Research site collaboration builds efficient database for CR sites
Research site collaboration builds efficient database for CR sites
REDCap saves time and is good fit
Electronic data capture (EDC) systems are supposed to make data collection and analysis easier and more accurate for clinical research (CR) teams. But often CR coordinators and investigators find the systems to be counterintuitive, inflexible, and difficult to navigate.
A new EDC being used by Vanderbilt University of Nashville, TN, and other institutions across North America resolves this dilemma since it's being designed by a group of CR professionals and informatics experts who specialize in research.
Called REDCap for research electronic data capture, the EDC system has been built by a consortium of research professionals, who are using the software and helping to improve it, says Paul Harris, PhD, research associate professor in the department of biomedical informatics and director of the clinical and translational science awards (CTSA) biomedical informatics operations at Vanderbilt University. Harris also developed StarBRITE, an electronic portal that assists researchers with finding information.
"We were finding that everyone now knows how important it is to have audit trails in your data, to have data capture, and to authenticate users and protect confidentiality, etc.," Harris says. "But I find that in basic clinical research studies there is nothing out there that's easy enough for researchers to use on their own."
With the goal of making data capture easier for investigators, Harris began work on an electronic tools project, which eventually became REDCap, in 2004.
REDCap is a user-friendly database for entering information and tracking people, says Gail Mayo, RN, a research services consultant in research support services at Vanderbilt University.
CR coordinators can use REDCap to track study participants, see what information is missing, and note where participants are as the study progresses, Mayo says.
"It allows multiple users on the same database," she adds. "So long as you're at an Internet-accessible computer, you can access information."
One of the ways the Internet and various open source software programs have evolved and improved over the decades is through sharing the systems with the people who will be using them. Everyone uses the electronic system, tweaks it, and submits their improvements for everyone else to use. Harris set up a similar process for REDCap through a consortium of research institutions that share the EDC system.
"We have an expectation that you help us contribute back to the project and make it better," Harris says. "This is why it's one of the best systems I've ever seen."
There are EDC systems available, but often they lack the flexibility required for CR work, he notes.
"Clinical research studies are different in scope and formulization of systems," Harris says. "We need something fast and flexible to support autism studies one day and diabetes or pain studies the next day."
While all EDC systems collect demographic data in similar ways, they need a great deal more diversity for some of the other details collected in clinical research, he adds.
"This system needed to be flexible and fast, and we wanted to offer it to anyone as a free level of support for their research, Harris says.
The Oregon Clinical and Translational Research Institute of Oregon Health & Science University in Portland, OR, got involved with REDCap after responding to a request for collaboration sent out by Harris.
"There was a small group of collaborators in the beginning, and that's been expanding," says Hannah Howard, informatics database analyst at the Oregon research institute. "We've been involved in the collaboration for one-and-a-half to two years."
REDCap has some limitations, which collaborating partners are working on, Howard notes.
"The collaboration works really well," Howard adds. "When there's some technical problem to iron out then smaller subgroups work on it."
The main advantage to REDCap is that it's an application that comes from people directly involved with human subjects research, says Brian Welburn, BS, clinical trials data manager at the University of Puerto Rico Comprehensive Cancer Center in San Juan, PR.
"All of these collaborators are people who provide services to people who have databases," Welburn says. "This system is specifically developed for research purposes, and I think it incorporates a lot of the experiments of all of these people who have been working in the research community for years."
The University of Puerto Rico became involved with REDCap after a programmer at the institution visited a research meeting for informatics professionals and learned about the system there, says José G. Conde, MD, MPH, professor in the school of medicine and an associate director of the RCMI program at the University of Puerto Rico Medical Sciences Campus in San Juan.
"I contacted Paul [Harris], and then we started to develop this collaboration," Conde says.
As part of the collaboration, the University of Puerto Rico is helping to translate the system into a Spanish version, says Brenda Nieves, BA, coordinator of research computing system at the Center for Information Architecture and Research at the University of Puerto Rico in San Juan.
"We're in the process of completing the translation," Nieves says. "Most of the program is translated and is already being used in Spanish."
Here are some of REDCap's features:
• Customization. As quickly as investigators can define their data, REDCap can be customized for their particular studies, Harris says.
"We can flip [the information] into a web-based database within five to 10 minutes per iteration," Harris adds. "We've had these take less than a day to go into production mode."
Most of the development time is spent as the research team tries to figure out what will be measured, Harris says.
"We support 140 studies right now in the REDCap project, and we take them as fast as we can get them," Harris says.
• Training research staff. Research teams receive one-hour training sessions on how to use REDCap.
The training sessions focus on using the tool, validating numbers going into the database, security, and audit trails, Harris says.
"We teach them about HIPAA [Health Insurance Portability & Accountability Act] security rules," Harris says. "We guide them through the process and talk about case report forms and data export in a number of statistical packages."
There's a brief demonstration and an opportunity to ask questions, he adds.
The training includes sending investigators home with an Excel template model and demonstration that instructs them on how to develop an operational database, Harris says.
"If they do the work on it we'll turn it into a real working model and send it back to them with a data dictionary," Harris says.
"We give all of the research team access to that database," Harris says. "We say, 'Work on this as a team.'"
• Makes informatics staff time more efficient. "REDCap allows us with our limited resources to help more people," Howard says. "There's not only the time it takes to develop something, but also the time we're waiting."
The research institution has only a few programmers to create custom applications for all research teams, Howard explains.
"We have six projects in production and three more are in development," she adds.
Also, since programmers now spend less time creating new databases for each research project, they have more time to devote to those projects in which REDCap cannot be used, Howard notes.
• It's web-based and easy to use once the hard decisions are made. REDCap is a web-based application that is flexible for all types of research and provides an intuitive interface for users to enter data.
"It's extremely easy to use," Welburn says. "The hardest part is agreeing on fields the investigator would like and the organization of those fields into easy form for optimizing the process."
"When we ask the research team to tell us about interesting parts of their process to create the database, everyone says it helps the science," Harris says. "They say it gives a good working model and everyone can participate."
• Works with a variety of statistical packages. "In the old system, we've usually custom-created an extract so they can pull data out," she explains. "This usually happens at the end of the development cycle."
This had to be created for each research project, but now with REDCap it has been created once and is automatic, Howard says.
"I don't have to create anything specific — it's automatically there," she says.
"The data can be extracted directly into statistical packages," Howard says.
This feature provides a good benefit to investigators, Conde says.
"They don't have to spend much time writing code and importing the data," Conde says. "You run it and import it into your statistical software."
REDCap supports exports to SPSS, SAS, R, Stata, and Excel, Harris says.
"This is one of our best features, as literally 30 seconds after finishing data collection, you can get a well-formed statistical file with data labels, coded variables, etc., already in place to start the analysis," Harris explains.
For investigators, REDCap is a more natural approach to database management, Welburn says.
"You can select which fields you want to have in your statistical database, selecting just what you want and not everything," Welburn says.
"This is important for investigators," he adds. "They can say, 'This is the database I got three months ago.'"
Electronic data capture (EDC) systems are supposed to make data collection and analysis easier and more accurate for clinical research (CR) teams. But often CR coordinators and investigators find the systems to be counterintuitive, inflexible, and difficult to navigate.Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.