Dr. Karla Hanson and her team were faced with the problem of bots attempting to enroll in their online research study. The study was aimed at collecting data in a rural community where a few enrollments per day were expected; however, suddenly there were hundreds of attempts to enroll in the online study. This was an unexpected outcome, given it was focused on a small community.

To address the issue, Dr. Hanson and her colleagues developed a protocol for reducing the number of enrollments by bots. The first step was to remove any enrollment attempts from IP addresses outside the geographic study area, which filtered out 25% of attempts.

Automated techniques only addressed part of the problem, so Hanson and colleagues used time-consuming manual methods by comparing submitted addresses to a postal database to ensure participants were real people. “It was very time-consuming and expensive to do all these active validation tests,” Hanson said. “And at each step, we found more fraudulent enrollment.”

Since participants were offered payments for participation, many people tried to enroll multiple times in addition to the bots. They found that 74% of the attempts were fraudulent. The protocol has been published so other researchers can adopt similar methods to ensure their data is reliable.