Bridging the gap between clinicians and computer scientists
Ask Michael Temple, M.D., M.S., what he likes most about his work as a senior data scientist at Fuse, Cardinal Health’s innovation engine, and he is likely to respond, “I just love data. I have a natural curiosity, and I ask ‘why?’ a lot: ‘Why is that?’ ‘Why does that person have this characteristic?’ Data provides lots of answers about people, health risks and well-being.”
Board certified in both pediatrics and clinical informatics, Temple describes himself as a unicorn. “There aren’t many data scientists who have a background like mine,” he said. “I can use my expertise in medicine and my knowledge of machine learning and artificial intelligence (AI) to bridge a gap: My job, in part, is to understand what clinicians want and then translate that into language the computer experts can understand. Together, we develop the technology solutions that can help achieve better clinical outcomes.”
A chance encounter prompts a career pivot
Temple didn’t start out with an interest in technology. “I didn’t write my first computer code until I was 38,” he said. Instead, he pursued his dream of practicing medicine, following a traditional path post-high school: undergraduate school, medical school and residency, and establishing a pediatrics practice that he managed for more than a decade.
Then, in 2009, the Health Information Technology for Economic and Clinical Health Act (the HITECH Act) prompted a significant career pivot. “The bill provided billions of dollars in incentives to hospital systems and clinics to adopt electronic health records (EHRs) instead of paper records,” Temple explained. “Like other health systems across the country, the system I worked for wanted to get on board; they chose my practice as the pilot site for the transition.”
At the time, Temple was doing some hospital administrative work in addition to managing his practice. On the way to a meeting, he ran into an executive who asked him to take the lead in transitioning to electronic records. “It was literally a 30-second meeting,” he said. “I had no idea what I was getting into.”
After getting some training to use the EHR tool the health system had selected, he realized that it wasn’t going to work. “We had seven medical specialties within our practice; the EHR we had was essentially a blank slate, with no templates to customize for different specialties,” Temple said. “So I got additional training to learn how to write computer programming, so that I could customize the tool myself.” Then, he worked closely with physicians throughout the practice to design specialty-specific custom content.
For the next three years, Temple continued his pediatrics practice parttime, devoting the rest of his time to leading the systemwide EHR implementation. “By the end of those three years, I was very interested in health care technology,” he said.
As it happened, at the end of those three years, Temple’s wife was offered a work transfer to Nashville, Tenn. Thinking it would be temporary, they decided to make the move, and “rather than start up a new pediatrics practice in Nashville, only to close it a couple of years later, I decided instead to go back to school.”
He chose biomedical informatics – which uses big data and scientific research to provide clinicians with insights to help them improve health outcomes – as his field of study. He was hired into a post-doctoral fellowship at Nashville’s Vanderbilt University, teaching and researching as he worked toward a master’s degree.
Meanwhile, his wife’s job kept them in Nashville for several more years, so, after earning his degree, he took a job as a data scientist with HCA Healthcare, a large health system headquartered there. He was there just over a year, helping develop technologies designed to improve patient care, then was recruited back to Vanderbilt, where he continued to work with the biomedical informatics team, even after he and his wife moved back to Ohio in 2019.
Joining the Cardinal Health team
Temple joined Fuse nearly two years ago. He and his data scientist colleagues at Fuse collaborate closely with the clinical team at Outcomes™, a digital ecosystem that delivers digital solutions to a large network of pharmacists to help them improve patients’ medication adherence, drive better health outcomes and lower the cost of care.
“I’ve gained so much appreciation for what pharmacists do,” he said. “They spend a great deal of time counseling and educating their patients to help them get the best health outcomes.”
Together, he and his colleagues use data to build predictive computer models to help identify patient populations who are most likely not to take their medications as prescribed, or who are at risk of medication side effects or drug interactions. The clinical team in turn provides this information to pharmacists so that they can give their at-risk patients additional coaching, education and other high-touch supports.
Temple was instrumental in the development of the Outcomes Clarity™ predictive modeling tool, launched last year, which uses machine learning to provide a detailed picture of factors that influence behavior and help pharmacists identify patients likely to become nonadherent to their medication and are in-need of actionable interventions.
Currently, he’s working on a project that will help identify patients who are most at risk for elevated health care costs. The tool, currently being piloted with a small pool of pharmacists, will evaluate several risk components to help pharmacists prioritize patients in need of targeted care or intervention to avoid more costly health care expenses in the future.
About his former pediatrics practice, Temple said, “I really miss working with children. But it’s an interesting tradeoff. I now have the opportunity, through the development of technology solutions, to help so many more people at once to achieve better health outcomes. Providing great insights that help improve health care – that’s a powerful way to make an impact.”