Precision medicine requires more data, sophisticated analytics
Healthcare’s one-size-fits-all approach to treating patients must be replaced with a personalized approach to medicine that focuses on individuals and the unique needs of each family member, says John Halamka, MD, chief information officer at Boston’s Beth Israel Deaconess Medical Center.
According to Halamka, co-author of a new book—Realizing the Promise of Precision Medicine: The Role of Patient Data, Mobile Technology, and Consumer Engagement published by Elsevier—the “goal of the precision medicine movement is to give clinicians and patients access to the kinds of information needed to create individually tailored programs to treat a variety of diseases and to ward off those that are preventable.”
However, he and co-author Paul Cerrato make the case that achieving these goals “will require the collection of far more data than clinicians now collect when they evaluate patients” as well as “more sophisticated analytic tools to glean meaningful insights from the data collected.”
Despite such barriers and limitations, healthcare can be re-engineered to overcome these challenges to create a system that tailors treatment to the individual, contends Halamka in his book, which examines how big data, electronic health records, genomics and health information exchange are helping to individualize medical care.
“I’m 55 years old, and at this age, you end up being the care traffic controller or navigator for all of your family members,” observes Halamka. “From that position, you really get a sense of where the U.S. healthcare system works and doesn’t work.”
In the book, Halamka recounts how his wife Kathy was diagnosed in 2011 with Stage III breast cancer and was able to leverage digital resources such as the Shared Health Research Information Network (SHRINE) and Informatics for Integrating Biology and the Bedside (i2b2) to help personalize her treatment. SHRINE is a network of computer systems affiliated with Harvard Medical School which gives users access to the EHRs of all of its member hospitals, and i2b2 is an open source platform that provides web-based access to a hospital’s EHRs—both of which are valuable informatics tools.
Specifically, Kathy Halamka’s providers queried i2b2 to see how many patients seen in all the Harvard-affiliated hospitals fit her profile. The tool found more than 17,000 and listed the medications they received, their average white blood cell counts, and their prognoses—among other information. Based on the data, her oncology team was successful in personalizing her treatment and melting away the tumor.
“Wouldn’t it be wonderful if in the middle of an electronic health record—using Epic, Cerner or athenahealth—you could push a button and going out to a cloud-hosted service provider you’re told, ‘You know, 10,000 patients like this one have these diseases, this treatment, and these concerns,’ ” Halamka asks. “That would be a whole lot better than what we have today.”
While he acknowledges that few patients currently have access to the digital resources his wife had, Halamka says that is gradually changing and becoming a reality. He also notes that healthcare stakeholders such as Health Level Seven (HL7) and several HIT vendors are working on the standards for incorporating genomic data into EHRs so biomarkers and genome information can be exchanged by providers at the point of care.
“The future of the next couple of years—we believe—will belong to the private sector creating innovative applications, modules and services which layer on top of existing EHRs,” Halamka concludes. “The next breakthrough will come from third party, cloud-hosted services that are offering this real-time decision support.”