Background: This article outlines procedures for the feedback of individual research data to participants. This feedback framework was developed in the context of a personalized medicine research project in Canada. Researchers in this domain have an ethical obligation to return individual research results and/or material incidental findings that are clinically significant, valid and actionable to participants. Communication of individual research data must proceed in an ethical and efficient manner. Feedback involves three procedural steps: assessing the health relevance of a finding, re-identifying the affected participant, and communicating the finding. Re-identification requires researchers to break the code in place to protect participant identities. Coding systems replace personal identifiers with a numerical code. Double coding systems provide added privacy protection by separating research data from personal identifying data with a third "linkage" database. A trusted and independent intermediary, the "keyholder", controls access to this linkage database.
Discussion: Procedural guidelines for the return of individual research results and incidental findings are lacking. This article outlines a procedural framework for the three steps of feedback: assessment, re-identification, and communication. This framework clarifies the roles of the researcher, Research Ethics Board, and keyholder in the process. The framework also addresses challenges posed by coding systems. Breaking the code involves privacy risks and should only be carried out in clearly defined circumstances. Where a double coding system is used, the keyholder plays an important role in balancing the benefits of individual feedback with the privacy risks of re-identification. Feedback policies should explicitly outline procedures for the assessment of findings, and the re-identification and contact of participants. The responsibilities of researchers, the Research Ethics Board, and the keyholder must be clearly defined. We provide general guidelines for keyholders involved in feedback. We also recommend that Research Ethics Boards should not be directly involved in the assessment of individual findings. Hospitals should instead establish formal, interdisciplinary clinical advisory committees to help researchers determine whether or not an uncertain finding should be returned.