The other nine technology trends featured in this issue involve topics that most CIOs are wrestling with in 2012. But the Healthcare Informatics editorial team thought it was important to highlight one trend with a slightly longer timeline: the convergence of emerging genetic medicine and electronic health records. The pioneering clinicians and healthcare informaticists we spoke to stress that CIOs should start paying attention now.
“Many of us believe that genomic information will inevitably transform healthcare beyond recognition,” says Christopher Chute, M.D., a Mayo Clinic bioinformatics researcher focusing on clinical and genomic data sources, management, standardization, and interpretation. “It will be a bigger breakthrough than antibiotics—not immediately, but in the next decade or two.”
Yet if the commonplace use of a patient’s personalized genetic risk information to make clinical care decisions is still a decade away, pilot projects and research studies are not. Dr. Chute is principal investigator on a study called the Electronic Medical Records and Genomics (eMERGE) Network, which brings together researchers from leading medical research institutions across the country. One focus is whether EHRs can serve as resources for complex genomic analysis of disease susceptibility and therapeutic outcomes. The next generation of EHRs, Chute says, will require at least three things: a facility to access genomic data about patients; access to a curated nationwide or international database that assigns genomic-clinical implications; and clinical decision support tools with a dynamic capacity to make use of that information.
“Let’s say a physician is about to order an antidepressant. A simple red, yellow, green indicator would strongly discourage that drug because the patient would metabolize it poorly or not optimally,” Chute explains. The doctor could then avoid it or switch drugs. “Mayo has already done this on a sample cohort of patients,” he says.
Elsewhere, the Coriell Institute for Medical Research, Camden, N.J., and the Ohio State University (OSU) Medical Center in Columbus are collaborating on a study involving OSU cardiologists and primary care physicians and 1,800 of their patients who have been diagnosed with congestive heart failure or hypertension. The patients’ genomic information is being attached to their EHRs. The study seeks to understand how likely doctors are to use the information when it is made available to them.
The project has run up against some constraints with the current generation of EHRs, says Scott Megill, Coriell’s CIO. “They currently have no fields ready to be populated by genetic data, which isn’t surprising,” he says. Coriell and OSU put the genetic risk reports in PDF files attached to the patients’ records, much like imaging files are attached.
Designing common ways for EHRs to access genetic data is a huge challenge, Megill says. It’s more complicated than just adding a few blanks fields to a screen. “There are many different types of pharmocogenomic results and ways that cardiac risks could be impacted by genetics,” he adds. “I think standards bodies have to get together and create standards that others will adopt.”
The H. Lee Moffitt Cancer Center in Tampa, Fla., is one of 18 hospitals taking part in the Total Cancer Care research study that follows cancer patients longitudinally and includes genetic data. “As we look at additional diagnostic and treatment therapies, researchers are starting to understand the nuances of different molecular profiles and they can target that for diagnosis and treatment,” explains Mark Hulse, Moffitt’s CIO.
Moffitt’s Cerner-based oncology-specific EHR called PowerChart Oncology does not yet have a way to alert the clinician that genetic data is available. “That is one of the next steps: to get the clinician information based on genetic data about whether a patient would benefit from front-line treatment or perhaps should go directly into a specific clinical trial,” Hulse says. “We need to develop clinical pathways to guide them to information about how their patient may benefit depending on how genes are expressed. Today this is primarily at the research phase, but clinical deployment is coming fast.”
Randall Lambrecht, Ph.D., vice president of research and academic relations at 15-hospital Aurora Health Care in Wisconsin, says that for CIOs, personalized medicine will necessitate increased networking across health systems. “It wouldn’t make sense for most to create their own biorepositories or genetic data repositories because they don’t have the volume of patients,” he says. “It makes more sense to join networks.”
Aurora, which has three million patients, has created a large biorepository with 100,000 samples. It uses bar codes to tie the samples to patients’ EHR data in a de-identified way. So far, the biorepository is being used strictly for research and is not used for treatment. “We want to work with universities and other academic medical centers on sharing this data,” Lambrecht says. “Researchers could view our cardiac population, for instance, and study how they might respond to different drugs.”
Moffett’s Hulse’s advice to other CIOs? Begin to educate yourself. “Personalized medicine is coming pretty quickly,” he says. For instance, there may be a dozen tests today about whether a certain drug for breast cancer would work based on a woman’s genetic profile. “But that number is going to scale massively as researchers recognize more and more subtle genetic variations,” he adds. “As leaders in the health IT space, we can influence vendors on how systems are developed to accommodate this huge change in diagnosis and treatment.”