Physician-led data analytics teams becoming more important in healthcare
Data input, extraction and analysis by fellow physicians can streamline processes and deliver cost value, says Dr. Amy Ho.
Dr. Amy Ho, senior vice president and chief of clinical informatics at Integrative Emergency Services, speaks at HIMSS22 in Orlando on Monday.
Photo: Jeff Lagasse/Healthcare Finance News
More than ever before, data drives the business of healthcare. The COVID-19 pandemic in particular brought statistics, data and analytics to the forefront. The healthcare industry saw real scrutiny on the sources of its data, as well as on the collection methods and analysis around it. This intensified the pressure to have accurate, streamlined and scalable data analytics capabilities for health systems, clinics and physician groups.
With this in mind, those in the business of healthcare need to consider how their data is collected and presented – and a data analytics team led by physicians may just be the best path forward.
Dr. Amy Ho, senior vice president and chief of clinical informatics at Integrative Emergency Services, in her session "How to Build a Homegrown, Physician-Led Analytics Department" at HIMSS22 in Orlando on Monday, made the case that the clinical workflow is only understandable, translatable and manageable by a clinician. Thus the importance of creating a physician-led data analytics team.
"Classically, informatics is heavily, heavily involved in EMR process and design," said Ho. "But not so much in what data is extracted and what it's used for.
"Information is translated at every stage. We translate that into an EMR, and hopefully it's engineered to make workflow easy, which flows into clinical measures, including reimbursement. The physician comes with all the knowledge to understand this transformation already. They just need the right skill set."
One of the reasons physician involvement is so important to the process is the insight they bring to deceivingly complex questions. For example, they may be asked how many patients a doctor has seen per hour. That may seem like a simple premise, but it brings up the question of what an hour truly means: Does it mean a physical hour? A pickup hour? A schedule hour? Only a clinician, said Ho, will truly know to even ask those questions, let alone know what the answers might be.
"It's hard to get physicians to put data into the EHR in a structured way, but with proper incentives and workflows, they will," said Ho. That means making sure data entry is faster, and aligning things in such a way that clinicians see the additional value, rather than feeling like it's just additional work.
To get there, however, physicians need a few extra skills in their arsenal, including understanding how electronic health records are built, and how and where the records are stored, so they can extract the data when they need to. They also need skills in analytics and data visualization to have the ability to extract meaningful information from the data.
While physicians can be trained in EHR models and data visualizations, a physician-led analytics team doesn't necessarily have to be grown internally. It can be hired out, especially when it comes to certain specific roles, such as data scientist, API analyst, and someone who keeps the physician on track, such as a manager or administrator.
"As a physician, because you have your own patients, you know what's happening with your patients, so it's helpful to be able to go into the data and extract things from it," said Ho.
One of the trickier aspects of getting a physician-run data team off the ground is making the value proposition. Part of what gives such a program value, in Ho's view, is performance management.
"Physicians don't want to be told they're doing poorly by anyone who's not a physician," she said. "They don't like it anyway, but it helps if it's coming from another physician. Also, because it's a physician that's doing it, you can look for systemic improvements in workflow or clinical processes, and implement those quickly."
Another benefit is in cost review, which can be mastered by data. Many health systems and practices have quality groups run by various levels of clinical nurses, but they often result in very time-insensitive chart reviews. That can be streamlined with smart data extraction, which translates into a huge cost value in terms of gaining back labor resources.
The goal in all of this, said Ho, is harnessing technology to improve processes specifically and health systems overall.
HIMSS22 Coverage
An inside look at the innovation, education, technology, networking and key events at the HIMSS22 Global Conference & Exhibition in Orlando.
Twitter: @JELagasse
Email the writer: jeff.lagasse@himssmedia.com
