When Do I Close a Study?September 2014 IssueBy John F. Ennever, MD, PhD, CIP, Director, BUMC IRB The short--but accurate--answer is, “When what you are doing no longer constitutes human subjects research.” The motivation for this article came out of a recent meeting of one of our IRB Panels. Several Board members, who themselves are active researchers, were unaware that once you have completed the data collection phase of a study and are only doing data analysis, that the study can actually be closed and the data analysis can continue without ongoing IRB approval if the data set are stripped of identifiers. It became apparent that more emphasis should be given to this important OHRP policy, since it is mutually beneficial to the Principal Investigator and the IRB to close studies which no longer require IRB oversight. In the previous edition of the CR TIMES, I wrote about what is and what is not human subjects research. In this month’s article, I will cover how a similar analysis can be used to let you know when it is proper to close a study, Once a study has been approved, oversight by the IRB continues until the investigator provides information that the study qualifies for closure. This oversight consists of continuing review and re-approval of a research project at least annually so long as the project continues to involve human subjects. OHRP considers a research project to continue to involve human subjects as long as the investigators conducting the research continue to obtain:
At first reading, you might think that once you have finished interacting with subjects and are no longer collecting identifiable private information, then at that point you are no longer doing human subjects research and, therefore, can close the study. This would be incorrect. This is because the OHRP also considers using, studying or analyzing identifiable private information or identifiable biological specimens to constitute human subject research. However, once the information or specimens are no longer identifiable, then using, studying or analyzing the information or specimens is not human subject research. There will be situations in which a data set may meet the definition of HIPAA de-identification, but the identity of the subjects would still be known to the investigators. In such a case, analysis of the data would constitute human subjects research, and continuing review by the IRB is required. For example, in a study involving interviews with a small number of individuals, the investigator who conducted the interviews can readily recall which subject provided which answers, even if the data set do not contain any of the HIPAA identifiers, because of the small number of subjects. The analysis being done by the investigator is still considered human subject research because the participants are identifiable, so continuing review by the IRB is required. For the rest of this article, let’s use the example of a research study where:
The only activity that remains is analysis of the data, which IS human subjects research if the data are identifiable; and is NOT human subjects research if the data are not identifiable. At this point, it should be determined whether the study is no longer human subjects research and is eligible for closure. Two obvious situations where data are no longer identifiable are (1) if all raw data have been erased; and (2) if data are coded (a study ID number replaces all direct identifiers), and the linking code has been destroyed. There will also be situations in which the data set contains one or more of the HIPAA identifiers, but the identity of the subjects is not readily ascertainable by the investigators. For example, a large data set from a clinical trial could still have dates of visits (a HIPAA identifier), but direct identifiers have been replaced by codes and the linking code has been destroyed. The use and disclosure of the data continue to be permitted by the HIPAA authorization signed by the subject (with no expiration date). Since the continuing analysis does not involve identifiable information and, thus, does not meet the definition of human subjects research, then the study does not require continuing oversight by the IRB and can be closed.
You can also close your study if you are not doing analysis but continue to maintain the data. Simply maintaining individually identifiable private information without using, studying, or analyzing such information is not human subjects research and, thus, does not require continuing oversight. Note that the maintenance/storage of identifiable private information must be consistent with the IRB-approved research plan, and the Investigator must continue to honor any confidentiality protections of the data. To close a study, submit a Final Report form in INSPIR prior to the expiration date, rather than using the Continuing Review form. The Final Report form asks questions related to the study closure which are not found in the Continuing Review form. The closure letter will indicate that the effective date of the closure is the date when the Final Report was submitted in INSPIR. Final Reports are processed on a routine basis; however, you can send an email to medirb@bu.edu if you have a time-sensitive need for the closure letter (for example, for the Sponsor or for funding purposes). If you have closed a study where you are maintaining but not analyzing the data, and at a later date, you wish to go back to do further analysis of the previously-collected data to answer a different study question, you would then have to submit a new protocol describing your plan. Alternatively, you can submit an amendment to your closed study to re-open it for further analysis.This Quiz applies to the current recertification period from July 1, 2013 to June 30, 2015. We recommend that you take this quiz now so you can stay up-to-date. Click here to login to My Account if you are a BUMC researcher and would like to take the quiz now.
|