NEW YORK — Results from genetic testing can often be difficult to find within an electronic health record (EHR), but an ongoing effort at the University of Pennsylvania aims to better integrate them with other patient information.
Genetic testing lab reports are typically returned as PDFs, which can be squirreled away in different parts of the chart and make them difficult to track down. Stephanie Asher, a genetic counselor at Penn Medicine, said she often had to scroll through hundreds of other test results in a patient's EHR to find the genetic test results she needed or had to poke around different parts of the chart because the results weren't well labeled.
Better integration of genetic information into EHRs is also thought to be a key step to realizing personalized medicine. The American College of Medical Genetics and Genomics, for example, said in a 2020 points-to-consider document that genetic testing results and clinician reports interpreting them should be easily accessible in EHRs, both by other clinicians and by patients.
The team behind the PennChart Genomics Initiative (PGI), which was launched in 2019, has enabled this by incorporating genetic test orders from four different providers into its EHR and by adding results into the EHR as discrete data points. According to a study the researchers recently published in Genetics in Medicine, these changes have reduced the amount of time spent both ordering genetic tests and managing their results.
"The whole reason we did it was because we think it's better for patient care," said senior author Katherine Nathanson, a medical geneticist at the Perelman School of Medicine at UPenn.
The first step toward integrating genetic testing results into the EHR was establishing a specific place in the chart for the results to go, a tab the Penn team labeled "precision medicine." That way, the results were not lost among the dozens of other lab test results but had their own place, making them easier to find. They additionally developed a dedicated document type for genetic testing results, so every time a result was scanned in as a genetic testing document, it would be placed under the new tab. Because they had previously developed standard labels for genetic testing reports, the PGI team was able to import more than 17,000 legacy reports into the tab.
The second stage aimed to integrate the results more fully into the EHR, as discrete data that can be used in clinical decision making. This step relied on building out the genomics module within the EHR system from Epic Systems, a commonly used electronic medical records system. That module, Nathanson noted, is akin to an Excel spreadsheet but with a Health Level 7 interface, a set of standards that allows for medical information sharing. While it is a place to put data, it needed customization and specific code in order to work the way they wanted, she added.
The team began by working with one genetic testing company, Ambry Genetics, before expanding to include three others — Invitae, OneOme, and GeneDx.
Now, clinicians can order genetic tests from within PennChart and receive the result in that system. Results come back initially as a PDF and then as discrete data that populate the genomics module. Pathogenic, likely pathogenic, or medically actionable variants are further labeled as "Genomic Indicators" and feature on the front page of the patient's chart.
All this, Nathanson said, required collaboration across disciplines, which could sometimes be tricky. "I don't always talk the same language that the IS [Information Systems] folks talk. You have to work in a very multidisciplinary way to be able to do it," she said, adding that the team also had supportive institutional leadership.
Their efforts were further helped by developing, at the outset of the effort, detailed standard operating procedures for documenting testing results.
After implementing the PGI's work, the team asked genetic counselors at the institution to estimate over the course of a month how long managing genetic testing took within the PennChart ecosystem versus outside of it. As they reported in their recent paper, ordering within the EHR took two minutes, as compared to eight minutes otherwise. The time spend managing genetic results also dropped.
While six minutes saved in ordering time does not sound like much, Asher noted that it adds up. "If you're seeing eight patients in a day, that's 45 minutes," she said.
Nathanson added that the time savings was an unanticipated benefit. "We went in for all these other reasons," she said. "It really was beneficial in that it ended up allowing our providers to be able to do things a lot more easily."
For instance, Asher said, the process is much more seamless and she no longer has to go back and forth to different testing lab portals to figure out which genes they have on the various panels they offer because they are all within the chart. Now she can submit the orders while seeing patients. "I don't have to carve out special time to be able to enter orders," she said.
Also, finding the reports now only takes two clicks. "It's so much easier," Asher said.
The ultimate hope is that making genetic testing information more easily accessible within EHRs will improve patient care. According to Nathanson, the new system will also enable researchers to see which clinicians are ordering genetic testing and ensure that their patients receive the proper testing support.
Toward that end, she and her colleagues recently received a grant from the National Human Genome Research Institute to conduct a large, randomized trial of 18,000 patients. Some of these will be referred to geneticists to order genetic testing while others will have their tests ordered directly by specialty physicians through one-click ordering in the EHR. According to Nathanson, the project aims to look at whether the testing process can be streamlined.
"Should — and I'm not sure I know the answer — I, as a geneticist, just be seeing people who are positive because so many people need testing?" Nathanson asked.
At the same time, further tweaks are also being made to the EHR system. With the onset of the COVID-19 pandemic and the push to limit in-person visits, the PGI team added an option in the interface to ask the genetic testing lab to send saliva specimen collection kits directly to patients at home.
Another feature both Nathanson and Asher said would be helpful to have in the futures is an automated way to send patient reminders. That way, patients get the screening they need and don't fall through the cracks. For instance, patients with Lynch syndrome could receive a reminder, or even a link, to make an appointment for their annual colonoscopy.
"This is why we're doing the genetic testing to begin with," Asher said. "So let's see it through all the way to the end and not leave it incomplete."
Additionally, as the PGI has initially focused on integrating germline testing, they are expanding their support of somatic and prenatal genetic testing.
The PGI is also developing a website with their SOPs, tip sheets, and instructional videos for clinicians so other institutions can learn from its efforts, though Nathanson noted the site is not yet fully live.