The FDA and document extraction specialist Captricity team up to speed up data entry while cutting costs.
What would bureaucracy be without miles and miles of paperwork?
The U.S. Food and Drug Administration intends to find out. On Thursday, Nov. 14, at Amazon’s Web Services conference, FDA Chief Health Informatics Officer Taha Kass-Hout announced an agreement between the administration and data extraction specialist Captricity to help decongest the FDA’s massive backlog of paperwork.
Unlike many data extraction companies, Captricity says its technology can understand human handwriting as accurately, or better, than its human counterpart.
"For the paper jam we experienced, it was originally estimated to take a very long time to overcome the backlog using the current approach of manual data entry,” Kass-Hout said. “Captricity was part of a solution that was timelier, and we were able to achieve very high quality data at the fraction of the cost."
The bulk of the paperwork had been submitted to the FDA’s Adverse Event Reporting System, or FAERS, a new database that contains reports from the public, health professionals and industry on drug side effects and medication errors linked with marketed drugs.
In a June release, the FDA reported it receives about 900,000 FAERS reports annually with 10 percent of those being written manually. That 10 percent (or 90,000 documents) had forced the FDA to hire additional data entry staff while prioritizing only the most critical reports such as those indicating a product may be harmful and reports related to emerging safety issues targeted by officials.
“The best big data tools out there don’t matter if we can’t get to the data in the first place,” Kass-Hout said.
The agreement, approved in August with services already under way, has saved both time and money for the FDA. In the way of numbers, Kass-Hout said that compared to the FDA’s previous methods it will be able to reduce costs by roughly eight times what it had previously paid. The savings, he said, would be coupled with an accuracy rate greater than 99 percent.
“We are currently evaluating this approach to deal with many other kinds of manual submissions to the agency,” he said.
Captricity CEO and Founder Kuang Chen hopes the agreement will set the stage for more federal service agreements. Chen said he’s already in discussions with other government entities and envisions the company — which started as part of Code for America’s 2012 Accelerator program — helping more agencies such as the U.S. Department of Homeland Security and U.S. Health and Human Services Department, both organizations dealing with heavy data entry loads.
Photo: Captricity CEO Kuang Chen
As with any federal IT agreement, Chen said it was one that took years of planning and research, not just to gain the necessary approvals that were required, for example HIPPA compliance, but also the need for intensive research that dates back to Chen’s dissertation at the University of California at Berkeley where he received his Ph.D. in computer science.
Looking at Captricity’s long-term benefits in government, Chen said he sees its major draw as more than being just a technological time saver, but as an efficient way to cut costs in today’s lean government structure that can’t afford to be as dependent on human capital. Chen estimated that depending on the documents' size, for agreements similar to the FDA's, private and public organizations can decrease cost by five to 10 times.
“For government right now I think this is a huge opportunity to save a lot of money,” Chen said. “You're getting human quality for the speed and cost of software.”