Volume XLIX, Number 2
Authors
- Fanny K. Ennever, PhD, CIP, Boston Medical Center
- John F. Ennever, MD, PhD, CIP, Boston Medical Center and Boston University Medical Campus
Introduction
Since 2013, institutional animal research facilities have been required to have a contingency plan that covers care in the event of common emergencies such as electrical outages, fires, and natural disasters (Animal Plant Health Inspection Service, 2013; Donaho, 2014). In May, 2018, the Office of Human Research Protection (OHRP) and the Food and Drug Administration (FDA) finalized guidance (first issued in draft form in August, 2016) containing 55 recommendations for what should be included in written procedures for Institutional Review Boards (IRBs) that oversee human subjects research. The 51st recommendation is:
- 51. Contingency plans for transferring oversight of one or more studies to another institution or IRB in the event that the IRB is unable to continue oversight of the study (e.g., the IRB closes, suffers loss due to fire, natural disaster). (OHRP & FDA, 2018, p. 13)
No additional recommendations or discussion were provided on how IRBs should develop contingency plans, and the final guidance omitted any mention of FDA guidance on transferring oversight (FDA, 2014) that had been referenced in the draft guidance (OHRP & FDA, 2016).
The draft guidance was issued during a time that the joint Boston Medical Center and Boston University Medical Campus Human Research Protection Program (BMC/BU Medical Campus HRPP) was in the process of overhauling the IRB policies in preparation for accreditation by the Association for Accreditation of Human Research Protection Programs (AAHRPP). The Institutional Officials (the individuals with the overall responsibility for the IRB at BMC and BU Medical Campus, respectively) both recognized the wisdom of developing a contingency plan, in the hope, of course, that it would never be used.
The contingency plan was developed with separate consideration of disruptions to electronic records and to personnel. Responses are calibrated to the scope and severity of the disruption, including assuring that electronic records could be reconstructed within one week and incorporating the option of having an independent IRB perform reviews according to the BMC/BU Medical Campus policies. The important components of the contingency plan are summarized in Table 1. This publication reflects on the outcome of the development of the plan, notably the way that planning for the transfer of oversight was expanded to include developing processes to respond to disruptions that would not, in fact, require transfer of oversight.
IRB Operational Responsibilities
Investigators at institutions receiving Federal funding must obtain IRB approval before conducting any activities that meet the definition of research with human subjects and that do not qualify as “exempt” [note, however that most institutions still require that IRBs perform an initial review even for exempt human subjects research (Loe, Winkelman, & Robertson, 2016)]. In addition, investigators may not make any changes in the approved research protocol without prior IRB approval, must report untoward events such as unanticipated problems and protocol deviations to the IRB, and must provide information for the IRB to re-approve the research protocol annually. [Note that this last requirement for annual review will only apply to research that poses risks that are greater than minimal under new Federal regulations that fully go into effect on January 21, 2019 (O’Rourke, 2017).]
Institutions may rely on an IRB other than their own to review research (called “ceding review”) by entering into a reliance agreement with the reviewing IRB. BMC/BU Medical Campus has reliance agreements for industry-sponsored multi-center drug and device studies to be reviewed by independent IRBs, one of which has established access to BMC/BU Medical Campus’s electronic system. Independent IRBs have taken over IRB functions for institutions in the past, either temporarily or permanently (Lis & Murray, 2008), and provide a key resource in contingency planning.
The consequences of disruptions to IRB functioning would be delays in the ability of investigators to start new research projects, to make changes to existing projects (including adding study staff), to continue projects that are near the expiration date of annual re-approval, and/or to receive assistance in responding to unforeseen events.
OHRP and FDA Guidance
Included in the draft recommendation about contingency plans (OHRP & FDA, 2016) was a reference to a 17-page document (FDA, 2014) with guidance on how IRBs should handle the following 8 steps involved in transferring clinical investigation oversight to another IRB:
- Identifying those studies for which IRB oversight is being transferred;
- Ensuring the availability and retention of pertinent records;
- Establishing an effective date for transfer of oversight, including records, for the clinical investigation(s);
- Conducting a review of the study(ies) by the receiving IRB, where appropriate, before it accepts responsibility for the study(ies);
- Confirming or establishing the date for the next continuing review;
- Determining whether the consent form needs to be revised;
- Notifying the key parties; and
- Updating IRB registration information.
The process of developing the plan for BMC/BU Medical Campus identified several additional issues that should be addressed:
- Specifying the criteria for deciding that transfer of oversight is necessary;
- Identifying and contracting with the IRB(s) that will receive the transfer of oversight;
- Determining whether the receiving IRB(s) will carry out reviews according to their own policies or according to the BMC/BU Medical Campus policies (i.e., acting as a panel for BMC/BU Medical Campus); and
- Deciding whether to take back oversight at the end of the disruption.
Addressing the first point above was particularly useful, leading to the recognition that plans could be put in place for managing many kinds of disruptions without the need to transfer oversight.
Disruptions to IRB Functioning
IRB functioning is dependent on staff and IRB members who have access to the records of communications with investigators. Disruptions to IRB functioning can be characterized by their scope and severity. The scope of a disruption depends on whether and to what extent the disruption
- Reduces or prevents access to records
- Makes some or all personnel unavailable to work
The severity of a disruption depends primarily on how long before operations return to normal. In addition to planning for the transfer of IRB oversight, the BMC/BU Medical Campus contingency plan prioritized developing ways of responding to disruptions quickly enough so that no transfer would be necessary.
Identification of Responsibilities
The plan designates the IRB Director as the individual who will lead the response to a disruption. If the IRB Director is unavailable, the responsibilities transfer to the Institutional Officials, the IRB Chairs, and IRB Administrators, in that order. A key responsibility is communicating to investigators about the reasons for, responses to, and anticipated duration of the delays in IRB operations.
Response to Disruption of Access to Records
BMC/BU Medical Campus uses an electronic system to manage records concerning submission, review, and approval of research. In August, 2016, the electronic system (both the software and the data) were already being backed up on a daily basis on a BU server that is located in a different building than the server that holds the production version of the electronic system (“local backup”). However, it was recognized that both versions could be damaged or destroyed by a wide-spread event such as a hurricane or earthquake. Thus, implementation of the contingency plan included arranging for another backup of both the software and the database to be made on tape on a daily basis. This involved adding the IRB records to the backup procedures already in effect through the BU Information Technology (IT) disaster recovery agreement with SunGard, wherein the tape is stored in a location in a different State (“remote backup”).
Recovery using the local backup is expected to take only a day or two to return to normal operations. The process is for the IRB Director make the determination that restoration from the local backup is required and for IRB staff coordinate with the BU IT department for the restoration of the electronic system and reconstruction of any records that were added to the system after the time of the backup.
For remote backup, incorporating the IRB records into the existing disaster recovery agreement involved choosing a timeframe for restoring the records. The decision was made to accept a period of one week after the date that disaster recovery was invoked for restoration of IRB records. The one-week goal was considered appropriate for IRB responsibilities because waiting an additional week for review of new protocols and amendments, while not ideal, would not be expected to significantly impede research. For annual renewal, investigators are expected to submit progress reports for continuing review to the IRB at least six weeks before the expiration date of the study. Thus, only investigators who had not met this expectation might be forced to cease study interventions (except for those required for the best interest of the already enrolled subjects) until records were restored. Investigators are also expected to submit initial applications well in advance of any need for IRB approval for funding purposes; again, a funding deadline might be missed if a submission was made less than a week before the funding deadline if records were unavailable. The one-week timeframe could potentially be problematic if an unforeseen event involving a fatal or life-threatening incident occurs during the disruption. However, the immediate response to such events could be accomplished without access to IRB records.
The process for remote backup is for the individual from BU responsible for disaster response to make the determination that a disaster requiring recovery has occurred, and for the IRB Director and IRB staff to coordinate with SunGard and the BU IT department for restoration of the electronic system as well as reconstruct any records of IRB actions after the time of the backup.
Because IRB records will be restored within one week, a disruption of access to records would not require transfer of IRB oversight.
Response to Disruption of Availability of Personnel
An unexpected lack of availability of some or all IRB staff and/or IRB members can be caused by multiple resignations, pandemics, and interruptions in electricity and/or internet service to work and/or home. Note that the inability of IRB staff and IRB members to travel to the IRB office location would only constitute a personnel disruption if electricity and internet access were widely unavailable, because IRB staff and IRB members are able to use the electronic system from home and participate in convened meetings via teleconference. In the specific instance that the IRB becomes subject to administrative actions by FDA under 21 CFR 56.120 or 56.121 or by OHRP under 45 CFR 46.103(e) that include limitations on the IRB’s authority to provide oversight, as has happened to other IRBs in the past (Lis & Murray, 2008), this would also be considered a personnel disruption for the purposes of the contingency plan, but would be likely to be known further in advance than other personnel disruptions.
To obtain external resources for responding to personnel disruptions, the contingency plan relies on the services of one or more independent IRBs which already provide oversight for some research at BMC/BU Medical Campus. The process for responding is for the IRB Director to decide whether or not a personnel disruption is likely to have a significant negative impact on IRB operations without external help, and to identify how soon the disruption is expected to be resolved (e.g., sick IRB members recover, additional staff are hired, new computers are purchased). If additional resources are needed, the IRB Director will initiate communication with one or more independent IRBs who are willing to perform reviews following the policies and procedures of the BMC/BU Medical Campus IRB. This would not constitute transfer of oversight to the independent IRB, because the independent IRB would be functioning as one of the panels for the BMC/BU Medical Campus IRB.
Some disruptions may be so severe that the IRB Director, in consultation with the Institutional Officials, will determine that investigators would be best served by transferring oversight to the independent IRB. In such a situation, the detailed FDA recommendations for transferring research oversight will be followed (FDA, 2014). The IRB Director will monitor the situation to determine when the disruption has been resolved and the services of the independent IRB are no longer required. If oversight of any studies has been transferred to the independent IRB, the IRB Director will decide whether or not to leave the studies with the independent IRB for the life of the study, taking into account the burden on investigators and the capacity of the IRB.
Reflection
The process of developing the contingency plan was a valuable exercise in identifying the steps that would need to be taken if IRB operations were disrupted. As it turned out, the planning process was particularly useful in thinking through responses to disruptions in ways that would minimize the situations where oversight would have to be transferred, including assuring that electronic records could be reconstructed within one week and incorporating the option of having an independent IRB perform reviews according to the BMC/BU Medical Campus policies. A similar process could be useful to other institutions as they incorporate this requirement into their IRB written procedures.
The hope is that none of the components of the plan will ever in fact have to be implemented, but thinking through the process certainly increased the sense of being prepared.
Authors’ Note
The authors gratefully acknowledge the help in developing the contingency plan that was provided by the members of the HRPP Advisory Committee at BMC/BU Medical Campus.
Fanny K. Ennever, PhD, CIP
Research Compliance Office
Boston Medical Center
Boston, MA 02118
(617) 638-8874
Email: Fanny.Ennever@bmc.org
John F. Ennever, MD, PhD, CIP
Director, Office of Human Research Affairs
Boston Medical Center and Boston University Medical Campus
Boston, MA 02118
(617) 358-5377
Email: ennever@bu.edu
References
Animal Plant Health Inspection Service, United States Department of Agriculture. (2013). Questions and answers: Final rule on contingency plans for regulated entities. Retrieved from https://www.aphis.usda.gov/animal_welfare/content/printable_version/contingency_ rule_faq.pdf
Donaho, J. (2014). Building the human component into contingency plans. Lab Animal, 43(1), 26-32. doi:10.1038/laban.423
Food and Drug Administration. (2014). Considerations when transferring clinical investigation oversight to another IRB. Retrieved from https://www.fda.gov/downloads/Regulatory Information/Guidances/UCM307779.pdf
Lis, J. M., & Murray, M. G. (2008). The ins and outs of independent IRBs. Journal of Health & Life Sciences Law, 2(1), 73-122.
Loe, J. D., Winkelman, D. A., & Robertson C. T. (2016). An assessment of the human subjects protection review process for exempt research. The Journal of Law, Medicine & Ethics, 44(3), 481-491. doi:10.1177/1073110516667944
Office for Human Research Protections & Food and Drug Administration. (2016). Draft Guidance. Institutional Review Board (IRB) written procedures: Guidance for institutions and IRBs. Retrieved from https://www.fda.gov/downloads/Regulatory Information/Guidances/UCM512761.pdf
Office for Human Research Protections & Food and Drug Administration. (2018). Institutional Review Board (IRB) written procedures: Guidance for institutions and IRBs. Retrieved from https://www.fda.gov/downloads/regulatoryinformation/guidances/ucm512761.pdf
O’Rourke, P. P. (2017). The Final Rule: When the rubber meets the road. The American Journal of Bioethics, 17(7), 27-33. doi:10.1080/15265161.2017.1329484
Have any questions?
Contact the editor »#VolumeXLIXNumber2