Infuse best practice strategies into managing multisite trials 

The key? Improve communication between sites

When clinical trials administrators are working on a study that is part of a multisite trial, the biggest challenge is communicating with everyone involved.

There are many areas in which best practices in communication might improve coordination of multisite trials, including at the beginning of the process when it’s important to make certain everyone is implementing the protocol the same way, says Mark Thornquist, PhD, senior staff scientist at Fred Hutchinson Cancer Research Center in Seattle.

"We set up systems to do a distribution of responses through an on-line system to which everyone involved in the study has access," says Thornquist. "We track inquiries and ensure the response is sent out to all sites to make sure all are interpreting the protocol in the same way."

When it’s necessary to hold teleconferences, time differences can form a barrier to effective communication.

"You have to be flexible because you might have sites throughout the country, and we have to be very aware of time differences," says Cim Edelstein, COMPASS coordinating center manager of Fred Hutchinson Cancer Research Center.

"You have to be careful to set up meeting and conference calls that can accommodate the people in different time zones," she adds.

When a study has trial sites around the world, the time difference is just one factor that makes communication more challenging. Another problem is dealing with multiple languages, says Jennifer Yahne, contracting projects manager at Fred Hutchinson.

Yahne works with the HIV Vaccine Trials Network, which coordinates clinical trials of potential HIV vaccine candidates at 30 sites in the United States, Africa, South America, the Caribbean, and Asia.

"We try to operate our particular part of the operation 12 hours a day from 6 a.m. to 6 p.m.," reports Banks Warden, MHA, chief operating officer of the HIV Vaccine Trials Network at Fred Hutchinson.

"We’ve tried to hire people who have [foreign] language skills, and we have the people who work in our travel office and conference call office also have significant language skills, " he says.

When one site is coordinating the multisite study, it’s important that the site’s IRB establishes effective communication with the other sites and IRBs involved, says Karen Hansen, director of institutional review office at Fred Hutchinson Cancer Research Center.

"If your site is the main coordinating center for a study, it’s really important for the IRB to communicate and work with lead investigators early on to identify the roles of all players and to make it clear who is participating in what level of activity," she advises.

Hansen and the other experts offer these additional guidelines for establishing best practices in coordination of multisite clinical trials:

1. Hire a central coordinator.

"Having a central coordinator for multisite trials is critical," Edelstein says. "You have to have a designated person who is the coordinator of the project, not multiple people coordinating activity."

Multisite trials need one project leader who is making sure everything necessary is done at all sites, Thornquist says.

A central coordinator or project manager might not have all the answers when inquiries arise, but will know who to contact to find the answers, Edelstein adds.

For instance, if a trial site encounters a programming glitch, then the project manager will talk with the information technologies staff; or if there’s a scientific question, the project manager will meet with the trial’s executive site committee, she explains.

Another good reason for having a project manager is because one designated person could establish relationships with investigators and staff at multiple sites and with sponsors, IRBs, and institutional staff, Edelstein says.

"I’ve been in this business for 20 years, and you find that you keep running into the same people. That’s helpful if you’re developing relationships where you know these people, and they know you," she adds.

2. Answer everyone when one person asks a question.

When multiple sites are conducting a clinical trial, it’s important to make the flow of information as efficient as possible by sending information and responses to questions to every site when one person involved in the project makes a request, Thornquist says.

One way to do this is through e-mail messages that are sent simultaneously to all involved.

For example, a research coordinator at one site might ask how and when a particular device is to be used. Once the answer comes from the principal investigator, it can be sent to every clinical coordinator and investigator, Thornquist explains.

Better yet, all inquiries and other updates could be stored in a central electronic location for referral throughout the trial, Thornquist says.

"Where there are clarifications to protocols or to modify protocols because the practice has been changed, we indicate a date of inquiry so from that date forward all information follows the date of that inquiry," he explains. "We call it our information management system."

The information management system tracks all internal review and stores questions and answers so the information will be useful when research staff need it, Edelstein says.

"Sites report they read these, and they use these as part of their training," Thornquist says. "They are required to go to the information management system daily."

Fred Hutchinson research staff had used a paper information management system before switching to an electronic version, and the electronic version is more efficient because it can get out common responses to inquiries much faster, he notes.

"Everyone has access to reading and submitting a question," Thornquist says. "The program looks through the database and pulls out those that may be similar to see whether the question has been asked before and whether it has been answered so it won’t be duplicated."

Get everyone on the same page

3. Create a manual of operation.

Fred Hutchinson’s institutional review office started a manual of operation for one study, and now it’s reviewed by the IRB each year, Hansen reports.

The institution keeps the manual current with information about all multisite clinical trial partners, standard operating procedures, sponsors, coordinating centers, IRBs, and it details all aspects of human subject protection and who the point person is for making certain all approval dates are in place, she says.

"It enables the IRB at the coordinating center level to have a solid footing on all activity that’s going on," Hansen explains.

The manual outlines all of the material that needs to be reviewed by the IRB, including every human subject protection measure that would be required by the Office of Human Research Protections (OHRP), she adds.

"This is a good mechanism when you have a multisite trial and your coordinator wants to have a handle on what’s happening at each site," Hansen explains.

Formatted in an Excel spreadsheet, the manual includes details, such as the OHRP memo of engagement, informed consent for subjects, regulatory operations center duties, the program medical officer’s role, and references, she adds.

4. Consider futures uses of collected data.

Too often in the past, clinical trial investigators and administrators had tunnel vision and thought only of the immediate uses of the data that would be collected during a course of a study, Edelstein notes.

Now, research professionals realize that sometimes the most important use of collected data might not be imagined at the time of a particular study, so it’s important to keep options open for future uses.

"For people who start any multisite trial, you have to look broad because the opportunities are very real," Edelstein says. "You have to think about that when you’re writing consent forms and think about that when you’re developing a tracking system."

Also, clinical trials staff will need to consider ownership issues and develop policies and procedures that can handle ancillary studies coming from a multisite trial, Edelstein says.

For example, a study of 10-20 years ago might have collected blood samples for the purpose of looking at the effect of certain vitamins on lung cancer, but then the potential to analyze genetic information occurred, and the same samples could be used for research that has more far-reaching potential, Edelstein notes.

"So you have to anticipate that science will change as you collect those samples," Edelstein says. "Twenty years ago, there was less concern about particular analyses being done on samples, but now it’s possible for consent to be open-ended, depending on the consents you allow."

Of course, all of this needs to be handled under the privacy rules of HIPAA, including obtaining IRB approval for all new uses of data, Thornquist notes.

"With HIPAA, the biggest impact has been ensuring that when you do share data or specimens that you do so with information or specimens being de-identified," Thornquist says. "The site receiving the information cannot determine specific individuals that the information came from."

Also, before investigators are given samples of data they are required to sign a privilege of received data form that will document what specific analyses they are permitted to do and that will inform them of the process they have to go through if they want to do additional analyses, Thornquist adds.

Another consideration is whether investigators will be able to use existing data effectively, and this might require a formal collaboration in the case of long studies in which the protocol has been modified many times over the years, he says.

"Sometimes, the only reasonable way to share data is through a formal collaboration so we can make sure they’re using the correct data elements, interpreting data right and interpreting it based on other data published in a consistent manner," Thornquist says.