UCMC pediatrician to streamline clinical trials

Professor and pediatrician awarded funding to create central database in order to reduce systematic redundancies in trials.

By Emma Dries

Sam Volchenboum thinks he’s found the cure to the inefficiencies of clinical trials.

Volchenboum, an assistant professor of pediatrics and the director of the informatics program at the University of Chicago Comer Children’s hospital, was awarded $40,000 by the Innovation Fund to develop a pilot program that will help streamline and automate the clinical trial process in June.

“From our survey of the field, his vision seems the most holistic,” said Eric Ginsberg, assistant director at University of Chicago Tech, in an e-mail. The Innovation Fund, run out of University of Chicago Tech, awards grants to product ideas developed at the University that would not otherwise be supported by traditional funding sources because they are too early in development.

“[Volchenboum’s] system starts by capturing the necessary elements of a clinical trial through a guided process which helps automate the downstream processes,” Ginsberg said, likening the process to the tax preparation software TurboTax, which helps taxpayers fill out tax forms usually filled out by hand, and puts the right pieces of data in the appropriate place.

According to Volchenboum, the overarching problem with clinical trials is how manual the process continues to be. Currently, the greater consortium running the trial will first put a static PDF protocol on its Web site for participating sites to download, look through, and develop patient criteria and eligibility on their own terms. Each hospital has an independent Institutional Review Board (IRB) that reviews the trial and ensures it is safe for its patients according to its unique rules and regulations.

“You can imagine that the opportunities for automation are tremendous and the possibilities for mistakes are huge,” Volchenboum said. “This was the frustration that I started with and I knew there had to be a better way to do this. The main problem is that [the protocol is] living as a static document. You can’t ask this document ‘What are the drugs?’ and ‘What are the doses?’ because it’s just a PDF file.”

Instead, Volchenboum said, having a database with a list of drugs, doses, timings, side-effects, labs, and so forth, would allow hospitals to automatically generate a protocol for carrying out the trial. Such a database would eliminate unnecessary and redundant paperwork, as well as minimize manual data input to and extraction from the Electronic Medical Record (EMR) system.

Unfortunately, there is no universal EMR. According to Volchenboum, there are 70 EMR companies, five of which control 80 percent of the market. One problem that Volchenboum foresees as he moves forward in developing this pilot program is finding a way to work effectively with the EMRs.

“Most of the time the EMRs are basically these ivory towers,” he said. “You might be able to pull data out, but you really can’t get anything in right now. The idea of sending a set of labs and orders into an EMR and have it read and then do something with it is something that we’re still not good at.”

However, that could soon change. According to Volchenboum, due to the Affordable Care Act, the government will start requiring a certain amount of interoperability within the EMR systems. But due to the inherent differences in every hospital’s IRB and individual protocol, developing a universal generator poses another set of challenges.

That being said, in building the system, “what I want to do is use those protocols that we started here but then work with these hospitals to see what is idiosyncratic about their system and try to build something that works everywhere,” Volchenboum said. “The whole point of doing it this way is to create something that’s agnostic to the institution.”