A debate is currently underway on whether local IRB review is better than external IRB review. Much of the debate surrounds the fact external IRBs have the advantage of quicker turnaround times, plus funds devoted solely to customer service. Many university IRBs have found it difficult to meet this challenge. (Whitney et al., 2008). The following inquiry examines how one IRB adopted external IRB practices and made themselves competitive with their external colleagues, settling some of the debate. By doing this, they were able to maintain high quality local review, while increasing efficiency and levels of service.
The concerns over IRB efficiency across the country have prompted a series of articles and opinion papers on the potential for a redesign of the IRB review system, utilizing a single IRB for all multi-site research. (Wechsler, 2007). Debates over whether a central IRB model is practical and whether the use of for-profit IRBs would decrease the quality of review due to the financial stake those entities have in the volume of reviews have been carried out in the literature. (Lemmens & Freedman, 2000). During the summer of 2011, an Advanced Notice of Proposed Rule Making (ANPRM) was released by the Office for Human Research Protections (OHRP) which included a call for input on how a central IRB model for multisite research might be designed in order to reduce the bottlenecks caused by lengthy IRB reviews at each research site. (Federal Register, 2011).
The national discussion on whether local or central Institutional Review Boards (IRB's) better serve to protect human research participants has raised questions as to whether the use of central boards to the exclusion of any local review would eliminate the consideration of local context and prevent institutions from evaluating the ability of their research program to safely conduct the study. (Wechsler, 2007). In addition, a long standing concern about the quality of review conducted by for-profit IRBs has made some in academic medicine skeptical of reliance of these entities for oversight of clinical trials. (Lemmens & Freedman, 2000).
Improving IRB efficiency and review quality has been a constant goal for the WFSM IRB, but in 2011 there were institutional priority shifts, and changes in leadership that focused the attention of research support leadership on finding ways to overhaul the support mechanisms immediately so that clinical research capacity at the institution could be increased without the need to hire new staff members or increasing budgets. It was also important that the quality of review and service to investigators remain high.
The Impetus for Change
In July 2011 Wake Forest School of Medicine (WFSM) welcomed a new Dean. Following the Dean's arrival, the priorities of the research program were reassessed in the light of a recently developed strategic plan for the academic medical center. One of the new priorities set by upper administration was the growth of the clinical trials research program. To make this possible, the Dean called on research administration to find ways of increasing capacity by decreasing turn-around time.
Turn-around time in July 2011 for initial human research submissions requiring full board review by the Wake Forest School of Medicine IRB was averaging approximately 45 days. This was comparable with the national average of 46 days cited by the Association for the Accreditation of Human Research Protection Programs (AAHRPP) as the average approval time of accredited IRBs who completed their latest performance metrics survey (AAHRPP, 2011). Most investigators were not dissatisfied with the review times at WFSM, but the Dean previously served in an Academic Medical Center that utilized a central IRB for industry sponsored clinical trials in order to achieve a top level of efficiency. WFSM IRB leaders knew that in order to meet the Dean's expectations, efficiency needed to meet commercial IRB levels without any reduction in the quality and thoroughness of protocol review. To meet this goal an examination of several commercial IRBs revealed that a common practice among for-profit boards was the use of a large number of panels and the conduct of frequent meetings. This approach allows for an even distribution of work, and a timely, thorough review.
Creating a New Review Model
The Wake Forest School of Medicine IRB operational model had traditionally consisted of four Boards with each one meeting once per month. Agendas for each meeting averaged 16 to 25 items for review. Boards were made up of about 18 regular members, which required a quorum of 10 people in order for business to be conducted. This was not an atypical model for an academic medical center. Inquiries into how similar organizations to ours operated the IRB review system revealed that many academic medical centers had three or four Boards that each met once per month. Problems with the traditional model included:
Tabled studies had at least a one-month wait for reconsideration, unless a special meeting was held;
Quorum was sometimes difficult to obtain or maintain during meetings especially since meetings required an entire afternoon and patient visits or emergencies often put clinical members behind for arrival or in need of early departure;
Agendas were so large that the primary reviewer system was relied upon more than we wished it to be (i.e. it was difficult for all members to personally review each submission to be discussed at the meeting);
Minutes from meetings were long, arduous documents to create and their creation was sometimes put off until just before the next meeting (which lessened the level of detail within the documents); and
The number of submissions on each agenda meant that memoranda to study teams explaining the results of the meeting were often not sent until the following day or the day after that given the complex compilation of notes that had to be put together for each submission.
After assessing what would be needed in order to gain the kind of efficiency necessary to be on par with commercial IRBs, eight boards was the number we thought represented the ideal. With eight boards that each met twice monthly we could solve several problems that arose from the traditional model without having to hire additional staff or increase the budget.
Tabled studies would be handled twice as fast with...