information about them is accessed. The elements of this review and the types of issues to be discussed during this review were detailed previously.

  • Amendment review: This occurs any time the researcher wants to change a procedure, consent process, or data collection form.

  • Adverse events or unexpected problems: A review should be done at the time any such events or problems occur.

  • Continuing review: Such a review is done at an interval specified by the IRB at initial review and is dependent on the risk level for the study (e.g., monthly or quarterly, but no less often than once per year).

Depending on the nature of an amendment or adverse event, and at the discretion of the IRB, it may be acceptable to use an expedited IRB review procedure for these types of reviews. Expedited reviews can be done by an IRB chair or experienced IRB member (rather than by the full committee), but they must include consultation with the prisoner representative. No research involving prisoner-subjects should be deemed exempt from IRB review. Even a low-risk questionnaire involving prisoner-subjects requires an IRB review.

There are other situations for which IRB review is required. A preliminary review can be conducted before funding has been obtained for the study. During this review, the IRB can assess at the most fundamental level whether prisoners can be ethically involved in the proposed research. If so, the adequacy of the protections are assessed and suggestions offered to the researcher about how to strengthen the protections. However, the preliminary review does not substitute for the initial review. No subjects can be contacted and no information can be accessed based on a preliminary IRB review. An IRB review is also needed when an investigator wants to use data for research purposes that was collected during an activity initially classified as treatment or QI (i.e., not research).


Although approval of research by the IRB or other independent review committee is a critical first step in protecting research subjects, it is not sufficient. Research involving prisoners, like other types of research, must be monitored throughout the course of the study to verify that study procedures are being conducted as approved and to detect adverse events or unexpected problems in a timely manner. Ongoing monitoring, then, is another key issue that must be considered in the new oversight requirements. The monitoring process may need to differ depending on the setting or type of study. For example, studies that take place in closed institutions that restrict subjects’ access to investigators or advocates may require more proactive (inside the institution) monitoring than those that take place in

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement