New web site centralizes multicenter submissions

IRBNet designed to improve communication

Developers of a new web site for researchers and IRB administrators hope it will help streamline review of multisite research by centralizing protocol creation and enhancing communication among research sites., which went on-line a little more than a year ago, was developed by Dartmouth College and the Children’s Hospital of Philadelphia.

Kristin Nowak, MPH, IRBNet project manager based at Dartmouth in Hanover, NH, says the goal of the site is to allow for cooperative review of multisite studies, which would help ease roadblocks for researchers and allow institutions to communicate among themselves regarding the studies.

"In these multisite studies, IRBs weren’t talking to each other at all," she explains. "You could have a concern that an expert on the committee at one site may bring up because they have more knowledge about the subject area, but at another site you may not have that sort of expertise, and the issue may never come up."

Another feature of the site allows a researcher to create a protocol and consent form on-line using a best practice model that follows International Committee on Harmonization guidelines. The web site then could serve as a distribution center for the study documents and an archive for past and current studies, allowing investigators to check on the progress of submitted studies at various IRBs.

Nowak says IRBNet was the brainchild of Elizabeth Bankert, MA, director of Dartmouth’s IRB, and Robert Nelson, MD, PhD, chair of the IRB at The Children’s Hospital of Philadelphia, who pooled NIH Enhancement Grant funds to start it up.

As Bankert gave presentations on IRB review, Nowak says she heard the same complaints from researchers about the cumbersome process of gaining approval for multisite clinical trials.

"Whenever Liz was going out, she was hearing from researchers what a big hassle it was to deal with IRBs," Nowak recalls. "When you’re only dealing with one IRB, you’re going to have a decent sort of sense of what they want. But if you’re the lead site of a multisite trial, every IRB is generally going to want things in their own format and they’re going to have their own schedule as far as their submissions and renewals."

They developed a plan for IRBNet, which still is evolving, Nowak says.

Currently, it features a three-phase process of study design, distribution, and communication:

  • Study Design — An investigator can either upload previously created protocol documents to IRBNet or create new study documents using IRBNet Study Designer, a document "wizard" that leads the investigator step by step through the process of creating a protocol or consent form. Because the Study Designer uses the ICH’s best practice model, it promotes standardization of studies.

"[Documents created by Study Designer] should contain everything that anyone would need to submit a document to their IRB," Nowak says, while acknowledging that some IRBs might have different institutional requirements. "Our hope would be that eventually everyone would accept documents produced by IRBNet, but we’re aware that at this point in time, that may not be entirely realistic."

  • Study Distribution — Using IRBNet, researchers can give their collaborators access to study documents during the design phase, distribute the documents to local researchers at various sites, or transfer ownership of the documents to another researcher or sponsor, all on-line.

Nowak says a researcher can create a database of all the studies he or she has created and all the documents that researcher has been working on.

"You can see whether they’ve been submitted, what the review status is, who they’ve been shared with," she says.

  • Communication — An important aspect of IRBNet is the ability it gives IRB administrators to communicate with each other regarding multisite studies they’re involved in, Nowak notes. An administrator at an IRB involved in a study can see the status of reviews of the study at other sites — including any comments that are available for review.

"They would be able to communicate with other IRBs if it was a multisite trial as to why a review status was as it was," she says. "If something was not approved, why it was not approved, if it needed revision in some manner or another."

Nowak says the site still is in the process of signing up a multicenter study that could be completely created using IRBNet. While there’s a lot of enthusiasm about the idea — IRBNet already has more than 400 registered users — finding a study that could be begun on the site, registering everyone needed and getting all the documents into the system has taken some time, she says.

"We really want to get a multisite study through the system because it will also help us see where some of our weaknesses lie," Nowak says. "I can go through and test the site all that I want and can think everything is fine, but it’s not until multiple people are actually using it as it’s intended to be used that certain things will come up."

Nowak says IRBNet has the capacity to customize forms or add new ones if an IRB was willing to use the study plan but wanted a different version of say, the consent form. The form could simply be available for download and submission, or in a more time-intensive process, IRBNet could put the form into the system so a researcher could be guided by a wizard as he or she filled out the forms.

"There are lots of possibilities for how IRBs could get what they want out of it, or how it can be made most useful and acceptable for IRBs," she says. "The goal is that this will streamline the process and also increase communication between IRBs. That would be the ultimate goal."

There’s also the issue of funding, since the NIH grants that started IRBNet will be ending soon. While applying for new grants, Nowak says the developers also are considering charging fees to sponsors or physician consortiums that run protocols through the system — perhaps customizing the site to meet their specific needs.

(Editor’s note: Registration at is free, but must be approved by a site administrator.)