Becker's Hospital Review

31 CIO / HEALTH IT Vanderbilt CIO: When customizing an EHR, don't get 'hyper focused on the tech' By Jackie Drees C ustomizing an EHR can be a large en- deavor for a hospital, so it is vital that everyone using the technology is on- board and involved in the process. Neal Patel, MD, chief health information of- ficer at Vanderbilt University Medical Center, has helped lead various EHR customizations at the Nashville, Tenn.-based health system. During these types of projects, Dr. Patel con- siders the people and the process aspects to be equally important as the technology itself. In January, Dr. Patel assumed his role as vice president of health IT and chief health infor- mation officer at VUMC. He has worked at the health system since 1997, serving in var- ious leadership roles including chief medical informatics officer. Here, Dr. Patel discusses some of his EHR expe- riences at VUMC as well as his advice for health systems looking to roll out a customization. Editor's note: Responses have been lightly edit- ed for clarity and length. Question: What is one initiative you've taken to customize your EHR system? Dr. Neal Patel: We leveraged our self-deployed, real-time alerts platform and our informatics ca- pabilities with patient data stored in our health data repository to quickly identify patients with a new blood stream infection and a cardiac im- plant. Oen, implants are not easily tracked in the EHR, but we are able to use natural language processing to review all text documents, PDFs and discrete data in a patient's record to pull any evidence of presence of an implant. Q: How does this EHR customization benefit patients? NP: One example is a patient we had who had a blood culture turn positive, which put him at additional risk for complications since he had a cardiac pacemaker. His problem list in our commercial EHR, however, didn't discretely denote he had the device — but the care team was quickly alerted of both the infection and in- creased risk due to him having an implant when our HDR auto-reviewed historical data from various sources, and we were able to amend our care approach on the spot. is is a great illus- tration of marrying in-house innovation with the best third-party tools that are available. Q: What advice do you have for oth- er hospitals looking to implement an EHR customization? NP: I'd strongly encourage that other hospi- tals do not get too hyper focused on the tech- nology itself and overlook the importance of people and process. Customizing an EHR requires a great deal of engagement with the providers, nurses and other people who will actually use the technology — did they have input into the customization process? Will the customizations make their work lives bet- ter and help them take the best care of their patients? Are they ready for the changes that the customizations will bring from a process standpoint? Oen the people and process as- pects of the equation require even more ener- gy and focus than the technology. Q: How do you promote innovation among your team members? NP: We look for every opportunity avail- able to connect our IT team members with our users throughout the medical center, from shadowing a provider, to sitting beside a patient care specialist who's checking in a patient, to watching a nurse discharge a pa- tient from the hospital. ese interactions can create those 'ah ha' moments where an analyst sees an opportunity for improvement. Q: What has been one of your most memorable moments as CIO? NP: In a single morning, we launched a whole new EHR across our entire medical center, which impacted three hospitals, hundreds of clinics, more than 135 physical locations and nearly 18,000 employees. ose first 24 hours were incredible, seeing everyone come togeth- er: our technology team members fixing bugs and making tweaks, our support staff helping their colleagues get accustomed to the system and our hospital leaders triaging and prioritiz- ing issues. It was challenging and invigorating and humbling — I'll never forget it. n Penn Medicine algorithm flags patients most in need of end-of-life planning By Andrea Park A machine learning algorithm accurately predicted oncology patients' six-month mortality, thus identi- fying those who would benefit most from a proac- tive discussion about end-of-life goals and values, a study from Penn Medicine found. Researchers from the Philadelphia-based institution devel- oped the algorithm, which used EHR data such as gender, age, comorbidities and lab and ECG results to predict the six-month mortality risk of cancer patients at two hospitals in the University of Pennsylvania Health System. Of the patients flagged as "high-risk" by the algorithm, just over half died within six months, and nearly 65 percent had died within a year and a half, compared to less than 8 percent of "low-risk" patients. A panel of 15 oncologists surveyed agreed that at least 60 percent of high-risk pa- tients would have benefited from immediate advance care planning meetings with their physicians. "Patients oftentimes don't bring up their wishes and goals un- less they are prompted, and doctors may not have the time to do so in a busy clinic. Having an algorithm like this may make doctors in clinics stop and think, 'Is this the right time to talk about this patient's preferences?'" said Ravi Parikh, MD, the study's lead author and an instructor of medical ethics and health policy at the University of Pennsylvania. n

• Cover