Skip to content Skip to navigation

Optimizing Care Transitions: Where Do Predictive Analytics Tools Fit In?

October 13, 2012
by Mark Hagland
| Reprints
At the five-hospital Inova Health System in northern Virginia, Daniel Rosenthal, M.D. is helping to lead a groundbreaking effort to better predict the need for interventions and optimize care transitions

Daniel Rosenthal, M.D., M.P.H., M.Sc., is director of healthcare analytics at the Falls Church, Virginia-based Inova Health System ( Health System. Inova encompasses five hospitals with more than 1,700 beds and 16,000 employees, as well as specialized care and research units, including the Inova Heart and Vascular Institute, Inova Translational Medicine Institute on Genomics, Inova Neuroscience Institute, and Inova Children’s Hospital.

Rosenthal, an internal medicine physician by clinical background, was most recently senior advisor in health IT at the National Quality Forum before recently coming to Inova in February 2011. He is also co-founder and president of WardManager (, a team-based sign-out system that wirelessly enhances care management and patient safety in inpatient hospital wards and in medical specialist groups. Inova is a client of WardManager, for its sign-out solution. Rosenthal spoke recently with HCI Editor-in-Chief Mark Hagland regarding his activities at Inova and his perspectives on the challenges and opportunities involved in working to improve caregiver transitions. Below are excerpts from that interview.

Can you explain your various professional roles—executive, clinical, and entrepreneurial?

My routine is to spend four-and-a-half days a week at Inova. I see patients on Friday afternoons at George Washington University Medical Center in DC in Urgent Care. And then everything I’ve done with WardManager Signout is on nights and weekends.

Daniel Rosenthal, M.D., M.P.H., M.Sc.

How long have you been practicing as an internist?

I finished medical school in 2001, and before my residency training, I did a National Institutes of Health informatics fellowship, so I spent three years at Mass General, and got a master’s in informatics and a master’s in public health, before finishing my residency in 2007; so I’ve been practicing since 2007.

How do your various professional roles relate to each other?

Each one of my different roles might seem somewhat related, but in reality, they’re fairly divorced from each other. As director of healthcare analytics at Inova, I don’t have a particular tasking of building an EHR, or convening a particular group, for example. Our job here is to help others ask the right questions and provide the right answers. On the clinical side, “I want a dashboard of our length of stay, and stratify it by complexity of disease or co-morbidity,” for example. So it’s everything from fulfilling requests for simple extracts—“can you tell me which inpatients right now have CHF?” And most everything we do is retrospective, but could be retrospective to the past 24 hours—and ranging to an interactive dashboard to be able to zoom in and out on population health.

The hot topic right now is predictive analytics. We’ve got a million patients; can we determine the predictors of particular situations and outcomes? So it sort of runs the gamut of your traditional operational, bread-and-butter things to more lofty questions like, can you predict who will have a bad outcome? But the real challenge is, what are you going to do about it? And a lot of people in IT are attracted to the dangling, shiny object. But when it comes to building predictive tools, the question is this: is your organization positioned to build on and act on uncertainty?

How big is your team at Inova?

We have seven people on my consulting team. I’m the only physician; we have two folks with a quality background; and we have another person with a strategic planning and organizational management background, so I'm used to dealing with clinical scenarios. And on the technical side, we have a traditional database administrator, an “all-rounder” systems engineer and then a systems analyst with a traditional mix of database and IT systems background. We have an additional eight report writers and reporting coordinator.

With regard to our organization, right now, we are in the middle of our go-live process with the Epic EHR solution. Meanwhile, our whole organization is sort of shifting from a “best- of- breed” organization to a focused one, with a centralized EHR and centralized analytics. The challenge for us is how we do this well, without outstripping our organization’s ability to consume the information, and without investing in too much vaporware, so to speak.

What projects have you been working on lately, especially patient safety-related ones?

One thing we’ve been working on is a real-time sepsis monitor. It’s a very simple list: here are the patients at risk. The challenge is identifying the appropriate mechanism for integrating into the clinical workflow. The easy part is building it technically, though that’s still challenging. But the big challenge is, once you have the answer, putting in place the processes for improvement. Because there’s a lot of, “Ooh, wouldn’t it be cool to do this?” But then, afterwards, did the people asking for this realize there would be a 40-50 percent false positive rate? And the same questions go into alerts and reminders. And this sepsis alert provides clinical decision support. It says, for these patients, you should consider ordering these extra lab tests, and extra screenings. But if you deliver high false-positive information to physicians, they just ignore that information. So you have to figure out how you implement the screening process, and which people need to act on it.

So, some of the lab tests in this area will come out as false positives?