An Institute of Medicine collaborative called Digitize (Displaying and Integrating Genetic Information Through the EHR) is working on a clinical decision support (CDS) rule to alert providers treating patients who have an allele that puts them at a high risk for a life-threatening reaction if prescribed the HIV treatment drug abacavir.
This is being done at a time when the laboratories and providers are on different systems from different vendors, and on either side, the infrastructure may have little capability specific to genetics. “Given that is the case, Digitize is starting to layer in some genetic-based CDS using existing infrastructure for the most critical needs,” reports Sandy Aronson, a Digitize co-chair and executive director of IT for Partners HealthCare Personalized Medicine in Boston. “All these stakeholders—EHR companies, laboratory information system vendors, providers, government agencies, and standards organizations—have to work collectively to solve this problem.”
But are these efforts to move genetic data between labs and EHRs and create the CDS rules about pharmacogenomics or family history tools confined to the rarified air of the largest academic medical centers and still decades away from relevance for community hospitals? Aronson says no. This is a tech trend that has relevance today as well as great potential for the future.
“We have explicitly said that we do not want to build recommendations that are only going to be implemented at a limited number of academic medical centers,” he says. “Our goal is to build things that are going to be helpful deep into the community.” He recommends that CIOs at community hospitals take a look at the Digitize implementation guide because it is designed to be accessible to them, and the problems are real today. “My understanding is that abacavir is being used in combination therapies, so you could have clinicians prescribing it who don’t realize that they are prescribing it and need to get the appropriate pharmocogenomic testing done.” The challenges for an academic medical center and a community hospital are not that different, he stresses. “It is a matter of coordinating with laboratories, but what needs to happen in terms of IT stuff is relatively simple.”
This type of genetic testing is indeed becoming more commonplace. For instance, patients who go to UPMC Presbyterian Hospital in Pittsburgh and receive a stent to treat clogged arteries, are now being screened to determine if they have a gene variant that makes them less likely to respond to a blood-thinning medication commonly prescribed after the procedure.
In another example, Penn Medicine in Philadelphia has created a new Center for Personalized Diagnostics, a clinical laboratory that does somatic tumor testing of targeted genes that are clinically significant and actionable. Their PennOmics data warehouse is proving of critical value to researchers studying clinical effectiveness by doing retrospective analytics on aggregated data. For example, a researcher studying a population of patients with the BRCA1 gene mutation can study treatment regimes and outcomes. “That all feeds back into the decisions you make in the clinical area about what is the best drug or treatment,” says Brian Wells, associate vice president of health technology and academic computing. “It all ties together. That is translational medicine at its heart.”
Rachel Liao, Ph.D., the coordinator for the Clinical Working Group of the Global Alliance for Genomics and Health, says providers and health IT leaders will see impacts at all levels in the next five to 10 years. First, the amount of genetic testing that might be done routinely in the clinic will increase dramatically. “Within a few years it will become standard practice to do genetic testing for many diseases, whether they have a hereditary basis or not,” she says. “There will be enough therapeutic options available based on genetics and enough understood about the genetic basis for even common diseases that it will be within the interest of even community medical centers to do some level of genetic testing, and all the labs are going to develop panels to test for whatever makes the most sense for the patient populations.”
How the results of genetic testing are stored and shared remains uncertain. The Global Alliance has an eHealth Task team focusing on how family history is collected in an EHR. The team did an inventory of tools available and is developing a standard of best practices for EHR vendors as they think about incorporating a tool, Liao says. (Studies have shown that the way family history is currently collected by providers leads to data that is not actionable. It is unstructured data and therefore unavailable to clinical decision support tools.) “The goal would be that EHR vendors such as Epic would begin to incorporate family history and pedigree data into their tools and consider these best practices we are setting up,” she says.
The Global Alliance also has a Phenotype Ontologies project to bring together existing international efforts to develop and promote standard language and tools for recording patient clinical phenotypes and exploiting phenotype data for diagnostics and translational research.
Data at the Point of Care
Speaking at HL7’s most recent Genomics Policy Conference, Joel Diamond, M.D., chief medical officer, genomics and precision medicine, for Allscripts and a physician at UPMC in Pittsburgh, told a story from his own practice. He said he knew a patient of his had seen an oncologist and had genetic testing. But as he looked through her records, he couldn’t find any sign of the results. “They are probably on a piece of paper in a file at the oncologist’s office,” he said.
“What we need to do is define a care plan and how it is distributed to everybody, and how we can take that genomic information and make it part of the record for diagnosis and ongoing care,” Diamond said. “Having that data at the point of care becomes important. Having the ability to support de-identified data so people can do research on regional and national levels is also somewhere we need to move quickly.”
Building the Airplane While Flying It
The teams working on genomics and EHRs have to decide whether to build things that can be shoehorned in to work with today’s infrastructure or for a future when many of today’s health IT systems will be obsolete. “One of the things we struggle with in Digitize is how much time to spend specifying future support and how much to spend working with existing support,” Aronson says. "We want to apply new and different forms of expertise to help existing standards bodies improve standards, ontologies, and codes so we can robustly transfer data to enable this kind of support."
"There are things that make dealing with genetic data difficult, but there are also things about it that make if far simpler than anything else we deal with in healthcare,” he says. “It’s inherently structured and new. We have the benefit of applying modern-day IT thinking from the beginning and, in my view, genetics can be used as a model for informing the rest of healthcare surrounding what’s possible.”