Internet guideline defines research, explores issues

Consider guideline a "living document"

When Sarah Fowler-Dixon, PhD, an education specialist in Washington University's Human Research Protection Office, began a project to develop the university's Internet research guideline for the university, she gathered a task force of IRB members, investigators involved in Internet research, and a technical advisor to help work through the complicated security issues involved.

"Part of the reason for having a task force is to get some buy-in as well," she says. "I primarily put the guideline together and then I ran it past the task force."

Fowler-Dixon says she searched for published guidelines on the topic, but didn't have much luck. She finally learned of a white paper developed at Duke University and used it, and other information to help put together a guideline that examined the range of possible uses of the Internet in research.

The guideline defines the various types of Internet research, including on-line surveys, questionnaires that could be downloaded and mailed in or e-mailed to participants, observation of behavior on web sites, recruitment of volunteers over the Internet, and the use of large public use databases.

In addition to exploring many ethical and practical issues involved in Internet research, the guideline also includes a detailed list of standards for any secure networks created at Washington University, to ensure privacy and confidentiality of protected health information (PHI).

One area the guideline doesn't address is issues surrounding e-mailing research participants, particularly if such communications involve the transmittal of PHI. Currently, Fowler-Dixon says, the university discourages the transmittal of PHI over the Internet.

"I'm actually working on a different guideline to address that," she says. "We didn't want to mix the two because then you're talking about a lot more biomedical research. The types of Internet research I've even heard talked about at national meetings are mostly social-behavioral type things that don't involve PHI."

Over the past several months, Fowler-Dixon has been tweaking the guidelines as new information becomes available.

"As we learn more, we want to make improvements to it — add a little bit more information to it," she says.

Other institutions considering developing their own guidelines should keep the "living document" principle in mind.

"It has to be flexible enough to be applied to a variety of studies," Fowler-Dixon says. "I think the issues that we address in these studies are pretty common to information being transmitted over the Internet — not Internet research only. Issues that are more specific to Internet research are going to be the transmittal of information and then the maintenance of confidentiality because that's going to be a little bit different."