Liaison efforts can improve compliance

Survey director worked at IRB

Compliance in research oversight has become a more prevalent concept in recent years, and this has led IRBs and research institutions to search for new ways to improve compliance communication and policies.

An IRB and the Institute for Social Research, both at the University of Michigan in Ann Arbor, have forged a new type of relationship with the common goal of improving the environment for conducting ethical research.

"We wanted to do more than improve compliance," says Steven G. Pennell, survey director for the Survey Research Center of the Institute for Social Research (ISR) at the University of Michigan. ISR is the world’s largest academic survey research organization and has been in existence for 20 years, conducting thousands of interviews every year with youths and adults.

"I like to think that we are improving commitment to the underlying ethical principles," Pennell says. "The idea is not to have people just do what they think the IRB wants them to do, in a rote way, but to think about the underlying ethical principles and proceed from there."

The IRB’s role in improving this commitment was to invite Pennell to work with the IRB, attending staff meetings and full board meetings for a couple of years.

"Over the course of that period, I spent at least half of my time here and half at ISR, Pennell says. He reviewed applications to identify areas that were not compliant with regulations, and he would work with investigators to find ways in which they could bring their applications into compliance.

"The IRB helped him have the insider’s view of why certain types of social science applications seemed to face difficulty within the IRB," says Judith Marie Birk, administrative manager of IRB Health/Behavioral Sciences at the University of Michigan.

Having Pennell as a liaison helped IRB members understand what ISR researchers knew about the IRB process and compliance. The IRB staff could then target areas in which researchers needed extra information and education, Birk says.

The IRB has had guidance documents available on line, but investigators probably did not refer to those when they needed help interpreting IRB comments, she notes. So investigators would be perplexed about some IRB requests. For example, when the IRB would ask for more information about research design, investigators would fail to see why that was relevant to the IRB, Birk says.

"If the IRB is just there for human subject protection, why would we be interested in the design of the study?" she says. "So we’d get a short, non-descriptive response to that question."

While working for the IRB, Pennell observed investigators’ responses, and he understood why the IRB was frustrated with the answers, so he worked at improving researchers’ knowledge of how the IRB process worked, Birk says. "Steve was able to be our interpreter. When the IRB says this is what they mean, he was able to help us spread the message."

Pennell improved the web site educational information that was available about the IRB process, including information that was specific to ISR. For example, one of the common reasons why research applications are returned to investigators is that one or more elements of informed consent are missing, Pennell says.

"The regulations are straightforward about what is required, and the IRB resources are available for people to review," Pennell says.

To bring the information closer to home, he worked on a guidance document that reviews best practices and informed consent practice, reminding researchers of what to do, step by step, he says.

"A lot of guidance documents were disseminated through a written internal newsletter originally, and then they were posted on an internal Intranet site that all staff could access," Pennell explains. "So they could go to the Intranet site and review information on best practices, informed consent forms, and look at large numbers of frequently asked questions."

Pennell also has provided direct application assistance to researchers. "I also had the opportunity to review an application before it was sent to the IRB, and so I could catch problems that way."

Many IRBs would benefit from having a liaison or experienced IRB staff person look at research protocols before they’re submitted to the IRB for review, Birk notes. "They can look at things with a critical eye and say, You may want to fix this, change this,’" Birk says.

"During the time I was a guest at the IRB, I identified various problems and approval of applications," Pennell says. "And we distributed the guidance documents through various communications we have at the institute, and we created the intranet site that included these guidance documents, as well as other areas that would help researchers better understand federal regulations."

There also were a series of brown bag lunch presentations on how federal regulations apply to social research, Pennell adds.

Having a liaison has improved outcomes because Pennell was able to contact the IRB office and discuss proposals that he had reviewed before they were submitted, Birk adds.

Pennell continues to provide this pre-submission review on an informal basis, at researchers’ requests, he says.

"One of the other things I did here while at the IRB is look at a lot of outcome data," Pennell says. "The creation of an application and review and approval of it creates enormous data, and we could look at the outcomes and establish some benchmarks of what the problems are, what issues resulted in an application being returned, and how long it took to approve an application."

By reviewing data, it was possible to compare the length of time it took to approve an application that didn’t have to be returned to the investigator versus the length of time to approve an application when the application had to be returned once or twice before all issues were adequately resolved, Pennell explains.

After changes were made as a result of the liaison relationship between ISR and the IRB, the percentage of expedited review process applications approved within two weeks or less increased, Pennell says.

The data show an increase from 56% in the Jan. 2003 to June 2004 time period to 78% in the Jan. 2005 to May 2005 period.

The data review showed that the most common area in which investigators make mistakes is with informed consent, Pennell notes.

ISR investigators use complex and innovative survey designs to address measurement and coverage issues, Pennell says. "Human subjects elements often have cost implications," he says. "For example, will we be asked to obtain consent, documentation from parents in order to speak with children at a school?"

It’s because of these cost issues that investigators need to know before submitting an IRB application how the IRB will address design issues that have human subject implications, Pennell explains.

Also, with survey research, there needed to be an acceptable and feasible consent process, so the IRB office and Pennell worked with investigators to try to craft language that contains the required elements of consent, but did so in a concise and cohesive way, he says.

These types of studies are low risk, Birk says. "We do a lot of research over the telephone, randomly, and speak with an adult in the household where we need to explain the nature of the study and eventually conduct a consent process," Pennell explains. "On the telephone the consent has to be quicker, so we worked with researchers to make sure they were meeting regulatory requirements, but also doing it in a way that minimized the likelihood that somebody would just hang up."