In their thoughtful Policy Forum on research ethics, “Certificates of Confidentiality and compelled disclosure of data” (14 November 2008, p. [1054][1]), L. M. Beskow, L. Dame, and E. J. Costello describe a case where research records, despite having this Certificate, were not fully protected from a court-ordered subpoena. Considering this case and our Institutional Review Board (IRB) experiences, we wonder whether the Certificate, unevenly required by IRBs and time-consuming to obtain, provides any additional, substantive protections to research subjects. Currently, Informed Consent and HIPAA authorization documents explicitly state how research data will be secured, with whom data may be shared, and what monitoring agencies may have access. Furthermore, consent documents often state which information (e.g., child or elder abuse) will be disclosed to authorities. These explicit data-sharing permissions and protections appropriately cover all research records, regardless of their supposed “sensitive” or “nonsensitive” content. Obtaining a Certificate of Confidentiality from NIH places burdens on investigators and lengthens informed consent documents already laden with confusing legal language. If time is critical, investigators may choose to remove sensitive questions or instruments to achieve more rapid IRB approval. Hence, the value of the research may be compromised. With the goals of enhancing subject recruitment and obtaining honest research data on sensitive topics, one must weigh the purported benefits of the Certificate against their potential impedance of established legal and judicial system processes. As imperfect as those processes may be, we believe that the effectiveness of legal arguments and the privacy rights provided to all research subjects by covered entities should govern final judicial decisions, and not a Certificate selectively issued by NIH. [1]: /lookup/doi/10.1126/science.1164100