Digital transformation and smarter data management — using artificial intelligence (AI) and predictive analytics accompanied by workflow redesign, standardization and redesigned products — can work together to fundamentally transform and improve healthcare delivery.
In a June Becker's Hospital Review webinar sponsored by Abbott, three healthcare leaders discussed staffing challenges, digital transformation and how AI and machine learning will transform the patient journey.
Panelists were:
- Jill Kalman, MD, chief medical officer, senior vice president and deputy physician-in-chief, New Hyde Park, N.Y.-based Northwell Health
- Joon Lee, MD, executive vice president, University of Pittsburgh Medical Center (UPMC) and president, UPMC Physician Services
- Richard Rapoza, PhD, DVP of research and development, Abbott's Vascular Business
Five key learnings were:
1. Health systems are pursuing short- and longer-term strategies to mitigate staffing challenges. Before COVID, Northwell's 12-month rolling average for open positions was about 9 percent. As COVID continued, it rose to 14 percent. To deal with staffing shortages, both Northwell and UPMC have used external agencies and created internal flex staff. They have focused on increasing retention by raising compensation and offering loan forgiveness and are working to build a pipeline of workers by partnering with local colleges. Northwell is looking upstream and considering building its own schools and certificate programs for high-demand areas. UPMC is also redesigning care models. However, despite these initiatives, "The statistics are sobering," Dr. Lee said. "At one point, we had 2,500 vacant nursing positions at UPMC, and the state of Pennsylvania only produces 5,000 nursing graduates per year."
2. Digital transformation must be aligned with workforce strategies. Previously, patients accessed the health system by going to the ER or a provider's office, or calling on the phone. Now, digital transformation is enabling an "omni-channel front door," Dr. Lee said. Technology makes possible online scheduling of appointments, gives patients access to medical information and can improve staff efficiency. However, just adopting technology is not enough. For example, moving a visit from in-person to virtual is a positive step but is not transformational. "The real revolution is figuring out how to integrate virtual visits and digital tools to deliver care in a different manner," Dr. Lee said.
3 Predictive modeling can improve workforce efficiency and the care clinicians provide. Dr. Rapoza commented that electronic health records provide a great foundation by aggregating large amounts of data. But, "What do you do with it?" he asked. Predictive modeling can turn data into useful insights. Data from the EHR and sources such as imaging can be combined to "give you really nice predictive models on what to do next," Dr. Rapoza said.
For example, predictive modeling can see patterns for when patient care will be needed, by season, week, day and time. These predictions can optimize staffing. Also, "AI and machine learning can help us manage patients and help with medical decision-making," Dr. Kalman said. Examples include using AI to quantify and predict the readmission risk for every patient as well as using AI to identify pregnant women at high risk for maternal mortality. These uses of AI give clinicians greater insight to guide treatment. Dr. Lee sees the need for "intelligence alerts." "It's not enough to convert analog information to data and give it to me faster and more," he said. "It has to be smarter and more organized. AI and machine learning can do that."
4. Digitization transformation can produce more patient-centered care. To date, care has been delivered episodically in an assembly line approach. With more data and technologies, it is possible to care for patients continuously on a personalized patient journey.
5. Workflow and product design need to change to fit in this new model. Just engaging in digital transformation is not enough. Dr. Rapoza said workflow optimization is also key, as is the need to design products to fit in this new model.
Technologies that manage data smarter and democratize the ease of interpreting results will move the needle in care delivery. They will provide clinicians with better insights to make more data-driven decisions, leading to improved outcomes and more personalized patient journeys.
To learn more from Abbott, click here.
To register for upcoming webinars, click here.