Research Integrity & Compliance:Data Retention and Destruction – Video and Audio Data

Definition of Data

“Data” refers to recorded information, regardless of form or the media on which it may be recorded.

Definition of Identifiable Data

Data is identifiable if the identity of the participant is known or may be readily ascertained. This includes, for example, indirect identification through the use of a code or other indirect identifiers.

Audio and Video (AV) Data

Audio and video recordings are always considered to be identifiable in the context of human subjects research. Options for de-identifying audio and video recordings include altering an individual's voice or blurring their face. Another option is to transcribe the recording, remove any potential identifiers from the transcript, keep the de-identified transcripts and destroy the original recording and any back-up recordings that have been made. Retention or destruction of video and audio recorded data may be subject to a number of considerations, including but not limited to:

Purpose of Retaining AV Data

What is the purpose and use of retention? For example, is the retention necessary to achieve the specific purpose for which the data were gathered and used? Is retention intended for, and necessary to achieve, secondary uses?

Type of AV Data Collected

Are they sensitive data? How may data conservation affect the participants if security is violated?

Nature of Research Under Review

Does the research field require AV data retention for the continuity of scientific research? What are the professional standards for the relevant discipline?

Informing Participants

Have the participants provided informed consent to the purposes, uses and retention of the AV data collected? Have they been informed of any potential for secondary use of such data?

Access to AV Data

Who is authorized to access the data? How will access be managed?

Data Storage, Security and Protection

Will the AV data obtained be stored securely and protected with all the precautions appropriate to the sensitivity of the data?