical time stamps or hash codes? This is a near certain indication that a document is a duplicate. However, the inverse is not necessarily the case — documents with distinct codes may, in fact, be duplicates (this occurs if the same document is scanned more than once).
A specific document is not applicable to all sites? Should it be filed once, or several times, in the eTMF? Preferably, the document should be filed in one location in the eTMF and linked to the appropriate locations. Program-level documents such as Investigator Brochures should be treated in the same fashion — filed in a singular location and linked to the relevant studies; for example, a new version of an Investigator Brochure would not be filed with studies that have been terminated.
The ideal strategy is to identify duplicate documents during the initial phase of creating an eTMF — upon the uploading of documents to the eTMF system. Preferably, the eTMF should be configured to notify the submitter of potential duplicates and provide the opportunity to determine whether to proceed with or cancel the upload. If the eTMF system does not permit this functionality, the submitter may be required to manually perform a quality check for identical documents prior to upload, which is considerably more time consuming. The least effective approach is to visually scan documents within an existing eTMF, a practice which is considerably more prone to error and cannot be performed completely and accurately until the study is closed and all documents are retrieved and uploaded to the eTMF.
As with all things eTMF, steering clear of duplicate documents requires driving best practices utilizing a combination of technology, development of comprehensive policies and guidance, and hands-on training for TMF clinical research professionals.