The new age of precision medicine and genetic testing demands a steep learning curve - for patients and physicians alike. Patients need to be aware of what tests they might benefit from, what the results might mean and what, if anything, can be done in light of them.
Physicians, many of whom learned little about genomics in their medical training, need to know what genetic tests to order and when - and, just as crucially, often how and where.
Technology may still have some ground to gain when it comes to truly enabling precision medicine techniques. But at NorthShore University HealthSystem careful calibration of electronic health records, provider workflows and clinical decision support tools have led to some useful strategies for helping physicians better prescribe genetic tests for their patient.
"One of the key challenges is that genetic and genomic information is rapidly coming at us from all different angles," said Peter Hulick, MD, director of the Center for Personalized Medicine at NorthShore University HealthSystem.
"Even today, traditional medical education cannot address it in enough detail to make every single provider an expert," he added. "Fortunately, we're also in an era where we're able to leverage tools like the EHR and really get doctors up to snuff for the genomic era of medicine - we can put the right amount of information in front of the busy clinician, at the right time, so that they have what they need to incorporate this information right at their fingertips."
Leadership at the Chicago-area health system recognized this is an important direction and allocated resources to modernize the electronic health record, said Hulick.
The aim, he explained, was to tweak the system to better enable things that might sound easy straightforward but in practice are complicated. Hulick pointed for example to the ability to track a genetic result in the EHR so clinicians know what the patient has been tested for and what the result is as a discrete variable.
When that data is trackable, "that means you have what we call in genomics a variant repository," he said. "Once you have that, you can start to link it to other information that's known about the patient: family history status, etc. And you can start to build an infrastructure around it and use some of the tools for clinical decision support that are used in other areas: drug/drug interactions, reminders for flu vaccinations, and you can start to build on those decision support tools but apply them to genomics."
Toward that end, NorthShore recently incorporated two new genetic testing programs that primary care docs could use during their annual visits with their patients.
One, called MedClueRx, is pharmacogenomics testing kit that can be mailed to patients in advance of the appointment to evaluate and identify genes they may have that impact how they metabolize certain drugs. The data from that test is integrated into NorthShore's Epic EHR and can help inform the patients' care plans.
Its Genetic & Wellness Assessment, meanwhile, takes stock of health and family history as a means to better indicate need (or not) for either preventative or specialty-based genetic testing. The results of the form, which is completed by patients, are integrated into the EHR.
So far, NorthShore has seen physicians appreciating the chance to learn more about pharmacogenomic testing and how it can be applied to clinical decision-making. The new kits are helping the PCP order up more precise - and relevant - tests, while also helping patients do better with medication adherence and treatment plans.
"We're not trying to make all of our primary care physicians into genomic experts," said Hulick. "That is a difficult strategy that isn't really scalable. But we're giving them enough tools to help them feel comfortable, recognizing when further investigation, such as a referral to one of our personalized medicine clinics."
NorthShore has about 10 clinics based around expertise in different areas, he said, "whether it be inherited cancer, cardiomyopathies, aortopathies, certain arrhythmias - different areas where there are genetic components and a champion in our group may have expertise but a primary care physician may not realize that or understand how to get that patient to the clinics."
Before their annual physical patients fill out some basic yes/no questions to help the PCP get a better understanding of their genetic background and family history.
"When the primary care physician reviews this information, they get what's called a best practice alert - right where they also see things like a flu vaccine reminder, a colonoscopy reminder - to let them know that this patient answered yes to these questions and that they may benefit from a referral or potentially a targeted genetic testing if it's indicated," said Hulick.
"And because most PCPs aren't going to know what tests to order, we've actually gone a bit further," he added. "We're starting to partner with labs who offer testing to make it more seamless in terms of teeing up the specific order based on the questions that are answered. So the primary care doctor just has to basically click on the suggestion."
The key is to do all in the normal workflow of the physician, of course.
"That's been one of our main goals,” Hulick said, "to bring genetics and genomics into routine care flow."
EHRs aren't typically designed to do this sort of thing out of the box, of course. "Fortunately we have a very innovative IT department and they've helped us maximize what Epic can do and customize it to get genomic data as a discrete variable, build a questionnaire that's deployed electronically, get that FHIR best practice alert that has the information embedded, so the clinician can see more recommendations behind it," said Hulick. "And then actually take action on it. You make this all seamless and minimize the clicks in Epic so it doesn't bog down the clicks for the primary care physician."
Close engagement with the docs has been key to making this work. "We had many educational meetings, brainstorming meetings, planning sessions with our physician leaders in the different areas of content. I myself am a geneticist, and there was administrative support from the medical group in terms of how do we practically deploy this,” Hulick said. "And a huge commitment from our health IT group to how we actually build it."
Amy Lemke, PhD, senior research scientist at NorthShore's Center for Personalized Medicine, said the work began with an exploratory implementation assessment to determine what physicians view as common challenges and perceived benefits of pharmacogenomics.
"That allowed us to use that feedback early on to modify and make some changes prior to full-scale rollout," Lemke said. "The same type of assessment, we're hoping to conduct with the genomic wellness assessment to get some of the provider and patient feedback so we can make the modifications and meet the needs of both."
Lemke added that the work also contributes to the evidence base.
"We'll look at these findings and see what some patients do - and that's going to feed back into physicians' experiences and they'll see the relevance to their practice and care," she said.
NorthShore is still at the very beginning of the project - it launched just in the past six months - but the hard work is already paying off.
"We've had more than 20,000 people complete this hereditary assessment," Hulick said. "We are identifying patients who wouldn't have necessarily been identified otherwise as having an indication, for example, for BRCA1 or BRCA2 testing. Some primary care physicians were looking for more structure, and so here we're making it more systematic.
That has proven advantageous at the physician- and enterprise-level particularly when it comes to identifying at-risk patients and understanding why some of they did not complete the appropriate screening or make it to the clinical referral.
As a large academic medical center, NorthShore is clearly better positioned than many to try a project like this. But Hulick hopes health systems of all sizes can learn from some of their experiences making smarter and more targeted use of genetic testing.
"If you try to design the perfect solution, you will never try anything," he said. "We took initial steps knowing it wasn't going to capture every patient we needed to capture. But if we keep moving the needle in increments we can make real headway. Real world experience can give you a lot more understanding of where are the potential breakdowns of the process and how do you fix them?"