Advantages to using web-based data system

Timely pay, ongoing data available

There are a number of advantages to using a Web-based data management system, including the following:

CT sites are paid on time: With the Web-based data management system it was very easy for people to be paid because they didn't have to generate invoices, says Richard Buchsbaum, senior data manager at the statistical analysis center, biostatistics department of Mailman School of Public Health, Columbia University of New York, NY.

Sites that are paid on a per procedure basis will have this information generated automatically by the system.

"The submission of data that answered the research question also recorded the answers from procedures for which they were reimbursed," Buchsbaum says. "It was all the same thing, so in some ways the workload was less for clinical trial sites because they only had to put in the data once."

Then once a month the system would generate an invoice based on the data captured, he adds.

Ongoing data are available: "Data collection could be viewed in real time, and most of the error tracking was programmed into the user interface, so there were very few possible queries to generate," Buchsbaum explains. "All were done on a regular basis, leaving very little collection or data cleaning to be done on the back end before giving the statistician the data for analysis."

Keeping data collection in real time creates inherent efficiencies. For instance, the closer data collection is to its source as you collect data from a procedure, the more likely it is to be accurate, Buchsbaum says.

"So part of training with the system is to have people enter the data very soon after it's collected and to immediately evaluate it for completeness and possible errors," he adds.

In some cases this data entering would be done within a few days, and the feedback on the data was immediate, Buchsbaum says.

"So the whole process of eliminating errors was close to the time of entry and completion," he says.

Hardest work is done by the trial's end: "So when we got to the final visit, everything up to that point was cleaned up, and we literally ticked off patients as they came across the finish line," Buchsbaum says.

"So when the final patients came in, there wasn't that much to do," he adds. "The efficacy dataset was pretty well complete."