CHICAGO (GenomeWeb) – With an eye toward providing a better communication tool of professional and consumer-level educational material for genomic tests, bioinformaticians at two major US institutions are developing a system to create and then deliver relevant documents to the point of care.
A web application called DocUBuild publicly emerged last year from within the National Human Genome Research Institute's Electronic Medical Records and Genomics (eMERGE) network to serve as a content management system for genomic knowledge. It is a template framework for providing structure and context to genetic knowledge.
DocUBuild takes advantage of a Health Level Seven International specification for "infobuttons," designed to speed "context-aware knowledge retrieval," according to Luke Rasmussen, a clinical informatician at Northwestern University's Feinberg School of Medicine.
Rasmussen created DocUBuild in collaboration with Casey Overby Taylor, an assistant professor of health sciences informatics at Johns Hopkins University. He presented the technology at the American Medical Informatics Association's 2018 Informatics Summit in San Francisco earlier this month.
"DocUBuild is built as a way to help supplement knowledge about what these test results mean and to help aid in the interpretation of results," Rasmussen explained in a subsequent interview with GenomeWeb.
"The goal is to help with easing the creation of documents that are compatible with infobuttons, but also creating resources that are relevant to clinicians without reinventing the wheel," Overby Taylor said.
"Given what we've seen within eMERGE, there was this need to be able to be more collaborative in the development, sharing, and reuse of these information and knowledge resources, providing a central spot where people could look at, review them, and then be able to provide that feedback. That was what we were going for," Rasmussen added.
DocUBuild is intended to be more collaborative than a typical content management system by offering a feedback mechanism for end users to suggest or directly make changes to improve the content and structure of an article, according to Rasmussen.
Organizations can create a central information repository to get at patient- and provider-facing material rapidly, allowing for quick responses to genetic test results, the developers explained.
"For example, if you configure it for only being seen with adult patients versus pediatric patients, then that context will be considered from the EHR and you should only see a resource that's relevant to the patient that you're seeing," Overby Taylor said. This context allows clinicians to link directly from the electronic health record to specific, relevant sections of documents, she added.
Rasmussen said he has been using infobutton technology since before it became an official standard in 2014. DocUBuild is an outgrowth of that.
"It's just very interesting and inspiring to know that we could avoid wading through millions of articles, or even tens of articles, to find the most relevant things that you're looking for," he said.
So far, Rasmussen and Overby Taylor have focused DocUBuild on use cases frequently brought up by participants in Phases II and III of eMERGE, many of which are related to pharmacogenomics.
"We're going to tell somebody they're predicted to be a poor metabolizer of clopidogrel," Rasmussen said, offering one real-world example he encountered at Northwestern.
"What does that mean? The genetic counselors who we were involved with are able to anticipate the types of questions that will come up as well as the types of questions that come up as they return results. We just kind of build and enhance the articles over time using that," he explained.
"When we're giving patients their genetic reports back as part of eMERGE, we're using the content from DocUBuild to provide them information within the context of that report," Rasmussen continued.
As eMERGE progresses, participants are seeing an increasing number of patients with rare genetic variants, making DocUBuild more useful, according to Rasmussen. "Having this centralized system is letting us add in content as we identify ... patients with variants that we need to report out," he said.
Still, the user base remains small. Not even Hopkins — which is not a direct eMERGE participant — is using DocUBuild in clinical practice yet. But the developers have high ambitions.
"We're hoping to have people think about the way that they're approaching their internal development and maintenance of content. Even though they may not necessarily be adopting DocUBuild today, they understand that there's another way to think about their content management and hopefully will consider that for the future," Rasmussen said.
"I hope that we can truly get to a point where we can centralize where knowledge lives and avoid duplicating, copying, and pasting across systems," Rasmussen continued. "I don't want to have to copy this block of text into my EMR because there is no other way for me to get that block of text in there."
Rasmussen admitted that it would take some time to get to that point, however.
"Really, the next thing is getting people to see the value of [DocUBuild] for their sites," Overby Taylor said. "If there is a way to start from something that already exists but also be able to make modifications and brand it the way that you need to for your institution, while also providing attribution to the folks who were the original developers, I think that the value of using a tool like this could be realized and might even go beyond the niche genomics use cases that we're considering now."