Abstract: De-identification, where personal patient identifiers are hidden so that individuals cannot be easily re-identified, is a crucial method of supporting multicenter research studies, allowing large volumes of data can be shared among institutions while minimizing the risk to patient privacy. While the U.S. Department of Health and Human Services has published general guidelines for de-identification, the literature does not report consistent best practice techniques among all modalities. This study reports the results of a multicenter survey on de-identification practices at different institutions. There was significant variability in de-identification processes within institutions, including approaches to structured patient data, medical text, and imaging. Variation also existed in constructing limited data sets and managing re-identification attempts. These findings suggest that there is a lack of standardization of de-identification practices among different institutions. The community may benefit from national best practice guidelines.

Describe the new knowledge and additional skills the participant will gain after attending your presentation.: Attendees will be able to better assess their institution's de-identification practices. Attendees will also recognize the need for national best practice guidelines to help standardize de-identification practices among institutions.

Authors:

Christopher Tenore (Presenter)
Icahn School of Medicine at Mount Sinai

Patricia Kovatch, Icahn School of Medicine at Mount Sinai
Joseph Finkelstein, Icahn School of Medicine at Mount Sinai