Good Clinical Practice Guide
Results 1 to 2 of 2

Thread: "Dictated" source data

  1. #1
    Forum Member
    Join Date
    Nov 2011

    "Dictated" source data

    We have recently had a site where the investigators routinely "dictated" their assessments/results of trial patient's visit via a dictaphone during the visit rather than making notes themselves, and the study nurses were afterwards typing out what the PI had dictated to file this typed page into the source data of the patient (signed and dated by the investigator obviously) as "the source".
    Following a recent audit, we had a finding defined because the original tapes from the dictaphone had not been kept and as such - according to the auditor - the original source data had been destroyed.
    Anyone have experience with similar cases? Would the recommendation be to treat the tapes as other source "documents" in terms of review/storage or would the typed document, signed & dated by the investigator, be considered as sufficient "original" source by the regulators?
    Any feedback possible from regualtor point of view in this particular case?

  2. #2
    Forum Member
    Join Date
    Nov 2011

    Dictated Source

    Original source documents/data are the cornerstone of GCP and therefore any destruction of original source documents should only be contemplated under conditions which will assure the veracity of the certified copies and the system to produce those copies. To create verified copies of source information there must be a process of verification and certification. A Dictaphone record may be somewhat different than other sources in that the typed version may not be an exact copy of the dictation, but rather the typed version must represent accurately the facts reported in the dictation. As a minimum, the individual generating the original dictation must satisfy themselves that the transcription is an accurate and complete representation of the dictation and then authenticate it via signature and date, or in an appropriate manner, in order to meet the burden of attributability and accuracy.
    It is recommended that when research source data is generated, via dictation, that the process involved be evaluated. Things that should be considered when evaluating the quality and its impact on the research, may be; the amount to which the dictation contributes to the total amount of source at the site; the importance of the data in the dictation; what Quality Control might be in place at the site; what processes or written procedures are available which describe the site's process; and the overall understanding of the meaning of the authentification done by the staff. These processes should be evaluated by the monitor before placing the study at the site. When an investigator uses an outside vendor to perform the transcription, the provisions of the contract with the vendor may detail the quality control steps taken by the vendor. Judgement will be necessary when evaluating the conditions present at the site.
    A high quality process or "best practice" would include a method to monitor the quality of the transcriptions. A person independent of the typist (this could be the originator of the record) would compare dictations, or a sample of dictations, back to transcribed records, assessing their accuracy. The level of accuracy, criticality of errors found, and most importantly a method to feed back results for continuous improvements, are important elements of a quality process.
    Also, it is expected to maintain a document to explain the reason of the destruction of the source data and the steps taken to ensure the transcript is fit for purpose.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts